WORK PSYCHOLOGY
Understanding Human Behaviour in the Workplace
7th Edition
John Arnold
Iain Coyne
Ray Randall
Fiona Patterson
WORK PSYCHOLOGY
At Pearson, we have a simple mission: to help people
make more of their lives through learning.
We combine innovative learning technology with trusted
content and educational expertise to provide engaging
and effective learning experiences that serve people
wherever and whenever they are learning.
From classroom to boardroom, our curriculum materials, digital
learning tools and testing programmes help to educate millions
of people worldwide – more than any other private enterprise.
Every day our work helps learning flourish, and
wherever learning flourishes, so do people.
To learn more, please visit us at www.pearson.com/uk
WORK PSYCHOLOGY
Understanding Human Behaviour in the Workplace
7th Edition
John Arnold
Iain Coyne
Ray Randall
Fiona Patterson
Harlow, England • London • New York • Boston • San Francisco • Toronto • Sydney • Dubai • Singapore • Hong Kong
Tokyo • Seoul • Taipei • New Delhi • Cape Town • São Paulo • Mexico City • Madrid • Amsterdam • Munich • Paris • Milan
With
Carolyn Axtell
Bernard Burnes
Samuel Farley
Tom Kinirons
Ilke Inceoglu
Joanne Lyubovnikova
Rebecca Pieniazek
Emma-Louise Rowe
Joanne Silvester
Bart Wille
Madeleine Wyatt
PEARSON EDUCATION LIMITED
KAO Two
KAO Park
Harlow CM17 9SR
United Kingdom
Tel: +44 (0)1279 623623
Web: www.pearson.com/uk
First published 1991 (print)
Second edition published 1995 (print)
Third edition published 1998 (print)
Fourth edition published 2005 (print)
Fifth edition published 2010 (print)
Sixth edition published 2016 (print and electronic)
Seventh edition published 2020 (print and electronic)
© Financial Times Professional Limited 1998 (print)
© Pearson Education Limited 1991, 1995, 2005, 2010 (print)
© Pearson Education Limited 2016, 2020 (print and electronic)
The rights of John Arnold, Iain Coyne, Ray Randall and Fiona Patterson to be identified as authors of this work have been asserted
by them in accordance with the Copyright, Designs and Patents Act 1988.
The print publication is protected by copyright. Prior to any prohibited reproduction, storage in a retrieval system, distribution or
transmission in any form or by any means, electronic, mechanical, recording or otherwise, permission should be obtained from the
publisher or, where applicable, a licence permitting restricted copying in the United Kingdom should be obtained from the Copyright
Licensing Agency Ltd, Barnard’s Inn, 86 Fetter Lane, London EC4A 1EN.
The ePublication is protected by copyright and must not be copied, reproduced, transferred, distributed, leased, licensed or publicly
performed or used in any way except as specifically permitted in writing by the publisher, as allowed under the terms and conditions
under which it was purchased, or as strictly permitted by applicable copyright law. Any unauthorised distribution or use of this text
may be a direct infringement of the authors’ and the publisher’s rights and those responsible may be liable in law accordingly.
All trademarks used herein are the property of their respective owners. The use of any trademark in this text does not vest in the
author or publisher any trademark ownership rights in such trademarks, nor does the use of such trademarks imply any affiliation
with or endorsement of this book by such owners.
Contains public sector information licensed under the Open Government Licence (OGL) v3.0. http://www.nationalarchives.gov.
uk/doc/opengovernment-licence/version/3/.
Pearson Education is not responsible for the content of third-party internet sites.
The Financial Times. With a worldwide network of highly respected journalists, The Financial Times provides global business
news, insightful opinion and expert analysis of business, finance and politics. With over 500 journalists reporting from
50 countries worldwide, our in-depth coverage of international news is objectively reported and analysed from an
independent, global perspective. To find out more, visit www.ft.com/pearsonoffer.
ISBN: 978-1-292-26943-6 (print)
978-1-292-26945-0 (PDF)
978-1-292-26947-4 (ePub)
British Library Cataloguing-in-Publication Data
A catalogue record for the print edition is available from the British Library
Library of Congress Cataloging-in-Publication Data
Names: Randall, Ray, author.
Title: Work psychology : understanding human behaviour in the workplace /
Ray Randall [and 14 others].
Description: Seventh edition. | Harlow, United Kingdom : Pearson, 2020. |
Includes bibliographical references and index. | Summary: “Work
psychology is about people’s behaviour, thoughts and emotions related to
their work. It can be used to improve our understanding and management
of people including ourselves) at work. By work, we mean what people do
to earn a living. However, much of the content of this book can also be
applied to study, voluntary work and even leisure activities”— Provided
by publisher.
Identifiers: LCCN 2019047294 | ISBN 9781292269436 (hardback) | ISBN
9781292269474 (epub)
Subjects: LCSH: Psychology, Industrial. | Work--Psychological aspects.
Classification: LCC HF5548.8 .A78 2020 | DDC 158.7—dc23
LC record available at https://lccn.loc.gov/2019047294
10 9 8 7 6 5 4 3 2 1
24 23 22 21 20
Front cover: YAY Media AS/Alamy Stock Photo
Print edition typeset in 9/11.5 pt Stone Serif ITC Pro by SPi Global
Print edition printed in Slovakia by Neografia
NOTE THAT ANY PAGE CROSS REFERENCES REFER TO THE PRINT EDITION
BRIEF CONTENTS
Preface
About the authors
Acknowledgements
ix
xvi
xvii
1 The discipline of work psychology: An initial orientation
1
2 Individual differences
32
3 Selection: Analysing jobs, competencies and selection methods
68
4 Assessing performance at work
111
5 Attitudes at work
137
6 Work motivation
177
7 Training and development
200
8 Work-related stress and well-being
228
9 Groups, teams and teamwork
276
10 Leadership
316
11 Careers and career management
355
12 Understanding organisational change and culture
397
13 The psychology of dispersed work
431
Glossary
References
Index
453
471
606
CONTENTS
Preface
About the authors
Acknowledgements
ix
xvi
xvii
1 The discipline of work psychology: An initial orientation
1
Introduction • Basic psychology and work psychology • Analysing qualitative data
• The changing world of work • Summary • Relevant websites • Suggested further reading
2 Individual differences
32
Introduction • Traditional models of cognitive ability • Systems models of intelligence •
Trait views of personality • Socio-cognitive approaches to individual differences
• Summary • Relevant websites • Suggested further reading
3 Selection: Analysing jobs, competencies and
selection methods
68
Introduction • The design and validation process in selection • Job analysis data •
Reliability • Validation processes • Financial utility • Selection methods • The impact of
selection procedures on applicants • Summary • Relevant websites • Suggested further
reading
4 Assessing performance at work
111
Introduction • Purposes of performance assessment • Defining work performance
• Measuring work performance • Multi-Source Feedback (MSF) • Technology and
performance assessment • Improving performance • Incentives • The future of
performance assessment • Summary • Relevant websites • Suggested further reading
5 Attitudes at work
Introduction • What is an attitude? • Measuring attitudes • Attitudes and behaviour
• Job satisfaction • Organisational commitment • Employee turnover • The psychological
contract • Summary • Relevant websites • Suggested further reading
137
Contents
6 Work motivation
vii
177
Introduction • Content theories • Context theories • Integrating content and context
theories • Process theories • Goal striving • Integrative process theories • Integrative
theories • Summary • Discussion points • Relevant websites • Suggested further reading
7 Training and development
200
Introduction • How training has changed • The training cycle • Training needs analysis
• Training design • Self-regulated learning • Training methods • Training evaluation
• Training and development in practice • Summary • Relevant websites
• Suggested further reading
8 Work-related stress and well-being
228
Introduction • Work-related stress • Models and theories of work-related stress
• The consequences and costs of work stress • Psychological well-being (PWB)
• Factors linked to employee stress and well-being • Interventions to tackle work
stress and promote well-being • The popularity and effectiveness of interventions
• Problem-solving approaches to intervention • Summary • Individual and group
discussion questions • Relevant websites • Suggested further reading
9 Groups, teams and teamwork
276
Introduction • Foundations of group behaviour • Social norms • Roles • Understanding
group performance • Group decision-making • Work teams in the wild • Defining real
teams • Types of work team • The input-mediator-output model of team effectiveness
• Inputs • Mediators • Outputs • Future challenges for teamwork • Summary
• Five individual/group discussion points • Relevant websites • Suggested further reading
10 Leadership
316
Introduction • Some important questions about leadership • Early leader-focused
approaches to leadership • Point of integration • Contingency theories of leadership
• Transformational leadership and charisma • The limitations and ethics of
transformational leadership • What attributes of leaders really matter? • Gender and
leadership • Global leadership • Summary • Outlook: Digital transformation and leadership
• Suggested exercises • Suggested assignments • Relevant websites
• Suggested further reading
11 Careers and career management
355
Introduction • Psychologists and careers • The context of careers • Career forms and the
boundaryless career • Career management in organisations • Career choice • Job search
• Summary • Discussion points • Relevant websites • Suggested further reading
12 Understanding organisational change and culture
Introduction • The importance of change management • Changing organisational
culture • The planned approach to organisational change • The emergent approach to
organisational change • Organisational change: approaches and choices • Summary
• Class discussion points • Relevant websites • Suggested further reading
397
Contents
viii
13 The psychology of dispersed work
431
Core features of dispersed work • Designing for dispersion • Summary
• Activities/discussion points • Relevant website • Suggested further reading
Glossary
References
Index
453
471
606
PREFACE
Work psychology is about people’s behaviour, thoughts and emotions related to their work. It can be used
to improve our understanding and management of people (including ourselves) at work. By work, we mean
what people do to earn a living. However, much of the content of this book can also be applied to study,
voluntary work and even leisure activities.
All too often, work organisations have sophisticated systems for assessing the costs and benefits of
everything except their management of people. It is often said by senior managers that ‘our greatest
asset is our people’, but sometimes the people do not feel that they are being treated as if they were
valuable assets. People are complicated, and their views of themselves and their worlds differ: you will see
a great many references to individual differences throughout this book. People do not necessarily do what
others would like them to do. One reaction to all this is for managers to focus on things that don’t talk
back, such as profit and loss accounts or organisational strategy. Another is to adopt a highly controlling
‘do as I say’ approach to dealing with people at work. Either way, the thinking behind how people in
the workplace function, and how they might be managed, tends to be rather careless or simplistic.
Work psychologists seek to counter that tendency by studying carefully people’s behaviour, thoughts
and feelings regarding work. As well as developing knowledge and understanding for its own sake,
this also leads to insights about motivation, leadership, training and development, selection and many
other people-related aspects of management. Work psychologists are also concerned about the ethical
use of psychological theories and techniques, and their impact on the well-being and effectiveness of
individuals, groups and organisations.
This book is designed to appeal to readers in many different countries, especially in Europe and
Australasia. Judging by the feedback and sales figures for previous editions, we seem to have generally been
successful in appealing to a range of people in a range of places. We have tried to make the book suitable
both for people encountering the subject for the first time and for those who already have some familiarity
with it. Specifically, and in no particular order, we intend that this book should be useful for:
■
undergraduate students in psychology, taking one or more modules with names such as work
psychology, work and organisational psychology, business psychology, organisational psychology,
occupational psychology and industrial–organisational psychology;
■
undergraduate students in business and management taking one or more modules that might have
titles such as organisational behaviour, managing people or human resource management;
■
postgraduate (MSc, MBA, MA) and post-experience students in psychology or business/management
taking one or more modules with any or all of the titles listed above;
x
Preface
■
students taking professional qualifications, particularly (in the United Kingdom) those of the Chartered
Institute of Personnel and Development (CIPD);
■
students on undergraduate or postgraduate courses in other vocational subjects such as engineering,
whose curriculum includes some elements to do with managing people at work.
We aim to give clear and straightforward – but not simplistic – accounts of many key areas of contemporary work psychology. More specifically, we try to achieve several objectives in order to make this book
as useful as possible to its readers.
First, we seek to blend theory and practice. Without good theory, practice is blind. Without good practice,
theory is not being properly used. We therefore describe key theories and evaluate them where appropriate.
We examine how the concepts described can be applied. To help readers apply the content of this text,
we provide case studies and exercises. These can be used as classroom exercises, or as assignments for
individual students. Some guidance and suggestions about how to use these are included on the website
for this book at go.pearson.com/uk/he/resources.
Second, we try to present material at a level the reader should find intellectually stimulating, but not too
difficult. It is all too easy to use a slick, glossy presentation at the expense of good content. There is always
the temptation to resort to over-simple ‘recipes for success’ that insult the reader’s intelligence. On the
other hand, it is equally easy to lose the reader in unnecessarily complex debates. We hope that we avoid
both these fates (and that you will let us know if we do not!).
Third, we try to help the reader to gain maximum benefit from the book by providing several more aids to
learning. Each chapter begins with clearly stated learning outcomes and concludes with suggested assignments that reflect these outcomes. Multiple choice questions (and answers) are given on the companion
website to this volume. At the end of each chapter we provide suggestions for further reading. Throughout
the text we specify key learning points that express succinctly the main message of the preceding sections
of text. We include diagrams as well as text, in recognition that pictures can often express complex ideas in
an economical and memorable way. We highlight key controversies and debates because it is not uncommon
for research findings to point in different, apparently contradictory, directions. To help the reader, we weigh
up the arguments. Because no topic in work psychology sits in isolation from others, we also point towards
some of the most natural connections between different parts of the book. At the end of the book there is
a comprehensive glossary explaining in a concise way the meaning of lots of key words and phrases. There
is also a very long list of references, to enable interested readers to find more material if they wish.
Fourth, we have chosen topics that we judge to be the most useful to potential readers of this book.
Some usually appear in organisational behaviour texts, whereas others are generally found in books of a
more specifically psychological orientation. We believe we have found a helpful balance between these
two overlapping but different worlds, so that there should be plenty of relevant material both for people
who want to be psychologists and those who do not. The topics we cover in chapters or parts of chapters
include individual differences, employee selection, assessing work performance, attitudes at work, training
and development, teamwork, work motivation, stress and well-being at work, leadership, careers, organisational change and culture, dispersed working, the nature of work psychology as a discipline and profession
and how to design, conduct and understand research studies in work psychology.
Fifth, we provide up-to-date coverage of our material. There are currently exciting advances in many
areas of work psychology, and we try to reflect these. At the same time, where the old stuff is best and still
relevant, we include it. There is nothing to be gained by discussing recent work purely because it is recent,
especially if that comes at the expense of better quality and more useful material.
Sixth, we attempt to use material from many different parts of the world, and to point out cross-national
and cross-cultural differences where these seem particularly important. Much of the best research and
practice in work psychology originate from North America, but it is possible to go too far and assume that
nowhere else has contributed anything. No doubt we have our own blinkers, but we try to include perspectives from places other than North America, especially the UK and other European countries. Nevertheless,
the USA and Canada provide much valuable material. We make use of good research and theory originating
from as many countries as possible.
Preface
xi
Developments from the sixth edition
Readers familiar with the sixth edition of this text, published in 2016, may find it helpful if we describe the
changes we have made. These are more evident in some parts of the book than others. Readers familiar
with previous editions will readily recognise this book as a direct descendant of the others but will also
notice quite a few differences. As the book has grown over the years, so has the workload involved during
its revision. We say this every time, but it has taken us longer than we (and our families) would have liked.
The changes from the sixth edition reflect the fact that quite a lot has happened in work psychology
over the last few years. Reviewers commissioned by the publisher (plus users’ comments made direct to us)
helped us to see where rethinks were required.
The main change you will notice is that we have moved towards an editor-contributor model, increasing
the number of authors in the volume. This has allowed us to include diverse perspectives on topic areas
and to solicit knowledge from experts in each of the topic areas. As with previous editions, Ray Randall
worked on the attitudes at work (5) and work-related stress chapters (8), John Arnold provided a discussion
of careers and career management (11), Bernard Burnes handled the organisational change chapter (13)
and Caroline Axtell the dispersed work chapter (14). This time around Fiona Patterson was joined by Tom
Kinirons on the individual differences chapter (2) and by Emma-Louise Rowe for the chapter on personnel
selection. Similarly, Joanne Silvester has co-authored the performance at work (4) and training (7) chapters
with Madeleine Wyatt. New contributors to the book are: Iain Coyne on an initial orientation to work psychology (1); Samuel Farley and Rebecca Pieniazek on work motivation (6); Joanne Lyubovnikova on groups and
teams (9); and Ilke Inceoglu and Bart Wille on Leadership (10). All contributors are excellent researchers and
practitioners with an international reputation in their field and with extensive experience of intervention
in organisations. This allows them to describe how theory can be put into practice. We are proud of this
edition, but as with all preceding editions, your opinion is the one that really matters!
We have reduced the length of the book through a process of concise writing, combining chapters and
removing chapters. Specifically, feedback received from reviewers has resulted in Chapters 1 and 2 from the
sixth edition being combined into one, more focused, chapter covering what it takes to be a work psychologist. We have moved away from trying to write comprehensive chapters on philosophical underpinnings
of research, research methods and statistical analysis – especially given the excellent texts already out
there on these subjects – focusing more on the key knowledge and skills needed within work psychology.
We welcome your feedback on our new approach.
Additionally, new features include the provision of an opening and closing case study which provides an introduction to and a summary of each chapter. Also, multiple choice questions are now available on the companion
website to this volume. Given their growing influence on the world of work, we also directed chapter authors to
embed within their discussions information on diversity, technology and internationalisation. These contributions
vary in size and scope and manifest as case studies, key debates or discussions within the text. Hopefully, when
reading the chapters, you can understand the impact each of these current influences has within a topic area.
Feedback from readers of previous editions clearly indicated that they appreciated the clarity of style
and the combination of theoretical and practical considerations. They also very much valued the substantial
list of references, many quite recent. Naturally, we have tried to preserve these features. The style remains
the same and the reference list has been revised and updated. We are grateful for the feedback we have
received and wherever possible we have reflected it in this edition.
We have chosen to keep the learning features from the sixth edition that we hope will continue to prove
useful. The ‘Research methods in focus’ feature is designed to give readers a deep and detailed insight into
how work psychologists go about their business (whether it be research or practice). Hopefully, it helps
to demystify research methods by showing how these ‘come to life’ when applied to investigate or solve
important issues. Reading about research methods and statistics can be daunting and difficult so we hope
that this feature makes the activity more accessible and enjoyable. When read in conjunction with Chapter 1,
the Research Methods in Focus feature is designed to help readers to develop a solid knowledge base that
can be used to interpret, use and critique any other research they encounter.
xii
Preface
Psychologists often disagree about theories, the meaning of research findings, what constitutes good data
and too many other things to list in this preface without doubling the length of the book. This can be confusing.
Who are we to believe? Is the weight of evidence in favour of one side or the other? We use the ‘Key debate’
feature to help these debates stand out from the rest of the text. Without appointing ourselves as judge and
jury, we use this feature to present the arguments as we see them and to comment on the amount and quality
of evidence available. We can’t always solve the controversy, but we try to describe why it exists and how it
might one day be resolved. We feel these issues have important implications for practice: too often promises are
made about interventions before uncertainties about the underlying research have been adequately resolved.
As work psychologists progress in their careers, they often become more specialised, focusing on particular topics, organisational problems or research methods. There’s nothing wrong with this and it certainly
helps us to avoid overloading ourselves or attempting to practise in areas we know too little about. That said,
we feel it is very important to understand and appreciate the linkages and connections between different
topics in our field. A simple question demonstrates the point. Is it worth training someone to do a job if
a good selection process would identify somebody who is already adequately skilled? The answer to this
question is not a simple one (excuse us for sitting on the fence, just this once) but it does involve considering the research on both selection and training. To keep the book neatly organised and easy to navigate
we keep these topics separate. We simply couldn’t do justice to them unless a chapter is devoted to each.
The Point of Integration feature is designed to give the reader a quick insight into how these connections
work – both in terms of research and theory. We also use them to show how people from different fields of
expertise can be brought together to develop effective interventions.
From positive feedback provided by reviewers we have kept the large number of exercises and case studies. The vast majority of case studies have been updated to highlight the relevance that psychology has when
dealing with contemporary issues in the workplace. We have kept several of the exercises that are familiar to
instructors, but many are updated to ensure their relevance to, and resonance with, our intended readership.
Where we feel the case studies from earlier editions still have a strong relevance these have been retained.
We have retained the Stop to Consider feature. These are designed to encourage students to pause to
reflect on their learning. Their content is designed to foster critical thinking and cement learning. We hope
these prove useful for students who wish to go beyond an understanding of content to attempt further
analysis of the issues described.
Chapter 1 we feel is particularly useful for those new to the study of psychology. It is also designed to
illustrate how the basic assumptions made by psychologists, and the approaches they follow as a result,
find their way into work psychology. This chapter is now considerably shorter than before. The feedback
we received indicated that the content on workplace trends should be integrated into the relevant parts
of the book. We still provide a brief introduction to important cross-cutting themes including culture and
diversity, but these are now dealt with more comprehensively by considering them throughout the book.
Contemporary issues are integrated into the content of each chapter (including in the Exercises and Case
studies) in order to provide clearer illustrations of how research and theory can be applied.
In the sixth edition we referred to our discussions about removing the detailed material on research
methods in the previous Chapter 2. Although we decided against it last time, here we have listened once
again to reviewers and have made a change. Chapters 1 and 2 from the sixth edition have now been combined by Iain Coyne into one chapter focused on helping the reader understand Work Psychology from a
conceptual, applied, ethical and methodological perspective. As a result we have removed a lot of material
on research methods and statistics – given we are not attempting to be a research methods book – and
condensed the material into the key knowledge we feel individuals need to know when embarking on their
research and/or practice in work psychology. The chapter provides an overview of the discipline of work
psychology (e.g. where does it lie within psychology as a whole), moving on to explaining the key skills a work
psychologist needs to be successful in her/his role. We envisage this chapter to be one which orients the
reader to many of the ways data are collected, analysed and used in our discipline and offers a foundation
on which the other, topic-focused, chapters build on. Chapter 1 is a ‘quick reference’ guide to help readers
make sense of the source material within other chapters and as an aid for interpreting research they
encounter in the future.
Preface
xiii
Fiona Patterson and Tom Kinirons build on the chapter in the sixth edition, updating the content
on individual differences. They continue to discuss research and practice in general mental ability,
personality, emotional intelligence, creativity and innovation. They have included newer research
on environmental factors on intelligence in children, added in Ackerman’s (1996) PPIK theory of
intelligence, provided a Key Debate section reviewing measuring emotional intelligence and added
recent research looking at the role of intrinsic work motivation, psychological empowerment and
locus of control. The personality section has seen a large update of material with recent research
on personality and OCBs, personality across the lifespan, ‘dark-side’ and ‘bright side’ personality and
social media and personality included.
Chapter 3 by Fiona Patterson and Emma-Louise Rowe has broadly similar coverage to previous editions,
but has been extensively updated in parts to reflect the fact that with increasing use of technology, selection
is a rapidly developing area of research and practice. New and important advances such as the promise
offered by situational judgements tests and the more widespread use of online testing get increased coverage to reflect their growing importance. Particularly, these authors have included debates and case studies
on the growing use of artificial intelligence (AI) in selection and the impact this is likely to have on future
recruitment practices, especially regarding diversity and fairness issues.
Chapter 4, by Jo Silvester and Madeleine Wyatt, on assessing performance at work, considers this key
topic in work psychology in the light of trends in organisations and technology. There are several new case
studies and “stop to consider” exercises. The traditional consideration of how to rate performance is still
there, but the tone of the chapter has shifted towards practical applications of what is known about performance assessment. There is more attention to the dangers of counter-productive work behaviour, the
role of performance assessment not just in rewarding performance but developing it, and how performance
assessment is a political issue in organisations, always subject to being manipulated by different parties
for their own ends. Coverage of the now widely used techniques of multi-source assessment (sometimes
called 360-degree assessment) is updated, and instead of asking only whether it ‘works,’ Madeleine and Jo
ask the more fundamental underlying question of what is it for.
Chapter 5 on attitudes at work now contains minimal coverage of basic social psychology and is designed
to complement content in other parts of the text. It still provides a brief general discussion of the nature and
measurement of attitudes but now focuses heavily on two specific work-related attitudes: job satisfaction
and organisational commitment. Because attitudes are connected to many topics covered in other parts of
the text (such as motivation, work design and organisational change) the focus is now on the psychological
mechanisms that influence individual differences in attitudes, and that shape the development of attitudes
over time. The section on the psychological contract has been adjusted to include more recent work on this
topic. The sections on underemployment and unemployment have been moved to Chapter 8 to reflect the
connections between these topics and well-being. The section on employee turnover has been developed
to give a more thorough and detailed account of this complex and important topic.
The work motivation chapter (Chapter 6) by Samuel Farley and Rebecca Pieniazek re-structures, abbreviates and updates the chapter on motivation and job design in the sixth edition. Adopting a classification
of motivation theories into content, context and process, Samuel and Rebecca articulate the key features
of these theories, and compare and contrast them. By classifying job design as a contextual approach to
motivation, they integrate this topic more fully than before with the other chapter material. There is more
coverage than before of self-determination theory (SDT), which reflects the increasing prominence of this
theory. They consider the role of the theory of planned behaviour (TPB), which is not so often applied to
work motivation. TPB also provides a good lead-in to goal-setting theory. There are new cases and exercises
based on recent organisational and individual events.
The previous Design at work chapter by Don Harris has been removed from the seventh edition, but the
sixth edition version of this chapter will remain available on the companion website to this volume.
Jo Silvester and Madeleine Wyatt’s chapter on training and development (Chapter 7) retains the extensive coverage of the various elements of the training cycle and theories of learning from previous editions.
There is however extensive updating with many references to recent studies and reviews of aspects of the
literature on training and development. Digital methods of training, and training of staff to respect and
xiv
Preface
utilise diversity receive more attention. There are several new case studies and exercises including one based
on the dramatic 2018 rescue of the boys’ football team from a flooded cave in Thailand. The applications of
social learning theory are more explicitly considered, which leads into a discussion of self-regulated learning.
Recent attempts to analyse the training cycle are compared with Kirkpatrick’s traditional one, reflecting the
way in which this well-established topic area is moving with the times.
Following on from the changes in the sixth edition, Chapter 8 on stress and well-being includes more
coverage of the benefits of well-designed and managed work. Naturally, the theories that explain the
effects of ‘bad work’ are still included but some receive a little less prominence as more sophisticated
and useful theories emerge. More in-depth coverage is given to strong theories and models that help
us to describe, explain and predict the impact of work on individual well-being (such as Conservation of
Resources). As a result there is less – but we hope still enough – coverage of extensive ‘lists’ of causes and
consequences of stress. As more evidence emerges, we continue to focus on the design delivery and
evaluation of a variety of interventions: these issues are complex and provide significant challenges
and opportunities for practitioners and researchers. The topics of underemployment and unemployment
now feature in this chapter.
Implementing teamwork is generally seen as being a good idea with potential benefits for the organisation and for the employee. Chapter 9 by Joanne Lyubovnikova now includes an extended critical evaluation
of the potential benefits and risks of implementing and managing teams in contemporary work settings.
There is a more in-depth discussion of the psychological mechanisms that underpin the effects of teamworking. To reflect the composition of teams in many organisations, an increased emphasis has been placed
on the discussion of the effects on team functioning of culture, diversity and individual differences. The
critical analysis of the experimental research that informs much of the applied work has been broadened,
but there is now much more analysis of the practical implications of the findings that have emerged from
this influential research. The input-mediator-output-input model takes more prominence in the chapter
and provides a framework for a detailed discussion of how the effects of teamwork unfold over time. In
particular, the discussion of mediators is developed in much more detail than in previous editions. Finally, the
differences between effective and ineffective teams (e.g. pseudo teams) now frame much of the content.
Bart Wille’s and Ilke Inceoglu’s leadership chapter (Chapter 10) retains the overall structure of the equivalent chapter in the sixth edition. However, there are significant changes and updates. The discussion of
some of the older topics like contingency theories has been shortened to reflect their declining influence.
A number of new studies have been used to flag up emerging issues. These include a closer look at the
different contexts in which leadership takes place and the consequences of contextual variables for the
conduct and consequences of leadership; a more empirically grounded analysis of the ways in which transformational leadership might produce positive outcomes, and the role of digital technologies and big data
in the activities and ethics of leadership.
John Arnold’s revisions in Chapter 11 (Careers and career management) have led to a somewhat slimmed
down account of a topic area that sits at the intersections between many others. A lot of the older material
has been summarised more succinctly or omitted. Many research articles published since the sixth edition
of this book are incorporated into the narrative, leading to substantial updating of what is known and theorised about core areas of career. There is increased coverage of theorisation of the effects of organisational
career interventions, the different perspectives or orientations people bring to their career, new ways of
classifying types of occupation, and international job assignments. New case studies and exercises highlight
the growing roles of social media and artificial intelligence in career management. There is also more explicit
consideration of whether career is an elite concept, and how the work lives of people who do not consider
themselves to have careers can be analysed.
Bernard Burnes’ chapter on change (Chapter 12) retains its structure from the sixth edition. It has been
very well-received in the past as an authoritative review of the field. Several updates reflect the contemporary challenges that workers and organisational decision-makers face when responding to rapid changes
in technology. It also considers in more detail the links between diversity in the workforce and various
approaches to organisational change and important new thinking around the topics of change and culture.
In several parts of the chapter the updated material provides new insights about how existing theories and
Preface
xv
models can be put to better use. The chapter also identifies where new approaches might be needed to deal
with the situations and pressures being faced by modern work organisations. Several promising emerging
theories are identified, discussed and evaluated.
Carolyn Axtell’s chapter on the psychology of dispersed working (Chapter 13) is once again the final
chapter because it integrates many of the issues discussed throughout the text. It is a good demonstration of the relevance and utility of psychological theory in contemporary work settings. The chapter
shows that if we are to make the best use of new ways of working, research from various different areas
of work psychology need to be applied. This topic also has its own research agenda and presents new
challenges for work psychologists. These have produced innovative and exciting approaches to research
and intervention.
As before, we welcome feedback and dialogue about this book. Please direct it to Iain Coyne, School of
Business and Economics, Sir Richard Morris Building, Loughborough University, Leicestershire, Leicester,
LE11 3TU, UK (i.j.coyne@lboro.ac.uk). Thank you for reading this preface, and please now carry on into the
rest of the book!
ABOUT THE AUTHORS
Editors
John Arnold is Professor of Organisational Behaviour, School of Business and Economics, Loughborough
University, UK.
Iain Coyne is Senior Lecturer in Organisational Psychology, School of Business and Economics,
Loughborough University, UK.
Ray Randall is Senior University Teacher in Work Psychology, Institute of Work Psychology, Sheffield
University Management School, UK.
Fiona Patterson is Founder Director of the Work Psychology Group, Visiting Researcher, University of
Cambridge and Visiting Professor, City University, London, UK and the University of Nottingham, UK.
Chapter authors
Carolyn Axtell is Senior Lecturer, Institute of Work Psychology, Sheffield University Management School, UK.
Bernard Burnes is Professor of Organisational Change, Stirling Management School, UK.
Samuel Farley is Lecturer in Organizational Psychology, Leeds University Business School, UK.
Ilke Inceoglu is Professor of Organisational Behaviour and HR Management, University of Exeter Business
School, UK.
Joanne Lyubovnikova is Senior Lecturer in Organisational Behaviour, University of Liverpool
Management School, UK.
Rebecca Pieniazek is Lecturer in Organisational Psychology and Organisational Behaviour, Leeds University Business School, UK.
Emma-Louise Rowe is a Consultant Psychologist at Work Psychology Group, Derby, UK.
Joanne Silvester is Professor of Human Resource Management and Organisational Behaviour in the
School of Business and Economics, Loughborough University, UK.
Tom Kinirons is a Consultant Psychologist at Work Psychology Group, Derby, UK.
Bart Wille is Assistant Professor in Industrial-Organizational Psychology, Department of Personnel
Management, Work and Organizational Psychology. Ghent University, Belgium.
Madeleine Wyatt is Senior Lecturer in Human Resources, Kent Business School, UK.
ACKNOWLEDGEMENTS
We thank Victoria Tubb and Dipika Rungta at Pearson Education for their immense patience in waiting for the
manuscript, and for doing everything they could to help it along without putting the authors under too much
pressure, even when we thoroughly deserved it. Iain Coyne would like to thank his son Joshua, partner Vicki and
his parents Vic and Doreen for their love, guidance and sense of fun. Ray Randall would like to thank his wife Kirsty,
children Izzy and Owen, and his parents Lal and Bob for their patience, advice, understanding and support. John
Arnold would also like to thank his parents Ann and Rev, and his wife Helen, for their love, wisdom and support.
Fiona Patterson would like to thank her family and loved ones for all their support and especially her team at
Work Psychology Group.
Publisher’s acknowledgements
Photo Credit(s):
1 Shutterstock: Pavlo S/Shutterstock; 32 Shutterstock: Lightspring/Shutterstock; 68 Shutterstock: Olivier Le
Moal/Shutterstock; 111 Shutterstock: Donskarpo/Shutterstock; 137 Shutterstock: Triff/Shutterstock; 177
Shutterstock: Studiovin/Shutterstock; 200 Shutterstock: Lightspring/Shutterstock; 203 Alamy Images: Allan
Cash Picture Library/Alamy Stock Photo; 203 Pexels GmbH: Pexels GmbH, https://images.pexels.com/photos/1204649/pexels-photo-1204649.jpeg?auto=compress&cs=tinysrgb&dpr=2&h=750&w=1260; 228 Shutterstock: Fotana/Shutterstock; 276 Shutterstock: Mauricio Graiki/Shutterstock; 316 Shutterstock: M.Stasy/
Shutterstock; 355 Shutterstock: Dimj/Shutterstock; 397 Shutterstock: Africa Studio/Shutterstock; 431 Shutterstock: This Is Me/Shutterstock
Text Credit(s):
4 P. Alex Linley: Linley, A.P. (2006) ‘Coaching research: Who? what? where? when? why?’, International Journal of
Evidence Based Coaching and Mentoring, 4, 1–7; 5 Academy of Management: Briner, R.B., Denyer, D. and Rousseau, D.M. (2009) ‘Evidence-based management: Concept clean-up time?’, Academy of Management Perspectives,
23, 19–32; 5 American Psychological Association: Lowman, R.L. (2012) ‘The scientist-practitioner consulting
psychologist’, Consulting Psychology Journal: Practice and Research, 64, 151–6; 7 The British Psychological Society: Adapted from British Psychological Society (2017) Standards for the Accreditation of Masters and Doctoral
Programmes in Occupational Psychology. Leicester: BPS; 8–9 British Psychological Society: Reprinted with permission from British Psychological Society (2018) Code of Ethics and Conduct. Leicester: BPS; 22 Sage Publications: Greene, J.C., Caracelli, V.J. and Graham, W.F. (1989) ‘Towards a conceptual framework for mixed methods
evaluation designs’, Educational Evaluation and Policy Analysis, 11, 255–74; 26 Taylor & Francis: Braun, V. and
xviii
Acknowledgements
Clarke, V., (2006) ‘Using thematic analysis in psychology’, Qualitative research in psychology, 3, 77–101; 33 The
Financial Times Limited: Richard Waters (2019) “Emotional Intelligences Takes Aicha Evans to Top of Silicon
Valley”, Financial Times, 18 January 2019 © Financial Times Limited 2019 All rights reserved; 36 The New Republic:
Boring, E.C. (1923) ‘Intelligence as the tests test it’, New Republic, 35, 35–7; 37 Association for Science
Education: Reproduced by permission of NFER originally published in Heim, A. W. (1968). AH5 group test of
intelligence. London: National Foundation for Educational Research.; 41 Elsevier: Bickley, P.G., Keith, T.Z. and Wolfe,
L.M. (1995) ‘The three-stratum theory of cognitive abilities: Test of the structure of intelligence across the life
span’, Intelligence, 20, 309–28; 42 Multi Health Systems, Inc: Bar-On, R. (1997) The Emotional Intelligence
Inventory (EQ-i): Technical manual. Toronto: Multi-Health Systems; 42 Penguin Random House: Goleman, D.
(1995) Emotional Intelligence. New York: Bantam Books; 43 John Wiley & Sons: Mayer, J.D., Salovey, P. and Caruso,
D.R. (2000) ‘Emotional intelligence as zeitgeist, as personality, and as a mental ability’, in R. Bar-On and J.D.A. Parker
(eds), Handbook of Emotional Intelligence. San Francisco, CA: Jossey-Bass; 49 American Psychological Association: Judge, T.A., Rodell, J.B., Klinger, R.L., Simon, L.S. and Crawford, E.R. (2013) ‘Hierarchical representations of the
five-factor model of personality in predicting job performance: Integrating three organising frameworks with two
theoretical perspectives’, Journal of Applied Psychology, 98, 875; 55 Hogrefe Publishing Group: Costa, P.T. and
McCrae, R.R. (2006) Revised NEO Personality Inventory (NEO PI-R) Manual (UK edition). Oxford: Hogrefe; 56 Pearson Education: Conn, S. and Rieke, M. (eds) (1994) 16PF-5. Technical Manual. Champaign, IL: Institute for Personality and Ability Testing; Matthews, G. and Deary, I.J. (1998) Personality Traits. New York: Cambridge University Press;
56–57 Pearson Education: Foley, B. and Cox, A. (2013) ‘Work organisation and innovation – Case study: Nottingham University Hospitals NHS Trust, UK’, Dublin: European Foundation for the Improvement of Living and Working
Conditions; 57 European Union: Foley, B. and Cox, A. (2013) ‘Work organisation and innovation – Case study:
Nottingham University Hospitals NHS Trust, UK’, Dublin: European Foundation for the Improvement of Living and
Working Conditions; 58 Sage Publications: Anderson, N., Potonik, K., & Zhou, J. (2014). Innovation and creativity
in organizations: A state-of-the-science review, prospective commentary, and guiding framework. Journal of management, 40(5), 1297–1333; 62 John Wiley & Sons: Pervin, L.A. (1980) Personality: Theory, assessment and
research, 3rd edition. New York: John Wiley; 64–65 The Financial Times Limited: Seb Murray, ‘Graduates with tech
and finance skills in high demand’, FT Case Study, © Financial Times limited 2019 All rights reserved; 69 The
Financial Times Limited: Murrary, H. (2018) UK recruiters use technology to offer ‘preemptive’ jobs, April 6 ©
Financial Times limited 2019 All rights reserved; 76 American Psychological Association: Flanagan, J.C. (1954)
‘The critical incident technique’, Psychological Bulletin, 51, 327–58; 83 The Financial Times Limited: Boyde, E.
(2017). Robo-recruiters are quick to replicate human bias. The Financial Times, December 7 © Financial Times
limited 2019 All rights reserved; 106 John Wiley & Sons: Patterson, F., Knight, A., Dowell, J., Nicholson, S., Cousans,
F., & Cleland, J. (2016). How effective are selection methods in medical education? A systematic review. Medical
education, 50(1), 36–60; 112 Harvard Business Publishing: Buckingham, M. and Goodall, A. (2015) ‘Reinventing
performance management’, Harvard Business Review, 93(4), 40–50; 114 Taylor & Francis: Bennett, W., Lance, C. E.,
& Woehr, D. J. (Eds.). (2014). Performance Measurement: Current Perspectives and Future Challenges. New York:
Psychology Press; 117 John Wiley & Sons: Bolino, M.C., Klotz, A.C., Turnley, W.H. and Harvey, J. (2013) ‘Exploring
the dark side of organizational citizenship behavior’, Journal of Organizational Behavior, 34(4), 542–59; 117 Taylor
& Francis: Organ, D.W. (1997) ‘Organisational citizenship behaviour: It’s construct clean-up time’, Human Performance, 10, 85–98; 119 Sage Publications: Schein, V. E. (1975). Relationships between sex role stereotypes and
requisite management characteristics among female managers. Journal of Applied Psychology, 60(3), 340; 122
American Psychological Association: Smith, P.C. and Kendall, L.M. (1963) ‘Retranslation of expectations: An
approach to the construction of unambiguous anchors for rating scales’, Journal of Applied Psychology, 47(2),
149–55; 130 Elsevier: Tomczak, D. L., Lanzo, L. A., & Aguinis, H. (2018). Evidence-based recommendations for
employee performance monitoring. Business Horizons, 61(2), 251–259; 135 Macmillan Publishers: Chris Hadfield
(2015) ‘An astronaut’s guide to life on earth’. London: Pan Books; Used with permission from Macmillan Publishers
139 The Financial Times Limited: Tim Smedley (2016), ‘Happy workplaces help companies perform better’, The
Financial Times, 26 February © Financial Times limited 2019 All rights reserved; https://www.ft.com/content/
6081b1fc-d0b2-11e5-92a1-c5e23ef99c77; 140 McGraw-Hill Education: Secord, P.F. and Backman, C.W. (1969)
Social Psychology. New York: McGraw-Hill; 141 Sage Publications: George, J.M. and Jones, G.R. (1997) ‘Experiencing work: Values, attitudes and moods’, Human Relations, 50, 393–416; 141, 144 SAGE Publications: Pratkanis,
A.R. and Turner, M.E. (1994) ‘Of what value is a job attitude? A sociocognitive analysis’, Human Relations, 47,
1545–76; 145 Elsevier: Adapted from Ajzen, I. and Madden, J.T., (1986) ‘Prediction of goal-directed behavior:
Acknowledgements
xix
Attitudes, intentions, and perceived behavioral control’, Journal of Experimental Social Psychology, 22, 453–74;
147 American Psychological Association: Judge, T.A., Weiss, H.M., Kammeyer-Mueller, J.D. and Hulin, C.L. (2017)
‘Job attitudes, job satisfaction, and job affect: A century of continuity and of change’, Journal of Applied Psychology,
102, 356–374; 148 Cary Cooper: © 1988 Cooper, Sloan and Williams. Reproduced with permission of the publisher ASE, College Lane, Hatfield, Herts, AL10 9AA, UK; 149 John Wiley & Sons: Taber, T.D. and Alliger, G.M. (1995)
‘A task-level assessment of job satisfaction’, Journal of Organisational Behaviour, 16, 101–21; 150 Sage Publications: Thompson, E.R. and Phua, F.T.T. (2012) ‘A brief index of affective job satisfaction’, Group and Organisation
Management, 37, 275–307; 152 American Psychological Association: Adapted from Timothy A. Judge, Carl J.
Thoresen, Joyce E. Bono and Gregory K. Patton (2001) The Job Satisfaction-Job Performance Relationship: A Qualitative and Quantitative Review. Psychological Bulletin, 2001, Vol. 127. No. 3. 376–407, Copyright (c) The American
Psychological Association, Inc.; 153 American Psychological Association: Hülsheger, U.R., Alberts, H.J.E.M.,
Feinholdt, A. and Lang, J.W.B. (2013) ‘Benefits of mindfulness at work: The role of mindfulness in emotion regulation,
emotional exhaustion, and job satisfaction’, Journal of Applied Psychology, 98, 310–25; 153–154 IOS Press:
Gronlund, A. and Oun, I. (2018) ‘The gender-job satisfaction paradox and the dual-earner society: Are women (still)
making work-family trade-offs?’, Work, 59, 535–45; 155 Elsevier: Mowday, R., Steers, R. and Porter, L. (1979) ‘The
measurement of organisational commitment’, Journal of Vocational Behaviour, 14, 224–47; 155 John Wiley &
Sons: Griffin, R.W. and Bateman, T.S. (1986) ‘Job satisfaction and organisational commitment’, in C.L. Cooper and
I.T. Robertson (eds), International Review of Industrial and Organisational Psychology. Chichester: John Wiley; 156
Elsevier: Ng, T.W.H. (2015) ‘The incremental validity of organisational commitment, organisational trust, and
organisational identification’, Journal of Vocational Behaviour, 88, 154–63; 157 Elsevier: Mowday, R., Steers, R.
and Porter, L. (1979) ‘The measurement of organisational commitment’, Journal of Vocational Behaviour, 14,
224–47; 157 John Wiley & Sons: Allen, N.J. and Meyer, J.P. (1990) ‘The measurement and antecedents of affective,
continuance and normative commitment to the organisation’, Journal of Occupational Psychology, 63, 11–18; 157
John Wiley & Sons: Warr, P., Cook, J. and Wall, T. (1979) ‘Scales for the measurement of some work attitudes and
aspects of psychological well-being’, Journal of Occupational Psychology, 52, 129–48; 158–159 Elsevier: Morrow,
P.C. (2011) ‘Managing organisational commitment: Insights from longitudinal research’, Journal of Vocational
Behaviour, 79, 18–35; 162 INFORMS: Hausknecht, J.P. and Holwerda, J.A. (2013) ‘When does employee turnover
matter? Dynamic member configurations, productive capacity, and collective performance’, Organisation Science,
24, 210–25; 163 American Psychological Association: Hom, P.W., Mitchell, T.R., Lee, T.W. and Griffeth, R.W. (2012)
‘Reviewing employee turnover: Focusing on proximal withdrawal states and an expanded criterion’, Psychological
Bulletin, 138, 831–58; 167 John Wiley & Sons: Robinson, S.L. and Rousseau, D.M. (1994) ‘Violating the psychological contract: Not the exception but the norm’, Journal of Organisational Behaviour, 15, 245–59; 167 John Wiley
& Sons: Rousseau, D.M. (2001) ‘Schema, promise and mutuality: The building blocks of the psychological contract’,
Journal of Occupational and Organizational Psychology, 74, 511–42; 167 Sage Publications: Herriot, P. (1992)
The Career Management Challenge: Balancing individual and organisational needs. London: Sage; 170 John Wiley
& Sons: Bal, P.M., Hofmans, J. and Polat, T. (2017) ‘Breaking psychological contracts with the burden of workload:
a weekly study of job resources as moderators’, Applied Psychology: An International Review, 66, 143–67; 170
Sage Publications: Rousseau, D.M. (1995) Psychological Contracts in Organisations. London: Sage Publications;
171 Elsevier: Griepa, Y. and Vantilborgh, T. (2018) ‘Reciprocal effects of psychological contract breach on counterproductive and organizational citizenship behaviors: The role of time’, Journal of Vocational Behavior, 104,
141–153; 174 The Financial Times Limited: Delphine Strauss (2019) ‘Employers face moves to strengthen gig
worker rights’, The Financial Times, 19th July, © Financial Times limited 2019 All rights reserved; https://www.
ft.com/content/602afc46-aa24-11e9-b6ee-3cdf3174eb89; 178 Mark Long: Mark Long, Director and Master
Distiller of Cuckoo Gin in 2016; 178 Taylor & Francis: Kanfer, R. (1990) ‘Motivation theory and industrial and
organizational psychology’, Handbook of Industrial and Organizational Psychology, 1(2), 75–130; 179 American
Psychological Association: Kanfer, R., Frese, M. and Johnson, R. E. (2017) ‘Motivation related to work: A century
of progress’, Journal of Applied Psychology, 102(3), 339, 340; 180 HarperCollins: Maslow, A. H. (1954). Motivation
and Personality. New York: Harper and Row; 180 SAGE Publications: Cooke, B., Mills, A.J. and Kelley, E.S. (2005)
‘Situating Maslow in Cold War America’, Group and Organisation Management, 30, 129–52; 181 Academy of
Management: Cropanzano, R., Bowen, D.E. and Gilliland, S.W. (2007) ‘The management of organisational justice’,
Academy of Management Perspectives, 21(4), 36; 181 John Wiley & Sons: Robinson, S.L. and Morrison, E.W.
(2000) ‘The development of psychological contract breach and violation: A longitudinal study’, Journal of Organisational Behaviour, 21, 525–46; 181 McGraw-Hill Education: Mowday, R.T. (1991) ‘Equity theory predictions of
xx
Acknowledgements
behaviour in organisations’, in R.M. Steers and L.W. Porter (eds), Motivation and Work Behaviour, 5th edition. New
York: McGraw-Hill; 183 American Psychological Association: Grant, A.M. (2008) ‘Does intrinsic motivation fuel
the prosocial fire? Motivational synergy in predicting persistence, performance, and productivity’, Journal of
Applied Psychology, 93, 49; 183 Elsevier: Ryan, R.M. and Deci, E.L. (2000) ‘Intrinsic and extrinsic motivations:
Classic definitions and new directions’, Contemporary Educational Psychology, 25, 54–67; 183 SAGE Publications:
Van den Broeck, A., Ferris, D.L., Chang, C.H. and Rosen, C.C. (2016) ‘A review of self-determination theory’s basic
psychological needs at work’, Journal of Management, 42(5), 1195–229; 184 John Wiley & Sons: Kauppila, O.P.
(2018) ‘How does it feel and how does it look? The role of employee motivation in organizational learning type’,
Journal of Organizational Behavior, 39, 941–55; 185 John Wiley & Sons: Sagie, A., Elizur, D. and Yamauchi, A. (1996)
‘The structure and strength of achievement motivation: A cross-cultural comparison’, Journal of Organisational
Behaviour, 17, 431–44; 185 John Wiley & Sons: Winter, D.G. (2010) ‘Why achievement motivation predicts success
in business but failure in politics: The importance of personal control’, Journal of Personality, 78, 1637–68; 185
Oxford University Press: Murray, H.J. (1938) Explorations in Personality. Oxford: Oxford University Press, 80–81,
quoted in Landy, F.J. (1989) The Psychology of Work Behaviour, 4th edition. Homewood, IL: Brooks/ Cole Publishing
Co, 73; 187 Elsevier: Hackman, J.R. and Oldham, G.R. (1976) ‘Motivation through the design of work: Test of a
theory’, Organisational Behaviour and Human Performance, 16, 250–79; Hackman, J.R. and Oldham, G.R. (1980)
Work Redesign. Reading, MA: Addison-Wesley; 187 HarperCollins: Taylor, F. W. (1911). Principles of Scientific
Management. New York: Harper; 189 John Wiley & Sons: Challenger, R., Leach, D.J., Stride, C.B. and Clegg, C.W.
(2012) ‘A new model of job design: Initial evidence and implications for future research’, Human Factors and
Ergonomics in Manufacturing and Service Industries, 22(3), 197–212; 190 Academy of Management:
Wrzesniewski, A. and Dutton, J. E. (2001) ‘Crafting a job: Revisioning employees as active crafters of their work’,
Academy of Management Review, 26, 179–201; 190 American Psychological Association: Kanfer, R., Frese, M.
and Johnson, R. E. (2017) ‘Motivation related to work: A century of progress’, Journal of Applied Psychology, 102(3),
338; 190 American Psychological Association: Kanfer, R., Frese, M. and Johnson, R. E. (2017) ‘Motivation related
to work: A century of progress’, Journal of Applied Psychology, 102(3), 342, 343; 190 American Psychological
Association: Tims, M., Bakker, A.B. and Derks, D. (2013) ‘The impact of job crafting on job demands, job resources,
and well-being’, Journal of Occupational Health Psychology, 18, 230–40; 191 American Psychological Association: Kanfer, R., Frese, M. and Johnson, R. E. (2017) ‘Motivation related to work: A century of progress’, Journal of
Applied Psychology, 102(3), 343; 192 American Psychological Association: Locke, E. A., Shaw, K. N., Saari, L. M.
& Latham, G. P. (1981). ‘Goal setting and task performance 1969–1980’, Psychological Bulletin, 90, 125–52; 193
Academy of management: Ordóñez, L.D., Schweitzer, M. E., Galinsky, A. D. and Bazerman, M. H. (2009) ‘Goals gone
wild: The systematic side effects of overprescribing goal setting’, Academy of Management Perspectives, 23(1),
6–16; 193 American Psychological Association: Kanfer, R., Frese, M. and Johnson, R. E. (2017) ‘Motivation related
to work: A century of progress’, Journal of Applied Psychology, 102(3), 338; 194 John Wiley & Sons: Montano, D. E.,
& Kasprzyk, D. (2015). Theory of reasoned action, theory of planned behavior, and the integrated behavioral model.
Health behavior: Theory, research and practice, 95–124; 195 John Wiley & Sons: Vroom, V.H. (1964) Work and
Motivation. Chichester: John Wiley; 195 Pearson Education: Ajzen, I., & M. Fishbein. (1980). Understanding Attitudes and Predicting Social Behavior. Englewood Cliffs, New Jersey: Prentice-Hall; 195 SAGE Publications: Vancouver, J.B. (2012) ‘Rhetorical reckoning: A response to Bandura’, Journal of Management, 38(2), 465–74; 196
Academy of Management: Locke, E.A. and Latham, G.P. (2004) ‘What should we do about motivation theory? Six
recommendations for the twenty-first century’, Academy of Management Review, 29, 388–403; 197–198 Pearson
Education: BBC. (29 June 2018). Timeline: How the BBC gender pay story has unfolded. Retrieved 6 May 2019,
from https://www.bbc.co.uk/news/entertainment-arts-42833551; 201 Apakorn Yukongkaew: Apakorn Yukongkaew, Navy SEAL Commander Rear Adm; 202 Sage Publications: Salas, E., Tannenbaum, S. I., Kraiger, K., &
Smith-Jentsch, K. A. (2012). The science of training and development in organizations: What matters in practice.
Psychological science in the public interest, 13(2), 74–101; 216 John Wiley & Sons: Baldwin, T. T., & Ford, J. K.
(1988). Transfer of training: A review and directions for future research. Personnel Psychology, 41(1), 63–105;
217–218 McGraw-Hill Education: Kirkpatrick DL. (1967). Evaluation of training. In Craig RL, Bittel LR (Eds.),
Training and Development Handbook (pp. 87–112). New York: McGraw-Hill; 220 Sage Publications: Sitzmann, T.,
& Weinhardt, J. M. (2018). Training engagement theory: A multilevel perspective on the effectiveness of
work-related training. Journal of Management, 44(2), 732–756; 229 The Financial Times Limited: Wang Xuegiao
and Tom Hancock, ( 2019) Overdoing it: the cost of China’s long-hours culture, 17 January © Financial Times limited
2019 All rights reserved; https://www.ft.com/content/d5f01f68-9cbc-11e8-88de-49c908b1f264; 232 Health
Acknowledgements
xxi
and safety executive: Work-related stress, Health and safety executive, www.hse.gov.uk/stress; 233 American
Psychological Association: De Lange, A.H., Taris, T.W., Kompier, M.A.J., Houtman, I.L.D. and Bongers, P.M. (2003)
‘“The very best of the millennium”: Longitudinal research and the demand-control-(support) model’, Journal of
Occupational Health Psychology, 8, 282–305; 234 Hachette Book Group: Karasek, R. and Theorell, T. (1990)
Healthy Work: Stress, productivity and the reconstruction of working life. New York: Basic Books; 237 Sage Publications: Halbesleben, J.R.B., Neveu, J.P., Paustian-Underdahl, S.C. and Westman, M. (2014) ‘Getting to the “COR”:
Understanding the role of resources in Conservation of Resources Theory’, Journal of Management, 40,
1334–1364; 238 Springer: Schaufeli, W.B., Salanova, M., González-Romá, V. and Bakker, A.B. (2002) ‘The measurement of engagement and burnout: A two sample confirmatory factor analytic approach’, Journal of Happiness
Studies, 3, 71–92; 239 Academy of Management: Wrzesniewski, A. and Dutton, J.E. (2001) ‘Crafting a job: Revisioning employees as active crafters of their work’, Academy of Management Review, 26, 179–201; 241 Thieme
Medical Publishers: Kuper, H., Hemingway, M. and Marmot, M. (2002) ‘Systematic review of prospective cohort
studies of psychosocial factors in the etiology and prognosis of coronary heart disease’, Seminars in Vascular
Medicine, 2(3), 267–314; 242 John Wiley & Sons: Maslach, C. and Jackson, S.E. (1981) ‘The measurement of
experienced burnout’, Journal of Occupational Behaviour, 2, 99–113; 244 Elsevier: Wright, T.A. and Cropanzano,
R. (2004) ‘The role of psychological well-being in job performance: A fresh look at an age-old quest’, Organisational
Dynamics, 33, 338–51; 245 Sage Publications: Fredrickson, B.L. and Joiner, T. (2002) ‘Positive emotions trigger
upwards spirals towards emotional well-being’, Psychological Science, 13, 172–5; 246 American Psychological
Association: Boehm, J.K. and Kubzansky, L.D. (2012) ‘The heart’s content: The association between positive psychological well-being and cardiovascular health’, Psychological Bulletin, 138, 655–91; 246 American Psychological Association: Nielsen, K. and Cleal, B. (2010) ‘Predicting flow at work: Investigating the activities and job
characteristics that predict flow states at work’, Journal of Occupational Health Psychology, 15, 180–90; 246 The
American Psychosomatic Society: Chida, Y. and Steptoe, A. (2008) ‘Positive psychological well-being and mortality: A quantitative review of prospective observational studies’, Psychosomatic Medicine, 70, 741–56; 249 Academy of Management: Greenhaus, J.H. and Beutell, N.J. (1985) ‘Sources and conflict between work and family roles’,
Academy of Management Review, 10, 76–88; 251 Elsevier: Paul, K.I. and Moser, K. (2009) ‘Unemployment impairs
mental health: Meta-analyses’, Journal of Vocational Behaviour, 74, 264–82; 253 Sage Publications: Salin, D.
(2003) ‘Ways of explaining workplace bullying: A review of enabling, motivating and precipitating structures and
processes in the work environment’, Human Relations, 56, 1213–32; 258 New Harbinger Publications: Flaxman,
P., Bond, F. and Livheim, F. (2013) The Mindful and Effective Employee: An acceptance and commitment therapy
training manual for improving well-being and performance. Oakland, CA: New Harbinger Publications; 259
NFER-NELSON: Schwarzer, R., and Jerusalem, M. (1995) ‘Generalised Self-Efficacy scale’ in J. Weinman, S. Wright,
and M. Johnston (eds), Measures in Health Psychology: A user’s portfolio. Causal and control beliefs (pp. 35–37).
Windsor, UK: NFER-NELSON; http://userpage.fu-berlin.de/~health/engscal.htm; 260 Association for Psychological Science: Judge, T.A. (2009) ‘Core self-evaluations and work success’, Current Directions in Psychological Science, 18, 58–62; 265 Taylor & Francis: Kröll, C., Doebler, P. and Nüesch, S. (2017) ‘Meta-analytic evidence of the
effectiveness of stress management at work’, European Journal of Work and Organizational Psychology, 26,
677–693; 266 Academy of management: Aeon, B. and Aguinis, H. (2017) ‘It’s about time: New perspectives and
insights on time management’, Academy of Management Perspectives, 31, 309–30; 271–273 The Financial Times
Limited: Darren Dodd (2019), Social networking and the increasing burden on mental health FT.com, 17 January
© Financial Times limited 2019 All rights reserved; 277–278 National Institute for Health Research: Adapted
with permission from West, M., Alimo-Metcalfe, B., Dawson, J., El Ansari, W., Glasby, J., Hardy, G. et al., (2012).
Effectiveness of Multi-Professional Team Working (MPTW) in Mental Health Care. Final Report. NIHR Service
Delivery Organisation Programme; 282 HarperCollins: Sherif, M. (1936). The psychology of social norms. New
York: Harper; 283 American Psychological Association: Bond, R. & Smith, P.B. (1996). Culture and conformity: A
meta-analysis of studies using Asch’s (1952b, 1956) line judgment task. Psychological Bulletin, 119, 111–137; 285
Cengage Learning: Zimbardo, P. G., Haney, C., Banks, W. C., & Jaffe, D. (1982). The psychology of imprisonment. In
J. C. Brigham & L. Wrightsman (Eds.), Contemporary issues in social psychology (4th ed., pp. 230–235). Monterey,
CA: Brooks/Cole; 286 Elsevier: Steiner, I. D. (1972). Group process and productivity. New York: Academic Press. p.
8; 293 Houghton Mifflin Harcourt: Janis, I.L. (1982a) Groupthink. Boston, MA: Houghton Mifflin; 295 Aston
University: Richardson, J. (2010). An Investigation of the Prevalence and Measurement of Teams in Organisations:
The Development and Validation of the Real Team Scale. Aston University, Birmingham, UK; 295 John Wiley & Sons:
Kozlowski, S.W.J., & Bell, B.S. (2003). Work groups and teams in organizations. In W.C. Borman & D. R. Ilgen & R.
xxii
Acknowledgements
Klimoski (Eds.) Industrial/Organizational Psychology, Vol. XII: John Wiley & Sons Ltd. p. 334; 298 Annual Review of
Psychology: Adapted from Ilgen, D.R., Hollenbeck, J.R., Johnson, M. and Jundt, D. (2005) ‘Teams in organizations:From input-process output models to IMOI models’, Annual Review of Psychology, 56, 517–43; 298 Journal of
Management: Adapted from Mathieu, J., Maynard, M.T., Rapp, T. and Gilson, L. (2008) ‘Team effectiveness
1997–2007: A review of recent advancements and a glimpse into the future’, Journal of Management, 34: 410–76;
298–299 John Wiley & Sons: Anderson, N.R., & West, M.A. (1998). Measuring climate for work group innovation:
development and validation of the team climate inventory. Journal of Organizational Behavior, 19, 235–58; 301
Sage Publications: Weiss, M., & Hoegl, M. (2016). Effects of relative team size on teams with innovative tasks: An
understaffing theory perspective. Organizational Psychology Review, 6(4), 324–351; 302 John Wiley & Sons:
Harrison, D.A., & Humphrey, S.E. (2010). Designing for diversity or diversity for design? Tasks, interdependence,
and within-unit differences at work. Journal of Organizational Behavior, 31, 328–337. p. 334; 303 Elsevier: Martins,
L. L., Gilson, L. L., & Maynard, M. T. (2004). Virtual teams: What do we know and where do we go from here? Journal
of Management, 30, 805–835; 304 Sage Publications: Stevens, M.J. and Campion, M.A. (1994) ‘The knowledge,
skills and ability requirements for teamwork: Implications for human resource management’, Journal of Management, 20, 503–30; 305 American Psychological Association: van Knippenberg, D., De Dreu, C.K.W., & Homan, A.C.
(2004). Work group diversity and group performance: an integrative model and research agenda. Journal of Applied.
Psychology, 89, 1008–1022; 306 Academy of Management: Marks, M.A., Mathieu, J.E. and Zaccaro, S.J. (2001) ‘A
temporally based framework and taxonomy of team processes’, Academy of Management Review, 26, 356–76;
312 American Psychological Association: Marks, M.A., DeChurch, L.A., Mathieu, J.E., Panzer, F.J., Alonso, A. (2005).
Teamwork in multiteam systems. Journal of Applied Psychology, 90, 964–971; 313–314 John Arnold: Case Study
written for this book by Joanne Lyubovnikova (2019). Adapted by John Arnold; 318 Sage Publications: House, R.J.,
Hanges, P.J., Javidan, M., Dorfman, P.W., & Gupta, V. (eds). (2004). Culture, Leadership, and Organisations: The GLOBE
study of 62 societies. London: Sage; 318 University of Cape Town: April, K., Macdonald, R. and Vriesendorp, S.
(2000) Rethinking Leadership. Cape Town: University of Cape Town Press; 319 John Wiley & Sons: Fiedler, F.E.
(1995). Cognitive resources and leadership performance. Applied Psychology: An International Review, 44, 5–28;
320 Sage Publications: Meindl, J.R., Ehrlich, S.B., & Dukerich, J.M. (1985). The romance of leadership. Administrative Science Quarterly, 30, 78–102; 322 American Psychological Association: Judge, T.A., Colbert, A.E., & Ilies, R.
(2004a). Intelligence and leadership: A quantitative review and test of theoretical propositions. Journal of Applied
Psychology, 89, 542–52; 323 McGraw-Hill Education: Fleishman, E.A. (1969) Leadership Opinion Questionnaire
Manual. Henley-on-Thames: Science Research Associates; 324 SAGE Publications: Gastil, J. (1994) ‘A definition
and illustration of democratic leadership’, Human Relations, 47, 953–75; 326 McGraw-Hill Education: Fiedler, F.E.
(1967). A Theory of Leadership Effectiveness. New York: McGraw-Hill; 327 University of Pittsburgh: Vroom, V.H.,
& Yetton, P.W. (1973). Leadership and Decision Making. Pittsburgh, PA: Pittsburgh Press; 328 Prentice Hall: Vroom,
V.H. and Jago, A.G. (1988) The New Leadership: Managing participation in organisations. Englewood Cliffs, NJ:
Prentice Hall; 329 Louis van Gaal: Louis van Gaal, Manchester united manager, August 2015; 335 American
Psychological Association: Bass, B.M. (1985) Leadership and Performance: Beyond expectations. New York: Free
Press; 336 John Wiley & Sons: Conger, J.A., Kanungo, R.N., & Menon, S.T. (2000). Charismatic leadership and follower effects. Journal of Organizational Behavior, 21(7), 747–67; 337 Annual Reviews: Antonakis, J., Bastardoz,
N., Jacquart, P., & Shamir, B. (2016). Charisma: An ill-defined and ill-measured gift. In F. P. Morgeson (Ed.), Annual
Review of Organizational Psychology and Organizational Behavior, Vol 3, pp. 293–319; 339 Sage Publications:
Wang, G., Os., I-S., Courtright, S.H., and Colbert, A.E. (2011) ‘Transformational leadership and performance across
criteria and levels: A meta-analytic review of 25 years of research’, Group and Organisational Management, 36,
223–70; 342 John Wiley & Sons: Derived from DeRue et al. (2011) ‘Trait and behavioural theories of leadership:
An integration and meta-analytic test of their relative validity’, Personnel Psychology, 64(1), 7–52 p. 33; 344
American Psychological Association: Barreto, M.E., Ryan, M.K., & Schmitt, M.T. (2009). The Glass Ceiling in the
21st Century: Understanding barriers to gender equality. New York: American Psychological Association; 346 Sage
Publications: Hofstede, G. (1980). Culture’s Consequences: International differences in work-related values. London: Sage Publications; 346–347 Elsevier: Den Hartog, D.N., House, R.J., Hanges, P.J., RuizQuintanilla, S.A., &
Dorfman, P.W. (1999). Culture-specific and cross-culturally generalizable implicit leadership theories: Are attributes
of charismatic/transformational leadership universally endorsed?. Leadership Quarterly, 10, 219–57; 348 Elsevier:
Dorfman, P., Javidan, M., Hanges, P., Dastmalchian, A., & House, R. (2012). GLOBE: A twenty year journey into the
intriguing world of culture and leadership. Journal of World Business, 47, 504–18; 349 American Psychological
Association: Bass, B.M. (1997). Does the transactional–transformational leadership paradigm transcend
Acknowledgements
xxiii
organisational and national boundaries?. American Psychologist, 52(2), 130–9; 349 John Wiley & Sons: Smith,
P.B., Misumi, J., Tayeb, M., Peterson, M., & Bond, M. (1989). On the generality of leadership style measures across
cultures. Journal of Occupational Psychology, 62, 97–109; 356 The Financial Times Limited: Adapted from an
article by Andrew Hill (2019) Finding your next job: Out with CVs, in with memes, The Financial Times, May 27 ©
Financial Times limited 2019 All rights reserved; 357 SAGE Publications: John, A. (1997: 16). Managing Careers
Into the 21st Century. SAGE; 358 Emerald Group Publishing Limited: Lyons, S.T., Schweitzer, L., & Ng, E.S. (2015).
How have careers changed? An investigation of changing career patterns across four generations. Journal of
Managerial Psychology, 30(1), 8–21; 358 John Wiley & Sons: Direnzo, M.S., Greenhaus, J.H. and Weer, C.H. (2015)
‘Relationship between protean career orientation and work-life balance: A resource perspective’, Journal of Organisational Behaviour, 36, 538–60; 360 Oxford University Press: Arthur, M.B. and Rousseau, D.M. (1996) ‘The boundaryless career as a new employment principle’, in M.B. Arthur and D.M. Rousseau (eds), The Boundaryless Career:
A new employment principle for a new organisational era. Oxford: Oxford University Press; 362 Emerald Group
Publishing Limited: Gubler, M., Arnold, J. and Coombs, C. (2014a) ‘Organisational boundaries and beyond: A new
look at the components of a boundaryless career orientation’, Career Development International, 19, 641–67
Reproduced from permission of Emerald Publishing; 363 Elsevier: Briscoe, J.P. and Hall, D.T. (2006) ‘The interplay
of boundarylessness and protean careers: Combinations and implications’, Journal of Vocational Behaviour, 69,
4–18; 363 SAGE Publications: Adapted from Hall, D.T. (2002) Careers In and Out of Organisations. Thousand Oaks,
CA: Sage Publications; 363 SAGE Publications: Hall, D.T. (2002) Careers In and Out of Organisations. Thousand
Oaks, CA: Sage Publications; 364 John Wiley & Sons: Schein, E.H. (1993) Career Anchors: Discovering your real
values, revised edition. London: Pfeiffer and Co; 367 The Work Foundation: Hirsh, W. and Jackson, C. (2004)
Managing Careers in Large Organisations. London: The Work Foundation; 368 Emerald Group Publishing Limited:
Baruch, Y. (2004) ‘Transforming careers – from linear to multidirectional career paths: Organisational and individual
perspective’, Career Development International, 9, 58–73; 368 SAGE Publications: Inkson, K., Dries, N. and Arnold,
J. (2014) Understanding Careers, 2nd edition. London: Sage; 368 Taylor & Francis: Arnold, J. and Cohen, L. (2013)
‘Careers in organisations’, in W.B. Walsh, M.L. Savickas, and P.J. Hartung (eds), Handbook of Vocational Psychology,
4th edition. Mahwah, NJ: Erlbaum; 369–370 British Psychological Society: Briner, R.B., Robertson, I., Patterson,
F., Anderson, N., Dietmann, A., Cassell, C. and Kandola, B. (2010) ‘Occupational psychology in a changing world’,
The Psychologist, 23, 892–99; 370 Elsevier: Adapted from Bagdadli, S., & Gianecchini, M. (2019). Organizational
career management practices and objective career success: A systematic review and framework. Human Resource
Management Review, 29(3), 353–370; 371: Adapted from Shen, Y., Demel, B., Unite, J., Briscoe, J.P., Hall, D.T.,
Chudzikowski, K. et al. (2015) ‘Career success across 11 countries: Implications for international human resource
management’, International Journal of Human Resource Management, 26(13), 1753–78; 376–377 The Financial
Times Limited: Adapted from story posted by Alicia Clegg (2015) The cleaner who soared all the way to the board,
5 January © Financial Times limited 2019 All rights reserved; 378 Sage Publications: Andreas Hirschi, Noemi Nagy,
Franziska Baumeler, Claire S. Johnston, and Daniel Spurk (2018) Assessing Key Predictors of Career Success: Development and Validation of the Career Resources Questionnaire, Vol. 26(2) 338–358, SAGE Publications; 380 American Psychological Association: Adapted from Duffy, R. D., Blustein, D. L., Diemer, M. A., & Autin, K. L. (2016). The
psychology of working theory. Journal of Counseling Psychology, 63(2), p. 129; 381 The Financial Times Limited:
Adapted and edited with permission from an article quoting Elizabeth Uviebinene, Financial Times, July 3 2019.
Uviebinene grew up in Peckham, London, and is co-author of the book Slay in your Lane: The black girl’s Bible. ©
Financial Times limited 2019 All rights reserved; 382 Houghton Mifflin Harcourt: Parsons, F. (1909) Choosing a
Vocation. Boston, MA: Houghton Mifflin; 382 PAR, Inc: Holland, J.L. (1997) Making Vocational Choices: A theory
of vocational personalities and work environment, 3rd edition. Odessa, FL: Psychological Assessment Resources
Inc.; 385 John Wiley & Sons: Arnold, J. (2004) ‘The congruence problem in John Holland’s theory of vocational
decisions’, Journal of Occupational and Organizational Psychology, 77, 95–113; 386 American Psychological
Association: Su, R., Tay, L., Liao, H. Y., Zhang, Q., & Rounds, J. (2019). Toward a dimensional model of vocational
interests. Journal of Applied Psychology, 104(5), 690–714; 386 John Wiley & Sons: Blustein, D.L. (1997) ‘A
context-rich perspective of career exploration across the life roles’, Career Development Quarterly, 45, 260–74;
388 John Wiley & Sons: Xu, Hui; Bhang, Cecile H. (2019) The Structure and Measurement of Career Indecision: A
Critical Review, Career Development Quarterly, Volume 67, Issue1, March 2019, p. 2–20; 390 Emerald Group
Publishing Limited: O’Neil, D.A. and Bilimoria, D. (2005) ‘Women’s career development phases: Idealism, endurance
and reinvention’, Career Development International, 10, 168–89; 390 SAGE Publications: Sullivan, S.E. and
Crocitto, M. (2007) ‘The developmental theories: A critical examination of their continuing impact on careers research’,
xxiv
Acknowledgements
in H. Gunz and M. Peiperl (eds), Handbook of Career Studies. London: Sage; 392 The Financial Times Limited:
Adapted with permission from Emma Jacobs (2019) “Mid-life does not have to be a crisis for careers” Financial
Times 31 January © Financial Times limited 2019 All rights reserved; 394 Veronique Biecques: Véronique Biecques,
Director of recruitment at Orange; 398–399 The Financial Times Limited: Adapted from Kuchler, H. (2017)
‘Airbnb’s gender diversity efforts put Uber in shade’, Financial Times, 6 April FT.com © Financial Times limited 2019
All rights reserved; 399 McKinsey & Company: Hunt, V; Prince, S; Dixon-Fyle, S and Yee, L (2018) Delivering through
Diversity. McKinsey & Company. Available at www.mckinsey.com; 399 The Financial Times Limited: Murray, S.
(2018) Billions of tech customers ‘don’t look what and middle class’. Financial Times, 14 November. Available
at FT.com © Financial Times limited 2019 All rights reserved; 401 Allen & Unwin: Eldridge, J.E.T. and Crombie, A.D.
(1974) A Sociology of Organisations. London: George Allen and Unwin; 401 Pearson Education: Brown, A. (1995)
Organisational Culture. London: Pitman; 401 SOUTHWESTERN Publishing Group: Cummings, T.G. and Worley, C.G.
(2001) Organisation Development and Change, 7th edition. Cincinnati, OH: South-Western College Publishing; 402
Penguin Random House: Handy, C. (1986) Understanding Organisations. Harmondsworth: Penguin; 402–403
Human Synergistics International: Jones, Q., Dunphy, D., Fishman, R., Larne, M. and Canter, C. (2006) In Great
Company: Unlocking the secrets of cultural transformation. Sydney, Australia: Human Synergistics; 403 Oxford
University Press: Hatch, M.J. (1997) Organisation Theory: Modern, symbolic and postmodern perspectives. Oxford:
Oxford University Press; 403 Sage Publications: Barratt, E.S. (1990) ‘Human resource management: Organisational
culture’, Management Update, 2(1), 21–32; 404 Human Synergistics International: Jones, Q., Dunphy, D., Fishman,
R., Larne, M. and Canter, C. (2006) In Great Company: Unlocking the secrets of cultural transformation. Sydney,
Australia: Human Synergistics; 404 Sage Publications: Meek, V.L. (1988) ‘Organisational culture: Origins and
weaknesses’, Organisation Studies, 9(4), 453–73; 404 Taylor & Francis: Watson, T.J. (1997) Sociology, Work and
Industry, 3rd edition. London: Routledge; 405 Pearson Education: French, W.L. and Bell, C.H. (1995) Organisation
Development, 5th edition. Englewood Cliffs, NJ: Prentice-Hall; 406 HarperCollins: Lewin, K. (1939) ‘When facing
danger’, in G.W. Lewin (ed.), Resolving Social Conflict. London: Harper and Row; 407 Henry Holt and Company:
Lewin, K. (1947b) ‘Group decisions and social change’, in T.M. Newcomb and E.L. Hartley (eds), Readings in Social
Psychology. New York: Henry Holt; 407 Massachusetts Institute of Technology: Lewin, K. (1947a) ‘Frontiers in
group dynamics’, in D. Cartwright (ed.), Field Theory in Social Science. London: Social Science Paper-backs; 409
Cengage Learning: Cummings, TG and Worley, CG (2015) Organization Development and Change (10th Edition).
Cengage: Stamford, CT; 410 Emerald Group Publishing Limited: Wooten, K.C. and White, L.P. (1999) ‘Linking OD’s
philosophy with justice theory: Postmodern implications’, Journal of Organisational Change Management, 12,
7–20; 410–411 The Financial Times Limited: Groom, B (2013) ‘Women stuck in executive slow lane’, Financial
Times, 11 December, , p 22 © Financial Times limited 2019 All rights reserved; 412 Geist Science: Pettigrew, A.M.
(1987) ‘Context and action in the transformation of the firm’, Journal of Management Sciences, 24(6), 649–70;
412 Harvard Business Publishing: Weick, K.E. (2000) ‘Emergent change as a universal in organisations’, in M.
Beer and N. Nohria (eds), Breaking the Code of Change. Boston, MA: Harvard Business School Press; 412 Springer:
Hayes, J. (2002) The Theory and Practice of Change Management. Basing-stoke: Palgrave; 412 Taylor & Francis:
Dawson, P. (2003) Reshaping Change: A processual perspective. London: Routledge; 413 Sage Publications:
Finstad, N. (1998) ‘The rhetoric of organisational change’, Human Relations, 51, 717–40; 413 Sage Publications:
Johnson, G. (1993) ‘Processes of managing strategic change’, in C. Mabey and B. Mayon-White (eds), Managing
Change, 2nd edition. London: Open University/Paul Chapman Publishing; 414 Harvard Business Publishing:
Kotter, J.P. (1996) Leading Change. Boston, MA: Harvard Business School Press; 414 Sage Publications: Benjamin, G.
and Mabey, C. (1993) ‘Facilitating radical change’, in C. Mabey and B. Mayon-White (eds), Managing Change, 2nd
edition. London: Open University/Paul Chapman Publishing; 414 Taylor & Francis: Wilson, D.C. (1992) A Strategy
of Change: Concepts and controversies in the management of change. London: Routledge; 415 Harvard Business
Publishing: Martin, R. (2000) ‘Breaking the code of change: Observations and critique’, in M. Beer and N. Nohria
(eds), Breaking the Code of Change. Boston, MA: Harvard Business School Press; 416 Pearson Education: Clarke,
L. (1994) The Essence of Change. London: Prentice Hall; 419 Sage Publications: Hendry, C. (1996) ‘Understanding
and creating whole organisational change through learning theory’, Human Relations, 48(5), 621–41; 419 Taylor
& Francis: Dawson, P. (2011) ‘The contribution of the processual approach to the theory and practice of organisational change’, in D.M. Boje, B. Burnes and J. Hassard (eds), The Routledge Companion to Organisational Change.
London: Routledge; 419 Taylor & Francis: Whittington, R. (1993) What is Strategy and Does it Matter? London:
Routledge; 420–421 The Financial Times Limited: Jackson, G (2018) ‘G20: Digital change and gig economy rewire
the world of work’, ‘Financial Times, 29 November FT.com © Financial Times limited 2019 All rights reserved;
Acknowledgements
xxv
422 Taylor & Francis: Stickland, F. (1998) The Dynamics of Change: Insights into organisational transition from
the natural world. London: Routledge; 427 Thomson Press: Sorge, A. (1997) ‘Organisation behaviour’, in A. Sorge
and M. Warner (eds), The IEBM Handbook of Organisational Behaviour. London: International Thompson Business
Press; 428–429 The Financial Times Limited: Adapted from Hill, A (2016) ‘Toppling Bureaucracy’, Financial Times,
15 April, p. 9. © Financial Times limited 2019 All rights reserved; 432–433 Haymarket Media Group: Carolyn Axtell,
Jo Wheeler, Malcolm Patterson and Anna Leach (now Meachin), ‘From a distance’, People Management, 25 March
2004, pp. 39–40. Reprinted with permission @ Haymarket Media Group; 435 Academy of management: Carlson,
J.R. and Zmud, R.W. (1999) ‘Channel expansion theory and the experiential nature of media richness perceptions’,
Academy of Management Journal, 42, 153–70; 435 Informs: Sproull, L. and Kiesler, S. (1986) ‘Reducing
social-context cues: Electronic mail in organisational communication’, Management Science, 32, 1492–512. 447
Elsevier: Ross, L. (1977) ‘The intuitive psychologist and his shortcoming: Distortions in the attribution process’,
Advances in Experimental Social Psychology, 10, 174–220; 449 Informs: Cramton, C.D. (2001) ‘The mutual knowledge problem and its consequences for dispersed collaboration’, Organisation Science, 12, 246–371;
CHAPTER 1
The discipline of work
psychology
An initial orientation
LEARNING OBJECTIVES
After studying this chapter, you should be able to:
1 describe the discipline of work psychology by being able to:
■
examine the relationship between basic and applied psychology;
■
understand the notion of pragmatic science;
■
specify the key skills of those practising work psychology.
2 understand research and practice in work psychology by being able to:
■
identify the main sources of information about work psychology research and practice;
■
distinguish between opposing philosophies in the conduct of psychological research;
■
describe the key features, advantages and disadvantages of different research designs
used by work psychologists;
■
specify the various methods of data collection used in research by work psychologists;
■
detail the difference between quantitative and qualitative data;
■
identify the main methods used for analysing both qualitative and quantitative data;
■
identify why diversity, technology and internationalisation are important cross-cutting
themes in work psychology.
2
Work Psychology
Opening case study
Assessing problems in recruitment and retention of nurses in
a hospital unit
You have been called in by the Clinical Lead of a unit operating within a hospital to help examine and
assess why they struggle to recruit and retain nurses. The unit operates across three sites – each
which deal with the same condition, although offer different services (e.g. provision for patients visiting just for treatment or wards for those with serious medical conditions). There is a high turnover
rate of nursing staff and, although the unit regularly advertises for replacement posts, these tend
to go unfilled. When filled, most new nursing staff leave within the first year. Additionally, trainee
nurses tend not to select this unit as part of their practical experience, further restricting the nurse
provision in the unit.
• What skills would you need to undertake this project?
• Who, in the hospital, would you need to consult with?
• How would you go about collecting information to help on this project?
• What ethical issues may arise?
• How would you design the research, and collect and analyse good data?
These questions will be considered in the current chapter. For now, just think about the questions
and the scenario. We will return to the case at the end of the chapter and offer an approach to
undertaking a project such as this.
Introduction
In this chapter we aim to help the reader gain a broad understanding of the nature of work
psychology and the context within which it operates before tackling more specific topics later
in the book. We start with a brief description of the discipline of psychology as a whole and
discuss the links between what we call basic and applied psychology, with work psychology
positioned as one branch of applied psychology. We frame work psychology using the notions
of pragmatic science and evidence-based management and give an account of the different
labels sometimes given to work psychology and the topics it covers. We then move to discussing the skills of a work psychologist with specific emphasis on ethical and critical evaluations
skills. Next, we consider the need for work psychologists to understand and enact research
skills to be able to function effectively. We focus this around several specific features: the best
sources of good knowledge about work psychology (apart from this book of course!); philosophical positions taken by work psychologists; methods used in research; design of research;
and analysis of research data. In the latter part of this chapter we look briefly at important
cross-cutting themes in work psychology: diversity, culture and technology and the potential
impact these issues will have on the research and practice of a work psychologist. These three
aspects feature throughout all the chapters in the book.
Basic psychology and work psychology
Psychology has been defined in various ways. Perhaps the simplest yet most informative
definition is that provided long ago by Miller (1966): ‘the science of mental life’. Mental life
refers to three phenomena: behaviours, thoughts and emotions. Most psychologists these
days would agree that psychology involves all three.
Chapter 1 The discipline of work psychology
Table 1.1
3
Basic areas of psychology and how they apply to the working context
Subdiscipline
Basic
Application to work
Reference
Biological
Biological bases of behaviour,
neuropsychology, evolutionary
psychology
The impact of simulated night shift
work on insulin sensitivity and risk of
Type 2 Diabetes
Bescos et al. (2018)
Cognitive
Thought processes such as attention,
memory, learning, perception and
language
The role of cued recall in recalling
non-routine complex cognitive skills
Frank and Kluge
(2018)
Developmental
The ways in which people grow and
change psychologically throughout life
Improve the working lives of the over
50s in the UK
Altman (2015)
Individual differences
How people differ from each other
psychologically and how those
differences can be measured
The relationship between personality
traits and exposure to bullying at work
Nielsen et al. (2017)
Social
How our behaviours, thoughts and
emotions affect, and are affected by,
other people
Virtual teams and the impact of
virtuality on team communication
Marlow et al. (2017)
The discipline of psychology can be divided into several subdisciplines, each with its own
distinctive focus. Collectively they can be termed basic psychology. There are several ways of
splitting psychology. Table 1.1 illustrates five ways in which basic psychology can be divided.
The table provides examples of applied work research that can be framed within each subdiscipline (see the Reference in column 4).
Key learning point
The five areas of basic psychology all contribute ideas and techniques to work psychology.
Work psychology is defined in terms of its context of application (see Figure 1.1), and an area
of applied psychology. As you will see throughout this book, work psychologists use concepts,
theories and techniques derived from all areas of basic psychology. Table 1.1 illustrated some
examples of research applying ideas from psychology to the working context. These areas
are not mutually exclusive: studying people at work from several different perspectives is
often necessary in order to understand fully the issue being examined. For example, Gomez
and Taylor (2018) explore the role of National culture (therefore using an individual difference approach) and in-group/out-group status (social) in strategies used to resolve group
conflict. Results indicated a Mexican sample of MBA students showed more of a preference
for confronting conflict using social influence and negotiating than a US sample – thereby
illustrating the impact of culture on conflict resolution approaches.
4
Work Psychology
Exercise 1.1
Think of a work-related topic or problem that interests you. It may be something that you have come
across in your studies or heard about in the media. Consider which of the basic areas of psychology
you might draw on if you were going to research this area. Remember, you are likely to draw on more
than one. Consider how each would help you to better understand the issue.
As shown in Figure 1.1, areas of applied psychology use ideas and information from basic
psychology. Conversely, they can also contribute ideas and information to the development
of basic psychology. Sometimes theory from basic psychology can directly contribute to the
solution of real-world problems. The need for solutions to difficult and complex real-world
problems can also stimulate developments in basic psychology. Applied psychology (rather
than basic psychology) offers theories and techniques directly applicable to practical problems and real-life situations. In fact, it might be argued applied psychologists are interested
in solving problems, while basic psychologists are motivated to develop knowledge for its
own sake. Thus, there may be a danger that the areas of applied psychology will fail to reflect
advances in basic psychology: some more theoretically inclined psychologists fail to take
sufficient account of work in applied psychology, or of current real-world issues.
The approach that many work psychologists aim for is what Anderson et al. (2001) refer
to as pragmatic science. This type of work addresses problems of practical importance and
does so using rigorous methodology. In pragmatic science, good research and practice are
almost indistinguishable. We have tried to ensure that, where possible, the vast majority of
the material cited in this text falls into this category. It refers to research that is done well,
that has been subject to review and critique, and stood up to tests of its quality. At the same
time, the research is useful and relevant: it helps organisations. Linley (2006: 3) summarises
the benefits of this approach:
Good research questions have the potential to bridge the academic-practitioner divide very
effectively, because they catalyse the interests, needs and aspirations of both parties through
delivering findings that are not only academically sound and valued, but that also offer practical
application and advancement.
Areas of basic
psychology
•
•
•
•
•
Areas of applied
psychology
•
•
•
•
•
•
Arrows represent flows of information and ideas
Figure 1.1
The relationship between areas of psychology
Chapter 1 The discipline of work psychology
5
Key learning point
Pragmatic science gives us the best of both worlds: good research that has clear practical relevance.
More recently the term evidence-based management (EBMgt) has been used to describe
better how theory and practice can be connected. Briner at al. (2009) argue that effective
decisions in organisations need to combine:
■
Evaluated good quality evidence. Research findings that have been subjected to critical review by independent experts, for example through systematic review and
meta-analysis of findings from numerous studies, can provide a solid evidence base for
decisions. The use of results from single studies to inform practice in new settings would
be especially risky.
■
The experience and judgements of practitioners who have some reliable insight, often
developed through experience and reflection, into the issue being addressed.
■
Input from those likely to be affected by the decision (stakeholders), including what is
important to them and what they prefer.
■
Information drawn from the organisational context, for example data held by the organisation about the issue being tackled, information about the pressures and opportunities
facing the organisation and so on.
The term scientist-practitioner is often used to describe people who integrate research
and practice to good effect. Lowman (2012) indicates that these are people who work with
important issues and measure important outcomes of their interventions. They are also good
at sharing their knowledge. This sounds relatively straightforward, but as Lowman points out,
‘the needs of clients do not necessarily derive from what research has chosen to study nor
does the path of science always focus on practical applications’ (2012: 153). No matter how
knowledgeable or qualified the psychologist appears to be, they will need to work with a range
of stakeholder groups if their advice is to have the best chance of success.
Key learning point
The terms evidence-based management and scientist-practitioner emphasise the need for psychologists to make good use of quality research and to connect with the various end-users of their work
and other knowledgeable professionals.
We should note that the effective application of theory and research is rarely straightforward. There are many tensions that can draw researchers and practitioners away from
a pragmatic science model, including policy priorities, organisational context, values and
demographics.
What is work psychology?
Work psychology has at least two distinct roots within applied psychology. One resides in a
pair of traditions that have often been termed ‘fitting the person to the job’ (FPJ) and ‘fitting
the job to the person’ (FJP). The FPJ tradition manifests itself in employee selection, training
and vocational guidance (e.g. see Chapters 3, 7 and 11). These endeavours have in common an
6
Work Psychology
attempt to achieve an effective match between job and person by concentrating on the latter.
The FJP tradition focuses instead on the job, and the design of tasks, equipment and working
conditions that suit a person’s physical and psychological characteristics (e.g. see Chapter 6).
You will see the influence of both approaches throughout this text.
The FPJ and FJP traditions essentially concern the relationship between individuals and
their work. The other root of work psychology can be loosely labelled human relations (HR).
It is concerned with the complex interplay between individuals, groups, organisations and
work. It therefore emphasises social factors at work much more than FPJ and FJP (e.g. see
Chapter 9).
Key learning point
There are two important traditions in work psychology. The concern of how jobs can be fitted to people (FJP) and how people can be fitted to jobs (FPJ); and the human relations approach emphasising
individuals’ experiences and interpretations at work.
One source of confusion you may experience is that the discipline of work psychology has
a lot of different names. The label often used in the USA is industrial/organizational psychology (or I/O psychology for short). In the UK, it is often called occupational psychology, but this
term is uncommon in most other countries. In the UK the title of Occupational Psychologist
is protected in law, meaning that only appropriately qualified persons can use the title.
Throughout Europe, increasing use is made of the psychology of work and organisations and
work and organisational psychology to describe the area. Just to confuse things further, some
specific parts of the field are given labels such as vocational psychology, managerial psychology,
business psychology and personnel psychology. Meanwhile, there are also some bigger areas of
study (e.g. human resource management and organisational behaviour) to which psychology
contributes greatly.
Our advice for the confused reader is: don’t panic! The differences between these labels do
mean something to some people who work in the field but should not unduly worry most of
us. In the main, the same content areas are covered regardless of the label used. We use the
term work psychology because of its simplicity, and because to us it encompasses the individual
and organisational levels of analysis.
With the confusion of labelling behind us, we can now be more specific in the areas
(or knowledge domains) in which work psychologists apply psychological ideas to work
and organisations. Figure 1.2 illustrates the five areas of work psychology as defined by
the Standards for the Accreditation of Masters and Doctoral Programmes in Occupational
Psychology published by the British Psychological Society (BPS, 2017). You will see
that all these areas are covered within this volume, some (e.g. psychological assessment
at work) with specific chapters and others (e.g. leadership, engagement and motivation)
crossing several chapters. Table 1.2 depicts the chapters in this text which relate directly
to the BPS’ knowledge domains. Interestingly, Chapter 13 is an example of a topic which
bridges more than one of the knowledge domains. In practice, most issues work psychologists research or practice in cross multiple knowledge domains. While they may be initially framed within one domain (e.g. development of a new selection system) the design
process will invariable mean considering other domains (e.g. the need to revise training
programmes given the quality of people now being selected into the organisation; the
differing motivational needs of these new people; design of work to ensure the new skills
are being used effectively).
Chapter 1 The discipline of work psychology
7
Diversity, gender, fairness, culture
Psychological
Assessment at Work
Learning, Training and
Development
Applying Psychology to
Work and Organisations
Leadership, Engagement
and Motivation
Research Design,
Advanced Data
Gathering and
Analytical Techniques
Well-being and
Work
Work Design,
Organisational Change and
Development
Figure 1.2
The five domains of work psychology
Table 1.2
BPS knowledge domains in work psychology and relevant chapters in this volume
Knowledge domain
Chapter(s) in this volume
Psychological assessment at work
Chapters 2, 3 and 4 are important when considering the relevant literature in
this knowledge domain
Learning, training and development
Chapter 7 is focused directly on training and development. Chapter 11 on
careers is also relevant in training and development
Leadership, engagement and motivation
Chapter 10 focuses directly on leadership and Chapter 6 on motivation.
Chapter 5 on attitudes and Chapter 9 on teams are also relevant for
understanding employee engagement and motivation
Well-being at work
Chapter 8 is focused specifically on well-being at work
Work design, organisational change and
development
Chapter 12 examines the key issues in organisational change
In addition to the content areas, the common and important themes of diversity, gender, fairness and culture need to be addressed in each content area. The BPS describes two
other areas that integrate with all domains. First, students and practitioners in occupational psychology need good knowledge of research design, data gathering and analysis.
Second, it is expected that qualified work psychologists will understand a range of specific
techniques they are likely to use in their practice. This involves the development of generic
skills such as questionnaire design, interviewing, report writing, presentation skills and
data analysis methods.
8
Work Psychology
The skills of a work psychologist
We are aware that some of you reading this book are commencing your studies in work psychology, whereas others are simply studying a work psychology module as part of other
non-psychology academic courses. It may therefore seem that a section on the skills needed
by a work psychologist might be of more relevance to the former than the latter. However,
while information on becoming a Chartered Psychologist or Registered Occupational Psychologist is mostly relevant to those of you with a psychology background, arguably the
skills inherent with work psychology are useful for everyone in education and practice. In
educational contexts, having the appropriate knowledge and skills will help you understand
better the literature you are researching for essays, dissertations or projects. In practice, some
of you may become leaders in your future careers, researchers in other disciplines or work in
related areas (e.g. human resources). Understanding the skillset of a work psychologist will
become a valuable asset for you in the future.
Clearly, given the discussion so far, understanding the five knowledge domains are key to
becoming an effective work psychologist. However, by examining the US-based O*Net website (https://www.onetonline.org/) and searching Industrial-Organizational Psychologists it is
evident that domain knowledge is only one part of the skillset of a work psychologist. Attributes include experience in a range of technology software, active listening, decision-making,
oral and written comprehension, mathematical ability and critical thinking. To make this
more concrete, let us detail how some of these attributes are important for studying, researching and practicing work psychology (Table 1.3).
The rest of this chapter will focus on three highly important skills inherent within a work
psychologist’s role. The following two short sections will consider ethics and critical analysis,
whereas research skills will be examined in the second part of this chapter.
Ethical conduct is crucial to operating as a work psychologist. As an example, in the UK all
practising psychologists are bound by a Code of Ethics and Conduct (British Psychological
Society, 2018). This code requires practising psychologists to be guided by four principles:
1
Respect: ‘Psychologists value the dignity and worth of all persons, with sensitivity to the
dynamics of perceived authority or influence over persons and peoples and with particular regard to people’s rights’.
Table 1.3
Selected work psychologist skills and examples of why they are important
Active listening
If interviewing an employee as a form of data collection, a competent work psychologist
should attend fully to what the employee is saying, ask appropriate questions and reflect
on points being made
Mathematical ability
In order to understand data in company reports, in research papers, in national surveys
etc., a good level of mathematical ability is needed
Oral expression
Being able to convey complex and technical ideas in psychology to non-psychology
audiences is an essential skill for a work psychologist. Mostly, clients will not be
cognisant of the theories, methods of analysis techniques
Statistical software
When collecting data, having knowledge of statistical analysis software packages helps
the work psychologist interpret the data
Problem solving
When presented with an organisational problem by a client, being able to review
information and evaluate options helps the work psychologist to develop actions plans
and solutions
Chapter 1 The discipline of work psychology
9
2
Competence: ‘Psychologists value the continuing development and maintenance of high
standards of competence in their professional work, and the importance of preserving
their ability to function optimally within the recognised limits of their knowledge, skill,
training, education, and experience.’
3
Responsibility: ‘Psychologists value their responsibilities to persons and peoples, to
the general public, and to the profession and science of Psychology, including the
avoidance of harm and the prevention of misuse or abuse of their contribution to
society.’
4
Integrity: ‘Psychologists value honesty, probity, accuracy, clarity and fairness in their
interactions with all persons and peoples and seek to promote integrity in all facets of
their scientific and professional endeavours.’
(From BPS Code of Ethics and Conduct, 2018: 5–7)
In practical terms, this means that psychologists are required to consider, among other
things, the following:
■
Consent: Participants in the research should normally be made aware beforehand
of all aspects of it that might reasonably be expected to influence their willingness to
participate.
■
Deception: Deception of those who participate in the research should be avoided wherever
possible. If deception is necessary for the effective conduct of the research, it should not
be the cause of significant distress when participants are debriefed afterwards.
■
Debriefing: After participation, the participants should be given any information and
other support necessary to complete their understanding of the research, and to avoid
any sense of unease their participation might have engendered.
■
Withdrawal from the investigation: The psychologist should tell participants of their right
to withdraw from the research (usually participants are not required to give their reasons
for withdrawing).
■
Confidentiality: Subject to the requirements of legislation, information obtained about a
participant is confidential unless agreed otherwise in advance.
■
Protection of participants: As much as is reasonably practicable, the investigator must protect participants from physical and mental harm during the investigation.
Key learning point
Work psychologists are required to demonstrate their academic and practical competence, and to
adhere to ethical principles. This is partly to protect the rights and well-being of people who pay for
their services and/or participate in their research.
Psychologists need to be skilled in the analysis, synthesis and understanding of information (textual or numerical). This critical analysis requires the individual to go beyond just
collating and describing information they have accessed. It involves evaluating the quality
of the information received; identifying patterns and relationships in the information; challenging assumptions/ideas offered using other information; posing questions; identifying
gaps in the information; and expressing an informed opinion. Critical analysis is regularly
a part of the assessment criteria in academic modules and evidencing this skill is often the
difference between an excellent and a good mark in an assignment. Frequently we get asked
10
Work Psychology
by our students: ‘how can we critically evaluate?’ Using the PEEL paragraph writing approach
is a useful aid to thinking and ultimately writing critically:
Position – state your position/point on a topic area.
Evidence – support your position with evidence obtained from the research literature.
Evaluate – evaluate how and why your evidence supports your position.
Link – link your points together across paragraphs to develop a coherent argument.
Like many skills, critical analysis is better practiced than described in words. Exercise 1.2
offers a practical approach to understanding and engaging on critical analysis.
Critical analysis
Exercise 1.2
The best way to understand critical analysis is to engage in critical analysis. This exercise will hopefully
allow you to understand what’s involved when critically analysing research data. It is in two parts:
1
Access the More or Less Podcast (https://www.bbc.co.uk/programmes/p04ymqbk) from 31
March 2017 on the prevalence of mental health problems. Listen to the critical approach taken
by the presenter and speakers in terms of the questions they pose, the evidence they offer and
the gaps in understanding they identify and discuss.
2
Using a similar approach, think about the sorts of information you would want to obtain and
the questions you would want to pose if faced with the following headline statistic: ‘One in five
employees are bullied at work.’
Hopefully, as a result of this exercise, you will be a little more critical of headline-grabbing statistical
findings. These headlines can be uncritical and overlook the subtleties and nuances of the research.
Understanding research in work psychology
Alongside ethics and critical analysis, research skills are vital to be a successful work psychologist. The rest of this chapter will focus on skills and introduce what you need to know when
researching and understanding work psychology. We do not profess to provide a comprehensive section on research methods and statistical analysis – there are plenty of specific texts that
offer this facility – but we do focus on the main areas when accessing or conducting research
within the sphere of work psychology. We present this section as a series of questions you
need to ask in order to understand a specific topic area within work psychology and activities/
methods to help you understand those questions.
How do I find resources on work psychology?
The starting point for anyone engaging in research or practice in work psychology, is to access
the current literature on that topic area. This is important for several reasons:
■
to obtain the current perspective on a topic area;
■
to help identify issues, themes and gaps in current understanding;
■
to guide the development of research questions;
■
to identify the methods used to study a topic;
■
to help in the development of practical interventions.
Broadly, such knowledge can be obtained from two sources.
Chapter 1 The discipline of work psychology
11
Academic literature
General texts on work psychology (and other similar terms) give a necessarily brief account of
major developments. Additionally, related textbooks on organisational behaviour and human
resource management will cover some of the same material as generic work psychology books,
albeit from different perspectives. These texts tend to cover a large range of topics within work
psychology, capturing the current literature and synthesising the main theories/research/models. Other specialist books devoted to particular topics (e.g. leadership, motivation and change)
and sometimes even to particular theories (e.g. Social Identity Theory) provide a deeper understanding of that topic, but lack the breadth associated with general texts. Therefore, as a student,
researcher or practitioner in this area you need to decide as to what knowledge you need at a
current point in time. For example, if starting your educational journey in work psychology
you may well commence with a general text to get a feel of work psychology and then move to
more specific texts on a topic in which you have interest or are planning to research/practice in.
Many new theoretical developments, and tests of established theories, can be found in
academic journals. The advantage of academic journals over books is that they tend to provide a more contemporary perspective on a specific topic. Their disadvantage is that they
may only offer a small snapshot of a topic area, often focusing on a very specific and limited
set of research questions. Also, they have a style to their writing which on initial inspection can seem complex and incredibly detailed. Journal articles have been subject to a peer
review process. This means two/three independent reviewers (plus the journal editor) have
read initial draft versions of papers, suggested areas for change and agreed a final version
which meets high quality standards in methodology, analysis and impact on science/practice.
This final, published version should not be viewed as a ‘definitive answer’ to a research topic
area (we contend that no research is perfect). Your newly developed skills in critical analysis
will always find avenues to question in any journal article. Indeed, the authors will present a
series of limitations of their work to illustrate the confines of the research. Other researchers
(which may include the original authors), then come along, critically review the research
and implement a new study to answer similar questions. Over time, the extant literature on
a topic area builds and builds allowing students, researchers and practitioners to develop an
evidence-based perspective on the research area.
Leading journals of work psychology include Journal of Occupational and Organizational
Psychology (published in the UK), Journal of Applied Psychology (USA), Journal of Occupational
Health Psychology (USA), Journal of Organizational Behavior (USA/UK), Human Relations (UK),
Work & Stress (EU), European Journal of Work & Organizational Psychology (EU), Personnel
Psychology (USA) and Journal of Vocational Behavior (USA). There are also other prestigious
journals which include work psychology along with other disciplines applied to work behaviour, including Academy of Management Journal (USA) and Academy of Management Review
(USA). Some other journals concentrate more on the concerns of practitioners; that is, people
who earn their living by supplying work psychology to organisations (e.g. The Industrial–
Organizational Psychologist [USA]).
It is worth noting that more authors are now making their articles available to all through
open access via the journal or through their online profiles and services such as ResearchGate.
Key learning point
Some topic areas have a large research base and it can be daunting to be faced with this volume
of reference material. To help, start by looking for review papers or meta-analyses on a topic area.
These provide a current perspective and, in the case of meta-analyses, statistical analysis of research
questions across a large number of studies (e.g. the extent personality predicts work performance).
For example, the Annual Review of Organizational Psychology and Organizational Behavior publishes
articles that can provide a quick insight into the key issues of a topic.
12
Work Psychology
Non-academic literature
Insights on a specific topic can also be found within non-academic contexts. While these
may not necessarily be subject to peer review along the lines of a journal article; they can
present analysis at a national/global level or discussions focused at a practitioner audience.
Such literature includes Governmental reports, research by professional bodies, practitioner
perspectives, public/private sector reports and online material. When accessing this research,
you need to evaluate the credibility and authenticity of the work before considering its use
in your own research and practice. Often, such literature provides a context for your own
research/practice and can prove useful in setting the scene (see Table 1.4 for some examples).
What do I need to consider when reading work psychology research?
Once you have decided which texts/articles/reports to read, your next set of questions revolves
around the approach, methodology and analysis of the research. Work psychology is eclectic
in the methods used to study phenomena and the approaches to analysing information from
these studies. Having insight and understanding of these features provides work psychologists
with an informed view as to the quality and limitations of the information they access. Let
us contemplate four considerations.
1. The philosophical position taken by the work psychology researcher
The assumptions about the nature of knowledge and theory influence how work psychologists go about their research, the kinds of data they obtain, and how they interpret their findings. Albeit, not necessarily espoused within a research paper, it will be evident from the way
they construct their introduction, the methodology adopted and the approach to analysis,
Table 1.4
Examples of non-academic literature on work psychology topics
Source
Example
Governmental
European Quality of Life Survey (Eurofound, 2016). Cross-country comparisons on
work–life balance
Professional bodies/trade unions
International Test Commission (https://www.intestcom.org/) guidelines on
test-related issues
Practitioner
Montefiori (2016). Games-based assessment: Face validity, fairness perception,
and the impact on employer’s brand image. Assessment & Development Matters,
8(2), 19–22
Public/Private sector reports
National Health Service Change model (2018). A model for change operated
within a national health care organisation
Online
Live Internet statistics (http://www.internetlivestats.com/). Provides live data on
number of people across the world using the Internet
Charities and NGOs
MIND offers research and resources on mental health at work positioned at individuals and organisations (https://www.mind.org.uk)
Chapter 1 The discipline of work psychology
13
the position taken by researchers in studying a topic area. This is important to acknowledge
and understand, because psychology is not a united or unified discipline and includes a multiplicity of views and perspectives. Indeed, scholars hold strongly to their position and defend
it vigorously. Knowing the position of the work psychologist allows you to understand the
assumptions made in the research as well as the limitations of the specific position adopted.
As already expressed, we are not writing a textbook on research methods here; hence we
will restrict our debate to the extreme philosophical positions. The most fundamental polarity has been described nicely by, among others, Easterby-Smith et al. (2018: Chapter 3) as
positivism versus social constructionism. Each of these positions has distinct and very different philosophical roots.
Positivism assumes that the social world exists objectively. This usually (but not always)
implies measuring things using quantitative data (i.e. numerical). Science is seen as advancing by making hypotheses (predictions) about laws and causes of human behaviour and then
testing those hypotheses, preferably by simplifying the problem of interest as much as possible. Progress is often made by trying to falsify these predictions, including ruling out alternative explanations for findings or for data that might provide evidence against the validity of
a theory. In the absence of such findings and data we can be more confident that a theory is
at least approximately correct or at least useful. To draw upon an often-used example: if you
want to prove all swans are white then it’s better to look for non-white swans (this is referred
to as falsification) than to spend your time identifying even more white ones (which would
be an uncritical, confirmatory approach). It is also assumed that the researcher can investigate without influencing what is being investigated: that is, their presence and actions are
assumed not to alter how people would naturally behave, think or feel.
The other extreme is labelled social constructionism. This viewpoint suggests that reality
is not objective. Instead, the meaning of events, concepts and objectives is constructed and
interpreted by people, through the complex and unfolding interplay of thought processes
and social interactions. Instead of measuring how often certain behaviours occur, the aim of
research is to examine the different ways in which people interpret and explain their experience. We cannot really gain direct access to the way participants view things because we must
interpret others’ views. If we are going to collect data about someone’s view of the world, we
will need to interact with them and interpret what they say or do. This means that the data
produced are always an interpretation of the participants’ experience. Therefore, data are
not viewed as some objective reality that exists independent of the view of the researcher or
the views of the research participants. In good examples of this research, the researcher will
provide an account of how they influenced all aspects of the research process: this is referred
to as reflexivity. The data produced by such research tend to be harder to obtain and to summarise than those produced by positivist research, but tend to be richer in meaning, detail
and explanation.
Key learning point
There is an important philosophical distinction in psychology between positivism and phenomenological approaches. The former emphasises objectively verifiable causes of behaviour, thoughts and
emotions; the latter focuses more on people’s subjective explanations and accounts and recognises
the impact of the investigator on the research process and findings.
These perspectives are bipolar opposites in their beliefs, assumptions and stance. In work
psychology several researchers/practitioners operate a more pragmatic stance in which
they see the benefits and need for both approaches to research – depending on the research
question(s) or practical work they are engaging in. While the academic literature in work
14
Work Psychology
psychology is still dominated by the positivistic approach, constructivist research is increasing and, in some areas (e.g. workplace bullying), scholars have called specifically for more of
this type of research to help understand and capture the diversity of participants’ experiences
and perspectives. Many topics in work psychology can be investigated from both perspectives
and all points in between (see exercise 1.3). As far as possible we try to reflect both positivist
and non-positivist research in this text as both have something to offer our understanding
of work-related issues.
Positivism vs constructivism
Exercise 1.3
A clear way to be able to evaluate if a work psychologist is taking a positivism or constructivism stance
is through the research questions posed. You can take a very different stance to examine the same
research topic, but your research questions will reflect that stance. Imagine you were interested in
studying the effects of workplace bullying on employee job satisfaction. Below illustrates research
questions posed from both perspectives.
Positivism – To what extent does experiencing workplace bullying behaviours relate to employee
ratings of satisfaction on the job?
Constructivism – What are employees’ experiences of workplace bullying?
Now you have a go. Think of each of the research areas below from the positivism and constructivism
perspectives and write research questions to reflect these differences:
■
Applicant reactions to internet-based testing.
■
Cross-cultural perceptions of commitment at work.
■
Career aspirations of older adults.
2. The methods used to collect data
Work psychologists use a variety of techniques in their research on human behaviour,
thoughts and emotions in the workplace. Research methods are the specific ways in which
information is gathered within the overall research strategy (the methodology). Understanding the methods available to gather information is useful as it allows the researcher, practitioner and student an insight into how research has been conducted, the problems associated
with the method of collecting information and the potential for future research using different research methods. The philosophical position adopted by the researcher will often dictate
the research methods used; i.e. questionnaires and objective data are synonymous with the
positivistic approach and interviews with the social constructivism approach.
Questionnaires and psychometric tests
Questionnaires are often used to assess a person’s attitudes, values, opinions, beliefs or experiences (see Chapter 5). Psychometric tests are normally employed to measure ability or personality (see also Chapter 3). Questionnaires and tests normally require a person to answer a series
of written questions presented on paper or on a computer screen. Answers are often multiple
choice; that is, the person must select the most appropriate response from a choice of several.
This kind of questionnaire is often referred to as structured, because both the questions asked
and the response options available to the person completing it have been predefined by the
researcher. Unstructured questionnaires, where questions are broader and people respond in
their own words, are much rarer. Responses are usually expressed as a number representing,
for example, a person’s intelligence, extroversion or job satisfaction. Some questionnaires
and tests need to be administered by the researcher in person or in a tightly controlled way
Chapter 1 The discipline of work psychology
15
(such as psychometric testing). Others are designed to be self-explanatory and can be filled
in by the respondent without supervision. Increasing use is being made of online administration of questionnaires and tests. Structured questionnaires are easily the most commonly
used research method in work psychology. They have the advantage of providing large quantities of data with relatively little hassle for researcher or respondents. Also, the data are usually relatively easily subjected to statistical analysis. They are often used in a positivist way.
This means that they may fail to reflect important aspects of respondents’ experiences and be
(mis) used by the researcher as a way of getting quick and easy information rather than truly
engaging with people in the setting being researched (the questions to which the participant
can respond is restricted by the researcher’s choice of questions).
Interviews
A work psychologist may conduct one or more interviews, normally with an individual, but
sometimes with a group of people. Group interviews are often designed to encourage discussion among interviewees about one or more topics and are often referred to as focus groups (see
Gray, 2018, Chapter 18). The work psychologist asks questions and records responses, either
by making notes and/or using a voice recorder. The questions may be specified in advance,
in which case it is a structured interview. On the other hand, the interviewer may define only
the general topic they wish to investigate and permit respondents to talk about whatever they
wish within that topic. This is an unstructured interview. Somewhere between the two extremes
is a semi-structured interview: questions are designed before the interview and used to guide
the discussion, but the interviewer may ask follow-up probe questions or adjust the schedule
in response to what the interviewee says or how they behave. This approach may also be used
to adjust the content of the interview as the research progresses and the researcher wishes to
explore new issues based on the data collected from other participants.
Psychophysiological and psychophysical measures
Psychophysiological and psychophysical measures involve assessing a person’s neurological, biological, physical or physiological state or performance. So, for example, in a study
of work stress, measures could include average heart rate, muscle activity, eye movements
and electrical activity in the brain (by electroencephalogram [EEG]). These methods of data
collection are less common in work psychology than in some other areas of psychology. This
is partly because collecting such data is invasive, and organisations and their employees are
often reluctant to engage in such research. Moreover, although research of this type yields
‘hard data’, organisations are rarely concerned about blood cortisol levels, and much more
concerned with issues such as performance, absence and turnover. The psychological meaning of these data can also be difficult to determine. For example, elevated heart rate could
result from excitement, fear, anger, frustration or any combination of emotions, or it may be
indicative of cardiovascular illness.
Observation
Work psychologists may observe people’s behaviour by stationing themselves as unobtrusively as possible, and recording the frequency, source and timing of behaviour. This can
be termed structured observation. Alternatively, work psychologists may participate in the
events they are studying. This is participant observation. Where people are being observed
in their workplace, they are normally informed or asked about it in advance. Their awareness
may itself affect their behaviour, but that is usually preferable to the alternatives of secrecy
or even deception (especially given ethical concerns). Clearly, a strength of observation is
that it allows the researcher to form impressions of what is said and done in a workplace at
first hand (without having to rely upon potentially biased data from employees). One possible
disadvantage of observation is that if people know they are being observed, they may behave,
think or feel differently from how they otherwise would (Gray, 2018).
16
Work Psychology
Key learning point
Observational data should not be interpreted as ‘hard facts’. This is because employees are likely to
be aware that they are being observed, and the observer will have their own perspective on what they
are observing.
Diaries
People may be asked to keep a diary of key events and/or their behaviour, thoughts and feelings. It is normally necessary to give people a fair amount of structure to help them to focus
their written comments, and to stay in contact with them as an encouragement to keep up the
diary-filling. One important advantage of the diary method is the ability to track the detailed
and fast-moving developments of people’s day-to-day lives. One disadvantage is that, almost
inevitably, some people on some occasions will forget to complete their diary, or simply not
bother.
Experience sampling methodology (ESM) is a mixture of questionnaire and diary methods.
Often ESM studies use electronic hand-held devices that are issued to participants or through
smartphone applications. Typically, software will trigger alarms on these devices several times
during the day: when the alarm sounds the participant answers questions presented on the
electronic device. This approach helps researchers to investigate the subtle sequential relationships between variables (for example, if a person reported feeling happy at work at lunchtime, might it be that their experiences at work in the morning reliably predict them going
on to report being happy?). Because data collection is more ‘instant’, arguably less invasive
and disruptive, and less affected by biases and flaws associated with human memory, the
data collected may be more accurate than data collected once a day (see Daniels et al., 2009).
Archival sources
As Bryman (2001) has pointed out, archival sources are a potentially rich but sometimes
neglected form of data. This is, strictly speaking, a source of data rather than a method of
collecting it. Archival information is anything that exists in organised form before, during or
after the work psychologist’s investigation. Examples include employee absence data, company accounts, productivity records, human resource policy documents, accident statistics,
minutes of meetings and many others. Data from archival sources are most often used either
to provide a context for a particular research project, or to investigate the impact of an event
on the functioning of an organisation.
Similar to the philosophical position debate, pragmatic work psychologists are likely to use
a variety of research methods to help them understand a particular issue – whether this is in
research or practice. Figure 1.3 illustrates a hypothetical research methods approach using different measures which may be adopted by a work psychologist to examine workplace stress.
The researcher is interested in the role demands play in workplace stress. Each of the methods
will provide different information to the researcher and each will be analysed using different
techniques (see later). However, overall the research can develop a more holistic picture of
the role job demands play in workplace stress and allow us to triangulate the findings from
one method of data collection with another.
3. The design of the research
Research designs concern the overall research strategy employed. This strategy depends on
the researcher’s beliefs about scientific investigation as well as the nature of the phenomena
being researched.
Chapter 1 The discipline of work psychology
17
Questionnaire
Employee ratings
of types and
intensity of job
demands at work
Diary
Daily diary over 2
weeks, where
employees record
extent of job
demands that day
Workplace
stress
Interview
Semi-structured
interview with
employees on their
experiences of
work demands
Physiological
Average heart rate
during demanding
and less demanding
work periods
Figure 1.3
Hypothetical research methods used by a work psychologist
The survey design
The key distinguishing feature of a survey is that its use does not usually disrupt naturally
occurring events, nor does its use control them. It simply takes a snapshot of what is happening, usually by asking people about it. The aim is usually to gather (mostly) quantitative
information about certain phenomena (for example, events, attitudes) from many people.
On occasions this will be done simply to ascertain the frequency of occurrence of a certain
event, such as feeling anxious at work. More commonly, a survey will be used to discover the
relationships of variables with each other – for example, whether anxiety at work tends to be
accompanied by low job satisfaction (this does not mean that anxiety causes satisfaction, but
just that the two tend to occur together).
Most commonly surveys involve questionnaires, although surveys can be conducted via
interviews (e.g. market research). Questionnaires cannot be just thrown together (as they are
in some popular magazines) if they are to do a proper job. What is included in the questionnaire and the way questions are worded should be driven by good quality research and strong
theories. Consistent with the positivist research philosophy, questionnaires must be carefully
devised so that they unambiguously measure what they are supposed to measure, i.e. so that
they are valid (see Chapter 3). It is important to collect data from an appropriate sample of
the relevant study population. Ideally, the respondents should be a representative sample of
all those people to whom the survey is relevant.
18
Work Psychology
The survey has both advantages and disadvantages (see Bryman, 2001). It can be used with
people directly involved in the issues to be investigated. It can investigate their experiences
in their day-to-day setting. It is normally easy to distribute and collect and makes relatively
low demands on participants’ time (something that is often an important consideration for
work psychologists). On the other hand, the survey does not involve any manipulation of
the variables being investigated. This makes it very difficult to establish cause and effect. For
example, if you have surveyed employees on their perceptions of their leader and their own
levels of task performance you cannot say that leader behaviour causes task performance.
They may well be related, but as you’ve not considered other variables nor actually tested a
prediction, cause and effect cannot be established. This is termed a cross-sectional survey.
Longitudinal surveys can provide some insight into cause and effect relationships. Here,
data are gathered on more than one occasion. Conditions pertaining at time 1 (A) may cause
those at time 2 (B) but not, presumably, vice versa. However, even if event A happens before
event B, that does not necessarily mean that A causes B, just that it happens first. To be more
confident of cause and effects, we would have to see if carefully planned adjustments to A
predicted B and rule out the effects of other variables on both and A and B. This, as you might
imagine, is difficult to do in complex real-life work contexts.
Surveys often comprise self-reported data on all constructs they assess. While this provides
insights, and is often the right way to assess a variable (e.g. if assessing perceptions then the
right approach is to ask an individual to self-report their perceptions), it has the potential to
lead to biases in responses which impact on the findings (e.g. they often lead to inflated relationships between variables). The work psychologist can reduce this impact to some extent by
using other sources of data (e.g. manager ratings, co-worker ratings or objective archival data)
for some of the variables being assessed. Taking the leader behaviour and task performance
example again, a researcher could ask an employee to rate his/her leader behaviour and obtain
co-worker ratings of the employee’s task performance.
Having said all of that, well-designed survey-based studies often provide a good starting
point for further research. They provide quick tests of relationships between variables that
can then be examined more rigorously with more complex research designs.
Key learning point
Survey research is relatively easy to conduct in work organisations. However, it is often difficult to
determine cause and effects from the data collected.
The experimental design
One key advantage of an experiment is that it allows the psychologist control over what happens in ways that can be used to rule out alternative explanations for the research findings. This
permits stronger inference about causes and effects. There are several key aspects of an experiment which need to be included for the research to be considered an experimental design:
■
Independent and dependent variables – the independent variable (IV) is the variable the
researcher manipulates in order to assess its impact on an outcome (or dependent variable, DV). Simply, this could comprise one IV (e.g. distraction) and one DV (error rates on
a clerical checking task) or could involve several IVs (e.g. visual or auditory distraction).
■
Experimental and control groups – the experimental group are the individuals who
receive the manipulation and the control group are those who do not receive the manipulation. People in the control group should as far as possible perform the same action as
those in the experimental group, apart from the manipulation. We then compare the outcome across both groups to assess if there are differences. For example, the experimental
Chapter 1 The discipline of work psychology
19
group will be subject to a distraction when completing the error checking task, whereas
the control group will not. This is often termed conditions. The IV has two conditions:
distraction vs no distraction.
■
Random allocation – participants should be randomly allocated to either the experimental or control condition. This is to reduce the potential for bias in allocation and avoid
groups being unequal in a specific psychological construct. For example, if we did not
randomly allocate people to the distraction vs no distraction condition, we might find
that those in the distraction condition tended to be more skilled in ignoring distractions
and hence the difference between the two groups could be minimal.
The psychologist’s control necessarily makes it an artificial situation because the real world
is rarely so neat and tidy. Unless the psychologist indulges in a huge (and unethical) deception, the experimental participants will know that they are not in a real job, and that the
experiment will last only a short time period. This could crucially affect their reactions to
the task. In running experiments something is gained – control – but something is lost – realism. Arguments rage over the use of experimental approaches to investigate organisational
phenomena. Some argue that it is only by finding out things in controlled settings first that
one can begin to study them in more applied settings. Others argue that the experimental
situation produces results that would not be discovered or hold true in the ‘real world’ because
they do not exist there.
Key learning point
Laboratory experiments allow the work psychologist to control and manipulate the situation in order
to establish whether there are causal relationships between variables, but it is often not clear whether
the same relationships would occur outside of laboratory situations.
Sometimes it is possible to conduct a field experiment instead. Quasi-experiments of
varying degrees of sophistication are often used in work psychology. These experiments take
place in a real work setting, probably with the people who worked there. Such experiments
rarely have ‘perfect’ designs. Even if managers and union officials at a factory were prepared to
allow the psychologist to create an experimental and control group on the factory floor, they
would probably not allow random assignment of participants to groups (a key feature of a
‘true’ experiment). Also, they probably could not arrange things such as identical supervision
and identical opportunities to interact with co-workers, even if they wanted to. However, you
may see true experiments designed using scenarios or vignettes. Here the researcher develops
a set of, more often written, vignettes on a work-related topic and within the vignettes manipulates the information so that the experimental group receives vignettes with the manipulation in and the control group does not. Participants are then asked to rate their perceptions
of the scenario as if they were involved and differences between groups on ratings can be
assumed to be due to the manipulation in the scenario. This removes the problem of trying to
set up an experiment within the field (as everyone reads the scenarios) but it relies on ratings
of intent and not actual behaviour: there can be large differences between these two things
(see Chapter 5).
Occasionally, it is possible for psychologists to conduct a field experiment using events
that are occurring anyway. This is sometimes called a natural experiment. For example, some
groups may engage in interventions to improve their working conditions while other groups
who will get the intervention some time later act as a ‘waiting list’ control (see Chapter 8).
From an ethical point of view, it is important to ensure that eventually all participants have
access to an effective intervention.
20
Work Psychology
Key learning point
It is occasionally possible to conduct experiments in real-world settings, though usually the work
psychologist has far less control over the situation than in laboratory experiments.
Qualitative design
Both surveys and experiments normally use numbers to express data (i.e. quantitatively).
They allow the people participating in the research little chance to express their opinions
in their own words, since the work psychologist investigates a limited number of variables
of their own choice, selected in advance. Hence surveys and experiments do not obtain a
detailed picture of any individual’s world. They both involve the psychologist in a detached,
quasi-scientific role, and tend to reflect the positivist research philosophy described earlier.
Qualitative research often (though not always) involves a much greater emphasis on seeing the world from the point of view of the people who participate in it (Cassell and Symon,
2004). That is, it tends to more often reflect the phenomenological research philosophy
described earlier. This normally means collecting detailed information using observation
and/or interviews from a small number of individuals or organisations – perhaps only one.
This information is intended to paint a picture rather than measure a limited number of
specific phenomena. Drawing on Gubrium and Holstein (1997), Silverman (2001: 38–9) has
identified four kinds of qualitative research:
1
Naturalism: The emphasis is on observing what goes on in real-life settings. This tends
to produce rich descriptions of behaviour and events, but little insight into how those
things are understood by the people involved.
2
Ethnomethodology: Focuses on a close analysis of interactions between people and how
these maintain and reflect social order. This can show how social groups and cultures
work but runs the risk of neglecting the role of broader contextual factors such as economic conditions.
3
Emotionalism: Here the primary interest is in establishing a close rapport with the people
being researched and finding out about their experiences and feelings. This differs from
the previous two categories in giving priority to people’s personal opinions, rather than
the researcher’s frame of reference, but runs the risk of overemphasising emotion.
4
Postmodernism: Rejects the notion that there is an objective truth, focusing instead on
how people portray themselves and their contexts in order to achieve personal goals and/
or affirm their sense of identity.
Key learning point
Qualitative research often involves an attempt to describe and analyse how individuals make sense of
the situations they are in: it often involves a broader and more flexible approach to data collection. The
focus may be on behaviour, social interaction, personal experience or self-presentation.
Qualitative research usually produces a large amount of data, which requires some editing
and interpretation by the researcher. It can be time-consuming and complex. Even obtaining
the necessary access to people in their workplace can prove impossible. It leaves researchers
vulnerable to the accusation that they have simply discovered in the data what they expected
to find.
Chapter 1 The discipline of work psychology
21
There are many qualitative research methods. Symon and Cassell’s (2012) text provides an
excellent description of many such techniques. Other techniques include:
■
Case study research, which involves the collection of data from several teams, organisations or individuals in order to better understand how the research context influences the
research findings. Case studies can be aggregated in order to identify some of the general
findings that apply across different research settings.
■
Ethnography where the researcher participates in the research setting in order to collect
data that reflect the meaning of events, behaviours or activities in the research setting
(rather than the meaning that might be attached to things by some ‘detached’ observer).
For example, one of the authors of this text spent some time, underground, with various
sewage workers in order to better understand their working conditions (and who said the
work psychologist’s life wasn’t very glamorous?).
■
Repertory grid technique, which is sometimes used to identify the knowledge, skills
and competencies of effective (and ineffective) employees in job analysis.
■
Attributional coding, which can be used, for example, to analyse how people describe the
causes of the events that occur in their working lives.
Action research design
Lewin (1946) coined the term action research to describe research where the researcher and
the people being researched participate jointly in it. Action research is intended both to solve
immediate problems for the people collaborating with the researcher, and to add to general
knowledge about the topic being researched. It involves not only diagnosing and investigating a problem, but also making changes in a work organisation based on research findings and
evaluating the impact of those changes. Gray (2018) illustrates action research as a cyclical
process involving planning (detailing the focus on the research), acting (developing actions
and gathering the information), observing (noting the impact of the actions) and reflecting
(evaluating the outcomes).
Action research can involve any of the research methods described earlier but, like qualitative research, is most likely to use interviews and participant observation (see Meyer, 2001).
Much more than other research designs, it focuses on a specific problem in an organisation
and its solution. It abandons the detachment of survey and experimental designs. Like qualitative research, it seeks to examine how people participating in the research see things, and
it usually covers fairly long periods of time. Unlike much qualitative research, it involves
attempting to solve a problem, and monitoring the success of that attempt.
From the point of view of the researcher, action research can be exciting, difficult and
unpredictable. It requires close involvement with an organisation because of its problem
orientation. In action research, the process of conducting the research can become as much
a focus of interest as the problem it was originally designed to address.
Key learning point
In action research, the psychologist and the people involved in the situation being researched work
together to define the aims of the research and solve practical problems. Chapter 8 includes some
specific examples.
Mixed methods approach
An increasing amount of research in work psychology is beginning to draw upon several methods used in combination, or sequence, to strengthen research designs. This mixed methods
approach is used to make the best of both worlds, drawing on the advantages of several different
22
Work Psychology
techniques to offset the disadvantages of each. It is common that this ‘mixing’ refers to the
use of both qualitative and quantitative methods in the same study. Gray (2018: 198, adapting
Greene et al., 1989) describes several ways this can be done and the benefits of each:
1
Triangulation: Convergence, corroboration, correspondence or results from different
methods.
2
Complementarity: ‘Seeks elaboration, enhancement, illustration, clarification of the
results from one method with the results from the other method.’
3
Development: ‘Seeks to use the results of one method to help or inform the other method.’
4
Initiation: ‘Seeks the discovery of paradox and contradiction, new perspectives, the recasting of questions or the results from one method with questions or results from the other
method.’
5
Expansion: ‘Seeks to extend the breadth and range of inquiry by using different methods
from different inquiry components.’
Key learning point
Research design refers to the overall strategy in conducting research, whereas research methods are
the procedures by which information is collected.
4. The way the data were analysed?
Data are outcomes of the research design and methods used by a work psychologist. Data can
be numbers-based (e.g. ratings on a scale, score on a test, average heart rate or reaction-time
on a computer task) or text-based (e.g. transcript of interview, synopsis of company records or
postings on an online message board). The philosophical position, research design and methods tend to dictate the type of data measured by the work psychologist. For example, the work
psychologist taking a positivistic position using a survey design with a questionnaire method
is likely to generate numbers-based data. Yet at this stage, all we are faced with is a series of
information which can be said to be in its raw form. For it to be useful, work psychologists
need to interpret the information and analyse it in a manner to make it more meaningful and
interpretable. This analysis is then used by work psychologists to help provide answers to the
research questions posed. It is beneficial for practitioners, because it allows them to gauge the
strength of a particular study and how useful it might be in their own practice.
Once again, we do not intend for this section to be a detailed discussion of approaches to
analysis. Rather, we detail the main types of analyses methods and how you may come across
them in a work psychology related context. Firstly, we will detail when data is numbers-based
(or statistical) and then when text-based (or qualitative).
Statistical analysis
When numbers are generated as data work psychologists need some way of collating the data
and interpreting it considering the research questions posed. They do this in two, interconnected ways – descriptive statistics and inferential statistics. Descriptive statistics, as their
name suggests, describe the patterns in the data, such as the mean, median, mode and standard deviation. Inferential statistics allow us to test the strength of our research findings and
infer whether we can have enough confidence in them (termed statistical significance).
They are interconnected because through running a specific inferential statistical approach
we can observe if the data shows notable differences or relationships; yet we always refer to our
descriptive statistics to see where our differences or relationships are. For example, look back
to our research study on distraction and clerical checking task error rates. We have two groups:
an experimental group of 50 employees who are distracted while performing a task; and a
Chapter 1 The discipline of work psychology
23
control group of 50 employees who are not. We obtain error rate scores for all 100 employees
and run an inferential statistic to see if there is a significant difference between the two groups
(later you will see which one is appropriate). Inferential statistics suggest there is a significant
difference and we then refer to the descriptive statistics (means error rates for both groups) to
view which group showed the higher error rates on average: from our research question we
would expect this to be the experimental group. Thus, descriptive and inferential statistics
work together during data interpretation.
Our research questions may mean that we are interested in examining differences on a
measure between two or more groups (as in the above example) or that we are more interested
in the relationship between two or more measures (e.g. job satisfaction ratings and absenteeism rates). As a result, the choice of descriptive and inferential statistics depends on the type of
data collected and research questions posed. Table 1.5 illustrates some of the main statistical
approaches used in work psychology. It is by no means an exhaustive list and does not delve
deep into the approach to analysis each takes. Rather it is meant to be an introduction to the
analyses you may come across when studying work psychology.
Table 1.5
Main statistical analyses seen within work psychology
Test
Features
Work-related example
Independent
t-test (t)
Assesses the significance of a difference between two
different group mean scores. Cannot be used when there
are more than two groups
Comparing mean scores in clerical checking error rates
between a group experiencing a distraction and one
having no distraction
Dependent
t-test (t)
Assesses whether the means of two sets of scores from
the same people differ significantly. Cannot be used when
there are more than two groups
Comparing the difference in means scores on knowledge
of health and safety before and after safety training
Analysis of
variance (F)
Extends the principles of the t-test to more than two
groups and can be used to identify interaction effects
Differences in error rates between groups given either a
visual, auditory or no distraction (independent)
ANOVAs can assess the difference between independent
or dependent groups as well as a combination of
independent and dependent variables (mixed)
Differences in ratings of seriousness of work-related,
personal and physical negative acts at work by the same
employees (dependent)
Differences in average heart rate between young
and older employees, pre and during a games-based
assessment (mixed)
Chi square (x2)
Used to test differences between groups in the frequency
with which group members fall into defined categories
Differences in the frequency of males and females likely to
leave the organisation within the next year
Correlation (r)
Tests the extent to which scores on two variables tend to
go together (i.e. covary)
Relationship between workload and employee burnout
Multiple
Regression (b)
Extension of correlation for more than two variables. Able
to examine the relative strength of each individual variable
in relating to an outcome variable. Can also test interaction
effects between variables
Relationships between workload, role ambiguity, conflict
and burnout. Tests the overall relationship and which of
the three predictor variables shows the strongest effect
Factor analysis
Exploratory factor analysis (EFA) examines the
relationships between variables and clusters them into
related factors
EFA of a 100-question measure to establish the factor
structure for personality
Confirmatory factor analysis (CFA) assesses the level of fit
of an already identified structure to the data provided
SEM
Examines the level of fit of the data to a model which
predicts relationships between variables
CFA of an already established five-factor model of
personality on a sample of people in a country where the
EFA above was not carried out
The fit of a model which hypothesises organisation
stress predicts negative emotion which in turn predicts
counterproductive work behaviour
24
Work Psychology
Meta-analysis
There is debate within psychology about whether we should steer away from using traditional
statistical approaches of establishing the significance of differences and correlations and
move towards analyses focusing on the actual size of the effect (effect size). These effect-size
measures (e.g. correlation, d-value and odds ratio) help establish how strong the relationship,
difference or likelihood is. Advocates argue they provide a better representation of the actual
findings from research than focusing on significance levels alone. For example, we might find
that the difference in error rates between our distracted and non-distracted groups, as tested
by an independent t-test, is significant. We’d probably run down the corridor of our academic
department whooping with delight if this was the case! However, referring to our descriptive
data (means error rates) for each group might show that the actual difference in units is quite
small (e.g. the experimental group on average makes 12.2 error rates and the control group
10.8 – a difference of only 1.4 errors). An effect size measure (in this case it would be a d-value)
would allow the researcher to fully understand the size of this difference (e.g. small, medium
and large) and therefore likely change the way they view their data – resulting in more of a
reflective smile whilst walking down the corridor! In journal articles you may see researchers
reporting statistical significance data and effect sizes, as this is the current default approach
for many journals.
One analysis approach which uses effect sizes as its data is meta-analysis. This type of
article is quite common, especially in topic areas where a lot of research has been conducted.
The aim of meta-analysis is to provide an overview and summary of what general conclusions
can be drawn from a body of research. A researcher conducting a meta-analysis extracts effect
size estimates (e.g. correlations or d-values) from relevant research studies and weights them
according to the sample size. They also control for several other artefacts within the data, such
as reliability of the measures used and restriction in the range of the data. The technique is
powerful because it focuses on looking at the average effect size and the spread of effect sizes.
This can help us to avoid the pitfalls associated with basing our practice on findings from
single, possibly unrepresentative studies, that each have their own limitations and flaws.
Meta-analysis has been criticised by some as being something of a blunt instrument that
cannot capture important details of how particular research studies were carried out. There is
a risk we may lose sight of some of the important contextual nuances that impact on the variables we are interested in. Exclusion criteria remove some studies from the research analysed,
even though these may contain interesting and important findings. Others have defended
meta-analysis. Rosenthal and DiMatteo (2000) have reviewed these arguments and conclude
that many of the criticisms are invalid if meta-analysis is properly conducted. An example of
meta-analysis that gives some insight into its usefulness is that reported by Kivimäki and colleagues (2015) into the links between long working hours and a serious illness, the experience
of a stroke (see Chapter 8). Their meta-analysis gathered up data from 25 studies from across
the world containing an eye-watering total of 528,908 participants. It would be difficult if
not impossible to conduct a single study of this size into such an important issue with such
a diverse participant group.
Analysing qualitative data
In this section we look at approaches to analysing text-based data. Just as there are many
techniques for analysing numerical data, so the same is also true for qualitative data. Just as
statisticians sometimes disagree about which statistical techniques are appropriate in which
circumstances, so qualitative researchers sometimes disagree about the best ways to analyse
qualitative data. Statistical arguments are usually about technical issues, whereas disagreements over how to analyse qualitative data are more often to do with the researcher’s philosophical and theoretical position and their impact on data analysis and reporting.
Chapter 1 The discipline of work psychology
Table 1.6
25
An excerpt from a research interview
A lawyer discusses assessment of work performance
Note: I = Interviewer, R = Respondent
I:
How is your performance evaluated here and what do you think of that?
R:
They have an annual PDR, which is Personal Development Review. You are given a form to complete a short time
before and then you meet with whoever’s undertaking it generally – the partner that you report to – and you sit in a
room and discuss it.
I:
And is that a fruitful discussion or is it going through the motions?
R:
For me it was fruitful because I think the . . . the difficulty I have with these is that more senior people can see the
benefit and the relevance of them. When you get down to the secretaries, they view it very, very much as a . . .
almost as a disciplinary matter. It’s their annual kicking from the boss because they haven’t been up to standard.
I:
But it doesn’t sound like you feel that way about your PDR?
R:
No. I think it’s the only true way of finding out your own standing and seriously expressing ambitions and what you
have to achieve to reach those.
I:
And is it sufficiently candid and open to achieve that?
R:
It was for me, yes.
I:
There’s a couple of indications that it might not be for everybody. Is that true? Or am I reading too much into that?
R:
No, you’re right. There are some people who I’ve been told on the quiet have reached as far as they will go but having
spoken to those individuals – not directly about their PDRs, but you know, just generally – I don’t think they’ve
actually been told that clearly.
I:
Okay.
R:
I think that’s more a function of whoever’s actually doing the PDR because I had two very forthright speakers who
will tell you exactly whereas others are a little bit more political.
I:
Okay. There would have been a time when having your performance assessed and working in a managed system was
a bit of an insult for a professional person, but it sounds like that comes with the territory now.
R:
It does. I’ve only had them since I’ve been here. It was a new concept to me. And yes, I was very nervous before it. But
unless you do get some form of independent view, I don’t think you’re the best judge of your own performance.
Table 1.6 shows an extract from a research interview conducted by one of the authors
of this text. The interview was one of 80 conducted by one of the authors and colleagues
as part of an investigation of how lawyers and architects see their work, and how they, as
professionals, view management and being managed. In this extract, the interviewer (I) and
respondent (R) (a lawyer working for a large firm of solicitors) discuss how the respondent’s
work is evaluated.
How might these data be analysed? What general statements can be made, and conclusions drawn based on data such as these? This depends heavily both on the research questions
being asked (together with the theoretical and/or practical basis for considering those questions important), and on the philosophical assumptions of the researcher. Excellent coverage
of the range of methods available for analysing qualitative data can be found in the texts by
Cassell and Symon (2004) and Symon and Cassell (2012).
One way of analysing this transcript would in effect involve turning qualitative data into
quantitative data. The researcher could attempt to assess how positively the interviewee felt
about the firm’s methods for assessing the work performance of employees, perhaps on a
numerical scale. While such a scoring procedure might be useful, it would of course be a huge
26
Work Psychology
waste of information if that was all that was done with these data. It is unethical to inconvenience participants by collecting data that will not be used. By going down an interview route,
the researcher is interested in understanding the individual’s experiences of a specific topic
and therefore requires a deeper level of exploration.
There are several different approaches to analysing textual data, including thematic analysis,
grounded theory, discourse analysis and interpretative phenomenological analysis. Each
differs in the extent analysis operates from a deductive (that is, deducing what types of response
are important on the basis of prior theory and research) or inductive (that is, starting with the
data and trying to develop theory on the basis of it) analysis approach. Arguably, the foundation
for most different qualitative analysis approaches is thematic analysis. Thematic analysis is often
used to identify informative, relatively consistent and accessible patterns in the data in which the
researcher sorts meaningful chunks of the text into themes reflecting the recurrent ideas which
can be detected in the material. This is an iterative process where the researcher engages with
the text, identifies initial patterns, collates thoughts into themes and reflects on the strength of
those themes. Thematic analysis can be used deductively or inductively.
Braun and Clarke (2006) propose a six-stage process for thematic analysis.
1
Familiarisation with the data. A researcher transcribes the interview data, reads it through
and re-reads it several times noting down initial ideas. As a result, including the actual
interview itself, the researcher has interacted with the text four or more times and hence
becomes immersed in the text. Familiarisation is central to understanding the perspective
of the interviewee.
2
Generate initial codes. Short labels (codes) are provided for sections of text within the
interview transcript that are perceived as important by the researcher (these codes are
usually modified several times throughout an analysis). Collate the codes with the accompanying text (quotes) that support them.
3
Search for themes. Examine the codes and quotes and consider if there are patterns of
meaning across the codes. These patterns are indicators of possible themes within your
data.
4
Review themes. Check if the themes generated tell the story of the experiences from your
interviewee. Themes do not necessarily have to be reflected in the transcripts of all your
interviewees; but should show an element of importance within the interviews (even if
this was just for one person).
5
Define themes. Provide a label for the theme which reflects the meaning inherent within
the quotes and codes under that theme. The researcher needs to compose a detailed
analysis of the theme and present the story each theme tells and how it fits in the wider
story of the data. From a deductive perspective your theme name will reflect or resonate
with a component from established theory. From an inductive perspective you generate
a theme from the data itself.
6
Report the findings. This will involve the description and discussion of each theme, providing enough relevant quotes to support the meaning of the theme and discussing them
in light of academic research.
Done well, qualitative analysis is not easy. It is easy to do poorly. Anyone choosing to
undertake qualitative research under the premise that it is simpler than statistics is either a
qualitative analysis genius or, more likely, is not fully cognisant of the nature of this form
of analysis. Even, what may be viewed as, a basic thematic analysis takes time and skill to
collect the data, transcribe the interview and code the interview. Then you repeat for the
other interviews you have conducted. Plus, you have the added complexity of continually
reflecting on your analysis thinking about the assumptions you held going into the coding,
how your own values, experiences of work, education and other parts of your life shaped
how you coded the data and whether the themes just reflected the questions asked. It is vital
the researcher reflects on how she/he may have influenced all stages of the qualitative data
collection and analysis.
Chapter 1 The discipline of work psychology
Exercise 1.4
27
Thematic analysis
Using the short transcript provided in Table 1.6, run a thematic analysis using Braun and Clarke’s
process. Just use the first five steps and see what theme(s) you develop. Reflect on any assumptions
you held going into the coding. How might your own values, experiences of work, education and
other parts of your life shape how you coded the data?
The changing world of work
It is probably true to say that nobody has ever lived at a time when they felt that not very much
was happening or changing in their world. Economic, political, demographic, technological,
global and consumer factors all impact on the nature of how an organisation runs: its mission
and goals; how it decides to organise work, how it is structured, how it manages individuals and
how it judges success/effectiveness. Contextual factors can enhance or inhibit organisational
functioning (Johns, 2006). Of course, factors need not always be external and can arise just
as easily from internal pressures (e.g. restructuring, diversification, response to engagement
surveys) but these are often triggered by change occurring in the wider environment. There is
no doubt that work psychologists engaging with one specific topic area need to ensure they
have one eye on external and internal contextual factors. These help to explain the historical
context of the problem or issues being tackled, why an organisation is where it is and why it is
going in a specific direction. A work psychologist trying to implement an intervention without
considering these contextual factors is likely to be faced with resistance, a lack of readiness to
accept the intervention and limited evidence of the effectiveness of an intervention.
In Table 1.7 we note some of the world of work changes under the headings of diversity,
technology and culture, and suggest some consequences for what are, or should be, hot topics
Table 1.7
World of work changes and their implications for work psychology
World of work changes
Implications for work psychology
Diversity
Ageing working population
Learning, performance and engagement with work of older people
Increasing labour market participation and
equality for historically disadvantaged groups
Further development of fair selection procedures; the work
experiences of members of disadvantaged groups; impact of
diversity on workplaces and organisational performance
Technology
More people working remotely (e.g. at home) using
information and communication technologies (ICT)
Computer and Internet-delivered forms of
assessment.
Supervision and leadership of people not physically present; impact
of isolation on work performance and satisfaction; effective virtual
communication and teamwork
Remote selection and assessment using the Internet; games-based
assessment; remote appraisal and online training
Culture
Increasing internationalisation of organisations
and markets
Cross-cultural comparisons of workplaces; working abroad;
interpersonal and intercultural influence
Selection and assessment in multi-national
organisations
Appropriateness of selection procedures across cultures;
translation and adaptation of assessments
28
Work Psychology
within work psychology. Throughout the text, you will notice chapter authors have included
these three components in various guises within their topic areas. This should allow you
to reflect on some of the implications of each trend within each of the specific topics. We
just highlight these factors here to advance the idea of the need to be aware of the potential
impact each of these factors has on work psychology research and practice. Chapter authors
offer more specific contexts and discussion in relation to their own topic areas. It is too simplistic to imagine each of these factors operates independently of the others. For example,
introducing a new selection system into an organisation involving initial screening via online
games-based assessments not only throws up technological considerations (e.g. robustness
of the Internet delivery, screen design and layout, security of data) but also diversity considerations (assessing for group differences in scores on the basis of demographics, access
for applicants with a disability) and cultural considerations (equivalence of the test across
different cultures, digital divide in access to the internet across countries).
Key learning point
The world of work is changing rapidly because of increased diversity, technological advances and
cross-cultural trends. Work psychology can be used to examine the human consequences of these
changes.
Summary
Work psychology is an applied approach to psychology and concerns both the interaction
between an individual and their work, and the relationships between people in the work
setting. This includes issues such as: psychological assessment; learning, training and development; leadership, engagement and motivation; employee well-being; work design; and
organisational change and development. It is diverse, with differing positions on how data
should be collected, analysed and reported, although many work psychologists adopt a pragmatic stance to their work. Work psychologists need to develop a varied skillset in order to
function effectively within their profession, as well as ensuring they engage in professional
development to maintain and update their skills. Three areas in which work psychologists are
particularly adept are ethical skills, critical analysis and research skills. It is also important to
for work psychologists to examine the context in which they are operating and not neglect
diversity, technological and cultural factors in their research or practice. As can be seen from
the topics addressed in this text, work psychology has a lot to offer organisations.
Closing case study
Assessing problems in recruitment and retention of nurses in
a hospital unit
Recall the case presented at the start of the chapter:
You have been called in by the Clinical Lead of a unit operating within a hospital to help examine
and assess why they struggle to recruit and retain nurses. The unit operates across three sites –
each which deal with the same condition, although offer different services. There is a high turnover
Chapter 1 The discipline of work psychology
rate of nursing staff and while the unit regularly advertises for replacement posts these tend to
go unfilled. When filled, most new nursing staff leave within the first year. Additionally, trainee
nurses tend not to select this unit as part of their practical experience, further restricting the nurse
provision in the unit.
Now let us work through how you might approach this case using the knowledge you have gained
within this chapter.
Skills
Active listening – you will need to listen appropriately to the Clinical Lead, lead nurses and other
stakeholders at the start of the process to fully understand their needs. Further during data
collection, active listening is vital in the data collection process.
Oral expression – you will need to convey psychological research/theory to the stakeholders in
a manner which maps their specific needs and requirements.
Problem solving – you will need to understand the problem fully and develop a set of actions on
how you are going to address the problem.
Evidence-based management – you will want to look at various sources of data (academic
literature, relevant stakeholders, organisational context, and practitioners) and plan actions
to obtain data from these sources.
Critical analysis – reflect critically on the information provided by the sources identified above.
Ethical practice – ensure throughout the whole project you conform to good ethical practice.
This is in your dealings with the Clinical Lead and in the data collection process. Think about
security of data, informed consent of research participants, ensuring confidentiality of information (especially that provided by nurses), plan for a debrief and how findings will be disseminated to all parties.
Research plan
You have chosen to adopt a qualitative approach for this project as you are interested in understanding the experiences of current nurses as a way of identifying why there is a problem in recruitment
and retention. Your plan is to run a series of interviews with nurses from the different sites, at
different levels, and analyse the findings via a thematic analysis methodology. At present you are
unsure of the approach to take and how to frame the interview. You therefore decide to complete
a review of research literature initially searching academic and nursing specific references for literature on problems in retention of nurses. Several specific journal articles as well as opinion pieces
are highlighted, and it is evident from these that work stress is the dominant explanatory mechanism
for understanding problems in recruiting and retaining nurses. So, what began as a recruitment
problem, is now moving more towards a well-being angle. You then focus the review within work
stress generally and come across the Job Demands-Resources Model (Bakker and Demerouti, 2007;
see Chapter 8) as a framework for positioning your data collection plan. As a result, you are focused
more within the constructivism position, albeit your interview and themes will be informed by the
JD-R model (this is taking a deductive approach). You can see here how a positivism and constructivism dichotomy is not always easy to visualise in practice. You report back to the Clinical Lead as well
as senior clinical staff your proposed plan of action, gain their approval and set a timeline for data
collection. In liaison with unit managers at the three sites you arrange for dates for data collection
to take place with nurses who have volunteered to be a part of the study. You compose an ethical
29
30
Work Psychology
protocol, design your interview schedule based on the JD-R model and send to unit managers for
their approval. You decide to trial the interview on a unit manager and as a result, make some minor
changes. Data collection then starts. You arrange to meet participants in a quiet room away from
the unit, explain the research is confidential and how the interview will run (plus how data will be
used going forward), detail that the interview is recorded but once transcribed the recording will
be deleted. You run your 10 interviews in each unit and start the process of transcribing the interviews. You then adopt Braun and Clarke’s stage approach to thematic analysis and develop an initial
thematic map illustrating the job demand and job resource themes within the role. You provide an
initial short report to the Clinical Lead and present your findings to the management team. You then
develop a more detailed report with specific recommendations that is sent to the Clinical Lead and
advise that the initial summary report is shared with all staff in the units. You offer to support the
recommendation made and suggest that a follow-up quantitative study testing the links between
JD-R and turnover intention is undertaken.
The role of an experienced work psychologist!
Discussion questions
1 Using an evidence-based management framework, discuss how you would approach
researching workplace bullying within an organisation. Think about the questions you
would want to ask the different sources of information.
2 Is work psychology any different to management consultancy? Think about this with
reference to the skillset of a work psychologist.
3 In pairs or groups take opposing philosophical stances (e.g. positivism or constructivism) and debate why your approach to research is the best one to adopt.
4 You have been asked to deliver a leadership training programme to a multinational
organisation’s top 100 future leaders. Think about the possible technological, diversity
and cultural factors which may come into play in designing, delivering and evaluating
this programme.
5 How may the world of work change in the future? What implications does this have for
the work psychologist?
Relevant websites
Professional psychology associations offer considerable information about the theory and
practice of psychology, including work psychology. Three good ones are the Society for Industrial and Organizational Psychology at https://www.siop.org/; the European Association
of Work and Organizational Psychology (EAWOP) at http://www.eawop.org/ and the Division of Occupational Psychology, British Psychological Society at https://www.bps.org.uk/
member-microsites/division-occupational-psychology. For those wishing to be an occupational psychologist in the UK the Standards for the Accreditation of Masters and Doctoral
Programmes in Occupational Psychology can be found at: https://www.bps.org.uk/sites/
bps.org.uk/files/Accreditation/Occupational%20Accreditation%20(2017).pdf. The Health
and Care Professions Council is also relevant (http://www.hcpc-uk.co.uk). Also of use is the
Association for Business Psychology (https://www.theabp.org.uk/home.aspx).
The Chartered Institute of Personnel and Development (CIPD) have an excellent website
that provides access to reports and surveys on a range of important workplace issues. It is at
http://www.cipd.co.uk.
Chapter 1 The discipline of work psychology
31
Suggested further reading
Full details for all references are given in the list at the end of this book.
1 Cathy Cassell and Gillian Symon’s books Essential Guide to Qualitative Methods in Organizational Research (Sage, 2004) and Qualitative Organizational Research: Core methods and
current challenges (Sage, 2012) are superb, focused introductions to a wide range of qualitative methods and how they are used in organisational research.
2 Mark Easterby-Smith et al., Management and Business Research, 6th edition (Sage, 2018),
gives a clear and practical guide to key issues in formulating and doing research.
3 David Gray’s book on Doing Research in the Real World, 4th edition (Sage, 2018), offers a
comprehensive guide to researching within applied contexts.
4 Discovering Statistics Using IBM SPSS Statistics by Andy Field (Sage, 2017) is a very accessible text that shows how to understand and use the SPSS software package. It is peppered
with a good amount of humour to lighten the load. You can also access material from
the same author via https://www.discoveringstatistics.com/
5 For an accessible reference on reflection and reflexivity please follow this link: https://
uk.sagepub.com/sites/default/files/upm-binaries/32441_01_Bolton_3e_Ch_01.pdf.
CHAPTER 2
Individual differences
LEARNING OBJECTIVES
After studying this chapter, you should be able to:
1 outline contemporary theories of human intelligence and intelligence testing;
2 compare conventional approaches to intelligence with system models;
3 describe the main theories and models of emotional intelligence;
4 describe the research evidence for emotional intelligence and how it is assessed;
5 define the ‘Big Five’ personality constructs;
6 describe the characteristics and behaviours associated with creativity and innovation at
work;
7 discuss how creativity and innovation might be assessed;
8 describe socio-cognitive approaches to understanding behaviour at work.
Chapter 2 Individual differences
Opening case study
33
Emotional intelligences takes Aicha
Evans to top of silicon valley
For someone who has spent most of her career
in the male-dominated, engineering-driven
chip industry, Aicha Evans slips very easily
into talking about emotions. ‘This is about
how we live, how we learn, how we love’ is
a phrase she is fond of repeating when discussing the significance of today’s advanced
technology. And, when asked at an industry event recently whether artificial intelligence will supplant humanity: ‘AI is going to
be additive. Nothing is going to replace the
human heart.’ After a 12-year rise that took
her into the top ranks at Intel, Ms Evans
stepped into a job that could help to define
how AI shapes the world, taking over as
chief executive of self-driving car start-up
Zoox. Even by the standards of autonomous
vehicles, Zoox’s soaring ambitions stand out.
It has raised more than $750m and plans to
have a commercial robot taxi service on the
road next year. Most others in the field have
chosen to specialise, but Zoox is trying to do
it all: develop the software for autonomous
driving, design and build its own cars, and
run a ride-hailing service to compete with
the likes of Uber. Heidi Roizen, an early
investor and board member, zeroed in on the
thing that made Ms Evans stand out from a
crowded field of contenders: ‘It’s her EQ – her
leadership ability, her ability to relate to people.’ But she cautioned against simplistically
equating these qualities with her gender.
‘I’ve met plenty of women without EQ,’ she
said. As a rare black woman at the top of a
high-profile start-up, Ms Evans is guaranteed
to attract outsized attention in Silicon Valley.
Ms Evans certainly has the record to match
against the most hardcore Silicon Valley
engineer. She talks warmly of her early engineering roles, including writing the code to
compress video so that it can be transferred
more easily over digital networks. And in her
most recent role, she has overseen a strategy that depends on diversifying faster away
from PCs into many new areas where masses
of new data are creating demands for new
types of semiconductors. Ms Evans herself
does not duck questions about her gender
and race, but also prefers to look past them.
Asked about her own experience in an industry notorious for the low representation of
women in senior roles, she said life has been
harder than it would have been for a man –
but then added that this has been balanced
by the advantages that come from having a
different perspective to most of her peers. Ms
Evans has described a personal technique she
turns to for dealing with her minority position: she tells herself that she is representing
the many other people like her who are not
present. ‘We’re the only one in the room so
we’re the only one who maybe has the possibility of bringing a different perspective and
a different set of ideas,’ she said in a coaching
video produced for the American Management Association. ‘I had to overcome a lot,
and women around me and behind me have
to. But I’m also asking us to look at the opportunity. We’re in one of those golden ages in
technology right now. . . ?We don’t want to
leave 50 per cent of the world behind.’ Swapping the big-company culture of Intel for the
start-up world, meanwhile, will bring new
challenges. ‘She was something of a renegade
inside Intel – she was known to be very creative, to chart her own course, to stand up for
[things],’ said Ms Roizen. With a prominent
position in trying to build a revolutionary
new industry, those are qualities that will be
much in demand.
Source: Richard Waters, “Emotional Intelligences Takes Aicha Evans to Top of Silicon Valley”, Financial
Times, 18 January 2019 © Financial Times Limited 2019. All rights reserved.
Introduction
The case study demonstrates that there is a growing interest in intelligence, personality and emotion in relation to job performance and other organisational outcomes. This is happening in
many different organisational settings across the globe. However, it also raises some important
34
Work Psychology
questions: Does emotional intelligence exist? Do different types of intelligence exist? What
do tests of intelligence actually measure? How does an individual’s personality relate to performance at work? Are some people naturally more creative than others? In this chapter we
explore theories of intelligence and personality and provide a brief historical overview of the
research literature in these areas. Traditionally, studies of intelligence and of personality have
been treated separately but more contemporary theories of individual differences have begun
to integrate these approaches in order to more accurately explain behaviour. As the opening
case study illustrates, emotional intelligence is more and more being considered a critical
factor in explaining performance at work. This has been a topic of fierce debate and current
research evidence is reviewed in this chapter. Given recent global economic uncertainty, the
topic of innovation at work has become an increasingly important part of the research agenda
for many work psychologists. In this chapter we also examine the individual-level resources
(such as intellect and personality) and social resources (e.g. teamworking) that are important
in predicting innovative working.
Point of integration
Individual differences are an important feature of theories and research across the various chapters of
this text. For example, they can be linked to how employees respond to stress (Chapter 8) and how they
form attitudes (Chapter 5). Here we focus on ability and personality, but there are many other individual
differences such as cognitive styles (or typical ways of thinking) that are explored elsewhere in this text.
Historically, in attempts to explain individual differences, psychologists have tended to emphasise the role of either internal (person) factors or external (situational) factors. The debate about
the relative importance of people and situations in determining behaviour is a long-standing and
controversial issue within psychology. Many approaches have focused on the structure and measurement of individual differences without emphasising the role of situational factors in determining behaviour. It would be a mistake, however, to conclude that the approaches described do not
recognise the potential for situations to influence how people behave.
Research has led to the development of sophisticated theories of how various individual
differences and characteristics of the work environment come together to shape workplace
behaviour.
Key learning point
Personal and situational factors interact to determine behaviour. In some theories the situation is
more important than personal factors. In other theories personal factors are dominant. More recently,
integrative theories have aimed to provide more complete insights into the determinants of behaviour.
Traditional models of cognitive ability
Some people are better at processing information than others. These differences can have a
significant impact on how they perform when faced with various tasks. We are constantly
exposed to evidence that, when it comes to cognitive (thinking) ability, there are significant and relatively stable variations among people. Thanks to good quality research it is
equally clear that some of these differences in intelligence are the result of differences in
Chapter 2 Individual differences
35
opportunities to learn. Makharia et al. (2016) found a wide range of environmental factors
can have an impact on children’s intelligence, including place of residence, physical activity, family income, parental education and paternal occupation. A review of the research
evidence concluded that the importance of environment for the development of IQ (intelligence quotient) can be seen in an average six-point increase in IQ for children adopted from a
working-class environment to a middle-class environment (Nisbett et al., 2012). Psychologists
have invested a great deal of time and effort into trying to understand and measure differences
in cognitive ability. One approach to the issues involved is the use of carefully designed tests
to assess people’s levels of cognitive ability.
Point of integration
Measures of cognitive ability are widely used to select people for work roles (see Chapter 3). This
is because there is good evidence that cognitive ability is linked to work performance across many
occupations.
In the early years of the 20th century, two French psychologists, Alfred Binet and Theodore
Simon, developed what is generally accepted as the first satisfactory test of human intelligence. France, in common with other industrialised nations, had introduced compulsory
education. Most of the children entering schools seemed well able to benefit from the regular
mainstream approach to education. Some seemed to require extra help but could still prosper
in the regular system. A third group, however, did not benefit from the regular system but
it was not always easy to identify these children who needed more specialised educational
interventions. The test developed by Binet and Simon was intended to help in identifying
these children by assessing their intellectual capabilities in a reliable and standardised way.
The approach adopted by Binet and Simon is still the basis for contemporary intelligence
tests. They considered that intelligence could be measured by assessing a person’s ability to
answer a carefully selected collection of questions. Although the questions in modern tests differ from the type of questions used by Binet and Simon, the principle of sampling test-takers’
behaviour on a carefully selected set of tasks is still at the core of most tests. Such tests need
to be administered under standardised conditions and require people to attempt to answer
questions, all of which have been evolved through a highly technical process of trialling
and analysis (see Table 2.1 for illustrative questions). The standardisation of tests is achieved
through careful and consistent administration (e.g. the time allowed, and instructions given
should be exactly the same for all test-takers) and test content (all test-takers answer the same
questions). This means that, in theory, everyone is given the same opportunity to perform.
The need for standardisation also brings some disadvantages. This approach cannot assess
people on their capacity to conduct everyday, real-life tasks. This is seen by some psychologists as a serious weakness and is a criticism that is gaining support within the literature.
Key learning point
Traditional tests of cognitive ability and intelligence usually need to be administered in a standardised
way. This allows for differences between individuals to be measured as all individuals are deemed to
have been given an equal opportunity to answer the questions because they all completed the test
under the same conditions.
36
Work Psychology
Sternberg (2006, Sternberg et al., 2007) studied the concept of ‘practical intelligence’,
which refers to the ability to solve real-life problems using information that is not necessarily
contained within a problem statement. In these measures of intelligence there are many possible solutions as opposed to the one ‘correct answer’ per problem as is the case for the majority of intelligence tests. Practical intelligence has been found to be only moderately correlated
with traditional IQ tests, yet it is predictive of significant amounts of variance in academic
and occupational achievement above and beyond that explained by IQ tests alone (in other
words, it shows incremental validity). Lievens and Chan, in their 2017 review, argue that
while practical intelligence may not be a unitary or distinct construct, it still has the potential
to add value.
This has led several authors (see, for example, Hunt, 2011) to conclude that measuring
practical intelligence could significantly improve the predictive power of intelligence testing.
Another important, but often overlooked, issue is that cognitive ability is often tested
within a maximal performance paradigm: test-takers are instructed to do the very best they can
do within a set time frame. This contrasts with the typical performance paradigm in which
information is gathered about what test-takers tend to do. Clearly, by sampling behaviour in
the maximal performance paradigm alone, one runs the risk of drawing false conclusions
about a person. Perhaps because of the particular questions asked, the circumstances in which
the test is taken, or various other reasons the person’s test performance may not reflect their
underlying abilities. That being said, careful test construction can help to moderate some of
this effect. We look at this issue in more detail throughout the chapter. It is worth pointing
out that Binet and Simon certainly recognised that test scores alone were not enough, and
they proposed that, before any decision about a child’s education was taken, other types of
assessment should also be made.
Intelligence testing
Since the pioneering work of Binet and Simon, psychologists have carried out a considerable
amount of work in their attempts to measure intelligence and to understand its structure. So
far, we have managed to discuss the underlying structure of intelligence without directly confronting the problem of what intelligence is! Defining intelligence presents problems for psychologists, and to this day there is no universally accepted definition. Many psychologists will
settle for the definition first proposed by Boring (1923: 35): ‘Intelligence is what intelligence
tests measure.’ In fact, this definition is not as meaningless as it seems at first sight. Tests of
general intelligence (which Spearman, 1927, labelled ‘g’) are designed to examine the innate
ability of people to carry out various mental operations. These include the ability to manipulate
data, solve problems, interpret information and so on as shown in the examples in Table 2.1:
don’t worry if you find them difficult – they are! The various tests of general intelligence are
all interrelated, and people tend to obtain similar scores in different well-constructed tests:
this is thought to be because to some extent the same underlying construct, ‘g’, has an impact
on performance across different tasks and tests. Thus, it has been widely argued that ‘g’ is a
quality that can be measured reliably and with some precision. Regardless of whether the
available tests measure what we think of as intelligence or some other qualities, several factor
analyses of models of intelligence have shown that the existence of ‘g’ is difficult to dispute
(Bickley et al., 1995).
There are a wide range of different tests available to measure ‘g’. There is also plenty of
evidence to indicate that ‘g’ determines performance across different job roles (see Chapter 3). However, it is important to recognise that ‘g’ does not provide a perfect prediction of
performance regardless of the type of task. Specific abilities have been shown to also influence
performance with their impact being dependent on the type of information being handled or
the specific tasks being completed. For example, in addition to ‘g’, the ability to work quickly
and accurately with numbers would be linked to test-takers’ success in dealing with problems
Chapter 2 Individual differences
Table 2.1
Q7
Some example questions from an intelligence test
Mountain is to molehill as valley is to . . .
1
2
3
4
hollow
chasm
hill
plain
5
mound
Q8
The third member of this series is omitted. What is it?
0.1, 0.7, . . . , 34.3, 240.1
Q9
Which one of the five words on the right bears a similar relation to each of the two words on the left?
1
2
3
4
5
Class; shape Rank
Grade
Analyse
Size
Form
Q10
Here are five classes. Write down the number of the class which contains two, and two only, of the other four classes:
1
2
3
4
5
Terriers
Mammals ‘Scotties’ Dogs
Canines
Q11
Sniff is to handkerchief as shiver is to . . .
1
2
3
4
blow
fire
catarrh
burn
5
sneeze
Q12
How many members of the following series are missing?
1, 2, 5, 6, 7, 11, 12, . . . , 20, 21, 22, 23
Q13
Which one of the five words on the right bears a similar relation to each of the two words on the left?
1
2
3
4
5
Stream; tolerate
Brook
Contribute Bear
Support
Pour
Working from the left, divide the fourth whole number by the fifth fraction:
Q14
37
8, 6, 5/7, 3, 9, 2/9, 3/8, 1, 17/31, 4/9
Source: The AH5 Test. Copyright © 1968, A. W. Heim. Reprinted by permission of the publisher ASE, College Lane,
Hatfield, Herts, AL10 9AA, UK.
involving money, weights, distances and other concepts expressed in a numerical way. These
core specific ‘primary’ ability factors tend to predict performance when they are ‘matched’
to the demands of the job role. For example, verbal ability may predict the performance of a
journalist, while numerical ability may more strongly predict the performance of an accountant. Indeed, measures of these primary abilities appear to predict variability in task performance that is not indicated by ‘g’ alone. Tests of both ‘g’ and more specific abilities have been
used with some success in personnel selection (see Chapter 3 for a more detailed discussion).
Key learning point
General intelligence ‘g’ has been linked to performance across a range of work-related tasks. Other
more specific abilities also appear to influence performance. The specific abilities required vary according to the demands of the task.
Tests of intelligence have been criticised on various grounds. One criticism is based on the
argument that intelligence tests do not measure pure underlying intelligence, but a mixture
of intelligence and of taught or acquired knowledge (sometimes referred to as crystallised
intelligence).
Proponents of intelligence tests believe that the development of ‘pure’ tests of intelligence
can be done; others consider that it has not been done properly and is probably impossible.
Furthermore, tests that focus more heavily on one or another form of intelligence might
pose difficulties for certain age groups due to differences in the rate of decline of each type
38
Work Psychology
of cognitive ability. For example, measures such as Raven’s Progressive Matrices (RPM Raven
et al., 1996, 2003) that focus on fluid intelligence – i.e. the capacity to think logically in new
situations and without acquired knowledge – will have scores that decline with age at a faster
rate than measures that focus on crystallised intelligence. While there is recent evidence that
fluid intelligence can be enhanced through training (e.g. Au et al., 2015), it is worth noting
that age differences may occur dependent on the type of test used.
Key debate
Are cognitive tests culturally biased?
In the personnel selection context, tests are also criticised because they are biased in favour of certain ethnic or
cultural groups. The argument of cultural bias asserts that the intellectual development that takes place naturally
is dependent on the specific environmental and cultural background in which a person develops. This means that
perfectly bright and intelligent people from certain socio-economic or ethnic backgrounds will have experiences that
impact (either positively or negatively) on the development of the qualities assessed in the tests. The consequence
will be that the test results will not accurately reflect their underlying intelligence.
The criticism that intelligence tests are biased against certain ethnic groups is, at least in part, based on the frequently replicated research finding that some ethnic minority groups obtain lower scores on cognitive tests than others
(see Sackett et al., 2008 for a review of evidence). This has led to much discussion and heated debate in the scientific
community. Despite the subgroup differences, many have argued that it is not unfair to use such tests for selection
decision-making. In essence, this conclusion is based on the finding that although there are consistent differences
between subgroups in mean scores, the accuracy of prediction of the tests (i.e. the prediction of future levels of work
performance) is the same for different ethnic groups.
There is now some evidence to the contrary. A meta-analysis examined ethnic differences in the criterion-related
validity of cognitive ability tests. This research examined many different high-quality studies of the correlations
between test scores and criterion measures such as task performance. The results showed that compared with the
average correlations for White individuals, correlations between test scores and performance were lower for Blacks
and Hispanics, but similar for Asians. This suggests that despite their standardisation, cognitive ability tests may not
provide a fair assessment of potential for every ethnic group (Berry et al., 2011). This is particularly important in selection scenarios, where test scores result in job offers or promotions. In such settings, the validity of intelligence tests
is vital, and therefore the finding that validity favoured White over Black and Hispanic test-takers across the majority
of studies in Berry et al.’s (2011) meta-analysis is very important.
Encouragingly, there is a consensus throughout the literature that there is no evidence
of gender differences in overall measures of ‘g’. Having said that, the sexes may have different propensities towards certain aspects of intelligence: males may display an advantage
towards visuospatial abilities (our ability to process visual information about the location of
objects) and females towards verbal abilities (Nisbett et al., 2012). Of course, there are also
many wider, societal issues to consider when examining the various influences on test performance. These include any links between demographic factors and access to educational
opportunities and any connections between socio-economic factors and opportunities for
intellectual development: these are complex and important societal issues but fall outside
the remit of this book.
A final criticism of intelligence tests is that while individual differences in ability are
assumed to exist, there is little research examining differences in test-taking motivation.
Duckworth et al. (2011) argue that although intelligence tests operate on a maximal performance paradigm, they do not always succeed in maximising effort. A common assumption
is that test-takers are alert, motivated, and fully focused on the task, and thus can do their
very best within the time limit. While this is truer of high-stakes settings (e.g. when the result
Chapter 2 Individual differences
39
is very important to the test-taker), research has indicated that for lower-stakes settings, this
is unlikely to be the case. This poses the question: are intelligence tests really measuring a
person’s best performance?
Recent studies indicate not. Findings have shown that IQ scores can be increased by
heightening motivation through material incentives (tangible rewards) in low-stakes conditions (Duckworth et al., 2011). Self-reported motivation has been found to be higher in
high-stakes settings. It is worth noting that this is typically measured after the test, so it is
possible that responses are clouded by test takers’ estimates of their performance as opposed
to an honest reflection of how hard they tried. Stress can also influence performance on
intelligence measures, since high levels of stress hormones can affect areas of the brain that
are important for regulating attention and memory. Perhaps the typical rather than maximal
paradigm would reflect a truer representation of test-takers’ intelligence in the absence of
stress and with optimal motivation. It is noteworthy that long-term stress can also impact on
intelligence more generally and may partially explain some of the ethnic group differences
reported in several research studies (e.g. Sharkey, 2010). Research suggests that the IQ gap
between ethnic groups may be explained by the finding that children from some groups are
more likely to live in stressful environments (for various reasons unconnected to their ability)
where they are consistently exposed to stressors. This is a relatively new avenue of research
and is an important area for future study.
Key learning point
There are consistent differences in the general mental ability test scores obtained by different subgroups. These may occur for several reasons and may not be linked to future job performance.
Systems models of intelligence
Unlike previous structural models of intelligence, systems models expand the concepts
underlying intelligence to include concepts other than cognitive abilities. Three prominent
approaches are the triarchic theory of intelligence (Sternberg, 1985), the three-stratum
theory of cognitive ability (Carroll, 1993) and Ackerman’s (1996) PPIK (Process, Personality,
Interests and Knowledge) theory. An emerging approach is that of emotional intelligence
(see Goleman, 1995; Bar-On, 2000; Mayer et al., 2001; Joseph and Newman, 2010).
Sternberg’s triarchic theory of intelligence
Robert Sternberg’s triarchic theory (Sternberg, 1985) builds on Spearman’s ‘g’ and the underlying information processing components of intelligence. The theory consists of three facets
(also known as subtheories) used to describe and measure intelligence. The three facets are:
analytical (componential), creative (experiential) and practical (contextual). Each is defined
as follows:
■
Analytical subtheory: Analytical intelligence refers to academic problem-solving (such
as solving puzzles) and reflects how an individual relates to their internal world. Sternberg suggests that analytical intelligence (problem-solving skills) is based on the joint
operations of ‘metacomponents’, ‘performance’ components and ‘knowledge acquisition’ components of intelligence. Metacomponents control, monitor and evaluate cognitive processing. These are the executive functions that help to organise strategies for
Work Psychology
40
performance and the process of knowledge acquisition. Performance components are
the basic operations that execute the strategies governed by the metacomponents. They
are the cognitive processes that enable us to encode information, hold information in
short-term memory, perform mental operations and recall information from long-term
memory. Knowledge acquisition components are the processes used in gaining and storing (memorising) new knowledge (i.e. the capacity for learning).
■
Creative subtheory: Creative intelligence involves the insights people have, their ability to
synthesise (e.g. have new ideas) and their ability to react to novel situations and stimuli.
Sternberg suggests that this experiential aspect of intelligence reflects how an individual associates their internal world to the external world. Creative intelligence comprises
the ability to think creatively and the ability to adapt creatively and effectively to new
situations.
■
Practical subtheory: Practical intelligence involves the ability to understand and deal
with everyday tasks. It is referred to as ‘real-world intelligence’. This contextual facet of
intelligence reflects how the individual relates to the external world. People with high
levels of this type of intelligence can adapt to or shape their environment. Sternberg and
Wagner’s test of Practical Managerial Intelligence includes measures of the ability to write
effective memos, to motivate people, to identify when to delegate and so on. In contrast to
structural models of intelligence, measures of practical intelligence go beyond mental
skills and include assessment of attitudes and emotional factors (Sternberg et al., 2000).
Carroll’s three-stratum theory of cognitive ability
General
(stratum III)
John Carroll proposed a three-stratum theory of cognitive ability, based on a re-analysis of
over 450 datasets of cognitive ability test results. He used factor analysis to identify consistent patterns across test-takers’ performance to see if there were clusters of components of ‘g’
arranged in a hierarchy. His analyses showed that cognitive ability could be reliably clustered
into three layers (strata), defined as representing narrow, broad and general cognitive ability
(see Figure 2.1). Stratum III most closely resembles ‘g’, reflecting one factor of ‘general intelligence’. Stratum II splits this general factor into broad abilities such as fluid or crystallised
intelligence, general memory and visual perception. Stratum I further narrows these into
Narrow
(stratum I)
Broad
(stratum II)
g
Gc
Gf
f1
Figure 2.1
…
f2
c1
Gy
…
c2
y1
Gv
…
y2
v1
Gu
…
u1
v2
Carrol’s three stratum theory of cognitive ability
Gr
…
u2
r1
Gs
…
r2
s1
Gt
…
s2
t1
…
t2
Chapter 2 Individual differences
41
specific abilities such as picture recognition and word fluency: each of these are aspects of the
general factors in Stratum II. This model has been described as ‘the most viable description of
human abilities’ (Bickley et al., 1995: 311), and can be thought of as a unifying model since
it combines and extends the dimensions of other established theories.
Ackerman’s PPIK theory of adult intelligence
Ackerman (1996) proposed a developmental theory of adult intelligence, referred to as PPIK
theory. This model of intelligence incorporates intelligence as process (P), personality (P),
interests (I) and intelligence as knowledge (K). A critical aspect of Ackerman’s theory is that
intelligence is expected to grow over time due to the acquisition of new knowledge.
The justification for this model was two-fold. The first justification drew on traditional
definitions of intelligence which focus on what an individual can perform or achieve
but these may only have a small impact on how that individual were to score on a traditional intelligence test. The second justification drew on explorations of knowledge-based
and skill-based differences among experts and novices in a variety of fields (e.g. in musicians; Ericsson et al., 1993) which point towards a knowledge-based view of intelligence.
This knowledge-based view of intelligence posits that intelligence is only possible with
knowledge. Simply, this means that intelligence itself consists of both the capacity for knowledge and actual knowledge.
Furthermore, Ackerman justifies the inclusion of knowledge and skill within his PPIK
theory by considering two theoretical approaches to adult intelligence; Vernon (1950) and
Cattell’s (1971, 1987) Investment Theory. Vernon’s (1950) theory of intelligence is essentially hierarchical with general intelligence at the highest node, two broad group factors at
the next node and unspecified lower node abilities. In addition to this, Vernon reviewed
the relationships between intellectual abilities and what is referred to as ‘educational attainments’ consisting of the knowledge and skills concerning factors such as ‘scientific subjects’,
‘literature’ and ‘vocabulary’ and are consistent with what is widely considered to be intelligence. Through extensive research, Cattell (1984) expanded and diverged intelligence into
two broad areas: fluid intelligence (Gf) and crystallised intelligence (Gc). Cattell’s Investment
Theory posits that Gc is influenced by the time invested in intellectual pursuits and historical
and current interests. In justifying the addition of personality to Ackerman’s model, he cites
research indicating that there are critical and general relationships between personality and
intelligence and poses that no theory of intelligence can be complete without considering
how personality, interests and abilities interact with each other to determine the knowledge
that an individual develops over their lifespan.
In essence, the inclusion of these four domains within Ackerman’s theory are based on
traditional theories of intelligence which have resulted in the concepts of crystallised intelligence
and fluid intelligence, the literature linking personality with intelligence and the premise that
intelligence is linked to an individual’s investment in their own intellectual interests.
Key learning point
Sternberg’s triarchic theory of intelligence redefines intelligence to incorporate a practical knowledge
component. Carroll’s three-stratum theory of cognitive ability unifies and extends earlier theories
by conceptualising intelligence using a three-layered model of general, broad and specific intelligence. Ackerman’s model of intelligence incorporates intelligence as process, personality, interests
and knowledge; a critical aspect of this theory is that intelligence is expected to grow over time due
to the acquisition of new knowledge.
42
Work Psychology
Emotional intelligence
Emotional intelligence (EI) is a concept that has generated a great deal of interest. Since Daniel
Goleman’s book Emotional Intelligence was published in 1995, many researchers have initiated
studies of EI, believing that EI fills a gap in knowledge. Not surprisingly, this has led to heated
debate among researchers and practitioners who are keen to define EI. This has prompted
questions about how it differs from established theories of intelligence and how it is best
measured. For practitioners, a key question is whether EI predicts aspects of job performance
that cannot be predicted by measuring cognitive ability alone (see Chapter 3 for a discussion
of validity in assessment).
Early research led to three main conceptualisations of EI appearing in the literature: one
proposed by Goleman, one by Bar-On and the third by Mayer and Salovey (e.g. Bar-On, 2000;
Goleman, 1995, 1998; Mayer and Salovey, 1997). More recently, Joseph and Newman (2010)
conducted a meta-analysis of existing research: this led them to propose a ‘cascading model’
of EI that explains the progressive effect of the different facets of EI on job performance.
Although there is some obvious commonality, there exists some significant divergence of
thought in these approaches. We now briefly review each in turn.
Goleman defined EI as:
abilities such as being able to motivate oneself and persist in the face of frustrations; to control
impulse and delay gratification; to regulate one’s moods and keep distress from swamping the
ability to think; to empathize and to hope.
(Goleman, 1995: 34)
Some have argued Goleman’s definition of EI is overly inclusive and repackages previous
literature on personality and intelligence without contributing anything new (Chapman,
2000). Matthews et al. (2003b) suggest that Goleman’s conceptualisation of EI rests on gathering up various aspects of what psychologists today would describe as cognition, motivation,
personality, emotions, neurobiology and intelligence.
Goleman has responded by insisting that EI is an ability that differs from other more established abilities (although there is limited scientific evidence to support some of his claims).
In later publications, Goleman (1998, 2001) suggests how his EI theory represents a framework of an individual’s potential for mastering the skills in four key domains: self-awareness,
self-management, social awareness and relationship management. Unlike other conceptualisations of EI, Goleman’s framework (2001) specifically refers to workplace behaviours and
is based on content analyses of competencies in several organisations. Rather than being
purely an innate ability, he defines an emotional ‘competence’ as ‘a learned capability based
on emotional intelligence that results in outstanding performance at work’. Each of the four
domains is viewed as a competency. For example, in considering self-awareness, this is the
extent to which an individual is accurate (competent) in their self-assessment of their personal strengths and limitations.
Equally influential in this area is the work of Reuven Bar-On (1997, 2000). He defines EI as
‘an array of non-cognitive capabilities, competencies and skills that influence one’s ability
to succeed in coping with environmental demands and pressures’ (1997: 14). He produced
the first commercially available measure of EI, based on a self-assessment instrument called
the Emotional Quotient Inventory (EQi). Bar-On (2000) presents his model as an array of traits
and abilities related to emotional and social knowledge: these influence our ability to cope
effectively with environmental demands. In this way, it is a model of psychological well-being
and adaptation. His model includes four domains, including the ability: to be aware of, to
understand and to express oneself (intrapersonal intelligence); to be aware of, to understand
Chapter 2 Individual differences
43
and relate to others (interpersonal intelligence); to deal with strong emotions and control
one’s impulses (stress management); and to adapt to change and to solve problems of a personal or social nature (adaptability). Bar-On has reported several validation studies of the
EQi and there is reasonable evidence that it predicts academic success and the diagnosis of
some clinical disorders. However, as with Goleman’s work, one of the main criticisms centres
around the question of whether it captures any construct that is unique and is not already
captured in existing personality measures (see Mayer et al., 2000).
The research of Jack Mayer and Peter Salovey (1997) has perhaps been the most influential. Unlike other approaches, they describe EI as extending traditional models of
intelligence. They define EI as ‘a concept of intelligence that processes and benefits from
emotions. From this perspective, EI is composed of mental abilities, skills or capacities’
(Mayer et al., 2000: 105). They suggest that traditional measures of intelligence fail to
capture effectively individual differences in the ability to perceive, process and manage
emotions and emotional information. Their approach defines EI as the ability to perceive
emotions, to access and generate emotions to assist thought, to understand emotions and
emotional knowledge and to reflectively regulate emotions to promote emotional and intellectual growth (Mayer and Salovey, 1997). For measurement purposes, Mayer et al. (1999)
have produced a Multifactor Emotional Intelligence Scale (MEIS) to assess facets of EI, comprising 12 subscales (e.g. feeling biases, managing self and managing others), all of which
are connected to a single underlying factor (an ‘emotional “g”’). Unlike other approaches,
this model attempts to measure EI as a distinct concept, i.e. something that is not captured
in established measures of personality. However, the evidence on the validity of the MEIS
as a distinct concept is mixed.
Mayer et al. (2016) revisited Mayer and Salovey’s (1997) model of EI; refashioning their
original ability model, clarifying elements of their model which were unclear and reflecting
on current research in the area. In addition to this, they suggested positioning EI amid other
hot intelligences such as personal intelligence and social intelligence. Mayer et al. (2016)
posit that the evidence base for EI points to it existing as a mental ability within the broad
range of hot intelligences.
There is the possibility that EI operates as part of a larger construct of intelligence rather
than a separate and unique type of intelligence which would mean that any measurement of
EI would be nothing more than a subscale of a broader test. However, Mayer et al. (2016) propose that EI is partly distinct from other intelligences and pose that in understanding more
about how people solve problems in the area of emotion, we can better teach people how to
do this. In a time when artificial intelligence (AI) is on the rise in many fields including work
psychology, understanding how this works could enable systems to be created that emulate
human reasoning and emotion.
Stop to consider
After reading about these conceptualisations of EI, ask yourself the following questions:
1
2
3
Do you think that these conceptualisations of EI are leaving anything out? If so, what?
What do you think are some of the challenges of attempting to measure EI?
What do you feel the impact would be, in general and for work psychology, of AI having the ability
to understand and simulate human emotion?
44
Work Psychology
One of the more recent models of EI was proposed by Dana Joseph and Daniel Newman
(2010). Their meta-analysis aimed to clarify the theoretical basis for EI, and generated a cascading model of emotional intelligence, in which a causal chain exists between three subfacets
of EI: emotion perception, emotion understanding, and emotion regulation; and job performance. The model also includes personality (specifically, conscientiousness and emotional
stability) and cognitive ability as important antecedents of various stages of the EI process.
Emotion perception refers to the ability to identify emotions in oneself and others; this influences emotion understanding, which refers to understanding how emotions evolve over time,
how emotions differ from each other, and which emotion is most appropriate for a given context;
this then influences emotion regulation, which is concerned with the process by which individuals influence which emotions they have, when they have them, and how they experience and
express these emotions. It is then emotion regulation in turn that influences job performance
(alongside aspects of personality and ability). Joseph and Newman (2010) suggest that emotion regulation is theoretically related to job performance through the generation of affective (mood) states that are beneficial to job performance (such as pleasure or excitement).
Positive affective states have been linked to favourable work behaviour, and those with a
greater ability to regulate their emotions would be more likely to create and maintain these
states over time and across situations. Similarly, they would be less likely to use ineffective
emotion regulation strategies such as ‘suppression’ of negative affect or ‘surface acting’. In
other words, the behaviour does not match the emotions being felt: these strategies cover
up rather than actually change a negative emotional state. Emotional stability is proposed
as an antecedent of emotion regulation since individuals reporting high levels of neuroticism have been found to regulate their emotions less effectively than those with high levels
of emotional stability. Gross and John (2003) found that individuals with a high level of
neuroticism do not engage in effective emotion regulation strategies as often as emotionally stable individuals (e.g. by re-appraising or re-interpreting the meaning or significance
of events).
Key learning point
Specific aspects of personality and ability are linked to some of the emotional regulation processes in
the cascading model of EI.
The cascading model also suggests that conscientiousness is positively related to emotion
perception since conscientious individuals (described as methodical, cautious and careful)
have shown above average levels of interpersonal functioning and an increased capacity for
self-conscious emotions (the emotions linked to how we see ourselves and how others see
us, such as pride). This means that conscientious individuals may be in a better position to
use emotional cues (such as facial expressions) to guide their need for controlled behaviour.
For example, they will be able to quickly and effectively determine when various behaviours
are appropriate or inappropriate. Cognitive ability has been linked to the emotion understanding phase: those with high levels of cognitive ability would have greater knowledge
of how to understand their own and others’ emotions. Meta-analyses drawn from lots of
different research projects and diverse research populations have shown a good fit with this
cascading model.
Clearly there are some similarities, but also some important differences, between the four
approaches. Bar-On has tried to develop a general measure of social and emotional intelligence associated with psychological well-being and adaptation. Mayer and Salovey attempted
Chapter 2 Individual differences
45
to establish the validity of a new form of intelligence, whereas Goleman’s approach is specific to behaviour in work organisations based on the ability to exhibit social and emotional
competencies. Joseph and Newman took a more integrative approach and explored the relationships between established aspects of EI and different factors including personality, intelligence and job performance.
Key debate
Do measures of emotional intelligence work?
Just how good are the measures of EI? Not surprisingly, those who develop the measures tend to be confident
about their quality and utility. Conte (2005) reviewed the research evaluating the most prominent measures of
EI, including Goleman’s Emotional Competence Inventory (ECI), Bar-On’s Emotional Quotient Inventory (EQi), and
Mayer and Salovey’s Multifactor Emotional Intelligence Scale (MEIS). He concluded there were ‘serious concerns’
regarding all three measures, for example with the way ability-based EI tests were scored (can there be unequivocally correct and incorrect answers for questions about the awareness and regulation of emotions?) and
the discriminant validity of self-report EI measures. As a result of his findings, Conte (2005) cautioned against
the use of EI measures for selection purposes, until further research has shown EI to be a valid concept. It is
also important to consider the way in which emotional intelligence is to be measured. As EI can be considered a
multidimensional construct with a crucial interpersonal component, should we be measuring EI through others’
perceptions rather than just self-reports? For instance, Elfenbein et al. (2015) conducted a large exploration of
self- and other-reports of emotional intelligence and found that predictions of performance were more accurate
when based on other-rated measures.
Point of integration
Measures used in work psychology need to demonstrate several important properties such as reliability, validity and fairness. These are considered in detail in Chapter 3.
One of the most important contributions of Joseph and Newman’s work is that it has shed
some light on some of the burning questions surrounding the utility of EI: is EI a unique and
valid concept? Does it add anything unique over and above personality and ability measures?
According to their meta-analysis, the answer is: probably not. Joseph and Newman studied
EI in two senses. One in relation to ability-based EI – that is, EI as a narrow set of constructs
pertaining to the recognition and control of personal emotion, and another in relation to
mixed EI – an umbrella term for a broad range of relevant constructs. Ability-based measures of EI were more theoretically grounded than mixed EI measures. However, scores on
ability-based EI measures did not predict job performance across job types: predictive validity was localised to specific professions and offered only a small incremental validity over
the Big Five personality traits and cognitive ability. Sub-group differences were also found,
with women and White individuals achieving higher scores on ability-based EI measures.
It is important to consider the added value of EI measures and consider the adverse impact
that these measures may create.
Moreover, large differences in predictive validity and subgroup differences were found
between the different types of EI measures. Unlike ability-based EI measures, results for
46
Work Psychology
mixed EI measures such as the EQi (Bar-On, 2000) show significant overlap with the Big Five
dimensions and largely support critics’ claims that the measure does not capture anything
not already captured by existing personality measures. Despite this, research has found
EI to be a predictor of success in different contexts. For instance, Libbrecht et al. (2014)
found that EI had the potential to predict performance in medical school over and above
conscientiousness and cognitive ability. With mixed results for the predictive validity of
EI, further research is needed in the area to explore this predictive validity in different
occupational contexts.
Key learning point
The existence of emotional intelligence as a new construct, distinct from other established constructs,
has been vigorously debated. Although there is an acceptance that EI is multifaceted, there is still
disagreement about what these facets are. The research does not support the widespread use of EI
measures in job selection processes due to their lack of practical utility and the inconsistencies in
sub-group differences between measures that may lead to adverse impact.
More research is required to answer many questions about EI. There is no doubt that the
debate associated with EI has begun to challenge previous assumptions of what leads to success in organisations, and indeed, many organisations have seen positive results come from
focusing on EI. Perhaps the concept is better used in a developmental context than from a
selection and assessment perspective?
Key learning point
Emotional intelligence is seen by many as an exciting concept because it encompasses thought,
emotion and interpersonal awareness more than earlier models of intelligence. While there is some
evidence from specific workplaces that improving EI can improve organisational outcomes like performance, researchers tend to advise caution around using EI measures for selection.
Trait views of personality
Psychoanalytic theories of human personality (e.g. the work of Freud) are often criticised
by other psychologists. They argue that these theories often lack scientific rigour or fail
to clearly define their key concepts. From a scientific perspective, the most fundamental
criticisms are that these theories either do not generate testable predictions about human
behaviour or, if they do, when predictions are made, they do not work out in practice.
Trait theories perhaps come closest to describing the structure of personality in a way that
matches our everyday use of the term personality. Trait theories use words such as shy, outgoing, tense and extroverted to describe the basic factors of human personality. These basic
elements – traits – represent tendencies to behave in certain ways, in a variety of different
situations.
Chapter 2 Individual differences
47
Point of integration
Trait-based theories have been prominent in leadership research. These indicate that certain levels of
particular personality traits help people to become, and function as, effective leaders. This approach
has attracted a great deal of criticism (see Chapter 10). There is considerable evidence that personality
traits such as neuroticism are linked to the experience of work stress: here the evidence of a link is
consistent. However, the extent to which personality is linked to stress when compared to the effects
of working conditions remains controversial (see Chapter 8).
Over several decades’ academic debate, research has produced some consistent evidence of the
existence of the so-called Big Five major personality factors. A fairly high degree of consensus has
emerged, and investigators have agreed that a five-factor structure represents an almost universal
template for describing the basic dimensions of personality (Digman, 1990; McCrae and Costa,
1990; Costa and McCrae, 1992). Costa and McCrae’s work found that within each of these dimensions there appear to be several facets (or subdimensions). These facets help us to understand the
breadth of each of the Big Five and to appreciate how the model attempts to measure the richness
of individual differences in personality. The Big Five dimensions and facets are listed below:
Openness to experience – fantasy, aesthetics, feelings, actions, ideas, values. People who
report themselves to have high levels of O tend to like working with ideas and possibilities
(as opposed to established methods), and they are ready to re-examine their attitudes
and values.
Conscientiousness – competence (i.e. feeling capable), preference for order, dutifulness,
achievement striving, self-discipline, deliberation (e.g. giving thorough and careful
thought to tasks). People who report themselves to have high levels of C tend to describe
themselves as being highly organised and thorough in their approach to tasks, implying
a desire to do things well.
Extroversion – warmth, gregariousness, assertiveness, activity (energy), excitementseeking and the experience of positive emotions. People who report themselves to
have high levels of E tend to describe themselves as outgoing, gregarious, lively and
sociable.
Agreeableness – trust, straightforwardness, altruism, compliance, modesty,
tender-mindedness. People who report themselves to have high levels of A tend to
describe themselves as being helpful to others and mindful of others’ feelings, and as
preferring cooperation to competition. A typically manifests itself in behavioural characteristics that are seen as kind, sympathetic and cooperative.
Neuroticism – anxiety, angry hostility, depression, self-consciousness, impulsiveness and
vulnerability. People who report themselves to have high levels of N tend to describe
themselves as prone to worry and self-doubt and being highly affected by their emotions
in stressful situations.
Key learning point
The Big Five has become a widely accepted template for understanding the structure of human personality. Tip: you can remember the Big Five dimensions using the acronym OCEAN.
48
Work Psychology
Substantial evidence exists that the Big Five structure is consistent across various
national groups. For example, McCrae and Costa (1997) reported results comparing six
diverse samples (German, Portuguese, Hebrew, Chinese, Korean and Japanese) showing all
to have substantial similarity in a Big Five structure when compared with a large American
sample. This and other evidence suggests very strongly that the Big Five structure is a useful general framework although, as McCrae and Costa acknowledge, this may be limited
to modern, literate, industrialised cultures. It also seems that the Openness to experience
dimension is less well defined than the others (Ferguson and Patterson, 1998). Lee and
Ashton (2004) suggest that there may be a sixth factor called honesty-humility (with facets of sincerity, fairness, greed avoidance and modesty), a measure of which is found in
their HEXACO personality questionnaire and subsequent revised HEXACO-PI-R (Lee and
Ashton, 2018).
The identification of the Big Five personality factors is an important development in
the trait-factor analytic approach. It is important to realise, though, that the establishment of the Big Five does not mean that other conceptualisations of personality become
redundant (Hough and Oswald, 2000). The Big Five provide a useful view of the minimum factors that must be included in any description of human personality. In many
circumstances it may make sense to use a more detailed set of dimensions. Hough and
Ones (2001), for example, make an important distinction between the use of personality
variables for description and their use for predicting job performance. In other words, while
the Big Five may be a useful framework for describing the different aspects of personality, they may not be as accurate when predicting job performance. Several studies and
meta-analyses have therefore examined the relationship between the broad Big Five factors and job performance, but debate persists regarding their suitability for predicting job
performance.
Research methods in focus
Lower-order personality traits
Researchers have aimed to provide a more comprehensive review of the role of personality in job performance by
focusing on lower-order traits as opposed to the broad factors (the Big Five). Judge et al. (2013) argue that weak
overall relationships with job performance may be masking significant relationships at the facet level, perhaps
because some subtle facets of personality are particularly important for effective work performance. As such,
using DeYoung et al.’s (2007) widely cited typology, they conducted a comprehensive test of all of the recognised
lower-order Big Five personality facets. Each of the Big Five dimensions were divided into two distinct lower-order
traits that in turn represented the six facets of each dimension described earlier (see the middle column of
Figure 2.2 – you could call these the ‘Big 10’).
This meta-analysis yielded several interesting and rather intricate findings about the links between personality
and work performance. Conscientiousness and its facets showed the highest correlation with performance, with
industriousness correlating slightly higher than orderliness. At the facet level, achievement-striving, dutifulness
and self-discipline showed the strongest correlations with job performance. Agreeableness overall had a weaker
correlation than conscientiousness with job performance, though still relatively high. Its lower-level traits did
not diverge much in their correlations with performance, but at the facet level there were larger differences,
particularly regarding modesty and tender-mindedness (tender-mindedness correlated more highly with job
performance). Rather expectedly, Neuroticism overall had a weak correlation with job performance (since one
might reasonably expect the frequent experience of negative emotions to adversely impact on performance). At
the lower-order trait level, however – although still weak – withdrawal correlated slightly higher than volatility.
In terms of Extraversion, the broad trait correlated quite highly with job performance, and lower-order traits
showed little difference in their correlations with job performance. At the facet level, there was distinctly more
variation, with very small effects for excitement-seeking but larger, albeit modest ones for positive emotions.
Chapter 2 Individual differences
NEO subfacets
DeYoung et al. facets
Big Five broad traits
Achievement striving
Competence
Industriousness
Self-discipline
Conscientiousness
Deliberation
Dutifulness
Orderliness
Order
Tender-mindedness
Altruism
Compassion
Trust
Agreeableness
Compliance
Modesty
Politeness
Straightforwardness
Angry hostility
Volatility
Impulsiveness
Anxiety
Neuroticism
Depression
Self-consiousness
Withdrawal
Vulnerability
Ideas
Intellect
Actions
Openness
Aesthetics
Aesthetic
openness
Fantasy
Feelings
Values
Gregariousness
Positive emotions
Enthusiasm
Warmth
Extraversion
Excitement-seeking
Activity
Assertiveness
Assertiveness
Figure 2.2
Hierarchical representation of the broad Big Five personality dimensions, to DeYoung
et al.’s lower-order traits, to the six subfacets of each dimension.
Source: Judge et al. (2013).
49
50
Work Psychology
Such findings indicate that personality-to-performance relationships at the facet level can be
obscured if the broad Big Five dimensions are examined alone. At the facet level, the Openness
to experience facets displayed the most variation of all the Big Five traits. Overall it correlated
moderately with job performance, and at the lower-order trait level intellect correlated higher
with performance than aesthetic openness. This research provides convincing evidence that
lower-order traits and facets can help us to predict organisational-relevant behaviours. The evidence is that the facets of each of the Big Five have different links to job performance.
Key learning point
The lower-order traits and facets of the Big Five have links to job performance that might be obscured
if we only examine the links between the Big Five broad traits and performance (see Figure 2.2).
Research has also indicated a role for personality in other organisational-relevant behaviour such as organisational citizenship behaviours (OCBs – see Chapter 4 for a fuller description of OCBs). High levels of conscientiousness and agreeableness have frequently been found
to be linked to OCBs. However, Chiaburu et al. (2011), in a meta-analysis of 87 independent
samples, found emotional stability (that is, the opposite of neuroticism), extraversion and
openness to experience also predicted OCBs, even after accounting for the effects of conscientiousness and agreeableness. Additional analyses compared the sizes of the relationships
between each of the Big Five dimensions and both OCBs and task performance. Findings
showed that conscientiousness, emotional stability and extraversion relate similarly to OCBs
and task performance, but openness and agreeableness relate more strongly to OCBs than to
task performance. This suggests that certain personality traits may be more (or less) important
for specific organisational behaviours and facets of work performance.
More recent literature in the area has begun to explore the mediating and moderating
effects of different factors on the relationship between personality factors and OCBs. For
instance, Phipps et al. (2015) investigated the moderating role of impression management on
the relationship between the Big Five and OCB while also proposing that counterproductive
work behaviour (CWB) has the same antecedents and moderator (albeit divergent) as OCB.
In simple terms, CWB is essentially the opposite end of the spectrum to OCB – undesirable
behaviours and actions in the workplace (Gruys and Sackett, 2003). The impression management perspective takes the view that individuals are motivated by how they are viewed
by others and suggests that their behavioural responses are influenced by what is deemed
socially desirable or acceptable and is not so much influenced by what is instinctive to them
(Rosenfeld, 1997). Therefore, it is plausible to suggest that their engagement with OCB is
influenced by their desire to impress others and gain favourability.
While Phipps et al.’s findings suggest that personality factors such as conscientiousness
and agreeableness are the strongest predictors in the manifestation of OCB and CWB, they
propose that a combination of personality traits appear to work together in influencing OCB
and CWB along with a number of different moderating factors including impression management. This may include contextual factors such as field of occupation, organisational culture
and demographic data.
Recent research has suggested that there may be a ‘dark side’ to these personality traits in
the workplace and equally a ‘bright side’ to those personality traits that are deemed to have a
negative effect on positive work behaviours. For instance, Ferguson et al. (2014) proposed that
this is dependent on changes in context. In order to explore this, the researchers looked at the
changes in context between pre-clinical years (classroom-based education) to clinical years
(practical application of skills) in medical students in the UK. The differences between these
two contexts is the learning that is required where a more methodical approach to learning
needed in the pre-clinical years and a more flexible approach to learning in the more practical
context of the clinical years. Their findings suggest that trait conscientiousness could enhance
learning in the pre-clinical years but have a ‘dark side’ in reducing clinical knowledge acquisition. This effect is believed to occur as conscientiousness enhances performance when the
Chapter 2 Individual differences
51
context requires methodical or ordered thought (as in classroom-based learning) but reduces
it when the context requires flexibility of thought (as in practical- or clinical-based learning).
Similarly, their findings indicate that anxiety has a ‘bright side’ in enhancing clinical skills
acquisition. This is thought to be due to the increased vigilance associated with anxiety which
can aid in clinical skill acquisition. In addition, they found a U-shaped association between
extraversion and pre-clinical learning. This research demonstrates that the relationship
between personality and performance is complex and can vary significantly across different
contexts and the lifespan (Ferguson and Lievens, 2017).
Point of integration
The links between personality and performance may depend upon the particular type of performance
being examined. Performance of core job tasks may be related to elements of personality that are not
as closely linked to organisational citizenship behaviour (and vice versa). Understanding the different
aspects of work performance (see Chapter 4) is important when making predictions about how individual differences may impact on behaviour at work.
Key debate
Does personality change through life?
Is our personality enduring and stable? Much of the research into personality has concluded that it is. However, research
is starting to emerge that challenges this assumption. Rantanen et al. (2007) conducted a meta-analysis into Big Five trait
stability in an adult sample, using structural equation modelling (see Chapter 1). Their findings show that between the ages
of 33 and 42, there was a significant increase in mean levels of Extraversion, Openness to Experience, Agreeableness and
Conscientiousness. Levels of Neuroticism fell between ages 33 and 42. In the same study it was found that Extraversion and
Neuroticism changed less over time in the male sample, compared to the female sample. Such findings may indicate that
job performance could be impacted by changes in self-reported personality over the lifespan, but more research is needed
before firm conclusions can be drawn. A more recent meta-analysis was conducted by Specht et al. (2011) and explored
the changes in personality across almost 15,000 individuals in Germany across all of adulthood. Their findings show that
Emotional Stability, Extraversion, Openness and Agreeableness all peaked between the ages of 40 and 60 before decreasing
afterward while Conscientiousness showed a continuous increase with age. They also found personality predicted several
major life events and were subject to change in reaction to experiencing these events. This finding, in line with Rantanen et al.
(2007), indicates that life events may have an effect on changes in our personality. In addition, Ferguson and Lievens (2017)
illustrated personality traits change across generations, the life span, and in response to environmental contingencies. In
discussing this topic, it might be helpful to consider some life events that you think might influence a change in personality.
Point of integration
The role of personality in personnel selection decision-making is discussed further in Chapter 3. For
the moment it is sufficient to recognise that the Big Five provide a level of description for personality
that dominates the research literature.
Personality measures
Although several studies have shown that the factors measured by many personality measures can
be related to the Big Five, the tests themselves provide data on a variety of personality dimensions.
Table 2.2 gives some information on the better-known personality questionnaires available.
Work Psychology
52
Table 2.2
Personality questionnaires
Instrument
Personality characteristics measured
Eysenck Personality Questionnaire (EPQ;
Eysenck and Eysenck, 1964)
Extraversion, neuroticism (or emotional stability), psychoticism.
The Sixteen PF5 (16PF; Cattell et al., 1970).
Several versions of short/long forms are
available and measure second-order personality factors etc., in addition to the 16 factors
mentioned here
Sixteen personality factors, e.g. submissiveness (mild, humble, easily led, docile, accommodating); self-assurance (placid, serene, secure, complacent); tender-mindedness (sensitive,
clinging, over-protected).
The Occupational Personality Questionnaire
(OPQ32r)
Thirty-two personality dimensions are measured using forced-choice questions in the OPQ
(the OPQ32n). The questions have their origins in descriptions of people’s behaviour in the
workplace. Other versions/models of the questionnaires are available. Illustrative personality
dimensions are: sociability (including the specific personality characteristics outgoing, teambuilding and socially confident), dynamism (vigorous, competitive, achieving, decisive) and
empathy (modest, democratic and caring).
The Hogan Personality Inventory
Its seven primary scales are: adjustment, ambition, sociability, agreeability, prudence, intellectance and scholarship. It was designed specifically for use in occupational settings and hence
questionnaire responses can also be scored on six occupational scales: service orientation,
stress tolerance, reliability, clerical potential, sales potential and managerial potential. The
measure is based on the Big Five model, but also socio-analytic theory. This means that the
questionnaire’s items are designed to measure how people present themselves to others
(rather than how they would describe themselves).
The International Personality Item Pool (IPIP;
Goldberg, 1999). A short version – the mini
IPIP – is also available (see Donnellan et al.,
2006).
The IPIP is a 50-item measure based on the Five Factor Model of personality. It measures each
of the Big Five with ten items each and has been found to be a reliable and valid measure
of these dimensions. A mini 20-item version was developed in 2006, designed for use in
time-critical situations. Research has showed reliability and validity to be comparable to the
parent measure, thus establishing the mini IPIP as a useful measure of the Big Five dimensions.
The revised NEO Personality Inventory (NEO
PI-R, Costa and McCrae, 1992)
The Big Five personality factors, plus facet scores for six subscales within each of the
Big Five domains. This measure is designed for a wide range of purposes (e.g. clinical,
research and occupational uses). Descriptions of each factor and their subfacets can be found
in Table 2.3.
The Myers-Briggs Type Indicator (MBTI)
This measure is based on a very different theory from the Big Five. It uses Jung’s theory of
psychological types, which discusses fundamental differences between people in terms of
dichotomies rather than positions on a scale. Research shows that while the MBTI does not
have a direct measure of Neuroticism, its constituent parts do correlate with the Big Five.
Extroversion–Introversion (EI) correlates with Extraversion but also taps into elements of
openness); Sensing–Intuition (SN) correlates with Openness to experience; Thinking–Feeling
(TF) correlates with Agreeableness; and Judging–Perceiving (JP) correlates with Conscientiousness and Openness to experience (Costa et al., 1991). It is important to remember that
although there are correlations with the Big Five there are some qualitative differences in
what the MBTI measures because it is based on a different underlying theory.
Wave Personality Questionnaires
The Wave questionnaires are driven by the Wave model which measure key characteristics
that underlie successful performance in the workplace. The model follows a hierarchical structure consisting of four clusters, 12 sections, 26 dimensions and 108 facets. The four clusters
are Solving Problems, Influencing People, Adapting Approaches and Delivering Results.
Depending on selection and development needs, there are several different questionnaires
that can be used.
HEXACO
The HEXACO model is based on six factors of personality: Honesty-Humility (H), Emotionality
(E), Extraversion (X), Agreeableness (A), Conscientiousness (C) and Openness to Experience (O). The most widely used measure is the HEXACO Personality Inventory-Revised
(HEXACO-PI-R) available as both a self- or -observer report instrument.
Chapter 2 Individual differences
Table 2.3
NEO facet
Descriptions of the NEO PI-R facets
Description
Conscientiousness
Competence
Sense that one is adept, prudent and sensible
Order
Neat, tidy and well organised; methodical
Dutifulness
Governed by conscience; ethical, fulfils moral obligations
Achievement-striving
High aspirations and works hard to achieve goals; driven to succeed
Self-discipline
Ability to begin and carry out tasks; self-motivating; persistent
Deliberation
Ability to think carefully before acting; cautious and deliberate
Agreeableness
Trust
Belief that others are honest and well intentioned; not sceptical
Straightforwardness
Sincere; unwilling to manipulate through flattery or deception
Altruism
Active concern for others’ welfare; helpful, generous and considerate
Compliance
Cooperative; seeks to inhibit aggression; forgiving; mild-mannered
Modesty
Humble and self-effacing
Tender-mindedness
Sympathy for human side of social policies; concerned for others
Neuroticism
Anxiety
Apprehensive, fearful, prone to worry, tense, jittery
Hostility
Quick to anger; easily frustrated and irritated by others; bitter
Depression
Depressive affect, guilt, sadness, hopelessness; prone to dejection
Self-consciousness
Shame and embarrassment, sensitive to ridicule
Impulsiveness
Inability to control cravings or urges; susceptible to temptation
Vulnerability
Susceptibility to experience stress; easily panicked
53
54
Work Psychology
Table 2.3
Descriptions of the NEO PI-R facets (continued )
NEO facet
Description
Openness to experience
Fantasy
Active imagination; tendency toward daydreaming; lost in thought
Aesthetics
Appreciation for art and beauty, moved by poetry and music
Feelings
Receptive to inner feelings and emotions; empathetic
Actions
Willingness to try different activities; preference for variety to routine
Ideas
Intellectual curiosity; willingness to consider new ideas
Values
Readiness to re-examine values; liberal; anti-tradition and anti-authority
Extraversion
Warmth
Affectionate and friendly; informal and unreserved around others
Gregariousness
Sociable; preference for company of others; ‘the more the merrier’
Assertiveness
Dominant, forceful and socially able; takes charge and assumes leadership
Activity
Prefers fast-paced life; high energy level; vigorous
Excitement-seeking
Craves excitement and stimulation; sensation-seeking
Positive emotions
Experiences joy; laughs easily; cheerful and optimistic; high-spirited
Source: Judge et al. (2013).
As with all kinds of psychological measures, personality questionnaires need to satisfy
various well-established psychometric criteria before they can be considered to be acceptable
measuring instruments. These criteria are concerned with assessing the extent to which the
test measures what it intends to measure (the issue of validity) and the precision or consistency of measurement that the test achieves (reliability).
With the ever-increasing prevalence of technology in society, it should come as no surprise that recent research has begun to explore the ways in which we can utilise technology in
work psychology. For instance, Park et al. (2014) investigated the potential for the language
that we use on social media to be indicative of our personality. They used open-vocabulary
analysis of written language from over 65,000 Facebook users alongside a self-report questionnaire of the Big Five to build a model of personality based on language. They then used
this model to predict the personality factors in a further set of Facebook users looking at
convergence of the personality profiles with self-report profiles of personality, discriminant
validity between predictions of different traits, agreement with personality profiles, correlations with external factors (e.g. political attitudes) and test-retest reliability. Their findings
Chapter 2 Individual differences
55
were supportive in suggesting that the language we use on social media can contribute to
reliable and valid profile of our personalities. They suggest that using an approach such as this
can complement the more traditional measures of personality and provide a measurement
methodology which can access large groups of people both quickly and cost-effectively. Further studies have shown that machine-learning algorithms can learn to translate Facebook
and Twitter activity into valid indicators of both personality and intelligence (e.g. Kosinski
et al., 2016). There has also been an increase in the exploration of games-based assessments
for measuring personality. However, there are some concerns about the accuracy of these
assessments for being indicative of personality (Leutner and Chamorro-Premuzic, 2018).
As a light-hearted illustration, Table 2.4 provides descriptions of the 16 Personality Factor
(16PF) scales together with examples of historical or literary figures who are supposed to
exemplify the personal qualities described. You may want to think of some other famous (or
infamous) individuals who display such qualities!
Stop to consider
In reading about the Big Five you may have reached the conclusion that some personalities are more
‘desirable’ than others. However, this is a dangerous assumption as so much depends upon the social
context within which the person is behaving. Most personality measures used in occupational settings
are measures of ‘normal’ personality: questionnaire responses not an absolute indicator of a problem
(e.g. people may score high on the Big Five Neuroticism scales without having a psychiatric disorder).
Costa and McCrae (2006) discuss the advantages and disadvantages of various scores on the Big
Five. They say about Conscientiousness: ‘On the positive side, high C is associated with academic and
occupational achievement; on the negative side, it may lead to annoying fastidiousness, compulsive
neatness or workaholic behaviour’ (2006: 17).
Have another look at the Big Five. Try to think of jobs in which low E, low O, high N, low A (etc.)
might be advantageous (hint: there are some). What personality might be well suited to being a librarian, an elite track and field athlete, a lawyer, a CEO, a counsellor/therapist (insert some more of your
own examples here)? Think about the lower-level facets as well as the broad dimensions. This will help
you to think about whether certain personality profiles really are inherently advantageous/disadvantageous. As you will see, much depends upon the demands of the situation.
Table 2.4
The personality traits assessed by the 16PF with examples of famous individuals exemplifying
the traits
Factor
Famous individuals
Trait descriptions
High
Low
Low
High
A
Outgoing
Warm-hearted
Reserved Detached
Falstaff
Greta Garbo
C
Unemotional Calm
Emotional Changeable
George Washington
Hamlet
E
Assertive Dominant
Humble Cooperative
Genghis Khan
Jesus
Work Psychology
56
Table 2.4
The personality traits assessed by the 16PF with examples of famous individuals exemplifying
the traits (continued)
Factor
Famous individuals
Trait descriptions
High
Low
High
Low
F
Cheerful
Lively
Sober
Taciturn
Groucho Marx
Clint Eastwood
G
Conscientious
Persistent
Expedient
Undisciplined
Mother Teresa
Casanova
H
Venturesome
Socially bold
Shy
Retiring
Columbus
Sylvia Plath
I
Tough-minded
Self-reliant
Tender-minded
Sensitive
James Bond
Robert Burns
L
Suspicious Sceptical
Trusting Accepting
De Gaulle
Pollyanna
M
Imaginative Bohemian
Practical Conventional
Van Gogh
Henry Ford
N
Shrewd
Discreet
Forthright
Straightforward
Machiavelli
Joan of Arc
O
Guilt prone
Worrying
Resilient
Self-assured
Dostoyevsky
Stalin
Q1
Radical
Experimental
Conservative
Traditional
Karl Marx
Queen Victoria
Q2
Self-sufficient
Resourceful
Group-dependent
Affiliative
Copernicus
Marilyn Monroe
Q3
Controlled Compulsive
Undisciplined Lax
Margaret Thatcher
Mick Jagger
Q4
Tense
Driven
Relaxed
Tranquil
Macbeth
Buddha
Note: Dimension B (intelligence) is omitted.
Sources: From Conn and Rieke (1994); Matthews and Deary (1998).
Creativity and innovation
Over the past two decades there has been an explosion of interest in innovation as organisations have recognised that creating new processes, products and procedures is vital for productivity and growth in all sectors. With more dispersed and virtual working, role innovation is
essential, since clearly defined job descriptions no longer exist for many jobs. The research literature on creativity and innovation is immense. The case study exercise below (Exercise 2.1)
illustrates the ‘business case’ for enhancing innovative working in organisations.
Chapter 2 Individual differences
Creativity pays, and not just financially
Exercise 2.1
A paper, published by Cornell University in 2013, highlights the important organisational and individual benefits of innovation. It reports the story of Nottingham University Hospitals Trust (NUH), a
public healthcare provider in England, set up in 2006.
The public healthcare sector in the UK has faced pressures to find new ways of working to decrease
costs. Based on evidence that innovation can be driven by a bottom-up employee process, efforts
were made to create changes with the goal being to become England’s best acute teaching trust by
2016. This was an ambitious agenda and would involve:
■
providing services that are good value for money
■
being the best in: clinical outcomes, patient experiences, staff satisfaction, teaching, training,
and research.
As a result, the National Health Service (NHS) Institute for Innovation and Improvement designed
and implemented the ‘Productive Ward’ programme, with the purpose of providing tools ‘specifically
developed to engage frontline staff in the initiation and implementation of change’. The programme
was initially rolled out on a voluntary basis over three years; the results now speak for themselves.
The Trust is one of the largest and busiest trusts in England, with 13,000 staff providing services to
over 2.5 million residents. The innovation programme has also resulted in several other individual
and organisational improvements, including:
■
new opportunities for staff involvement – at least 50 per cent of staff are now involved in
implementing change;
■
more opportunities to offer feedback – monthly meetings include the discussion of ideas for
change, which managers have said is good for encouraging them to think about and challenge
old ways of thinking, leading to the implementation of better working strategies;
■
staff development – delegating implementation responsibilities resulting in more active engagement, increased confidence and enhanced CVs;
■
improvements in teamworking;
■
more efficient patient care and working practices such as the ordering of equipment;
■
improved experience and safety of patients;
■
an overall improvement in performance at the hospital;
■
organisational culture change – the organisation now listens to employees, breaks down hierarchy and allows staff to be change agents.
(Based on Foley and Cox, 2013)
Suggested exercise
Evidently, there is a business case for innovation. The above case study has illustrated how implementing
an innovation programme has helped this organisation to develop an effective and efficient workforce,
enabling them to overcome challenges such as spending cuts and new health threats. Imagine you were
the chief executive of NUH. How would you ensure that creativity and innovation are valued by employees
and managers? How would you encourage creative thinking in your employees? Do you think there are
some individuals who are more predisposed to innovate? And how might you attract, select and hire these
creative thinkers? What challenges might there be in managing creative people?
57
58
Work Psychology
The concepts of creativity and innovation are often confused and used interchangeably
in the literature. The main distinction can be seen as novelty. Creativity is concerned with
generating new and entirely original ideas. Innovation is a broader concept because it also
encompasses the application and implementation of new ideas to produce something new
and useful (in the context of groups, organisations or societies). Innovation is often referred
to as a process, because implementing new ideas necessarily involves influencing others
(whereas creativity could be achieved in isolation).
In 2014, Anderson, Potocnik and Zhou proposed a new integrative definition of creativity
and innovation:
Creativity and innovation at work are the process, outcomes, and products of attempts to develop
and introduce new and improved ways of doing things. The creativity stage of this process refers
to idea generation, and innovation to the subsequent stage of implementing ideas toward better
procedures, practices, or products. Creativity and innovation can occur at the level of the individual, work team, organization, or at more than one of these levels combined, but will invariable
result in identifiable benefits at one or more of these levels-of-analysis.
(Anderson et al., 2014: 4).
Authors commonly agree that innovation in organisations is a complex, iterative process.
Numerous approaches and process models have been proposed, but two main stages are
common to all. First, there is a suggestion phase and second, an implementation phase.
Innovation is not a linear process: it involves several cycles of activities such as initiation,
reappraisal, adaptation, implementation and, finally, stabilisation. Innovation is a result of
both person-level resources and various environmental factors such as feedback, leadership
style, resource availability and organisational climate. In terms of the measurement of innovative potential, there are various measures available that claim to assess the propensity to
innovate, such as Patterson’s (2002) Innovation Potential Indicator.
Innovation involves multiple components at the individual level. In her reviews on individual differences, Patterson (2002, 2004) suggests that the research literature can be classified
into studies of the links between innovation and intelligence, knowledge, motivation and
personality. Later research indicated a role for additional factors including emotional intelligence, mood states and values.
Key debate
Is creativity equivalent to genius?
Is creativity determined by or even equivalent to high intelligence? The best-known researcher in this field is Guilford. In his
theory of the Structure of Intellect (SI), first published in the 1950s (see Guilford, 1996), he claimed that creative thinking
was a mental ability, involving divergent production (i.e. thinking that goes off in different directions). Many researchers
followed Guilford’s work by producing evidence that ideational fluency (i.e. the quantity of new ideas) underlies divergent thinking test scores. This is akin to Edward De Bono’s ‘lateral’ thinking concept. However, review studies have cast
doubt on this conclusion: divergent thinking scores often fail to correlate significantly with various indices of innovation.
Some authors doubt whether divergent thinking tests are adequately measuring the abilities actually involved in creative
thinking as it is the quantity, not quality, of ideas that are considered. Since innovation can only be useful if it adds value
to an organisation, it is idea quality that is of the highest importance (Wycoff, 2004).
Other authors suggest that genius, as the most obvious manifestation of high intelligence, is closely tied to the
propensity for innovation. However, despite the large amount of research carried out, there has been a substantial lack
of evidence to support a direct relationship between innovation and intelligence. More recent studies conclude that
intelligence and innovation potential are moderately related, but once IQ scores go over 115 the relationship is near
zero. This finding has been described as ‘threshold theory’: once intelligence reaches a certain point, its relationship to
Chapter 2 Individual differences
innovation breaks down. It seems that instead of being twin or even sibling constructs, intelligence and innovation
potential may be more like ‘cousins’.
Finke et al. (1992) suggest that in order to understand the role of cognitive abilities in idea generation, we must
draw upon cognitive psychology. These researchers have used experimentally based observations of the processes
that underlie generative (creative) tasks. Their work follows a framework called the ‘geneplore model’. The model
proposes that many creative activities can be described in terms of:
• an initial generation of ideas or solutions; followed by
• an extensive exploration of those ideas.
They suggest that individual differences in idea generation occur due to variations in the use and application
of these two generative processes, together with the sophistication of an individual’s memory and knowledge in
the domain they are working in. In summary, findings showed that the capacity for creative cognition is normally
distributed among the general population and therefore, highly creative people do not have minds that operate
in any fundamentally different way from other individuals.
Exercise 2.2
Test your creative thinking
Task 1: The nine dots puzzle
Look at the nine dots below. Draw no more than four straight lines (without lifting the pencil from
the paper) which will cross through all nine dots.
Task 2: One-minute idea generator
Your task is to ask yourself the following question: How many uses are there for a teaspoon?
Go for quantity not quality of ideas. Write down every idea. Do not judge or criticise! Stay relaxed,
playful, even silly. Adapt your point of view. For example, look at the teaspoon as if you were an
insect or as a designer or as an electrician or in the middle of the desert. Ask yourself ‘What if . . . ?’
questions. What if the teaspoon was flattened out and one end was sharpened to a point? What if
several of them were linked together somehow? Time yourself for one minute.
Stop to consider
After reading the section on creativity and innovation and trying the two tasks above, ask yourself the
following questions:
1
2
3
What individual-level factors do you feel influenced your performance on the above tasks?
What group-level factors may have influenced your performance if you had been working as part
of a team in an organisation to complete the tasks?
Do you think these are robust tests to assess an individual’s creativity? If not, why?
59
60
Work Psychology
A common criticism of intelligence research is that traditional measures of intelligence
do not fully explain ‘real-world’ intelligence. Part of the issue here has been that intelligence (similar to innovation) is often viewed as a unitary concept. Previous theories of
intelligence have tended to over-emphasise cognitive abilities and underestimate the role
of knowledge-based intelligence (Silvia, 2008). However, knowledge is a key variable in both
generative thinking and innovation. An essential condition for innovation is being immersed
in domain-specific knowledge, as one must develop an accurate sense of the domain before
one can hope to change it for the better. For example, it might be difficult to design the
world’s best car if you did not know what a car was! However, on the other hand, the literature
highlights that too much expertise in one area can also be a block to innovation within that
domain. This is illustrated by research by Simonton (2004), who studied the lives of over 300
eminent people to explore lifespan development of innovation. They found that both a lack
of, and an excess of, familiarity within a subject domain, can be detrimental to innovation.
It seems that knowing too much can stifle innovation.
Most studies on the motivation to innovate rely on the well-established typology of human
motivation which differentiates between intrinsic and extrinsic motivation (see Chapter 6).
Motivation is said to be intrinsic if an activity is valued and performed for its own sake and
the individual has a genuine passion and interest in innovation itself. By contrast, a person
performing a task with the aim of obtaining a reward separable from the task (e.g. pay) is
said to be extrinsically motivated. Research has generally found that extrinsic motivators are
related to innovation only under certain circumstances. Intrinsic motivators can therefore be
said to have a stronger and more consistent relationship with innovation (Hammond et al.,
2011). It should not be ignored, however, that constructive evaluation (i.e. feedback that is
informative, supportive and recognises accomplishment) can enhance innovation. Individuals who receive positive feedback given in an informational style, and who work in a high
task autonomy environment, generate the most innovative solutions (see Zhou and Shalley,
2008). Research has also suggested that intrinsic and extrinsic motivators may be important
at different stages of the innovation process. However, given the complexities, Zhou and
Hoever (2014) suggest a perspective that emphasises the actor-context interactive effects is
needed to fully understand creative behaviour at work.
From several decades of research on the association between innovation and personality,
a consistent set of characteristics has emerged. Innovative people tend to be imaginative,
inquisitive, have high energy, a high desire for autonomy, social rule independence and high
self-confidence. Of the Big Five, Openness to experience is perhaps the most important personality dimension to predict the propensity for innovation. Research also shows that low
conscientiousness is associated with innovation. Defined by terms such as fastidious, ordered,
neat and methodical, individuals high on Conscientiousness are more resistant to changes
at work, and are more likely to comply with current organisational norms. The relationships
between the five personality domains and innovation are more complex at the facet level,
and although this requires further research, it seems that assessing personality at the facet
level could prove more useful than assessing at the domain level in terms of predicting innovation. Similarly, a recent review of the literature found that personality characteristics such
as creative and proactive personality and core self-evaluations have the highest propensity
to predict innovative job performance. However, these are not characteristics that are usually
assessed during personnel selection (Potocnik et al., 2015).
Interest has shifted to the role of EI in innovation at the organisational, leadership, team
and individual levels. For example, in a review of the role of emotions in transformational
leadership behaviour, Ashkanasy and Tse (2000) suggested that EI involves the ability to
use emotions in a way that allows flexible planning and creative thinking. Similarly, leaders’
levels of EI are likely to accentuate the employees’ inclination to engage in the innovation
process. From a team perspective, Barczak et al. (2010) found that emotionally intelligent
teams create both cognitive and affective team trust, which in turn builds a collaborative
culture and facilitates higher levels of creativity and innovation. For individual employees, it
Chapter 2 Individual differences
61
has been found that those who show high levels of EI also display higher levels of readiness to
create and innovate (Suliman and Al-Shaikh, 2007), and are likely to benefit more from both
positive and negative creativity-related feedback. Furthermore, it is thought that EI may be
beneficial when innovative employees have to persuade others to support them during the
implementation phase of innovation. Specifically, being able to perceive others’ emotions
and to regulate them appears to be important in the persuasion process. More recent research
has attempted to understand the relationship between emotional intelligence and creativity
further by exploring the moderating factors. For instance, Jafri et al. (2016) found emotional
intelligence resulted in highest levels of employee creativity when an employee displayed a
more proactive personality and when the organisational climate was favourable for creativity.
Point of integration
Leadership behaviour may have a significant impact on employees’ propensity to innovate (see Chapter 10). It is a good example of the interactions that occur between individual differences and the
situation. In this case leader behaviour can moderate (i.e. turn up or down) the ‘volume’ of employees’
potential to innovate.
Exercise 2.3
Assessing the propensity to innovate at work
A long-standing problem for HR managers has been how to select and develop innovative individuals
effectively. Getting the right raw material in the first place is a critical issue. The propensity to innovate is often listed as a key competency in person specifications, but there has been little available
material to assess it in a reliable way. Traditional measures of creative thinking have been of the kinds
that ask, ‘How many uses for a paper clip can you think of?’ in, say, three minutes. Such measures
may indicate some level of conceptual thinking but are of limited use in identifying individuals who
are innovative in the workplace.
Previous literature has been confused as to whether the propensity to innovate is predicted by an
individual’s level of intelligence or whether it is more concerned with personality. Early research
suggested that innovative potential was an aspect of general intelligence. However, many authors
have demonstrated that aspects of intelligence may be a necessary, but not a sufficient, condition
for innovation to occur. Later research focused on the propensity to innovate as an aspect of personality. Researchers have noted that motivation and job-specific knowledge are key components in
predicting the propensity to innovate at work. While there has been a vast amount of work examining
these characteristics, this has lacked an integrative model with which to understand how all these
aspects interrelate – although more recent theories are beginning to integrate the literature (see
Patterson et al., 2009; Sears and Baba, 2011).
The technology and consulting company IBM spoke out about the problem of assessing innovation
potential, agreeing that identifying innovative individuals in a recruitment process is one of the most
difficult responsibilities of HR professionals. Nevertheless, based on their expertise in sophisticated
HR tools and practices, they propose several strategies for assessing and recruiting innovative individuals. First and foremost, the use of behavioural assessments that can identify desirable individual
differences is key. Trait-based measures that directly assess the propensity to innovate, such as the
Innovation Potential Indicator (IPI; Patterson, 1999) can be invaluable for assessing one’s capability
of implementing ideas in specific contexts. Other behavioural assessments include: measures of
62
Work Psychology
personality variables associated with innovation such as novelty, creativity and initiative; intelligence;
motivation; biographical data; and situational judgement tests (SJTs). According to IBM, ability tests
that focus on assessing things like word fluency (see Carroll’s three stratum theory, 1993) are more
appropriate indicators of innovation potential than measures of general intelligence. Other methods
like SJTs can assess one’s innovative approach to various situations, and behaviour-based interviews
can be an indicator of individuals’ previous innovating experiences.
Suggested exercises
1
Create a list of the person-level variables that might influence the propensity to innovate at work.
How do you think these variables interrelate?
2
Suggest how these person-level variables might be best measured. What further research could be
conducted to learn more about innovation at work?
3
Do you think innovation potential is more useful in a selection context or for employee development?
Why is that?
Socio-cognitive approaches to individual differences
The beginnings of the socio-cognitive approach were evident in the debate about whether
it is individual differences or the situations that people find themselves in that determine
people’s behaviour. To most of us it is clear from everyday experience that our behaviour
is not completely at the mercy of situational influences. There is some cross-situational
consistency in how we behave from one setting to another, particularly in terms of key
features of our psychological make-up, such as extroversion, agreeableness and anxiety.
On the other hand, most people will behave quite differently at a lively party and at a very
important formal business event. The relative influence of person and situation variables
has been a topic of considerable controversy. Some people have argued very strongly for the
predominance of situational influences, suggesting that stable individual differences in psychological make-up have a relatively small role to play (e.g. Mischel, 1968). Despite these
historical differences of opinion, modern psychology allows for the influence of both person
and situation variables. Eysenck captured the essential futility of trying to identify a single
cause by writing:
Altogether I feel that the debate is an unreal one. You cannot contrast persons and situations in
any meaningful sense . . . No physicist would put such a silly question as: which is more important
in melting a substance – the situation (heat of the flame) or the nature of the substance?
(Quoted by Pervin, 1980: 271)
Situations influence our behaviour because we think about how we should respond to
them. We could argue that work psychologists are equally guilty of ignoring the thought processes that underlie behaviour. For example, most selection research has sought to determine
whether various selection methods predict job performance, rather than why these methods
predict. For example, we know that personality traits predict work performance (see Chapter 4), but we know much less about how personality traits influence behaviour. Moreover, it
is highly unlikely that there is a direct relationship between personality traits and behaviour
(Skarlicki et al., 1999). After all, individuals will have their own way of appraising the situation
that they find themselves in and deciding how they should best respond.
Chapter 2 Individual differences
63
Point of integration
Cognitive appraisal of working conditions is an important element of theories of work-related stress
(see Chapter 8) and features in many theories of career choice (Chapter 11). These theories help us to
explain why different individuals have different responses to the same situation.
Hodgkinson (2003) found that senior managers, particularly CEOs and senior executives, all have their own way of explaining key organisational events (such as company
performance) and this way of thinking influences strategic decisions (Sparrow and Hodgkinson, 2002). This is just one example of how individual differences in cognitive style
may play a significant role in job performance. A number of cognitive-based personality
characteristics have been developed and investigated over the years, including proactive
personality, personal initiative, self-efficacy, attributional style, locus of control and
psychological empowerment (e.g. Bandura, 1982; Frese et al., 1997; Silvester et al., 2003;
Li et al., 2015). Interestingly, all of these appear to relate to motivation – a construct that
has been notoriously difficult to explain in terms of personality traits. Kanfer and Ackerman’s (2002) work on motivational traits acknowledges the importance of cognition in
motivation. This raises the intriguing possibility that the impact of personality traits on
work performance will be mediated by individual differences in cognitive style. Evidence
for this can be found in work examining the relationship between attributional style and
the performance of sales personnel. For example, Corr and Gray (1996) found that male
insurance sales agents who attributed positive outcomes (such as making a sale) to internal, stable and global causes (such as their own personality) were more successful than
individuals who externalised the cause to more unstable causes (e.g. luck, or being in a
good mood that day).
Silvester et al. (2003) also found that individuals who perceived themselves to have more
control over sales outcomes (both successful and unsuccessful) were rated as more successful
by their managers. These findings indicate that even in the case of failure, individuals who
attribute the outcome to more internal controllable causes (e.g. using the wrong sales strategy)
will be more likely to maintain a higher level of motivation and effort than sales staff who
typically put the failure down to external uncontrollable causes (e.g. the customer had seen
a cheaper version of the product elsewhere).
Li et al. (2015) explored the relationship between psychological empowerment and
employee performance while looking at the moderating effect of locus of control and the
mediating effect of intrinsic work motivation. Their results indicate that psychological
empowerment is related to job performance in several different areas; task performance, contextual performance and innovation performance. The relationship between psychological
empowerment and contextual and innovation performance was moderated by an individual’s locus of control. In addition, intrinsic motivation was found to partially explain the relationship between psychological empowerment and all three work performance indicators.
Clearly, there is tremendous potential for researchers to explore how individual differences
in sense-making and cognition impact upon work performance.
Key learning point
Stable individual differences in cognitive style represent another potential source of variance in job
behaviour.
64
Work Psychology
Summary
Individual differences in personality and cognitive ability have been the subject of much
psychological research. Research has uncovered stable, underlying structures for both cognitive ability (a general factor and specific abilities) and personality (five major factors).
Measures have been developed for cognitive ability and personality. To be of value in work
psychology, such measures need to be both reliable and valid. At the moment there is a
great deal of interest in exploring the concept of emotional intelligence and whether it
might offer something new in terms of our understanding of behaviour at work. There
has been an increase in research into possible biological, neurological and physiological
correlates of intelligence (e.g. Beaujean, 2005; McDaniel, 2005).
Looking to the future, the closing case study illustrates the growing use of artificial
intelligence in the workplace, which has important implications for the role of human
intelligence in effective workplace behaviour. This case study also highlights there is
increasing recognition that understanding individual differences means that we also need
to understand how factors surrounding human-machine interaction will impact on work
behaviour in future.
Closing case study
Graduates with tech and finance skills in high demand: the
workforce will have to adapt as companies use artificial
intelligence for more tasks
Seb Murray, FT Case Study 2019
By 2030, some 400,000 full-time jobs in capital markets – including asset management, investment
banking and brokerage – will be lost as technology and artificial intelligence (AI) supplant humans,
predicts Opimas, a research consultancy. ‘There will be blood on Wall Street if the workforce fails to
adapt’ says Axel Pierron, co-founder and managing director of Opimas. If a role involves doing calculations, such as equity trading, he adds, a machine will be able to do it faster and more accurately.
But there are opportunities for Masters in Finance (MiF) graduates. The blend of technological and
financial knowledge is a sweet spot, as is a growing Asian economy and new opportunities in areas
such as risk management. Mr Pierron predicts 66,000 new technology and data science jobs will
be created by 2030 in roles such as computer programming. Mr Pierron says technology is freeing
financiers of routine tasks and improving their competitiveness. Hedge funds and asset managers
are mining new information sources to beat the markets. For example, changes in the number of
jobs listed on a company’s website can hint at financial performance before results are officially
released. Heidi Pickett, assistant dean of the MiF programme at MIT Sloan School of Management
in Massachusetts, US says ‘Computer programming skills are becoming a must-have. ‘Don’t bother
putting Excel or PowerPoint on your résumé. Financial institutions are looking for R, Python or
another programming language.’
Yet soft skills are increasingly essential for the financial workforce, says Gary Baker, managing director
for Europe, the Middle East and Africa at the CFA Institute, an international association for investment
professionals. In a report by the institute, industry leaders ranked ‘T-shaped skills’ – which include subject expertise, the ability to understand diverse perspectives and being comfortable with technology
Chapter 2 Individual differences
65
– as the most important aptitudes for the future, above even technical abilities. ‘Some think this is
wishy-washy softer stuff, but that’s very difficult for any machine to pick up,’ Mr Baker says. Humans
will always be retained in financial institutions, he adds, because people still need to interact with
clients and make investment decisions. In Europe, the regulation that followed the financial crisis
has made students with technical and legal knowledge a prized asset, says Olivier Bossard, executive
director of the MSc finance programme at HEC Paris business school. ‘There is a growing demand for
our students who can calculate and communicate risks,’ he says, including at banks and consulting
firms. Finance is a fast-changing industry, and knowledge and skills acquired today might not guarantee a job in future, says Alex Stremme, assistant dean of MSc finance programmes at the UK’s Warwick
Business School. Financial training can develop resilience to cope with that change, he adds. So, while
the future is uncertain, business schools can at least help students come to terms with the ambiguity.
Suggested exercises
1 Using your learning from this chapter, how would you describe the differences between artificial
intelligence and human intelligence in this setting?
2 Using psychological theory, how would you describe the so-called ‘soft skills’ required in the finance
sector in future?
3 How will technology impact other sectors in future and what are the implications for research in individual differences?
Suggested discussion exercises
1 Discuss whether different types of individual differences (e.g. cognitive ability, intelligence, personality and creativity) can predict behaviour and performance at work.
■ Examine the criticisms that these theories do not generate testable predictions
about human behaviour
■ Consider how certain factors can be considered to have both a ‘bright side’ and a
‘dark side’
■ Consider how certain factors (e.g. emotional intelligence) may be considered a predictor of success in different contexts
■ Reflect on how other factors, such as social resources (e.g. teamworking), might also
be important in predicting innovative working
■ Examine how different aspects of work performance (e.g. organisational citizenship
behaviours) are important when making these predictions
2 Discuss how to best measure individual differences for the purposes of workplace selection, policy and interventions.
■ Research some of the most popular tools and consider their properties; including
reliability, validity and fairness
■ Consider the need to examine the constructs that each tool purports to measure
■ Consider how this might fit with the competencies included in HR person
specifications
■ Examine how measuring personality at the facet level could prove more useful than
at the domain level for predicting certain traits (e.g. innovation)
■ Discuss the risk of using maximal performance paradigms alone in drawing false
conclusions
■ Examine how the results could be used to better understand the workforce and make
accommodations
66
Work Psychology
3 Discuss what evidence there is that emotional intelligence (EI) is distinct from traditional conceptualisations of personality and intelligence.
■ Consider how researchers (e.g. Goleman) often aggregate these well-researched concepts in their definitions
■ Reflect on how tools which purport to measure EI (e.g. the Emotional Quotient
Inventory) may not capture any unique constructs not already found in established
measures of personality
■ Research the validity and reliability of tools (e.g. Multifactor Emotional Intelligence
Scale) which attempt to measure EI as a distinct concept
■ Examine how EI can predict aspects of job performance that cannot be predicted by
cognitive ability measures alone
4 Do you agree that leaders can moderate (i.e. turn up or down) the ‘volume’ of employees’ personalities?
■ Examine the interactions that occur between individual differences and the situation in hand
■ Explain how other’s perceptions of one’s personality may also impact work behaviours through persuasion
■ Consider the importance of emotional intelligence, the ability to perceive other’s
emotions and regulate them, in doing so
■ Reflect on how displays of personality may depend also on wider organisational factors (e.g. organisational climate)
5 Discuss whether you agree with Seligman and Duckworth that an individual’s IQ
accounts for ‘only a third’ of any difference in academic performance when compared
with peers.
■ Reflect on why IQ may not fully explain variation in academic performance
■ Consider whether qualities such as perseverance, self-discipline, hard work, creativity and luck might instead influence academic performance
■ Consider whether these factors are a result of nature (e.g. genetic inheritance, biology and hormones) or nurture (e.g. parenting, education and experiences)
Relevant websites
A very useful site for learning about psychometric testing from the points of view of both the
tester and the person being tested is provided by the British Psychological Society at http://
ptc.bps.org.uk/
If you are interested in intelligence testing, Indiana University, USA, hosts a site with material you may find helpful. It is not specifically about the workplace, but much here is relevant
to the workplace. Find it at http://www.intelltheory.com/index.shtml
The MBTI is a widely used personality measure in work psychology. A good source of information about the measure can be found at http://www.myersbriggs.org/my-mbti-personality-type/mbti-basics/
The International Personality Item Pool at http://ipip.ori.org/ contains lots of questionnaire
items designed to measure the Big Five personality domains (including the facets).
A very useful collection of basic information and materials for the HEXACO Personality Inventory-Revised which measures the six major dimensions of personality can be found at
http://hexaco.org/
Chapter 2 Individual differences
67
Suggested further reading
Full details for all references are given in the list at the end of this book.
1 Sternberg and Kauffman’s Handbook of Intelligence (Cambridge Handbooks in Psychology)
(Cambridge University Press, 2011) provides a comprehensive overview of theory and
research in the field of human intelligence covering a full range of topics of interest.
2 Ziegler-Hill and Shackleford’s book Handbook of Personality and Individual Differences
(Sage, 2018) provides an excellent overview of the research literature in personality and
individual differences, focusing on various contexts and applications including psychology in the workplace.
3 Zeidner’s book called What We Know About Emotional Intelligence: How it Effects Learning,
Work, Relationships and our Mental Health (MIT Press, 2012) provides an in-depth analysis of the evidence on emotional intelligence.
4 Ashton’s book Individual Differences and Personality (Academic Press, 2013) provides a
comprehensive overview of research in personality; exploring the main approaches
to personality, basic principles in its measurement, personality traits and dimensions
of personality, along with examining important individual differences such as mental
ability, vocational interests, and religious and political attitudes.
CHAPTER 3
Selection: Analysing
jobs, competencies and
selection methods
LEARNING OBJECTIVES
After studying this chapter, you should be able to:
1 outline the selection design and validation process;
2 understand approaches to job analysis and competency modelling, and describe how the
outputs are used to design assessment criteria and methods in selection;
3 in the context of selection, define reliability and five different types of validity: (i) faith,
(ii) face, (iii) content, (iv) construct, (v) criterion-related validity;
4 outline the criterion problem and some practical difficulties associated with conducting
validation studies in organisations;
5 understand several selection methods and the strengths and limitations of each;
6 understand candidate reactions to selection practices;
7 explain how some selection techniques could lead to bias and
unfairness to some subgroups (e.g. gender, ethnicity);
8 explain some of the advantages and challenges associated with
the increasing use of technology and social media in selection.
Chapter 3 Selection: Analysing jobs, competencies and selection methods
Opening case study
69
UK recruiters use technology to offer
‘pre-emptive’ jobs
UK recruiters are developing technology,
including artificial intelligence (AI) that
allows them to offer people jobs before they
even start to look, mindful that if they do not
keep innovating, their own business models
face redundancy. Some of the biggest recruiters in the world have set up partnerships with
business network LinkedIn that allows it to
access some of the company’s data, such as
when a user adds a skill or profile picture.
With this sort of information, AI can identify
someone who is about to begin a formal job
hunt, allowing the company to pre-emptively
contact them with relevant roles. Alistair
Cox, Hays chief executive, stated: ‘Data and
artificial intelligence can give you some
pretty accurate insights as to whether an
individual would be receptive to a job that
you’ve got on your books.’ ‘[It’s part of] a race
to get very strong candidates in front of your
client before the competition.’ Hays is just
one of the large global recruiters, several of
which are headquartered in the UK, investing in technology in an effort to match jobseekers and roles more swiftly and efficiently.
‘What we want is a consultant not having to
trawl through hundreds of candidates to
identify the one that’s right for the job,’ said
Steve Ingham, PageGroup’s chief executive.
Technology is already replacing some of the
more mundane tasks that head-hunters typically undertake, such as screening CVs and
manually sifting through candidates to draw
up a shortlist. Other developments include
recruitment chatbots, a trend towards video
CVs and using AI to profile job applicants.
But there is also the risk that the technology
recruiters are adopting could cannibalise
their own business, by automating their role
as intermediaries and the swaths of jobs they
would typically find candidates for. ‘Technology is having a disruptive impact on the
recruitment sector – and having an immediate effect today,’ said Rahim Karim, support
services analyst at Liberum. But recruiters
also argue that the art of job matching will
never be fully replaced by science. ‘What
will happen is recruiters will actually end up
being much more like business consultants
rather than doing desktop research,’ said
Albert Ellis, chief executive of Harvey Nash,
a recruiter that specialises in technology.
Murrary, H. (2018) UK recruiters use technology to offer ‘pre-emptive’ jobs, April 6 © Financial Times
Limited 2019. All rights reserved.
Introduction
The research and practice of work psychology have important contributions to make in many
areas of organisational life, but selection constitutes an area where arguably the biggest and
most consistent contribution has been made. Given the centrality of psychological measurement to psychological science in general, psychologists are recognised as experts in the design
and validation of high quality selection and recruitment processes.
Our opening case study shows how technology (AI and social media) is now rapidly
transforming selection and recruitment methods and practices. In this chapter we begin by
reflecting on the issues relating to these changes and how this impacts the design of new
selection tools and processes. Traditionally, selection has involved the matching of people to
the requirements of jobs. Now the field has progressed to also explore attraction and recruitment issues. In order to define the attributes required in any job role, an important first step
is to conduct a job analysis: in this chapter we discuss the various techniques and methods
that work psychologists use in practice. We then concentrate on topics concerned with the
development of procedures for predicting future job aptitudes; specifically, validation studies
and associated issues regarding the practicalities of conducting research in organisations.
70
Work Psychology
Next, we describe the selection methods that are used frequently in organisations and examine their relative accuracy at determining who is ‘the best person for the job’. After examining
the evidence concerning the validity of each of the main methods, we explore the extent
to which the various methods are used and how they are viewed by applicants. There is a
growing need to understand issues regarding fairness and how adverse reactions could lead
to legal challenges. Throughout this chapter, we reflect on the role of technology in selection
and discuss some of the advantages and challenges technology may bring to the recruitment
process in future.
Stop to consider
Imagine you are working in Human Resources (HR) of a large, global corporation. On top of your typical
duties, you also have to select new recruits for the graduate scheme. Thousands of applications are
pouring in from candidates who all have similar experience, skills, qualifications, and all are keen to get
the job. What will the selection process look like? How will you find the best candidates amongst the
sea of applications? How long will the process take? How will you make sure the best candidates select
your grad scheme over a competitor’s?
Think about jobs you have applied for and the process you went through, all the way from finding
the job advertisement to receiving a job offer. With the knowledge you currently possess, create a
graduate scheme selection process, taking into consideration available technology and the need to
attract people from anywhere in the world who will be able to work anywhere in the world. After reading this chapter, come back to the process you originally described and make improvements based on
what you have learned.
The design and validation process in selection
Two main principles underpin selection procedures in organisational settings. The first
principle is that there are individual differences between people in aptitude, skills and other
job-related personal qualities (see Chapter 2). This simple principle leads to the very important conclusion that people are not equally suited to all jobs and suggests that procedures for
matching people and jobs could have important organisational benefits. The second principle is that future behaviour is, at least partly, predictable from evidence of past and current
behaviour. The goal of selection activities is to match people to jobs and use procedures (e.g.
interviews, psychometric tests) to provide a means of estimating the likely future job performance of candidates; the belief that future job performance can be estimated is an important
facet of this principle.
There are several main elements involved in designing and implementing a selection procedure. The process begins with a job analysis to define a competency model and create a person
specification. This information is used to identify the selection and assessment criteria and may
also be used for advertising the job role. The job analysis information is used to decide which
selection methods to use to assess applicant behaviour related to the selection criteria. Selection
methods can then be piloted and refined and, where possible, validated to see if an individual’s
performance on the methods correlates with job performance. This can be done by asking
current employees to try the methods as ‘candidates’ or by asking actual candidates to try the
methods but excluding their results from selection decision making. Piloting the methods
allows for data to be gathered to then assess the accuracy of the methods. Those assessed
during the pilot should also be asked for their feedback on the methods.
Chapter 3 Selection: Analysing jobs, competencies and selection methods
71
In attracting a pool of applicants, prospective candidates will engage in self-selection where
they can make an informed judgement about whether the particular role suits their skills
and abilities. Once candidates have applied, a selecting out process may take place based on
eligibility (e.g. essential criteria such as educational qualifications are checked) before the
selecting in process is undertaken using the chosen selection methods. Following the selection
decisions, the accepted applicants enter the organisation (assuming they accept the organisation’s offer).
After some time has elapsed, information on the work performance of job holders (i.e.
measures of performance related to the original competency model) can then be used to
examine the validity of the selection methods and decisions (i.e. whether high and low scores
on the selection instruments are associated with good and poor work performance respectively). Results from validation studies can be used to reassess the original thinking behind
the job analysis and choice of selection methods and can help inform changes to the original
competency model. Best practice selection is an iterative process whereby feedback is used
continually to improve the selection system and enhance accuracy and fairness. Feedback
to candidates (both successful and unsuccessful), which is useful for development, should
be provided – although many organisations have practical and legal reasons for not always
giving detailed feedback.
Key learning point
Most selection processes require candidates to go through several stages before selection decisions
are made. There is typically first a ‘selecting out’ phase, where candidates who do not meet essential
criteria (e.g. academic qualifications or experience) are filtered out of the process. The subsequent
stage(s) then ‘select in’ to the organisation using the chosen selection methods.
A thorough job analysis is the foundation of an effective selection process and is used to
guide the design and choice of selection methods. The outputs from a job analysis should
detail the tasks and responsibilities in the target job and provide information about the particular behavioural characteristics required of the job holder. The analysis may suggest, for
example, that certain personality and dispositional characteristics are desirable, together with
specific previous work experience, technical qualifications and levels of general intelligence
and specific cognitive abilities (e.g. numerical ability).
The next stage in the process is to identify and if necessary, design selection instruments
(e.g. psychometric tests, work simulation exercises, interviews, application forms) that can
be used to examine the extent to which candidates display the required characteristics. These
instruments are then used to assess candidates and selection decisions are made based on
candidates’ performance in these tasks.
Empirical validation studies are needed to monitor the quality of the selection process.
Here, the psychologist is focusing on how useful (and accurate) the selection procedure was
in identifying ‘the right person for the job’ by testing the strength of the links between performance during selection and subsequent job performance. In addition, it is important to
assess the candidates’ reaction to the process too. Poorly run selection processes are likely to
result in candidates having a negative first impression of the organisation. Consequently,
the best candidates for the job could decide to work for another (possibly competing) organisation. Historically, selection has tended to be a ‘buyers’ market’. However, there has been
a recognition in some organisations that selection is a two-way process: in order to attract
the best candidates, being seen to have professional, fair and effective selection procedures is
very important. This recognition is also enhanced by the increase of social media and public
72
Work Psychology
review in which organisations place importance on applicant perceptions to avoid public
slander. This is especially important for organisations where the applicant pool is also their
potential customer.
Unfortunately, validation studies are rarely conducted in organisations because they can
be difficult to do well and are time-consuming and costly. Validating a selection process often
means tracking the performance of new recruits over several months or years. Validation
studies are usually only possible with large-scale recruitment programmes and many organisations do not recruit large numbers of people into one specific job role to provide enough
data for statistical validation. The result is that the best-practice methodology of (i) conducting a thorough job analysis, (ii) designing selection tools based on a job analysis output and
(iii) validating the process, tends to be reserved for large organisations and is near impossible
for small to medium-sized enterprises (SMEs) to achieve (Wyatt et al., 2010). Organisations
therefore often draw upon research evidence from other organisations by employing work
psychologists to advise on best practice procedures in selection.
Key learning point
Best practice selection is an iterative process. It involves thorough job analysis and using feedback
from validation studies to continually improve accuracy and fairness. Validation can be difficult
to achieve.
Job analysis and competency modelling
Job analysis procedures are designed to produce systematic information about jobs, including
the nature of the work performed, responsibilities, technology/equipment used, the working conditions and the position of the job within the organisation. The outputs from a job
analysis are often used to generate a job description and a person specification for the job
role. It is worth noting that satisfactory job analyses are prerequisites for many decisions and
activities that have a crucial influence on the lives of employees within the organisation.
These include not only the design and validation of selection procedures, but also training
and career development schemes, job design or job redesign activities, job evaluation processes and health and safety interventions. Voskuijil (2017) provides a detailed discussion of
job analysis research and methods that will be introduced in this chapter.
Traditionally, job analyses have been used to identify the tasks and responsibilities associated with a job role. Job analysis is an umbrella term for many different analyses and involves
analysing job tasks. In practice, it is sometimes difficult to translate the information into
describing observable behaviours that underpin the tasks and responsibilities. For example,
it is difficult to accurately describe the behaviours associated with writing a complex financial report. As a result, competency analysis has become very popular in organisations to
help define the person-focused assessment criteria. A competency can be defined as the cluster
of specific characteristics and behaviour patterns a job holder is required to demonstrate
in order to perform the relevant job tasks with competence. Many organisations now use
competency modelling to identify the required knowledge, skills and behaviours that are
essential to perform a specific job role. The main aim of these analyses is to derive a competency model for the target role. A competency model comprises a comprehensive list of all
the relevant competencies associated with a given job role. For more detailed discussions
see Sanghi (2016).
Chapter 3 Selection: Analysing jobs, competencies and selection methods
73
Point of integration
Competencies can be used as assessment criteria which have a variety of different applications including selection, but also performance appraisal, management development, careers counselling and so
on, as you will see in several chapters in this book. One of the key advantages in using competencies
as assessment criteria is that competencies can become a common language shared by employees in
the organisation to describe the desired (and undesirable) behaviours during performance appraisal
and career development activities. In this way, the competency modelling outputs can directly inform
the design of the assessment criteria in selection.
Competency modelling and defining behavioural indicators
Exercise 3.1
Communicating with clarity is a key competency for the job of a teacher. An initial job analysis indicates that a teacher must be able to impart accurate information and be receptive to others. Some
example behavioural indicators for this competency are:
■
listens effectively and checks to ensure understanding;
■
communicates accurately and without ambiguity;
■
shows a genuine interest in other people’s views;
■
always approachable and receptive.
Example negative behavioural indicators are:
■
withholds or provides inaccurate information;
■
unapproachable;
■
uses inappropriate language or tone;
■
tends not to listen to others.
Suggested exercise
Think about your own experiences in education. What do you remember about your good teachers and
your less effective teachers? Describe other competencies that might be associated with the job role of a
teacher. For each, suggest some example positive and negative behavioural indicators.
Although the use of competencies and competency modelling has become commonplace,
there is considerable confusion about what competencies are and how they differ from other
job analysis outputs. There are three distinctions that can be made:
■
job-oriented (or task-oriented) analysis procedures focus on the work itself, producing a
description in terms of the technology/equipment used, the end results or purposes of
the job, resources and materials utilised;
■
worker-oriented (or person-oriented) analyses concentrate on describing the psychological or behavioural requirements of the job, such as communicating, decision-making
and reasoning (see Moscoso et al., 2015);
74
Work Psychology
■
and competency models that categorise and define behavioural indicators or behaviour patterns that impact on performance, such as actively listening to customers and
attempting to understand their needs. However, in many organisations competency
models have been developed without using appropriate psychological techniques to
identify precisely the required behavioural indicators associated with a specific work
role. This is important because it is these indicators that can be targeted for measurement purposes in selection, and other contexts. Further, some organisations use generic
competency frameworks: this often comprises a set of core competencies relevant to many
job roles in the organisation. This approach is appealing for cost purposes and ease of
interpretation, but generic competency frameworks often lack sufficient behavioural
specificity. In other words, the behavioural indicators in a competency model should
be directly relevant to the job role and the context within which the employee operates. By adopting a tailor-made approach, the assessment criteria can be more accurately
defined, and more useful during selection.
Key learning point
Competency models define the key behaviours associated with performance in the target job role.
While generic competency frameworks may be a tempting and cost-effective option, bespoke is best
as every job role (however similar they may appear) has context-specific behavioural indicators that
could be missed on a generic framework.
It is important to note that job analysis outputs and competency models are not static
documents: best practice selection is characterised by a continuous improvement cycle. Over
time, as the needs of the organisation evolve (possibly in response to a range of external
influences such as changes in legislation, technology used or customer demand), the job
requirements and competencies required for a specific role may also change. For example,
in the UK general practitioners have experienced significant changes in their role: familiarity with information technology, financial acumen and legal awareness are now core
requirements for the job of running an effective medical practice (Patterson et al., 2013).
The evidence suggests that competency models and outputs from job analyses should be
regularly reviewed to evaluate the original assessment criteria. There are now legal reasons
for ensuring accurate selection procedures are used and it is essential for compliance with
current local legislation. In many countries, organisations could be asked to demonstrate
that they have used fair and accurate selection processes that assess candidates on competencies that are relevant to the job role.
Job analysis data
Sources of job analysis data may be divided into four categories: written material, job holders’
reports, colleagues’ reports and direct observation. Examples of job analysis and competency
modelling techniques can be found in Table 3.1. There is probably no one best way to conduct
a job analysis. Indeed, it is generally recommended that using multiple methods is the best
way of capturing the most comprehensive range of perspectives on the target job role and
account for future challenges of the role.
Chapter 3 Selection: Analysing jobs, competencies and selection methods
Table 3.1
Examples of job analysis/competency modelling techniques
Method
Administration
Appropriateness
Questionnaire/Survey
• Online, mail or face-to-face
Useful for gathering data from a large sample in
different contexts and locations; often used for
validation of role analysis outputs arising from the
collection of other data.
• Structured and/or open questions
• Bespoke solution for a specific context
Interviews/Focus Groups
• Face-to-face, via phone or teleconference
• Structured and/or semi structured interviews
• Individual or group interviews
Conducted with subject matter experts (such
as those in the role, supervisors, members of
the multi-disciplinary team, customers). Often
combined with other objective data.
• Focus groups
Observation methods
• Direct and indirect observation of a role, e.g. the
work environment, tools and equipment used,
interrelationships with other workers, handling
of critical incidents
• Data collected can include both quantitative
(proportion of time spent) and qualitative
(demands of different tasks)
Job roles are often complex, involving diverse
behavioural responses, novel situations, creative
problem solving, multi-disciplinary teamwork and
use of technology. Therefore, direct observation can
be very helpful in unpicking some of the required
abilities.
Task analysis
• Breaking down the detailed components of
how a task is performed including various
elements such as: duration, frequency,
complexity, cognitive requirements, manual
handling, environmental conditions, role of
team members, necessary equipment, safety
requirements and any other unique factors.
If a job role is very complex, breaking the task down
into smaller components can be very helpful in
defining the abilities required.
Analysis of existing policy
documentation
• Organisational charts
This is a key activity to begin with, often appropriate
when conducting job analysis. A mapping exercise
may be helpful once a new competency framework
has been defined, to ensure that no behavioural
indicators have been overlooked.
• Strategic plans
• Task inventories
• Pre-existing competency frameworks
• Pre-existing selection material
• Application forms
• Job descriptions
• Professional body requirements and standards
• Documentation for other relevant professions
Work records
• Human resource records and statistics (e.g.
absenteeism, attrition rates)
• Staff engagement and culture survey data
• Workplace health and safety data
• Job applicant test scores correlated against job
performance
Particularly useful when looking to understand
the organisational factors (e.g. culture, staff
engagement, turn over, safety) that impact on a job
role. Supports the understanding of the job factors
required within a particular context, such as public
sector roles and private sector roles.
Research methods in focus
A procedure for collecting job analysis information when interviewing job holders that remains popular involves
the use of Flanagan’s (1954) procedure known as the critical incident technique (CIT). In brief, using this technique involves asking the interviewee to recall specific incidents of job behaviour that are characteristic of either
75
Work Psychology
76
highly effective or highly ineffective performance in the job. A critical incident is defined as ‘any observable human
activity that is sufficiently complete in itself to permit inferences and predictions to be made about the person
performing the act’. By describing the behaviours and characteristics that led to the incident, the analyst can
get an excellent insight into the relevant job behaviours that are important. Not only will this information enable
the analyst to differentiate between good and poor job holders, it can provide some contextual information that
might be helpful in designing the actual section methods.
CIT is a versatile technique that can be conducted either during interviews with individuals or with groups
of job holders. A group-administered CIT can be very useful to get an initial ‘cross-validation’ of information,
where job holders might have experiences or witnessed similar incidents (see Patterson et al., 2013). One
potential drawback with using this method is that it might lead to job holders focusing on the extreme
behaviours (good or bad) that influence performance, at the expense of a detailed analysis of the more
routine but important elements of job performance. However, a skilled practitioner can use CIT in a way that
avoids the problem.
Key learning point
No single method of job analysis is perfect. Each approach has different strengths and limitations.
Job analyses are usually best developed by using as many different approaches and means of data
collection as possible and pooling the results.
As occupational psychologists, ensuring the selection methods you are using are accurate
and predictive are key objectives to practice. Two elements that measure the effectiveness
of selection methods are reliability and validity. Any measuring instrument used in a selection procedure must be both reliable and valid. The intricacies of reliability and validity are
explained next in this chapter.
Reliability
The reliability of a selection instrument is an extremely important characteristic and refers
to the extent to which it measures consistently under varying conditions. If the same candidate
produces very different scores when they take a test on two different occasions, the reliability
of the test must be questioned. In technical terms, if a selection tool has high reliability it is
relatively free from errors of measurement (i.e. the score that we obtain is close to the true
score). There are several different types of reliability, including internal and external reliability. Internal reliability assesses the consistency of results between questions that make
up the test, whereas external reliability assesses consistency of the test results from one use
to another. Test reliability may be calculated and expressed in a numerical form, with zero
indicating total unreliability (e.g. in a totally unreliable test someone’s score on a second
administration could not be predicted with any accuracy from their score on the first). See
Table 3.2 for forms of reliability and their descriptions.
Chapter 3 Selection: Analysing jobs, competencies and selection methods
Table 3.2
External forms
of reliability
Internal forms
of reliability
77
Forms of reliability
Test-retest reliability
Test is completed by a sample at time one and then again by the same
sample at a later date (time two) – ensuring the testing conditions are
same on both occasions. High reliability would mean the scores on the test
do not differ drastically between time one and time two – the construct
being measured is relatively stable.
Parallel forms
of reliability
Two forms of a test measuring the same construct are created meaning
they have equal difficulty levels and similar content. You would expect
performance on Form A and Form B to be similar.
Cronbach’s
coefficient alpha
A statistical analysis to ensure the questions of a test measuring the
same construct are producing consistent results. It is a measure of the
consistency of response to questions designed to measure the same thing.
You would expect an alpha to be around 0.7 or above, to indicate good
internal consistency.
Split-half method
The form is split into two (odd/even numbers or top half/bottom half)
and examine if the scores produced are similar.
Kuder-Richardson 20
A special form of the Cronbach’s alpha that is used for questions that have
right or wrong answers (dichotomous answers).
Validation processes
The pivotal stage in the selection process occurs when the selection decision is taken, and a
candidate is either offered a position within the organisation or turned away. At this point
various pieces of evidence concerning the current or past performance of candidates (e.g.
behaviour at an interview, psychological test scores or references), usually referred to as
predictors, are used to decide whether or not a candidate is suitable for the job in question.
Although a job analysis may suggest that certain candidate characteristics might be desirable,
on its own the job analysis data cannot prove that candidates with these characteristics will
do better than others on the job. Validation is about finding out whether the various parts
of the selection process were effective. There are several ways of examining the effectiveness
of a selection process.
Criterion-related validity
Criterion-related validity is the strength of the relationship (or the correlation) between the
predictor (e.g. psychological test scores or interview ratings) and the criterion (e.g. subsequent
work performance). Criterion-related validity is high if candidates who obtain high predictor
scores obtain high criterion scores and candidates who obtain low scores on a predictor also
obtain low criterion scores. See Figure 3.1 to explore the meaning of correlations.
The ideal way to collect criterion-related validity data is to use a predictive (or follow-up)
design. This design involves collecting predictor information for candidates during the selection process and then following the candidates (e.g. during their first year of employment) to
gather work performance criterion data (see for example Hoffman et al., 2015). What might be
Work Psychology
78
High
Average number of
units per day
Average number of
units per day
High
Low
High
Job knowledge
test score
(a)
Figure 3.1
High
Low
(b)
Cognitive ability
test score
Some hypothetical predictor and criterion scores
Key learning point
The criterion-related validity of a selection procedure is normally indicated by the magnitude of the
validity coefficient (the correlation between the predictor and criterion scores).
Research methods in focus
The meaning of correlations
The strength of the relationship between predictor scores and criterion scores is usually expressed as a correlation
coefficient (referred to as a validity coefficient). Perfect positive correlation between two variables will produce a
correlation of 1. No correlation at all will produce a validity coefficient of zero.
When the correlation between two variables is high it is possible to predict the score of one when supplied
with the score of the other. Consider, for example, Figure 3.2, where the predictor–criterion relationship is
perfect (i.e. a validity (correlation) coefficient of +1.0). In the case of a new candidate, it would be possible to
obtain their score on predictor A and thus predict a score on the criterion. If the organisation wished to select
staff who would produce an average of at least 75 units per day, what should be done?
As Figure 3.2 shows, if only people who obtained a score of above 80 on selection test A were offered jobs
by the organisation, all future employees would be very likely to produce at least 75 units per day. In practice,
selection can rarely be conducted in such an idealised and clear-cut fashion and there are several practical
problems that must be considered.
Research has shown that it is most unusual in practical situations to obtain validity coefficients much in
excess of +0.5 let alone the coefficient of + 1.0 that will allow perfect prediction (e.g. Patterson and Ferguson,
2007; Salgado et al., 2001). Nevertheless, validity coefficients of considerably less than + 1.0 can provide a basis
for improved selection.
Chapter 3 Selection: Analysing jobs, competencies and selection methods
79
High
Criterion (B) units per day
100
75
50
25
0
0
50
100
150
Low
200
High
Selection test (A)
Figure 3.2
Prediction of criterion scores from selection test scores
assessed as the predictor criterion needs careful consideration (Lievens and Sackett, 2008; Van
Iddekinge et al., 2016,). Task performance, citizenship performance and counterproductive
work behaviour are the three major domains of job performance (see Chapter 4). Cognitively
loaded predictors (e.g. cognitive ability test scores) tend to be the strongest correlates of task
performance and non-cognitive predictors tend to be the best predictors in the citizenship
and counterproductive behaviour domains. This indicates that careful attention needs to be
paid to the criterion of interest to the organisation: this should be a critical determinant of the
eventual make-up and success of a selection system (Lievens and Sackett, 2017; Sackett and
Lievens, 2008). Clearly it would be unwise to have a selection system that predicted perfectly
aspects of performance that were irrelevant to the organisation.
Similarly, it is important to consider the role humans play in the measurement of performance (Hauenstein and McCusker, 2017, Ingold et al., 2018). When selection processes are
new, the way performance measures are identified and designed improves predictive validity,
but as the selection process matures, the qualities of those rating performance (raters) tend to
become increasingly important. There is now some research into the effects of interventions
designed to standardise and quality assure rater involvement to ensure raters are assessing candidates objectively and limiting unconscious bias. In a high-stakes selection process, Patterson
et al. (2014) developed competency models for the role of assessors. Competencies that were
identified as required skills included communication skills and openness to learn and develop.
These can be valuable for (i) recruiting suitable assessors, (ii) measuring the performance of
assessors and highlighting areas for development and (iii) training assessors effectively; all of
which are likely to raise the validity and reliability of the process they are assessing.
Ideally, scores on selection tasks should not be used to make selection decisions until
after a validation study has been conducted. This creates a difficulty: until the relationship
80
Work Psychology
between predictor and criterion is firmly established, candidates should be offered employment regardless of their performance on the predictors. If job analysis data suggest that the
use of certain predictors should improve their selection decisions, in many organisations,
managers would not be willing to permit candidates with low scores on these predictors
to enter employment. Sometimes it is possible to convince them not to use the results of
potential predictors and to continue to use their existing methods while a validation study
is carried out. Often, however, they cannot accept the constraints of a complete predictive
validity design (because it takes considerable time to follow up enough candidates) and some
compromise is needed.
Another design problem with the predictive validity design is that of ensuring that the
predictor results obtained by new employees are not revealed to other members of the organisation before a validity study has been conducted. Obviously, a supervisor who is allocated
a new employee with either a high or low predictor score could be affected by these results.
The supervisor’s behaviour towards the new employee might be influenced and/or estimates
of the new employee’s work performance could be biased.
An alternative design for conducting criterion-related validity studies is the concurrent
design. In the concurrent design, predictor data are obtained from existing employees on
whom criterion data are already available. Existing employees’ predictor scores are correlated with their criterion performance. One advantage of the concurrent design is that the
organisation is not required to collect predictor data from job applicants and then somehow
ignore that data during selection decision-making. Predictor data are collected from existing
employees only. A second advantage of the concurrent design is that there is no time delay
between the collection of predictor and criterion data. The criterion data are likely to be readily available. There is certainly no need to wait for the lengthy follow-up period. Figure 3.3
compares the two designs in a hypothetical situation and indicates the considerable differences in timescales and data collection effort that can occur.
However, this approach has disadvantages. It is very likely that the concurrent sample
is incomplete and not representative of the potential workforce. Current job holders have
already been selected in using existing company selection procedures and may have adapted
their work behaviour after being with the organisation for some time (i.e. did employees
have strong arithmetic skills before entering the job role or did they acquire their arithmetic
skills whilst working in the role?). No data are available on people who were not hired by
the company, or on those who were hired but have subsequently left. It would not be fair or
informative to test new applicants for skills that could only be acquired through training and
experience within the company. Such problems make it hard to be certain about the actual
predictive value of results derived from a concurrent validity study.
Key learning point
Predictive validity designs are more rigorous than concurrent validity designs, but their implementation usually presents several practical problems.
As we have already seen, once people are in the job their performance may be measured
across several different criteria. Therefore, validation studies can quickly become very complex practically because organisations rarely use one single predictor to make selection decisions. Given the multifaceted nature of job analysis information, recruiters are likely to design
multiple selection tools to assess these criteria. Therefore, recruiters must decide whether a
job applicant must score highly on all assessment criteria (a non-compensatory approach to
selection) or whether high scores on some criteria can make up for low scores on another
Chapter 3 Selection: Analysing jobs, competencies and selection methods
Predictive
Concurrent
Test candidates as they
are recruited (e.g. 12
per week)
1 year
Test 100 current employees
Sample of approx. 450
( 2/3 of candidates
start work)
Collect work performance
data on current employees
(2 months)
81
Sample of approx. 300
Collect follow-up data on
work performance during
first year
(12 months)
( 1/3 of candidates leave
during first year)
Total time
Total number tested
Final sample size
Sample characteristics
Figure 3.3
Calculate validity
coefficient
Calculate validity
coefficient
18 months
450
200
Satisfactory
2 months
100
100
Incomplete and possibly
biased
A comparison of predictive and concurrent validity study designs
(a compensatory approach). In practice, recruiters might assign different weightings to various
assessment criteria, depending on the nature of the job role (e.g. it might be decided that
customer service skill is the most important criterion and applicants who do not achieve a
certain score will not be considered further).
Most selection systems combine several predictors such as an applicant’s score on an interview and on a situational judgement test. In validation studies, a key question is how much
does adding another predictor increase the predictive power of the selection process? This
is known as incremental validity. In other words, recruiters might want to know if using the
selection tool adds value to the selection process by providing further insight into candidates’
potential job performance. Information on the incremental validity of a specific selection
82
Work Psychology
tool is extremely valuable as it allows organisations to conduct a cost–benefit analysis of using
additional tools. Research including meta-analysis (see Chapter 1 for detail on meta-analysis)
has attempted to estimate the incremental validity of one predictor over another (like a
‘league table’ of selection methods). This research attempts to better understand the relationships among predictors and dimensions of job performance to enhance the practical choice
of selection methods (see Sackett and Lievens, 2008, for a summary of incremental validity
studies and limitations to these). The typical levels of predictive validity for various selection
methods are examined later in this chapter.
Criterion-related validity is the most important type of validity as far as selection is concerned because the main purpose of selection procedures is to select people who are likely to
perform well in the role. However, there are also other important types of validity. See Table 3.3
to explore the types of validity and their importance to personnel section. Exercise 3.2 gives
an opportunity to reflect upon your own experiences of selection processes, particularly in
relation to your perceptions of the validity of the process.
Table 3.3
Different types of validity in selection
Faith validity
This is ‘blind’ faith that a selection method works because someone plausible said so
Face validity
The selection tool content appears relevant to the target role (determined by the
applicants)
Content validity
The content of the selection tool is judged to be directly relevant to the target role by
subject matter experts
Criterion validity: Concurrent
A form of criterion-related validity in which data on the predictor and criterion are
obtained at the same time. High correlations between predictor and criterion scores
indicate concurrent validity
Criterion validity: Predictive
This is the extent to which a predictor measure (e.g. a selection test score) is correlated
to a criterion measure (e.g. work performance). High predictive validity indicates that
a selection measure gives an accurate indication of candidates’ future performance on
the criterion
Incremental validity
This is an empirical issue to determine how much additional value using another
assessment provides
Construct validity
An indication of the extent to which the test or procedure measures the construct that
it is intended to measure (such as empathy, integrity, etc.)
Political validity
An indication of the extent to which various stakeholders and stakeholder groups (such
as employers, parents, government departments, society etc.) consider the tool(s) to
be appropriate and acceptable for use in selection
Point of integration
Validity – and the various types of it – is an important property of all measures and methods used in
work psychology, not just in selection. For example, considerable time and effort goes into the validation
of attitude measures and measures of leadership behaviour. The same applies to the issues of reliability,
fairness and candidate perceptions discussed later in this chapter.
Chapter 3 Selection: Analysing jobs, competencies and selection methods
83
Key learning point
To be effective, selection methods must be both reliable and valid.
Exercise 3.2
Robo-recruiters are quick to replicate human bias
Advances in artificial intelligence and the use of
big data are changing the way many large companies recruit for entry level and junior management positions. These days, graduates’ CVs may
well have to impress an algorithm rather than an
HR executive. While algorithms supposedly treat
each application equally, experts are divided about
whether so-called robo-recruitment promises an
end to human bias in the selection process – or
whether it may in fact reinforce it. The problem,
experts say, is that to find the best candidates an
algorithm has first to be told what ‘good’ looks
like in any given organisation. Even if it is not
fed criteria that seem discriminatory, an efficient
machine-learning system will quickly be able to
replicate the characteristics of existing workers. If
an organisation has favoured white male graduates
from prestigious universities, the algorithm will
learn to select more of the same. Many algorithmic tools are most likely perpetuating bias. The
good ones should have auditing processes in place
and focus on evidence-based methods. An example of this is of a hiring platform who anonymise
the tests completed by candidates and human
assessors rate each question. Every stage of the
process has been designed to strip out bias. With
the same aim, another high-volume recruitment
company developed a game-based test in which
candidates are scored on their ability to take risks
and learn from mistakes, as well as on emotional
intelligence. The process has increased the ethnic
diversity of its shortlisted candidates and has been
more successful at selecting candidates who will
eventually be hired.
Adapted from Boyde, E. (2017). Robo-recruiters are quick to replicate human bias. The Financial Times, December 7
© Financial Times Limited 2019. All rights reserved.
In this exercise you are asked to reflect upon the above text regarding artificial intelligence in selection
processes and its impact on diversity.
Suggested exercises
1 What are the key concerns of AI regarding diversity in selection?
2 Describe the elements of the recruitment process. From the perspective of the applicant, how might AI
and technology-based selection methods demonstrate face validity and content validity?
3 In general, what might make AI inaccurate or unreliable? What methods would you choose to ensure
these selection processes are valid?
Estimating the validity of selection procedures
Predictive or concurrent validation processes may be used to estimate the criterion-related
validity of a selection procedure. Most of the selection procedures mentioned so far in this
chapter have been examined in this way: investigators have conducted a huge number of
validation studies on many selection procedures in many industries. To estimate the validity of a selection procedure more than one study is needed, so that any bias due to specific
features of any study will not have an unduly large influence on the findings about that
84
Work Psychology
selection procedure overall. An example will help to illustrate the way in which the results of
validation studies may be cumulated and summarised to more accurately estimate validity.
A researcher may be interested in the extent to which an individual difference characteristic,
such as intelligence or achievement motivation (call this factor X) is predictive of managerial
performance. Any individual study, using a sample of managers, will not give a definitive
result for the validity of factor X (see Table 3.4).
As Table 3.4 makes clear, some of the problems are caused by sampling error (item 1), imperfect reliability in the selection method or the criterion measure (item 2), and range restriction
in the selection method scores (item 3). Consider these sources of error in relation to our example of the predictive validity of factor X. The sources of error are usually referred to as artefacts
because they are not part of the natural relationship under investigation but are a consequence
of the particular investigative procedures used in the study (carrying out research in functioning organisations is very different from carrying out research in the laboratory setting). If the
researcher in our example was able to identify 10 studies of the predictive validity of factor X,
each study would have inaccuracies due to the artefacts that impacted upon that study. Sampling error would be present because in each study the sample involved would not be perfectly
representative of the population (although studies with larger samples would of course be less
prone to sampling error, because the sample is a bigger percentage of the available population).
As far as unreliability and range restriction are concerned, these artefacts will adversely
affect observed validity coefficients (i.e. the validity will appear to be less than it is).
Table 3.4
1
Major sources of distortion in validation studies
Sampling error
The small samples (e.g. 50–150) used in many validation studies mean that the results obtained may be unduly
influenced by the effects of small numbers of people within the sample whose results are unusual. As sample size
increases, these irregularities usually balance each other out and a more reliable result is obtained.
2
Poor measurement precision
The measurement of psychological qualities at both the predictor (i.e. selection method) and criterion (i.e. job
performance) stage of the validation process is subject to unsystematic error. The result we get from a selection
method comprises the candidate’s true score plus error. The error (unreliability) in the scores obtained will reduce
the maximum possible observed correlation between predictor and criterion: the error is unsystematic and random;
thus, this element of the predictor or criterion score will not correlate systematically with anything. This means that
as reliability decreases, the maximum possible correlation between predictor and criterion will decrease.
3
Restricted range of scores
The sample of people used in a validation study may not provide the full (theoretically possible) range of scores
on the predictor and/or criterion measures. This occurs in practice because data are collected on predictors in a
restricted population, i.e. individuals that have already been selected. The results are then generalised to the whole
applicant pool, i.e. an unrestricted population. This has a straightforward statistical effect: it limits the size of the
correlation between two variables. So, just like unreliability, range restriction in a sample serves to reduce the
magnitude of the observed correlation coefficient, making the predictor appear less valid than it is.
Key learning point
Studies with larger sample sizes are less prone to sampling error and more likely to give reliable results.
This is because they represent a larger proportion of the population than a smaller sample would.
Chapter 3 Selection: Analysing jobs, competencies and selection methods
85
Financial utility
Using selection procedures with good predictive validity is always important but, unfortunately, procedures with good predictive validity alone do not guarantee that a selection procedure will be cost-effective. Two important factors determining cost-effectiveness (usually
referred to as utility) are:
1
the selection ratio, i.e. number of jobs/number of candidates;
2
the financial benefit of improved job performance.
Let us now examine in a little more detail the role that validity, selection ratio and the
other factors play in determining utility. Figure 3.4a shows the situation for a validity coefficient of 0.1. The shaded areas (B and C) identify people who will be hired by the organisation.
Notice, however, that although all these people achieve the minimum acceptable score on the
predictor (the cut-off score), only a proportion of them (B) also show satisfactory work performance. Similarly, although all the people in the unshaded areas (A and D) fail to achieve
the cut-off score on the predictor, some of them are capable of satisfactory work performance
(those in area A). Areas B and D represent correct selection decisions; area D contains people who have justifiably been rejected, known as true negatives. Area C contains the false
positives; that is, people who would be hired but produce unsatisfactory work performance.
Area A contains the false negatives; that is, people who would not be hired but are capable of
satisfactory work performance. Areas A and C represent errors in the selection process. Note
that unless the validity coefficient is 1.0 there will always be errors of selection.
As Figure 3.4b shows, when validity increases (0.5), the proportions in areas A and C
decrease and the proportion of correct decisions (areas B and D) increases. Thus, as validity
increases, the quality of selection decisions also increases.
When calculating the financial benefit of improved job performance, basic utility guidance developed by Hunter and Schmidt (1983) applies a value of 20 to 40 per cent of salary to
each standard deviation of performance. Then, the cost of the selection procedure needs to
be subtracted and account taken of the expected tenure of recruits and the number of people
selected.
Criterion
score
Criterion
score
Satisfactory
Satisfactory
A
B
D
C
D
Reject
Figure 3.4
C
Unsatisfactory
Unsatisfactory
(a)
B
A
Hire
Selection score
Reject
(b)
Hire
Selection score
The effect of different validity coefficients on the proportion hired: (a) validity coefficient = 0.1;
(b) validity coefficient = 0.5
Work Psychology
86
Selection methods
There is over a century’s research literature exploring how best to select a person for a job.
Work psychologists have had a major influence on the way selection processes are designed
and implemented across many organisations in all sectors of industry across the world. Over
the past three decades, the research shows that not all the methods are equally useful or
appropriate in all selection processes.
How well do selection methods work?
Before it is feasible to consider how well selection methods work it is necessary to be clear
about what it means for a selection method to work. Table 3.5 lists a number of these features.
Table 3.5
Evaluation standards for selection procedures
Evaluation
How can this be evidenced?
1. Reliability
A selection method is reliable if it is consistent in how it assesses
candidates under varying conditions.
• Psychometric evaluation using statistical analysis
2. Validity
The selection tool measures what it claims to measure, it should be
relevant, precise and accurate.
• Psychometric evaluation using statistical analysis
3. Fairness, promotes diversity
This is based on three principles: 1) valid selection criteria; 2) accurate
and standardised administration by trained staff; 3) monitored
outcomes.
• Evaluation questions posed to candidates
• Analysis of sub-group differences
4. Legality
The extent to which the design of a selection process and the
decisions generated is legally defensible. Selection processes that are
perceived as unfair are more likely to result in legal case initiation.
• HR experts in employment law
5. Scalability for high volume recruitment
The extent to which a selection process can be scaled up or down and
remains efficient and effective for different volumes of applicants.
• Data modelling with costs of implementation and validity of
selection methods
6. Efficiency
The costs involved, and the time taken in developing and
implementing the selection tool(s).
• Analysis of costs by recruiters and employers
Chapter 3 Selection: Analysing jobs, competencies and selection methods
Table 3.5
87
(continued )
Evaluation
How can this be evidenced?
7. Utility
The costs involved, and the time taken to develop more accurate
adequate procedures need to be balanced with the potential benefits
(e.g. improved performance).
• Statistical analysis of the predictive validity of a selection tool
compared to the costs to implement the tool (using established
utility calculation methods)
8. Expertise required for analysis & interpretation of information generated by the tool
Some selection tools (for example personality tests) require an
appropriately trained individual to administer, score and provide
feedback. Similarly, assessors in selection centres must also be
appropriately trained in how to evaluate a candidate in a work sample
test for example.
• For psychometric tools there are specific licensure guidelines (e.g.
from the British Psychological Society)
9. Positive applicant perceptions
The extent to which applicants react positively to the selection
process and each selection method within that process.
• Evaluation questionnaires of candidate perceptions
10. Assessor/Interviewer training
The extent to which assessors and interviewers are trained
appropriately and the methodology is designed to ensure they are
calibrated effectively.
• Evaluation of assessor/interviewer training
When candidates apply for a job with an organisation it is likely that they will complete
an application form and other documents in which they are expected to provide certain
verifiable biographical information (such as previous employment, personal career history
and education). The main purpose of collecting biographical data is that information can be
gathered on several candidates and correlated with subsequent job performance. Items of
information that predict subsequent performance can then be identified and perhaps used
in selection decision-making. Items of information chosen on this basis do not necessarily
have any obvious link with the job; it has merely been demonstrated, on a statistical basis,
that they predict future performance.
When using predictors such as interviews, psychometric tests or work-sample tests, the
specific areas to be investigated in the interview or test should be derived from job analysis
data. The information from the job analysis provides a basis for deciding which factors (e.g.
numerical ability, communication skills, organisational skills) might be important for job
success.
In most practical situations it is not appropriate to base selection decisions on the use
of one predictor only or even on one type of predictor. As you will see in this chapter, each
predictor has its own strengths and weaknesses. When several different predictors are used
together, the aim should be to ensure that the various predictors complement one another
rather than duplicate each other (i.e. incremental validity).
88
Work Psychology
Table 3.6
1
Summary of selection methods
Interviews
Many involve more than one interviewer. When several interviewers are involved, the term panel interview is
used. Some interviews are now conducted virtually and recorded for assessors to grade afterwards (these are
known as Asynchronous Video Interviews). The most important features of an interview are the extent to which a
pre-planned structure is followed and the proportion of questions that are directly related to the job.
2
Psychometric tests
This category includes tests of cognitive ability (such as general intelligence, verbal ability, numerical ability),
self-report measures of personality* and situational judgement tests. Such tests can be completed online or in
testing facilities.
3
References
Usually obtained from current or previous employers, these are used often in the final stages of the selection
process. The information requested may be specific or general and open-ended.
4
Biodata
Specifications of biographical information about a candidate’s life history. Some biodata inventories may contain
many (e.g. 150 + ) questions and ask objective questions, such as which professional qualifications are held by the
applicant, and more subjective ones, such as preferences for different job features.
5
Work-sample tests
Such tests literally use samples of the job in question (e.g. the contents of an in-tray for an executive position,
specific kinds of typing for a secretarial post or interacting with a customer for a sales position). The applicant is
given instructions and then a specific amount of time to complete the tasks.
7
Assessment centres
This procedure involves a combination of several of the previously mentioned techniques (e.g. psychometric
tests, interviews, work samples). Candidates are usually dealt with in groups and some of the techniques used
require the candidates to interact (e.g. in simulated group decision-making exercises or group presentations).
8
CVs/résumés and personal statements
These are written descriptions of the applicant’s professional experiences, education and any other relevant
information.
9
Application form
A form or collection of forms that an applicant must fill out as part of the process of informing an employer of
their availability and desire to be employed. They will usually ask for information on the applicant’s background
and why they believe they are suited to the job they have applied for.
*The term personality assessment is used here for simplicity. However, measures of personality are not really tests
because they do not have ‘right or wrong’ answers. Personality questionnaire and personality measure are also
appropriate terms.
Chapter 3 Selection: Analysing jobs, competencies and selection methods
89
Key learning point
In most situations it is best to use a combination of several selection techniques to ensure fairness,
accuracy and bolster validity.
Interviews
Interviews have long been the most popular form of selection (e.g. Dipboye et al., 2012; Macan,
2015; Ryan and Ployhart, 2014). They are used by nearly every organisation and for selecting
employees at all levels. Despite this popularity, interviews have been criticised for being subjective, unreliable and vulnerable to bias. However, with knowledge about how selection decisions are reached, and the potential sources of error and bias, there is the potential to design
and execute better interviews. There have been several reviews of the literature on selection
interviews over the years (e.g. Dipboye, 2017; Posthuma et al., 2002; see Levashina et al., 2014
for a summary and critical analysis of the last two decades’ research on the selection interview). These reviews show that interviews can be structured, and interviewers trained, in ways
that considerably enhance their reliability, fairness and validity. Therefore, a critical point
to remember is that the employment interview can represent an extremely important and
valid means of selecting employees if it is structured to ensure that (i) interviewer questions
are based on a job/competency analysis, (ii) questions are consistent across interviewers and
interviewees and (iii) the interviewers use a consistent set of criteria to evaluate interviewees’
responses. Meta-analytic data have consistently found strong evidence for the superiority of
structured versus unstructured interviews.
While there is considerable evidence to support the use of structured interviews in selection, there is much less agreement as to why selection interviews predict work performance.
The majority of research focuses on (i) analysing the way interviewers process information,
(ii) the social processes involved in the interaction between the interviewer and the candidate
and (iii) the quality of interviewer assessments. In their review of the last century of selection, Ryan and Ployhart (2014) highlight that there has been little research into the structure
and content of interviews, apart from how one might improve structured interviews (e.g.
Melchers et al., 2011). Instead, research has continued to focus on issues such as impression
management (e.g. Roulin and Powell, 2018; Swider et al., 2016). Investigations of these issues
have found that non-verbal and verbal cues and impression management can play an important role in determining the evaluation that a candidate receives in the interview. To combat
these issues research suggests that great care needs to be taken in the design and execution of
interviews for them to have a chance of exhibiting good reliability and validity.
It is suggested that factors such as job complexity, however, have been found to influence the validity of situational interviews. In examining whether situational or behavioural
interviews show incremental validity, findings have been mixed. Some research shows that
situational interviews display incremental validity over behavioural interviews, while other
research shows the opposite pattern. Despite this, both still display good validity, and can
both be used with confidence, although Levashina et al. (2014) suggest that behavioural interviews may be marginally better for more complex jobs. Perhaps actual experience becomes
more important as job complexity increases? Research has also indicated that the different
types of interview may be suited to assessing different types of performance. For example,
because situational interviews force candidates to think about how they would behave in
certain situations, they can predict maximal performance (the best a person can do) as well
as typical performance (what a person typically does). On the other hand, behavioural
90
Work Psychology
Table 3.7
Question type
Examples of interview
questions
Example
Behavioural
Give an example of a challenge that you faced as a graduate student and handled
effectively.
Situational
Suppose you notice your supervisor has given the wrong instructions to your new
colleague. What would you do?
Background
What previous experience do you have in the consulting sector?
Knowledge
How would you conduct a job analysis for a competency framework?
Opinions and attitudes
What is your opinion on emotional intelligence?
Goals and aspirations
Where do you see yourself in five years?
Self-descriptions and self-evaluations
How do you handle pressure?
interviews focus only on candidates’ actual experiences and therefore are likely to predict
only typical performance. A recent meta-analysis found structured behavioural interviews
to be free of adverse impact on gender and when designed properly can be used to combat
gender discrimination (Alonso et al., 2017).
Candidates can now be expected to attend interviews virtually or complete asynchronous interviews in which you are asked questions in writing or via pre-recorded video and
then record your answer. These answers can be assessed later by human raters or by AI. Early
research regarding these methods of interviewing have indicated that their use can impact
positively the recruitment of distant applicants, the selection process, and the method’s validity due to their heavily structured approach (Torres and Mejia, 2017; Torres and Gregory,
2018). Researchers must now work to keep up with the advances in technology to inform the
public of the validity of these methods and considerations.
Stop to consider
Which type of interview is best?
Not only are there many types of interview questions, there are also many forms of interviews, especially
with the help of technology. At what times might you use the different types of interview questions (see
Table 3.7 for examples)? How do you think candidates would perceive video interviews compared to
face-to-face interviews? How might asynchronous (video-recorded) interviews benefit organisations?
Psychometric tests and measures of personality
Psychometric tests are broadly a standardised sample of behaviour which can be described
by a numerical scale or category system. In other words, psychometric tests offer a quantitative assessment of some psychological attribute (such as verbal reasoning ability, general
Chapter 3 Selection: Analysing jobs, competencies and selection methods
91
intelligence, numerical reasoning ability, etc.). For selection purposes psychometric tests
may be divided into two categories: cognitive ability tests (e.g. general intelligence, spatial
ability, numerical ability) and personality measures (see Chapter 2 for more information
on specific cognitive and personality tests). Cognitive ability tests tend to measure maximal ability, whereas personality tests are measures of typical performance – they indicate
preferences.
Cognitive ability tests in selection
For many practitioners conducting large-scale recruitment programmes, ability tests are
relatively cost-effective selection tools, especially now they are computer-based and can be
completed via the Internet. Unlike most other selection techniques, to purchase and administer psychometric tests in many countries those administering tests need to demonstrate a
specific level of competence in administration, scoring tests and interpreting test scores (e.g.
the British Psychological Society’s Qualifications in Test Use).
In general, cognitive ability tests have been shown to be the single best predictor of job
performance (e.g. Sackett et al., 2017; Schmidt and Hunter, 1998) with validity coefficients
generally shown to be approximately 0.50 (i.e. explaining an impressive 25 per cent of the
variance in the test-taker’s future job performance). Large-scale research has demonstrated
consistently that cognitive ability accurately predicts some proportion of future job performance across almost all occupations and organisations. Many researchers argue that cognitive ability tests are perhaps the best and most widely applicable predictor in selection because
all jobs require some cognitive ability to learn the job and perform effectively. Some tests are
tailored to assess specific abilities such as mechanical comprehension or spatial ability. The
predictive validity of such measures depends upon the extent to which these specific abilities
are required to perform job tasks effectively. Tests measuring specific abilities tend to have
more face validity and generate positive candidate reactions in a selection context.
Research on the validity of cognitive ability testing has been scarce in recent years, possibly
due to cognitive ability having been so firmly established as a predictor of job performance in
earlier research (Ryan and Ployhart, 2014). However, there is merit for expanding the focus
of cognitive ability to include procedural knowledge (i.e. domain-specific knowledge) and
declarative knowledge (knowing facts and job relevant information), for example, in addition to the traditional ‘g’ (Schneider and Newman, 2015). Indeed, research has found knowledge tests to be one of the best predictors of job performance (Lievens and Patterson, 2011;
Schmidt and Hunter, 1998).
Key learning point
Cognitive ability tests remain consistent in producing good criterion-related validities for a wide range
of jobs.
In summary, the research findings demonstrate clearly the validity of cognitive ability tests
for selection purposes. However, there are two main issues to consider for future research.
First, while selection research is focusing on a unitary trait of general mental ability, research
into intelligence itself is moving away from this simple conceptualisation and developing
multifaceted models of intelligence (see Chapter 2). Second, of the research evidence available
there is a gap relating to the pace of advancements in technology and its impact on selection
methods. Evidence available suggests that applicants have more positive reactions to cognitive ability tests administered via the Internet than they do to ‘pencil-and-paper’ equivalents
(Khoshsima and Hashemi, 2017).
92
Work Psychology
Personality measures
The personality measures used in selection activities are usually based on a trait-factor analytic model of personality (described in Chapter 2). For more than two decades the status
of personality measures as predictors of performance was low. However, some authors have
argued persuasively that personality is an important determinant of behaviour at work
and there has been a resurgence of interest in personality assessment for the purposes of
selection.
Results from various meta-analytic studies suggest that of the Big Five Personality
model, conscientiousness is a valid predictor of performance across most jobs and organisational settings, with an average criterion-related validity of 0.23 (Van Aarde et al., 2017;
Leutner and Chamorro-Premuzic, 2018). Openness to experience tends to be positively
correlated with training performance across many job roles. Emotional stability has been
shown to be positively associated with job performance across many organisational settings. Extroversion is generally found to correlate positively with performance in jobs such
as sales, where performance is judged within an interpersonal environment (Leutner and
Chamorro-Premuzic, 2018). Researchers highlight that validity of personality variables vary
according to the type of performance being measured and job type. For example, conscientiousness correlates most highly with overall job performance, however, conscientiousness
may be negatively correlated with creativity. In this way, depending on the job, the facets
of conscientiousness are differentially important (Diekmann and Konig, 2015; Judge and
Zapata, 2015).
To provide a more comprehensive review of the role of personality in job performance,
researchers have begun to focus on lower-order traits (e.g. facets of Extraversion include
sociability and assertiveness) as opposed to the broad factors (such as the Big Five). Judge
et al. (2013) argue that weak overall relationships with job performance may be masking
significant relationships at the facet level of personality. Their meta-analysis revealed that
while the broad Big Five factors followed the typical pattern (i.e. conscientiousness, extraversion, and agreeableness being highly correlated with job performance, openness correlating moderately with job performance, and neuroticism having a low correlation with job
performance), things were more complex at the facet level. For example, achievement striving, dutifulness and self-discipline were highly correlated with job performance, but other
conscientiousness facets like deliberation were not. Similarly, intellect was highly correlated
with job performance, but the more aesthetic facets of openness did not correlate as highly;
and the excitement-seeking aspect of extraversion had a much lower correlation with job
performance than the positive emotions facet. This suggests that there is some basis for the
argument that the facets of the Big Five dimensions may differentially predict job performance, however analysis at the Big Five level may underestimate the potential for measure
of personality to predict performance.
Other research suggests that rather than looking deeper at the facet level, we should
perhaps be looking more broadly at the relationship between personality and job performance. It has been suggested that a general factor of personality (GFP) exists that sits at the
top of the hierarchical structure of personality, and in personality terms is comparable to
Spearman’s ‘g’ – the general intelligence factor. Empirical evidence in favour of the existence of this GFP has been growing (e.g. Van der Linden et al., 2016; Davies et al., 2015),
and for the first time, was linked to job performance by Van der Linden et al. (2010). Their
research showed that the Big Five factors could be categorised as two factors instead –
stability (or a), consisting of conscientiousness, agreeableness and neuroticism; and
plasticity (or b), consisting of openness and extraversion. These two factors could then
be categorised as one single factor – the GFP. Following this, Van der Linden et al. (2010)
found that the GFP had a significant relationship with performance indicators, namely
supervisor-rated job performance.
Chapter 3 Selection: Analysing jobs, competencies and selection methods
93
Research has also indicated that the relationships between personality factors and
job performance may not be straightforward. Curvilinear relationships have been found
between both conscientiousness and extraversion and task performance, organisational
citizenship behaviours, and counterproductive work behaviours (see Walmsley et al., 2018;
Wihler et al., 2017; Carter et al., 2018; Le et al., 2011). Earlier findings suggested that in
terms of the links between personality and performance, there can be ‘too much of a good
thing’. However, more recent research indicates that the extreme ends of personality scales
are unlikely to affect job performance, especially when balanced with other job factors.
As job performance is a multi-faceted construct, an extreme preference is likely to have
minimal effect when paired with a combination of other skills and traits. See Ferguson and
Lievens (2017) for an exploration of myths of personality within a selection context.
Key learning point
Conscientiousness tends to predict overall job performance over a range of occupational settings.
However, when exploring validity at the facet level, results vary according to the measure of work
performance used. For other personality factors, their relationship with job performance tends to
depend upon the nature of the job being carried out.
Whether focused on trait-specific personalities or GPF, there is concern that candidates
can fake their personality preferences for selection purposes (Sackett et al., 2017; Schermer et
al., 2019). Advances in technology and common access to the internet have made personality
assessment a valuable tool in the early phases of large recruitment schemes. Though personality research has found traits to be stable over time and maintain their levels of validity against
social desirability, computerised personality assessments have allowed for matured response
formats and profiling techniques to reduce fake-ability (Leutner and Chamorro-Premuzic,
2018). Examples of such techniques include adaptive sequential questioning and consistency
scores in response patterns. Such advances have created a cost-effective option for selecting
out candidates using a valid method earlier on in the selection process than ever before. However, with heightened use of personality measures on a global scale, researchers are adamant
that test publishers provide detailed information regarding how they norm such measures
to ensure issues of diversity are addressed as selection decisions have long-lasting effects on
individuals (Gaddis et al., 2015).
The digitally enhanced world and increased levels of screen time have elicited a general belief that candidates require highly engaging tests to elicit better applicant reactions.
To do so, traditional assessment methods are being transformed into gamified versions, or
game-based assessments (GBAs), where gaming elements such as story lines, levels or points
are combined with pre-existing psychometric properties to evaluate behaviourally based
constructs such as creativity or leadership. Gamified assessments have piqued the interest
of many HR departments, especially those with interest in graduate recruitment with the
belief that young, technology-savvy candidates require stronger levels of engagement than
in the past; now being dubbed ‘Recrutrainment’ (Korn et al., 2017). However, researchers
continue to evaluate the validity of such assessments in comparison to their traditional
counterparts as well as question if GBAs do in fact improve candidate reactions (Kato and
de Klerk, 2017; Armstrong et al., 2016). Some conclusions suggest elements of game-design
such as storylines can improve feelings of relevance if the context of the game relates to the
job tasks and foster levels of engagement through complex problem solving (Sailer et al.,
2017; Hummel et al., 2017).
94
Work Psychology
Stop to consider
Technology and Human Resources trends are advancing faster than work psychologists can research
the reliability and validity of new selection tools and methods. Using the reliability and validity tables
presented earlier on in this chapter, what initial reflections do you have on the validity of tools such as
Game-Based Assessments (GBAs)? For example, would candidates for an accountancy position consider
a GBA to be face valid if it contained pirates as characters in the game? What information would you
use when determining the validity of these tools?
Assessment centres
An assessment centre (AC) is a popular and effective method of recruitment consisting of a
multifaceted assessment process with large, well-resourced organisations. In a typical AC,
numerous applicants are invited together for between half a day and three days of assessment.
Assessment methods can include: work sample exercises, group exercises, presentations,
in-tray exercises, role plays, practical skills, interviews and psychometric tests; the particular
combination depends on the target job role. ACs profile an applicant’s ability across a range
of competencies and job-related contexts. The appeal of ACs lies in their generally good levels
of criterion-related validity and face validity (Hough and Oswald, 2000). Although there has
been much debate regarding the predictive validity of ACs, meta-analytic studies show the
average validity of ACs to be very good (Hoffman et al., 2015).
Key learning point
Assessment centres assess an applicant on multiple competencies using multiple job-related exercises
and multiple trained assessors.
The design of an AC reflects the need to assess the extent to which applicants can demonstrate a range of competencies. Consequently a series of exercises and assessment tools are
developed that: (i) are able to elicit the required behaviours; (ii) reflect the actual content
of the role; (iii) assess applicants’ performance in a variety of job-related situations; and (iv)
allow for different assessors to assess required behaviours over different exercises. As candidates are observed by several different assessors the impact of assessor bias is also reduced. In
most AC designs, a consensus discussion is held at the end of the session to discuss candidate
ratings with all assessors (International Taskforce on Assessment Center Guidelines, 2015).
However, more recent research is suggesting not to include such sessions in AC design as
these discussions can be unfocused or hurried (Dewberry and Jackson, 2016).
Typically, competencies are assessed multiple times within each AC, using different exercises. Each exercise assesses multiple competencies (sometimes referred to as dimensions in
the AC research literature). For instance, a group exercise may be assessing leadership skills
and communication skills. However, an exercise that assessed all the competencies would
overload assessors with too much information (Sackett and Tuzinski, 2001). Certain exercises
are best suited to assessing particular competencies. For example, a communication skills
competency can be better assessed in a presentation exercise and a group exercise than it can
be in a psychometric test. However, there is large debate around the use of competencies and
the impact of general performance in each exercise on overall assessment scores (Hoffman
et al., 2015). See Key debate, below.
Chapter 3 Selection: Analysing jobs, competencies and selection methods
Key debate
What do assessment centres measure?
Do ACs measure stable, underlying traits that influence performance across situations, or do they identify candidates’
ability to do well in the specific situations that are part of the AC? There is a substantial body of research debating
the traditional assumptions that dimensions (competencies) are central to how ACs work. Lance (2008) conducted
a review of 25 years of research into this so-called ‘construct validity problem’. This problem has plagued the AC
literature and is as follows. It has generally been assumed that assessors look for evidence that candidates can or
cannot demonstrate competence in each of the dimensions across the varying work-related situations in different
exercises. Unfortunately, a common finding is that performance ratings on individual AC component exercises often
reflect factors associated with the exercises, as opposed to the dimensions, or ‘constructs’ that the AC is designed to
measure. This relates to the problem of convergent validity versus discriminant validity: performance tends to be more
consistent across competencies within the same exercise, than it is within a competency across different exercises
(Jackson et al., 2016; Lee et al., 2017; Robertson and Smith, 2001).
The problem can be illustrated as follows. If dimensions / competencies are of central importance in assessor
decision-making, one would expect relatively high correlations between different assessors’ ratings of a candidate
for the same competency (say communication skills) across different exercises. This would mean that the applicant
demonstrates good communication skills in a presentation, in a group exercise and in an in-tray exercise. Naturally, they
may not score highly for all competencies within an exercise (they could communicate poor knowledge very clearly!).
Cross-situational consistency across exercises rather than within exercises indicates that the AC has discriminant
validity, i.e. that different competencies are evaluated separately in each exercise. Another meta-analysis of 24 studies
into the construct validity of ACs was conducted by Bowler and Woehr (2006), who found that AC exercises explain
more variance in candidate performance than do AC dimensions. Similar results have been found in more recent
research (e.g. Hoffman et al., 2015).
This poses a problem: are ACs measuring what they are supposed to measure (i.e. a dimension or competency that
drives performance across a range of situations)? If not, does it matter if they predict job performance? What other
mechanisms might contribute to the validity of ACs?
Some argue the tasks assessors must undertake when observing and rating a candidate’s behaviour are complex and demanding and in general it is easier to provide an overall evaluation of effectiveness in a particular
exercise. Arthur and colleagues (2008) responded to Lance (2008), suggesting that dimension-based AC construct validity should not be abandoned. Instead, a more representative sample of constructs must be taken to
represent the job content in that the constructs must be specific and well defined. They argue that the AC is theoretically sound and that the construct validity of AC dimensions may still stand up if the appropriate constructs
(i.e. competencies that are relevant to the job role) are measured in the AC, and they are defined in the appropriate way. This is an important point as it represents a possible explanation for the problems associated with
AC construct validity. Rather than viewing construct validity as problematic for ACs, there may be good reasons
why performance on the same dimension can differ between exercises. A person may be much more effective at
communicating on a one-to-one basis than presenting information in a public-speaking context: a competency/
dimension such as communication skills might be assessed in both contexts, with the same individual achieving
different scores in the two different contexts. It may be that defining the competency differently (e.g. defining
presentational communication skills and individual communication skills as two separate dimensions) might lead
to greater consistency across exercises.
Kuncel and Sackett (2014) proposed the argument that the construct validity problem may not actually be the
big deal it has been conceived to be over the past three decades! According to their research, we are forgetting
that competency scores for each exercise are only a step towards an overall final AC rating for each competency.
They argue that if we shift our focus to this overall AC competency rating, we will see that when ratings for each
competency across all of the AC exercises are combined, we can see from the data that exercise-specific effects
95
96
Work Psychology
are no longer the dominant source of final AC ratings. This may be because variations of competency performance and rating errors do not unduly influence the process of arriving at an overall evaluation for each
competency.
This debate is still heavily argued between work psychologists today. In recent discussions work psychologists suggest that competencies may act as the crucial framework to build the AC, however once built the
competencies are no longer needed as the AC is now able to, metaphorically, stand on its own. It is the key
competencies that help design the exercises to ensure the correct attributes are assessed and helps to guide
assessors in rating candidates, however it is the candidate’s performance in each exercise that matters – a
task-based approach to ACs.
Key learning point
When designed appropriately, assessment centres are valid predictors of job performance. Construct
validity is enhanced by ensuring the content is directly relevant to the target job role.
ACs are generally assumed to have good predictive validity because repeated assessment
is based upon direct observation of job-relevant behaviours. This enables assessors to predict
how candidates will behave in the job by observing them engaging in job-relevant behaviour.
Following best practice guidelines, ensuring strong training of assessors and using behavioural indicators in assessing, many researchers have found ACs to have strong predictive
and incremental validity (Meriac et al., 2008; Slivinski, 2008; Herd et al., 2015; Drasgow et
al., 2016). In a recent meta-analyses, Sackett and colleagues (2017) found ACs to demonstrate
stronger criterion-related validity than cognitive ability tests in 15 of the 17 comparisons of
studies involving candidate populations across the globe varying from internships to management positions.
The use of technology has influenced AC methodology and exercise content similar to
other selection methods. Audio and video recordings within exercises, computer-based simulations and assessment software platforms are now common-practice in most ACs. Despite
the name, some ACs are completed virtually using virtual assessment portals where candidates can communicate with each other, download materials and complete their tasks online
(Howland et al., 2015). Tippins (2015) argues ACs using technology demonstrate stronger levels of fidelity as the exercises are more realistic to the job, such as receiving emails to prioritise
and deal with during a work simulation exercise. Though it is argued that technology within
ACs can enhance candidate experiences, aid efficiency, and lower costs, further research is
essential to monitor the validity of these tools and the impact of technologically advanced
ACs on diversity (Kleinmann and Ingold, 2019).
Situational judgement tests
An example of an assessment method with lower physical fidelity (meaning a method
that does not directly simulate a work task) are situational judgement tests (SJTs). SJTs are
designed to assess an applicant’s judgement regarding situations where hypothetical
work-based scenarios are presented to applicants, who make judgements about the possible responses presented, and these responses are then assessed against a predetermined
scoring key (Rockstuhl et al., 2015; Krumm et al., 2015; Lievens et al., 2008). SJT scenarios
Chapter 3 Selection: Analysing jobs, competencies and selection methods
97
reflect work-related situations that candidates are likely to face in the job, assessing key
attributes associated with successful performance in the role. Research shows that SJTs
can be used reliably in selection to identify a range of non-academic or professional
attributes such as empathy, teamworking and integrity (Patterson et al., 2015; 2016;
Wyatt et al., 2010).
Applicants can be required to make their judgements in the format of choosing the best
option, worst option, or asked to rate the effectiveness of each suggested response (see Exercise 3.4 for an example). Their responses are typically scored by comparing them to the
judgements of experts in the respective field (Oostrom et al., 2015). Some similarities can be
seen between SJTs and situational interviews, assessment centre exercises and work samples,
however there are key differences in terms of presentation (usually because SJTs are written),
how options are presented, how options are scored (i.e. SJT predetermined scoring from a set
of pre-determined options), and the type of responses given.
The popularity of SJTs is seen most in large-scale selection, often at the shortlisting stage
(Patterson et al., 2016). SJTs are relatively easy and cost-effective to develop, administer and
score. SJTs are also versatile in their presentation; written or video-based. However, although
SJTs have become popular for practical reasons, their construct validity has remained elusive:
here remains uncertainty over how SJTs work and what they measure. Recent research has
attempted to classify SJT constructs concluding that the use of cognitive diagnosis modelling
can help clarify how variables relate to each other in SJT scoring and suggesting response
options should be written in line with a specific construct varying in degree of correctness
(see Guenole et al., 2017; Sorrel et al., 2016). That said, SJTs are a measurement method for
job-relevant attributes that reflect complex situations and events, are tailored to the particular
context, and can be designed to measure a variety of cognitive and non-academic constructs.
This is important because it means that their content and format can be amended to fit the
job and specification of the test.
Correlations between SJTs and personality are to be expected, due to the types of situations included in SJTs (e.g. many situations include interpersonal aspects of situations and
tap into work preferences/styles). McDaniel and Nguyen (2001) conducted a meta-analysis
and found that SJT scores are significantly related to the Big Five personality traits (0.31
between SJTs and Emotional stability; 0.26 for Conscientiousness; 0.25 for Agreeableness).
Other researchers have found similar but not entirely consistent patterns where SJTs are
measuring personality-related knowledge, or personality-based behaviours and habits
(Olaru et al., 2019). This is likely to reflect the wide variation in content of the SJTs and the
range of contexts in which they are used. Meta analyses have identified modest correlations
between SJTs and cognitive ability tests of between 0.34 and 0.42, although they note that
this masks a wide variation of results (Lievens and De Soete, 2015). Important factors that
influence this correlation include the format (video-based versus written scenarios), the
methodology of the SJT writing process (whether a job analysis was performed) and the
level of detail in the questions and the type of response instructions (knowledge or behavioural tendency related questions). In summary it appears that SJTs are not just measuring
cognitive ability.
With regards to predictive validity, meta-analyses show moderate to good criterion-related
validity of SJTs. Research evidence suggests that SJTs can predict job performance across a
range of different occupations (Gardner and Dunkin, 2018; Lieviens and De Soete, 2015;
Patterson et al., 2017; Stemler et al., 2016; Wyatt et al., 2010). Importantly, SJTs have incremental validity over ability and personality measures (Lieves and De Soete, 2015). SJT validity
is often a debate derived from the fact SJTs are a tool that can be used quite broadly and come
in a variety of forms: validity is strong for specific SJTs in some selection processes but less
so for other SJTs in different selection processes. Some reasons behind differences in validity
mentioned previously highlight the importance of following proper design processes and
methods when constructing an SJT.
98
Work Psychology
Designing an effective SJT
Exercise 3.3
You are a psychologist working in a firm that specialises in Situational Judgement Tests. You are
approached by a local hospital asking you to develop an SJT to recruit junior doctors. The situations
you create in your SJT need to present challenging, but realistic situations to the medical profession
that measure non-academic attributes.
An example question using a ranking answer response format is:
You are reviewing a routine drug chart for a patient with rheumatoid arthritis during an overnight
shift. You notice that your consultant has inappropriately prescribed a drug called methotrexate
7.5mg daily instead of weekly.
Rank in order the following actions in response to this situation (1 = Most appropriate; 5 = Least
appropriate)
A
B
C
D
E
Ask another doctor if the consultant has made any other drug errors recently
Correct the prescription to 7.5mg weekly
Leave the prescription unchanged until the consultant ward round the following morning
Phone the consultant at home to ask about changing the prescription
Inform the patient of the error
Suggested exercises
1
What non-academic attributes would you want to measure in the SJT? Why are these attributes
important to being a doctor?
2
Try to create another scenario that would be relevant for doctors. What do you find difficult in developing the question? What knowledge and skills would you need to write a good SJT question?
3
What processes would you follow to ensure the SJT is free from bias and valid? Who might you ask to
help develop SJT scenarios? Who might you ask to review the scenarios?
In terms of fairness and adverse impact, findings show there are generally smaller subgroup
differences (e.g. ethnicity) when using SJTs than when using cognitive ability tests (De Leng
et al., 2017; Lievens et al., 2016; Jensen, 1998, as cited in Lievens et al., 2008). Research also
shows that fairness and adverse impact may vary with the way that constructs are measured,
for example it may vary in accordance with the presentation format. Technologies advancements in video-streaming has led to interest of the impact of video-based SJTs on fairness (see
Naemi et al., 2016; Podsakoff et al., 2013). Research has found that video-based SJTs show
lower levels of cognitive loading as candidates are no longer required to read large amounts
of text (Oostrom et al., 2015). Though now easily streamed over the internet, video-based SJTs
may become more costly to design because of the need for production and / or animation
expertise. Another consideration regarding fairness of the SJT is the use of the tool in multiple
countries where scenarios would likely need to be revised to avoid cultural differences and
language issues (Weekley et al., 2015).
One area that SJTs have been particularly useful in is values-based recruitment. You will
see in the closing case study that hiring candidates whose values are aligned to those of the
organisation can be very important for certain roles. Since values are non-academic factors
that are socially constructed, they are difficult to assess with traditional selection methods.
For example, panel interviews are often criticised for their potential to be biased or for their
lack of standardisation, while personality tests offer lower face validity and are less acceptable
to candidates as a selection tool. SJTs are a standardised way of more objectively assessing a
broad range of attributes for large numbers of applicants, while being face valid to candidates since the scenarios used in SJTs are based on job-relevant situations. Indeed, there is
Chapter 3 Selection: Analysing jobs, competencies and selection methods
99
good evidence to show that SJTs are a useful methodology to evaluate a range of professional
attributes for selection into fields such as medicine and dentistry (see Patterson et al. 2015 for
a review). Long-term follow-up studies have shown an SJT measuring empathy, integrity and
resilience (used to select candidates applying for training in UK general practice) to be the
best single predictor of subsequent job performance compared with other selection methods
(Lievens and Patterson, 2011; Patterson et al., 2018b). This is likely due to SJTs being constructed differently, adapting the format depending on the requirement of the role. SJTs are
designed to assess an applicant’s judgement regarding situations encountered in the workplace, targeting professional attributes (e.g. teamworking).
Key learning point
Situational judgement tests show evidence of validity as a selection method. Applicant reactions tend
to be positive. Further research is needed to explore the construct validity of SJTs.
References
References are widely used methods for obtaining information on candidates, although
it seems likely that in many situations potential employers take up references only when
they are about to make a job offer. In general, the validity evidence for references is not
particularly good and there has been no analysis of whether references add any incremental
validity over other methods. With more heightened concern for litigation in many countries, references are now more formal, simply stating if the information presented in a candidate’s CV is true – such as how long they worked for a company or their pay. With the
use of social media platforms, organisations now consider a candidate’s internet presence
via sites intended for professional or non-professional use. Some believe such sites provide a stronger insight into a candidate’s persona than references or tailored CVs. However,
Chamorro-Premuzic and colleagues (2016) found online profiles are no more accurate than
their traditional counterparts as candidates alter such profiles to ensure a certain image,
which can be tailored to the job.
Stop to consider
In the preceding sections we have considered the properties of a range of different selection methods. What do the best selection methods have in common? What do the less effective methods have
in common? Is good selection just about the predictive validity of a measure? If not, what other factors
contribute to the effectiveness of a selection method?
The impact of selection procedures on applicants
Fairness
Until recently, there has been relatively little work in selection which has looked at the issues
involved from the perspective of candidates. The main area of work concerns the extent
to which selection procedures are fair to different subgroups (such as ethnic minorities or
women) of the population. This issue has stimulated a large amount of research. A variety of
100
Work Psychology
terms such as bias, adverse impact, fairness and differential validity are used in the literature on this issue, and a clear grasp of the meanings and definitions of some of these terms is
crucial to an understanding of the research results.
First, it needs to be made clear that a test is not unfair or biased simply because members of different subgroups obtain different average scores on the tests. Men and women
have different mean scores for height; this does not mean that rulers are unfair measuring
instruments. However, it would be unfair to use height as a selection criterion for a job, if
the job could be done by people of any height: it is important for selection criteria to be
job-related.
Unfortunately, it is possible for tests to appear to be valid and yet be biased against some
subgroups. This may happen if the relationship between the test score and job performance
is not the same for the two subgroups. For example, it is possible to imagine that the link
between job performance and certain personality or ability factors could be different for
two subgroups of the population (e.g. the personality factor predicts performance for men
but not for women). A validity study based on a mixed sample of people would produce
results that were somewhat incorrect for both subgroups; if the results were used to develop
a selection procedure the predictions of candidates’ work performance would be in error. For
example, an international study found immigrant populations tend to score half a standard
deviation lower than non-immigrant populations on General Mental Ability tests, however
this difference did not then translate into differences in work performance (Robie et al.,
2017). Unfair direct discrimination is where the selection process results in the less favourable treatment of candidates because of their gender or ethnic group, for example. Indirect
discrimination, however, is usually unintended and difficult to prove. It occurs when an
employer applies a requirement for applicants (e.g. score on a test) which one group (defined
by gender/race/socio-economic background, etc.) finds it considerably harder to comply
with, i.e. a larger proportion of one group cannot meet this requirement. This is known as
adverse impact.
Key learning point
Indirect discrimination occurs when an employer applies a requirement for applicants (e.g. score on a
test) which one group (defined by demographic variables) finds it considerably harder to comply with.
A procedure is biased or unfair when it shows different validity for different groups.
In using a selection test, the test designers would need to ensure that the difficulty level
of items (percentage of applicants getting the item right) is not significantly different
across ethnic groups sitting the test. If differences are found, then there is said to be differential item functioning. There are sophisticated statistical procedures that can be applied
to assess differential item functioning for tests. In general, the research evidence suggests
that for professionally developed selection tests there is little or no evidence of differential
item functioning across ethnic groups. The fact that the scientific research provides little
evidence of differential validity for well-established selection procedures does not imply
that all selection methods are unbiased, nor does it imply that unfair discrimination does
not take place. For example, cognitive ability testing is the method that has created the
most frequent problems with adverse impact in selection processes. Research has shown
that there are significant ethnic subgroup differences in validity of cognitive ability tests,
with validity favouring White candidates over other ethnic minority candidates (Berry
et al., 2014; Dahlke and Sackett, 2017; Hausdorf and Robie, 2018). In other words, there is
a stronger connection between tests scores and performance in White candidates than in
ethnic minority candidates. Similar research has shown significant differences in validity
Chapter 3 Selection: Analysing jobs, competencies and selection methods
101
of cognitive ability tests also exist between candidates of different ages (Fisher et al., 2017;
Klein et al., 2015).
Other selection methods such as SJTs have also shown certain subgroup differences.
De Leng and colleagues (2017) found that Whites scored better than ethnic minorities,
with differing effect sizes dependent on the method of scoring. Similar findings were found
for gender where females tended to score better than males and older individuals tended
to score higher than younger individuals in some domains (Schripsema et al., 2017). Suggestions for mitigating potential adverse impact in SJTs often includes decreasing the
amount of cognitively-loaded questions in the tests (see Lievens and De Soete, 2015;
Patterson et al., 2017).
ACs make use of several selection methods and this may minimise the impact of bias
in one component. Most literature assumes low subgroup differences effects in ACs.
However, Buckett and colleagues (2017) found stronger ethnic group differences when
tasks involved written content and analytical, or problem-solving skills (which are more
cognitively-orientated). Regarding gender differences in ACs some evidence suggests a negative subgroup difference between female and male individuals, with females tending to
receive higher performance ratings in ACs (Dean et al., 2008). Such findings can be taken
alongside other studies of diversity and fairness to illustrate the mixed picture of evidence
that exists regarding potential discrimination arising from the use of ACs. It is also worth
remembering that candidates taking part in an AC are a small proportion of the general
population, that is they have chosen to apply for the job and have been selected for the
AC based on the information provided in their application form (and perhaps some other
pre-AC assessments). Therefore, it might be that some of the effects observed at ACs are a
result of events that occur before the AC. For example, it may be that only very able female
candidates apply for the job and make it through the pre-AC assessments (that may be biased
towards males). In contrast, males of all levels of ability may apply for the job and some of
the pre-AC assessments may favour males, thus resulting in males from across the ability
spectrum being assessed at the AC.
Increasing usage of online selection tools magnifies the critical question of equivalence
and legal concerns are of importance. Oostrom et al. (2013) argue that recruiter characteristics
(such as personality and computer self-efficacy) may affect whether new technology is incorporated into selection practices. They found that recruiter characteristics (e.g. personality)
were significantly related to perceptions of the usefulness and ease of use of technology, and
in turn, this related to intentions to use new technologies. Recruiter characteristics were also
found to be more important in determining whether recruiters will use new technologies
than perceived face and predictive validity and fairness.
Research investigating the equivalence across different selection methods has produced
mixed results. With the wide acceptance of technology and consistent use of internet platforms across the globe, test publishers face new challenges when creating valid and reliable
psychometric tools. Due to the use of laptop computers, tablets and mobile devices, psychometric tools must be valid and reliable across the varying screen dimensions. In fact, there
is an increasing demand for selection tools to be tailored for small screens and quick completion. Research has found small differences in performance linked to screen size and the
amount of time to complete the test, reinforcing the message that test developers must ensure
they take this into consideration for future test development (Brown and Grossenbacher,
2017; Dages et al., 2017).
As you may imagine, there are potential dangers involved with the use of ‘unproctored
(unsupervised) Internet testing’. Some methods to combat this issue include video monitoring where the candidate is recorded from their webcam whilst completing the assessments.
Retesting can also be used: candidates can be asked to complete the assessments again in person at a controlled centre in which their results are then compared with their original scores
in search for dramatic differences. It is suggested that unproctored internet testing be part of
the first selection procedures within a larger recruitment scheme, especially for recruitment
schemes with large applicant pools, such as graduate schemes.
102
Work Psychology
Adverse impact in selection
Exercise 3.4
A large organisation was found to have discriminated against ethnic minority applicants after an
employment tribunal. Following a restructuring of the organisation, 100 new middle management
posts were created. Approximately 30 per cent of employees are of ethnic minority origin, and currently, 3 per cent of management are of ethnic minority origin. The HR department decided to use
a structured interview and two cognitive ability tests (a measure of verbal reasoning and a measure
of numerical reasoning).
Of the 600 employees who applied for management posts, 30 per cent were ethnic minority applicants. However, only 10 per cent of the job offers were made to ethnic candidates. The tribunal ruled
that the tests were inappropriate in terms of the time allowed, the level of difficulty, the skills tested
and the content covered. There was evidence of adverse impact and unlawful discrimination. The
organisation conceded there was unintentional and indirect discriminatory impact on ethnic minority
candidates and suggested compensation.
Suggested exercises
1
What is the difference between fair and unfair selection testing?
2
How can unfair discrimination be recognised?
3
What needs should be considered in evaluating the fairness of the test?
Applicant reactions
There has been more emphasis placed on applicants’ reactions as candidates may choose
to work for a competitor or divulge negative reactions on social media platforms, harming
the organisation’s brand and revenue (Lievens and Slaughter, 2016). In an extreme case, an
applicant who has a negative reaction to a selection procedure could make a legal challenge
based on unfair discrimination. Of course, perceptions of injustice can depend partly on
whether the person is successful in the selection process. However, a recent study found applicants who perceived the selection process as fair were more likely to view the organisation
as an attractive employer and recommend it to others, regardless of the selection outcome
(Schinkel et al., 2016).
Research has tended to explain the different factors that affect applicant reactions using
theories of organisational justice. Distributive justice focuses on perceived fairness regarding equity (where the selection outcome is consistent with the applicant’s expectation) and
equality (the extent to which applicants have the same opportunities in the selection process). Procedural justice refers to the formal characteristics of the selection process such as
the level and quality of information and feedback offered, job-relatedness of the procedures
and methods and recruiter effectiveness.
Point of integration
Procedural and distributive justice have their origin in research into employee relations and worker
motivation. They have provided a useful framework for investigating the factors that influence worker
satisfaction with a range of human resource management processes and interventions (see Chapter 6,
specifically Table 6.1).
Chapter 3 Selection: Analysing jobs, competencies and selection methods
103
An early review of these theories by Anderson et al. (2001a) and Gilliland (1993) remain
relevant today in which four main factors seem to account for positive or negative applicant
reactions to selection methods. These are: (i) the selection appears job-relevant; (ii) the selection method is less personally intrusive; (iii) the method does not contravene procedural or
distributive justice expectations; and (iv) the method allows applicants to meet the recruiters.
Other literature suggests that applicants prefer multiple opportunities to demonstrate their
skills, that they prefer assessment that allows them to talk about their potential (not just their
past), and that the selection system is administered consistently for all applicants. Table 3.8
outlines the procedural justice rules related to applicant reactions of selection methods. A
more recent review by McCarthey and colleagues (2017) explores the impact of technology
on how work psychologists assess applicant reactions. They discuss other theories such as
expectations theory, fairness heuristic theory, and attribution theory that are commonly used
in applicant reactions research suggesting that a conceptual model incorporating all theories
will help to understand individual differences to determine why applicants have reacted in
a selection scenario.
Many studies have used a measure originally developed by Steiner and Gilliland in 1996 to
assess applicant reactions to ten widely used selection methods. The consensus, on a global
scale, is that interviews, CVs and work-sample tests are among the most favourable methods
(Zibarras, 2018). In an age in which many companies operate internationally and must extend
their selection systems beyond their home countries it has become increasingly important
Table 3.8
Procedural justice rules underlying reactions to selection processes
Rule
Description
1. Job relatedness
The extent to which selection methods appear to measure content that is relevant to
the job role or appears to be valid
2. Opportunity to perform
The extent to which candidates can demonstrate their knowledge, skills and abilities
3. Reconsideration opportunity
The opportunity to review test results or challenge scores, or to be able to re-test
4. Administration consistency
The degree to which selection processes are consistent or standardised across people
and over time
5. Feedback
Provision of feedback that is timely and informative regarding test results and selection
decision
6. Selection information
Information on, and justification for, the use of selection methods and decisions made
7. Honesty
The extent to which communication with candidates is candid
8. Interpersonal effectiveness
The extent to which candidates are treated with respect during the selection process
9. Two-way communication
The opportunity for candidates to ask questions during interpersonal interaction
throughout the selection process
10. Propriety of questions
The degree to which the questions asked of candidates during the selection process
are appropriate
104
Work Psychology
to investigate the hypothesis that applicant reactions may vary as a result of cross-cultural
differences. In a review, Nikolaou and colleagues (2015), found cross-cultural variations in
applicant reactions are minimal suggesting the reason behind favouring interviews and work
sample tests is due to the opportunity to perform, the job-relatedness and the inter-personal
opportunity. These results suggest there may be a core component in applicant reactions that
is stable across cultures and countries.
This international stability allows companies to extend their selection practices across
countries without a differing impact on applicant reactions. It is important to acknowledge,
however, that research on applicant reactions tends not to be conducted during live selection
processes, and because of the lack of studies focusing on important organisational outcomes
(e.g. applicants actually choosing to withdraw from selection processes, or refusing to take
the job offer), there are some weaknesses in this field of research, especially when it is applied
to real selection settings.
With globalisation and technology advancements, classic methods such as interviews or
CVs have been revised into new formats, such as video interviews or asynchronous interviews, and websites such as LinkedIn. Due to the nature of these new selection tools, the loss
of interpersonal warmth may hinder the receptiveness of these methods. A meta-analysis of
recent literature has found applicant reactions to be less favourable for technology-mediated
interviews due to feelings of frustration due to restrictions on opportunity for impression
management (Blacksmith et al., 2016). More specifically, asynchronous interviews showed
stronger negative reactions due to the lack of two-way communication and concerns for
privacy and transparency about what is being assessed as the information is recorded and
assessed later (Langer et al., 2017). Other research has found positive applicant reactions to
asynchronous interviews in line with other non-interactive selection methods as long as a
face-to-face interview is used further on in the selection process (Brenner et al., 2016; Guchait
et al., 2014; Zibarras et al., 2018). The use of social media platforms in selection seem to
have mixed reactions (Chamorro-Premuzic et al., 2016) dependent on age and gender, with
those younger and male being more in favour of social media use (Aguado et al., 2016). Overall, applicants are more accepting of vetting professional platforms such as LinkedIn over
non-professional platforms such as Facebook. On the other hand, applicants have shown
positive reactions to recruitment advertisements and brand interactivity on social media platforms (Frasca and Edwards, 2017).
Konradt et al. (2013) suggest that the very nature of technologically-based selection means
that applicant reactions towards these methods may be influenced in a different way than
traditional methods of selection. For example, personal contact is reserved for the latter stages
of selection; online tests are highly standardised, which could positively influence applicants’
perspective of their opportunity to perform or on job-relatedness; and applicants generally
use their own equipment when completing online exercises, meaning that the experience
may not be the same for every candidate due to differences in their environment (including
perceptions of unfair Internet speed). Some argue that candidates are in favour of web-based
tools that they can complete at their discretion without having to take time off work or travel,
which is also of financial benefit to the organisation hiring.
Key learning point
Organisational justice theories are used to understand applicant reactions to selection methods. Applicant reactions to selection methods are generally stable across countries. Technology has a large
impact on selection methods that can aid applicant reactions to selection methods if used wisely.
Chapter 3 Selection: Analysing jobs, competencies and selection methods
105
Stop to consider
An important factor of applicant reactions is the opportunity for feedback. As technology has led to
the automation of much of the early stages of recruitment, feedback can now be provided both quickly
and easily. However, the feedback provided must be meaningful to elicit positive applicant reactions
(Patterson et al., 2018a). Feedback provided must be thorough, easy to understand, and allow for further communication with the candidate in case of questions. Feedback must not only be provided to
those who succeed, but also to those who do not succeed at all stages of the selection process. Though
technology can be an extremely useful HR tool that creates an efficient selection process, it can lead
to an impersonal and transactional approach for candidates (Stone et al., 2015).
One of the most cited meta-analysis studies is by Schmidt and Hunter (1998), who based their
results on 85 years of research in selection. This study separated the results of selection studies
into two categories: (i) training performance used as criteria and (ii) overall job performance
used as criteria to examine employee performance. The results showed that the predictive
validity for each selection method is broadly similar in predicting both training and overall
job performance criteria. The significant research evidence on each of the techniques listed in
Table 3.9 is discussed next in this chapter. Also included is an estimate of the extent of usage,
and likely applicant reaction to, each technique.
Stop to consider
You will see within this chapter that the best selection methods are not always the most widely used.
Also, there are some quite weak methods that are widely used. Why do you think this might be? Think
about how different stakeholders in the selection process might view the various selection methods
(e.g. might there be some that managers prefer to use?). Tables 3.3 and 3.9 might also help you with
this task. Is it always practical or necessary to use the selection methods with the highest validity – and
if not, why not?
Table 3.9 outlines a summary on the validity of selection procedures. We can see from
Table 3.9 that methods like references have typically been found to have low criterion-related
validity, yet they are highly used. There continues to be a significant scientist-practitioner gap
when it comes to the choices organisations are making regarding their selection procedures.
Despite a wealth of research on the utility of different selection methods, organisations tend
to use methods with low predictive validity (Schuler and Jackson, 2007). But why is this? Ryan
et al. (2015) are some of only a handful of researchers that have begun to answer this question
(see also König et al., 2010). In a global survey, Ryan et al. (2015) found that decision-makers
in organisations consider the cost and commonality of test-use within the field when selecting a test over the tool’s legal defensibility or validity.
A recent examination into the legal considerations around selection amongst 17 countries found a few countries in which choices of selection methods are affected by strong legal
requirements (e.g. the USA, South Africa), whereas organisations operating in other countries (e.g. Switzerland) are not bound by such legal pressures (Shen et al., 2017). In line with
Work Psychology
106
Table 3.9
A summary of studies on the validity of selection procedures (Patterson et al., 2016)
Selection method
Reliability
Validity
Candidate acceptability
Promotes diversity
Academic records
High
High
High
Low
Structured interviews
Moderate to high
Moderate to high
High
Moderate
Situational judgement
tests
High
High
Moderate to high
High
Aptitude testing
High
Various
Moderate
Moderate
Personality tests
High
Moderate
Low to moderate
N/A
Unstructured interviews
Low
Low
High
Low
Personal statements
Low
Low
High
Low
References
Low
Low
High
Low
*Note that there are international differences in the extent of usage for various techniques. For example, assessment centres are used
more frequently in the UK than in the USA.
previous research, concerns around fairness and discrimination remains pertinent in many
countries worldwide. An emerging concern in most countries surrounds the use of new technologies and social media in selection and how this impacts citizens’ privacy rights.
Stop to consider
Ryan et al. (2015) found some similarities and some differences to the findings of Köing et al. (2010).
Why do you think these changes might have occurred in the space of five years? How might the topical concerns of today and the influences of technology impact these findings further? For example,
how might today’s global interests and advancements affect considerations such as legal defensibility,
cost, applicant reactions, organisational promotion and predictive validity? How do you believe these
considerations could be influenced in the future? How might organisational considerations differ from
what applicants feel are important in selection methods?
Summary
Selection processes are designed on the assumption that there are job-relevant, individual
differences between people, which can be assessed. The job analysis defines the key behaviours that underpin the successful performance in the target job role and set the basis for
the assessment criteria.
Chapter 3 Selection: Analysing jobs, competencies and selection methods
107
Various selection procedures are available for use in organisational settings. Research
over the past decades has provided a much clearer picture of the criterion-related validity of
these procedures. Some of the methods, such as cognitive ability tests, seem to have broad
applicability across a range of situations. As well as examining the validity of selection
procedures, research has also concentrated on the impact of the procedures on candidates.
One area that has been reasonably well researched involves an examination of the fairness
(to different subgroups) of the various techniques.
Selection methods and processes are adapting rapidly to the technological advancements readily available. However, researchers are struggling to keep pace. Though tools
such as asynchronous interviews and Game-Based Assessments are available on the market,
work psychologists must continue to determine the reliability and validity of these tools
as well as the impact on fairness and applicant reactions.
Closing case study
Recruiting for values: the value of values-based recruitment
The notion of values-based recruitment – that is, assessing and selecting suitable candidates based
on the appropriateness of their values for the job for which they have applied – has gained both
research and practical interest in recent years. Values are guiding principles about how individuals
ought to behave; they govern one’s choices and are predictive of behaviour. But how valuable are
values to selection? Research suggests that they can be a useful part of a selection process, particularly in an era where success at work is increasingly based on factors beyond ability alone (although
a bit helps, of course!).
One sector that has recently drawn public attention to the importance of values at work is the healthcare sector (see Patterson and Zibarras, 2018). It is widely agreed that irrespective of job suitability
‘on paper’, a set of values including compassion, empathy, respect and dignity are also important for
all healthcare professionals; these have been linked with high-quality care and favourable patient
outcomes, and are held in the National Health Service (NHS) constitution in the UK. The absence of
these values, however, can have disastrous consequences.
The Francis Report (2013) reported findings from an inquiry into a failing UK NHS hospital trust.
Shockingly, the inquiry uncovered a severe lack of compassion, care and committed nursing that
resulted in the appalling suffering of many of the hospital’s patients. The same year, the Cavendish
Review was published, which focused heavily on the importance of making sure healthcare workers
have core values in place that ensure the safety of patients and the success of workers (Cavendish,
2013). In fact, the key recommendation was that employers must define a thread of values common
to all healthcare professionals, and this must be considered a core attribute in recruitment if cases
like that of the Francis inquiry are to be avoided.
Other hospitals and healthcare organisations have thankfully been more successful. The Cavendish Review revealed that a core theme across top-performing healthcare providers was an
emphasis on values in addition to traditional selection measures. For patients, values-directed
care provides that ‘something extra’ over and above good job performance; for organisations
in the UK and the USA, lower staff turnover and sickness absence, and higher performance and
quality of patient care are just some of the many potential benefits of well-designed and executed
values-based recruitment.
108
Work Psychology
But how do we assess values? Recent research has examined the utility of several selection methods for assessing non-academic attributes, including interviews, CVs, personal statements and
situational judgement tests (SJTs). Generally, structured interviews and SJTs are effective for
values-based recruitment, while personal statements and CVs are less effective. SJTs stand out as
being a good addition to any selection process that wishes to assess values, as they have been found
to have high face, content and predictive validity, even over methods like personality tests. SJTs assess
applicants’ judgements about various work-related scenarios by asking them to identify appropriate
responses to several written or video-based scenarios. The advantage is that SJTs can be extremely
versatile: they can be constructed using different formats and can capture several different domains
dependent on the requirements of the role.
Clearly, in the healthcare sector at least, there is definite ‘value’ (pardon the pun!) in values-based
recruitment. But this is not unique to healthcare recruitment – recruiting based simply on academic
ability or test scores is becoming a thing of the past, while non-academic attributes are becoming
increasingly important. Values-based recruitment can provide a reliable and valid indication of the
underlying motivations behind candidates’ behavioural choices and ensures that they are in line with
those desirable for the organisations to which they apply.
Is it more important to assess values than any other aspect during recruitment? Perhaps not. But
are they an important addition to a selection process that is predictive of both organisational and
individual success? Most definitely.
Sources: Francis, R. (2013) Report of the Mid Staffordshire NHS Foundation Trust Public Inquiry:
Executive Summary (vol. 947).
Edwards, H., Kinirons, T. and Patterson, F. (2018). Values-Based Recruitment. In Selection and
Recruitment in the Healthcare Professions (pp. 275–297). Palgrave Macmillan, Cham.
Suggested discussion points
1 Outline the key stages in the selection process. Discuss the practical problems associated with validating selection procedures in organisations.
■ Pre-assessment: application forms, personality tests, SJTs
■ Video interviews/structured interviews/unstructured interviews
■ Assessment centres
■ Issues with reliability and validity of technology-based selection methods –
under-researched area
■ Validation studies are rarely conducted because they can be difficult to do well,
time-consuming and costly, i.e. validation studies track the performance of new
recruits over several months/years – what impact might this have on the validity of
the selection methods proposed?
■ Issue that validation studies are only used in large-scale recruitment programmes,
however a lot of organisations do not recruit large numbers of people into one specific job role
2 Critically review the validity evidence for three of the following contemporary selection methods: personality assessment, tests of cognitive ability, interviews, situational
judgement tests, assessment centres.
■ Examples include:
i
Unstructured interviews have demonstrated low reliability and validity yet high
candidate acceptability versus structured interviews that are much higher across
these domains
Chapter 3 Selection: Analysing jobs, competencies and selection methods
109
ii
Personality tests have high reliability, moderate validity and low to moderate
candidate acceptability
iii Cognitive ability tests have received high reliability, mixed validity scores and
moderate candidate acceptability
iv Personality tests have little/no adverse impact and high reliability; however,
validity scores have been varied in predicting job performance
3 Discuss the advantages and disadvantages of using new technology for selection and
recruitment.
■ Lower costs associated with technology-based selection methods
■ Aid efficiency of selection process – usefulness and ease of use of technology, i.e.
completing tests from own tablets in a location convenient/comfortable for the
candidate
■ Candidate reactions/experience: feedback may lack meaning and insight, negative
reactions due to impersonal nature of technology
■ Technology may enhance candidate experiences especially individuals that are
‘tech-savvy’
4 What do situational judgement tests measure? What are their advantages and disadvantages compared to other selection methods?
■ Consider the wealth of validity studies that support SJT use in selection (i.e incremental validity over interviews)
■ Can be conducted online and thus saves time and costs associated with paper-based
traditional versions
■ SJTs are popular and used widely in numerous settings
■ Relatively easy to develop – content of the SJT can be amended and tailored to
the particular context and job specification assessing a wide range of job-relevant
characteristics
■ Promotes diversity
■ Construct validity has remained elusive – how do SJTs work and what do they actually measure?
5 Given the increased accessibility to public opinion via social platforms, discuss the
implications of applicant reactions on selection methods. How might work psychologists support organisations throughout the selection process?
■ Offering feedback
■ Fair assessments
Relevant websites
The British Psychological Society has published a set of Standards that describe best practice
in the use of assessment centres: https://tinyurl.com/y2jt43uq
The International Test Commission have published guidelines that describe best practice
when performing large scale assessments across culturally diverse populations: https://www.
intestcom.org/files/guideline_diverse_populations.pdf
A guide to the application of selection methods published by the Society for Human Resource
Management (SHRM) can be found at https://www.shrm.org/hr-today/trends-and-forecasting/special-reports-and-expert-views/documents/selection-assessment-methods.pdf
110
Work Psychology
Suggested further reading
Full details for all references are given in the list of references at the end of this book.
1 For more depth on the issues raised in this chapter, consult a comprehensive text on
selection, such as Evers et al.’s The Blackwell Handbook of Selection (Wiley-Blackwell
Publishing, 2019) or Cook’s Selection: Adding value through people – A Changing Picture
(Wiley-Blackwell Publishing, 2016)
2 A very detailed description of validity is provided in Chapter 6 of the Handbook of Personnel Assessment and Selection written by Sackett and colleagues (2012)
3 An excellent review of selection over the last 100 years within the Journal of Applied
Psychology can be found in Ployhart et al.’s (2017) article ‘Solving the Supreme Problem:
100 years of selection and recruitment at the Journal of Applied Psychology’.
4 Tippins’ (2015) ‘Technology and assessment in selection’. This article provides a review
of the impact of technology on the selection process. It provides a strong discussion regarding the pros and cons of using technology within selection that Work Psychologists
should consider in practice.
5 Schmidt and Hunter’s (1998) review article explores the validity and utility of selection
methods over 85 years of research. It analyses a huge amount of data to examine the
links between performance during selection and in the job itself.
6 Ryan and Ployhart’s (2014) ‘A century of selection’ presents a discussion of recent advances in designing, implementing and evaluating selection systems.
7 Sanghi’s (2016) handbook of competency mapping provides an in-depth understanding
of the use of competency models within organisations.
CHAPTER 4
Assessing
performance
at work
LEARNING OBJECTIVES
After studying this chapter, you should be able to:
1 recognise the different reasons for measuring performance;
2 understand the importance of defining work performance;
3 identify how bias can influence interpersonal judgements when assessing performance;
4 understand the qualities of different performance rating scales;
5 describe the properties of multi-source feedback systems and understand the strengths
and weaknesses of these systems;
6 recognise how performance
assessment can be subject to
competing political goals;
7 understand how technology
can be used to collect information on performance;
8 identify how to improve performance based on assessment data;
9 be aware of current chal-
lenges and future directions
in performance assessment
research.
112
Work Psychology
Opening case study
Performance assessment in a VUCA environment
Deloitte, one of the largest professional services firms in the world, recently calculated that their
managers and employees were spending a whopping 2 million hours a year on performance management, including setting objectives, completing forms, creating ratings and holding meetings to
discuss and compare employees. And while their employees liked the predictability of the process,
they did not find much value in the once-a year conversations about past performance.
Taking these factors into account, Deloitte decided to radically redesign their performance assessment system. They wanted a process that focused on improving future performance and enhancing
employee strengths rather than assessing past behaviour, and they wanted it to do so in a dynamic
way so that employees could quickly adapt their performance to respond to the volatile, uncertain,
complex and ambiguous (VUCA) environment that they operate in.
As a result, they scrapped once-a year reviews and replaced them with agile, real-time ‘performance
snapshots’. They also shifted focus from managers rating employees’ past behaviour, to managers
identifying how they themselves will work with employees, e.g. ‘Given what I know of this person’s
performance, I would always want him or her on my team’. Deloitte argue this reduces rater error
because people may be more accurate assessing their own, rather than others’ behaviour, and it
keeps performance assessment focused on the future rather than the past. Deloitte also introduced
weekly check-ins, involving brief but regular conversations with employees about how to enhance
their strengths so their performance remains accurate and relevant to their environment.
It remains to be seen whether weekly check-ins and regular performance monitoring help to reduce
the overall time spent on performance management. However, real-time, ongoing performance assessment systems are certainly growing in popularity as organisations contend with a VUCA environment.
Source: Buckingham, M. and Goodall, A. (2015) ‘Reinventing performance management’,
Harvard Business Review, 93(4), 40–50.
Introduction
Assessing performance at work can be a contentious topic that has important implications
for individuals and their employers. It is a complex and fascinating activity that poses many
challenges for researchers and practitioners. Many factors can threaten the validity of systems
and procedures that organisations use to assess employees. Yet, there is also a substantial body
of research that can help inform the design of more reliable, accurate and fair performance
assessment. Not surprisingly, performance assessment constitutes a core area of research and
practice for work psychologists.
This chapter sets out important challenges associated with performance assessment and
how work psychology can contribute (e.g. what is work performance and how do we measure
it?). It explores why, for example, performance appraisals continue to receive criticism, and
methods that can be used to improve the delivery and effectiveness of performance assessment procedures. We begin by considering how performance assessment contributes to wider
organisational systems of performance management. Next, we examine what is meant by
work performance, and how it can be defined and measured. We then explore how interpersonal assessments can be subject to bias and distortion, and how findings from work psychology research can help design systems that deliver more reliable and valid decisions. The
chapter concludes by looking at the challenges and future trends of performance management in research and practice.
Chapter 4 Assessing performance at work
113
Purposes of performance assessment
The broad purpose of performance assessment is to evaluate how well an individual is performing in relation to the requirements of their role and compared to other colleagues in their
work group or wider organisation. Traditionally, this has been accomplished using performance
appraisal, where a manager is expected to observe and record how well an employee is performing, collate and compare with observations from other sources (e.g. team members and clients),
evaluate their observations against pre-determined job-related performance criteria, then meet
with the employee to provide feedback. The appraisal meeting traditionally occurs annually, or
bi-annually; although as we discuss towards the end of this chapter, many organisations are beginning to question the form and frequency of performance assessment. Importantly, information
gathered from performance appraisal contributes to complex organisational systems, including
human resource management (HRM). As such it is important that all responsible for developing
and implementing appraisals understand the different reasons for assessing performance, which
include for administrative, developmental and strategic purposes (Iqbal et al., 2015).
When performance assessment is used for administrative purposes it helps decision-making
about how to manage, reward, promote, retain and terminate personnel. For example, a manager can use information gathered from appraisals to decide how to allocate performance
related pay, such as employee bonuses. Administrative purposes are therefore aligned with
HRM philosophy: this typically views employees (or personnel) as an essential business
resource (often referred to as ‘human capital’) that requires careful management and monitoring via reliable and accurate performance assessment, in order to achieve the best return
on investment (ROI: Schleicher et al., 2018).
Performance assessment used for developmental purposes focuses on how and where an
organisation needs to build the capability of employees. This includes identifying training
needs, providing developmental feedback and helping to select future work assignments that
will challenge employees. For these purposes it is particularly important that there is opportunity for two-way dialogue so both managers and employees can reflect on performance
and identify development needs and future goals. These systems are important for individual
employees, because they impact on opportunities for career progression and reward. For example, ‘talent pipelines’ are used by many large organisations to identify talented individuals at
junior levels (e.g. from recent graduate recruits) and provide additional targeted support to
facilitate skill development and help them progress through the ‘pipeline’ to senior leadership
positions. Of course, ensuring initial judgement about potential is accurate is vital to avoid
resources being directed inefficiently (i.e. individuals being judged to have leadership potential
who later fail to progress) and, perhaps more importantly, to ensure that talented people with
potential are not overlooked or lost to the company if they leave and move to a competitor.
Accurate and fair assessment is therefore essential to ensure that employees receive the most
appropriate reward, support and development, and that human capital is not wasted.
Thirdly, data from performance assessment systems can also inform strategic
decision-making: performance data collected from different parts of a company can be useful
when making decisions about investing in, or supporting, human capital. For example, if
there are reliable data that employees in one part or function of the company are not performing well then remedial interventions (e.g. training or improved selection processes) can be prioritised. Advances in information technology now routinely allow performance to be tracked
over time, and comparisons to be made about performance across different work groups,
different parts of an organisation and sometimes between different companies. We explore
this in more detail later in the chapter. However, while electronic performance monitoring
is becoming easier and more prevalent, most performance assessment still relies on managers’
interpersonal ratings about the work of the people they supervise. Not surprisingly, the psychological processes influencing the quality of these ratings have been extensively studied.
Clearly, an important advantage of performance assessment is its versatility of purpose.
However, this can be problematic if those responsible for developing and implementing
114
Work Psychology
performance assessment fail to clarify objectives. For example, appraisals that aim to support learning and development may be undermined if they also closely scrutinise past performance and are linked to pay (Evans and Tourish, 2017). Few individuals are likely to be
open and frank about areas they need to improve if they know that information from the
performance appraisal process will also be used to determine their salary.
Defining work performance
Understanding what constitutes work performance and how it can be assessed is one of the
most important challenges facing work psychologists (Campbell and Wiernik, 2015). Yet
assessing work performance is a surprisingly complex activity. At its core lie two fundamental questions (i.e., ‘what do we mean by performance at work’ and ‘how do we measure it?’)
Together these have come to be known as the ‘criterion problem’ and have generated a substantial body of research and discussion (Bennett et al., 2014).
In short, if performance is measured accurately, it first needs a clear definition about what
needs to be measured, but this can be rather tricky. Take the case of a violin player: it may
seem obvious that good performance is being able to play music well, but what exactly do we
mean by this? What criteria would we use to indicate someone is playing well and how would
we assess this? We might try to define ‘good performance’ by rhythm and pitch but also how
enjoyable it is to listen to, yet this relies on subjective evaluations about what is ‘enjoyable’ and
might vary depending on audience members’ knowledge and preferences of different music
styles. In many jobs there are often many ways that role incumbents can demonstrate good
and poor work performance. The potential variety and complexity of judgements involved in
assessing performance means that it is very important to have clarity for workers and managers
about what a job incumbent is expected to do and the standards they are expected to achieve.
Exercise 4.1
Rating your lecturer
Most universities routinely ask students to rate their lecturers as a way of capturing information
(feedback) about the relative quality of different courses and teachers. Think about the questions
that you are asked about your lecturers’ performance. Now consider whether these questions capture all aspects of the work undertaken by a university lecturer. What other areas of performance are
likely to be important and how might these be assessed? What other performance criteria might be
used to provide a comprehensive assessment of how well a lecturer is performing his or her work?
Generic models of work performance
One way in which researchers have sought to deal with the complexity of work performance
as a construct has been to investigate whether there are underlying components of work
performance common to all job roles. This approach is similar to the efforts of psychologists
to understand the latent structure of personality or intelligence. For example, personality
theorists (e.g. Goldberg, 1993) have identified five traits that appear to explain personality
(see Chapter 2).
Bartram (2005) examined competency frameworks developed by many different companies
for different and similar jobs. Broadly, competencies relate to the observable skills or abilities that an individual requires in order to perform a task or role effectively, and competency
frameworks describe the behaviours associated with good and poor performance for each competency (see Chapter 3). In a meta-analysis of 29 validation studies that used line managers’
ratings of performance, Bartram investigated whether there was any similarity between the
Chapter 4 Assessing performance at work
Table 4.1
115
Job performance factors
Bartram (2005)
Campbell (2012)
1 Enterprising and performing
1 Technical performance
2 Interacting and presenting
2 Communication
3 Analysing and reporting
3 Initiative, persistence and effort
4 Creating and conceptualising
4 Counterproductive work behaviour
5 Adapting and coping
5 Hierarchical leadership
6 Supporting and co-operating
6 Hierarchical management
7 Leading and deciding
7 Peer/team leadership
8 Organising and executing
8 Peer/team management
competencies identified as important across these independent sources. He found evidence for
eight ‘great’ factors of job performance that were common across different jobs and companies
(see Table 4.1). Using a different approach, Campbell (2012) has also identified eight ‘general’
factors of job performance, each of which is made up of several more specific factors.
As Table 4.1 shows, used together these two models provide a comprehensive description
of the breadth and depth of work performance. However, there are intriguing differences. For
example, unlike Bartram, Campbell’s model does not highlight adaptability and creativity,
factors that have become increasingly recognised by practitioners as important to job performance in a VUCA environment (Harari et al., 2016). Yet, Campbell does distinguish between
hierarchical and team-based relationships; the latter including behaviours like helping and
cooperating with others that reflect more collaborative work requirements of flatter organisations. This behaviour and the addition of counterproductive work behaviour (discussed
below) show that Campbell’s model looks beyond how performance might be described in a
traditional role description to include behaviour that might be considered extra to the role.
Key learning point
Although different work roles have different performance requirements, there appear to be categories
of work activity that are common to most work roles.
Point of integration
The techniques and methods used to investigate and analyse work performance are very similar to
those used in the early stages of the design of a selection process (see Chapter 3). Indeed, results from
competency analysis are often used to shape both performance assessment and selection processes.
116
Work Psychology
Extra-role performance
In fact, there is a substantial body of research which focuses on aspects of performance that
are not captured in traditional conceptions of work performance. For example, Borman and
Motowidlo (1997) draw a useful distinction between ‘task’ performance (defined as the effectiveness with which an employee completes a set of technical or role-specific objectives) and
‘contextual’ performance (which impacts indirectly on organisational climate, perceived
support and improved work satisfaction). Borman and Motowidlo identify seven types of
behaviour associated with contextual performance, which include volunteering to take on
responsibilities not formally part of one’s role, supporting and defending organisational decisions, persisting with enthusiasm, and following rules even when they may be personally
inconvenienced.
Building on this distinction, Griffin et al. (2007) created a framework of work performance
that includes proficiency (i.e., the extent to which an individual meets the formal requirements
of his or her role), adaptivity (i.e., the extent to which an individual adapts to change in work
systems or roles) and proactivity (i.e. the extent to which an individual takes self-directed
action to anticipate or initiate change in work systems or roles). More recently, this model
has been used to classify and describe how 97 existing performance constructs contribute at
individual, team and organisational levels (Carpini et al., 2017).
Stop to consider
Research by Michael Frese and colleagues (see Frese and Gielnik, 2014) identifies proactivity as important aspect of ‘active performance’; characterised by workers who (1) use their initiative, (2) self-start
goals and (3) persist in attaining them. They have found that these qualities are particularly beneficial
for entrepreneurs. What type of roles would benefit from ‘active performance’? How might you assess
active performance?
The extra role behaviour that has received most attention from researchers is ‘organisational citizenship behaviour’ (OCB: Organ, 1988; 1997). OCB was originally regarded as
voluntary behaviour not requested directly by managers or listed on job descriptions and
appraisal forms but identified as important in helping to make the workplace a better environment. OCB is often categorised into behaviours targeted towards individuals (OCB-I),
such as general courtesy or helping teammates when they are absent or have heavy workloads, and behaviours that influence organisations (OCB-O) like adhering to informal rules
or being careful with organisational property (Ocampo et al., 2018; Williams and Anderson,
1991). In a meta-analysis of 168 samples, Podsakoff et al. (2009) found that OCB was related
to managers’ ratings of employee performance and decisions about how to allocate rewards,
as well as to productivity, efficiency and reduced costs.
Point of integration
OCB is a recurrent theme in work psychology. In this text you will also find OCB mentioned as a feature
of engagement at work and as an outcome in theories of motivation (Chapter 6). Perceptions of justice
appear to play a particularly important role in motivating OCB and, interestingly, there is little evidence
that pay incentives increase the likelihood or frequency of different OCBs.
Chapter 4 Assessing performance at work
117
Despite originally being conceptualised as extra-role behaviour, OCB can be considered
an important aspect of task proficiency in many roles, particularly where employees work
interdependently (Carpini and Parker, 2017). For example, nurses and doctors who are punctual, look after hospital property and help team members would be considered as engaging
in activities crucial for team success, rather than ‘going the extra mile’. There are also cultural differences in what might be deemed ‘extra-role’. For example, managers in collectivist
cultures, such as China, are more likely than those in more individualistic cultures to value
OCB over task performance (Mishra and Roch, 2017). To reflect workplace and cultural characteristics, Organ (1997: 91) redefined OCB, describing it not necessarily as ‘extra-role’ but
rather ‘contributions to the maintenance and enhancement of the social and psychological
context that supports task performance’. Despite this reconceptualisation, OCB remains a
challenge for performance assessment because it is not normally made explicit in job descriptions. Consequently, it can be unclear whether managers expect OCB from employees and
thus whether raters are expected to assess it.
Another form of extra-role performance is counterproductive work behaviour (CWB:
Sackett and DeVore, 2001), but unlike OCB, CWB involves intentional acts that cause harm
to organisations, such as stealing property, ‘cyberloafing’, or ‘un-green’ behaviour like not
recycling, or to individuals, such as endangering, gossiping or blaming co-workers (Dilchert
et al., 2018; Marcus et al., 2016; Robinson and Bennett, 1995). Although it is tempting to consider OCB and CWB as opposites, research suggests that employees may engage in both types
of behaviour at the same time. CWB may even be motivated by OCB demands (Bauer et al.,
2018; Fox et al., 2012); for example, employees who feel forced to help co-workers may react
negatively by withdrawing from, or sabotaging, other projects. CWB can also result from factors like personality, feeling overqualified, feelings of injustice or role conflict (Ones, 2018). In
practice, observing and accurately assessing CWB is challenging because most employees who
engage in these types of behaviour do so covertly. Managers are therefore likely to depend on
reports from other employees or electronic monitoring to identify CWB (Gruys et al., 2010).
Clearly, there is much to be explored in relation to defining and operationalising work
performance, and, as Carpini et al. (2017) argue, there is still a lack of integration between
different performance constructs. Future research therefore needs to look at broadening both
conceptualisations of job performance and methods of measuring it.
Stop to consider
Some researchers have suggested that OCB and CWB may be political in nature. For example, helping a co-worker might be regarded as a positive OCB behaviour, but it could also be perceived as an
act of political self-interest, impression management or favour-exchange, for example, ‘I’ll help you
now but I expect to be helped in return’ (Bolino et al., 2013). Likewise, employee behaviour deviating
from expected norms may be construed as CWB or constructive, depending on the perspective of
the observer (e.g., whistleblowing or expressing voice: Vadera et al., 2013). Therefore, the motive or
intention behind CWBs and OCBs and differences in how these are interpreted by others increase the
complexity of assessing these types of behaviours.
Key learning point
For performance assessment to be successful in any work context two questions need to be answered:
(1) what constitutes good and poor performance, and (2) how can it be measured?
118
Work Psychology
Measuring work performance
Our ability to define work performance is important because it impacts on how performance
is measured and assessed. In the case of the violin player, if audience perception is used as a
criterion of good or poor performance, we need to ask the audience to rate the musician. But
we also need to consider what rating scales might be used and how we can be sure that different
members of the audience are rating performance in a similar way. There are many different
criteria that can be used to assess performance, and some criteria are easier to quantify than
others. For example, performance criteria for a retail salesperson might include the number
of successful sales they achieve, their customer satisfaction ratings, absenteeism rate or how
many mistakes they make. These types of information are often viewed as more objective
measures of performance than ratings provided by managers, because the latter can be influenced by bias or whether a manager has sufficient opportunity to observe a worker.
In reality, no performance criterion is likely to be completely accurate. Volume of sales can
depend on the product area for which an employee is responsible (for example, in a retail pharmacy store there may be fewer opportunities to sell foot spas than hair products), or whether
an employee is working in a store located in a relatively deprived or economically vibrant area.
In order to overcome problems associated with any one type of performance data, data from
several different performance criteria may be collected allowing findings to be triangulated (i.e.,
examined together) to build a more comprehensive picture of how an individual or work group
is performing. To date, however, the most common form of performance data is still generated
from managers’ direct or indirect observation of individuals in the workplace (Brown et al.,
2018), so it is important to understand how these can be improved.
Observing and judging performance
Managers routinely observe and informally judge the performance of their subordinates on a
day-to-day basis, but there is considerable evidence that these interpersonal judgements can
be selective and biased. There is a substantial body of literature in social and work psychology
concerned with common forms of bias that influence managers’ (and others’) interpersonal
judgements. They include ‘halo and horns’, where individuals tend to make generally favourable or unfavourable evaluations of people based on the limited information initially available
to them (Belle et al., 2017). For example, a job applicant who answers a first question well can
sensitise the interviewer to other positive features of the applicant (halo effect), or to negative
features if they answer badly (horns effect).
Managers can also be vulnerable to primacy and recency effects. This is where information
presented early or late in a sequence dominates memory and judgement. For example, observations made by a manager in the week before an appraisal might play an overly significant role in
shaping the manager’s appraisal rating, and if performance is only assessed once or twice a year
this might be particularly problematic. Another common bias is the fundamental attributional
error, whereby individuals habitually ignore the situational influences on others’ behaviour, and
often attribute it to internal dispositional causes (e.g., personality or ability). This bias may also
become more likely when the wider economic context is prosperous because managers may feel
employees have a greater generalised ability to control their own behaviour (Sirola and Pitesa,
2018), leading managers to underestimate the importance of environmental factors on individual performance; including the impact of their own actions on the performance of others.
Key learning point
Several different unconscious biases can undermine the reliability and validity of performance ratings
made by colleagues, managers and clients.
Chapter 4 Assessing performance at work
119
The biases discussed so far are likely to apply to whomever managers are rating, but some
sources of bias can have a detrimental effect on certain groups of employees compared to
others. For example, similarity–attraction theory (Byrne, 1971) predicts that the more similar
an observer (e.g., manager) perceives another person (e.g., subordinate) to be to themselves
in terms of personality, attitudes, values or demographic characteristics, the more prone
they are to like, trust and interact with them. This means that employees who are similar
to their line manager may be rated more favourably and allowed more opportunities to
work on important assignments, thereby increasing their potential to progress (Wyatt and
Silvester, 2015).
These types of bias often result from unconscious cognitive shortcuts (better known as
cognitive heuristics) that we use to make sense of, and respond quickly to, complex situations, people and surroundings. Another example is stereotyping, which arises when we rely
on automatic processing of cognitive schemata that develop over time from experience and
cultural learning (Fiske and Tablante, 2015). While these stereotypes can be useful in speeding up the way we make decisions about people and actions, this comes with two important
costs, namely accuracy and fairness. Stereotypes are often too simple and can reduce the
accuracy of judgements about performance (e.g. Li et al., 2017). For example, a manager
might hold a stereotype of new mothers returning to work as consultants after maternity
leave as ‘unreliable and likely to prioritise family over client needs’. This stereotype may
make the manager more likely to award new business contracts to employees without children, and possibly more likely to attribute poor performance to lack of motivation rather
than lack of opportunity.
In many organisations, performance judgements are used to make decisions about whether
an individual has potential to achieve future leadership roles, and biased performance
evaluation can have important implications for employees from minority groups (Giumetti et
al., 2015; Roth et al., 2003). According to Lord and Maher (1991) becoming a leader depends
on being perceived to be a leader by others. But, as Virginia Schein’s (1975) work on ‘think
manager – think male’ has shown, women and members of minority groups can be disadvantaged because leadership prototypes are typically white and male (Braun et al., 2017).
People judged to have leadership potential are usually singled out for additional training and
development and greater access to mentors and networks (Murrell and Blake-Beard, 2017),
therefore it is important to train managers to be aware of, and motivated to address, different
sources of bias. Another way to make managers more careful and less likely to engage in automatic stereotypic thinking is by holding them to account for the assessment decisions they
make. For example, evidence shows that simply asking assessors to record their observations,
and explain their decisions to employees, results in them engaging in more controlled and
effortful processing of information (Harari and Rudolph, 2017).
Key learning point
To maximise fairness and validity, raters need to be aware of the ways in which bias can influence the
way they perceive and rate the performance of others.
Point of integration
Raters can be trained to be more aware of unconscious biases that may affect ratings of ‘out-group’
members, but the impact of this training on behaviour has mixed support (see Chapter 7).
Work Psychology
120
Rating performance
The last section focused on naturally occurring sources of bias during interpersonal decision making, but another equally significant source of variation in assessment occurs when
there are no clear or shared criteria regarding good and poor performance (Longenecker and
Fink, 2017). This section considers an area where work psychologists have made an extremely
important contribution, namely the design of rating scales for performance assessment.
Graphical rating scales, like the ones shown in Figure 4.1 are one of the most popular formats for assessing performance (Gorman et al., 2017). Yet, like many forms of psychological
measurement they can vary in their reliability and validity. For example, scales should provide
(a)
Very
good
Very
poor
(b)
Average
(c)
Consistently
good
performance
Performance
generally
better than
average
Satisfactory
performance
with average
number of
errors
(d)
1
(e)
10
Never
satisfactory
Figure 4.1
Usually
unsatisfactory
Consistently
poor
performance
Overall performance level
Performance
Errors
Wastage
Quality
Safety
Usually
below
average
performance
Unacceptable
Acceptable
Good
20
30
40
Sometimes
unsatisfactory
Graphical performance rating scales of various kinds
Usually
satisfactory
50
Always
satisfactory
Chapter 4 Assessing performance at work
121
a clear indication of the meaning assigned to each point on the scale (validity) so that the rater,
and anyone who needs to interpret the rating on the scale, can make a valid inference. Clear
and unambiguous interpretation is impossible with the scale in Figure 4.1a, since it provides
so little information. It is also important that scales are used consistently, either by different
raters, or by the same rater on different occasions. The scale in Figure 4.1b poses problems,
because with so much subjective judgement required judgements may well change from rater
to rater, or from trial to trial. Although the scale in Figure 4.1c provides a better basis for validity and reliability, such graphical rating scales are still vulnerable to many possible sources of
error. One of these, leniency, is a characteristic of the person doing the rating. Some people
appear to be ‘easy’ raters – tending to provide a greater proportion of high scores (positive
leniency), whereas others tend to be harsh or severe raters (negative leniency). Leniency effects
can often be observed when the results of two or more assessors are compared. Although raters
may be unaware of their leniency preferences (the effects may be the result of unconscious
bias), there is also a potential for motivated leniency which occurs when a rater is motivated
to favour one assessee over another. Motivated leniency might occur when a manager wishes
to promote a particular employee rather than another, or if they want to reduce any anxiety
that might result from providing negative ratings (Rosen et al., 2017).
Other sources of error include the halo effect, discussed above, and the central tendency
error, which occurs when raters are reluctant to provide ratings at the extremes of a scale and
their ratings therefore tend to group around the mid-point of the scale, resulting in a potential restriction of range. All of these can influence the accuracy and fairness of decisions that
result from ratings made using scales, but one straightforward way of helping to minimise
error in any type of rating scale is to use multiple questions (items) for each performance
area being rated. As any individual item is open to some degree of misinterpretation by
the rater, using a set of items relevant to the construct being rated will help to reduce overall error. This is because the misinterpretations and errors related to individual items will
probably be random in their effect and therefore likely to balance each other out (although
an exception to this will be the case of motivated bias). Consequently, a score based on the
average of all items will almost certainly be more reliable than a score derived from a single
item (see Figure 4.1).
Key learning point
As well as unconscious biases, deliberate manipulation of performance ratings may occur.
Behaviourally Anchored Rating Scales (BARS)
As we have seen, a common problem with unsatisfactory rating scales concerns the failure
to define sufficiently the ‘anchor’ points. Broad, generalised descriptions of anchors such as
‘average’, ‘good’ or ‘excellent’ make it impossible to be sure that everyone who uses the scales
will interpret them similarly. One popular means of providing unambiguous anchor points
on a scale is to use behaviourally anchored rating scales (BARS), where anchors describe a specific
behaviour that is critical in determining a particular level of job performance (Smith and
Kendall, 1963). BARS are developed using a four-step procedure:
1
With the aid of a group of ‘experts’ (i.e. people who understand the role in detail such as
employees, supervisors or senior managers) define the factors needed for successful job
performance.
Work Psychology
122
2
Use a second group of ‘experts’ to provide examples of specific behaviour associated with
high, average or low performance on the factors.
3
A third group takes the examples from step 2 and independently matches them with the
factors from step 1. This ‘retranslation’ acts as a cross-check on the two previous steps.
Examples that are not assigned correctly to the aspect for which they were written do not
provide unambiguous behavioural anchors for that aspect of job performance and should
not be used.
4
The final step involves using more ‘experts’ to assign scale values to the surviving items,
which then serve as the behavioural anchors for the scale.
Multiple item
Strongly
agree
Strongly
disagree
1 The lecturer speaks clearly
1
2
3
4
5
6
7
2 Overhead projector slides are visible and
contain a small number of relevant points
1
2
3
4
5
6
7
3 etc.
Single item
Strongly
agree
1 The lecturer presents material well
1
2
Strongly
disagree
3
4
5
6
7
Behaviourally anchored rating scale (BARS)
Presentation
9
8
7
6
5
4
3
2
1
Could fail to define some
(non-central) concepts
Could be expected to speak audibly, without
hesitation, and respond to student reaction
Could be expected to use overheads in correct
order and give summary at end of lecture
Could fail to be audible on several occasions
Behaviour observation scale (BOS)
Prepares overhead projector slides
and key points in advance of lecture
1
Never
2
Seldom
3
Sometimes
4
Generally
Forced choice rating
Choose two out of four statements that best describe the employee’s performance
1 Successfully persuades customers
2 Demonstrates knowledge of the industry
3 Is confident with customers
4 Develops ongoing relationships with customers
Figure 4.2
Multiple item BARS and BOS rating scales
5
Always
Chapter 4 Assessing performance at work
123
Behaviour observation scales (BOS), which follow similar development procedures as
BARS, measure frequency that behaviours occur (Wiersma and Latham, 1986: see Figure 4.2).
Research in the late 1970s suggested that BARS produce results that are only slightly better
than well-constructed graphic rating scales (Landy and Farr, 1980). However, BARS remain
a popular method of assessing performance (Gorman et al., 2017). This is possibly because
a well-constructed graphic scale may require an amount of effort equal to that involved in
constructing BARS. Furthermore, with planning, it is possible to combine the process of eliciting information required to compile a BARS with undertaking a job analysis (see Chapter 3),
which minimises both cost and effort.
Other scale development procedures include forced choice rating formats, where raters
are presented with several statements that describe behaviour, but rather than providing
ratings for each one, they are asked to choose those that most accurately reflect the employees’ performance. Using Item Response Theory, raters’ selections are then scored against the
underlying performance dimension (Brown, 2016). For example, in Figure 4.2 raters would
be asked to choose two out of four statements that describe the performance of a salesperson.
If statements 1 and 2 best represent good performance, then choosing statements 3 and 4
might result in a lower performance rating. Because the statements are equally desirable,
and the scoring mechanism is not transparent to raters, this format of rating can reduce halo
effects and motivated leniency (Brown et al., 2017). However, this lack of transparency also
means forced choice systems are less popular with managers, because they experience less
control over the assessment process and its outcomes, and it is also difficult to identify areas
for employee development. Despite the considerable attention paid by work psychologists
to reducing measurement error in ratings, somewhat disappointingly no procedures have
been able to produce scales immune to rating error. Rating behaviour remains one of the
most difficult tasks that managers must undertake when assessing employee performance.
Key learning point
Rating scales can be improved by forced choice rating formats, but managers may not like not knowing
how their responses alter the performance score.
Key debate
Should we get rid of performance ratings?
Although decades of studies have examined the structure and utility of performance ratings, there is growing debate
amongst work psychologists (summarised by Adler et al., 2016) about whether we need performance ratings at all.
On one side of the debate reported by Adler et al. (2016), psychologists Colquitt, Murphy and Ollander-Krane argue
that we should give up trying to improve the validity and reliability of performance ratings, because despite a considerable amount of research raters often disagree when evaluating the same employee. For example, there are often
weak relationships between employees’ actual performance and the ratings they receive, and factors other than job
performance (e.g. rater biases, motivation) can impact ratings (Spence and Keeping, 2013). Another argument is that
providing ratings places managers (and often colleagues) under too much pressure to accurately observe individuals
and produce time consuming reports; likewise, employees often find the process too bureaucratic, and neither motivating nor valuable. They suggest that removing ratings will allow managers more time to provide feedback informally,
encouraging richer conversations around performance, rather than having to focus on justifying scores. Employees
may also experience less anxiety and increased engagement with their performance.
124
Work Psychology
Another group of psychologists disagree. Adler et al. (2016) report that Adler, Campion and Grubb suggest that
we should continue to strive to improve rating accuracy due to encouraging evidence that ratings can be valid and
reliable. They argue that even in organisations that focus on development, managers still make informal judgements about performance (e.g. ‘you did well today!’) and it is important that these are structured and documented
to render transparent and fair any decisions that are based on informally judged performance. Importantly, they
claim that ratings are required to establish an increase (or decrease) in performance, which employees need to
manage their own development. This can be especially valuable for attracting and retaining high performers who
likely want to be rewarded for their individual efforts. They also provide a useful reminder that few alternatives to
ratings exist when it comes to motivating, promoting and rewarding employees, and for guarding organisations
against lawsuits.
Overall, both sides agree that performance management needs to be improved and that perhaps work psychology has focused too heavily on the mechanics of rating scales and as a discipline we may need to consider
the performance management process more holistically and practically.
Rater training
One of the oldest methods of improving the quality of performance ratings is to train the people who provide ratings. Rater Error Training (RET) involves telling raters about the common
errors they might make (e.g. halo effect) and encouraging them to avoid doing so. Technology
can now provide raters with real-time feedback about potential bias as they complete performance assessments (Bracken et al., 2016), though in practice, training that tells raters what
not to do has not been well supported (DeNisi and Murphy, 2017).
Better outcomes have been found when using Frame of Reference (FOR) training
(Hauenstein and McCusker, 2017), which aims to provide raters with a common idea of what
performance looks like. FOR training typically provides raters with examples of what constitutes good, average and poor performance, then asks them to observe and rate a hypothetical employee by watching samples of the employee’s performance on video and identifying
behaviours that justify their rating; trainers then discuss why a rater was accurate or inaccurate in their assessment. This is likely to be more effective than RET because it provides raters
with a theory about why and how behaviours might vary, making them less susceptible to the
types of bias discussed earlier. Roch et al. (2012) found clear evidence in their meta-analysis
that FOR improved rater accuracy across different work situations. They suggest that it has
become the method of choice for rater training.
Multi-Source Feedback (MSF)
Multi-source feedback (MSF) emerged in response to criticisms levelled at traditional PA and
is now a standard part of many performance management systems. It is often referred to as
360-degree feedback, because the performance of a target individual is rated anonymously
by many different people at various levels of seniority who are able to observe different
aspects of their work behaviour. These raters normally include the target’s manager, their
work colleagues, the people they supervise (direct reports), and can include clients or customers (see Figure 4.3), allowing MSF to provide a more rounded and arguably more valid
assessment of an individual’s performance. For example, work colleagues may be better
able to assess how well someone engages with team members, customers can provide a
perspective on service interactions, and direct reports can comment on their manager’s
ability to manage.
Chapter 4 Assessing performance at work
125
Manager
Client
Client
Colleague
Target
Direct
Report
Figure 4.3
Colleague
Direct
Report
Direct
Report
Sources providing multi-source feedback for target employee
Individuals typically complete an online questionnaire which asks them to rate a series of
behaviours that are relevant to different competencies that the target individual is required to
perform in their job (e.g., ‘responds quickly to customer needs’ or ‘provides clear instructions to
colleagues’). Because these ratings are from multiple people, and they are anonymous, they have
several advantages over feedback provided from a single person, such as a manager. MSF can
be better suited to capturing information about extra role behaviour, such as OCBs, or CWBs,
which managers may not typically observe; multiple raters also means the impact of individual
raters’ biases may be reduced, and targets may also be more accepting of developmental feedback
if they receive consistent messages from multiple sources (Campion et al., 2015).
Figure 4.4 illustrates how feedback from different sources (self, manager, colleagues and
subordinates) might be summarised and presented in a report, in this case for the competency
Managing Performance. Here the 5-point rating scale ranges from 1 ‘well below average’ to 5
‘well above average’. The bottom bar (blue) shows that the individual has rated him or herself
3. The other bars represent the mean scores of the combined responses provided by members
of different rater groups. For example, the mean score from the target’s four colleagues is
3.3. The aim of these reports is to provide target individuals with anonymous feedback from
different groups in order to understand how they might need to change their behaviour. For
example, Figure 4.4 shows that the target rates themselves higher on managing performance
Managing performance
Managers
Colleagues
Direct reports
Self
1
Figure 4.4
2
3
Comparison of ratings from different sources
4
5
126
Work Psychology
than his/her direct reports, so this is an area s/he may want to improve. It may also bolster
an employee’s confidence and motivation when they find others rated them higher than
the employee has rated him or herself (Atwater et al., 2007). These discrepancies between
self- and other-ratings can occur because individuals lack self-awareness (Lee and Carpenter,
2018), therefore feedback enables them to calibrate their own efforts and performance more
effectively against the expectations of others.
Key learning point
Multiple performance ratings from different sources can provide a more rounded picture of an individual’s work performance, but care must be taken not to assume that any one source provides the
most accurate or correct rating.
Key debate
What should be the purpose of MSF?
MSF has traditionally been used for employee development, but psychologists have questioned whether it is wise
to also use MSF for administrative purposes (Pulakos et al., 2015). Tying MSF to pay or promotions might mean that
colleagues offer less candid feedback and adjust their ratings depending on how much they want the target to succeed, or managers change their ratings to reward employees they get on with. Targets may also be less accepting of
the feedback they receive if it is linked to pay, although there is a danger that they ignore feedback all together if it
is solely provided for development (Campion et al., 2015). In a meta-analysis of 24 longitudinal studies, Smither et al.
(2005) found that MSF used only for developmental purposes encouraged individuals to focus on acquiring new skills
and resulted in better gains in performance compared to when MSF was used for administrative purposes.
A key challenge when using MSF for administrative purposes is that different patterns of ratings often emerge
from different feedback groups. Semeijn et al. (2014), for example, studied MSF ratings for managers in a Dutch
organisation and found that while peers and supervisors agreed in their ratings, there was disagreement between
peers and subordinates. It is possible that each set of ratings is an accurate reflection of performance, because each
group has a slightly different perspective. For example, a manager’s idea of what excellent customer service is might
be that which boosts sales, whereas customers might consider good service to be about helping and politeness. While
these differences might be useful to reflect upon for development, when used for administrative purposes it requires
decision makers to choose which groups’ voice is more legitimate, powerful and credible.
Despite these challenges, in a study of 253 South Korean organisations, Kim et al. (2016) found that MSF systems
that were dual purpose (i.e. used for development and administration), had a stronger impact on employees’ knowledge
sharing and the firms’ financial performance. They suggest this is because such systems balance motivating employees’
skill development with holding them to account for behaviour change. Bracken et al. (2016) also advocate dual purpose
MSF because it encourages managers to be more involved in the process so that they discuss results with employees,
plan and prioritise employee development.
However, it is used, it is important that those who develop and implement MSF decide its purpose during the design
phase to ensure that it is applied consistently and is sympathetic to stakeholder expectations and the organisational
climate.
Finally, it appears that follow-up support also enhances the likelihood that performance
will improve among feedback recipients. Providing post-MSF coaching has been found to
improve manager self-awareness, satisfaction, commitment and organisational performance
(Luthans and Peterson, 2003). Although in a study of 469 managers who either attended a
single post-MSF feedback workshop, or attended the workshop and received several coaching
Chapter 4 Assessing performance at work
127
sessions, Nieminen et al. (2013) found that although those who received coaching felt more
confident in their own leadership behaviours, both groups improved similarly as rated by
peers, direct reports and supervisors. Clearly MSF systems provide an opportunity to collect,
analyse and compare large volumes of performance data relatively quickly and easily, and
as such they are likely to continue as popular methods of performance assessment. Future
research is therefore needed to further our understanding of factors that influence successful
use of MSF with different groups and in different organisational settings.
Research methods in focus
Using mixed methods to define work performance
In many work roles it is not easy to define good work performance when different observers may have different
points of view. And if we can’t agree on what good performance is, how do we measure it? Mixed methods research
into the performance of politicians illustrates how work psychologists have attempted to pull together data from
a range of different sources and to analyse it with a range of different methods to tackle such difficult questions
(Silvester, 2008; Silvester et al., 2014; Silvester and Wyatt, 2018).
Although political scientists often use election performance as a proxy for role performance, voters can
choose whatever criteria they like when they decide whether to vote for a candidate (e.g. whether they find
them likeable or personable). We also know that electoral performance might also be influenced by how well
a candidate’s political party is performing nationally or how much that party is supported locally. Electoral
performance also takes no account of how well, when elected, a politician performs the day-to-day aspects of
their role in government or their local constituency. Silvester et al. argue that unlike in other work roles, where
performance is normally defined by managers according to organisational objectives, political performance is a
contested construct. This is because (i) different people and groups vary in what they think a particular politician
should do in office, and (ii) as democratically elected representatives, politicians – unlike employees – have a
legitimate right to define their own work roles.
In their study Silvester et al. examined whether politicians shared a latent mental model of political
performance (in other words, whether politicians had a shared common understanding of what good performance
was). They developed behavioural indicators of performance from semi-structured critical incidents technique
interviews with 32 local government politicians. These interviews elicited examples of specific behaviours that
the interviewees believed were important indicators of good performance in different aspects of the politicians’
role (e.g. when working with political colleagues, the public and local government workers). These incidents were
analysed using thematic analysis to group the behaviours into several themes (or competencies). The coding of
the final themes was then checked with a group of subject matter experts (the politicians themselves and people
who worked closely with them).
These results were used to develop a questionnaire which was used to collect 360-degree feedback, i.e. selfratings from the politicians and anonymous observer ratings from their political colleagues and the non-elected
local government employees they worked alongside. Factor analysis was then used to identify clusters in the
ratings, i.e. groups of behaviours that were rated in a similar way. The exploratory factor analysis of the data from
the self-ratings revealed five common dimensions, or clusters, of performance-related behaviours: representing
people, resilience, politicking (reversed as integrity), analytical skills and relating to others.
But did this set of clusters just reflect the view of the politicians themselves? To test this, confirmatory factor
analysis of the observer ratings (from political colleagues and civil servants) was carried out. This revealed a
similar cluster of performance dimensions. In other words, the different rating groups tended to hold a common
view of the underlying factors that were important when considering the overall effectiveness of a politician.
The authors argued that analysis of 360-degree feedback data provides an important method for determining
whether there are underlying shared latent constructs of performance in roles where stakeholders can hold very
different and potentially conflicting views about what constitutes good and poor performance.
128
Work Psychology
Politics of performance assessment
Much of what we have discussed in this chapter so far has concerned how appraisal processes
can be structured to reduce biases and achieve valid, fair and reliable ratings for individual
performance. Despite such efforts, performance assessment can be vulnerable to political
activity (Spence and Keeping, 2013). For example, managers may intentionally underrate
an employee during appraisal, because they want to make an example of them, to provide
evidence to manage them out of the organisation, or because Machiavellian managers may
want to protect their own power and status. Alternatively, they might overrate performance
to curry favour, reward and encourage favourite employees, to avoid confrontation or loss of
friendship, and even to make themselves look like competent and capable managers (Adler
et al., 2016). Employees’ own political behaviour can also influence performance assessment:
some are more skilled at managing impressions or using tactics like ingratiation (e.g. flattery)
to influence their manager’s ratings (Lee et al., 2017). Likewise, employees with extensive
networks may receive better ratings, because their managers want access to the resources and
connections they control (Bizzi, 2018).
Research has shown that performance assessment can become politicised for several
reasons. Rosen et al. (2017) found that the organisational context is important, with supervisors more likely to deviate from formal procedures and make appraisal decisions based
on informal criteria (e.g. extra-role behaviour) when the wider organisational context is
perceived as political and managers are less likely to be held accountable for their decisions.
Political activity is also likely to increase when managers have limited resources. Pulakos
et al. (2019) theorise that managers may ‘retrofit’ ratings (i.e. decide who gets pay rewards
before rating them) to help ensure they can allocate the limited rewards they control in
ways that best engages their team members and maintains productivity, regardless of individuals’ actual performance. Cultural values also influence the extent to which political
behaviours are considered acceptable. For example, leniency and modesty rating biases
are more likely in Asian cultures that value group-harmony over merit-based assessment
(Aguinis et al., 2012b).
While it may be impossible to remove political activity from performance assessment
completely, it can be reduced if those designing rating systems hold raters to account for
the decisions they make, and consult closely to ensure that assessment processes reflect the
reward strategies managers regard as important for their teams (Rosen et al., 2017; Wyatt,
2017). With relatively few empirical studies in this area, the impact of power and politics in
assessment undoubtedly deserves further attention from work psychologists (Silvester and
Wyatt, 2018).
Key learning point
Although formalising performance assessment can help to reduce biases, competing political goals
can influence the accuracy of appraisals.
Exercise 4.2
Assessing and rewarding team members
You are the manager of a team of five web designers responsible for creating and maintaining websites for clients. Team members interface with client users to understand their needs. The team are
ultimately responsible for making sure that websites are user-friendly and experience no problems or
glitches. Most of the work involves final stage testing of websites and client liaison, either remotely, or
Chapter 4 Assessing performance at work
129
by travelling to the client’s location, which can be anywhere in the world. A computer program ‘erratrak’
tracks all their coding and assigns each team member an overall score out of ten (1 = very poor,
10 = excellent) based on their accuracy, number of glitches and completed projects.
Danny (erratrak score = 4) is a new member of staff who has little work experience so while he
is developing quickly his coding is inefficient. Malaika (erratrak score = 6) has reasonable coding
skills and works very well with a range of clients, she is regarded as a ‘safe pair of hands’ in the team.
Erin (erratrak score = 9) can be difficult with clients and often fails to listen to their requirements,
however her coding is excellent. Alex (erratrak score = 5) is innovative with his code but is lazy
when it comes to checking it for errors, something that has resulted in websites having to be taken
down and client complaints. Teo (erratrak score 5) is usually an excellent coder, but since he became
a father in the last month his performance has suffered.
This year the company has decided to allocate a bonus of £30,000 to be divided between you and
your team. Your task is to decide how to divide this bonus up and explain why you have reached your
decision. How would you make the process fairer next year?
Technology and performance assessment
One way to reduce the impact of politics on assessment is to use measures that do not rely
on the subjective judgements of raters, such as number of sales, or speed in responding
to customers. This type of information can be gathered through electronic performance
monitoring (EPM), which is increasingly widespread. Tomczak et al. (2018: 251) claim that
nearly 80 per cent of organisations use some type of EPM, which involves using technology
to gather, store and analyse data on employee behaviour. Common methods of EPM include
camera systems, computer or phone monitoring, GPS tracking and even information on
physical activity using fitness trackers, such as heart rate and the number of steps employees
have taken.
EPM is commonly used for delivery drivers whose work is tracked by GPS and vehicle
monitors to determine factors like whether they take suitable routes to customer addresses,
their speed, how safely they drive, the number of times they brake or reverse and their fuel
consumption. Handheld digital devices also track how quickly drivers move from their vans
to the customer’s front door, how long they spend with customers and how quickly they are
on the road again. EPM is also used to ask customers about what they thought of the delivery
driver; for example, they may be asked to provide a 1–5-star rating about how courteous
they were.
EPM has advantages over traditional performance monitoring by managers, because it
enables employers to collect and compare performance data from large numbers of workers over periods of time. This ‘big data’ makes it easier to find patterns of employee behaviour to improve efficiency, such as suggesting new ways for delivery drivers to unload
goods to reduce time spent at each address. It can also answer questions like, ‘Does the
performance of employees improve after a training programme?’ or ‘Are employees following safety procedures correctly’. With client permission, consultants may also be able
to collate and benchmark organisational data, that enables them to provide individual
companies with information about the relative efficiency of their workforce. Data collected from EPM are therefore an important source of information for making strategic
business decisions such as whether to invest in training and development or redesign
selection procedures. EPM can also be used to make employees more aware of their own
performance so they can self-regulate their behaviour (e.g. showing you how much time
you really spend on social media!).
130
Work Psychology
Stop to consider
Organisations are increasingly likely to apply machine learning to big data; for example, using complex
algorithms to automatically schedule performance reviews and training for specific employees or flag
up those in need of health and wellbeing interventions (Nunn, 2018; PWC, 2017). However, machine
learning is open to algorithm bias (Howard and Borenstein, 2018). For example, if algorithms are created and ‘trained’ on data primarily about the behaviours of male employees then the conclusions they
reach about the performance of women may be biased.
As an employee how would you feel if your performance and training expectations were identified
by an algorithm? Now put yourself in a manager’s position. How would you justify decisions based on
machine learning to employees? Would you even tell them? In your opinion, are there any ethical concerns with EPM more broadly and how might these be addressed?
EPM has prompted concerns about potential negative impact on employees. Studies
have shown that employees who are electronically monitored can feel it is overly intrusive (McNall and Stanton, 2011), that they lose control over their work (Jensen and Raver,
2012) and have less trust in management (Holland et al., 2015). However, researchers
point to evidence that performance data collected over time is generally more reliable,
resistant to rater error or manager bias, and less vulnerable to the effects of one-off events
like employee illness or a problem at work. EPM has also been suggested to encourage
OCBs and curtail CWBs such as theft, although evidence for this has been mixed. Bhave
(2014) investigated the consequences of EPM in a longitudinal field study which revealed
organisational citizenship behaviour was higher among employees whose managers used
EPM to monitor performance. However, Martin et al. (2016) found that in a sample of
3,200 Australian workers, and depending on their general attitude to surveillance, EPM led
to increased counterproductive behaviour. The authors suggest that this may be a consequence of individuals seeking to retaliate against close surveillance by altering equipment
and pretending to work.
Like any management tool, the success of EPM depends on how it is implemented. Tomczak et al. (2018) provide the following guidelines for effective use of EPM:
■
only utilise EPM for work-related activities – and only when necessary;
■
aim to use less invasive monitoring wherever possible;
■
be transparent with employees about how, where and why they are being monitored;
■
use EPM to develop, rather than punish employees;
■
consider the type of role (e.g. manual, technical, knowledge worker) when deciding the
type of EPM required.
One area that deserves more attention from work psychologists concerns potential practical and ethical risks associated with collecting and storing vast amounts of performance
data about individual employees. In practical terms organisations are faced with a growing
need to recruit and manage highly specialised data science teams to deal with the big data
they collect. Likewise, managers and employees (including researchers) need to understand their responsibilities in relation to capturing, storing and accessing EPM data. While
big data provides big opportunities for analysis and research, work psychologists need to
incorporate and adhere to guidelines about appropriate ways to store, use and potentially
sell performance data (e.g. GDPR in the EU), which means that they also need to explore
how these complex ethical issues can be incorporated into best practice and embed this
in training.
Chapter 4 Assessing performance at work
131
Key learning point
Electronic performance monitoring can yield reliable, dynamic and sensitive data about employee performance, but it may not be welcomed by employees and may require specialist teams able to analyse
and interpret data effectively and ethically.
Improving performance
A key purpose of performance assessment is to provide information that can help identify the best strategies to improve individual and, ultimately, organisational performance.
As part of wider HR systems, performance assessment procedures in many organisations
require managers to provide feedback to those being assessed, identify their development
needs (see Chapter 7), set goals for future performance and provide incentives tied to improved
performance (DeNisi and Murphy, 2017).
Providing feedback
Providing appropriate feedback is crucial if individuals are to understand what and how they
need to change. Whether employees act on feedback though depends on how fair they perceive the performance assessment system to be (Sharma and Sharma, 2017). This can include
perceptions of how fair the process of arriving at appraisal decisions was (i.e. procedural justice), the distribution of rewards (i.e. distributive justice), how this was communicated (i.e.
informational justice) and how they were treated during the process (i.e. interactional justice: Colquitt, 2001). Importantly, the feedback interview itself can influence perceptions of
fairness and performance outcomes because it allows managers to justify their decisions and
provide support for development and gives employees opportunity to have their voice heard.
Research by Jacobs et al. (2014) found that employees who perceived higher interactional
justice, such as being treated with respect, in their last appraisal were more likely to engage
in pro-organisational behaviours, such as doing more than expected and trying to make a
difference at work.
Providing negative feedback can be a challenge for managers because employees may
react with anger or denial (Niemann et al., 2014). Managers may also ‘hold back’ from providing critical feedback to certain groups of employees, such as Millennials, who, despite
research suggesting generational differences are often exaggerated (Lyons and Kuron, 2014),
are widely reported to be particularly sensitive to criticism and likely to react negatively if
feedback emphasises past performance rather than development (Anderson et al., 2016),
or to black Asian and minority ethnic employees and women for fear of appearing racist or
sexist (Croft and Schmader, 2012). However, avoiding negative feedback means employees
understand less about what they can do to improve and develop. Managers therefore need
training so that they are confident in delivering high quality (i.e. positive and negative)
feedback and do so considerately to ensure positive reactions (Wallace et al., 2016). Training
implications of research on feedback suggest the following key actions (Aguinis et al., 2012a;
Meinecke et al., 2017; Rubin and Edwards, 2018):
■
ensure that managers prepare for feedback interviews so that they appear credible in their
knowledge of the employees’ job requirements and performance;
■
ask employees to prepare for the feedback interview, such as reflecting on their expectations and development plans;
132
Work Psychology
■
explain to employees how the performance assessment was conducted, what information
was considered, how was it rated and by whom;
■
structure feedback on specific, factual information;
■
focus on employees’ strengths primarily and how to continue using their strengths at
work;
■
ensure negative feedback focuses on skills, which can be developed, rather than talents
or traits, which are less easily changed;
■
make sure sufficient time is set aside to discuss a development plan with opportunities
for follow-up;
■
encourage employee participation so they can shape the interview process and outcomes
themselves;
■
train managers in coaching skills so they can support employee development over time.
Goal setting
As part of the feedback process managers will need to identify how employees are best
motivated in order to achieve the highest levels of performance. Goal setting theory
(Locke and Latham, 1990) suggests that in order to improve performance, SMART (specific, measurable, aligned, realistic and time-bound) goals are useful to direct employees
in how to divide their time and focus their efforts. As outlined in Chapter 6, SMART goals
can be more effective than asking employees to ‘do their best’ and they can also be linked
to organisational goals so that each employee is aware of how their work feeds into strategic objectives.
Despite these advantages, managers need to consider the content and timing of goals
carefully. Focusing on narrow and long-term performance goals, for example, may mean
other areas of work get ignored or that the goal becomes irrelevant if there are changes in
the environment. Therefore, mixing long-term, ‘distal’ goals with more frequent ‘proximal’
or short-term goals has been advocated to help employees adjust their performance quickly
in response to environmental changes (Pulakos et al., 2019). Although, too many goals set
in succession, or targets that change too frequently can result in employees losing focus and
motivation, which may lead to CWB or unethical behaviour (Welsh and Ordonez, 2014). For
example, a car mechanic who is set a goal to sell tyres, and then set progressively higher sales
targets may resort to unethical behaviour such as lying to customers or damaging customers’
tyres in order to meet his goal.
The appraisal interview can therefore be an important opportunity to set goals in collaboration with employees, rather than imposing them. Employees who can participate in setting
goal expectations and have support from managers to identify strategies to self-regulate their
goal striving, i.e. maintaining focus, motivation and persistence, are more likely to attain
their goals (London et al., 2004). This might include comparing their current performance
with the outcomes they want to achieve and identifying plans to bridge the gap (e.g. implementation intentions: Budworth et al., 2015).
Incentives
As we have discussed, there are often disadvantages to using performance appraisal for development and reward, but the performance appraisal meeting can provide a useful opportunity
for managers to learn about what motivates staff and, perhaps, how they can be rewarded in
ways that do not involve increased pay (see Table 4.2).
Chapter 4 Assessing performance at work
Table 4.2
133
Examples of rewards
Social context
Job-related
Personal
Private office
Flexible working hours and/or breaks
Saying ‘thank you’
VIP parking spot
Job with more responsibility
Compliments on work in progress
Company parties
Control over work content (e.g. choice
of clients, projects)
Friendly greetings
Club privileges
Access to a mentor or coach
Soliciting advice, views or suggestions
Access to personal learning
opportunities (not job-related)
Home-working
Non-verbal recognition (smile)
Introduction to senior personnel
Job rotation
Formal recognition of achievement (letter, in-house
journal)
Membership of ‘high potential’ group
LinkedIn recommendations
Point of integration
When considering how to use performance data, it is important to consider carefully theories of motivation (Chapter 6). Goal-setting theory and the job characteristics model in particular provide several
important insights into how such data can be used to enhance employee performance.
Gamification is increasingly used to provide employees with incentives to improve performance. This involves aligning performance goals to game features, such as points, levels,
badges or leaderboards. For example, a leaderboard may be used to compare the performance
of members of a sales team. Gamification can provide employees with immediate feedback
on how they are performing, how it compares to others and whether they need to adapt their
behaviour to improve performance. Achieving badges or points can give employees a sense of
accomplishment and increase their engagement at work. However, gamification is likely to
be better received if it is used sparingly to avoid employees losing interest and with employees who are more competitive and mid–high-level performers, because poor performers may
consistently lose out on badges/points and score low in group competitions leaving them
despondent (Cardador et al., 2017).
The future of performance assessment
In recent years researchers and practitioners alike have argued that the traditional annual
review process may not be fit for purpose because it often fails to capture the dynamic nature
of the contemporary workplace (Pulakos et al., 2019). Importantly, it has also been reported
that neither managers nor employees are satisfied by traditional forms of performance
134
Work Psychology
assessment (Cappelli and Tavis, 2016). These criticisms reflect two emerging trends in performance assessment: (1) an increase in the frequency of performance assessment, and (2) an
enhanced focus on employee development.
First, as we discussed in our opening case study, organisations are moving away from traditional annual appraisals because there is potentially a long delay (i.e. up to 12 months) before
employees are assessed on their performance. This means that annual appraisals may not
reflect employees’ current abilities and the feedback provided may not be particularly useful.
Instead, shorter and more frequent and informal ‘check ins’ have been advocated to better
reflect the fast-paced nature of the workplace, where people work on short-term projects
and switch between teams (Kuvaas et al., 2017). This allows managers to provide ‘real-time’
feedback to correct performance and set ‘agile’ shorter-term goals so that organisations can
ensure they are proactive and reactive to changes in the competitive market (Pulakos et al.,
2019). For example, a manager might provide feedback with how employees dealt with clients
at the end of a project, before they move onto the next. Because frequent feedback enables
employees to see clearly the link between their performance and how it has been assessed,
they are likely to find the process more satisfying and fairer (Pichler et al., 2018). However, frequent performance assessment may place increased burden on managers unless technology
is used to gather objective performance assessments or crowd-source feedback from others.
Second, organisations are increasingly using performance assessment to drive employee
development rather than holding employees to account or allocating pay. This trend has been
attributed to economic changes, such as the global economic downturn giving managers less
freedom to distribute pay rewards, and organisational changes, where flatter organisations
with team-based work mean that cooperation is valued more than competition (Cappelli and
Tavis, 2016). Some organisations have responded to these changes by removing performance
ratings entirely, although those that do so are still likely to grapple with how to align individual performance and company strategy, manage reward and identify poor performers (CEB,
2016). What is apparent is that organisations are moving towards performance assessment
becoming a two-way process, where employees have a greater say in identifying their own
development needs.
Summary
This chapter has explored the contribution of work psychology to the field of performance assessment at work. As Adler et al. (2016) note in their debate about performance
ratings, there are many potential flaws with performance assessment. These can occur
because performance measures have been badly designed, or the flaws may be caused
by the unconscious or conscious biases of those evaluating employee performance. The
good news is that research has provided some excellent insights into how these issues can
be managed. The bad news is that such problems are unlikely to be eliminated entirely
through the proper design of measures and the use of effective rater training. It is therefore unlikely that any one assessment method in isolation (e.g. performance appraisal,
360-degree review, EPM) will be sufficient to overcome all potential threats to validity
and reliability.
It is important that work psychologists look beyond individual forms of assessment to
the wider responsibilities of performance management systems. There is a growing body
of research that can inform evidence-based decisions about the timings of assessment,
the nature of feedback and the support provided to employees for their development and
which systems are most suited to different contexts. More importantly, such knowledge
can be used to improve understanding among managers and employees about how to make
performance assessment more effective.
Chapter 4 Assessing performance at work
135
Closing case study
From Chris Hadfield (2015) An Astronaut’s Guide to Life on Earth. London: Pan Books.
The Canadian astronaut Chris Hadfield logged nearly 4000 hours in space during his time as an astronaut, and in his recent book An astronaut’s guide to life on earth he provides fascinating reflections
on the performance culture at NASA. Read what he says below and consider how his comments fit
(or possibly contradict) what we have learned about performance assessment and feedback in this
chapter.
‘At NASA, everyone is a critic. Over the years, hundreds of people weigh in on our performance on a
regular basis. Our biggest blunders are put under the microscope so even more people can be aware
of them. . . often we’re scrutinized and evaluated in real time (and) quite a few simulations involve a
crowd. . . when we’re simulating de-orbit to landing, for instance, dozens of people turn up, hoping
that something new – a flaw in a standard procedure, say, or a better way of doing something – will
be revealed’ (p. 77).
Not only does Chris stress the importance of seeing criticism as potentially helpful advice (rather
than a personal attack), he argues that the ability to depersonalise negative feedback is a basic
survival skill for many who work at NASA due to the high-risk nature of their work for self or others:
‘if you bristle every time you hear something negative – or stubbornly tuned out the feedback – you’d
be toast.’
However, he goes further in emphasising the relationship between feedback and organisational culture: ‘At NASA we’re not just expected to respond positively to criticism, but to go one step further and
draw attention to our own missteps and miscalculations. It’s not easy for hyper-competitive people to
talk openly about screw-ups that make them look foolish or incompetent. Management have to create
a climate where owning up to mistakes is permissible’ (p. 79).
Questions:
■
What can we learn from NASA’s approach to scrutinising and evaluating the performance of a few
astronauts in a group setting?
■
Why is NASA’s approach so different?
■
Would traditional performance appraisal work with astronauts?
■
How might you go about building an organisational culture where admitting mistakes is permissible –
even encouraged?
■
Are there any types of organisation or work role where you think NASA’s approach would not work?
■
According to Chris, feedback can ‘sting’ – even for a highly experienced astronaut. Is it realistic to
think that performance feedback can ever be an entirely positive experience?
Individual or group discussion points
1 How can bias be reduced in performance assessment?
■ What biases typically impact performance assessment?
■ What types of rating scales are most effective at reducing biases?
■ What type of training might be useful to reduce bias? Is it effective?
136
Work Psychology
2 Should organisations get rid of ratings from their performance assessment?
■ What are the advantages and disadvantages of ratings?
■ What can replace ratings when making decisions about reward?
3 What are the potential advantages of MSF over traditional performance appraisal?
■ What happens if different sources rate an individual’s performance differently?
■ Which rating source is most valid?
4 How does politics influence performance assessment?
■ What type of political goals might managers have when rating performance?
■ What are the political goals of employees when having their performance rated?
■ How can the impact of politics on performance assessment be reduced?
5 Is electric performance monitoring (EPM) ethical?
■ What are the different types of EPM?
■ Are there limits to what EPM should be used for?
■ How can organisations ensure they are using EPM ethically?
Relevant websites
The O*NET program is the primary source of occupational information for job roles in the
USA. The O*NET database is a free resource that contains information on hundreds of standardised and occupation-specific descriptors: http://www.onetonline.org/
The CIPD publishes a range of free resources relating to performance assessment, including
the following on performance review: http://www.cipd.co.uk/hr-resources/factsheets/
performance-appraisal.aspx
Suggested further reading
Full details for all references are given in the list at the end of this book.
1 Mueller-Hanson, R.A. and Pulakos, E.D. (2018) Transforming Performance Management to
Drive Performance: An Evidence-based Roadmap. New York: Routledge provides a detailed
overview of actionable methods to improve performance management.
2 Tziner, A. and Rabenu, E. (2018) Improving Performance Appraisal at Work. Cheltenham:
Edward Elgar gives an excellent insight into the many practical issues associated with
running successful performance appraisals in the workplace.
CHAPTER 5
Attitudes at work
LEARNING OBJECTIVES
After studying this chapter, you should be able to:
1 define an attitude and briefly describe the components of attitudes;
2 specify two functions of attitudes for the person who holds them;
3 describe the theory of planned behaviour;
4 explain why attitudes and behaviour are not always consistent;
5 describe three features of attitudes that increase the probability that they will influence
behaviour;
6 define job satisfaction and identify three general propositions about what affects it;
7 describe the evidence indicating that a person’s job satisfaction is not simply a function of
the nature of their work;
8 define organisational commitment and its component parts;
9 specify the factors that appear to strengthen a person’s organisational commitment, and
outline its consequences;
10 understand the main
causes of employee
turnover;
11 explain the psychological
contract, understand
how it develops, and
discuss how breaches of
it affect employees and
employers.
138
Work Psychology
Opening case study
Happy workplaces help companies
perform better
Companies with a reputation for strong
employee engagement and creating a happy
workplace generate good publicity and cachet
for their chief executives. But these companies are finding that what makes a great
employer can also lead to business success.
In fact, the link between employee attitudes
and business performance has been known
for decades.
A series of influential experiments from
the 1930s, known as the Hawthorne studies,
showed the impact of improved workplace
environments on overall factory outputs. One
finding was that reducing the working day by
half an hour saw productivity improve. Later,
James Worthy, a sociologist-turned-executive,
wrote in 1950 of increased autonomy for
employees leading to improved morale and
productivity.
While some have argued that successful companies make employees happy, not the other
way around, academic studies have steadily
discredited this view. One of the most comprehensive studies on the topic is Gallup’s
2012 research of 192 organisations in 49
industries and 34 countries, covering 50,000
business units and 1.4m employees. It found
that business divisions scoring in the top half
of reported employee engagement had nearly
double the performance outcomes compared
with those in the bottom half.
The authors concluded: ‘Financial performance is best viewed as a downstream outcome . . . Employees with positive attitudes
toward their workplace are likely to carry
those attitudes over to customers and to
engage in the discretionary effort it takes
to serve . . . at a high level.’ Yves Morieux,
director for the Boston Consulting Group’s
Institute for Organization, believes the
debate is now over. ‘Happy employees create
high-performing organisations . . . because
there are more opportunities to satisfy their
aspirations, job security, chance of promotion, career development [and] wages,’ he
says. ‘That is obvious . . . if the company is
very productive but does not maintain satisfaction at work then in a few years performance will drop.’ But many employers have
yet to cotton on.
Employee engagement and happiness at
work is in decline. Separate studies by the
Conference Board, Gallup and Quantum
Workplace found that employee engagement
has been falling for the past decade or more.
Mr Morieux attributes this to a ‘proliferation
of cumbersome processes, systems, scorecards,
metrics, meetings – what I call “complicatedness”. People spend between 40 per cent and
60 per cent of their time wasting their time,
working on less and less value-added activity.’
By contrast, best practice employers are
‘turning the engagement process into a lot
more than an annual event,’ says Jim Harter, Gallup’s chief scientist for workplace
management and wellbeing. ‘They embed it
in how people think about their day-to-day
work . . . education and training, development that focuses on the natural talents
and strengths in the team. Helping people
feel like they are the future of the organisation and helping them do what they do best.’
Teamwork and collaboration are consistently
linked to increases in innovation and discretionary effort, shaping how some companies
now think about reward packages. ‘You must
be very careful with reward and bonuses,’
says Mr Morieux, ‘because if the bonus is very
significant then your goal becomes to earn
the bonus. And then you will do everything
you can to earn the bonus, including hiding,
exaggerating and why not cheating? These
strong incentives are counterproductive if
you want people to co-operate.’
Netflix, the digital broadcaster, pays only salary and not bonuses. It allows staff to choose
their working hours. A seemingly dull set
of 124 PowerPoint slides outlining Netflix’s
employment culture has become a surprise
viral hit, now viewed close to 14m times.
Mr Morieux believes the Netflix model has
become ‘an archetype’. He adds: ‘The Netflix values include management that is about
‘context, not control’, creating the right environment that inspires people, that provides
direction and transparency – as opposed to
top-down decision making, approvals processes and committees.’ Lego, the Danish toy
manufacturer, also credits employee empowerment as central to its success.
Chapter 5 Attitudes at work
The Lego group chief executive, Jorgen Vig
Knudstorp, is widely quoted as saying ‘blame
is not for failure, it is for failing to help or to
ask for help’. Mr Morieux says: ‘When you use
this principle it changes everything because
suddenly it becomes in people’s interest to
be as transparent as possible about their real
weaknesses, their real forecast, as opposed
to hiding or protecting themselves . . . The
way [Lego] describe it is, “don’t think less
of yourself, but think of yourself less”. This
is precisely to promote co-operation.’ Lego,
along with the likes of Google and WL Gore,
a US manufacturing company, has a flat management structure. Layers of management
are effectively stripped out, leaving only
the employees, a cadre of team leaders and
the executive team. WL Gore’s 10,000-plus
employees are divided into self-managing
139
teams of eight to 12 people who set their own
work and pay. Staff also elect the company’s
chief executive. What does this say, then, for
the future of managers?
Yves Duhaldeborde, director of organisational surveys and insights at consultants
Towers Watson, argues that managers – the
ones that are left, at least – actually become
even more important in flatter structures.
‘What people need from managers is not
someone they get permission from, but the
person who coaches and gives meaningful
feedback,’ he says. ‘Good people management
is key to having happy people in an organisation.’ The best managers build ‘a sense
of trust and autonomy’ in teams, adds Mr
Harter of Gallup. ‘Getting that right leads to
higher performance.’
Source: Tim Smedley (2016), ‘Happy workplaces help companies perform better’, The Financial
Times, 26 February © Financial Times Limited 2019. All rights reserved. https://www.ft.com/content
/6081b1fc-d0b2-11e5-92a1-c5e23ef99c77
Introduction
To simplify a little, attitudes indicate how positively or negatively we feel about something.
The opening case study is just one example of many that describes relatively simple links
between positive employee attitudes and organisational success. The basic premise is that
if we like our job and we like the company we work for, there is a good chance we will stay
with the company and perform better as a result: a ‘win-win’ situation. While this argument
is intuitively appealing, research shows that the links between employee attitudes, retention
and performance are more complex than this. Positive employee attitudes might be desirable for all involved but do not guarantee success. There might be other explanations for the
connections between attitudes and performance. For example, as the case study suggests,
might it also be that workers who are performing well go on to develop a positive attitude to
their employer because they have been given the opportunity to perform to the best of their
ability? Is the evidence that attitudes cause performance (and not the other way around) as
convincing as the article suggests? In this chapter we analyse the evidence relating to the
causes and consequences of job satisfaction and organisational commitment. The concept
of employee engagement is discussed in Chapter 8.
Certainly, how satisfied we are with our job is not just determined by how much we
get paid. As the article points out, using pay to improve employee attitudes also carries
significant risks. The article focuses instead on how attitudes are shaped by, for example,
the content of work tasks and how work activities are managed and scheduled. Teamwork,
autonomy and empowerment are highlighted as determinants of attitudes. Importantly,
these are aspects of the job that can be changed significantly by the actions of managers
and by adjusting organisational structures. Overall, the case study indicates that our attitudes are subject to change according to the circumstances we encounter. But might some
employees be pre-disposed to be more positive about their situation than others, making
140
Work Psychology
employee attitudes somewhat resistant to changing circumstances? This is an issue left
undiscussed in the case study but one that has been of interest to psychologists. In this
chapter we examine in detail these important dispositional influences on attitudes and
behaviour, and how these interact with other influences found both in and outside of the
work environment.
Employers are interested in good quality information about the causes and consequences
of workers’ attitudes. If satisfied workers behave in a way that means they are more likely to
be productive, innovative and energised, attitude survey data could be used to develop powerful interventions. The opening case study identified changes to organisational structure,
line managers’ behaviour, access to training and development activities and the redesign of
work activities as interventions that could change employee attitudes for the better. In other
words, attitudes can be influenced by both the work that people do and the wider organisational context.
Data on employee attitudes are usually gathered using questionnaire surveys. These measure both the strength and direction of attitudes. The questions asked often require employees to provide a subjective evaluation of how they feel about things. This might include
measuring whether they feel very satisfied, very dissatisfied or somewhere between the two
extremes. Or the questions might focus on how much they like or dislike something or someone. Whatever the measure, an attempt is being made to gather reliable information about
private judgements and bring them into the public domain, something that it is not easy to
do well. We examine the challenges of attitude measurement early in this chapter.
A great deal of research has focused on obtaining data about employees’ attitudes towards
specific features of their job or towards the employing organisation. There are sophisticated
analyses of the nature of attitudes, how they change and how they predict, or fail to predict,
behaviour. In this chapter we first examine attitudes as psychological constructs. Second, we
briefly describe the ways in which attitudes can be measured. Two of the attitudes that are
particularly important in work psychology are job satisfaction and organisational commitment. These have the potential to influence lots of organisational outcomes including
performance, attendance at work and employee turnover (Harrison et al., 2006; Judge et al.,
2017). Attitude change might not be very important if it is not reflected in behaviour change,
so a recurrent theme in this chapter is the connection between attitudes and behaviour such
as performance and turnover.
Then attention turns to the question of what factors affect the formation and change of
attitudes at work. A particularly important concept in work psychology is the psychological
contract, so we look at this in some detail. It provides an insight into how the promises that
employees and employers believe that they have made to each other influence work behaviour. For example, in the case study, some employers allow employees to have high levels of
autonomy in return for innovative and creative work.
What is an attitude?
Attitudes were defined by Secord and Backman (1969) as ‘certain regularities of an individual’s feelings, thoughts and predispositions to act toward some aspect of his [sic] environment’.
Attitudes are evaluative and reflect a person’s tendency to feel, think or behave in a positive
or negative manner towards the object of the attitude. Evaluative dimensions of attitudes
include, for example, good–bad, harmful–beneficial, pleasant–unpleasant and likeable–
dislikeable (Ajzen, 2001). Attitudes refer to a particular target, or object – a person (e.g. boss),
group of people (e.g. close colleagues, senior management), object (e.g. work equipment)
or concept (e.g. performance-related pay). The objects of our attitudes that influence our
behaviour can vary from one person to another (e.g. some of us are more influenced by our
peers than by our boss).
Feelings represent the affective component of an attitude, thoughts the cognitive component
and predispositions to act are the behavioural component. The affective component can be
Chapter 5 Attitudes at work
141
reflected in a person’s physiological response: working with inspiring colleagues might raise
our heart rate. We can also voice or report how we feel about the object. The cognitive component refers to a person’s perceptions of and beliefs about the object: we might say our
colleagues are exciting to work with. The behavioural component is reflected by a person’s
observable behaviour towards the object: we might seek out our colleagues when we seek
excitement. An attitude is usually taken to imply the cognitive and affective components.
Behaviour is most often construed as an outcome of attitudes.
Key learning point
Attitudes are a person’s predisposition to think, feel or behave in certain ways towards certain defined
targets. Attitudes are different to personality because they focus on a particular object/target.
Cognitive and affective components do not always correspond. It is possible for a person
to feel positive about their job overall (affective component) but to simultaneously report
that it has only a few attractive elements (cognitive component). Responses to a questionnaire measure of an attitude could be influenced more by affect than cognition, or vice versa
(Schlett and Ziegler, 2014). For example, ‘How do you feel about your computer?’ and ‘How
reliable do you believe your computer to be?’ are two very different questions. Which attitude
is expressed, and which one will influence behaviour can depend on whether the person is
concentrating on emotions or on cognitions (Millar and Tesser, 1989). Pratkanis and Turner
(1994) describe an attitude is stored in memory as a ‘cognitive representation’, which consists
of three components:
1
An object label and rules for applying it. For example, if one is concerned about attitudes to colleagues (the object label), it is necessary to be clear about who counts as a
colleague.
2
An evaluative summary of that object, i.e. whether it is broadly ‘good’ or ‘bad’.
3
A knowledge structure supporting the evaluative summary. This may include technical
knowledge, arguments for or against a given proposition, or a listing of the advantages
and disadvantages of the object.
It is also important to distinguish attitudes from other related concepts. Values are a
person’s beliefs about what is good or desirable in life. They are long-term guides for a
person’s choices and experiences. A worker might not like their boss but they value the
money that the job provides, and therefore a harmonious working relationship with their
boss is relatively unimportant to them. Moods, on the other hand, are ‘generalized affective states that are not explicitly linked to particular events or circumstances which may
have originally induced the mood’ (George and Jones, 1997: 400). Values are relatively
stable over long time periods and focus on how things should be in the future. Attitudes
change but evolve steadily over time. Moods are how feel in the present moment and can
fluctuate rapidly and often, but like values moods guide our behaviour generally and are
not focused on a specific target or object.
Although different, values, attitudes and moods can influence each other. Engaging in
work activities that are congruent with one’s own values (self-determined behaviour) has
been shown to be linked to more positive work attitudes (Glomb et al., 2011). Over time
attitudes can change values. For example, being dissatisfied with one’s job for a long time
may cause a person to re-appraise the importance of work in their life. However, attitudes,
values and moods are not always congruent: a person may have negative attitudes towards
their colleagues but still help out others at work because they place a high value on being
cooperative.
142
Work Psychology
Exercise 5.1
Attitudes to this book
Consider your attitude to this book. How positively or negatively do you feel about it (affective
component)? Do you believe some or all of it to be a good support for your studies (cognitive component)? How do you behave towards it, e.g. how often do you read it and for how long (behavioural
component)? Consider whether the extent to which you use this book is influenced by your values,
attitudes and moods. What other factors affect your behaviour towards the book?
To summarise attitudes are useful because they help us to:
1
Make sense of our environment and act effectively within it.
2
Define and maintain our sense of self-identity (who we are) and self-esteem.
3
Maintain good relations with other people, particularly those who have the power to
reward or punish us. By holding and expressing attitudes similar to another person, we
can often make ourselves more attractive to them, and more able to understand and
empathise with them.
Measuring attitudes
Attitudes are often measured using self-report questionnaires. Attitude measurement usually depends upon what people are prepared to reveal about their feelings, beliefs and/
or behaviour towards the particular object in question. This is their explicit attitude. The
result of attitude measurement is normally a measure of how extreme a person’s attitude is:
for example, whether they strongly like, mildly like, are indifferent towards, mildly dislike
or strongly dislike studying work psychology. Attitude strength is influenced by a range of
factors including: the certainty people feel about their attitude; its importance to them;
how intensely they hold the attitude; how knowledgeable they are about it; how likely
the attitude is to come to mind; the intensity of the emotions attached to it; and the link
between the attitude and one’s moral values and convictions (Howe and Krosnick, 2017).
It appears that the link between attitudes and behaviour is a function of attitude extremity
and strength.
Research methods in focus
Measuring attitudes using Likert-type scales
In this approach to measuring attitudes, the psychologist selects a large number of statements that relate to the
attitude object concerned. Respondents indicate their agreement or disagreement with each. A five-, seven- or
ten-point response scale is utilised for each statement (e.g. from strongly agree to strongly disagree).
Some statements are worded negatively to check that the respondent is not simply agreeing or disagreeing
with everything for the sake of it, or not reading the items properly. For example, in a questionnaire to measure
attitudes to overtime, a positively worded item could be ‘overtime is the best part of my job’ and a negatively
worded item could be ‘overtime is a real nuisance’. Respondents might have the options ‘strongly agree’ (score 4),
‘agree’ (score 3), ‘neither agree nor disagree’ (score 2), ‘disagree’ (score 1) and ‘strongly disagree’ (score 0).
Chapter 5 Attitudes at work
143
Statements are included in the final scale only if they (i) tend to be responded to in the same way as others
covering apparently similar constructs (internal consistency reliability), and (ii) elicit the same responses on two
occasions (test-retest reliability). The person’s overall attitude score is the sum or average of their responses
to the statements. In the case of the overtime example used above, because the scores for negatively worded
statements have been reversed, a high score indicates a favourable attitude.
These measures can be subject to the social desirability effect – that is, respondents giving the socially
desirable answer, such as, for instance, ‘Of course, I have never driven faster than the speed limit’ when, in
fact, they have. Actually, almost every driver has. There are techniques available to minimise this effect. Social
desirability scales contain questions that almost everyone responds to in the same way if they are telling the
truth. Almost everyone who drives has broken a speed limit at least once, so the vast majority of people who drive
say they have broken the speed limit if they are telling the truth. Therefore, if the responses a person gives to a
series of questions like this consistently goes against that of the majority, there is an increased likelihood they
are trying to present themselves in a socially desirable way on the questionnaire. Forced choice items can also
be used to address the problem. In these questions, the respondent chooses between two desirable responses.
An example would be: ‘Is it better to be (a) helpful to colleagues or (b) well-organised in your approach to work?’
Judge and Kammeyer-Mueller (2012) argue that because attitudes develop from our experiences, they are
likely to change in response to our everyday experiences, emotions and moods. Therefore, it is a good idea to
measure them often because a single ‘snapshot’ measure cannot show how attitudes develop over time. Rapid
and frequent measurement also helps researchers to identify the events that reliably precede and follow changes
in attitudes. Experience Sampling Methodology (ESM) involves the frequent completion of brief measures,
several times per day over several days or weeks. Data collection in ESM can be carried out using inexpensive
smartphone applications with minimal disruption to the participants’ everyday activities. Researchers can track
intra-day fluctuations in individuals’ attitudes and identify the experiences and thought processes that reliably
precede such fluctuations.
Key learning point
Attitude measurement is a positivist approach (see Chapter 1), it is usually assumed that attitudes are
real and exist independently of attempts to articulate or measure them.
Research suggests that beliefs that are important to a person are remembered more completely and more quickly than those that are not (Van Harreveld et al., 2000). Does the same
logic apply to the measurement of attitudes? Implicit Association Tests (IATs) are used to
measure the speed of reaction to a stimulus. In work psychology IATs are appealing because
some argue that these identify attitudes that participants may try to hide in their responses
to questionnaires (e.g. a bias against certain groups of employees). However, responses to
IATs and questionnaire measures of attitudes appear to capture different things. Karpinski
and Hilton (2001) suggest that IATs are a measure of what people have been exposed to in
their environment rather than a measure of the views they will express and act upon, or their
explicit attitudes. Take, for example, somebody who frequently saw male colleagues being
discriminated against at work. The effects of these experiences, or environmental associations, could mean that an IAT would reveal that they have subconscious negative attitudes or
bias towards their male colleagues. Such implicit biases are malleable. In their meta-analysis
of the results of experiments, Forscher et al. (2019) found that several interventions, such as
exposure to evidence that ran counter to the implicit bias, could change it. Building on the
example above, this implies that if an employee was exposed to examples of male colleagues
144
Work Psychology
being treated fairly, their implicit attitudes to male colleagues could become more positive.
It appears, however, from Forscher et al.’s (2019) extensive and thorough study that it is
explicit biases, the attitudes that we deliberately bring to mind, that have more impact on
behaviour. Judge and Kammeyer-Mueller (2012) point out that our explicit attitudes do not
exist independently of the context in which they are expressed: there may be something to
be lost or gained expressing an attitude. Attitudes are made explicit with a purpose in mind
in an attempt to achieve one’s goals (Verkuyte, 1998). These might be to convince others of
one’s correctness, morality, acceptability, rationality and so on. For example, senior manager’s explicit attitude towards a trade union representative might emerge from a desire to foster
harmonious working relationships between management and staff. An employee’s explicit
attitudes towards the same representative could be motivated by their need for help and
support to achieve better working conditions.
Attitudes and behaviour
Research shows that attitudes predict behaviour to some extent, but not always and not
always by very much. There could be a number of reasons for this. Social pressures of various kinds, such as laws, societal norms and the views of others can all prevent a person
behaving consistently with their attitudes. Individual differences in skills, abilities and
confidence can impact on the extent to which what we like doing or believe in influences
our behaviour.
In their influential work, Pratkanis and Turner (1994) identified four of the factors they
suggest will increase the correspondence between attitudes and behaviour:
1
When the object of the attitude is both well-defined and salient. An example of a poorly
defined object would be where a person was not sure whether their immediate supervisor should be classed as a close colleague or as a member of the management team. This
would make it uncertain whether that person’s attitudes to management would affect
their behaviour towards the supervisor. Salience concerns the extent to which the object
is perceived as relevant to the situation at hand.
2
When attitude strength is high as this means the attitude comes easily to mind.
3
When knowledge supporting the attitude is plentiful and complex. This increases a person’s certainty about what they think, as well as their ability to act effectively towards
the object.
4
When the attitude supports important aspects of the self. For example, an accountant may have positive attitudes towards other accountants because they believe that
accountants (and therefore by extension themselves) perform an important role in
society.
Key learning point
The strength of the link between an attitude and behaviour depends on several features of the attitude
but also a range of other individual differences and environmental influences.
The cognitive processes concerning in the link between attitude and behaviour are articulated in the theory of reasoned action (TRA: see Ajzen and Fishbein, 1980). It includes an
extra step in the attitude–behaviour link: intention. In TRA actions are predicted by intentions. These intentions are determined by a person’s attitude and their perception of social
Chapter 5 Attitudes at work
Behaviour
145
Perceived behavioural
control
Intention
Attitude
Subjective norm
Correlative importance to
person of attitude and
subjective norm
Outcomes of
behaviour and value
of those outcomes to
the person
Figure 5.1
Beliefs about how favourably/
unfavourably people
would react if one performed
the behaviour, and motivation
to comply with their views
Theory of planned behaviour
Source: Adapted from Ajzen, I. and Madden, J.T., (1986) ‘Prediction of goal-directed behavior: Attitudes,
intentions, and perceived behavioral control’, Journal of Experimental Social Psychology, 22, 453–74.
pressure. Perception of the subjective norm involves an individual taking into account
both the opinions of others and their wish to comply with those opinions. The subjective
norm allows us to consider the influence of factors such as culture in determining thought
and action.
The theory of planned behaviour (TPB) (Figure 5.1) also includes the concept of perceived behavioural control: the extent to which the person believes that they can perform
the necessary behaviours in any given situation. It can affect both intention and behaviour.
In the TPB, attitude is defined as the beliefs and personal values about the consequences of
a specific behaviour. In summary Ajzen (1991) argued that intentions to perform particular
behaviours are predicted by a combination of attitudes towards the behaviour, subjective
norms and perceived behavioural control. In TPB intentions and perceived behavioural
control are presented as proximal influences on a person’s actual behaviour, with attitudes
being a little further back in the causal chain. TPB also acknowledges that people vary as to
whether it is their attitude or the subjective norm that acts as the key driver in determining
their intentions.
Point of integration
The theory of planned behaviour takes ideas from cognitive theories of motivation (see Chapter 6)
in proposing that actions are the product of attitudes, social pressures and intentions. TPB is also a
process theory of motivation
Ajzen also draws attention to some additional points. First, the relative importance of the
various factors in predicting intention can vary for different behaviours and across different
146
Work Psychology
situations. Second, perceived behavioural control is important but if that belief is mistaken no
amount of belief will translate an intention into behaviour. Third, the influence of subjective
norms on intentions and behaviour is often quite weak. Ajzen suggests that some people may
be much more responsive to their own attitudes than to the opinions of other people, at least
in individualistic Western cultures. In these cultures, people may be unwilling to admit to the
influence on them of other people’s opinions. We may also surround ourselves with people
who generally agree with us (see Chapter 9), so that the subjective norm is very similar to our
own attitude.
Armitage and Conner (2001) conducted a meta-analysis of tests of the theory. Based on
a total sample of nearly a million people, they found that the combination of attitude, subjective norm and perceived behavioural control correlated 0.63 with behavioural intention.
Intention plus perceived behavioural control correlated 0.52 with actual behaviour. These are
impressive findings. A more recent systematic review by McEachan et al. (2011) looked at the
results of 237 longitudinal studies that used the theory to predict health-related behaviours
(such as smoking, alcohol consumption and healthy eating). Overall, the theory accounted
for just under 20 per cent of the variation in measures of health-related behaviour. Intention
was the strongest predictor of behaviour and the theory worked best for short-term studies.
It did not predict long-term behaviour particularly well, especially outside of controlled laboratory settings. Studies using observational measures of behaviour rather than self-report
measures provide less convincing evidence for TPB. Sniehotta et al. (2014) also point out
that the theory does not include any feedback loops: a person’s behaviour and its outcomes
might have an influence on their subsequent attitudes. Evidence of the extent to which TPB
is linked to motivation is discussed in Chapter 6.
Another issue concerns the evaluation of alternative courses of action. In a research study,
a person may score quite high on intention to pursue action A, but what we do not know
because we did not ask is that they score even higher on intention to pursue action B. There
is considerable debate about the nature of perceived behavioural control. Is it, in effect,
self-efficacy, or does it include external factors such as lack of opportunity? Finally, TPB
ignores factors such as the salience of the attitude, the extent to which it is supported by
knowledge and any unconscious influences on behaviour.
Applying for a promotion
Exercise 5.2
The Theory of Planned Behaviour suggests that an employee’s attitude towards applying for a promotion will be determined partly by:
1
Their assessment of the potential consequences (both good and bad) of applying for promotion.
2
The probability that the consequences of applying for the promotion will be good / bad.
What might some of the consequences be for the employee in this example? Identify two good
consequences and two bad consequences.
Imagine that the individual considering applying for the promotion has low self-confidence. In TPB,
the consequences that matter the most to an individual are thought to have the strongest impact on
their attitude. Which type of consequences are likely to influence most their attitudes about applying
for the job: the good or bad consequences?
Subjective norm reflects whether other people thought it was a good idea to apply for promotion
and how much the employee cared about their opinions. In this example, whose opinions might the
employee care about? How motivated might they be to comply with others’ views?
Finally, is perceived behavioural control high or low in this situation? Given your answer, is intention
likely to lead to behaviour in this instance?
Chapter 5 Attitudes at work
147
Despite such potential shortcomings TPB can be applied to a number of workplace issues,
but the links between attitudes and behaviour are often more complex (see also Chapter 6).
Sniehotta et al. (2014) point to the obvious perils associated with trying to offer reliable explanations for behaviour using models and theories that contain a relatively small number of
variables. Even if theories do not offer complete explanations of the attitude–behaviour links,
they point to some important factors that influence both. We now look at two concepts of
great importance in work psychology: job satisfaction and organisational commitment. In
the former the object of the attitude is the job, in the latter the object of the attitude is the
organisation. In discussing both we draw attention to factors that interact with and work
alongside attitudes to influence employees’ behaviour.
Stop to consider
From what you have read so far, how concerned do you think managers should be about their employees’ attitudes? To what extent might employees be able to influence others’ attitudes and how could
they do this?
Job satisfaction
Judge et al. (2017) define job satisfaction as an attitude about one’s job as a whole or about
various specific facets of it (2017: 357):
At its core, job satisfaction refers to the overall evaluative judgment one has about one’s job
(Weiss, 2002). That is, satisfaction is the assessment of the favorability of a job, typically arrayed
along a continuum from positive to negative.
At the facet-level, pay, working conditions, colleagues and boss, career prospects, job demands
and many other situational variables can all influence job satisfaction (see Table 5.1). Some
authors also distinguish between expectation-based job satisfaction and performance-based
job satisfaction. The former refers to whether a person’s expectations are being met in their
work role; in contrast the latter is a measure of the extent to which the employee is satisfied with their own level of performance (see Avery et al., 2015). Job satisfaction is seen as
important for two main reasons. First, there is good evidence that it is linked to a person’s
psychological well-being (Dirlam and Zheng, 2017). Second, it is often assumed, though
not always proven, that job satisfaction will lead to motivation and good work performance
(Judge et al., 2017).
Point of integration
Relatively strong links have often been found between job satisfaction and subjective well-being. The
two appear to exert a mutual influence on each other. This is one of the reasons why measures of job
satisfaction are often used in research into work stress (see Chapter 8).
Judge et al. (2017) have suggested that there are three prominent but different viewpoints
on the causes of job satisfaction. The evidence relating to these viewpoints is discussed later
(see: Causes of job satisfaction).
Work Psychology
148
Table 5.1
A job facet-level satisfaction measure
How you feel about your job
Very much satisfaction
6
Much satisfaction
5
Some satisfaction
4
Some dissatisfaction
3
Much dissatisfaction
2
Very much dissatisfaction
1
1
Communication and the way information flows around your organisation
654321
2
The relationships you have with other people at work
654321
3
The feeling you have about the way you and your efforts are valued
654321
4
The actual job itself
654321
5
The degree to which you feel motivated by your job
654321
6
Current career opportunities
654321
7
The level of job security in your present job
654321
8
The extent to which you may identify with the public image or goals of your organisation
654321
9
The style of supervision that your superiors use
654321
10
The way changes and innovations are implemented
654321
11
The kind of work or tasks that you are required to perform
654321
12
The degree to which you feel that you can personally develop or grow in your job
654321
13
The way in which conflicts are resolved in your company
654321
14
The scope your job provides to help you achieve your aspirations and ambitions
654321
15
The amount of participation which you are given in important decision-making
654321
16
The degree to which your job taps the range of skills which you feel you possess
654321
17
The amount of flexibility and freedom you feel you have in your job
654321
18
The psychological ‘feel’ or climate that dominates your organisation
654321
19
Your level of salary relative to your experience
654321
20
The design or shape of your organisation’s structure
654321
21
The amount of work you are given to do, whether too much or too little
654321
22
The degree to which you feel extended in your job
654321
Source: © 1988 Cooper, Sloan and Williams. Reproduced with permission of the publisher ASE, College Lane, Hatfield, Herts, AL10 9AA, UK.
Chapter 5 Attitudes at work
149
1
Job satisfaction is a relatively stable disposition that is either the learned product
of experience (Griffin and Bateman, 1986; Staw et al., 1986) or an attitude that
stems from a person’s genetic inheritance (Judge et al., 2012). This the dispositional
approach.
2
The social information processing approach suggests that job satisfaction is determined
by experiences and information provided by others at work (Salancik and Pfeffer, 1978;
O’Reilly and Caldwell, 1985). In this approach job satisfaction is at least in part a function
of how other people in the workplace interpret and evaluate what goes on. For example, if
your colleagues show that they are satisfied at work then their opinions have an influence
upon your own satisfaction levels.
3
The cognitive information-processing approach suggests that job satisfaction is the result
of the accumulation of cognitive information about the workplace and one’s job. In other
words, job satisfaction is influenced most directly by the characteristics of the job (see
also Chapter 6), and the extent to which those characteristics match what that employee
wants.
Key learning point
Job satisfaction can be seen as a function of: (i) a person’s general personality or disposition; (ii) the
opinions of other people in the person’s workplace; or (iii) the features of a person’s job. It is likely to
be the result of the interactions of all three.
Measuring job satisfaction
Examples of measures of job satisfaction include the very widely used and extensively validated Job Descriptive Index (JDI; Smith et al., 1969), the Job Satisfaction Scales of Warr et
al. (1979), the Overall Job Satisfaction scale (Brayfield and Rothe, 1951); the Job Satisfaction
Survey (Spector, 1985); and the job satisfaction scale of the Occupational Stress Indicator
(OSI; Cooper et al., 1987). When completing some measures respondents can be asked to
indicate what they think and/or feel about their job as a whole (global satisfaction). In others
that may be asked about specific aspects of it (facet-level). Likert scaling (see earlier in this
chapter) is often employed.
It is generally assumed attitude measurement that it is better to ask lots of questions rather
than only one general question: longer and repeated measurement tends to provide more reliable data. The facet-level approach to measurement focuses on the cognitive component of job
satisfaction: people are asked to evaluate in detail various job conditions in a logical and rational
manner. Many researchers have found that non-work factors (such as family problems) can
influence overall job satisfaction and these are rarely measured directly in facet-level job satisfaction scales. ‘Adding up’ cognitive evaluations of different aspects of a job might not capture
very accurately an employee’s overall feelings. As Taber and Alliger noted (1995: 118):
Perhaps workers form a gestalt – a perception of pattern – about their jobs that is not a simple
linear function of task enjoyment . . . a worker might perform 15 different enjoyable tasks; nevertheless, the worker’s global job satisfaction still could be low if the 15 tasks were so unrelated
to one another that the total job was not meaningful, or did not relate clearly to the mission of
the organisation.
Shorter measures with more general questions about how the person feels about their work
overall or in general may better capture the affective component of the attitude (Thompson
and Phua, 2012). Many definitions of job satisfaction emphasise this component so this
150
Work Psychology
would seem important. Thompson and Phua (2012: 301) have developed a four-item measure of job satisfaction designed to tap directly into this component:
1
I find real enjoyment in my job
2
I like my job better than the average person
3
Most days I am enthusiastic about my job
4
I feel fairly well satisfied with my job
Response categories: 1 = Strongly disagree, 2 = Disagree, 3 = Neutral, 4 = Agree,
5 = Strongly agree. Note: the measure also includes distractor items available in the article.
The coverage of attitudes earlier in this chapter suggests that job satisfaction is likely to be
used by people to help make sense of their work, and to define who they are, or are not. Locke
(1995) pointed out that job satisfaction will depend partly on how well people’s tasks fit their
long-term purposes, how much their self-esteem depends on their job and which job experiences are processed most thoroughly in their memory. These cognitive influences might also
therefore play a role in determining workers’ responses to overall measures of job satisfaction.
Key learning point
Job satisfaction is more than how much the person enjoys the job tasks. It also depends on how important the job is to the person, and how well it fits in with their long-term aims.
Questionnaire measures of job satisfaction might not travel well across cultures. Translations may be imperfect: people in different countries may understand the same words or
phrases in ways that the questionnaire designer had not anticipated or intended. In addition,
the interpretation of response scale options such as ‘quite’ or ‘often’ may vary accordingly to
cultural norms. These problems might matter to managers who want to know, for example,
whether employees working in a factory in one country are more satisfied than those in a factory in another country. In a study of 14,446 employees from 24 different and diverse nations,
level of income, independence and good working relationships were found to be linked to job
satisfaction to a similar extent across all of the cultures studied (Hauff et al., 2015). However,
national cultures seemed to moderate a little the strength of the links between job satisfaction
and job security, and between job satisfaction and the extent to which the job was interesting. For workers in individualistic cultures having interesting work to do was more strongly
linked to overall job satisfaction than it was in collectivist cultures. In cultures where status
differences were marked (high power–distance) job security was more strongly linked to job
satisfaction than it was in low power–distance cultures. Job security was also less important in
determining job satisfaction in countries with high uncertainty avoidance. At first this may
seem odd: a person who avoids uncertainty might seem likely to value job security. Hauff and
colleagues argue that in these countries the prevailing culture means that there tends to be
better support systems for people who find themselves without a job. Chapter 12 contains
fuller definitions of the features of these cultures.
Key debate
Is a satisfied worker a productive worker?
Job satisfaction is a non-specific attitude in terms of action, so it could be a common cause for a link to many work
behaviours. Drawing on a wide range of published research, Judge et al. (2001) estimated the average correlation
between job satisfaction and job performance was 0.30.
Chapter 5 Attitudes at work
151
Harrison and colleagues (2006) found that job satisfaction was not only linked to job performance but also with
withdrawal behaviours such as absence and lateness. However, the links between job satisfaction and absence are generally quite small and inconsistent. Absence is also influenced by organisational culture and norms and the individual’s
situation outside of work (e.g. how important the job is to them). Workers’ personality may also have an impact on the
extent to which dissatisfaction translates into absence. For example, the links between dissatisfaction and absence
from work are especially weak among those with a propensity to feel guilty about not meeting others’ expectations of
them (see Schaumberg and Flynn, 2017).
Returning to the link between job satisfaction and job performance Judge et al. (2001) identify six possible reasons why
these might be related (see Figure 5.2). Demonstrating that two variables are correlated is not the same as demonstrating
that one causes the other. Riketta (2008) re-examined the satisfaction–performance relationship but his meta-analysis
included fewer (14) studies that used strong longitudinal research designs. After controlling for initial levels of performance, this revealed a much weaker link between job satisfaction and later performance than the relationship between
job satisfaction and performance found by Judge et al. (2001). This may be because Judge et al. had included many studies
with cross-sectional survey designs. There was also no evidence that good performance consistently caused higher job
satisfaction as Judge had suggested. Interestingly, Riketta also found that the link between satisfaction and performance
was strongest when there was a short time gap between the measurement of satisfaction and performance, indicating
that any effect of satisfaction on performance could be short-lived. This may occur because high performers develop
higher expectations of the rewards that are due to them and when these are unmet, performance levels may soon drop.
Of course, such findings do not rule out the possibility that in some situations the satisfaction–performance link is strong,
but it does not appear to be consistently strong across a range of different longitudinal studies.
Others have looked at the connection between collective levels of job satisfaction among a group of employees
and the performance of that group. This approach makes sense because productivity, customer satisfaction, absence,
employee turnover and so on result from collective efforts. This tests the ‘satisfied workplace–productive workplace’
hypothesis. Imagine if you were a high-performing and satisfied individual but most of your colleagues frequently
expressed a dislike of their own work situation and were performing badly. Over time you might end up sharing their
perceptions and as a consequence their performance levels.
Whitman et al. (2010) carried out a meta-analysis of 73 studies to test the group-level links between satisfaction
and various organisational outputs. They found a relatively strong connection between group-level satisfaction and
measures of output including productivity, client satisfaction and absence. The satisfaction–output links were strongest
when there was good consensus among employees regarding their level of job satisfaction. When a group of employees
are in agreement with each other about their levels of job satisfaction, it is most likely to be linked to their collective
outputs. One possible implication of this finding is that interventions should be aimed at lifting satisfaction for a group
of employees to a similar shared level.
It may also be that employees are likely to be more satisfied working for a company that is doing well. Ford and
Wooldridge (2012) tested this hypothesis by looking at how much various sectors of US industry had grown their revenue over a three-year period leading up to a questionnaire survey. Workers in growing industries reported job enrichment (doing new and varied work tasks, getting good opportunities to be creative at work, having more autonomy
and opportunities to use their skills, making more decisions and having good opportunities to grow and develop their
competencies). The authors argue that this enrichment mediates the relationship between company performance
and job satisfaction. It is also feasible that job enrichment could be a cause of both job satisfaction and improved
company performance.
Causes of job satisfaction
As predicted by the cognitive-information processing approach, many determinants of
global job satisfaction derive from the intrinsic features of the work. These are most commonly based on the Hackman and Oldham (1976) core constructs of skill variety, task identity, task significance, autonomy and feedback (see Chapter 6 for a detailed account of these
factors). Plenty of studies have established that various features of work design are linked
Work Psychology
152
1.
JS
JP
Job satisfaction causes job performance, i.e. people tend
to work harder and/or better because they like their job
2.
JS
JP
Job performance causes job satisfaction, i.e. people tend
to like their job because they are successful at it
3.
JS
JP
Job satisfaction and job performance cause each other,
i.e. both 1 and 2 apply
4.
JS
JP
Job satisfaction and job performance are correlated, but
only because of another variable (C) that affects them
both. For example, the clarity of the job’s requirements
may help both satisfaction and performance
JP
Job satisfaction and job performance may be causally
linked, but the strength of this link depends on some
other variable (C), for example, the extent to which
successful performance is rewarded
JS
JP
Job satisfaction and job performance are specific
instances of more general constructs of positive feelings
and personal effectiveness, and these have a causal
impact on each other
Affect
Performance
C
5.
JS
C
6.
Figure 5.2
Some possible relationships between job satisfaction (JS) and job performance (JP)
Source: Adapted from Judge et al. 2001. The American Psychological Association.
to job satisfaction. As predicted by the social information processing approach factors such
as perceptions of distributive justice (the fairness with which rewards were distributed in
the organisation) also predict job satisfaction (Agho et al., 1993). In addition, O’Reilly and
Caldwell (1985) demonstrated that both task perceptions and job satisfaction of workers
were influenced by the opinions of others in their workgroups.
Point of integration
Work design has been extensively linked to job satisfaction (see Chapter 6). The material on job characteristics such as skill variety, task identity, task significance, autonomy, feedback (etc.) is especially
relevant to the understanding of the causes of job satisfaction. Therefore, it is important to consider
in detail the material on job design in Chapter 6 when analysing the determinants of job satisfaction.
There is also a good amount of evidence that some people are simply more satisfied than
others by their nature (the dispositional approach). For some people their stable disposition
to feel happy or sad, or positive or negative, drives the affective component of job satisfaction.
Chapter 5 Attitudes at work
153
Evidence suggests that the job satisfaction levels of people with a high need for affect (a tendency to approach situations that are charged with emotions for themselves or others) tend
to be determined more by their prevailing emotions than by their ratings of various features
of their work environment (Schlett and Ziegler, 2014). High levels of extraversion tend to be
associated with high levels of job satisfaction (Avery et al., 2015). Arguably, this is because
extraverts tend to be prone to experience positive mood. The enthusiasm facet of extraversion,
the tendency to be friendly, sociable and to exhibit a positive demeanour, appears to drive
this link between extraversion and job satisfaction (Harari et al., 2018). Mindful individuals
might also experience higher job satisfaction because they tend to view their experiences at
work in ‘a receptive, non-judgmental way, they observe stressful events more objectively and
refrain from attaching a meaning or evaluation to it’ (Hülsheger et al., 2013: 312). Mindfulness can be developed and appears to change the way that people interpret their experiences,
something that may have implications for their mood and job satisfaction (see Chapter 8).
Core self-evaluations (CSEs; see Chapter 8) have been found to predict job satisfaction
in a number of studies (see for example Keller and Semmer, 2013). This relationship appears
to work in both directions. Using data from a longitudinal study of almost 6000 workers, Wu
and Griffin (2012) found that the experience of high levels of job satisfaction also predicted
CSEs. For some workers at least, extended periods of high levels of job satisfaction could
have a positive impact on the way they see themselves in terms of their effectiveness, sense
of self-worth and capability.
There is little doubt that disposition and situational factors combine to determine job
satisfaction. Bowling et al. (2006) found that job satisfaction was fairly stable (which suggests disposition is important), but that it changes when people change jobs (which suggests
the situation is important). Perhaps more significantly they found that when people change
jobs, people with a positive disposition tended to experience gains in job satisfaction, while
those with a negative disposition tended to experience losses in job satisfaction. It may be
that people with a positive disposition are more likely to look for, be aware of and remember
positive experiences (Harari et al., 2018). By studying young workers over a period of five
years early in their careers, Keller and Semmer (2013) found that growth in both CSEs and an
important job characteristic – job control – predicted job satisfaction. In this case the effects
of both situational and dispositional factors were roughly similar. What they also found was
that when CSE or job control was stable for a period, job satisfaction actually decreased. They
argue that, at least among younger workers, sustained personal development and continuous improvements to work situation are significant influences on job satisfaction. The ‘fit’
between a worker’s personality and the demands of their work role can also influence job
satisfaction. Christiansen et al. (2014) found that when workers were asked to carry out tasks
requiring high agreeableness and conscientiousness, those with high levels of these traits
experienced the highest levels of satisfaction.
Researchers have also attempted to go beyond self-reported disposition to isolate the
genetic influences on job satisfaction. In a sophisticated study of both identical and
non-identical twins, raised either apart or together, Judge et al. (2012) found that 35 per cent
of the variance in job satisfaction between these Swedish individuals appeared to be due to
genetic factors. However, twin studies estimate the effects of stable individual differences on
job satisfaction as that which is not accounted for by situational factors. This assumes that
all of the important situational factors have been taken into account – perhaps an optimistic
assumption.
The links between gender and job satisfaction are similarly complex. A number of studies
have found a gender paradox, i.e. women tend to be over-represented in work roles with
relatively poor pay and promotion prospects yet report higher levels of job satisfaction than
men (Clark, 1997). Summarising the research, Gronlund and Oun (2018) hypothesise that
this may be due to: (1) women having lower expectations than men, something that may
disappear in time as opportunities in the labour market become more equal, or (2) an acceptance by women of a work–family trade-off, in which taking a job with lower demands and
154
Work Psychology
more flexible work arrangements (e.g. hours and schedules) compensates for low pay and
poor career prospects. In their large study of Swedish workers Gronlund and Oun (2018)
did not find any evidence that women had lower expectations or were simply accepting the
work–family trade-off described above. Zou (2015) argues that there is evidence that women
value the support associated with good working relationships at work more than men, and
that the provision of flexibility around work hours and schedules is often interpreted by them
as support. Therefore, it may be that the strength of the link between some working conditions (e.g. support) and overall satisfaction differs between men and women.
Changes in job satisfaction over time
An extensive meta-analysis by Ng and Feldman (2010) identified that increases in job satisfaction with age were reliably observed in a wide range of different studies from across the globe.
Why might job satisfaction tend to increase fairly steadily through working life? There are
several possible reasons (see Dobrow Riza et al., 2018). First, older people may be in better jobs
than younger ones, since they have had longer to develop their skills and to find a job that
suits them. Second, older people may have changed their expectations of what is important
to them over the years, so that they more readily draw satisfaction from what they experience
at work. Ng and Feldman (2010) suggest that older workers may be inclined to focus on emotionally fulfilling work activities – and worry less about acquiring new skills. This could be a
result of them becoming increasingly aware that their remaining time at work is limited and
a desire to ‘make the most of it’. In both mechanisms, older workers are taking steps to adapt
to and cope with changing circumstances. Truxillo et al. (2012) argue that mechanisms such
as these are the result of proactive self-regulatory processes. Older workers may be adjusting
the way they work, or seeking out different jobs that better fit with their capabilities and goals
(e.g. older workers might seek out complex work tasks as these require the experience and
knowledge that they have developed).
The general trend for an increased use of teamworking may also boost the job satisfaction
levels of older workers as the autonomy and interdependence on offer allow them to be more
selective about how they use their skills as well as providing the opportunities for social interaction (Truxillo et al., 2012). Two other mechanisms might also come into play. Older people
might always have been more satisfied than younger ones – a so-called cohort effect. Finally,
dissatisfied older people may be more likely than younger ones to opt for early retirement or
voluntary redundancy, so that those remaining in employment represent a biased sample of
satisfied older people.
Key learning point
Although age and tenure are often correlated, they appear to have different relationships with job
satisfaction.
Dobrow Riza et al. (2018) analysed data from a large longitudinal study that revealed that
as tenure increased job satisfaction decreased and that this effect was independent of age.
This can occur because of the association between tenure and boredom at work or because
tenure brings with it an increased awareness of the flaws of the organisation and subsequent
sense of disillusionment. Their study also showed that changing jobs provided a boost to job
satisfaction. Unfortunately, some of the boosts to job satisfaction that occur with changing to
a different occupation with a new employer may represent a ‘honeymoon period’. Zou et al.
(2017) found that after such a change satisfaction with the work itself dropped over time, and
Chapter 5 Attitudes at work
155
dropped more than it did for those changing employer but not occupation. They hypothesise
that this effect occurs because changing employer and occupation can damage and disrupt
important sources of job satisfaction (such as social networks and one’s professional identity)
and can shatter overly optimistic expectations about the new role.
What about factors outside of work: might these spill over, psychologically, into work?
One possibility is that positive life events improve subjective well-being, something that is
linked to job satisfaction. Analysing longitudinal data from almost 16,000 UK adults, Georgellis, Lange and Tabvuma (2012) found that job satisfaction went up significantly prior to a
person’s first marriage and, to a lesser extent, prior to the birth of a first child. Perhaps overall
levels of life satisfaction can spill over into the work domain during the anticipation of these
happy events. After the birth of the first child, job satisfaction levels were found to drop
significantly with the researchers concluding that this could be due to added financial and
time pressures and work–life conflict outweighing the boost to life satisfaction associated
with starting a family. A smaller drop in job satisfaction was observed after a first marriage:
increased work–life conflict might also account for this finding. It is also possible that culture
plays a role in such findings. Haar et al. (2014) found that work–life balance had a bigger
impact on job and life satisfaction in individualistic cultures than it did in collectivist cultures. The authors suggest that this could be because in individualistic cultures, responsibility
for achieving a good balance falls very much on the individual worker and that it is a particularly critical aspect of their life. Those that achieve a good balance are therefore more likely
to derive satisfaction from their achievement.
Key learning point
Job satisfaction is partly determined by individual differences, but not so much so that it is constant
over a person’s working life.
Organisational commitment
What is organisational commitment?
Organisational commitment is what many employers say they want from employees. It
implies loyalty and the willingness to go the ‘extra mile’. It has been suggested that commitment will lead to so-called organisational citizenship behaviours (see Chapter 4). We look
more closely at the validity of these claims later.
Organisational commitment was defined by Mowday et al. (1979) as ‘the relative strength
of an individual’s identification with and involvement in an organisation’. It fits nicely with
the earlier definition of an attitude: it is targeted towards an object and has strength and
direction. This concept is often thought to have three components (Griffin and Bateman,
1986): (i) a desire to maintain membership in the organisation; (ii) belief in and acceptance
of the values and goals of the organisation; and (iii) a willingness to exert effort on behalf of
the organisation.
Judge et al. (2017) point out that commitment is less about the pleasure that one gets
from interacting with the organisation (as reflected in job satisfaction) and more about the
fulfilment of one’s values through employment. If a person is committed to an organisation,
therefore, they have a strong identification with it, value membership, agree with its objectives and value systems, are likely to remain in it, and, finally, are prepared to work hard on its
behalf. The concept of organisational commitment used in contemporary research is slightly
differently from that described above. For example, with regard to point (ii), Ng (2015) draws
156
Work Psychology
a distinction between organisational commitment and identification, with the latter being a
‘deep-level psychological phenomenon’ (2015: 155) somewhat stronger than an attitude. This
is characterised by the employee coming to believe that their values match those espoused
within the organisation.
Allen and Meyer (1990) argued that there were three types of commitment. We use this
very influential model throughout this section as it remains widely used by researchers and
practitioners:
■
Affective commitment: essentially concerns the person’s emotional attachment to their
organisation or their liking for it (‘I want to stay’).
■
Continuance commitment: a person’s perception of the costs and risks associated with leaving their current organisation (‘I need to stay’); this includes both the personal sacrifice
that leaving would involve, and a lack of alternatives available to the person.
■
Normative commitment: a moral dimension, based on a person’s felt obligation and responsibility to their employing organisation (‘I feel I ought to stay’).
Critics of the model have argued that it is a little muddled because it combines an attitude
towards a target (the organisation) with attitudes toward a behaviour (leaving or staying).
Given its focus on decision-making about whether or not to stay evidence shows that it is best
used as a means of predicting turnover (Judge et al., 2017; Solinger et al., 2008). Other observers have pointed out that people feel multiple commitments at work – not only to their organisation, but also perhaps to their location, department, work group or trade union (Barling
et al., 1990). Employees distinguish between different targets of their commitment, i.e. commitment to the organisation, their colleagues and their supervisors. Wasti and Can (2008)
showed that commitment to the organisation predicted organisational outcomes (such as
turnover), and commitment to the supervisor predicted citizenship behaviours (OCBs).
Becker et al. (1996) also found that commitment to the supervisor, especially when it goes
deeper and the employee internalises the manager’s values, is more highly correlated with
job performance than the less specific organisational commitment. Interactions between
these different commitments can be important. Vandenberghe and Bentein (2009) found
that low affective commitment to the supervisor was a good predictor of employee turnover,
but that when coupled with low affective commitment to the organisation the chance of
employee turnover was significantly increased. There is also a wider issue here: what exactly is
the organisation that the employee is committed to? Complexities such as parent companies
and franchises can make it difficult to identify exactly to which organisation one belongs.
Most organisations consist of various factions with somewhat different and possibly even
contradictory goals. It seems that most people think of the term organisation as meaning top
management when thinking about commitment, unless they are explicitly asked to consider
other objects (such as their supervisor or work team).
Key learning point
Organisational commitment concerns a person’s sense of attachment to their organisation. The object
of the commitment (e.g. supervisor, colleagues, senior management) may be important in determining
how commitment impacts upon their behaviour.
Measuring organisational commitment
A number of questionnaires have been developed to measure the various aspects or theories of
commitment discussed above. Allen and Meyer’s (1990) measure has stood up well to psychometric scrutiny. The Organisational Commitment Questionnaire (OCQ), which was developed
Chapter 5 Attitudes at work
157
by Mowday et al. (1979), is a 15-item questionnaire which has been used as a total commitment scale. It has also been broken down into subscales by various researchers (Bateman and
Strasser, 1984). The OCQ comprises items such as, ‘I feel very little loyalty to this organisation’,
‘I am willing to put in a great deal of effort beyond that normally expected in order to help
this organisation be successful’, and ‘I really care about the fate of this organisation’. The OCQ
was designed before the distinctions between different types of commitment were articulated
in the literature and chiefly reflects affective commitment. Warr et al. (1979) developed a
nine-item scale that includes the item: ‘I feel myself to be part of the organisation’. This measure also tends to concentrate on affective commitment. Most of these measures are used to
provide a ‘snapshot’ of employee commitment. Some argue that attitudes such as commitment
are very dynamic and fluctuating and can change quickly in response to events that change an
employee’s mood (such as an argument with a supervisor or a customer). Affective event-based
approaches to the measurement of commitment are designed to capture these fluctuations:
this approach involves frequent measurement using brief measures (Judge et al., 2017).
Exercise 5.3
Three different types of commitment
Example questions from Allen and Meyer (1990) include: ‘I think that people these days move from
company to company too often’; ‘Right now, staying with my organisation is a matter of necessity
as much as desire’; and ‘I enjoy discussing my organisation with people outside it’. For each of these
three items, try to identify which type of commitment is being measured: affective, normative or
continuance.
These measures ask employees to evaluate their own commitment. Perhaps other people
could provide a more dispassionate view of how committed we appear to be. Goffin and
Gellatly (2001) found that self-ratings of commitment were only moderately correlated
with ratings given by supervisors and work colleagues. Ratings of people’s commitment
made by their supervisors and colleagues were more similar to each other than either was to
self-ratings. This suggests that an external perspective is providing something over and above
self-perceptions. It may be that commitment is a private psychological experience not fully
reflected in behaviour or that people are not good judges of how their commitment translates
into overt behaviour. It is also likely that the word commitment means different things to different people. Singh and Vinnicombe (2000) have shown that, for many people, it has more
to do with taking a creative and assertive approach to one’s work than with involvement in
and loyalty to one’s employing organisation. These differences in perceived meaning could
translate into different behaviours.
Key learning point
Organisational commitment is usually measured using self-report questionnaires containing multiple
items and Likert scales.
Causes of organisational commitment
There are some complex models and theories of the possible causal connections between
commitment, satisfaction, motivation, job characteristics and other variables (see, for example, Eby et al., 1999 and Judge et al., 2017). Figure 5.3 represents our attempt to summarise
what is known, and what seems likely.
Work Psychology
158
Organisational commitment
Affective
Characteristics
of the
organisation
Nature of
the job
A person’s
experiences
at work
Belief in the values
of the organisation
Emotional
attachment to the
organisation
Normative
Sense of loyalty
or obligation to
the organisation
Continuance
Characteristics of
the person
(including his or her
motives for joining
the organisation)
Organisational
citizenship
behaviours
Intention to
stay in the
organisation
Investments in
the organisation
Unavailability of
alternatives
Labour
market
constraints
Note: Dotted arrows indicate weak effects; solid arrows indicate strong effects.
A model of organisational commitment
Figure 5.3
There are various approaches to understanding commitment. The behavioural approach
suggests that people examine their own behaviour and conclude that since they did something with significant consequences in full view of others, when they could have chosen
not to do so, they really must be committed to it (Mabey, 1986). Discretionary investment
of effort is a predictor of affective commitment: when employees invest lots of resources in
a task they need to find a reason for doing so. Flynn and Schaumberg (2012) found that people who experience high levels of guilt when they get things wrong (guilt-proneness) tend
to make extra efforts to put things right. Their results indicated that liking the employing
organisation (reporting high affective commitment) gives guilt-prone employees a justification their extra efforts.
More commonly, however, it has been suggested that people’s commitment can be fostered by giving them positive experiences at work: this is the social exchange approach. The
person is essentially saying, ‘if this work organisation is nice to me, I will be loyal and hardworking’. Many researchers have tried to identify exactly which pleasant experiences matter
most for organisational commitment. Morrow (2011) summarises five categories of experiences that could, in theory, influence commitment:
■
Socialisation practices. These occur around the point of entry to the organisation.
High-quality selection processes designed to identify the right employees have been
shown to foster good person–organisation fit. Giving new employees good experiences
Chapter 5 Attitudes at work
159
during induction and early tenure also fall into this category of experiences. These
include clear work roles, challenging tasks and good support from a mentor.
■
Organisational changes. These include the introduction of teamwork, using technology to
make work more interesting (for example the automation of mundane or repetitive tasks),
and high worker involvement/participative management practices (see for example
Chapters 8 and 12).
■
Human resource practices. These include performance appraisal, performance management
and training practices that are designed to develop and reward commitment.
■
Interpersonal relations. These include experiences of working with inspiring and motivating leaders, good mentoring and being around co-workers who are helpful and
courteous.
■
Employee–organisational relations. This category of experiences centres on the extent to
which employees perceive that the organisation is supportive of their personal goals,
values their work efforts and takes an interest in their well-being.
Morrow felt that more evidence was needed in order to draw firm conclusions about which
of these experiences reliably leads to improvements in commitment. For some experiences
(e.g. teamwork) there was evidence of both negative and positive effects on commitment.
This may be a consequence of the way that the implementation of teamwork is handled (see
Chapter 9). Some employers offer various accommodations and adjustments to employees
according to their particular needs (e.g. additional training and development, special levels
of pay, opportunities for promotion) in order to stimulate commitment. Ng and Feldman
(2010) point out that these idiosyncratic deals are costly to employers but may foster trust and
excitement and, in turn, bolster affective commitment. By gathering data from 375 managers
they tested this proposition and found that around 13 per cent of the variance in affective
commitment reported after eight months could be attributed to contract idiosyncrasy. Such
effects are not huge and it remains to be seen whether the benefits of increased commitment
outweigh the costs of these tailored arrangements.
Immediate supervisors’ actions can play an especially important role in determining
organisational commitment when employees see the supervisor as the embodiment, or representative, of the organisation. They may interpret what their supervisor does as an indication
of how much they are valued by the organisation. In such circumstances, the quality of that
relationship (the Leader-Member Exchange: see Chapter 10) impacts on affective commitment (Eisenberger et al., 2010). Findings such as these also suggest that when the object of
an attitude is unclear, as may be the case with an ‘organisation’, employees’ will look to other
relevant objects such as their supervisor when forming their attitudes.
Key learning point
Organisational commitment can be fostered by positive experiences at work, and to some extent by
the circumstances in which the person joined the organisation.
Cognitive approaches to commitment focus on identifying the features of the job that
are reliably linked to commitment. Factors intrinsic to the job (e.g. challenge, autonomy)
are generally found to be more important in fostering affective commitment than extrinsic
factors such as pay (Arnold and Mackenzie Davey, 1999). Stressors associated with poorly
designed or managed work, such as high demands, conflict and low autonomy have all
been linked to lower levels of affective commitment (Cheng and Chan, 2008). In contrast, continuance commitment is heavily influenced by the person’s perception of their
160
Work Psychology
past contributions to the organisation and their present attractiveness to other potential
employers (Meyer et al., 1989). Frone (2018) found that during a recent recession in the
US, significant drops in affective commitment were reported and also linked perceptions
of lower job security. This may be because employers interpreted the loss of job security
as a significant breach of their psychological contract by their employer (see later in this
chapter).
Fisher’s (2014) huge study of over 6000 employees of a multinational company spread
across 18 countries sheds some light on the factors that might protect commitment, even in
the face of the very stressful working conditions associated with high workloads. He found
that when employees reported working in a cooperative climate the negative impact of overload on commitment was less keenly felt. Empowerment also seemed to help but only in
some cultures, namely those where status differences are less pronounced (where sharing
responsibility and power is the norm).
Employees’ perceptions of how supportive the organisation is of their family also seem
to be important. This support can include: the employer giving members of staff time to
attend to family issues, allowing them to deal with personal issues during work time and not
requiring employees to perform work roles that have a negative impact on their life outside
of work. Wayne et al. (2013) found that when this support was high, employees were more
likely to report that their family life had improved. They then attributed the cause of this good
outcome to the source of the support (the work organisation): this attribution strengthened
commitment.
As with job satisfaction it has been argued that commitment is also partly a function of
the person. Longitudinal studies have shown that organisational commitment is relatively
stable over long periods of time (Bowling et al., 2006). Perhaps some people, through their
personality or disposition to feel positive emotions, feel highly committed. Positive affect
(defined in Chapter 8) tends to be associated with high levels of organisational commitment,
and negative affect with low levels of commitment (Thoresen et al., 2003).
There are some important links between the Big Five personality traits and organisational commitment. Choi et al.’s (2015) meta-analysis of these relationships found that
those high in agreeableness may be particularly motivated to fulfil their need for trust and
loyalty through committing to the organisation. They may also look to develop and maintain good close working relationships with colleagues, making their work more pleasant
and enjoyable thus further strengthening their liking for their employer and their sense
of moral obligation. The effects of agreeableness on commitment were stronger in collectivist cultures where group outcomes tend to be valued more than individual ones. Those
with high levels of emotional stability, extraversion and openness to experience tended to
show lower levels of continuance commitment. Compared to most people, such individuals are likely to be less worried or apprehensive about moving to a new employer, more
motivated to meet new work colleagues and keener to engage in new work activities or
to adjust their values to fit with a new work setting. However, some perceived organisational values seem to influence commitment regardless of individual differences. Finegan
(2000) found that the extent to which the organisation was perceived to value ‘humanity’
(e.g. courtesy, fairness) and ‘vision’ (e.g. initiative, openness) was correlated with affective
commitment, while the value of ‘convention’ (e.g. cautiousness) was correlated with continuance commitment.
Key learning point
A good amount of evidence indicates that several stable individual differences are linked to organisational commitment. Commitment is not just a result of what people experience at work.
Chapter 5 Attitudes at work
161
Key debate
Is organisational commitment linked to performance?
Meyer et al. (1989, 1993) and Goffin and Gellatly (2001) found that workers reporting high on affective commitment
to their organisation tended to be better performers than those reporting low affective commitment. The opposite
pattern of results was observed for continuance commitment. This makes sense: high continuance commitment is
based partly on a perceived lack of employment options, and one reason for people lacking options may be that they
are not very good at their job.
Other research has failed to find these links between commitment and performance. Many factors intervene
between attitude and behaviour. One of these might be ability. A person is unlikely to perform well at a task that they
are not able to do, even if highly committed to the organisation. Therefore, one might expect a stronger link between
commitment and performance for aspects of performance that depend more on motivation, such as being a pleasant
or supportive colleague, than they do on ability.
Consistent with this, Harrison et al. (2006) reviewed relevant longitudinal research and found a significant link
between commitment and subsequent contextual performance (see organisational citizenship behaviours, Chapter 4)
of about 0.28, compared to 0.16 for core job performance. A meta-analysis of five longitudinal studies (Riketta, 2008)
suggests that the correlations are somewhat smaller than that found in earlier studies, with the links between organisational commitment on the one hand and focal and contextual performance on the other hand being about the same.
Meta-analyses draw together lots of different studies in order to identify consistent and reliable effects. In some
circumstances the links between attitudes and performance are stronger than in others. For example, some people
feel an especially strong motivation to carry out a particular job or profession because of their particular talents. This
is sometimes referred to as a ‘calling’. Someone with a strong calling might say, for example, ‘I was always meant to be
a priest’ and we might reasonably expect that the strength of this calling has implications for their performance. Kim
et al. (2018) found that those with a strong calling tended to report more affective commitment to the organisation,
with this commitment then being linked to supervisor rated performance. In other words, their positive attitude to
their employing organisation played a key role in determining their subsequent performance.
A relatively consistent finding in the research literature is that a person who does not feel committed to their employing organisation is more likely to want to leave it, and actually to do so, than a person who feels more committed (Judge
et al., 2017). Intention to leave the organisation is the strongest and most often reported correlate of low organisational
commitment (Solinger et al., 2008). However, intention to leave does not necessarily translate into actual leaving: much
may depend on the alternatives available to an employee entering the labour market (see Theory of Planned Behaviour).
Exercise 5.4
Satisfaction and commitment
Using your own experience of work, what factors most affect your job satisfaction? And what factors
affect how committed you feel towards an employer? Are they the same things? Consider the extent
to which you think people can be ‘made’ more satisfied and/or committed by changing things in their
work environment. In your experience, are some people just contented (or not) by nature, almost no
matter what happens to them?
Employee turnover
Research has shown that intention to leave is not only influenced by attitudes towards the act
of leaving, but also attitudes towards objects: whether the person likes the job itself and their
employing organisation. This is the affect-initiated pathway to voluntary turnover: employees
who are dissatisfied with their job or their employer, start looking for other opportunities and
162
Work Psychology
leave when they have secured a good alternative (Hom et al., 2017). This pathway has some
validity, but things are rarely so simple. Research has consistently found that job satisfaction
and organisational commitment are better predictors of intention to leave than they are of
actual employee turnover. Various forces within and outside the organisation, as well as stable
and fluctuating individual differences, exert significant influences on employees’ decisions
to leave their job.
Employee turnover is an issue that people within organisations like to be able to monitor
and to try to control. It has costs and can present opportunities. Costs include hiring and
training new staff, disruption to teamwork, organising people to cover the work done by
people who have left and so on. It can also lead to an influx of fresh talent, new ideas and
enthusiasm, as employees who are not performing well or do not feel that the job is right for
them leave a company.
A comprehensive meta-analysis of the effects showed that higher levels of turnover were
most strongly associated with lower levels of customer service and measures safety, with there
being weaker but still significant relationships with productivity (Hancock et al., 2013). These
effects were not consistently large. Hausknecht and Holwerda (2013) conclude that particularly problematical turnover occurs when:
■
experienced people leave many parts of the organisation, leaving high proportion of
relative newcomers working across the organisation
■
many people leave in a short period of time
■
people leave who have especially important roles within the organisation
■
the knowledge, skills, abilities and experience of newcomers does not sufficiently replace
that possessed by those who have left.
One of the problems with collecting data on turnover is that once employees have left
the organisation, there is little that can be done to retain them. We might, therefore, want
to collect data on whether they are considering leaving. A measure of intention to leave can
provide the organisation with some early warning signs and trigger interventions to prevent
the types of turnover identified by Hausknecht and Holwerda (2013). The theory of planned
behaviour indicates that when employees are thinking about leaving the organisation or
beginning the search for a new job there is an increased chance of the employee actually
leaving. In other words, attitudes to the act of leaving can be one of many influences on
intention to leave the company.
Collecting data on these attitudes involves asking employees whether they agree with
statements such as ‘leaving the organisation would be good for me’. Subjective norms
would also come into the equation. This might mean finding out whether other people
who were important to the employee (e.g. partner, colleagues, friends, etc.) thought it was
acceptable for the employee to leave the job. Breukelen et al. (2004) point out that the
state of the employment market is a good indicator of actual behavioural control. Labour
market theories of turnover set to one side the psychology of employee turnover, instead
arguing that people are all essentially the same and turnover is mostly influenced by labour
market conditions (Morrell et al., 2001). However, a person’s perception of the amount of
employment opportunities available to them is also important: it is an indicator of perceived behavioural control.
A logical prediction from the theory of planned behaviour is that the better people see
their chances in the labour market, the stronger their intention to leave, the more active
they will be in their job search and the more likely that their intention to leave will lead to
them handing in their resignation. Researchers have found that both internal and external
‘push’ and ‘pull’ forces come together to influence the links between intentions and subsequent behaviours. For example, by studying workers in a Finnish hospital and university,
De Cuyper et al. (2011) found that perceived employability was not significantly linked to
turnover intention unless employees were dissatisfied with the level of control they had in
their work.
Chapter 5 Attitudes at work
163
Key learning point
The theory of planned behaviour has been used to understand how intention to leave develops, and
how this intention translates into actually leaving the organisation.
Individuals’ perceptions of the external forces that bind them to their job and employer
are important influences on turnover behaviour. The concept of job embeddedness is now
widely used to describe these forces. Lee et al. (2014) describe this as the level of ‘stuckness’
that an employee experiences because of the way their job or occupation connects them
to the social systems both at work and in their community. It is a concept that may help to
explain why it is that intention to leave tends to weaken with tenure (e.g. Breukelen et al.,
2004). Embeddedness has three components (see Lee et al., 2014):
1
Links: the connections that employees have to other people and activities that in some
way depend upon them staying in their current job or occupation.
2
Fit: the extent to which a person’s work and where they live fits well with other aspects
of their life.
3
Sacrifice: what a person would have to give up if they moved to a new employer.
All three can operate at the individual and community level. What this means, for example, is
that a job can link someone to a network of friends who are also their colleagues. It could also be
that an individual’s job locates them conveniently near to lots of their non-work friends and family members. Therefore, embeddedness is not just about the job itself but also captures the effects
on turnover of forces outside of the organisation. As a consequence, Hom et al. (2017) argue that
the concept may be especially important in understanding turnover in cultures where non-work
influences on behaviour are strong (e.g. collectivist and high power-distance cultures: see
Chapter 12). In such cultures the impact of changing jobs on a worker’s capacity to care for
family members may be a very important factor for many employees when decide to look for, or
accept, a new job. It is important to note that for some employees their embeddedness is more
about their ‘stuckness’ to their occupation. Highly skilled professionals (e.g. surgeons, accountants, architects etc.) are likely to have many links with other professionals that work in other
organisations, and may see moving jobs not as a sacrifice but as a way of enhancing their skills.
For these individuals, occupational embeddedness, not job embeddedness, is likely to influence
their behaviour (see also Chapter 11 for a more detailed discussion of career decision-making).
Although it is a relatively new concept, there is already plenty of evidence that embeddedness,
alongside other factors discussed in this section, plays a significant role in predicting both turnover intention and actual turnover behaviour (Jiang et al., 2012).
Returning to attitudes, there are some significant but modest links between organisational
commitment and job satisfaction on the one hand, and turnover on the other hand (e.g.
Griffeth et al., 2000). Recent research has revealed important nuances in these relationships.
Attitudes appear to predict turnover intention and behaviour more reliably for some workers
than they do for others. In a very comprehensive review of the turnover literature, Hom et
al. (2012) found substantial evidence that not every employee’s considered decision to leave
is triggered by the same set of factors. They identified different ‘proximal withdrawal states’
(2012: 831) influenced by their own preferences, personal circumstances and their perceptions of choice and control. These motivational states can be used to place employees in four
categories that help us to better understand the various mechanisms that drive turnover:
■
Enthusiastic leavers. People who are dissatisfied with their work and actively seek employment elsewhere, or those who experience a significant change in life circumstances that
leads to them looking for alternative employment.
■
Reluctant leavers. This category includes people who would like to stay but see themselves
as potential candidates for being made redundant (e.g. if there was a downsizing or a
164
Work Psychology
merger). Those who are nearing retirement and those who leave a job they enjoy to fulfil
caring responsibilities or to meet cultural expectations (e.g. to stay at home to look after
children) also fall into this category.
■
Enthusiastic stayers. These fall into two subcategories: ‘engaged stayers’ and ‘slackers’. The
former are people who enjoy the work and like being around their work colleagues; their
competencies fit well with the demands of the work; they are loyal and engaged. The
latter do not tend to enjoy the job but for them the benefits that come with the job such
as security and financial reward are highly desirable; they do just enough work to get by.
For these people it is only likely to be ill-health or retirement that precipitates leaving.
■
Reluctant stayers. There are significant barriers to leaving for these people. ‘Trapped stayers’ do not enjoy their work, or do it particularly well, but feel that leaving would be
disruptive to them or their family (e.g. they might have to relocate) or that there are
insufficient opportunities in the job market. ‘Contractual stayers’ are those who feel that
the job is not what was promised during recruitment. They perform the work well so as
not to risk being fired or to harm their long-term prospects, but will stay only until the
duration of the employment contract ends.
In a thorough test of the effects of these states, Li et al. (2016) found that psychological variables traditionally linked to turnover (such as embeddedness, job satisfaction and intention to
leave) did not predict job search and turnover in the same way for all four groups. For those
with less control over exercising their preference for leaving or staying (the reluctant leavers and
stayers) attitudes and intentions were poor predictors of actual turnover. In contrast, the same
variables were significant predictors of the behaviour of those who felt they had more control over
their behaviour (enthusiastic stayers and leavers). Turnover behaviour of some groups of employees is particularly strongly linked to their attitudes: interventions to change attitudes are likely
to have more of an impact on enthusiastic stayers and leavers. The existence of various proximal
withdrawal states may also help to explain the lack of consistency in the effects of turnover on
company performance such as those reported by Hausknecht and Holwerda (2013). Some turnover is undesirable for all involved while other forms of turnover might represent a ‘win–win’. For
example, at the end of a ‘reluctant stayer’s’ tenure they may move to a more satisfying role and
be replaced by someone who is better suited to the role (an ‘enthusiastic stayer’).
Key learning point
Different factors influence turnover for different employees. There are some common causes but also
lots of idiosyncrasies that are linked to features of the person and the situation they find themselves in.
Other approaches to turnover focus on how discrete events, rather than attitudes, can
be specific triggers of intention to leave and turnover behaviour. Lee and Mitchell (1994)
argued that particularly significant events occurring within and outside the organisation were
frequently powerful triggers of thoughts of leaving. This is the unfolding model of turnover
and downplays somewhat the importance of employee attitudes (see Holtom et al., 2008).
Instead it uses image theory to propose that people develop habitual ways of perceiving
events (schemas) and that we tend to see what goes on around us in a way that fits with these
schemas. For example, if you feel the organisation is a very good employer, but one of your
colleagues says something at work that upsets you, it might not be enough to challenge your
view of the company as a good employer. If something more extreme (or shocking) happens at
work or away from it, this might be enough of a shock to break this habitual interpretation. For
example, if you hear that a friend has a similar job that pays double your salary, that may be
a significant enough shock to change your view of your current job. This might then get you
thinking about leaving your current employment (i.e. changing your habitual behaviour).
Chapter 5 Attitudes at work
165
Morrell et al. (2004) found that negative shocks at work (e.g. a perceived unfairness after
being passed over for promotion) tended to lead to people starting a search for an alternative
job. Positive shocks from outside the workplace (such as the opportunity for a new challenge
elsewhere) were linked to people actually leaving the organisation. The implication of this is
that organisations may be able to intervene to stop employees leaving as a result of some types
of work-related shocks, but also that turnover can be a consequence of events that happen
in people’s lives. Positive career shocks from within the organisation can prevent turnover.
An unexpected promotion, success in a work project or a salary increase, have been shown to
override employees’ intentions to leave work to pursue their ambitions in higher education
(Seibert et al., 2013). A high level of embeddedness (see earlier) may also buffer against the
effects of negative shocks (Burton et al., 2010).
What about the content of the job itself? Overall satisfaction with the characteristics of
the job has frequently been found to link people’s experiences at work and their intentions to
leave (e.g. van der Aa et al., 2012). Podsakoff et al. (2007) carried out a meta-analysis of studies that had looked at the links between positive and negative aspects of jobs and employee
turnover. Positive job characteristics are those that employees often interpret as opportunities
to develop (challenge stressors). These include the level of attention required by the job, the
demands of the role, having to work quickly and dealing with interesting tasks. These factors
were largely unrelated to turnover intentions and actual turnover. The results were very different for hindrance stressors (the constraints that the work situation placed on employees,
hassles, organisational politics, role conflict, role ambiguity and lack of resources: see
Chapter 8). These were linked to both turnover intentions and actual turnover.
In terms of the context within which people work, a lack of fit between the values and
needs of employees and what can be provided by the organisation also seems to play a
small, but significant, role in employee turnover (Hoffman and Woehr, 2006). This can have
some interesting consequences. For example, Pierce and Snyder (2014) argue that unethical
employees tend to enjoy long tenure in companies when customers demand unethical services from those organisations. In other circumstances, unethical behaviour would tend to
be linked to shorter tenure.
Point of integration
Changes in procedural justice (see Chapter 6) over time seem to have a particularly strong link to the
development of turnover intentions (Hausknecht et al., 2011). Organisational policies and procedures
need to be implemented fairly and equitably.
Individual differences also play a role in determining employee turnover. Ng and Feldman
(2009) present a huge meta-analysis of the relationship between age and turnover, finding a
small but significant relationship between the two (r = - 0.14). On average older workers are
slightly less likely to leave a job than their younger counterparts. This relationship is a little
stronger among those with few, or no, formal educational qualifications. Ng and Feldman
argue that there could be a number of reasons for these findings, including:
■
Age tends to be linked to tenure, and when tenure is higher the relationship between the
employee and the employer tends to be stronger.
■
Compared to younger adults, older adults tend to have less intense negative emotional
experiences in the face of problems.
■
Older adults also tend to have lower self-efficacy than younger adults and hence have
more doubts about their ability to adjust to a new job role.
■
Compared to younger adults, older adults strive for high-quality interpersonal relationships which can be provided by long-term employment.
166
Work Psychology
More stable individual differences also appear to be important. In his meta-analysis, Zimmerman (2008) found a significant relationship between emotional stability (or neuroticism, see Chapter 2) and turnover intention: the higher emotional stability, the less likely
the intention to leave. This is logical because people who report low levels of emotional
stability experience and display more negative emotions than most. Almost always there is a
certain level of stress involved in being a new employee, something that those with low levels of emotional stability may be keen to avoid. Interestingly, Zimmerman’s analysis showed
this effect of personality on turnover intention to be direct, meaning that the increased likelihood of turnover was not because people with low emotional stability were performing less
well or were less satisfied with their job. Agreeableness and conscientiousness were related
to actual turnover (the higher levels of these, the less likely people were to leave). Those low
on agreeableness are more likely to act on impulse and have less concern about the impact
that their departure will have on others. People who have high levels of conscientiousness
prefer to see things through and tend to make careful plans when decision-making. The
link between agreeableness and conscientiousness and actual turnover was not dependent
on turnover intentions: those low on agreeableness and conscientiousness are more likely
than most people to act on impulse and leave their job without planning. This challenges
the theory of planned behaviour and is consistent with research using the unfolding theory
that distinguishes between those that act on impulse and those that leave with a clear plan
in mind (Maertz and Campion, 2004).
Rather less has been written about the links between cognitive ability and turnover. Individuals with high levels of ability may be in strong demand in the labour market and knowing
this they may also be more likely to seek out more challenging and well-paid opportunities. Those with low levels of ability find themselves underperforming and ‘pushed’ out of
work roles that demand high ability. This suggests that turnover will be highest for those
in low-ability and high-ability groups. Maltarich et al. (2010) found that this seemed to be
broadly true for cognitively demanding work and that those with higher levels of job satisfaction were less likely to choose to leave cognitively demanding jobs. Using good job design
to drive high satisfaction could be especially important when seeking to retain the most able
employees to do very complex jobs. Things were not quite as simple as this for jobs with lower
cognitive demands: those with high cognitive ability were less likely than might be expected
to leave jobs that required low cognitive ability. This suggests that choosing to do a job that
matches one’s values and circumstances can sometimes override any problems associated
with doing work that is not intellectually fulfilling.
Key learning point
Causes of employee turnover can include some aspects of personality, ability, age and job design, as
well as attitudes to the job and the organisation and the state of the labour market. There are numerous different psychological processes through which these factors have an effect. Many are more
complex and nuanced than they at first appear.
Stop to consider
From what you have read so far, how important is it for organisations to have a satisfied and committed
workforce? Are there certain circumstances when it becomes especially important for employers to
worry about such issues?
Chapter 5 Attitudes at work
167
The psychological contract
So far in this chapter we have looked at what attitudes are, how they form, and why they are
important in the workplace. We have also explored what work means to people and why what
they do at work it is important to them. In this section we bring these various issues together
by looking at the psychological contract. This has been defined in several slightly different
ways. We will use the following definition (Robinson and Rousseau, 1994: 246):
An individual’s belief regarding the terms and conditions of a reciprocal exchange agreement
between that focal person and another party . . . a belief that some form of a promise has been
made and that the terms and conditions of the contract have been accepted by both parties.
To add a little more detail to this definition, the ‘psychological contract comprises subjective beliefs regarding an exchange agreement between an individual and, in organizations typically, the employing firm and its agents’ (Rousseau, 2001: 512). The use of the word
‘subjective’ is especially important here: the influences on individuals’ interpretations and
sense-making feature heavily in psychological contract research. Herriot (1992) describes it
as ‘the invisible glue which binds individuals to the organisation over time’ (1992: 6). It is not
the legally binding written employment contract (although inevitably there is some overlap
between the two, especially around issues such as pay and working hours). This does not make
it any less important. Research shows that when it is working well for both parties good things
happen – but when the contract breaks down, there are likely to be significant problems.
In the psychological contract there is a belief that the agreement is mutual and binding.
From the employees’ point of view, this is the agreement that they think they have with
their employer about what they will contribute to the employer via their work, and what
they can expect in return. So if the employee believes that working late will help them to get
promoted, they might expect that the number of times they had worked late would be viewed
favourably. When an employee makes these efforts they expect the employer to stick to their
promises: they perceive the employer to have an obligation to give something back, to reciprocate their efforts. The contract is promissory: it about making and keeping promises and
these are kept (or broken) through exchanges between the employee and the employer. The
employee does something based on the employer’s promise to them, and gets something back
in return. As you might expect, broken promises can have a serious negative impact. Rousseau
(1998) argued that these promissory exchanges provide a framework for understanding the
important psychological processes that influence individual attitudes and organisational performance. The psychological contract is very flexible and inclusive: any aspect of the working
relationship and a whole host of individual differences can shape the psychological contract
as it unfolds over time.
Types of psychological contracts
Of course, not everyone wants the same thing from their employment. Money might be more
important to some than it is to others. Some people might take a job for the training opportunities it offers, some might be especially interested in the social aspects of work. The same
applies to the motives and needs of managers and employers. Short-term contract workers
might be hired to just do the job without being given any of the training or other benefits that
went with being a permanent employee. The work of Rousseau (1990, 1995) and Herriot and
Pemberton (1995) defines four different types of psychological contracts, or social exchanges,
commonly in operation in organisations:
■
Relational contracts offer the most mutual trust and stability. Employees offer loyalty,
conformity to requirements, commitment to their employer’s goals and trust in their
employer not to abuse their goodwill. In return, the organisation offers security of
168
Work Psychology
employment, promotion prospects, training and development, and some flexibility
about the demands made on employees if they were in difficulty.
■
Transactional contracts are based more on money and well-defined specific performance
terms (for both parties). These are much more like a short-term economic exchange.
The employee offers longer hours, broader skills, tolerance of change and ambiguity and
willingness to take more responsibility. In return the employer offers (to some) high pay,
rewards for high performance and, simply, work when work is available. This type of
exchange is more common when people are aware that they have little security of tenure
or if the amount of work offered to them can vary from one day to the next.
Key learning point
Different employees want different things from their work. Different employers want different things
from their employees. The psychological contract describes what each party promises and expects
from the relationship and this contract can change and develop over time.
Two other types of social exchanges, transitional and balanced exchanges, sit somewhere
between the two. Transitional contracts are typical of an eroding relationship between the
employer and the employee. This can happen when the company can no longer offer promises about future employment. In a balanced contract, mutual expectations are flexible and
dynamic, and there is a mutual understanding that performance expectations are likely
to change: the employer will make efforts to train and support the employee when things
change. This is a bit like the relational contract being enacted when things are less stable for
all involved. Guest (1998) argued that this variety of social exchanges is consistent with the
ever-changing nature of employment relationships and helps to highlight who has power
as demand for labour and employment ebbs and flows with changing economic conditions
and working populations.
The development of the psychological contract
The psychological contract starts to form as soon as a prospective employee hears about a
company or a job opportunity and develops further during recruitment and selection as both
parties find out a lot more about each other. Robinson and Morrison (2000) found that problems with the psychological contract were more likely when an employee had less interaction
with the employer before being hired. Giving prospective employees a realistic job preview
can reduce levels of employee turnover near the beginning of tenure. This can be achieved
through using good recruitment and selection procedures that allow prospective employees
to get a good insight into the job role and the organisation. Work sample exercises in assessment centres or situational judgement tests (see Chapter 3) provide pre-employment experiences of what it might be like to work for the company. Proper induction and socialisation
processes can also help. Woodrow and Guest (2017) found that both promise strength (the
perceived clarity of promises) and promise fulfilment (the extent to which employers were
perceived to deliver on their promises) mattered to newcomers. In a longitudinal study of
new employees across their first three months of employment, they found that the amount
of knowledge about the work role, work team and wider organisation that employees felt
they had acquired was linked to both promise strength and fulfilment. Importantly, they
also found that newcomers tended to report lower stress and higher commitment when these
promises were fulfilled.
Chapter 5 Attitudes at work
169
If the psychological contract is not to their liking, new employees may take it upon themselves to take action to try to put things right, rather than waiting for managers or colleagues
to try to repair the situation. Bankins (2015) found that newcomers could sometimes feel that
they were given tasks that were too simple for them. To remedy this, several approached their
manager to identify and self-initiate opportunities for being involved in more challenging
work. More often, however, newcomers coped by focusing on the future, and reappraised
the problems as transient ones. They accepted the less than challenging work as a temporary
situation that would soon improve.
While important, these early interactions are not everything. Taris et al. (2006) examined
the expectations of 1500 newcomers to an organisation. They found that unmet expectations
did have an impact on important outcomes such as turnover and motivation. However, the
newcomers’ expectations changed over time, with some of the unmet expectations becoming
less important. This illustrates why it is that a psychological contract is often referred to as an
‘unfolding process’: promises tend to change over time.
Key learning point
Psychological contracts change or ‘unfold’ over time. What people want and what an employer can
offer are both very dynamic.
Much of the research into the psychological contract has looked at what happens when
it is broken, or breached. These breaches can be quite common. In their early work Robinson and Rousseau (1994) found that 70 of their sample of 128 managers thought that their
employer had breached their psychological contract in the first two years of employment.
Breaches most commonly concerned failure to deliver on promises about training and development, pay and benefits, and promotion opportunities. There are three items below adapted
from Robinson and Morrison’s (2000) measure of breaches (employees would be asked to
what extent they agreed with these statements):
1
I feel that my employer has come through in fulfilling the promises made to me when
I was hired (agreement = no breach).
2
So far my employer has done an excellent job of fulfilling its promises to me
(agreement = no breach).
3
My employer has broken many of its promises to me even though I’ve upheld my side of
the deal (agreement = breach).
Key learning point
A breach of the psychological contract occurs when it is perceived that promises are broken.
Figure 5.4 illustrates the ‘violation process’ (e.g. Robinson and Morrison, 2000). A breach
is often thought of as the realisation that what has been promised has not materialised (i.e. a
perceived discrepancy). Violation is associated with the triggering of strong negative emotions such as anger or frustration (this is sometimes referred to as an affective event). Most
researchers agree that these negative emotions occur after an individual makes sense of why
the breach occurred (Bal et al., 2017). If, during that sense-making, there is a perception that
a breach was deliberate, purposeful and reflected unfair treatment it is more likely to then
trigger the strong negative emotional reactions associated with a violation.
170
Work Psychology
Employee’s track
of whether psychological
contract is being broken
1
Observation and recognised
extent of contract violation
committed by employer
Employee’s level of inspiration
and commitment to the
organisation
Gravity of
perceived violation
1
2
Figure 5.4
2
1
2
Plausible explanations
for violation
2
Employer’s efforts
to rectify violation
2
Recognised procedural
justice
Recognised
voluntariness of
violation
The process of psychological contract violation
Source: Adapted from Rousseau (1995: 118), reprinted with permission.
Unsurprisingly the experience of an intensely unpleasant experience such as bullying at
work is frequently interpreted as a violation of the psychological contract (Kakarika et al.,
2017). However, perceptions of breaches and violations can result from more subtle shifts
in work relationships and work conditions. Employees who see themselves as having a
long-term career with their current employer and have established a strong working relationship with their supervisor may feel especially betrayed when they perceive a breach in
the psychological contract (Doden et al., 2018). This appears to be because the breach poses
a threat to desired and valued long-term relationships. In contrast, those who want to move
frequently between employers throughout their career appear less likely to feel betrayed in
these circumstances. Instead, they see the relationships with the supervisor and employer as
temporary rather than binding: keeping these in a satisfactory state as less important to the
fulfilment of their long-term, self-orientated, career goals.
Working conditions can also set the scene for breaches. Periods of high demand appear to
deplete workers’ psychological resources and prime them to: (i) have strong emotional negative reactions to problems at work and (ii) attribute the causes of those events to the actions
of their employers (Bal et al., 2017). This means that after a busy or hectic period, employees
are more likely to interpret events they usually see as fairly innocuous as breaches of the psychological contract. After periods of low demand, or when plenty of resources such as support and autonomy were available to help employees deal with high demands, perceptions
of breach appeared less likely. This is the resource-depletion approach to understanding the
effects of breaches. It is somewhat different to the social-exchange mechanisms described
in early interactions of the theory (see also Chapter 8 for a detailed discussion of conservation of resources theory). Using this approach, Kiazad et al. (2014) examined the impact of
breaches on innovation among university employees in Australia. Some employees actually
reacted positively to a breach by engaging in more innovative behaviours. This makes sense
if we consider how innovation can help people recover some of their losses associated with
the breach. Kiazad and colleagues argue that innovation can involve: strengthening existing
collaborative links with colleagues, taking on more challenging tasks and a re-appraisal of
Chapter 5 Attitudes at work
171
the significance of their work situation (i.e. because of the breach they have ‘little to lose’ by
innovating). All of these actions can help with the development of resources.
The effects of breaches, violations and fulfilment
Herriot and Pemberton (1995) describe typical behavioural reactions to violation of the psychological contract as get out, get safe or get even – or, to put it another way: to leave, to stay
and protect your interests, or to stay and take your revenge. When employees felt that their
employer had violated the psychological contract, they were inclined to feel less sense of
obligation and less commitment to their employing organisation and also, perhaps, to their
customers and clients (Deng et al., 2018).
Performance, turnover and organisational citizenship and attitudes such as job satisfaction and organisational commitment can all be impacted by violations. Conway et al.
(2011) found that the consequences of a breach on attitudes such as job satisfaction and
organisational commitment were bigger than the consequences of fulfilment. The effort and
care required to fulfil a psychological contract can be justified by the costs that are avoided.
Conway and Coyle-Shapiro (2012) carried out a complex longitudinal study of sales advisors
and collected data on changes in performance over time. As time passed it appeared that performance loomed larger in people’s evaluations of the promises being made. This indicated
that when employees had been with the organisation for some time, when they performed
well they were more likely to experience fulfilment of their psychological contract (and vice
versa). Earlier on in their tenure the support they received from others mattered more, even
if they were less than successful at achieving performance targets. The fluctuating state of the
psychological contract over time is a common theme in longitudinal studies.
The evidence for the impact of violations of the psychological contract continues to
mount. In a meta-analysis of the research Zhao et al. (2007) looked at 51 well-designed studies. The analysis revealed that the strongest links between the psychological contract and
organisational outcomes occurred when employees had a strong emotional response (e.g.
feelings of violation or mistrust) to the breach. These reactions were particularly strongly
linked to turnover intention, organisational commitment and job satisfaction. There were
weaker, but significant, links to job performance and organisational citizenship behaviours.
However, there was no significant link to actual turnover, most likely because turnover is
a very complex, multi-causal phenomenon, as discussed earlier in this chapter. The type
of contract breached did not seem to make too much difference to the outcomes, except
that breaches of transactional contracts had stronger links to organisational commitment
than breaches of relational contracts. Something that may be a concern for organisations
is that employees who perceive violations may remain in their job and engage in a steadily
increasing amount of covert counterproductive work behaviour aimed at their employer (see
Chapter 4). By collecting data frequently over time, Griepa and Vantilborgh (2018) found that
as perceptions of breaches accumulated over time, feelings of violation appeared to intensify
into a negative spiral of ‘escalating negative interactions’ (2018: 150).
In their longitudinal study, Ng et al. (2010) found that organisational commitment was
harmed by breaches and that this had knock-on detrimental effects on levels of innovation.
They also found that after initial perceptions of breaches, employees held stronger perceptions about the breach – there is a risk that the effects of breaches can grow over time. Breaches,
whether caused by the manager or not, seem to have a particularly damaging effect on line
manager–employee working relationships with this, in turn, leading to long-term detrimental effects on employees’ career success (Restubog et al., 2011). Breaches of the psychological
contract have also been linked to an increased likelihood that employees will report working
with less vigour and energy after a breach then they did before it (Rayton and Yalabik, 2014).
These effects may extend beyond the organisation and impact on the quality of work done
for clients (Deng et al., 2018).
172
Work Psychology
Breaches appear to be a feature of working relationships in complex, constantly changing
organisations. As a result, researchers have started to look at what factors determine whether
a breach triggers an emotional response and whether the negative outcomes of breaches can
be averted. For example, a person who starts out favourably disposed towards their employer
might be less likely to notice breaches than somebody who already mistrusts the employer.
They might also be less likely to construe breaches as violations or at least have less strong
psychological reactions to them. The intensity of psychological responses to breach is the
focus of a lot of research. For example, in a study of over 6000 British Royal Air Force personnel, Clinton and Guest (2014) found that it is the important felt consequences of breach
that appear to drive its links to turnover. Their results suggest that if breach resulted in a
perception that trust in the employer had been significantly damaged, or that the employee
felt that their exchange with the employer was unfair (e.g. they got out less than they put in)
then breach would be linked to turnover.
How a breach is handled is important in determining whether it turns into a violation.
Lester et al. (2007) found that open communication with employees about the nature of
the psychological contract was important to keeping the contract fulfilled. The provision of
credible, legitimate and consistent reasons for breaches can help: employees are less likely
to feel the contract has been violated or to develop feelings of mistrust. Zhao et al. (2007)
argue that employers need to guard against breaches by making efforts to keep abreast of
their employees’ needs. If it is felt that the breach was neither the employer’s fault, nor
intended, then the impact on the employee’s loyalty is likely to be small, particularly if it is
put right quickly.
Individual differences and the psychological contract
There is some evidence that age moderates the impact of a breach. Longitudinal research
indicates that older workers react less intensely than younger workers to breaches with less
impact on their job satisfaction and performance (Bal et al., 2008, 2013). This effect appears
not to be a consequence of tenure and more likely to be linked to age-related differences in
emotional regulation (more constructive regulation may be practised by older workers). Age
appears to bring with it a wider repertoire of strategies for coping with and appraising the
meaning of emotions.
It also appears that psychological breaches can leave employees with some ‘baggage’. Robinson and Morrison (2000) found that experiences of breaches of the psychological contract
with former employers were linked to the likelihood of breaches occurring with the current
employer. Karagonlar et al. (2015) have found that some employees remain more wary than
others of the promises made by employers, even if their experiences indicate that psychological contracts are fulfilled.
Those who feel that promises have been broken might engage in deviant or counterproductive work behaviours (see Chapter 4). Restubog et al. (2015) studied workers in the hospitality and measured ‘revenge cognitions’: workers may consider exacting revenge but not
actually act on their thoughts. They found that if the organisational culture was aggressive
then such thoughts were more likely to translate into actions. Their findings indicate that
organisational culture needs to be managed carefully if some of the more unpleasant effects
of breaches are to be minimised. Those higher in self-control felt the effects of psychological
contract breaches as keenly as anyone else but, compared to those with low self-control,
they were less likely to attempt to exact revenge.
What is most important to people in their psychological contracts can vary quite a bit.
Restubog et al. (2007) found that breaches have bigger effects for people who attach high
value to the tangible benefits of employment such as money (outcome-oriented people)
Chapter 5 Attitudes at work
173
than it does for those who attach high value to the benefits of relationships at work
(relationship-oriented people). Outcome-oriented employees tended to attempt to ‘even
the score’ by extracting revenge. Relationship-oriented employees appear less likely to do
so at such actions can place at risk relationships with co-workers and supervisors.
Key learning point
There are likely to be very significant differences between individuals in the kind of promises they
expect to be kept, and the value they place on the rewards they might receive for their efforts.
Lambert (2011) argued that the way most people see their psychological contracts can be
broken down into four components: promised inducements, promised contributions, delivered inducements and delivered contributions. Her results showed that much depends upon
how each of these elements fulfils the specific needs of each employee at any given time:
elements proximal (close) to personal need fulfilment appear to matter most. For some (e.g.
those who have a personal need to pay their rent) delivered inducements (e.g. pay) may matter
most. Others’ needs might be closely linked to their desire to achieve long-term goals and
therefore promised inducements could be more likely to match their needs (e.g. the promise
of a well-defined path to a top job sometime in the future). In this case they might be happy
to exert effort in return for a promise from the employer.
Key learning point
Interventions can stop breaches of the psychological contract leading to the strong negative emotions
associated with violations.
Summary
Like most social psychological phenomena, attitudes are more complicated than they
first seem. A person’s attitudes may predict their behaviour quite well in some circumstances, if the right attitude is assessed, and if the person’s perceptions of social pressures and their own capabilities are also taken into account. Job satisfaction concerns a
person’s evaluation of their job, while organisational commitment refers to the extent
to which a person feels attached to their employing organisation. They can both be
measured satisfactorily, are influenced by the nature of the person’s job and individual
differences and appear to have quite complex connections with a range of behaviours
and other attitudes at work. However, they may allow us to make some imperfect predictions about if a person stays in their employing organisation and their job performance.
Of course, the causes of performance and turnover are complex and many other factors
aside from attitudes influence these behaviour. The psychological contract provides a
useful framework for understanding how the employment relationship develops and is
maintained. It also tells us quite a lot about how attitudes develop over time and what
people want to get from their work.
174
Work Psychology
Closing case study
Employers face moves to strengthen
gig worker rights
UK employers will have to compensate workers if they cancel shifts at short notice under
proposals intended to give people in the gig
economy more control over their working
lives. The measures, set out in a government
consultation on Friday, are part of a broader
effort to align employment regulation and
practices with the changing world of work.
They are also a response to concerns that the
UK’s jobs boom – with employment at record
highs and many workers benefiting from a
steadily rising minimum wage – masks precariousness and ongoing abuses at the bottom end of the labour market.
They aim to address the problem of ‘one-sided
flexibility’ – where employers misuse flexible
working arrangements, leaving workers with
unpredictable hours, insecure income and
little ability to assert their rights – that was
identified two years ago in a review led by
Matthew Taylor, chief executive of the Royal
Society of Arts.
The government is proposing that workers
should be entitled to reasonable notice of
work schedules, and compensation if shifts
are cancelled at short notice. It also wants to
bolster protection for workers who are penalised for refusing shifts offered at the last
minute. The measures are in line with recommendations made last year by the Low Pay
Commission (LPC), which advised against
the proposal made in the Taylor review, that
workers should receive a higher minimum
wage for working non-guaranteed hours.
Ministers have resisted calls from unions
and other campaign groups to ban the use of
zero-hour contracts, which are often blamed
for leaving workers at the mercy of employers who give them little notice of when they
will work and leave them with fluctuating incomes. However, the latest proposals
include confirmation that the government
plans to legislate to give workers a right ‘to
potentially move towards a more predictable
and stable contract’, with more certainty
over the number of hours or the fixed days
on which they work.
Bryan Sanderson, chair of the LPC, said the
government’s proposals ‘have the potential
to improve work and life for hundreds of
thousands of people’, while allowing flexible workers to maintain their autonomy and
businesses to tailor their workforce to peaks
in demand. Ben Willmott, head of public policy at the CIPD, the body for HR professionals,
welcomed the proposals but said businesses
would need clarity on the circumstances in
which employers could refuse a request to
switch to a more predictable work pattern.
Although UK employment is at record highs,
and low paid workers have benefited from
steady increases in the minimum wage, ministers are under increasing pressure to do more
to improve the quality of work and address
ongoing abuses of employment rights. Earlier this week, the government set out plans
to strengthen enforcement of labour market
rules by creating a single body with a mandate to stamp out workplace exploitation.
Source: Delphine Strauss (2019) ‘Employers face moves to strengthen gig worker rights’, The
Financial Times, 19th July, © Financial Times Limited 2019. All rights reserved. https://www.ft.com/
content/602afc46-aa24-11e9-b6ee-3cdf3174eb89
Suggested exercises
1 What types of psychological contract best describes the situation typically experienced by workers
in the ‘gig economy’? Are the contracts most likely to be relational, transactional, balanced or
transitional?
2 Before the implementation of the proposed legislation, what breaches of the psychological contract appear to be frequently occurring? Consider this question from both the perspective of the
employer and the perspective of the employee.
3 Do you think the proposed legislation will reduce the number of breaches? Explain your answer.
4 Why might some employees find working in the gig economy a satisfying experience while others
see it as highly unsatisfactory?
Chapter 5 Attitudes at work
175
Individual and group discussion questions
1 What are attitudes and in what circumstances do attitudes determine behaviours at
work?
■ In your own words define an attitude.
■ What are different components of attitudes?
■ What factors influence attitude strength?
■ Why is it important that the object of an attitude is clearly defined?
2 How much is known about that factors that determine either job satisfaction or organisational commitment.
■ Define the attitude being discussed and identify its object.
■ What are the situational influences on this attitude?
■ What individual differences might play a role? Consider both stable and unstable
individual differences.
3 Should managers in organisations care about results of surveys of staff satisfaction and
commitment?
■ How good are the data collected from these surveys? Consider the reliability and
validity of the data and identify any problems associated with the measurement of
these variables.
■ What is the strength of the links between these variables and outcomes that managers and employers typically value? How consistent is this evidence and does it come
from good quality research?
4 Is the concept of the psychological contract useful?
■ What is the psychological contract?
■ Identify two different types of psychological contracts and describe their key
features.
■ When and how does it start to form?
■ What is a breach of the contract and how might that turn into a violation?
■ Identify two possible consequences of a violation. How can violations be avoided
even if a breach has occurred?
Relevant websites
Examples of how public opinion survey companies work and present their findings can be
found at https://www.ipsos.com/ and https://yougov.co.uk/ Here you can find a summary
of opinion surveys carried out in various countries about attitudes to a very diverse range of
topics.
National Health Service Trusts in the UK now carry out annual staff attitude surveys. The
website provides some many examples of how employee attitude surveys can be conducted,
interpreted and reported. https://www.nhsstaffsurveys.com
Extensive and accessible guidance on the design of attitude surveys can be found in the Harvard
Business Review article ‘Getting the Truth into Workplace Surveys’ by Palmer Morrell-Samuels.
This is available at: https://hbr.org/2002/02/getting-the-truth-into-workplace-surveys
The UK CIPD carries out large annual survey of employees’ views on many aspects of their
job. Summaries of the findings and details of the methodology used can be found at https://
www.cipd.co.uk/knowledge/work/trends/uk-working-lives (or search the CIPD website for
UK Working Lives Survey).
Deloitte’s report on it Millennial Survey is an interesting example of how survey methods
can be used to capture data from across the globe. Its results also make interesting reading
https://www2.deloitte.com/content/dam/Deloitte/global/Documents/About-Deloitte/
gx-2018-millennial-survey-report.pdf
176
Work Psychology
Suggested further reading
Full details for all references are given in the list at the end of this book.
1 The papers by Judge et al. (2017) on employee attitudes (see also http://www.timothy-judge.com/), and Hom et al. (2017) on turnover (both in the Journal of Applied
Psychology) contain good overviews of key developments and findings in these complex
topic areas.
2 The paper by Howe and Krosnick (2017) in the Annual Review of Psychology provides a
good, critical summary of the links between attitudes and thoughts and behaviours.
3 Ajzen’s 2001 article in the Annual Review of Psychology describes the influential theory of
planned behaviour, as well as other research on how attitudes work. Not an especially
easy read, but very informative about attitude theory.
4 The paper by Harrison et al. published in 2006 in the Academy of Management Journal
does a good job of summarising the various possible links between attitudes, individual
behaviour and subsequent organisational outcomes. The introduction is particularly
useful.
5 The book Understanding Psychological Contracts at Work: A critical evaluation of theory and
research, by Neil Conway and Rob Briner (Oxford University Press, 2005), provides an
accessible and in-depth discussion of a high-profile topic in work psychology.
CHAPTER 6
Work motivation
LEARNING OBJECTIVES
After studying this chapter, you should be able to:
1 define the three component parts of motivation;
2 describe the two types of content motivation theories;
3 critically evaluate at least one trait-based theory of motivation;
4 compare and contrast normative and trait-based theories;
5 describe and critically evaluate the job characteristics model in explaining motivation and
performance at work;
6 outline the way in which socio-cultural forces at work motivate employees;
7 describe process theories of motivation;
8 critically evaluate goal-setting theory in explaining motivation and performance at work.
178
Work Psychology
Opening case study
Launching Cuckoo Gin
What makes a person trade in a secure, well-paid job for the daunting task of starting up a business?
That was the question facing Mark Long, now Director and Master Distiller of Cuckoo Gin in 2016,
when he decided to quit his job as a Fine Food Accounts Manager to pursue his dream of running a
gin company. ‘I think the aspiration to do something on my own was the main driver’, said Mark when
asked what prompted the decision, ‘I could also see that potentially me and my family could have a
good life together quicker than if I stayed in a 9-5 job that could have seen me working for 40 years’.
Being an entrepreneur is often portrayed in a romanticised manner; it conjures up the image of
James Dyson tinkering with vacuum cleaner parts in a shed, or Anita Roddick tracking down ethical
skincare products before opening the Body Shop. However, the realities of starting a company can
be brutal, it involves long hours and can be financially and emotionally draining. ‘It was difficult to
leave a job I actually enjoyed and colleagues that I had grown to know very well’, Mark commented
when reflecting on the difficulties of starting out on his own, ‘going out on your own also takes away
the comfort blanket of a guaranteed monthly wage and the support of people who have helped you
to develop your skills’.
Motivation is one of the key factors needed when entrepreneurs begin a new venture. ‘I needed
to be self-motivated to achieve a better life, otherwise I would have stayed in a comfortable job’,
reflects Mark. Now in its third year as a company, Cuckoo Gin is thriving. The company manufactures
three different types of award-winning gin and Mark has taken on five employees to help with the
workload. However, Mark knows the importance of staying motivated ‘More motivation and more
pressure comes from the fact that we employ people, they are working hard for us so we want to
make sure our employees are growing with the company’.
Introduction
A sound motivation theory must (1) identify an outcome behaviour, (2) identify what is driving that behaviour and (3) explain how the drivers of motivation work together in a casual
chain (Kanfer, 1990). As such, motivation theories show how a person acts, what propels
them to act, and why motivation leads to action. Motivation theories explain why employees do (or do not) demonstrate specific behaviours within the workplace. These can include
positive behaviours, such as citizenship and pro-environmental behaviour, as well as negative
behaviours, such as theft and bullying. Regardless of the specific behaviour of interest, motivation theories must consider the three components of motivation which include:
■
the direction (choice) of the behaviour: what a person is trying to do; the ‘direction’
creates a ‘vision’ which drives the following two aspects;
■
intensity of action (effort): how hard a person is trying;
■
persistence of the behaviour over time (duration).
While these motivating components cannot directly be observed, the behavioural outcomes
of their effects can. Understanding factors which initiate behaviour, and how these factors
are related, in turn explains their motivational consequences (i.e., the outcome behaviour).
As with many important concepts in psychology, there is no single universally accepted definition/theory of workplace motivation. As such, we should understand motivation as ‘an
umbrella term meant to capture the dense network of concepts and their interrelations that
Chapter 6 Work motivation
179
underlie observable changes in the initiation, direction, intensity, and persistence of voluntary action’ (Kanfer et al., 2017: 339).
In this chapter, we will examine the following types of motivation theory: ‘content’ theories describe factors ‘within people’ that motivate human behaviour at work. These consist
of factors which are common across people (as per normative theories) or vary between individuals (as per trait theories). Second, context theories describe factors within an employee’s
work environment that motivate behaviour. Third, we move away from a static perspective of
factors which exist either within people or the environment, to dynamic ‘process theories’
which explain the cognitive processes involved in selecting and persisting at the behaviour.
In summary, the content and context theories articulate ‘what’ factors have motivational
properties, whereas process theories explain ‘how’ behaviour is planned and enacted. We close
the chapter with an integrative perspective which aims to provide a more holistic explanation
of work motivation, arguing that all the approaches are all worthwhile in their own way!
Key learning point
The outcome of workplace motivation includes: (1) displaying a behaviour at work, (2) enacting effort
towards the behaviour and (3) persisting with it over time.
Content theories
Content theories describe factors within people that motivate behaviour. Within content
theories, there are normative and trait approaches. Normative theories highlight universal
drivers of motivation that are consistent across people, such as innate desires for shelter,
company and recognition. Comparatively, trait theories focus on differences between people
that explain why some individuals display incredible motivational energy, while others are
happy to avoid exerting effort. Moreover, differences in motivational instigators do not only
occur between people but also within a person at different points in time, and thus we explore
how age influences motivation.
Normative theories
Normative theories highlight motivational factors which are relatively enduring across people and time; such as specific human desires, wants and needs. Hence, they are also referred
to as universal motive theories. The normative theories we review examine people’s desire for
need fulfilment, fairness, autonomy, relatedness and competence.
People’s desires to fulfil their needs
Needs are defined as ‘internal forces that are essential for supporting life and growth’ (Kanfer
et al., 2017: 340). Humans are driven to fulfil their biological and psychological needs, such as
those for food, shelter and company. Need theories state that when our needs are unmet, we
direct motivational energy to fulfilling them. The influence of biology is clear, as similar to
some animals which instinctively go from hunting, to mating, to hibernating, need theories
argue that humans exert motivational energy when they feel the instinctual tug of a need.
Maslow’s Hierarchy of Needs (1943, 1954) states that there are five human needs, which
people seek to fulfil sequentially:
1
Physiological: to eat, sleep, drink and remain sufficiently warm.
2
Safety: to feel physically and psychologically safe.
180
Work Psychology
3
Belongingness: to receive support, affection, love and interpersonal warmth.
4
Esteem: to feel valued and respected.
5
Self-actualisation: to fulfil one’s potential, to develop one’s capacities and express them.
When one need is satisfied, the next need up the hierarchy then becomes the main behavioural drive (see Figure 6.1).
Despite dominating work motivation for years, Maslow’s Hierarchy of Needs has received
the following criticisms (e.g. Rauschenberger et al., 1980; Salancik and Pfeffer, 1977; Wahba
and Bridwell, 1976):
■
The needs have been insufficiently described. Rowan (1998) argues for two types of
self-actualisation: one where a person can express their real self, and one which is more
mystical, incorporating a sense of closeness with God or humanity.
■
The theory does not reliably predict or explain when particular needs become important.
■
The theory does not adequately show how specific behaviours are predicted by key needs,
because a behaviour can be instigated by different needs.
■
The over-emphasis on needs as biological phenomena ignores the capacity of people to
construct their own perceptions of needs and how they can be met (Cooke et al., 2005).
Nevertheless, using data from Canadian, French, German, Israeli, Japanese and British
employees, Ronen (1994) identified support for the universal validity of Maslow’s hierarchy.
Self-actualisation
needs drive
behaviour
Esteem
needs drive
behaviour
If satisfied
Belongingness
needs drive
behaviour
If satisfied
Safety
needs drive
behaviour
If satisfied
Physiological
needs drive
behaviour
If satisfied
Note: start at the bottom!
Figure 6.1
Never entirely
satisfied
Maslow’s hierarchy of needs
Chapter 6 Work motivation
181
Furthermore, Taormina and Gao (2013) developed a robust questionnaire of need satisfaction
using Maslow’s hierarchy. In line with the theory, the authors demonstrated that satisfaction of
a need went hand in hand with satisfaction of the need immediately below it in the hierarchy.
People’s desire for fairness
People’s actions are also driven by a need for fairness. Justice theorists claim that people’s
natural desire for fairness determines their behavioural choices. Hence, justice theories are
classified as cognitive theories of motivation. Given the reference to cognition, some may
classify justice theories as a process approach (see later). However, Kanfer et al. (2017) classify
justice theories as needs-based, rather than process-based, since the drive to achieve fairness
is structural and static, rather than a product of thought processes.
Equity theory postulates that individuals evaluate their relationships with others by
assessing their outcomes gained relative to their inputs and comparing their ratio to that
of their colleagues (Adams, 1965). If an employee feels their gains have been smaller than
another person’s, despite similar levels of input, they experience ‘underpayment inequity’. This
can lead to compensatory behaviour such as absenteeism, lowered performance, and even
theft (Ambrose and Kulik, 1999). These are known as restoration techniques and include
cognitively distorting inputs or outcomes, adjusting the amount of effort devoted to a task,
changing the comparison other, or terminating the relationship. Research in the 1960s and
1970s provided some support for equity theory (e.g. Pritchard, 1969), especially in laboratory
studies where the key aspects of the theory could be controlled and measured.
While a more thorough review of equity theory is offered by Pritchard (1969), we note the
following criticisms:
■
It does not apply consistently across contexts, despite claiming to be based on universal
drivers of behaviour. For example, when people are overcompensated (rather than undercompensated) the theory does not explain behaviour (Mowday, 1991).
■
It does not apply consistently across cultures. Li and Cropanzano’s (2009) meta-analysis
concluded that organisational justice was a greater motivational mechanism in individualist cultures of North America than collectivist cultures of East Asia.
Equity theory focuses on peoples’ drive for distributive justice, namely whether people believe
they have received (or will receive) fair rewards (e.g., in the context of hiring, promotion
and raise decisions). However, two other judgements about fairness in the workplace exist
(Colquitt et al., 2001; see Table 6.1). Procedural justice refers to the extent to which employees
perceive fairness in procedures used to determine outcomes. For example, has the process
Table 6.1
The need for three types of justice at work
Distributive justice
(appropriateness of outcomes)
Procedural justice
(appropriateness of the allocation process)
Interactional justice
(appropriateness of the treatment received)
Appropriateness of outcomes can be
judged in various ways:
For a procedure to be just it must:
For interactions to be just, they must be:
Not be biased against any individual or group,
i.e. consistent process.
Interpersonal – people must be treated with
dignity, courtesy and respect.
Make decisions based on accurate
information.
Informational – all employees must receive
the relevant information.
Equity – employee outcomes depend on
their contributions.
Equality – everyone gets the same
reward
Need – people get what they require.
Follow ethical codes.
Correct errors or injustices via appeal or
review procedure.
Source: Adapted with permission from Cropanzano et al., 2007: 36.
182
Work Psychology
been applied consistently across employees, and have colleagues had an opportunity to voice
their input? In practice, managers should highlight the importance of giving people every
opportunity to show what they can do during selection and promotion processes. Finally,
Interactional justice includes both interpersonal and informational justice. Interpersonal
justice refers to whether employees feel they have been treated with dignity and respect.
Informational justice refers to whether someone feels they have been fairly informed about
organisational decisions.
Contemporary research has found evidence for the physiological and neurological processes underlying drives for justice, reinforcing their role as universal need drivers (Dulebohn
et al., 2015; Yang et al., 2014). Furthermore, Ryan and Wessel’s (2015) meta-analysis found
that justice perceptions motivate behaviours such as citizenship behaviour (see Chapter 4),
withdrawal and performance within the workplace. Given the role of justice in motivating
behaviour, researchers have begun examining the factors that influence a person’s drive for
fairness – such as individual differences in charisma (Scott et al., 2007; 2009).
So how can we explain these relationships? Some argue they are simply a product of our
desire for human dignity. However, social exchange theorists state that we are grateful to
organisations for treating us fairly, and respond by working harder, almost as a ‘thank you’,
Another, relational, explanation for why the desire for fairness triggers behaviour, results
from how the perception of being treated fairly signals that you are a valued member of a
community, thus validating your personal and social identity. Finally, instrumentally, being
treated fairly reduces fear of exploitation.
However, even in the best-run workplaces, there will be times when some people think
something is unfair. This is partly because, although people often distinguish between what
is good for them and what is fair, there is some overlap between outcome favourability and
perceptions of distributive justice (Fortin, 2008).
Point of integration
A key area in which perceived justice is important is the assessment of work performance (see
Chapter 4). If assessment methods are not thought to be accurate or fairly administered, they are likely
to be considered unfair. You may also notice some similarities with effort-reward imbalance models
of work-related stress (see Chapter 10) which describes how increases in effort to obtain reward can
threaten well-being.
Key learning point
People have an inherent need to cognitively appraise situations for fairness. When this need is not
being met, it triggers action (i.e., the need for fairness is a motivational mechanism).
Desire for autonomy, competence and relatedness
Self Determination Theory (SDT) (Ryan and Deci, 2000) describes how people are universally driven to fulfil the psychological needs of autonomy, competence and relatedness (see
Figure 6.2). Employees experience autonomy when they have choices and can self-determine
aspects of their work competence when they achieve mastery over tasks, and relatedness when
they feel they are being challenged and have the ability to connect interpersonally with colleagues and other stakeholders, including demonstrating care towards others and feeling like
others do the same towards them. Van den Broeck and colleagues’ (2016) meta-analysis of 99
Chapter 6 Work motivation
Nonselfdetermined
behaviour
Selfdetermined
behaviour
Extrinsic
motivation
Amotivation
Nonregulation
Figure 6.2
183
External
regulation
Introjected
regulation
Identified
regulation
Intrinsic
motivation
Integrated
regulation
Intrinsic
regulation
Self-determination theory
studies investigating SDT at work, confirms that when these needs are satisfied, individuals
experience psychological growth, internalisation and well-being. According to SDT, psychological growth is the experience of curiosity and exploratory engagement when employees
conduct activities they find interesting and enjoyable in the absence of external reinforcement (Van den Broeck et al., 2016). Internalisation refers to the human inclination to convert external reasons for engaging in a behaviour into internalised and integrated reasons for
engaging in a behaviour (Deci and Ryan, 1985). Finally, well-being refers to increased positive
and reduced negative affect, life satisfaction, and physical and mental health.
Work tasks that satisfy one or more of the three psychological needs are usually experienced as enjoyable, which heightens intrinsic motivation. Intrinsic motivation is ‘the desire
to expend effort based on interest in and enjoyment of the work itself’ (Grant, 2008: 49). It is often
compared favourably to extrinsic motivation which refers to motivational energy expended
to obtain rewards unrelated to the work itself, such as financial incentives and status. Intrinsic
motivation is also increased when the choice of behaviour is perceived to be self-determined,
which means that individuals act according to their wishes, rather than having behavioural
choices imposed upon them (Deci et al., 2017).
Interestingly, it has been suggested that one of the factors that undermines intrinsic motivation is the use of extrinsic motives, such as pay, bonuses and benefits (Deci, 1975). We
discuss this idea in greater detail later, but it is important to be aware of extrinsic motivation
because it is highly prevalent in work environments. One of the major contributions of SDT
is to break down extrinsic motivation into several forms:
■
External regulation: whereby a person performs a behaviour in order to satisfy an external
demand, which could be material (the need for pay) or social (the need to please others).
They experience this as externally controlled and not what they would do by choice.
■
Introjected regulation: whereby a person has internalised the external demand enough for
it to matter to their self-esteem. They would feel pride, shame or guilt if they engaged, or
did not engage, in the behaviour (Van den Broeck et al., 2016).
■
Identified regulation: whereby a person accepts an external demand or reward as being
of personal importance, and therefore also uses self-regulation in order to perform the
required behaviours, which often align with their personal values.
■
Integrated regulation: this is similar to identified regulation, but the person not only
accepts external rewards or requirements as important, but also as an expression of self.
184
Work Psychology
According to SDT, a context with no extrinsic or intrinsic rewards would not trigger behaviour.
This is referred to as amotivation. The presence of extrinsic rewards leads to controlled motivation, where a person’s behaviour is governed by rewards other than the joy of the task itself.
Comparatively, autonomous motivation arises from intrinsic, integrated and identified forms
of motivation. The autonomous motivations generally lead to better levels of performance and
well-being than the controlled ones (e.g. Bono and Judge, 2003; Welters et al., 2014).
In support for SDT, Milyavskaya and Koestner (2011) surveyed 203 adults, asking them
about their needs and what they termed autonomous motivations, which captured both the
autonomous pathway and the controlled pathway. In line with SDT, they found that autonomous motivation was positively associated with need satisfaction (comprising the three
psychological needs). Consistent with SDT being a universal theory, they found this was the
case whatever the domain examined.
Despite receiving significant empirical support, SDT has received the following criticisms:
■
Some argue SDT is a ‘western theory’ because autonomy is an important element within
western but less so in eastern cultures (Markus and Kitayama, 1991). However, Deci et al.
(2017) refute this, suggesting that it is based on a definition of autonomy as an independent, detached and individualistic concept, rather than as a volitional and concurrent concept. Cross-cultural research seems to support this assertion. A sample of South Korean,
Russian, Turkish and American employees were found to be psychologically healthier
when they acted autonomously (Chirkov et al., 2003).
■
Some challenge the universal assumption that all individuals inherently have the natural
tendency to fulfil their potential, as described by the need concepts.
■
There is a lack of clarity over how the three needs (autonomy, competence and relatedness) relate to behaviour, and whether they do so in the same way. Bindl et al. (2018) show
they relate differently to three main types of behaviour, which happen to be task crafting,
skill crafting and relational crafting.
■
SDT states intrinsic work motivation positively relates to employees’ learning new things,
rather than maintaining the status quo. However, intrinsic motivation is not the only
motivational mechanism at play here, as research shows the exploratory behaviour of
learning new things is reduced when employees try to portray a positive self-image (i.e.,
self-enhancement; Kauppila, 2018).
Key learning point
Employees are driven to achieve three psychological needs (autonomy, competence and relatedness).
There are two pathways to realising these needs and their associated employee outcomes: the autonomous pathway based on intrinsic motivating factors, and the controlled pathway based on extrinsic
motivating factors.
Trait-based theories
Trait theories of motivation identify person-based motivational factors which differ across
people, such as need for achievement, personality, general self-efficacy and age.
Need for achievement
Contrary to the section above, some argue that not everyone is driven by the same universal
needs. Popular between 1960 and 1980, the need for achievement (NFA) theory concerns
the desire ‘to overcome obstacles, to exercise power, to strive to do something difficult as well and as
Chapter 6 Work motivation
185
quickly as possible’ (Murray, 1938: 80–81, quoted by Landy, 1989: 73). Sagie et al. (1996) state
those high on NFA seek tasks:
1
involving uncertainty rather than sure outcomes;
2
which are difficult rather than easy;
3
involving personal responsibility, not shared responsibility;
4
involving a calculated risk, rather than no risk or excessive risk;
5
requiring problem-solving or inventiveness, rather than following instructions;
6
that gratify the need to succeed, rather than ensuring the avoidance of failure.
McClelland (1961) argued a nation’s economic prosperity depends partly on the level of NFA
in its population. He based this argument on a statistical relationship between the economic
performance of countries and the prominence of themes of achievement in popular children’s stories in each country. An individual’s NFA also relates positively to career success
(Spangler, 1992), and status/wealth, professional fulfilment, and contribution to society
(Parker and Chusmir, 1991).
Exercise 6.1
Are there downsides of being driven to achieve?
Based on the six task preferences by Sagie et al. above, in what respect is NFA a good thing in your
kind of work or study, and in what respect could it be a bad thing?
NFA has received the following criticisms:
■
NFA is a Western and individualistic concept, of little relevance to some cultures. When
analysing economic growth over 50 years, Beugelsdijk and Smeets (2008) questioned
McClelland’s theory after finding no relationship between growth and a country’s NFA
score. At the individual level, Sagie et al. (1996) found that Americans usually score higher
on NFA, followed by people from the Netherlands and Israel, with those from Hungary
lower and those from Japan lower again.
■
The need for achievement is not the most important personality trait. In politics and management, the need for power predicts success better than NFA. This is because individuals
in these roles are not working alone, but need others to act upon decisions (Winter, 2010).
Personality
From the 1980s to the present day, researchers have linked personality to levels of motivation. Meta-analyses found that conscientiousness (positively) and neuroticism (negatively)
relates to job performance, whereas extraversion, agreeableness and openness to experience
only related to performance in certain jobs, such as extraverts performing better in sales jobs
(Barrick and Mount, 1991; Barrick et al., 2001). Researchers also demonstrate how personality
related to performance by influencing self-efficacy (Barrick et al., 2002; Judge and Ilies, 2002).
In line with trait theories, people vary in their behaviour as a result of their reactions to
positive, negative and uncertain situations. Some individuals are hyper-responsive to positive
environmental cues, referred to as ‘approach orientation’ (Davidson, 1984; Gray, 1987). They
report more positive work attitudes (Cropanzano et al., 1993), are more extraverted in social
interactions (Eysenck and Eysenck, 1975), and hence responsive to positive feedback/information (Larsen and Ketelaar, 1989). Approach-avoidance types, however, respond deeply to
negative stimuli (Davidson, 1984; Gray, 1987), and hence experience significant negative
emotion when receiving negative feedback at work (Larsen and Ketelaar, 1989). Promotion
186
Work Psychology
(as opposed to avoidance) focused people demonstrate greater productivity and well-being
because striving for gains, rather than focusing on minimising losses, elicits stronger positive
emotions (Ferris et al., 2013). Whether someone has an approach or avoidance motivational
trait has direct implications for the goals they set and the strategies they pursue.
General self-efficacy
People differ in terms of their degree of general self-efficacy (their belief in their ability to
perform well in different situations) and hence differ on performance (Hodgkinson and Healey, 2008). Judge and Bono’s (2001) meta-analysis shows however, that the extent to which
self-efficacy predicts behaviour/performance, over and above other individual differences (e.g.,
mental ability, Big Five personality traits and experience) is not as large as first anticipated.
Some have grouped self-efficacy together with self-esteem, locus of control and emotional
stability, to create core self-evaluation (CSE; see Judge et al., 2004). In their meta-analysis,
Chang et al. (2012) evidence the role of CSEs in increasing work outcomes through the motivational variables of effort and persistence. However, Bandura (2012) critiques the utility of
trait self-efficacy because it does not explain how behaviours are produced, nor does it explain
how a general self-efficacy can be reliably interpreted when it varies across situations.
Age
Given employees are retiring later, there is increased age diversity in the workplace. This has
implications for motivating people at work because older people are motivated differently to
younger colleagues. Hence, factors instigating motivational effects do not just vary between
people, but also within a person over time (Kanfer and Ackerman, 2004). Such differences
in motivation relate specifically to people’s age as opposed to their generation (e.g., Baby
Boomers, Generation X). Aside from age relating to declines in cognitive and intellectual
capabilities, Kooij et al. (2011) and Inceoglu et al. (2012) find age is positively associated with
intrinsic motives, and negatively to extrinsic motives. While they found older adults were less
motivated to achieve job-related growths, Kooij et al. (2011) did find certain groups of older
adults wanted to grow in terms of their social relations.
Normative or trait theory?
One argument is that researchers should focus on either normative or trait theories of motivation. Van Iddekinge et al. (2018) meta-analysis shows state measures of motivation better relate
to performance than trait measures. Another argument is that both perspectives are relevant.
Sampling students across the academic year, Holding et al.’s (2019) investigation of the two SDT
pathways in the normative SDT found increased autonomous motivation and decreased controlled motivation related to trait self-control, even after controlling for the Big Five. Hence,
traits interplay with the motivational mechanisms articulated within SDT. A final argument
is that both perspectives are relevant, but one more so. Based on 464 employees working in a
large retail organisation, O’Neill et al. (2011) evidenced the role of both normative and trait
motivations in explaining workplace deviance (see Chapter 4), by focusing on justice and personality. Personality explained deviance behaviour significantly more than justice perceptions
alone, thus emphasising a more influential role of personality in driving deviant behaviour.
However, whether this is the case for other behaviours is yet to be investigated.
Key learning point
Both normative and trait theories play a role in explaining the motivational mechanisms leading to
behaviour.
Chapter 6 Work motivation
187
Context theories
So far, we have suggested that motivation is a property of the person. An alternative view is
that factors within an employee’s work environment propel the employee to act. In this section we review environmental factors that influence motivation, including task, job and team
specific aspects. More broadly we also explore the role of social and technical facets within an
organisation that motivate behaviour.
Characteristics of jobs
Job design theory focuses on how managers design jobs for employees (Berg et al., 2010). An
early specification of how jobs should be designed was introduced during the industrial revolution in the early 20th century, and is referred to as ‘scientific management’, ‘simplification’
or ‘Taylorism’. As a machine-shop foreman, Taylor (1911) re-designed employees’ work with
the aim of improving their efficiency. He used the following strategies:
■
systematically compiled information about the work tasks required;
■
removed workers’ discretion and control over their own activities;
■
simplified tasks as much as possible;
■
specified standard procedures and times for task completion;
While Taylorism made for a well-ordered world, studies found its techniques were associated
with employees having negative attitudes towards their job and poor health (e.g. Kornhauser,
1965; Turner and Lawrence, 1965). Instead it was believed that work tasks which were naturally
more enriching than controlling, would better motivate employees (Herzberg, 1966). Herzberg
(1966) suggested that drivers of dissatisfaction were not necessarily the inverse of drivers for
satisfaction. What he termed ‘hygiene factors’, such as adequate pay, appropriate conditions
of employment, a safe working environment, and other features extrinsic to the work activities
themselves, did not cause satisfaction, but could cause dissatisfaction if they were not present.
In contrast, ‘motivation’ factors included job challenge, recognition and skill use. These factors
led to satisfaction and their absence produced not dissatisfaction, but a lack of satisfaction.
Building on the notion that employees are motivated by tasks which are enriching, Hackman and Oldham (1976, 1980) presented the Job Characteristics Model (JCM) which identifies
five core job characteristics which all have potential to motivate employees (Figure 6.3):
1
Skill variety (SV): the extent to which the job requires a range of skills.
2
Task identity (TI): the extent to which the job produces a whole, identifiable outcome.
3
Task significance (TS): the extent to which the job has an impact on other people, either
inside or outside the organisation.
4
Autonomy (Au): the extent to which the job allows the job holder to exercise choice and
discretion in their work, such as freedom to choose work methods, scheduling and tasks.
5
Feedback from job (Fb): the extent to which the job itself (as opposed to other people) provides information on how well the job holder is performing.
The reason why these job characteristics motivated employees was because they relate to
‘critical psychological states’ within an employee. The first critical state is ‘experienced meaningfulness’ of the work, which results from at least one, or a combination, of the first three
job characteristics. The second state is ‘experienced responsibility’ for work outcomes, which
is triggered by autonomy. The final state is knowledge of the results of the work activities, as
triggered by feedback. When enacted these critical psychological states produce increased
employee motivation, satisfaction and performance. In other words, the JCM shows how
the five job characteristics result in work outcomes (e.g., motivation) by activating the critical psychological states within people. However, the JCM does not solely draw on environmental factors as the determinants of motivation, but also trait factors. Specifically, the JCM
188
Work Psychology
Core job
characteristics
Critical
psychological
states
Outcomes
Skill variety
Task identity
Experienced
meaningfulness
of the work
Task significance
Autonomy
Feedback from the job
Experienced
responsibility for
work outcomes
Knowledge of
results of work
activities
High intrinsic
motivation
High job
satisfaction
High work
effectiveness
Moderators:
Growth-need strength
Extrinsic rewards
Figure 6.3
Hackman and Oldham’s job characteristics model
argues that the three-step process is increased (i.e., moderated) in employees who place more
importance on their growth needs. Therefore, individuals with stronger needs for personal
growth and accomplishment at work will display even greater motivation than their peers
when their critical psychological states are triggered by job characteristics. Accordingly, the
JCM integrates environmental motivators with Maslow’s propositions regarding esteem and
self-actualisation needs. However, it should be noted that research has not universally validated the moderating influence of growth need strength (Tiegs et al., 1992).
Key learning point
Technology has an important role to play in providing a motivational work environment. Trist and
Bamforth’s (1951) study highlights the motivating role of autonomy. Originally, it was believed that
providing miners with technology to cut coal more quickly would make their work easier and increase
their motivation. However, the miners felt that the technology controlled them which inhibited autonomy and motivation. Therefore, when employees cannot control how they complete tasks, they should
have discretion over decision making for what the system for organising work should be (Parker, 2014).
Chapter 6 Work motivation
189
The JCM continues to provoke empirical research (e.g. Millette and Gagné, 2008; DeVaro
et al., 2007), and does specify a lot about what is important in terms of how to design jobs
so people are enriched and motivated (Humphrey et al., 2007). It has also recently been
extended to explain motivation within virtual working (Gibson et al., 2011). However, the
JCM has received the following criticisms:
■
While the job characteristics and resulting psychological states have been shown to
correlate with motivation and satisfaction, the relationship to performance has been
weaker. Hence, the job characteristics relate more to employees’ work attitudes than to
their actual work performance.
■
Researchers have pointed to the insufficiency of the specified job characteristics. Warr
(1987) stated the importance of additional aspects, such as availability of money, physical security, interpersonal contact and valued social position. Morgeson and Humphrey (2006) and Humphrey et al. (2007) state we must also consider the knowledge
requirements of jobs as well. For example, does a person’s work require deep specialist
knowledge of an area and problem-solving? Parker (2014) points out that in increasingly
knowledge-based economies there is a difficult balance to be struck between quality
control processes that guarantee consistent performance standards, and the capacity
to innovate and exercise expert judgement. This ‘ambidexterity’ requires skilful handling to ensure that reliability of performance is high while also offering jobs that highly
skilled people want to do, and which make use of their expertise.
■
The single directional nature of the relationship between job characteristics (i.e., content)
and performance, through the mechanism of motivation, is also questioned as others
have found that in turn performance relates through various mechanisms, such as trust,
back through to job content (Challenger et al., 2012).
An implication of JCM could be for organisations to re-assign tasks, which cannot benefit
from increases in the five job characteristics, to artificial intelligence because employees
would want to conduct different tasks that they find more motivating. Furthermore, advances
in technology can be used to improve task significance. For example, technology provides
virtual interactions between service employees and their customers, regardless of geographical location (Grant et al., 2011). Moreover, advances in technology increase opportunities
for immediate and timely feedback, which JCM describe as a contextual factor driving motivation (Grant et al., 2011). For a helpful overview of the aspects of the work environment
which drive behaviour at work, see Parker and Ohly (2008).
Social and technical drivers
Relational aspects of people’s jobs increase their ‘prosocial motivation’ by connecting them
with the beneficiaries of their work (Grant, 2007). Prosocial motivation is defined as the
desire to expend effort to benefit other people (Batson, 1987). In order to increase this form
of motivation, Grant advocates the design of work to include frequent and in-depth contact
with beneficiaries. This, he says, will increase the person’s perceptions of their impact on
beneficiaries and foster a sense of commitment to them. Prosocial motivation has been
linked to several positive outcomes, including citizenship behaviour that benefits others,
performance ratings from supervisors, willingness to act upon feedback and higher life
satisfaction (Bolino and Grant, 2016). Grant acknowledges that the notion of impact on
beneficiaries overlaps with the JCM core job characteristic of task significance, but with
more of a focus on the social aspect, and the implications for how relationships at work
can be designed.
Furthermore, Gibson et al. (2011) showed that the JCM insufficiently explains behaviour
when colleagues are not working together face to face but in virtual environments, which
are becoming increasingly common following technological advances (see Chapter 13).
190
Work Psychology
In terms of technology, people have different views on an objective amount of (1) electronic
dependence (e.g., some see technology as controlling whereas others view it as a facilitator
of autonomy), and (2) co-presence (the degree to which people feel a psychological degree
of closeness to colleagues, despite how many miles are between them). Such variations in
perceptions across people relate to their perceptions of social aspects between themselves
and colleagues, including intimacy (i.e., closeness of social relationships between colleagues),
and identification (self-categorisation process in which common identities between people
are found).
Motivational propensities of social interactions are also important when considering the
role of teams (see Chapter 9). Increasing globalisation and changes in the organisation of
work has seen an increase in employees working in teams. However, ‘more research is needed
to investigate how team-level processes (e.g., planning and coordination) influence an individual’s task goal,’ (Kanfer et al., 2017: 342) ‘individual motivation in teams has been relatively sparse and concentrated largely on the performance effects of team-level phenomena
(e.g., social loafing)’ (p. 343).
Job crafting
We now consider how employees themselves determine their environments by job crafting (see Chapter 8). Job crafting is the mechanism through which employees cognitively
and behaviourally shape their job roles to enhance their sense of meaning and motivation
(Wrzesniewski and Dutton, 2001). In the modern workplace, employees are often afforded
greater latitude to customise, modify and emphasise their work tasks. However, job crafting is by no means restricted to employees with high levels of autonomy. Wrzesniewski
and Dutton (2001) describe a study in which hospital cleaners were interviewed about the
tasks they conduct outside of their job descriptions. The researchers found that the cleaners ‘crafted’ their jobs in a variety of ways. One cleaner stated that a central part of his job
involved making patients and family members feel comfortable by listening if they wanted
to talk, or by cheering them up if they felt down. Another described how he ignored his
supervisor’s wishes by neglecting to vacuum the visitor’s lounge when family members
were taking a nap. By engaging in job crafting, the cleaners were able to bring a sense of
meaning to their jobs.
Here, we are interested in whether job crafting improves motivation. Yet, most of the
research on job crafting outcomes has focused on work engagement (Harju et al., 2016;
Rudolph et al., 2017; Van Wingerden et al., 2017), defined as ‘an active, positive, work-related
state characterized by vigor, dedication, and absorption’ (Tims et al., 2013). While engagement is related to motivation, it is a distinct construct (although this has been debated, see
Inceoglu and Fleck, 2010). Hence, we should not generalise the role of job crafting in relation
to engagement to the role of job crafting in motivating employees. Recently however, Slemp
and Vella-Brodrick (2013) showed that all elements of job crafting (task, cognitive and relational) related to intrinsic goal-setting at work.
Key learning point
Job crafting describes the propensity employees have with regard to acting to shape their behavioural,
relational and cognitive aspects of their jobs to alter their workplace experiences and motivations
(Wrzesniewski and Dutton, 2001).
Chapter 6 Work motivation
191
Integrating content and context theories
Barrick et al. (2013) argue that both trait and contextual factors influence motivation. Their
theory of purposeful work behaviour suggests personality traits drive purposeful goals, yet
the link is stronger when job characteristics facilitate meaningfulness (see Figure 6.4). Different personality traits relate to people striving for different purposes, including connection,
status, autonomy and achievement. Similarly, job characteristics relate differentially to the
purposes people strive for. Job characteristics create motivational mechanisms (a sense of
meaningfulness in work), which in turn predicts work performance. In sum, instead of seeing
work characteristics as key motivational drivers with personality affecting their impact (i.e.,
a moderator), as is typical within work design theory, Barrick et al. (2013) believe personality
is the origin of motivation, and is expressed in the goals people think are most important.
As such, job characteristics help or hinder the achievement of these personality-based goals.
Marescaux et al. (2013) also integrate work environment and normative drivers. They
show how perceptions of high-performance work systems (HPWS) positively predict organisational commitment and turnover intentions, through the motivational mechanism of
need satisfaction. Similarly, Martinaityte et al. (2019) found that employees’ perceptions of
creativity-oriented HPWS positively relates to employees’ creativity performance through
the mechanism of need satisfaction. This highlights the importance of needs, but also how
motivation can flow from the work context (i.e., creativity-oriented HPWS).
Process theories
Process theories describe the cognitive mechanisms behind selecting goals and striving
towards their accomplishment (Kanfer et al., 2017). Content and context theorists explain
how we get behaviour as a result of people or environments, whereas process theories explain
how we get behaviour as a result of a rational thought processes. Process theories therefore take a different perspective in explaining behaviour. Drawing on cognitive processes,
some Process theories explain why people choose a behaviour over another (i.e. goal choice
theories), for example Expectancy theory), and others explain why people keep persisting
at a behaviour (i.e. goal striving theories), for example Self-regulation theory. There are also
integrative theories, such as Goal-setting theory, which explain both the situation of choosing a behaviour and persistence at it (Kanfer et al., 2017).
Goals are ‘internal representations of desired states that direct attention, organize action,
and sustain effort aimed at achieving those states’ (Kanfer et al., 2017: 343). We review two
theories of how we select goals.
Goal striving
Self-regulation theories
While the theories above explain what behaviour an employee selects, Goal striving theories
describe the processes through which goal intention is maintained, alongside perseverance,
revision or abandonment of goals – a process called self-regulation (Kanfer, 1977). There are
three cyclic steps in self-regulating including: self-monitoring (i.e., paying attention to events,
behaviours and feedback related to the goal); self-evaluating (comparing the goal state to the
current state, or goal progress); and self-reacting (emotionally feeling the consequence of
self-evaluations, whereby discrepancies between desired and goal states trigger affective and
motivational responses).
192
Work Psychology
Integrative process theories
Goal-setting theory
Goal-setting theory (GST) has been a highly influential motivation theory, as it integrates the
cognitive process of choosing a goal with the striving to deliver it. GST was developed in the
1960s by Edward Locke and his associates and is based on the premise that intentions shape
actions. Within GST, a goal is defined as ‘what an individual is trying to accomplish; it is the
object or aim of an action. The concept is similar in meaning to the concepts of purpose and
intent’ (Locke et al., 1981: 126). Figure 6.4 represents goal-setting theory.
The most fundamental proposition of goal-setting theory is that difficult goals which are
expressed with a clearly defined performance level, produce higher levels of performance than
other goals, or no goals at all. Difficult and specific goals have this effect by focusing a person’s
attention on the task, increasing the amount of effort they put into it, increasing the length of
time they exert effort, and encouraging the person to develop strategies for goal achievement.
These ideas have found expression in popular management practice, particularly in the form
of so-called ‘SMART’ – specific, measurable, agreed, realistic and time-based – goals. Reviews
by Locke and Latham (1990) and Mento et al. (1987) arrived at several conclusions, most of
which fully or substantially supported goal-setting theory. Locke et al. (1981) reported that
in goal-setting field experiments, the median improvement in work performance produced
by goal setting was 16 per cent.
Goal-setting interventions have increased energy-efficient behaviour (Loock et al.,
2013) and productivity and persistence on analogue work tasks (Tammemagi et al., 2013).
Goal-setting theory has also been successfully applied to diversity training (Lindsey et al.,
Moderators1:
Goal commitment
Goal importance
Self-efficacy
Feedback
Task complexity
Goal core:
Specificity and
difficulty (e.g.
performance and
learning goals,
proximal goals)
Willingness to
commit to new
challenges
Performance
(e.g. productivity,
cost improvement)
Satisfaction with
performance and
rewards
Mechanisms2:
Choice/direction
Effort
Persistence
Strategies
1
2
factors which strengthen or weaken the effects the aspects of the goals have on performance.
factors which enable the core aspects of the goals to drive changes in performance.
Figure 6.4
Goal-setting theory
Chapter 6 Work motivation
193
2015), and has closed achievement gaps between men versus women and minority versus
majority ethnic groups (Schippers et al., 2015).
Despite the allure of findings on difficult and specific goals, research has identified that the
process is not always that simple. People and circumstances vary, sometimes in ways which
affect the goal-setting process and applying GST to a work situation is more complex than
it initially seems. Some of these complexities are becoming well-understood, and they are
shown as ‘moderators’ in Figure 6.4.
Goal commitment, the unwillingness to abandon goals (Wright et al., 1994), signals the
effect of goal setting on behaviour is greater when goals matter to a person. If individuals are
not committed to a goal, they will not exert effort in its pursuit. Commitment can be thought
of as another word for motivation – if I am committed to a goal, then that could be another way
of saying I am motivated to achieve it. One way to increase goal commitment is for people to
discuss what goals to set (Erez, 1986). However, others suggest this is not necessary if people see
their goal as challenging but not impossible (Locke and Latham, 1990). This poses an important practical question: can goal assigners set specific and challenging, but attainable, goals
without consultation with those trying to achieve them? (Haslam et al., 2009). For example,
people with low self-efficacy typically perceive goals to be too difficult. Hence, some authors
still recommend that goals should be set collaboratively as opposed to being assigned to people.
Ordonez and her colleagues (2009) raised several issues with GST in the wonderfully titled
article ‘Goals gone wild’. However, we do note that Locke and Latham (2009) naturally defend
their theory against these claims which they argue were based on anecdotal, rather than
empirical evidence. GST has received the following critiques:
■
Goal-setting theory is not always simple to implement in the workplace. In most organisations there are multiple goals to be achieved. For organisations, and for individuals
and groups within them, there may be some conflict between goals. Achievement of one
may be at the expense of another. For example, achievement of reductions in staff costs
may be at the expense of quality of customer service. So, when a manager is given goals
to reduce costs by 10 per cent and to increase customer service by a similar amount, each
goal on its own may be achievable but in combination they are not. This may well negate
the positive effect of goal-setting.
■
Goal-setting can narrow people’s focus in a way that makes them blind to issues that
appear unrelated to the goal, such as health and safety (Ordonez et al., 2009).
■
GST offers a theory of pursuit of behaviour but is not helpful when there is a choice.
When set multiple goals, individuals tend to concentrate on just one, which may be the
easiest to achieve (Ordonez et al., 2009).
■
GST can create a culture of competition, which encourages risk-taking and unethical
behaviour and to induce dissatisfaction when goals are not met (Ordonez et al., 2009).
■
Another criticism of GST, and process theories in general, is that they assume people
make rational decisions, when they may be significantly influenced by emotion (Kanfer
et al., 2017).
■
A final criticism relates to the static nature of GST. Fried and Slowik (2004) show how time
influences goal-setting processes and job performance in work organisations, making it a
more ecologically valid explanation of behaviour in complex, continually evolving, work
environments.
Key learning point
The setting of performance goals that are specific and difficult (but not impossible), and to which the
person feels committed, is likely to improve their work performance. This is especially the case if the
person receives feedback on progress, and rewards for successful performance.
Work Psychology
194
Theories of reasoned action and planned behaviour
The Theory of Reasoned Action (TRA) (Fishbein, 1967) demonstrates the importance of attitudes, beliefs and intentions in choosing a goal. Before the development of TRA, many scholars had recommended eliminating attitudes as motivators of behaviour, as there was little
evidence that they influenced motivation. However, Fishbein distinguished attitudes towards
a specific behaviour from attitudes towards the objects (or goals) at which the behaviour is
directed (Montano and Kasprzyk, 2015). Fishbein showed that attitudes towards the behaviour
is a much stronger behavioural predictor than attitudes towards the outcome of the behaviour
(Fishbein and Ajzen, 1975). For example, an individual’s attitude towards running will be a
much better predictor of engagement in running activities than the object at which running
behaviour is directed (e.g. completing a marathon or getting in shape). TRA states that behavioural intention is the main cause of whether the individual will perform a specific behaviour
(see Figure 6.5). Behavioural intention is a composite of two factors: (1) the person’s attitude
towards the behaviour, and (2) subjective norms. An attitude is defined as an individual’s
general feeling of favourableness or unfavourableness for that behaviour (Ajzen and Fishbein,
1980). When attitudes towards the behaviour are favourable, an individual will have stronger
behavioural intentions. Comparatively, subjective norms are an individual’s perception that
most people who are important to them think that they should perform the behaviour (Ajzen
and Fishbein, 1980). When subjective norms are strong (i.e. an individual believes that important others think they should perform the behaviour) behavioural intention is greater.
TRA breaks down attitudes and subjective norms into two further component parts (see
Figure 6.5). Attitudes comprise ‘behavioural beliefs’ – the belief that performing the behaviour
will lead to certain outcomes, as well as ‘evaluations of behavioural outcomes’ – the desirability
External factors
Other individual
difference factors
Recognised
power
Control
beliefs
Recognised control
Attitudes toward
targets
Personality features
Motivation to
comply
Normative
beliefs
Subjective norm
Demographic
factors
Assessing
behavioural
outcomes
Behavioural
beliefs
Attitude
Intention to perform the behaviour
Behaviour
Figure 6.5
Theory of reasoned action and planned behaviour
Note: The unshaded boxes show the Theory of Reasoned Action; the entire figure shows the Theory of Planned
Behaviour. Figure adapted from Montano and Kasprzyk (2015).
Chapter 6 Work motivation
195
of the likely outcome. Subjective norms are comprised of ‘normative beliefs’ regarding whether
the salient referent thinks they should perform the behaviour (Ajzen and Fishbein, 1980)
and ‘motivation to comply’ with the referent’s wishes. However, it was realised that one of the
main factors influencing the explanatory power of TRA was whether people could exercise
control over the behaviour. For example, an individual with strong behavioural intentions to
go running may not be able to due to work and family commitments. Therefore, Ajzen (1991)
added ‘perceived control’ to the model, which created the Theory of Planned Behaviour (TPB).
Perceived control is a composite of ‘control beliefs’, which refer to the presence of facilitators
and barriers to behavioural performance and the ‘perceived power’ of these facilitators/barriers
(Montano and Kasprzyk, 2015).
One of the strengths of TPB is that it provides a framework to demonstrate the behavioural
and normative beliefs that influence behaviour. This allows interventions to be targeted
towards changing the unhealthy beliefs that influence attitudes and consequently behavioural intentions. Accordingly, TPB has been used to explain why people engage in unhealthy
behaviours, such as smoking and alcohol use. It has also been used to inform health promotion efforts towards exercise, HIV/STDs, cancer screening and oral hygiene (Montano and
Kasprzyk, 2015). There is a substantial body of evidence which supports the propositions
of TPB (Montano and Kasprzyk, 2015) and a review of intervention studies concluded that
interventions conducted publicly with groups were more successful than those conducted
privately with individuals (Steinmetz et al., 2016).
TPB has been used to explain the motivations behind pro-environmental behaviour at
work (e.g. switching off one’s PC and recycling; Greaves et al., 2013), intentions to start
cycling to work (Lois et al., 2015), and a willingness to pursue specific types of career (Giles
and Rea, 1999). However, the theory has not been used extensively in explaining people’s
motivations towards their work tasks. This is possibly due to the fact that a central element
of the model pertains to perceived control, which many workers do not have over their work
task since they are often formally outlined in job specifications.
Expectancy theory
Vroom’s (1964) Expectancy theory states people select goals based on a cognitive process of
determining: (a) the satisfaction expected from achieving the goal (valence), (b) the likelihood
of achieving the outcome based on a particular level of performance (instrumentality) and
(c) the likelihood they can themselves actually reach that level of performance (expectancy).
The more favourable these evaluations, the more likely the person is to select the behaviour.
Given Expectancy theory explains goal choice, these three factors of valence, instrumentality
and expectancy, have been referred to as the antecedents of intentions (Fitzsimmons and
Douglas, 2011). Expectancy theory has received some empirical support (e.g. Georgopoulos
et al., 1957; Mitchell and Nebeker, 1973), and allows for differences between people to be
considered, unlike normative content theories.
■
Scholars critique the assumption that humans are rational in their decision making and
choices (Kanfer et al., 2017).
■
Scholars question the multiplicative process (Van Eerde and Thierry, 1996). For example,
Fitzsimmons and Douglas’ (2011) study on entrepreneurial behaviour, does not support
the claim that when expectancy and value are both high, motivation is strongest. Contrary to expectancy theory, an entrepreneur will act entrepreneurially if either perceived
desirability or perceived feasibility is high.
■
Expectancy theory is classified under ‘process theories’ meaning that by definition it
should be dynamic and not static, yet it is not. It explains ‘one-time’ decisions, whereby
properties of expectancy and value are treated as constants throughout the decision forming process, which is unrealistic. In other words, expectancy theory is insufficiently fluid
to explain situations in which pursuing one goal prevents or detracts from another at a
point in time (Vancouver, 2012).
Work Psychology
196
Nevertheless, Expectancy theory is still influential. For example, Kopetz et al. (2019) use
Expectancy theory to explain why people enact risky behaviour to achieve a goal, rather
than finding an alternative, non-risky way.
Integrative theories
The early 2000s saw the beginning of content theories being linked to process theories. Barrick et al. (2002) and Judge and Ilies (2002) explained how personality traits indirectly relate
to performance through goal-setting. Building on this, Locke and Latham’s (2004) integrated
model (see Figure 6.6) explains how motivational mechanisms driving behaviour do not just
come from within people, or the environment (i.e., context theories), or the rational decision
processes people make (i.e., process theories), but through a combination (also see Latham
and Pinder, 2005). Locke and Latham state ‘needs’, expressed within personality, are the
underlying basis of motivation. These affect the kinds of goals people deem to be important.
These factors, in addition to the core features of GST, explain when and how goals instigate
action. Most of the links in the model have at least some empirical support and others have
a great deal.
However, the model lacks feedback loops; one might expect performance, rewards or
satisfaction to influence goal choice and goal behaviour as well as self-efficacy. Moreover,
integrative perspectives, such as this one, do not explain how employees deal with multiple
goals, or what motivates them to select some tasks over others (for a conceptual discussion
see Unsworth et al., 2014).
Key learning point
Integrative theories of motivation show how individual differences in needs or personality influence
the kinds of goals and job characteristics people want.
Needs
(need
theory)
Values,
personality
(personality
theory)
Incentives,
e.g. money
(extrinsic
motivation)
Figure 6.6
Goal choice
(goal-setting
theory;
expectancy
theory)
Self-efficacy;
self-identity
(social cognitive
theory;
goal-setting)
An integrated model of work motivation
Behaviour
aimed at goalachievement
(goal-setting
theory)
Performance
Goal moderators,
e.g. feedback,
commitment
(goal-setting
theory)
Satisfaction/
dissatisfaction
Rewards for
performance
Work
characteristics
( job design
theory)
Chapter 6 Work motivation
197
Summary
Throughout this chapter we have covered how different factors and processes creating motivation result in behaviour. First, we reviewed theories supporting motivation as a product of
human needs, which afford behaviour (i.e., normative content perspective). Following SDT,
we described different ways in which external requirements to perform tasks can be incorporated into people’s sense of self, blurring the distinction between extrinsic motivation
(doing something for the rewards it brings) and intrinsic motivation (doing something just
for the joy or interest in doing it). In contrast, we reviewed factors which vary across people
in affording action (i.e., trait content perspectives). Second, we considered motivation as a
property of, not the person but, the social and technical features of the work environment
(i.e., context theories). Third, we reviewed process motivation theories, outlining how people select and strive towards goals. Finally, we described how more recently scholars have
integrated different perspectives on motivation to better explain predictors of behaviour.
Closing case study
Gender pay gap at the BBC
Reference: BBC. (29 June 2018). Timeline: How the BBC gender pay story has unfolded. Retrieved 6
May 2019, from https://www.bbc.co.uk/news/entertainment-arts-42833551
In July 2017 the BBC released a list of its best-paid stars. Of the top ten highest earners, only Claudia
Winkleman was female and her salary bracket (£450,000–£499,999) was substantially lower than the
BBC’s highest earning man (Chris Evans; £2,200,000–£2,249,999). The publication sparked a chain
of events which revealed that the average pay gap between men and women at the BBC was 9.3%,
meaning that for every £10 the average man earns, the average woman earns £9.07. It also led Carrie
Gracie, the BBC’s China editor, to resign due to pay inequality with male colleagues. Gracie stated she
was dismayed to learn that two male international editors at the corporation earned ‘at least 50%
more’ than their two female counterparts.
The BBC is not alone in paying men more than women. In 2019, it was revealed that out of 10,443
British firms, 8,124 pay men more, while only 1,424 pay women more. How does a gender pay-gap
affect the motivation of female employees? We seek to answer this question by recapping several of
the motivation theories covered in this chapter.
Perhaps the most pertinent motivation theory regarding the gender pay gap is organisational justice
theory. This suggests that women will be concerned with whether their pay is a fair relative to that
received by men. They will also want to see fair procedures for allocating pay. If procedures are not
seen as fair, the attitudes and motivation of female employees will be adversely affected, and the
organisation is also likely to suffer. Bloom (1999) studied the pay of North American baseball players
over the years 1985 and 1993 and found that the more equal the pay of the different players in the
team, the better individuals and teams performed. This suggests that the pay of others matters and
that pay equality is linked to organisational performance.
Equity theory also suggests that the gender pay gap could have serious motivational implications.
In many industries women are underrepresented in senior leadership positions. As such, a female
employee who has been overlooked for a promotion, despite putting in similar or greater amounts
198
Work Psychology
of effort than a male counterpart, may engage in compensatory behaviour to maintain the balance
between her inputs and outputs. The compensatory behaviour may involve absenteeism, reduced
performance or even leaving the organisation for a different job.
Recently, Rynes et al. (2004) have argued forcefully that the importance of pay as a motivator is often
underestimated by human resource managers. They argue that people’s behaviour suggests that pay
is more important to them than they say it is, but particularly so when the distribution of pay seems
unjust. Indeed, Carrie Gracie felt so strongly about the issue that she resigned from her job.
Interestingly, other theories place less emphasis on pay. According to SDT, pay will be most important
in cases of external regulation and to some extent introjected regulation, when extrinsic factors, such
as pay, have a stronger influence on motivation. This suggests that a gender pay gap within a firm
would have a greater influence on a woman’s attitude, motivation and performance when her job
does not satisfy the psychological needs of autonomy, relatedness and competence.
Herzberg (1966) and others also tend to argue that pay is not a key motivator at work, though it is
acknowledged that this conclusion depends on some basic level of pay being provided in order to meet
basic needs. For Maslow, pay would be a motivator only for people functioning at the lower levels of
the hierarchy of needs. However, advocates of need for achievement point to the fact that pay, and
other material rewards, often signal success. Therefore, women with higher NFA are likely to be more
dissatisfied with a gender pay gap, because it signals that they are less successful than highly paid men.
We have focused on how pay affects motivation within a job, yet pay procedures can also affect
peoples’ motivations to apply to work at a specific company. Currently, all UK companies with 250
or more employees are required to publish an annual pay report. Therefore, potential applicants can
research the gender pay gap at their organisation of choice. This should incentivise organisations to
close their pay gaps, or risk missing out on talent who value gender equality.
Discussion points
■
Discuss whether trait theories of motivation offer better explanations of motivation
than normative content theories.
For tutors:
Students could either choose to (1): Argue for trait theories. Describe and explain them.
Highlight their strengths. Then describe normative theories and highlight weaknesses.
Or (2): Argue for normative content theories. Describe and explain them. Highlight
their strengths. Then describe trait theories and highlight weaknesses. A third approach would be to argue that neither theory is superior than the other, and that
both play a role, and in particular they may play a combined role. (Note, this argument would involve more reading beyond this chapter.)
■
Compare and contrast content and context theories of motivation.
For tutors:
Students should define both categories of theories and describe how they differ based
on their definitions. Students should then consider, what do each of the theories
explain? What do each of the theories not explain? Are there differences between the
two in terms of these answers? How do each of the theories account for implications
of an ageing workforce, development of technology, diversity in the workforce etc?
What are the implications of both of the theories for the motivating with pay debate?
Are there differences between the two theories?
Chapter 6 Work motivation
■
199
It is often claimed that goal-setting is a theory of motivation which works. Examine
whether it works better in some circumstances than others.
For tutors:
Students should argue that ‘yes’ it does explain the process of motivation, in driving
outcomes, better in some situations than others. Students would need to define GST
and describe its various components. They would then need to provide evidence supporting its role in positive outcomes. Students would then need to review the ‘moderator’/’contingency’ factors which influence how effective GST is on different outcomes.
■
How can motivation theories be combined to get the best ideas from each?
For tutors:
Students should define the various categories of motivation theory, alongside presenting a convincing argument that no single theory is superior given they focus on different aspects. Students would then need to describe and justify at least one integrative
theory of motivation. As part of this they would need to argue why it is better than a
single theory, and why they have selected the integrative theory of motivation they
have. They should also end with a discussion of what we still do not understand based
on the current understanding of integrative theories of motivation. (Note, being able
to answer this question well would involve more reading outside this chapter.)
Relevant websites
These sites describe training courses in motivational techniques (for managing self or others)
and describe some well-known motivation theories, usually the oldest and simplest ones:
■
http://humanresources.about.com/od/motivationsucces3/a/lead_motivation.htm is
part of a managers’ self-help site. This item encourages managers to take on responsibility for the motivation of the people who work for them.
■
http://changingminds.org/explanations/theories/a_motivation.htm gives an index of
different theories of motivation. Clicking on a theory brings you a little more information about it.
■
https://www.mindtools.com/ offers a long list of motivation topics which you can find
by typing motivation into the search facility. To access the full articles, you need to join
mindtools.
Suggested further reading
1 Kanfer’s (1990) chapter reviews theories and integrative approaches, which existed prior
to 1990.
2 Steers, Mowday and Shapiro (2004) review motivation theories and articulate future
research.
3 Ramlall (2004) reviews motivation theories and their implications for employee
retention.
4 Gary Latham’s book Work Motivation (Sage, 2007) offers a theoretical and practical
discussion of motivation.
5 Expanding your understanding of integrative approaches to theorising motivation,
Ed Locke and Gary Latham’s 2004 article attempts to specify an integrated theory of
motivation. It is quite heavy going but well written nevertheless, and a good way of
expanding both your knowledge of theories and your understanding of how they might
fit together.
CHAPTER 7
Training and development
LEARNING OBJECTIVES
After studying this chapter, you should be able to:
1 identify and explain the three components of the training cycle;
2 describe and explain the purpose of three different levels of training needs analysis;
3 explain how learning theories can inform training design;
4 understand how technology can facilitate training delivery;
5 explain why teams have specific training needs;
6 identify barriers to learning
transfer and how these can
be addressed;
7 understand the importance
of training evaluation for
determining cost effectiveness and utility of training
interventions;
8 outline how leaders engage
in training and development;
9 describe how training is
used to support diversity in
organisations;
10 identify how an increase
in learner control poses
challenges for learning
organisations.
Chapter 7 Training and development
201
Opening case study
Thai cave rescue: training needs for extreme conditions
In 2018 a group of 12 Thai young soccer players and their coach became trapped in the Tham Luang
cave in northern Thailand when sudden rising water forced them deep into the cave, leaving them
unable to escape. After a huge search effort, the group was located but getting them out of the
complex cave network required a large-scale operation between the Thai government, Thai military
and international cave divers, which lasted 18 days.
Training played a key role in bringing the boys out alive. Once they were located in the caves, divers
started training the boys to swim and understand basic principles of diving so they could be led out
of the cave safely. The coach was also able to train the teenagers in meditation, important not only
to keep them calm in the dark, wet and confined conditions, but also slow their breathing to reduce
oxygen consumption.
However, early in the rescue mission it became apparent that the Thai Navy SEALs were not sufficiently trained to deal with the conditions. Although they were experts in sea-diving, they did not
have the knowledge or skills to swim through the murky water, tight passages and cope with being
unable to return to the surface when encountering difficulties navigating the complex cave system.
Instead they had to turn to a team of international divers who had cave diving experience.
After the success of the mission, the Navy SEAL Commander argued ‘SEALs need cave-diving training’ and has since added cave diving to the SEALs’ training regimen to prepare for such situations.
This case highlights the importance of taking time to analyse what training needs are, but also to
examine how different conditions or contexts might influence how skills are transferred into the
workplace.
Introduction
Training and development activities have the potential to benefit individuals, organisations
and society as a whole. Engaging in training means that employees can develop a portfolio of
skills that can improve their earning potential, chances of promotion, self-confidence and
opportunities to participate in more interesting work. Now recognised as an important contributor to employee performance, organisations spend more than $360 billion globally on
training and development each year (Training Industry, 2017).
In economic terms, a country’s skill base is linked to its national competitiveness: the
existence of a substantial pool of skilled workers is not only important for local business, but
also essential for attracting investment from international sources like global companies. For
example, in recent years Mexico has become an important location for car production for
manufacturers like Toyota, BMW and Audi, in part because the country has trained its workers
in engineering skills that cater specifically to the automotive industry.
Governments therefore encourage organisations to invest in the training and development of staff to ensure a national workforce can successfully compete in international markets. Singapore, for example, experienced a significant skills shortage after rapid growth in
the digital economy. However, instead of relying on foreign workers to fill the skills gap, the
Singaporean government introduced incentives for hiring locals, thereby increasing pressure
on organisations to upskill and reskill home-grown talent by training their staff effectively.
Work psychology has played a very important part in improving the effectiveness and
impact of training and development initiatives. Research shows that skilled individuals
202
Work Psychology
perform work faster and more safely, make fewer errors, produce higher quality work, and that
training can lead to improved job performance and other benefits like improved efficiency
and lower stress among employees (Aguinis and Kraiger, 2009; Bell et al., 2017). There is also
a great deal of research and theory providing good insight into how employees learn, the
training methods likely to result in successful transfer of learning to the workplace, and how
training can be evaluated to ensure maximum return on investment (Salas et al., 2012). In
this chapter we draw on this evidence base to introduce core concepts (i.e. the training cycle,
training-needs analysis, training design and training evaluation) and describe methods
used to support employee learning. We also reflect on the emergence of learning organisations and consider future opportunities and challenges for the science of training. To begin
with, however, it is important to clarify what we mean by training and how it compares to
the concept of learning.
Point of integration
The learning organisation is an important concept in organisational change (see Chapter 12). This
underlines the importance of training and development before, during and after episodes of organisational change.
Learning is a much broader concept than training. We learn all the time by observing and
listening to others, particularly when we encounter new demands like taking on a new role
or joining a new company. For example, organisational newcomers often spend a lot of time
watching and listening to others in order to learn the social norms and values that influence
how people work in that setting. This learning is described as ‘informal’ because it is not
structured or guided by the organisation: the learner decides what might be important to
learn. In contrast, training is a subcategory of learning that involves formal, structured learning activities that meet specific business needs.
For this chapter we draw specifically on the definition provided by Salas et al. (2012: 77),
which describes training as ‘planned and systematic activities designed to promote the acquisition of knowledge (i.e., need to know), skills (i.e., need to do) and attitudes (i.e., need to
feel)’ that have the goal of creating ‘sustainable changes in behaviour and cognition so that
individuals possess the competencies they need to perform a job’. We also follow Aguinis
and Kraiger (2009) in using ‘training’ to refer to both training and development activities.
How training has changed
Over recent decades there have been many changes to the ways in which people work that
have also impacted on how training has evolved (Cascio, 2017; Kraiger, 2014). The photographs in Figure 7.1 illustrate a few of these changes. The group of 1950s engineering apprentices with their tutor in photograph A shows one of the most common means by which
workers were trained by companies until the early 1970s: the apprenticeship. Most apprentices spent several years with a company learning a particular skilled ‘craft’, and as job tasks
and demands did not change very much apprentices often stayed with the same company
in a ‘job for life’. Such stability meant that companies could afford to invest time and money
in lengthy and relatively expensive training programmes because there was every likelihood
that their investment would be repaid. However, job demands and careers have changed, and,
as shown in Photograph B, today’s workers are more likely to work in flexible groups and use
digital technology to monitor continual streams of information while also communicating
with colleagues and customers who might be working in locations across the globe.
Chapter 7 Training and development
Figure 7.1
203
The changing context of training (left: c.1950; right: modern day)
Source: left: Allan Cash Picture Library/Alamy Stock Photo; Right: Pexels GmbH, https://images.pexels.com/
photos/1204649/pexels-photo-1204649.jpeg?auto=compress&cs=tinysrgb&dpr=2&h=750&w=1260.
In most workplaces, the content of training has therefore had to change: this has involved
a shift in focus from developing motor skills required to produce a physical product, to an
emphasis on the cognitive skills needed for knowledge work, innovation and digital technology, and the social skills required for working in culturally and spatially diverse groups and
teams. The delivery of training has also become much more flexible and fast paced in response
to volatile global markets, and short product cycles, driven by technological change. Moreover, an increase in flatter organisations (i.e. fewer levels of management and multi-skilled
employees) has also meant that training delivery is much less likely to be manager or
instructor-led; instead employees are expected to guide their own development through
e-learning activities such as virtual classrooms and short digital lessons (Brown et al., 2016).
Importantly, as ‘jobs for life’ have declined there has been a trend towards learner-centric
training where employees have greater control over, and more responsibility for, what and
how they learn. While potentially beneficial for employees, organisations may struggle to
engage workers in the training they require to meet specific business aims (Sitzmann and
Weinhardt, 2018). Similarly, in the case of highly marketable skills, employers can find it
less financially viable to invest in long periods of training because they lose employees to
competitors (CIPD, 2017). As a result, organisations are having to look for ways to make the
training they offer more desirable so that they can attract and retain employees. In doing so
training and development has become an important tool for organisations to distinguish
themselves from competitors as part of their employer brand (Cascio, 2014).
Point of integration
The way that workers shape their careers (see Chapter 11) has changed significantly over recent years.
This has consequences for the way that employers make decisions about the type of training made
available. Employee attitudes such as organisational commitment and job satisfaction (see Chapter 5)
can also be heavily influenced by the availability of training.
204
Work Psychology
1. Training needs
analysis
3. Training evaluation
Figure 7.2
2. Training design
The training cycle
Employability skills
Think about your own training and development learning. What training have you received from employers, and how is the course you are studying likely to contribute to your learning for future jobs? Looking
ahead, what other training might an employer expect you to engage in? What training will you expect
from your future employer? How will you ensure you have the skills in order to compete in the employment market?
The training cycle
Training has the potential to deliver many benefits, but it is important to note that training
per se may not necessarily be beneficial. Training that is not well developed or well executed
could impact staff morale, lead to poor performance, risk employee and customer safety and
be costly to put right. For training to be effective, it must follow three components of the
training cycle (see Figure 7.2): Training needs analysis (TNA), which provides the information
needed for training design (TD), where appropriate development activities are implemented,
and training evaluation (TE), which establishes whether the training has been successful
and should be used to provide feedback to improve future training activities. Each activity
is equally important and, like a tripod, if one leg is poorly constructed or missing, then the
training is likely to fail. Each of these elements (TNA, TD and TE) is considered in more detail
in the following sections.
Key learning point
Training needs analysis (TNA), training design (TD) and training evaluation (TE) constitute the three
stages of the training cycle
Exercise 7.1
If training development does not follow the training cycle it can lead to several costs for organisations. What might be some of the costs associated with poorly implemented training in the following
sectors: Hospitality, healthcare, financial industry?
Chapter 7 Training and development
205
Training needs analysis
Before starting any systematic training or development activity, those responsible should
be confident that it will produce worthwhile results. Unfortunately, detailed TNA does not
always happen. It may be that managers believe that training (any training) is a ‘good thing’.
On occasions, training programmes can take on a life of their own; occurring regardless of any
clear and established need for them, and employees may find themselves on a training course
simply because their manager has their own performance target to ensure staff participate in
a set number of training days per year. Alternatively, those who control budgets or resources
may prefer to rely on personal ‘instinct’ rather than undertake what might be perceived as
a costly TNA. However, a poorly constructed training programme can be substantially more
costly for an organisation. Money and time are needed to develop, implement and manage
the training, and there may be a decline in productivity while employees are away from their
work so, not surprisingly, senior managers are likely to ask for a justification of expenses in
terms of return on investment (ROI).
Point of integration
There is frequently a trade-off between the need for training and selection activities. If well-qualified
and appropriately skilled employees already exist, then good selection processes may off-set the need
for extensive TNA activities. If, however, there is a shortage of ready-equipped workers then TNA takes
on greater importance, and individuals’ potential to benefit from training (their trainability) takes on
added importance.
The primary aim of TNA is to establish training objectives (TOs) or, put simply, what the
training needs to achieve. Training objectives influence how training is designed, delivered
and evaluated, and must be specific to ensure training is as focused as possible. The TNA
should identify who needs to be trained, what knowledge and skills need to be developed,
when and where training needs to happen and examine factors that may help or hinder
training effectiveness, such as level of resources, or support from management. To capture
this information, it is beneficial if training needs are analysed at three basic levels (i) organisational analysis, (ii) job/task/role analysis and (iii) person analysis.
1 Organisational analysis
The purpose of undertaking TNA at the organisational level is to understand where training
activities fit into wider organisational systems and how they relate to the wider organisational
strategy. The question being asked is: ‘What are the training needs of the whole organisation?’
This begins by identifying the organisational objectives that are likely to have a training solution through examining the company’s strategic aims and policy statements and discussing
these with senior personnel. Training needs often arise as a consequence of planned change
(e.g., a manufacturer entering a new market) and unplanned change (e.g., a clothing chain
facing a sudden drop in sales), but it is important to note that training is not a panacea for
the challenges facing organisations. Other interventions may be more suitable; for example,
the clothing chain may find that updating the clothes they sell to meet current fashions or
improving their website to make it more user-friendly are better ways to improve sales.
Organisational TNA is also necessary to determine whether investment in training is appropriate across different parts of an organisation. Senior managers usually control spending on
206
Work Psychology
training and development and in order to justify an investment: as such they need to know
why the training is needed, how much it will cost, and how it will help the organisation
achieve key performance indicators. When training is seen as important for achieving organisational goals, managers at all levels are more likely to endorse staff training, allocate time to
attend it, and align training participation with other procedures like performance appraisal
(Ostroff and Bowen, 2000). Moreover, if the strategic importance of training is communicated
well to employees, they are more likely to consider it relevant to their role, be more receptive
and engaged with it, and less likely to resent the training or exert minimal effort (Sitzmann
and Weinhardt, 2018). Tying training objectives to organisational strategy is therefore important for its effectiveness.
In larger organisations, HR managers are often responsible for annual reviews of training needs across different business areas. Information pertinent to an organisational review
is therefore routinely collected as part of other HRM systems, including performance management, succession planning and recruitment. Occasionally, there is a need for a ‘global
review’ which involves examining every job category in a company in order to create or revise
job specifications, then an assessment of every employee against the new specifications for
their role, with any shortfall indicating a potential training need. However, this costly and
time-consuming form of organisational review is comparatively rare and most likely used if
a company is undergoing radical change, such as a sudden need to downsize or restructure.
Not all organisations are large enough to have big HR functions, and smaller firms and
new start-up companies often require more responsive and focused methods of organisational TNA. One such method is Critical Incident Analysis. This involves initial interviews
with members of the core management team to examine specific incidents of good and poor
organisational performance. These incidents can be used to identify how and where training
could help to achieve business objectives, and to identify likely barriers and facilitators to
implementing training successfully. This method is also useful in larger organisations when
there is a need to understand how future training activities might need to adapt to changing
business objectives and unpredictable business environments.
2 Task analysis
In many ways, TNA at the task level is similar to job or role analysis (see Chapter 3). But instead
of defining the knowledge, skills and abilities (KSAs) an employee needs to perform a role,
task analysis identifies what training is needed in order to develop the requisite KSAs. Task
analysis has its origins in the early 1900s scientific management movement, where tasks were
broken down into specific operations in order to identify how they could be completed most
efficiently.
Today, hierarchical task analysis (HTA: Annette and Duncan, 1967) is a method that is
still used widely to divide a job or job component into its tasks and subtasks. By way of illustration, consider the role of sales assistant in a clothes shop. The overall task can be considered ‘staffing the sales desk’. Sub-operation one might be ‘sales transaction’, and this can be
subdivided further into: (i) taking the garment from the customer; (ii) locating and scanning
the barcode; (iii), asking for payment; (iv), processing payment; (v) asking customer if they
want a bag, (vi) folding the garment; (vii) bagging the item; (viii) providing the receipt;
(ix) thanking the customer. HTA is beneficial when tasks have clear observable components
and are relatively unchanging, but it is less effective at capturing the cognitive work required
in a role. For example, the sales assistant might need to engage in small talk with the customer, persuade the customer to sign up for a store card, or make decisions about how to
respond to customer complaints.
Cognitive task analysis (CTA: Zsambok and Klein, 1997) looks beyond the physical tasks,
to examine the mental activities necessary to perform a role. CTA refers to a series of methods, such as interviews, focus groups and simulations that involve taking subject-matter
Chapter 7 Training and development
207
experts through a series of detailed and systematic steps where they are asked to verbalise how
they analyse information, make decisions, and cope with unexpected problems (Dachner
et al., 2013; Militello and Hutton, 1998). These methods are worthwhile for knowledge and
innovation roles where most of the work is predominantly cognitive and cannot be directly
observed, like computer programming.
Both HTA and CTA can be time consuming and costly, so organisations may look to other
methods such as questionnaires to collect job-related information quickly. The Position Analysis Questionnaire (PAQ; www.paq.com) for example, measures 300 job characteristics and
can be used to gain insight into a role from different perspectives, including customers, managers, new or experienced-job holders. This can be useful for capturing information about
the wider context of a job and establish which tasks are team-related. The sales assistant,
for example, may need to work with other members of staff to manage queues efficiently.
Establishing the interdependence of roles with a team task analysis is particularly suitable
for occupations that rely heavily on teamwork, such as those in healthcare or the military
(Shuffler et al., 2018).
3 Person analysis
Unlike task analysis, which determines the skills needed to perform a task or job, person analysis involves identifying who needs training and what kind of training they require. A person-level
TNA may be a routine element of performance management, or it may be in response to
changing business needs, such as deciding whether a group of employees needs to learn how
to use a new piece of software. It is also part of the broader philosophy of the learning organisation, where individuals are encouraged to engage in continual development.
Point of integration
Person analysis is also an ongoing component of most performance review procedures, with managers
and employees discussing current and future training needs during appraisal, often using feedback
collected using 360-degree review (see Chapter 4).
Traditionally, responsibility for person-analysis lies with line managers who might
identify employee skills gaps as part of regular performance appraisals. This approach has
advantages in that, as line managers are likely to have frequent contact with employees,
they are most knowledgeable about their strengths and weaknesses; it also encourages
managers to set and support training objectives. However, fuelled in part by the growing
emphasis on employee-centred learning by employers, employees are, not surprisingly,
seeking more control over their learning (Karim and Behrend, 2014). This can create conflict if employees and managers disagree about the nature and importance of training
needs, or availability of training resources. For example, while an employee might believe
she needs to attend an in-depth leadership development programme, her manager might
prefer a less costly webinar.
To increase the effectiveness of a person-level TNA, managers must also be trained to determine whether an employee (and the organisation) will benefit most from improving the
employee’s existing skills – and therefore enhancing performance in their current role – or
from expanding the skillset to develop capacity for future roles. Managers must also be aware
of the suitability of individuals for different types of development activity, including assessing their personality and motivation for training to ensure that training is effective (Greco
et al., 2018; Woods et al., 2016).
208
Work Psychology
Key learning point
TNA identifies the objectives of training, and can occur at three levels: organisational, job or task, and
individual.
Stop to consider
In deciding who can and cannot receive training, managers hold significant power over employees’ development. This can be appropriate if managers behave professionally and fairly, but managers can be biased
against certain groups and/or demonstrate favouritism when choosing who to develop. For example,
research has found that line managers may block access to training for ethnic minority employees, which
may slow down their career progression and is likely to be a contributing factor to the low representation
of ethnic minorities at senior organisational levels (Wyatt and Silvester, 2015). As a work psychologist,
how would you ensure that line managers are fair in their allocation of training opportunities?
Training design
Training design (TD) is the second component of the training cycle. Salas et al. (2012) argue
that training design is the single most important element of an effective training programme,
because it involves asking questions about what methods will be used, how feedback will be
provided to trainees, and how the wider organisational context can support trainees learning
and put their learning into practice. In practical terms, TD needs to: (i) set the training aims
and objectives, (ii) determine an appropriate training strategy and (iii) plan and implement
the training.
Training aims and objectives
Training design (TD) builds on TNA by translating information about why the training is
needed into what a training programme needs to achieve and how. Training aims involve
general statements of intent such as: (i) ‘This programme sets out to provide participants with
“a grasp” of the basic principles of management accountancy’ and (ii) ‘The course aims to give
trainees an awareness of the relevance of industrial psychology to the management process.’
By contrast training objectives are much more specific and precise. They focus specifically
on what trainees should be able to do at the end of training and are crucial for determining
whether or not the training has been effective at a later point. They can be expressed in the
following three-component form:
1
The criterion behaviour is a statement of what the trainee should be able to do at the end of
training, often referred to as behavioural objectives. Importantly, they are not a description
of training content, for example ‘participants will gain experience of various personnel
selection interviewing techniques’ would not be acceptable as a behavioural objective,
because it describes what happens during the training – not what the trainee will know
or be able to do at the end. In this example, ‘participants will be able to ask job-related
questions in a structured selection interview scenario’ could be an appropriate objective
of the training.
Chapter 7 Training and development
209
2
The conditions under which the behaviour is to be exhibited. This might involve a specification of the equipment available to support the behaviour and the context within which
it will need to occur. For example, after training, a mechanic trained to service a new
computer-controlled engine will need the appropriate tools, the opportunity to practise
on new engines and support from their manager to trial new patterns of working.
3
The standard of performance of the behaviour. This is important, because it provides guidance to trainees, managers and those providing the training about what is expected and
provides a means for evaluating whether the training has achieved its objectives.
Key learning point
Training objectives give a clear description of what the trainee should be capable of doing at the end
of training.
Once the training aims and objectives have been identified, the next step in design is to decide
how employees will be trained. This means considering what methods are likely to achieve
the desired learning and performance in this particular context. Designing training can be
complicated. Not only are there many different training methods to choose between (e.g.
lecture-based, online and/or self-directed learning, role play and virtual simulations) and
multiple ways of presenting, sequencing and rehearsing information, trainees vary considerably in their ability and motivation to engage in learning, and practical matters like cost,
time, where trainees are located and the availability of qualified personnel, are important in
training design. Fortunately, work psychology research and theories provide an evidence base
to help us make these decisions.
Theories of learning
Understanding how people learn is a good place to start when thinking about training design,
because theories of learning and skill development can help to inform decisions about the
most appropriate methods of instruction. Learning theories represent ‘ideas’ about how
learning occurs – they are also an area where psychological research has been very influential. From a historical perspective, the most prominent approach to learning derives from
the behaviourist tradition, which proposes that learning results from strengthening stimulus
(S)-response (R) links, and reinforcing desired behaviour by giving trainees rewards and, in
some cases, ‘punishing’ inappropriate behaviour (though in practice this is rare in the training environment for a host of good reasons).
The behaviourist tradition has been important in the evolution of training and is often used
in development activities that employ ‘gamification’, for example, where trainee behaviour
is reinforced using points, badges or digital tokens to represent achievement (Landers, 2015).
However, not all trainees view these rewards as valuable, and often the behaviours that trainers
want to reward, like in-depth knowledge or meaningful participation, are difficult to observe.
A common criticism of the behaviourist approach is that it ignores the cognitive processes
involved in learning, including how the learner feels about what he or she is learning, how they
process and understand new concepts, and how this new knowledge influences the choices
they make.
Social learning theory
Social learning theory (SLT), which was developed by Bandura (1986) in response to criticisms
of behaviourism, allows a bigger role for internal mental processes in learning. These include
trainees’ expectancies about learning outcomes, and the capacity of individuals to learn
210
Work Psychology
without direct experience, for example by observing someone perform an activity. In terms
of employee training, SLT emphasises a need for trainees to engage actively in the learning
process and to think about the training experience rather than simply receive information
passively. One of the most important aspects of this theory is the emphasis on how experiencing success helps a trainee develop a belief in their ability to do the task. This is referred
to as self-efficacy and is discussed in more detail later in this chapter.
SLT has had an important impact on the design of training, and typically, a SLT-based
training intervention has three stages. First, trainee attention is focused by using a ‘model’ to
perform the target behaviour (e.g. demonstrating how to engage a customer effectively). Second, the trainee observes a sequence of behaviours where the model is rewarded for performing appropriate behaviour (i.e. interacting with customers in a desired manner) or punished
for inappropriate behaviour (i.e. responding to a customer in an undesirable way). Third,
trainee learning is strengthened by providing opportunities to rehearse and practise the target behaviour. The following example describes what might happen in a SLT-based training
intervention aimed at improving transfer of information in a cross-cultural team:
1
The trainer begins by explaining and emphasising the importance of communication
skills for working in cross-cultural teams.
2
Trainees then watch a video of a model effectively listening and facilitating communication between team members during a group meeting.
3
The trainer draws out key points about the model’s behaviour and encourages group discussion between trainees and trainer.
4
Trainees participate in a role-play exercise where they are required to perform a group task
and receive feedback post-exercise from the trainer and other group members.
5
Trainees return to work where they begin to practise their new behaviour with members
of their own teams.
6
The trainer and trainees meet two weeks after training to discuss their experiences and
how they dealt with them.
Thus, trainees are told why effective communication is important, they observe the required
skills, then practise and receive feedback. Finally, they attempt to generalise their learning to
other situations when they return to work. A large body of research has documented the effectiveness of SLT as a training method (e.g. Lee and Lee, 2018). Key to this is that SLT identifies
straightforward and relatively easy ways to improve training success. For example, we know
that learning improves if the role models used for training are friendly, helpful, high status,
and of the same race or gender as the trainees (Bandura, 1977). Similarly, role models who
are powerful and able to control desired resources are also more effective, which means that
how managers behave is often an important source of learning for their direct reports (e.g.
Koch and Binnewies, 2015). Modelled behaviour should also be vivid and memorable, sufficiently detailed, and structured sequentially to allow least difficult behaviours to be modelled
first, building to more complex behaviour. A good illustration of how principles of SLT have
been applied to training can be found in the popularity of commercial ‘off-the-shelf’ training
materials featuring films of well-known actors using humour to model how (and how not) to
perform activities like appraisal interviews.
Stop to consider
‘Presenteeism’ (people turning up for work when they are ill), and ‘leavism’ (workers using leave or
working beyond contracted hours to complete work) are being blamed for falls in workplace productivity and employee wellbeing (CIPD, 2018). How could SLT help to explain how informal learning within
organisations might be helping to create workplace cultures that foster such work patterns, and why
might it be important in determining interventions to reverse these trends?
Chapter 7 Training and development
211
Baldwin (1992) makes an interesting and important observation in relation to SLT and
the use of modelling in training. He argues that trainers providing trainees with examples of
how tasks should be performed, ignore the fact that trainees may already be proficient in performing a task – albeit in a different or less desirable way. For example, a barista working a coffee
shop may be extremely proficient at making a cappuccino, but not in the most cost-efficient
or standardised way expected by their employer. In this case the barista may need to unlearn
their old pattern of behaviour before he/she can learn the new way. Models that show both what
should and should not be done are therefore sometimes used in training (e.g. Bledow et al.,
2017). Moreover, because old learning can interfere with new learning, experienced trainees
can take longer to learn a task than novices. Taken together, these factors mean that, in order to
be motivated to learn, the barista first needs to be persuaded why his/her effort is worthwhile.
While SLT is extremely useful in the design of training, it is not useful for all training
because not all tasks can be modelled (Gist, 1989): for example, an aspiring author is unlikely
to learn very much by observing an experienced author writing a book – all they would see is
someone typing a sequence of words! For this reason, learning theories derived from cognitive
psychology have become important for training design, because these help training designers
understand the factors that influence how trainees learn cognitive skills, and those that can
impede learning. Cognitive theories have also helped identify how trainees can be encouraged to become more engaged in their learning by guiding their own training experiences.
Key learning point
Employees can learn new skills from observing and imitating the people around them
Self-regulated learning
As employees have been granted greater control over (and responsibility for) their development,
researchers have paid more attention to how employees self-regulate their learning. Self-regulation
refers to the cognitive and emotional strategies that trainees adopt during the learning process,
including the training goals they set, and the attention and effort they devote towards learning
(Sitzmann and Ely, 2011). For instance, when learning to play the piano, a student might set their
own training goal (e.g. to become a concert pianist, or to finish six lessons), decide how often and
for how long they practice, and remind themselves that they are only just a beginner when they
are not able to play Beethoven’s Moonlight sonata after just a handful of lessons.
Many researchers have identified learner self-efficacy – a trainee’s belief in their ability to
complete tasks – and cognitive ability, as particularly important in how trainees self-regulate
their learning (Panadero, 2017). Someone who believes they can influence training outcomes
and acquire skills quickly, is likely to set more challenging learning goals. Likewise, whether
a trainee is motivated to attain skills in order to develop him- or herself (a mastery orientation)
or is merely attempting to maintain job performance (a performance orientation), will differentially impact the training goals they set and how they strive to achieve them (Greco et al.,
2018). In practical terms, this means that whoever is designing training needs to have a good
understanding of the individual differences among their target group of trainees.
Research has identified several ways training can be designed to help facilitate self-regulated
learning, including interventions to develop ‘metacognitive’ strategies that encourage
workers to exert control over their own thinking (Flavell, 1979). These metacognitive strategies typically target three stages. First, a focus on planning seeks to influence the content of
trainees’ goals, including when, where and for how long they will focus on training. Second,
training design can encourage monitoring, so that trainees check the effort and attention they
devote towards achieving their goals. Third, trainees can be supported in evaluating their
learning. Strategies to enhance self-regulation are more likely to be successful if they also
212
Work Psychology
consider the wider learning context, such as organisational and managerial support for training (Sitzmann and Weinhardt, 2018). The following examples describe how training can be
adapted to improve trainee self-regulation.
■
The trainer begins by emphasising proficiency rather than capability when introducing
training objectives to encourage trainees to set mastery goals.
■
The trainer explains the link between training objectives and organisational strategy to
make training more meaningful and increase trainee motivation.
■
Managers are kept informed of training activities so that they can allocate development
time to staff, allowing trainees to prioritise training goals.
■
Trainees are encouraged to make a conscious effort to track their own progress towards
training goals.
■
Trainees are provided with flexible access to training resources, so they have control over
when and where to revise training content.
■
The trainer introduces information that highlights trainee knowledge gaps, to encourage
trainees to reflect on their training and to maintain effort to achieve goals.
■
Managers are encouraged to reward or discipline trainees to encourage them to persist in
attaining training goals.
■
The trainee and trainer meet during and after training to reflect on feedback from peers
and managers.
■
The trainee is asked to evaluate whether they have met training goals and how they will
continue their learning.
Although encouraging trainees to engage in self-regulation can be beneficial for their learning
(Sitzmann and Ely, 2010) it is important to consider when such interventions will be most
valuable. In his theory of skill acquisition, Anderson (1987) suggested that when learning
a new skill, learners first learn facts about what is involved in a task (i.e. declarative knowledge), for example by reading an operating manual for a bicycle. They then compile that
information into knowledge about how to accomplish a task (i.e. procedural knowledge),
such as learning how to balance on a bike and use its pedals. Finally, with practice, or tuning,
performance becomes efficient and automatic. The type of self-regulation activity required by
individuals is likely to vary across these stages: too much too soon may backfire. For instance,
trainees in the ‘declarative’ stage may benefit more from interventions that target exploratory self-regulation, where they are encouraged to review their knowledge gaps by reading
information, asking questions and experimenting with new skills, whereas interventions that
target metacognitive activities, such as encouraging trainees to reflect, adjust and refine their
learning, may be more beneficial in the ‘tuning’ stage (Hardy et al., 2018).
In practice, it may not always be possible to design training that fosters self-regulation in
every learner at every stage of their skill development. Self-regulation activities are likely to
be most effective when training encourages learner control and targets complex skills that
require continual learning, such as learning a new computer programming language. They
may be less relevant for training that is highly structured and closely guided by trainers, for
example when training workers on a production line.
However, using theories of learning to identify factors that can influence training experiences, and to identify the training methods most likely to bring about successful learning,
is important for creating effective training programmes that deliver the training objectives.
Key learning point
Employees may need support to regulate their feelings, thoughts and behaviours to improve training
outcomes.
Chapter 7 Training and development
213
Training methods
Training methods are structured activities that are designed to bring about desired changes
in trainee behaviour by facilitating acquisition of knowledge, skills and attitudes. There
are many different ways to present learning materials, involving different training media
(e.g. class-based, video or information handouts), instructional settings (e.g. lecture theatre, multi-media learning environment, or the learner’s home) and ways of sequencing of
material (e.g. learner-controlled pace, structured learning programmes). Training research
provides an evidence base to inform decisions about what methods are likely to be most effective with different trainees, and in different settings. For example, a meta-analysis of studies
investigating leadership training found that training programmes that were located within
organisations were more effective than those run off-site (Lacerenza et al., 2017).
Digital training
Possibly the biggest influence on training design in the last decade has been the use of technology to support learning. Industry reports suggest that over 80 per cent of organisations use
technology to train and develop staff and the online learning industry is forecast to be worth
$325 billion by 2025 (CIPD, 2016; McCue, 2018). Digital methods of training include e-learning
(or web-based learning) where learners access training via virtual learning environments (VLEs)
and websites at home or work, including webinars, video sharing and social networking sites,
wikis, blogs, virtual reality simulations, and m-learning, where learners use mobile devices such
as tablets and smartphones to access learning material, often via mobile applications or ‘apps’.
Digital training is popular with employers for several reasons: it can reduce travel and classroom costs, enable organisations to implement standardised training across a large number of
employees, and enhance learner control thereby reducing the need for hands-on instructors.
The accessibility of digital learning means employees can also learn skills in bite-sized chunks
(known as ‘micro-learning’), which reduces time away from other work activities (Cascio,
2018) and permits ‘just-in-time’ learning so that training has an immediate impact on productivity. For example, a salesperson can learn about a new product in the morning and sell
it successfully in the afternoon. Digital training can also transport trainees to contexts that
are otherwise hard to replicate: an example is the supermarket chain Wal-Mart where virtual
reality has been used to train new recruits in how to manage aggressive ‘Black Friday’ sales
crowds (Fink, 2017). Augmented reality can also allow users to see the real world but with
digital information overlaid, so that trainees can be guided through procedures in real-time,
such as repairing complex machinery (Limbu et al., 2018). Trainees can therefore explore and
experiment with new skills, when mistakes in the real world would be costly in financial or
human terms (e.g. high-risk scenarios such as police hostage rescue, and astronaut training).
Although research comparing the effectiveness of digital and face-to-face learning shows
that both methods can be equally effective (Sitzmann et al., 2006), those designing training
should beware the temptation to simply digitise traditional training content; careful consideration needs to be given to the type of information that can be digitised and how it is best
presented. Poorly designed interfaces that are difficult to use, or suffer technical issues, can
impair learning (Sitzmann et al., 2010). Likewise, digital training can be expensive to develop
and implement well, and smaller organisations in particular may have less access to the technical expertise needed to design bespoke digital training platforms or to keep content up-to-date.
Key learning point
Designing digital training should not simply be a case of digitising traditional training content.
214
Work Psychology
Keeping employees engaged in digital training can also be challenging. For example, a
study of nearly 4000 people signed up to a MOOC (Massive Online Open Course) found that
only 30 per cent actually completed it (Greene et al., 2015). Drop-out rates for online resources
are likely to be high if employees do not perceive them important for their day-to-day role.
This means that, like traditional training, digital learning needs to be carefully aligned with
organisational strategy and goals, and managers need to communicate its importance and
relevance. Trainees can also be easily distracted if training is provided via the same mobile
devices used for other activities like work email and notifications. Trainees are likely to need
encouragement and incentives to plan and self-regulate their learning, for example, by concentrating and monitoring their learning progress (Sitzmann and Johnson, 2012).
Digital training can also be an isolating experience for trainees (Johnson et al., 2008). One
method of enhancing ‘social presence’ has been to encourage virtual communities of practice
via social media or chat-based platforms, which allow trainees with different levels of experience to interact with each other and share knowledge. ‘Stack Overflow’ is an example of an
online community that allows computer developers to ask questions and share programming
knowledge. However, there is always a risk that in these unstructured learning environments
the information shared by fellow trainees may be inaccurate or misleading.
‘Blended learning’ which involves the combined use of digital and traditional face-to-face
formats, is an approach that is gaining traction in educational contexts. It is thought to overcome potential disadvantages associated with using each method of training in isolation, possibly because it provides confident, independent learners opportunity for more self-directed
learning, while less confident learners benefit from face-to-face learning and more opportunity of support and feedback (Noe et al., 2014).
Game-based training
Technology based training has also seen growing interest in ‘serious games’ and gamification.
Serious games are not just for entertainment, but usually involve immersive simulations that mirror the technical and social interactions found in real-life work situations. The French military,
for example, use 3D-virtual environments to train soldiers to care for casualties when under-fire
(Planchon et al., 2018). Gamification, in contrast, involves the inclusion of game-based elements
into training design to make learning more engaging (Deterding et al., 2011).
Gamification can be applied to training in various ways: common methods include the
use of badges or points to reward achievement, adjusting the level of difficulty of activities
to match trainee skill level, offering surprise tasks or rewards as part of the training, and
using narratives or storylines to enrich content (Bedwell et al., 2012). For example, trainers
aiming to liven up health and safety training might design a mobile game app where trainees
role-play a warrior guiding citizens out of danger as they travel through a fantasy kingdom
by identifying and solving health and safety violations, and earning points and rewards for
each solved successfully. However, it’s worth noting that not all gamified training requires
technology: a leader-board comparing trainees’ performance written on a flip-chart would
also constitute an element of gaming design!
Research has found that gamification has a positive impact on learning because it is interesting and enjoyable, offers immediate feedback, and motivates trainees (Hamari et al., 2014).
Like digitalisation, however, gamification is not a panacea for poorly performing training
programmes. Training may fail, not because trainee motivation is low, but because the learning content is inaccurate or there is a lack of organisational support (Armstrong and Landers,
2018). There is also evidence that gamification is more successful for trainees who are familiar
with technology, are more competitive and perform well. Indeed, those at the bottom of a
leader-board or who fail to collect badges or points may well become demoralised, demotivated and therefore less likely to learn (Cardador et al., 2017). It is therefore important to
consider different methods of gamification, and the reasons for using them, in order to avoid
the charge that it is simply an expensive novelty (Landers, 2015).
Chapter 7 Training and development
215
Key learning point
Adding game elements can make training more interesting and enjoyable but it can also be a distraction if not used appropriately.
Team-based training
Teams (i.e., individuals working together to produce collective workplace goals (see Chapter 9))
present specific challenges for training design, because team members are likely to play different roles, have different professional backgrounds and level of work experience, and perform
tasks that are interdependent. This means that team training needs to focus on developing an
individual’s knowledge skills and abilities in relation to the task they are set, how they coordinate their actions and communicate with others, and who they need to go to for specific
information or expertise (Salas et al., 2018). To complicate things further, teams are usually
in a state of flux, changing over time as members join and leave, and increasingly likely to be
distributed across different locations, culturally diverse and reliant on communicating virtually (Lacerenza et al., 2017). Thus, to be effective, team training design must accommodate
longitudinal and dynamic perspectives.
Point of integration
Training design and training outcomes need to consider key features of teamwork, as this is the context
in which many employees carry out work tasks.
Research on team training has evolved significantly in recent years, though most studies
still tend to focus on teams involved in high-risk work (e.g. aviation, medicine and even space
exploration – see closing case study in Chapter 4.) where poor performance can have a critical
impact on the safety and well-being of others (Hughes et al., 2016; Littlepage et al., 2016; Salas
et al., 2015). In their meta-analysis, McEwan et al. (2017) found that team training improves
team processes (e.g., communication and trust between teammates) and team performance
across a variety of work settings, tasks and team types. However, team training interventions
were most effective when targeted not just at improving execution of a task, but at regulation
strategies across the whole team, such as setting goals for team outcomes, and monitoring
and reflecting on progress towards team goals. The researchers also found that training was
more useful where participants were encouraged to learn and practise teamwork through a
variety of experiences.
Team training designs often include strategies to encourage team coordination and adaptation (e.g. Henrickson Parker et al., 2018). For example, team members may be ‘cross-trained’, or
trained in the tasks of their teammates, in order to improve their knowledge of team processes
and the team’s ability to adapt to changing work demands or unexpected events. Team members may also be taught how to discuss and anticipate potential changes in their situation. For
example, in simulation or event-based training, medical teams are often trained using high-tech
patient simulators that trainers program to recognise drugs, excrete bodily fluids and demonstrate varying cardiac rhythms and blood pressure: thus, by replicating often highly challenging
and fluctuating requirements of medical emergencies, teams learn how to work together efficiently while coping with stressful and changing demands (e.g. Härgestam et al., 2016).
A critical aspect of effective team training is psychological safety, where trainees feel
sufficiently safe to express their concerns, share ideas, ask questions, and make mistakes
216
Work Psychology
without fear of negative consequences (Lacerenza et al., 2015). Consequently, team training
requires skilled facilitators who can create a positive climate for sharing ideas, and coach team
members to discuss their progress, model effective feedback skills and provide constructive
feedback on their own and others’ development (Smith-Jentsch et al., 1998).
Exercise 7.2
Training for teamworking
National football teams often struggle to come together as a team because they comprise members of competing domestic football teams. They therefore need time and support to learn to
work together effectively. How would you design training for a newly formed national football
team whose members do not know each other or who have existing rivalries? What content might
you cover and what training methods would you use? Are there any practised behaviours (such as
those specific to their domestic teams) they might need to unlearn in order to work as effectively
as possible in a team?
Transfer of learning
So far, we have focused on the content of training and how it can be delivered. Equally important, however is the need to consider how learning from the training environment can be
transferred back to the work context. If transfer of learning is not considered, even the best
designed training can ultimately fail to change trainee behaviour. Baldwin and Ford’s (1988)
original and influential model of learning transfer differentiates between three areas: (i) learning and retention – what the trainee learns and remembers from the training programme; (ii)
generalisation of learning – the extent to which trainees can adapt what they have learned and
apply it to the specific needs of their own workplace; and (iii) maintenance – trainees continuing to use and apply this learning over time. The model also provides some useful explanations of why individual, training and environmental factors may have different effects on
trainee learning, and how learning transferred to the workplace eventually translates into
improved performance.
Personality is an important individual-level (i.e. trainee) factor that might affect transfer. For example, individuals who are conscientious and proactive are more likely to take
the initiative to apply new skills (Roberts et al., 2018). As discussed earlier in the chapter,
self-efficacy and mastery orientation are also important individual-level factors: trainees who
are confident in their ability to learn and engage in training for self-improvement are more
likely to regulate the transfer of their learning and persist in the face of setbacks (Grossman
and Salas, 2011; Huang et al., 2016). A meta-analysis of 89 studies (Blume et al., 2010) found
that self-efficacy and motivation, and environmental context were particularly important
in training to develop open skills (i.e. where trainees have a high degree of choice over how
they use the skills learned in training in their job), but, these factors were less important for
transfer of closed skills (i.e. newly acquired skills involving more prescribed behaviour and
less employee choice over how they can be used in work). Transfer of open skills can also be
enhanced by encouraging trainees to form implementation intentions, or plans about when
they will practise and apply their KSAs (Friedman and Ronen, 2015).
Manager support is considered an organisational-level factor that can impact on transfer of
learning. It might involve making sure that trainees have adequate resources to practise their
skills when they return from a training programme, providing reminders about what has been
learned, plus frequent feedback and rewards for putting the new skills into practice (Sitzmann
and Weinhardt, 2018). However, managers may themselves need training in order to understand the importance of their role in facilitating others’ development. This is one reason why
Chapter 7 Training and development
217
training programmes are often cascaded down organisations, with managers trained first to
appreciate the goals and relevance of the training, and their responsibility for modelling new
behaviours and providing support for new learners (Ghosh et al., 2015).
In their Dynamic Transfer Model, Blume et al. (2017) describe transfer as a cycle: trainees
start with an attempt to transfer new skills to the workplace, they then collect feedback about
how well the new skills have transferred, and subsequently use that feedback to guide further
transfer attempts. Importantly this model recognises that the cycle can create positive outcomes (i.e. where trainees persist in trying to apply their training) or negative spirals where a
trainee may eventually decide to dismiss or discard their learning. As such, it highlights the
importance of support and feedback immediately after training in order to ensure trainees
maintain their motivation to persist in attempts to transfer their new KSAs. The model also
recognises that successful transfer may not be immediate; often requiring time and practice
to successfully apply learning in the workplace.
Key learning point
For training to be successful, employees must transfer their learning to the job; changing their behaviour by using their new KSAs and maintaining this change over time. Managers play an important role
in creating a supportive work climate for this transfer to take place – but some trainee characteristics
are also important.
Training evaluation
The third component of the training cycle is training evaluation (TE). This is concerned with
establishing whether or not the training has worked (i.e. that the learning objectives have
been achieved), and that the organisation’s investment in training has been worthwhile. TE
also collects information to explain why the training has or has not achieved its intended
outcomes in order to show how future training can be improved. It is striking that many
companies do not evaluate their training programmes (Salas et al., 2012). Indeed, surveys
suggest that just 20 per cent of organisations assess whether learning has transferred to the
workplace, and only 7 per cent examine the wider impact of training on the organisation
(CIPD, 2015). One reason for these low figures is that TE can be difficult and expensive to
undertake. Practical challenges include measuring behaviour change post-training: trainees
from a company may undertake training at different times, and then disperse to many different locations, making it difficult to capture data in sufficient quantity and quality to link
trainee performance to organisational performance.
Models of training evaluation
Probably the most widely used model of training evaluation was developed by Kirkpatrick
(1967) and involves four levels of data collection: (i) reaction, (ii) learning, (iii) behaviour and
(iv) results (see Figure 7.3).
The first and minimum level of evaluation involves collecting reaction data, which relates
to trainees’ views about the training: it is generally assumed that trainees need to have a positive reaction to the training experience for training to be effective. Collecting reaction-level
data typically involves asking trainees to complete a post-training evaluation questionnaire
with questions about the perceived quality and enjoyability of the training (you may have
218
Work Psychology
Level
Figure 7.3
Method of assessment
1 Reaction
Post-training feedback sheets (e.g. ‘did you enjoy the
training?’; ‘was the training useful?’)
2 Learning
Tests of knowledge acquired (e.g. a test of knowledge
of the procedures for dealing with customer complaints)
3 Behaviour
Observation of skill performance and job behaviour,
tests of speed and accuracy (e.g. observing an
employee dealing with a customer complaint)
4 Results
Assessment of team performance, levels of absenteeism,
overall organisational effectiveness, productivity and
profitability (e.g. customer satisfaction with the way the
employee dealt with a variety of complaints)
Kirkpatrick’s model of training evaluation
come across something similar in the form of course feedback collected in universities to
monitor the quality of teaching).
Level two involves collecting learning data and assumes that effective training depends
on trainees acquiring the desired KSAs. Learning data is collected to determine whether trainees have attained the objectives of the programme, and might involve asking trainees to
complete pre- and post-programme written tests of knowledge, or testing how accurately or
quickly they can perform new skills. This allows the trainer to check participants’ understanding of material covered during training. Learning data are typically collected at the end or
immediately after training. However, testing whether a trainee has retained material over a
longer period is also important (e.g. six months post-training). Just like reaction data, learning data are useful but incomplete, because although trainees may have learned the relevant
knowledge or skills there is still a question of whether they will use them in their roles.
Therefore, Kirkpatrick’s third level of evaluation concerns changes in behaviour; specifically, the extent to which trainees demonstrate and apply skills they have learned during
training once they return to work. Behavioural data are often captured using managers’ assessments of a trainee’s performance once they return to work: as such they can be vulnerable to
distortion and bias. For example, ratings can be influenced by a manager’s expectation that
training will be effective or ineffective (e.g. ratings might be higher if the training was the
manager’s idea, or lower if they think the training is a waste of time). It is possible to collect
more objective assessments of behaviour change, such as a reduction in errors, or improved
quality of work by a trainee. However, as Kirkpatrick’s third level concerns behaviour change,
data are more difficult and time consuming to collect, because this needs to happen before
and after the training. Another challenge is that as initial performance often declines when
a trainee begins to practise a new skill, post-training behaviour should ideally be assessed at
several time points (Huang et al., 2016). Consequently, it is just as important to decide when
to collect behavioural data for training evaluation as it is to decide what type of data to collect.
The fourth and final level is described as ‘results’ and concerns the extent to which the
training has had an impact on organisational effectiveness. Although conceptually clear,
results criteria for this level are extremely difficult to assess in a controlled fashion. For example, it can be almost impossible to determine with certainty that improved organisational
effectiveness – even at a team or functional level – has been brought about by the training
rather than a combination of other factors like a general improvement in the economy or
Chapter 7 Training and development
219
the company’s launch of a successful new product. In practice, results-level data are the most
difficult and expensive to collect and interpret (Mabey and Ramirez, 2005).
Key debate
It is popular, but is Kirkpatrick’s model a good one?
Criticisms of Kirkpatrick’s model have focused on two aspects: (i) methods used to collect and assess data at each
level, and (ii) the hierarchical assumptions of the model where success at one particular level is viewed as dependent
on successful achievement of each previous level (Alliger and Janak, 1989; Alliger et al., 1997). For example, where
successful learning (level 2) is seen as dependent on an individual enjoying the course (level 1) or where behaviour
change in the workplace (level 3) is seen as reliant on successful acquisition of knowledge (level 2).
Many training interventions rely on level one reactance data to evaluate training. While trainees’ views are important, and post-training questionnaires provide a relatively easy and inexpensive source of information, this type of data
can give a misleading impression of the value of a training programme and an incomplete view of training effectiveness.
For example, spicing up dull training content with jokes, videos or memes is likely to make trainees more enthusiastic
about a training programme. Trainees may also provide poor reports for training when their instructor is less entertaining, or they need to work harder than expected. But none of this feedback tells us whether the training was effective at
producing new learning (Sitzmann and Johnson, 2014). While reaction-level feedback is often referred to dismissively
as ‘happy sheets’, studies show that there are relatively simple ways in which it can be substantially improved (Weber,
2015). For example, Warr and colleagues (1999) found that including questions about how difficult and how useful
trainees found the course increased the correlation between reactance-level data and subsequent learning.
More recently, Sitzmann and Weinhardt (2018) have argued that Kirkpatrick’s model fails
to account for the dynamic nature of learning over time or how similar measures of effectiveness can be assessed at different levels (see Table 7.1). To address such challenges, they
propose four types of indicators for training evaluation: utilisation indicators (e.g., how many
people enrol and leave the training), affect indicators (e.g., self-efficacy, motivation and reputation of the training programme), performance indicators (e.g., learning and behaviour
change) and financial impact (e.g., return on investment). Importantly, they argue that these
four indicators need to be considered at the within-person, between-person and macro level.
For example, evaluations focusing on performance might examine how individual trainees
develop their learning over time (within-person), how their learning compares to those who
did not receive training (between-person), while also looking at the organisation’s human
capital or collective capability – the knowledge and skills of all employees that provides the
organisation with competitive advantage (macro-level). This multilevel training evaluation
framework therefore has the advantage of allowing investigation of the causes and outcomes
of learning in much greater detail than Kirkpatrick’s model.
Clearly, training evaluation can be complex and, in general, no single method is able to
provide detailed and comprehensive information about all aspects of a training programme’s
effectiveness. The most successful and useful evaluation requires a multi-method approach
involving different types of data to build up an overview of whether the training has achieved
its objectives and how it might be improved.
In today’s dynamic and complex work environment a key challenge for those evaluating
training is how information can be collected quickly and accurately. This has driven a growth
in rapid evaluation strategies that aim to evaluate training in real-time by asking trainees for
immediate feedback during and after training using methods such as interviews, focus groups
or questionnaires (McNall and Foster-Fishman, 2007). Not surprisingly, technology is also
becoming more common; for example, audience response systems ask trainees to answer
questions or vote in polls using mobile devices, providing immediate feedback to trainers and
Work Psychology
220
Table 7.1
Multilevel framework of training effectiveness, Sitzmann and Weinhardt (2018)
Training utilisation
Affective indicators
Performance
indicators
Financial impact of
training
Macro level
Numbers of employees
who enrol and leave
training
Training reputation
Organisational /team
performance
Return on investment
Between-person level
(stable comparison across
different employees)
Which employees enrol,
complete/drop out
Satisfaction,
self-efficacy,
motivation
Learning, training
transfer
Personal return on
investment
Within-person level
(responses that are
dynamic and change over
time)
Decision to enrol
Satisfaction,
self-efficacy,
motivation
Learning, training
transfer
allowing them to adapt the training in real time. The large amounts of data held by organisations about their employees, known as ‘Big data’, can also help to identify training needs,
monitor and evaluate learning and performance over time. Predictive analytics, for example,
can be used to identify which type of training is likely to be most successful and tailor training
provision accordingly before training has even begun.
Exercise 7.3
It can sometimes be important to evaluate the delivery of training separately from the content. For
example, to ensure that game elements are not a distracting novelty in a training programme they
need to be examined in their own right. Similarly, training effectiveness overall might be influenced
by trainees’ familiarity with the technology or their attitudes towards gaming rather than their ability
to learn (Landers and Armstrong, 2017).
Imagine you are evaluating a gamified training programme for car salespeople. The training involves
employees assuming an avatar and walking through a 3D virtual car showroom where they encounter
customers who provide them various challenges, such as difficult demands and surprising reactions.
Employees are rewarded with points when they communicate successfully with customers.
How might you evaluate this training to determine whether gamification added value to the training
design?
Designs for training evaluation
To ensure that the conclusions made about the effectiveness of training are valid, it is important
that the design of training evaluation draws on traditional psychological research methods.
Threats to drawing valid conclusions are usually referred to as internal or external threats to
validity. Threats to internal validity concern factors that make it appear like training is responsible for changes in learning, behaviour or results, when the changes are due to other causes.
For example, while a pre- and post-training test shows that trainee knowledge of a new piece of
software has increased over a six-week training course, it is possible that this is due to time spent
Chapter 7 Training and development
221
using the software rather than the content of the training course. Threats to external validity
relate to whether evaluation results generalise to other groups of trainees and situations. The
same software training could be more effective for new recruits than for existing employees,
who can have an additional need to unlearn knowledge and skills relating to ‘old’ software.
To control for various threats to the validity of training evaluation, investigators can make
use of experimental designs. Although as most training evaluation takes place in real organisational settings it is often not possible to obtain conditions necessary for perfect experimental designs. In such circumstances it is common for quasi-experimental or pre-experimental
designs (Campbell and Stanley, 1963) to be used (see Chapter 1 and box below).
Research methods in focus
Pre-experimental, experimental and quasi-experimental designs
Two of the best-known training pre-experimental training evaluation designs are shown in Figure 7.4, together with
more complex designs that overcome some of the problems inherent in the pre-experimental designs.
Clearly, the one-group, post-test-only design controls for none of the possible threats to internal or external
validity and it is not possible to interpret the data. It is impossible to tell whether scores are better after training
than before – let alone whether training or some other factor is responsible for any changes. The one-group,
pre–post measure design goes some way towards resolving the problems by making it possible to measure
change over time. Nevertheless, this design cannot assess whether training has caused any difference.
Establishing whether the training has made a difference may only be done if there is also an untrained control
group, the members of which are similar to the trained group and whose performance has also been measured
at the appropriate times. To conduct a true experiment, trainees should be assigned to the experimental and
control groups on a random basis since systematic differences between groups before the experiment could
bias the results. Often this degree of control is impossible in field research and the kind of quasi-experimental
design shown in the non-equivalent control group example of Figure 7.4 is the best that can be done. Typically,
pre-existing groups in the organisation, such as all the members of a particular job group, region or unit, form
the groups. This is administratively much more convenient than random assignment and does control for some of
the main threats to validity. Even the pre–post measure control group design is subject to some threats to validity,
and for totally unambiguous results more complex designs are needed (see Campbell and Stanley, 1963).
A difficulty with pre- and post-test designs is that they might fail to capture information about how training
transfers over a longer time period (Huang et al., 2016). Recent advances in statistical modelling, such as
latent growth curve analysis and multilevel modelling have provided sophisticated methods of tracking trainee
development over multiple time-points. These techniques can also be used to assess the impact of training on
several dependent variables at the same time and across different levels of the organisation (e.g. individual
learning, team performance and return on investment). For example, Kim and Ployhart (2014) used growth
models to examine the performance of 359 firms over 12 years. They found after controlling for firm size and
industry type, firms who used more internal training were more profitable than competitors before the 2007–
2009 recession and recovered more quickly after the recession.
Key learning point
Pre-experimental designs, such as the one-group post-test-only design, are common in the training
world but they are not capable of providing clear, unequivocal results about training effectiveness.
Practical constraints often make it difficult, or impossible, to execute the most complex evaluation
study designs that have high levels of external and internal validity.
Work Psychology
222
One group, post-measure only
X
M2
One group, pre-measure/post-measure
M1
X
M2
R
M1
X
R
M1
Pre-measure/post-measure control group
M2
M2
Non-equivalent control group
M1
X
M2
M1
M1a
M2
M1b
M1c
M1d
X
M2a
M2b
M2c
M2d
Key:
M1 = pre-measure (administered prior to training)
M2 = post-measure (administered after training)
X = training programme
R = random assignment of people to groups
Figure 7.4
Pre-experimental, experimental and quasi-experimental designs for training evaluation studies
Training and development in practice
In this section we consider how training and development activities have been used to address
three specific business needs including (i) leader development, (ii) improving diversity and
inclusion and (iii) creating organisational learning cultures.
Training and development for leaders
Leadership development is an important area for many organisations. A huge amount of
money is spent annually on efforts both to develop potential leaders and enhance the skills
of incumbent leaders. Indeed, leadership development is now often treated as a reputational
benchmark for companies competing to hire the most talented people. Harvard Business Corporate Learning publishes a biennial review of leadership development. In their 2018 review
they found that companies who built leadership development into their strategy reported
being more successful in their ability to transform in the face of change, had a stronger market position than their competitors, felt it was easier to promote leaders within their own
companies and had leaders who felt they excelled in the role.
In a meta-analysis of 335 training studies, Lacerenza et al. (2017) found that leadership
training is effective in improving trainee reactions, learning transfer and organisational and
subordinate outcomes. Among a range of factors, successful programmes were more likely
Chapter 7 Training and development
223
to incorporate opportunities for feedback, be delivered on-site and face-to-face rather than
off-site or virtually and spaced weekly rather than delivered over shorter intervals. They also
found that practice-based training methods, such as role-play or simulations were more
valuable than lectures, presentations or demonstrations, although a combination of these
methods was most effective. Leadership training that targeted business skills, such as problem solving or budget monitoring had most effect on learning and transfer, but training that
targeted softer skills, such as interpersonal and leadership styles, was more important for
organisational outcomes, such as return on investment and absenteeism as well as subordinate outcomes, including subordinate job satisfaction. This supports the growing evidence
that interpersonal and political skills, such as influence, negotiation and networking are vital
for leadership development (Ferris et al., 2017).
Point of integration
There is a longstanding debate about whether leaders are ‘born or made’ (see Chapter 10), but training
and development research indicates that people can acquire some of the important knowledge, skills
and other qualities that impact on leader behaviours.
Senior leaders may be reluctant to engage in traditional training activities for fear of stigma
– often because taking part in training implies a development need, which can be construed
or revealed as a weakness to others (Silvester and Wyatt, 2015). For this reason, much of the
development activity of senior personnel is likely to involve one-to-one bespoke learning
activities such as coaching: this has the appeal of occurring in private ‘behind closed doors’
where potential development needs can be discussed without fear of prejudice. The coaching
provided to organisational leaders often involves an external coach who works one-to-one
with the leader to identify and help address a specific development need, frequently in relation to leadership ability and style. These types of development activities are extremely difficult to observe and learning objectives (if they exist) are much less likely to be formalised
or measured compared with traditional training activities. This means that although there is
evidence that coaching can be worthwhile for leaders, less is known about how and why it is
beneficial (Athanasopoulou and Dopson, 2018).
Despite the effectiveness of formal development initiatives, leadership is often thought to
be developed through experience, such as hardships, mistakes and successes over the course
of a career (Hammond et al., 2017). Employees may also rely on informal relationships to
facilitate their leadership development, such as mentoring. Mentoring is particularly good
for providing leaders with an understanding of the social norms, cultural expectations and
insight into the political side of organisations (Blass and Ferris, 2007). For example, this may
include information about who knows what and who has the power to make decisions within
the organisation. However, employees from minority ethnic backgrounds may find it especially difficult to access this type of political information because they are outside privileged
powerful groups (Wyatt and Silvester, 2015). Organisations may therefore need to provide
support for employees in developing relationships with powerful role models and mentors
in order to improve the number of minority ethnic employees reaching leadership positions.
Key learning point
Leaders’ development often takes place outside of the formal training environment, for example
through one-to-one coaching.
224
Work Psychology
Diversity training
Globalisation and an increase of women, minority ethnic and older workers in the workforce has meant that organisations are increasingly turning to diversity training to enhance
employee awareness, and develop attitudes and skills to improve how individuals with different backgrounds work together. Implicit or unconscious bias training has proven particularly
popular. For example, responding to a widely publicised incident in 2018 where two black
men were arrested for merely sitting in a Starbucks outlet without ordering, the coffee chain
closed 8,000 stores in the US to ensure all staff attended mandatory unconscious bias training.
This type of training typically involves taking a test, commonly the Implicit Association Test
(IAT: Project Implicit, 2018), which measures reaction times when participants are asked to
match stimuli, such as a black or a white face, to attributes, such as the words ‘good’ or ‘bad’.
Participants are then debriefed and educated on unconscious bias theory and ways to reduce
biased behaviour are suggested, such as removing names from CVs and actively questioning
biases during selection procedures (Atewologun et al., 2018).
There is evidence that unconscious bias training can increase awareness of bias and reduce the
strength of bias measured by post-training IATs (Forscher et al., 2018), but it suffers from a ‘transfer
problem’ because changes are often short-lived, sometimes lasting just a matter of days. Importantly, there is also limited evidence that reducing the strength of bias also leads to a reduction in
discriminatory workplace behaviour (Dobbin and Kalev, 2018). While some scholars have gone
so far to suggest that unconscious bias training is ‘pointless’ (Noon, 2018), in practice, it is difficult
to evaluate the effectiveness of bias training because the training objectives are often very broad
(i.e. ‘to increase awareness of bias’). Likewise, as discrimination is often hidden and difficult to
observe, participants are unlikely to fully report discriminatory incidents pre-and post-training.
Key learning point
Widespread adoption of diversity training demonstrates that political concerns can trump validity concerns when it comes to companies needing to present as ethical. In this case through commissioning
company-wide employee training, despite comparatively little evidence of its impact on behaviour change.
To improve diversity training, researchers have advocated moving away from mandatory
programmes that may result in backlash from participants if they feel they are being indirectly
blamed for inequality (Dobbin and Kalev, 2016). However, while voluntary programmes can
be effective because the participants are already motivated to address bias, relying on volunteers may mean that training bypasses those who may be most discriminatory and unwilling
to change. Consequently, decisions about who should be trained ought to be based upon a
TNA (Kulik et al., 2007). Moreover, diversity training that is implemented over time, develops
skills (e.g., verbal or non-verbal communication), provides opportunities for practice, like
role plays and simulations, and offers participants feedback, is most likely to have a positive
impact on diversity in practice (Bezrukova et al., 2016; King et al., 2010). Importantly, as with
any development initiative, to be most effective, diversity training should be supported by
an organisational strategy, with action plans, diversity committees and devoted staff roles,
in order to ensure commitment and accountability (Dobbin and Kalev, 2015).
Training and the learning organisation
Although discussion in this chapter has focused on training as an essential activity in its
own right, it is only proper that we end by discussing training as part of the wider organisational system of strategic human resource management, with relevance for concepts such
Chapter 7 Training and development
225
as organisational learning (OL) and knowledge management (KM). Generally speaking,
OL and KM are discussed in fields like organisational behaviour and change, rather than
psychology because they examine the wider complex and dynamic learning systems that
characterise successful organisations, rather than focus on the specifics of training design
and evaluation.
Point of integration
Organisational change (see Chapter 12) can often be a stimulus for learning. Collaborative change
processes often involve different groups of employees working together to plan change or to solve
problems.
The past few decades has seen an emerging tension in how organisations approach training and development. On one hand, organisations increasingly recognise training as a
core business function, and the need to become ‘learning organisations’ so that they can
adapt to the change and instability of global markets and remain competitive by attracting and retaining talented staff. Central to the concept of the learning organisation is
that employees are encouraged to take up development opportunities and view learning
as an ongoing feature of their job (Easterby-Smith et al.,1999). Therefore, organisations
promote learning and development and encourage a positive attitude towards training
opportunities.
On the other hand, organisations are under increasing pressure to cut costs and training
is often regarded as an ‘unaffordable luxury’ where expenditure can be reduced (Roche et al.,
2011: 47: Sheehan, 2014). Lower investment in training, combined with advances in technology, such as m-learning, has meant organisations increasingly shift responsibility for learning
and training onto individual employees, who may have to develop knowledge and skills in
their own time and with their own resources. This perspective puts learning organisations at
risk because individual employees hold the power about whether they pass on and share the
knowledge they have developed.
As a result, there has been increasing focus on how learning and development systems
can be ‘turned on their head’ to reduce costs and also ensure that unstructured or informal
employee learning is captured, shared and dispersed so that it can benefit organisations (Millar et al., 2017). For example, many organisations encourage employees to form internal communities of practice so that information is shared across people in similar roles (Randhawa
et al., 2017). Likewise, organisations may also fund individual employees to attend training
and then ask them in exchange to become the trainer by sharing their newly acquired knowledge with their team upon their return.
This illustrates a substantial shift in our thinking about learning and development activities. There has been a move away from seeing training as a top-down process where the
organisation identifies the knowledge and skills required by an employee and controls access
to them through formal, structured learning opportunities. Instead, there has been growing
interest in knowledge as a social construction, co-created by individuals through their interactions with one another (or between mentor and protégé as we saw earlier) and the environment. Informal learning, driven by the interests and proactive engagement of an employee
is recognised as an ongoing feature of working life. What this type of learning emphasises,
however, is the inherently political nature of learning and development. Employees do not
passively acquire learning, but exchange, barter and gift knowledge as a commodity in order
to gain power and influence with colleagues and their employer (Konstantinou and Fincham,
2011). Although relatively unexplored by work psychologists, these political aspects of learning and development undoubtedly provide a fertile and fascinating area for future research
(Silvester, 2008).
226
Work Psychology
Summary
Constant social and technological change provides a context for organisational life. This
coupled with individual growth and career development means that training has an important, if not core, role to play in organisations. Effective training needs analysis (at organisation, task and person levels) provides the basis for the design and implementation of training
activities. Although the analysis of needs, together with a clear statement of training aims
and objectives, is important, there is still a certain degree of judgement involved in choosing
appropriate training methods. This essential subjectivity can be checked and assessed by the
application of systematic procedures for evaluating the effectiveness of training. Evaluation
at reaction, learning, behaviour and results levels provides a way of determining the overall
value of training and assessing necessary improvements. In general, research has shown that
training is an effective way of bringing about behaviour change, although there are often
problems in ensuring that the potential for change provided by training activities actually
transfers to the work setting. In fact, organisations today are far more likely to see training as
a continuous process of organisational learning that incorporates formal, structured training activities as well as informal learning opportunities, where employees are expected and
encouraged to share their knowledge and learning with each other.
Closing case study
Can you train a politician?
Politicians have some of the most important roles on the planet – they make decisions that affect
the lives of billions of people, control resources that impact business, wealth and environments
globally, and create laws that govern how societies act. Yet, politicians rarely receive any training to
perform their roles. Unlike most other work roles, preparation for, and support in, political office relies
overwhelmingly on informal socialisation and knowledge sharing between peers. In the case of newly
elected British Members of Parliament, this usually means a two-day formal induction programme
to introduce them to the Palace of Westminster (home to the Houses of Parliament), and rules and
procedures for claiming expenses.
Although the House of Commons recently introduced a buddying programme for new MPs, where they
are paired temporarily with members of House Staff, MPs must still rely on their own efforts to learn how
to navigate their new work environment, and understand and decide how to perform their new roles.
This reliance on informal, self-directed learning has been criticised for failing to provide new MPs
with clarity about their roles and how to perform them. Likewise, new MPs often complain that the
Clerks, whose role is to ensure procedure is followed in political proceedings such as the debating
chamber and House committees, are reluctant to help them learn. Likewise, new MPs complain that
more experienced colleagues can be slow to help and reluctant to share information.
Questions
■
What training do you think a new MP is likely to need in order to perform their role?
■
Why do you think there is so little formal training for politicians?
■
There are similarities between new MPs and new CEOs, both of whom often receive little formal training when they take on their role. What can this tell us about the prevalence of training at different
organisational levels?
Chapter 7 Training and development
227
Individual/group discussion points
1 How does an increase in learner control pose challenges for learning organisations?
■ Why are employees wanting more control over their learning?
■ How can organisations encourage employees to develop the skills organisations
need?
2 Why is training needs analysis important?
■ What are the levels at which TNA can occur?
■ Why is it important to link TNA to selection and performance management
systems?
■ How does a TNA impact other phases in the training cycle?
3 Why is it important that learners use self-regulation strategies?
■ How might managers and trainers facilitate self-regulation among trainees?
4 Is gamification just an expensive novelty?
■ How does gamification differ from serious games?
■ What are the advantages and disadvantages of gamification?
■ What is important to consider when evaluating gamification?
5 What can be done to enhance training in a very diverse workforce?
■ How would you identify who needs to be trained?
■ What would be the content of training?
■ What can be done to enhance training transfer?
Relevant websites
The UK Health and Safety Executive provide detailed guidance for employers regarding training in the use of work equipment. This material provides a good insight into how research
and theory in training can be translated into effective practical interventions: http://www.
hse.gov.uk/work-equipment-machinery/training-competence.htm
The Chartered Institute of Personnel and Development provides a range of resources to help
employers develop and implement effective training and employee development. Their
Learning and Development Survey provides an excellent insight into the challenges being
faced by employers: http://www.cipd.co.uk/hr-topics/learning-development.aspx
One of many examples of excellent online training courses for workers provided by the
National Institute for Occupational Safety and Health (NIOSH) can be found at: http://www.
cdc.gov/niosh/topics/violence/training_nurses.html Their publication on training effectiveness is also useful: see http://www.cdc.gov/niosh/docs/99-142/pdfs/99-142.pdf
This video provides an accessible summary of the role of unconscious bias in workplace decision making: https://royalsociety.org/topics-policy/publications/2015/unconscious-bias/
Suggested further reading
1 Ellingson, J.E. and Noe, R.A. (2017) Autonomous Learning in the Workplace. London: Routledge is a thorough overview of the shift to learner-centered training and
self-regulation.
2 Salas, E. et al. (2015) Team Training essentials: A Research-Based Guide. London: Routledge provides an accessible overview of evidence-based best practice for team training
in organisational settings.
CHAPTER 8
Work-related stress
and well-being
LEARNING OBJECTIVES
After studying this chapter, you should be able to:
1 define work-related well-being;
2 describe structural and transactional approaches to work-related stress;
3 describe conservation of resources theory and the job demands-resources model;
4 distinguish between theories of work-related ill-health and theories of positive well-being;
5 describe the impact of the content and context work on a range of well-being outcomes (from very negative to very positive, and from individual-level outcomes to
organisational-level outcomes);
6 discuss the quality of the evidence linking work to the outcomes described in 5, above;
7 identify some of the individual differences that are linked to work-related well-being;
8 list and evaluate ways of assessing work-related well-being;
9 distinguish between the activities
associated with, and the objectives
of, a range of different interventions
designed to improve work-related
well-being;
10 discuss the effectiveness of
different intervention strategies.
Chapter 8 Work-related stress and well-being
Opening case study
229
Overdoing it: the cost of China’s
long-hours culture
Gina Wang, a brand manager at Chinese
internet company Tencent, starts work most
days at 10am and generally finishes at 10pm,
but sometimes is still working at midnight.
Like many others in China’s tech industry,
she reports working long hours of overtime
for years. The 29-year-old thinks the burden
of work might be contributing to her stomach
problems and periods of depression. She is not
alone. Jing Li, 25, an accountant at China’s
largest telecoms equipment maker, Huawei,
also complains that all she does is work. She
feels that she too has succumbed to depression. ‘I’ll stay if I can change my position or
location next year. If I can’t, I will leave,’ Li
says, but she faces hard choices if she does.
She earned the equivalent of around $40,000
last year, about four times the average annual
urban wage in China. At the moment Li feels
it is not worth it. She says that although she
earns more than many other Chinese workers, ‘you have to work as if you are four people’. China’s manufacturing employees have
for decades worked long hours in often risky
environments, but staff and analysts say the
long-hours culture has spread to China’s
office workers, thanks in part to the country’s
rising internet and technology sector. Staff at
tech companies often cite the phrase ‘996’ to
describe their working life – starting work at
9am, leaving at 9pm and working six days a
week. Besides generous pay, there are perks –
some of China’s largest tech companies, such
as Baidu and Tencent, routinely provide free
transport or meal subsidies for employees
who leave after 9pm, according to employees.
What the companies do not do, however, is
provide the option to work fewer hours.
The problem of long and stressful working
hours can be a hidden one. The average Chinese employee spends 44.7 hours a week at
work, according to the 2017 China Labour
Dynamics Survey, published by Sun Yat Sen
University in Guangzhou, covering 21,086
people nationwide. China’s labour law states
that workers should not work more than
eight hours a day, or 44 hours a week, and that
anything more than this should be classed as
overtime, which can only be imposed after
consultation with worker representatives. At
first glance, therefore, the Sun Yat Sen study
appears to show that most workers surveyed
have a good work–life balance. But the limits
are loosely enforced, and more than 40 per cent
of respondents to the survey reported working
longer than 50 hours a week. The question of
overwork among Chinese white-collar workers, in the tech industry in particular, became
a talking point in 2016 when Zhang Rui, the
44-year-old founder of a mobile health app
start-up, died suddenly of a heart attack. Chinese media reports linked his death to his punishing routine of working late into the night
– often still sending emails at 3am. ‘Manufacturing sector working hours are declining:
you are not seeing the same level of excessive overtime that you used to,’ says Geoffrey
Crothall, a spokesman for Hong Kong-based
charity China Labour Bulletin. He cites falling
demand for manufacturing labour in the country as it transitions to a more service-based
economy. But, he adds, ‘in white-collar jobs,
working hours continue to escalate’.
Tech companies are leading the trend. In 2016,
Didi Chuxing, China’s biggest ride-hailing
platform, investigated urban white-collar
overwork, based on taxi bookings from business buildings to residential areas between
9pm and midnight on working days. Its data
showed that employees of internet, finance
and media companies were most likely to be
working late into the night. There is strong
evidence linking long hours to poor health.
People who work for more than 55 hours a
week face an increased risk of stroke and
coronary heart disease compared with those
who work between 35 and 40 hours a week,
according to a study based on data from more
than 600,000 individuals, published in 2015
in The Lancet medical journal. A survey in
Shanghai last year of patients with cardiovascular disease found a significant incidence
of arrhythmia – an irregular heartbeat that
can be a prelude to more serious disease –
in patients aged 21 to 30. Sun Baogui, executive vice-chairman of the Chinese Heart Failure Society, says this finding was consistent
with their reported lifestyle of long working
hours and getting little sleep. Some 85 per
cent of white-collar workers in China have
230
Work Psychology
to work overtime, with more than 45 per cent
reporting overtime of more than 10 hours a
week, according to a survey last year by job
recruitment website Zhaopin.
China is not the only country in the region
where long working hours have become a
cause for concern. Karoshi, as death from
overwork is called in Japan, has become a
recognisable phenomenon across east Asia,
according to an academic paper published
in 2014 – although China appears to have
lagged behind neighbouring countries such
as Japan, South Korea and Taiwan in providing proper recognition and compensation for bereaved families. Mental health is
another risk. Working long hours is associated with depression and anxiety, which
can lead to disturbed sleep and worse. A
national health and wellness survey in
2017 by Kantar Health, a consulting firm,
found that nearly 20 per cent of adults in
China who worked more than 51 hours a
week reported feelings of anxiety. Employees of technology companies who spoke
to the FT said they expected a long-hours
culture when they entered the industry.
However, although Chinese law requires
overtime work to be compensated at 1.5
times ordinary pay, they said they did not
generally receive such payment, because the
work was classified as ‘voluntary’, meaning it falls outside the official definition of
overtime. A Chinese lawyer, who wishes to
remain anonymous, says there is no mechanism to protect employees’ rights if they
have agreed to ‘voluntary overwork’.
The lines between voluntary and enforced
overtime are often unclear. ‘[When supervisors] tell you to attend a meeting, or prepare
materials, can you say no?’ says Huawei’s
Li. She says that in her experience much of
the overtime work she feels she has to do is
a result of inefficient management. Working
in a support role, she often has to wait for
signatures from different levels of the company bureaucracy. ‘Extreme cases were 10–20
signatures,’ she says. Huawei and Tencent
declined to comment for this article. Some
companies, however, are trying to change
the overwork culture. Feng Dahui, founder
and chief executive of Nocode Technology, a
medical research engine start-up, says mental work requires creativity and ‘overwork
does not achieve better results’. His employees usually start work at 10am and finish by
7pm, he says.
Source: Wang Xuegiao and Tom Hancock (2019) Overdoing it: the cost of China’s long-hours culture,
17 January © Financial Times Limited 2019. All rights reserved. https://www.ft.com/content/d5f01f6
8-9cbc-11e8-88de-49c908b1f264.
Introduction
The information in the case study suggests that the consequences of badly designed and
poorly managed jobs can be far from trivial. Occupations, such as teachers, ambulance staff,
police, social workers (Johnson et al., 2005), senior executives and those working in the technology sector (UK HSE, 2018) are often believed to have the highest stress levels. These jobs
have positive features: they tend to be varied, offer opportunities for challenge and creativity
and the chance to make important and influential decisions. Yet these benefits may be lost
if employees have little time to recover from their efforts and to engage in other activities.
The effects of stress may not be apparent in high levels of staff turnover: employees may be
reluctant to forego financial security and the opportunities afforded to them by their job.
Unfortunately, prolonged exposure to sources of stress carries significant risks.
The case study also indicates that the consequences of work-related stress are many and
varied. National laws mean that long working hours may not appear in employment contracts, but employees may face pressure to be present before and after the end of the contracted working day. This pressure may be overt, or it may be transmitted through subtle
indicators that imply long hours might be rewarded or are expected. Long working hours
can also be symptom or consequence of problems such as unclear expectations, inefficient
working practices or excessive bureaucracy.
Chapter 8 Work-related stress and well-being
231
Longitudinal studies have revealed findings that fit with many of the experiences reported
in the opening case study. These studies have provided powerful information about the possible causal links between the workers’ conditions and serious illnesses such as heart disease
(CHD). A consistent finding from the ‘Whitehall Studies’ – studies of workers employed in
the UK Civil Service (see Ferrie et al., 2005; Griffin et al., 2007; Kuper and Marmot, 2008) –
was that employees reporting high job demands combined with a lack of opportunities to
use their skills and to make decisions (referred to as high strain jobs) were at a significantly
increased risk of developing CHD. Among those with pre-existing cardiometabolic disease,
working in a high strain job has been linked to reduced life expectancy (Kivimaki et al., 2018).
There is plenty of evidence that high strain jobs also increase the risk of the development of
symptoms of depression (Theorell et al., 2015).
Such findings paint a rather gloomy picture. However, well-designed and managed
jobs can have a significant positive impact on employees. Workers facing the everyday
challenges presented by chronic diseases tend to remain in work longer if they report
experiencing high levels of social support and can decide how and when to use their skills
(Fleischmann et al., 2018). Work can trigger and sustain positive emotions that facilitate
employee learning and personal development (Fredrickson, 2013). Concepts such as thriving at work have led researchers to investigate what it is about work that leads to feelings of
being energised and alive (vitality) and a sense of continuous learning and improvement
(Porath et al., 2012).
This chapter begins with work-related stress. Sources of stress include job demands, low
control at work, conflicting or unclear or unreasonable work roles, workplace bullying, and
the tensions between work and non-work demands. To provide balanced coverage we then
examine the positive aspects of work. We consider several individual differences that play a
role in determining how employees respond to their experiences at work. We then examine
the content and effectiveness of different interventions for reducing stress and promoting
well-being.
Work-related stress
What is stress?
Cannon (1929) defined stress in terms of the internal physiological state of subjects exposed
to threatening or exciting situations. This includes the raised and helpful adrenalin secretion
in the well-known fight or flight reaction. Hans Selye (1946) used data from experiments
to devise a model of the human response to stressors: the General Adaptation Syndrome. It
has three stages:
1
Alarm reaction, in which an initial phase of lowered resistance is followed by countershock, during which the individual’s physiological defence mechanisms become
active.
2
Resistance, the stage of maximum adaptation and, hopefully, successful return to
equilibrium for the individual. Responses in these two stages are useful as they provide the energy that individuals need to deal with stressors. If the stress agent continues or the defence mechanism does not work, the individual will move on to a third
stage, namely:
3
Exhaustion, when adaptive mechanisms collapse and harm starts to occur. Prolonged,
frequent or intense stress responses can, over time, cause overactivation and impair the
functioning of physiological systems (Eddy et al., 2016). For example, opportunities
for recovery that allow symptoms of stress to dissipate are an important contributor to
employee well-being (Sonnentag and Fritz, 2015).
232
Work Psychology
The psychological mechanisms associated with these physiological changes are now better
understood. Stress is not an illness in itself: it resides in the intervening psychological processes that link exposure to work-related problems to the negative impact of those problems.
Stress is associated with acute negative emotions and extended periods of low mood. The UK
Health and Safety Executive’s definition captures the various scientific definitions: ‘the process that arises where work demands of various types and combinations exceed the person’s
capacity and capability to cope’ (see: www.hse.gov.uk/stress). Stress is not high workload or
work-related absence, rather it is the psychological processes that link stressors to outcomes.
Stressors are those aspects of work that have the potential to cause negative and unpleasant
and harmful outcomes.
Individual perception or appraisal plays a role in the stress process. Some students enjoy
examinations. Others (probably most) find exams to be stressors. Students who feel in control of events and have completed plenty of good quality revision will tend to feel fewer
negative emotions because they have resources (e.g. their knowledge, good exam technique
etc.) available to meet the impending demands (the exam). In other words, they feel they
can cope. They may even see preparing for an exam as challenge that provides them with
opportunities to develop. Such individual differences have significant implications for how
stress is measured and managed.
Key learning point
Stress is a state that can occur when an individual’s perception of their ability to cope is not
well-matched with the demands they perceive in their environment.
Models and theories of work-related stress
Most models and theories describe stress as resulting from a lack of fit between an individual’s
capacity and the demands they anticipate and encounter. These help us to understand why
one person flourishes while another suffers in the face of similar stressors. Some focus on
the factors that determine how an individual chooses and evaluates their responses to stressors. Contemporary models and theories offer quite detailed explanations of how workers’
thoughts and behaviours result in them changing the way they respond to work demands.
To simplify things a little, most can be split into three categories: structural approaches, transactional approaches and resource-based approaches.
Structural approaches
Structural approaches to stress focus on describing the features of work that are likely to
lead to stress for most workers. These are the aspects of work (often referred to as psychosocial
working conditions) that are reliably linked to significant physical or psychological health
problems. The job demands control-model (JDC-M; Karasek, 1979) is widely researched.
Demand refers to the amount and intensity of work a person does, and includes time
pressures, the amount of mental and physical effort involved in the job and its difficulty.
Quantitative demand is the amount of work. Qualitative demand is the type and difficulty
of work. Most people expect high workload equals high stress. This is true up to a point.
A meta-analysis by Bowling et al. (2015) found that high perceived workload is consistently
Chapter 8 Work-related stress and well-being
233
and strongly linked to impaired physical health. However, the links between workload and
measures of depression were weaker. In in the American Psychological Association’s 2017
Work and Well-Being Survey, a minority (39 per cent) of workers report that having too
heavy a workload was either a significant or a very significant work stressor for them (low
salary and lack of opportunities for enhancement were more frequently cited). Control also
appears to be especially important.
Karasek’s (1979) original definition of control included two components. Skill discretion is
the extent to which the job allowed the employee to use their skills and capabilities. Decision
latitude is the amount of control the employee has over their work situation through their
decision-making. Added together as separate entities, control and demand have implications
for well-being. Beyond this they may also combine and interact with each other to produce
especially stressful environments known as high- or low-strain jobs (De Lange et al., 2003).
This is known as the strain hypothesis (Daniels et al., 2009). By combining demands and
control, four categories of job have been identified:
1
passive job (low demands – low control);
2
high-strain job (high demands – low control);
3
low-strain job (low demands – high control);
4
active job (high demands – high control).
Employees classified as being in high-strain jobs were more likely, over time, to develop
various significant health problems (Karasek and Theorell, 1990). Further, high strain jobs
are a significant risk factor for stroke later in life (Fransson et al., 2015) and for depression (Theorell et al., 2015). Control is thought to be so important for two reasons (Fox
et al., 1993; Parker, 2014). First, it provides employees with opportunities to deal better with
demands (e.g. by having the freedom to be proactive, to prioritise tasks and to use their
skills to get the job done well in the most efficient way). This is the buffering hypothesis but
for it to work there needs to be a good match between the type of control available to the
employee and the type of demand they are facing (Hausser et al., 2010). Second, it fulfils a
basic human need for control that has consistent and strong implications for our well-being
(Deci and Ryan, 2000). Low levels of control can have significant negative consequences
by causing frustration as we want to feel competent and in control. High levels can yield
significant benefits by providing us with opportunities to have autonomy over the development of our competence.
Johnson and Hall (1988) added a third factor: social support (the JDCS-M). Support
encompasses the helpful interactions with supervisors and co-workers (Daniels et al.,
2009). As well as a low support being a stressor in its own right (see later in this chapter), high levels of social support can help employees faced with high demand situations.
Emotional support, practical help to do the job (instrumental support), and information
from others (informational support) can reduce exposure to work demands (O’Driscoll
and Brough, 2010). Given similar levels of demands and control, jobs with higher levels
of social support will tend to be healthier. Recent studies indicate jobs with higher level
of support tend to have lower levels of demands (Fila et al., 2017). However, an important
finding is that providing high-dose social support is not a panacea for or vaccine against
the likely consequences of high strain jobs (van der Doef and Maes, 1999; Viswesvaran
et al., 1999).
High control and social support when accompanied by high job demands can also create
active jobs. In this situation workers have the autonomy they need to use existing skills and
to learn new skills to adapt to demands that lead to improved motivation and satisfaction.
This is the learning hypothesis (De Witte et al., 2007). The benefits of this type of work may be
at risk when other aspects of the job are not well-managed: long and unpredictable working
hours can mean that active jobs become saturated (Holman, 2013).
234
Work Psychology
How stressful are these jobs?
Exercise 8.1
Below is a list of job roles. Karasek and Theorell (1990) describe how several different jobs were categorised according to the levels of demands and control in each job. In which of the four categories
of job is each job role likely to be most appropriately classified?
■
call-centre phone-call handler;
■
elite team sports athlete;
■
qualified medical doctor;
■
self-employed delivery driver delivering parcels for a large internet retail company;
■
company chief executive;
■
warehouse operative for large internet retail company.
Suggested exercises
1
Do you think that some of these jobs could be inherently stressful and therefore likely to be seen as
stressful by most people?
2
In your current work, or in your studies, how and why does having control help you when you are faced
with high demands?
Karasek’s four combinations of demand and control still make sense for lots of work roles.
Holman (2013) argues that modern workplaces bring with them additional risks and benefits:
■
Senior managers and professionals may have saturated jobs. These are similar to active
jobs but may place worker health at risk because they have a very high workload, high
demands for interaction with others and long and unusual working hours.
■
Team-based jobs involve lots of collective decision-making such as the allocation of tasks
to team members, coupled with lots of discretion and support. The work is usually carried
out during fixed, standard hours.
■
Insecure jobs are low-paid roles in which employees are uncertain about the security of
their tenure (e.g. those working in internet-based services, the ‘gig economy’, delivery
staff). These work roles can have low levels of control and support, few opportunities for
training and progression and may not provide workers with the sense of social identity
associated with secure jobs (Selenko et al., 2017). It is not just the content of the job where
stressors can be found: job insecurity is an issue not considered in structural models.
In his vitamin model Peter Warr argued that structural models over-simplify things by
making linear the more complex links between work and health (see Warr, 2009; Warr and
Nielsen, 2018). Working conditions, like vitamins, can have a positive impact on health,
but ‘mega-doses’ could be harmful. Having lots of control at work might not be helpful for
employees who do not have sufficient knowledge, confidence or skill to exercise that control
in a productive way. As we discuss later, feeling overwhelmed with unsolicited or unwanted
support from colleagues might not be a good thing.
Key learning point
Demands, control and support at work are particularly important in structural models of work stress
but there are many other potential sources of stress.
Chapter 8 Work-related stress and well-being
235
Other models have looked at a wider range of sources of stress and placed more emphasis
on describing the impact of employees’ appraisals of the ‘fit’ between their own skills, abilities and preferences and the demands of the job. The Michigan model (Caplan et al., 1975)
and the person–environment fit model (French et al., 1982) propose that each employee’s
skills, abilities, knowledge and attitudes need to match the demands of their job. Stress is
likely to occur if there is a mismatch. Subjective perceptions of work stressors are particularly
important and these include factors that are particularly relevant in contemporary work
roles, such as job insecurity, lack of participation and having a poorly defined role at work.
Lack of fit allows us to consider the role of skills, ability, personality and other individual differences in the stress process. For example, many psychologists would find it very stressful to
have to repair a broken-down car: this is a challenge that many skilled mechanics would relish. Watson et al. (2018) identified several interventions focused on developing job-related
capabilities that impacted positively on well-being, but there were also many that did not,
although none did harm. There are so many factors that can influence person–environment
fit that it is difficult to make specific predictions from these complex models (Mark and
Smith, 2008). With its more easily testable predictions the JDCS-M remains widely used in
the research literature.
Key learning point
The amount of ‘fit’ between the employee and their work environment plays an important role in
determining how stressful the employee finds their job.
Transactional approaches
Through any observation of an individual at work it becomes clear that they react to,
respond to and transact with the work environment: we adjust our thinking and behaviour,
our approach to tasks and the environment within which we work. By coping or responding, workers change their situation and then respond to the changed situation through an
unfolding series of transactions. At the centre of these models are the worker’s dynamic and
ever-changing perceptions of the balance between the demands they perceive and their evaluations of their own coping resources. Judgements about the frequency, severity and intensity of
the stressor become important. These are cognitive theories of stress and coping: workers
make sense of their experiences through evaluating working conditions (e.g. ‘is my workload
high?’); their emotional responses (‘is the way I am feeling unpleasant?’); and make decisions
about their coping and behaviour (e.g. ‘Is what I am doing reducing my workload?’). Transactional approaches also highlight the importance of workers’ evaluations of the success of their
attempts to respond to demands (i.e. workers look for and evaluate feedback). These private
processes are not considered in detail in structural models.
Lazarus and Folkman (1984) proposed that individuals evaluate the extent to which a situation poses a threat to them or presents them with opportunities for gain: this is a primary
appraisal. The appraisal might be that the situation is a problem (a harm-loss appraisal), could
be a problem (a threat appraisal) or presents some potential for gain (a challenge appraisal).
There is then a secondary appraisal: if a person thinks they can cope with a situation they are
less likely to appraise it as a threat and more likely to appraise it as a challenge. Together, these
primary and secondary appraisals are the basis of what the person chooses to do in response to
the situation (i.e. their transaction with the work environment). These appraisals are also linked
to the person’s emotional response and, in turn, their physiological responses (such as changes
in heart rate or blood pressure). If threat appraisals occur often, there are likely to be negative
236
Work Psychology
outcomes for the employee such as negative emotions and low mood. Challenge appraisals
are likely to yield positive emotions (e.g. excitement) and be motivating. Various individual
differences mean that different people may make dissimilar appraisals of similar events.
Workers evaluate the outcomes of the coping response/the action taken – a tertiary
appraisal (Cox, 1993). If they feel their action was effective, then they may become immediately less anxious and, in the future, perceive similar problems as being less stressful. Tertiary
appraisals feed back into subsequent primary and secondary appraisals. For example, focusing
on and completing important tasks might help employees to reduce subsequent perceived
demand. Some coping choices are maladaptive (e.g. unhealthy levels of alcohol consumption
after a demanding day at work). These choices can be influenced by individual differences:
some people might have high levels of self-efficacy and therefore feel very able, allowing them
to appraise complex or new tasks as challenges. Others may be less confident and focus on
dealing with unpleasant emotions. Feedback on the success of the employee’s actions might
not always be accurate or available. For example, bosses and customers might not always tell
an employee when they have done well or badly. In addition, people do not always interpret
feedback correctly.
Key learning point
Transactional approaches consider employees’ perceptions, individual differences and coping decisions. These also consider the ways in which the individual receives and responds to feedback on their
experiences.
Siegrist (1996; 2017) argued that two factors are particularly important in many transactions: perceived efforts and perceived rewards. In the effort–reward imbalance (ERI) model
high cost and low gain situations for the employee are linked to the experience of negative emotions. This is because the employee perceives unfairness or experiences lowered
self-esteem when they feel that their efforts are not reciprocated in terms of money, esteem
and recognition. Those in precarious work or difficult financial circumstances, may be particularly dependent on the rewards, however unfair, and thus have no choice but to remain
in a stressful situation (Siegrist, 2008). Such workers’ difficulties may be further compounded
by the effects of job insecurity and exposure to poor psychosocial working conditions (Benach
et al., 2014). Others might be motivated to increase work effort, or over-commit, in a desire
to obtain reward and recognition, perhaps accepting their current situation as a price worth
paying for better rewards in the future. Therefore, ERI indicates that poor career development
and job insecurity can be powerful sources of stress. This model also indicates that self-esteem
can be at the heart of the stress process: some might be driven by an underlying need for
approval at work and thus not accurately assess the balance between efforts and rewards. The
individual then gets into a cycle of working harder and harder, thus experiencing more stress
at work and potentially reducing the quantity and quality of recovery from work. Unscrupulous managers might be tempted to exploit this effect by withholding rewards from hardworking employees. Of course, not every worker will tolerate such a situation especially if
other options (such as early retirement) are available to them. In a prospective study of over
a thousand individuals Hintsa et al. (2015) found that ERI significantly raised the risk of exit
from the labour market (at 61 years or younger). There is quite a lot of evidence to support the
main predictions of this model and these various mechanisms (Siegrist et al., 2014; Tsutsumi
and Kawakami, 2004). For example, a large well-designed 14-month longitudinal study of
UK University workers showed that lower security (job security and promotion prospects)
rewards predicted lower mental health (Kinman, 2019).
Chapter 8 Work-related stress and well-being
237
Resource-based approaches
Conservation of resources (COR) is based on the premise that employees are proactive and
motivated to identify, protect and develop resources (Hobfoll, 1989; Hobfoll et al., 2018).
Resources are the personal characteristics and environmental conditions that are widely
valued because they can be used to deal with demands. While especially valued resources
can vary from one person to another, Hobfoll argues that individuals value conditions (the
worker’s situation such as having good security of tenure or a senior role), personal resources
(such as locus of control, optimism and self-efficacy) and energies (including time, knowledge
and skills). Social support helps people to acquire and preserve these resources (e.g. a good
network of friends can help people to get things done, achieve promotion and bolster their
self-esteem).
Hobfoll et al. (2018) argue that in contrast to appraisal theories, COR allows us to predict
what is stressful before a stress-inducing event occurs. In addition, some events (e.g. the possibility of becoming unemployed) would be stressful to almost everyone, downplaying the
importance of idiosyncrasies in appraisals. The resources provided by work highlight the risks
associated with unemployment. Jahoda (1979) argued that work supports structuring of time,
social contact outside the family, personal status/identity and regular activity. Warr (1987)
identified several features of ‘psychologically good’ employment (e.g. money, variety, goals,
skill use and development, security, valued social position etc.).
Employees are more likely to experience stress when resources are lost or when there is a
threat of loss. Very high workload might place well-being at risk because it uses up resources
and can make it more difficult to replenish them and acquire new ones (Bowling et al., 2015).
Gradual loss of job security, erosion of confidence in one’s ability to do the job, or reduced
time to complete tasks could all raise the prospect of the depletion of resources over time. In
fact, as Halbesleben et al. (2014: 6) argue, resources can be ‘anything perceived by the individual to help attain his or her goals’.
The COR model provides a novel insight into proactivity. Employees act to obtain and
secure resources that matter to them. If a resource is lost or threatened there is likely to be
a strong response. These effects are stronger, felt more quickly and longer-lasting than the
positive effects of gains in resources: resource losses have a disproportionate effect (Hobfoll
et al., 2018). In the face of a threat of resource losses, employees may act to replace that
resource or rethink its importance. For example, after missing out on a promotion at work a
worker might re-evaluate the importance of their career progression.
The job demands–resources (JD-R) model (Bakker and Demerouti, 2007; Demerouti
et al., 2001; Van den Broeck et al., 2013) applies to the workplace principles from Hobfoll’s
COR and elements from other approaches mentioned in this section. Resources can be
used to achieve work goals, deal with and reduce job demands and stimulate personal
growth and development. Bakker and Demerouti (2007) argue that structural and transactional approaches over-emphasise the importance of a relatively narrow range of potential
work stressors given that job demands are context dependent. They cite the high level
of information-processing demands faced by air traffic controllers working with technology, a role with threat avoidant vigilant work (Belkic et al., 2000). This is different to
the demands to show empathy faced by a nurse. What are the differences in the types of
demands people typically face in each work role in Exercise 8.1?
The JD-R is a dual process model. Psychological health impairment processes deplete
employees’ resources and can lead to exhaustion. Demands can trigger this process, especially those hindrance demands that obstruct effective performance or prevent the worker
from learning and developing. Challenge demands, in contrast, provide workers with opportunities to succeed, learn and grow. Job resources can activate these motivational processes
because they help employees achieve their goals and can have a positive impact on employee
238
Work Psychology
attitudes and competencies. They can lead to high levels of work engagement, which has been
described by Schaufeli et al. (2002) as having three components:
1
Vigour: Having high energy and mental strength to exert a high degree of effort and to do
these things even in the face of difficulties.
2
Dedication: Holding the view that your effort ‘means something’; feeling inspired, enthusiastic and welcoming new/persistent challenges.
3
Absorption: Being so engrossed that it is difficult to ‘let go’ of tasks; attachment to tasks at
the exclusion of more mundane concurrent events (such as the passing of time).
Key learning point
The JD-R model helps to explain why features of work environments can be linked to well-being and
good work performance.
Demands can moderate (weaken or strengthen) the links between resources and motivation: in a job with very high demands, resources may not be as motivating as they might
otherwise be. Similarly, when resources are very high, the link between demands and strain
might not be particularly strong. According to Bakker and Demerouti (2007) resources can
also act as buffers against demands. Research also shows that, over time, demands are not
strongly linked to engagement: rather that engagement appears to be a product of abundant resources (Lesener et al., 2018). This suggests that interventions that increase resources
can help to both reduce health impairment and promote motivation. Autonomy, feedback
and social interaction have all been identified as particularly important and common job
resources (Crawford et al., 2010; Nielsen et al., 2017a). Job autonomy can give employees
opportunities to develop their own effective ways of responding to work demands without
having to deplete their resources too much (Parker, 2014). Good feedback on performance
can help employees to adjust their work efforts to use their resources more efficiently. In
their meta-analysis Nielsen et al. (2017a) found that HR practices (e.g. provision of training
and career support) and transformational leadership behaviour were sources of resources
that were reliably linked to both employee well-being and performance. There seems to be
good evidence for the predictions of the JD-R as being applicable across a range of occupations (Brauchli et al., 2015).
Large studies of engagement also show that those in managerial and professional roles
tend to exhibit higher levels of engagement than those in occupations requiring a lower
level of skill, perhaps because such jobs tend to provide fewer organisational resources
(Hakanen et al., 2019). The authors of this study also suggest that the low levels of engagement reported by those working without a permanent contract might be symptom of their
exposure to poorly designed jobs and their lack of access to resources such as training and
career progression.
Personal resources and non-work resources have also been added to the model (Bakker
et al., 2014; Bakker and Sanz-Vergel, 2013). These are a cluster of individual differences related
to an individual’s own perceptions of their ability to successfully control and to have an
impact on their environment (e.g. self-efficacy). Non-work resources include support from
family and friends. The demand-induced strain compensation model (DISC; see de Jonge
and Dormann, 2006; van Veldhoven et al., 2005) includes three categories of resources and
demands (emotional, cognitive and physical). In this model, the matching of each type of
resource to each type of demand is important – if an employee has high levels of cognitive
resources this might not help them much if the physical demands of the job are very high.
Chapter 8 Work-related stress and well-being
239
Stress is more likely to occur if there is a mismatch between the type of demand and the type
of resource available.
Key learning point
Individuals appear motivated to develop and protect various psychological, organisational and
non-work resources that help them to meet the demands of work.
The JD-R model also suggests that employees will take proactive steps to change their
own work environment. Job-crafting encompasses ‘the physical and cognitive changes individuals make in the task or relational boundaries of their work’ (Wrzesniewski and Dutton,
2001: 179). Individuals strive to shape a job role in which they can use their skills (Tims
and Bakker, 2010) and identify and use resources that they value (Demerouti et al., 2001).
This can involve shaping work activities to reduce demands, enhance control or maximise
available social support (Holman et al., 2010). This portrays workers as being proactive in
the way they seek to adjust their work activities (Petrou et al., 2012). There is some evidence
that activities such as taking steps like reducing one’s social contact with others at work may
allow employees to perform better at core tasks, but perhaps at the expense of job satisfaction
(Nielsen et al., 2017a).
Key debate
Which explanation of stress is best?
Structural approaches to stress offer relatively simple and universal predictions about the impact of work on
employee well-being. This simplicity is not always borne out by research findings. However, there is plenty of strong
research showing that features of work such as demand, control and support have significant implications for employee
health. Transactional theories of stress provide a sophisticated account of the dynamic psychological mechanisms
of stress and point to the impact of individual differences. This added complexity makes them more difficult to test
and validate. Resource-based models have done much to further our understanding of the proactive steps that
individuals take to prepare themselves to respond to demands and on how work can have a positive impact. The JD-R
model offers a flexible means of capturing workers’ proactive behaviour and the idiosyncrasies of the stress process.
It also provides insight into both the risks and benefits of work (the dual processes).
The consequences and costs of work stress
The immediate physiological effects of stress include high levels of activation of the
sympathetic-adrenal-medullary system and the hypothalamic-pituitary-adrenal system
of the brain (Nixon et al., 2011). These trigger elevated adrenaline levels, blood cortisol,
heart rate and blood pressure (Sonnentag and Fritz, 2015). This can be very useful, as Selye
proposed, in preparing the body for action. Prolonged or frequent stress responses can
directly damage physiological systems and hinder their repair through mechanisms such as
chronic inflammation (Liu et al., 2017). Kivimaki and Steptoe (2018) highlight the evidence
linking stress to cardiac electrical instability, disruption to blood flow and the formation of
blood clots. Many of the symptoms of stress can be detected before the onset of stress-related
illnesses.
240
Work Psychology
Behavioural and psychological symptoms include: acute changes in emotions (such as
anger), prolonged change in mood (such as sadness), difficulty in making decisions, an inability to concentrate, loss of sense of humour, and feeling unable to cope. Physical symptoms
include: changes in appetite, disturbed sleep, elevated blood pressure and tiredness. These
symptoms indicate that stress plays a role in significant health problems including cardiovascular disease, migraine, asthma, peptic ulcers, menstrual difficulties, skin disorders and
depression. However, it is often difficult to isolate with precision the extent to which stress
plays a part.
Key learning point
Stress can trigger physiological changes and have an impact on how people think and behave.
These changes can have an impact on individual well-being and outcomes valued in organisational
settings.
The relationships between stress and behaviour at work are not simple. For example,
absence is a complex form of illness-related behaviour influenced heavily by a host of psychological, physical, societal and cultural factors (Johns, 2011; Whitfield, 2009).
Four categories of measures are used to examine work-related health and well-being:
1
organisational;
2
psychological and physical health (sometimes referred to as subjective well-being as data
are often collected through questionnaire measures distributed to employees);
3
physiological;
4
health-related behaviours.
Organisational measures
Measures valued by employers include absence levels, measures of job performance and staff
turnover. Absenteeism is widely researched because of its costs and because it is evidence
that an employee is experiencing a serious problem. Various causes of absence such as home
and family responsibilities, personal problems and drug and alcohol problems may also be
stress-related (EU-OSHA, 2014). A combination of high demands and low control has been
shown to place employees at significantly greater risk of developing common mental health
problems and diseases that cause prolonged absence (Ferrie, 2004; Stansfeld and Candy,
2006). Stress is indirectly linked to other causes of absences from work or early retirements
(EU-OSHA, 2014). Poor working conditions can also have an adverse impact on worker attitudes (e.g. job satisfaction and commitment) and therefore reduce motivation to attend
(Johns, 2011).
Inaccurate recording of absence as stress-related makes it difficult to be precise about how
much it causes absence. Stress-related absence tends to be costly because it tends to be lengthy
and managing employees’ return to work complex. Hassard et al. (2018) in their systematic review found the majority (70–90 per cent) of these costs to be productivity losses (a
function of the duration of absence and the employee’s salary) with the remainder being
healthcare and medical costs. A more difficult to detect problem is presenteeism. This occurs
when people are attending work when they are ill (Johns, 2010). Ahlstrom et al. (2010) found
that presenteeism is a strong predictor of future absence, making it a useful early warning
signal. The antecedents of presenteeism include both positive motivational reasons such as
Chapter 8 Work-related stress and well-being
241
enjoyment of and commitment to work, maintaining a social network and a strong sense of
obligation to colleagues. Negative reasons are associated with health impairment and stress,
such as fear of losing one’s job or loss of earnings (personal resources), strict organisational
absence policies and the pressure of work (Johansen et al., 2014; Miraglia and Johns, 2016).
Johns (2010) suggested that two working conditions linked to presenteeism are: adjustment
latitude (this allows employees to modify the way they do their work or their work outputs
when they are feeling unwell); and ease of replacement (the extent to which the work missed
by an employee while absent can be done by others).
Employee turnover has been shown to be linked to working conditions in several studies (see Bond et al., 2006). Estimates vary but there is reasonable evidence indicating that
around one in five cases of staff turnover is stress-related (see also Chapter 5 for a more
complete discussion of turnover). High rates of employee turnover raise training costs,
reduce overall efficiency and disrupt other workers (EU-OSHA, 2014). Taking data from
many different organisations, Taris (2006) and Taris and Schreurs (2009) found significant,
modest negative correlations between employees’ work-related exhaustion and their performance. In general, higher levels of exhaustion are associated with lower performance;
but the modest correlation also means that the effect is not consistently strong, nor does it
hold true for all workers.
The JD-R model indicates that some demands provide opportunities for growth: hence perhaps the argument that ‘a little bit of stress is good for you’. This statement conflates the stress
bought about by difficult working conditions with the challenges provided by well-designed
work. Difficult working conditions might lead to increased performance under some circumstances (e.g. if employees perceive that their position is under threat if they do not perform or
there is a pressing and important task to be completed). After reviewing the evidence, O’Driscoll and Brough (2010) concluded that as jobs get more stressful, there is a greater negative
impact on employee well-being and performance.
Key learning point
The links between individual well-being and outcomes valued by organisations (e.g. low absence) are
not simple and stress is only one of many influences on these outcomes.
Indicators of psychological and physical health
Research has revealed links between psychological well-being and three major categories of
physical health: cardiovascular disease; sugar/fat metabolism; and immune system functioning. Kuper et al. (2002) concluded that, ‘based on prospective epidemiological data, there was
evidence for an association between depression, social support and psychosocial work characteristics and (coronary heart disease) aetiology and prognosis’ (2002: 267). Similar research
has also established links between psychological stress, obesity and diabetes (Brunner, 2007;
Heraclides et al., 2009) and compromised functioning of the immune system (Cohen et al.,
2001). Stress may also play a role in the development of musculoskeletal problems through
its impact on biomechanical load (e.g. time pressures may lead to workers lifting in a hurry
and adopting unsafe postures) or through chemical and biological processes associated with
a stress response, such as muscular tension, reduced blood flow to the extremities and poor
muscular repair (Hauke et al., 2011).
Data on the incidence of disease may be available at an organisational level in company
records of staff absences and return to work processes. Self-report questionnaire measures of
physical and psychological symptoms can be more sensitive to the early signs of poor health.
242
Work Psychology
Some measures have threshold scores: people scoring in excess of these levels are significantly
more likely to already have a diagnosable psychological illness (e.g. clinical depression). For
example, the General Health Questionnaire (GHQ) (Goldberg and Williams, 1988) measures symptoms of anxiety, depression, social dysfunction and loss of confidence. The Brief
Symptom Inventory (BSI) (Derogatis and Melisaratos, 1983) assesses a wide range of physical
and psychological symptoms. The World Health Organisation’s Well-Being Index (WHO-5)
measures subjective well-being that is quick and simple to complete and has been shown to
be linked to measures of depression (Krieger et al., 2014). More specific work-related measures (e.g. work engagement) can provide an excellent insight into the positive outcomes of
work (Salanova et al., 2014; Schaufeli et al., 2002). The Maslach Burnout Inventory (MBI) (see
Maslach and Jackson, 1981) measures psychological symptoms of emotional exhaustion and
cynicism that are particularly important to those in the, often regarded high-stress caring
professions. These are:
1
Emotional exhaustion: feelings of being emotionally over-stretched and exhausted
by work.
2
Depersonalisation: an unfeeling and impersonal response to clients.
3
Personal accomplishment: feelings of competence and successful achievement at work.
Measures of recovery from work (e.g. Oerlemans and Bakker, 2014) are increasingly widely
used because recovery appears to help explain how it is that some people in very demanding
jobs not only retain their health by reducing symptoms of strain but also continue to experience enjoyment and fun (Sonnentag and Fritz, 2015).
Key debate
Is it acceptable to use questionnaires in stress research?
Some argue that self-report data are heavily contaminated with the errors and biases that impact on human
memory and judgement. It may be that personality factors such as neuroticism (see Chapter 2) can influence the
reporting of symptoms of illness (Watson and Pennebaker, 1989). However, questionnaires with good reliability
and validity can capture data about the immediate effects of the work environment on the individual such as
negative emotional experiences, psychological distress and dissatisfaction. Self-reports reveal these very personal and subjective experiences. It could also be argued that questionnaire measures are less impacted by the
multitude of factors that impact on organisational-level measures (e.g. absence and turnover) and measures
of disease. Self-report data become even more powerful if secured by triangulation and the use of longitudinal
research designs.
Longitudinal studies can reveal problems with self-reported psychological health that are developing over a period
of time and show whether such changes are associated with fluctuations in levels of work stressors. If these variations
are also accompanied by, for example, increases in absence levels or diagnosed illness, then we might be more likely
to accept that the self-reported psychological health data are indicating that intervention is needed.
Physiological measures
Changes in our physiological functioning are some of the first things that many of us sense
when we experience stress (e.g. elevated blood pressure, increased sweating, heart rate and
muscular tension). These changes can be measured and provide relatively ‘instant’ indications that the body is responding to something. Collecting data about these changes is not
simple and raises significant ethical issues associated with the invasive processes involved.
Chapter 8 Work-related stress and well-being
243
Moreover, the construct validity of these data is not always clear. Many fitness trackers
collect heart-rate data: but elevated heart rate could indicate any combination of engagement in strenuous physical activity, excitement or anxiety. The relatively infrequent use of
these measures in applied stress research in organisational settings is an indication of these
limitations.
Health-related behaviours
Health-related behaviours include sleeping patterns, exercise levels and maladaptive coping
behaviours such as poor diet, smoking, drug and alcohol use. These data are usually captured through self-report questionnaire measures or records from medical records. Sometimes health-related behaviours are referred to as maladaptive coping patterns. These are
a means of coping with work stress that itself causes harm. The use of maladaptive coping
strategies often indicates that an employee has already been ‘damaged’ in some way by their
work. As such, these problems are often dealt with through tertiary interventions, which
are examined later in this chapter. The relationship between these behaviours and working
conditions is not always obvious. Long working hours seem to play a role in alcohol use
(Virtanen et al., 2015) with there being some evidence that workers use it to self-medicate
when they feel the effects of fatigue at work, low mood and reduce tension (Frone, 2016).
Gimeno et al. (2009) found those in jobs with low demands coupled with low control were
significantly more likely than others to engage in heavy drinking (but were not likely to
drink more frequently than most). Sikora et al. (2008) found, during downsizing, alcohol
use can also be a cause of work stress (e.g. impacting work performance) as well as a consequence of it.
Stop to consider
Theories and models of work stress are designed to help explain the problems with organisational and
individual health. How could you test whether these theories did help to explain these problems? Have
a look back at Chapter 1 and the Research methods in focus features of this text. Consider if any of the
research methods described there could be of use. Thinking about this issue will help you to see how
stress research is generally conducted.
Links between outcomes
The link between employee health and organisational outcomes may be influenced by attitudes and culture. Väänänen et al. (2008) found that a group of employees’ tolerance for
absence behaviour among their colleagues and their cohesiveness (or ‘togetherness’) influenced individual workers’ attitudes towards taking absence. Employees can find their job
satisfying and be reluctant to take time off (perhaps because of felt obligations to clients and
colleagues) while still finding the job to be exhausting (see Miraglia and Johns, 2016). Absence
can also trigger improvements in psychological health because it allows the employee to
escape from stressors, forget about work, do other things, and for negative mood to dissipate and recovery to occur (Sonnentag and Fritz, 2015). Interestingly, retaining some contact
with the workplace does seem to help employees manage a successful return after absence
(Mikkelsen and Rosholm, 2018).
244
Work Psychology
Psychological well-being (PWB)
For most workers both stress, and positive emotions and moods occur during a working day.
Studying only the absence or presence of stress overlooks issues such as enjoinment or thriving at work. The extent to which a person is experiencing pleasant positive affect is independent of the extent to which they are experiencing unpleasant negative affect (Agho et al.,
1992). It is also possible to distinguish between high arousal and low arousal. This creates four
quadrants of affect and emotion (see Figure 8.1).
Key learning point
The concept of stress may be too narrow to capture the emotions people have at work. Positive
well-being is not simply the absence of ill-health.
Positive psychology emerged from a dissatisfaction with a focus on the healing functions
of applied psychology (Seligman and Csikszentmihalyi, 2000). It is focused on developing a
better understanding of how people flourish when all is well (see for example the COR theory earlier) and of how people develop and grow in difficult times. Wright and Cropanzano
(2004: 348) noted that:
Employee PWB (Psychological Well-being) has both theoretical and applied relevance in today’s
society. Using the Positive Psychology/Positive Organisational Behaviour (POB) framework, it
seems evident that promoting employee PWB is an intrinsic good for which all should work. If this
approach promotes better job performance, which the findings strongly suggest is the case, then
so much the better.
Seligman (2011) argued that well-being has five components, all of which have already
appeared in slightly different guises in this chapter: engagement, relationships, meaning
and accomplishment all feature in addition to positive emotions. Ryff (2014) argues that PWB
has six dimensions: autonomy; environmental mastery; personal growth; positive relations
with others; purpose in life; and self-acceptance. Robertson and Flint-Taylor (2008) point
to two factors that underpin these components: a purposive eudemonic element and an
emotional hedonic element. In other words, it is the affective and purposive psychological state that people experience while they are at work. The hedonic element associates
well-being with the experience of positive moods and emotions. Over the long-term most
people feel at their very best not when totally relaxing or enjoying themselves (hedonic),
High arousal
Panic
Worry
Excitement
Passion
Unpleasant
Pleasant
Boredom
Sadness
Contentment
Satisfaction
Low arousal
Figure 8.1
Four types of emotion
Chapter 8 Work-related stress and well-being
245
but also when something worthwhile has been achieved (eudemonic). Sense of purpose
appears to be associated with positive emotions.
Point of integration
In the workplace, ‘sense of purpose’ is most obviously related to the job role that people hold – and its
goals and objectives. Goal-setting theory (see Chapter 6) shows that when people are committed to
goals that are clear, specific and challenging they tend to perform well. People also tend to experience
PWB when they have a clear sense of purpose, i.e. clear, specific and challenging work goals.
The experience of positive emotions puts individuals in a psychological state that allows
them to grow new knowledge and skills by broadening thought-action repertoires (Fredrickson, 1998; Fredrickson and Joiner, 2002). This is the ‘broaden-and-build’ theory. Its authors
argue that when individuals experience negative emotions they remain narrowly focused on
a restricted range of thoughts and actions. Positive emotions broaden the scope of attention,
thought processes and actions, building physical, intellectual and social resources. Fredrickson (2013) argues that these connections and new resources can be quite specific. For example:
■
experiencing joy tends to make workers get more involved in their tasks, to treat work as
play and to gain skills via experiential learning;
■
experiencing interest can help workers to explore new possibilities and learn, gaining
knowledge;
■
experiencing hope can encourage people to plan for a better future, developing
optimism.
Thus, positive emotions can create a positive spiral in which: ‘positive affect and broad-minded
coping serially enhanced one another: positive emotions initiate upwards spirals towards
enhanced emotional well-being’ (Fredrickson and Joiner, 2002: 172).
Research has shown that relatively small everyday self-initiated experiences can bring
about shifts in people’s overall level of PWB. Lyubomirsky (2007, 2013) identified committing acts of kindness for others, or expressing gratitude or optimism as examples. Grant
and Gino (2010) carried out experiments that showed that when workers were thanked for
their efforts, they experienced greater feelings of self-worth. In turn these feelings predicted
increased work performance. Some argue that counterproductive work behaviours (see Chapter 4) are linked with negative emotions such as resentment and envy (Aquino et al., 2001).
Gratitude is associated with the experience of positive emotions that tend not to co-occur
with feelings of envy and greed and may therefore work as a vaccine against other more negative impulses (Solomon and Flores, 2001).
Well-designed work can provide opportunities for people to experience high levels of
eudemonic well-being. Parker and Bevan (2011) point out it is impossible to think of work as
‘good’ if it appears meaningless to the job holder. The specific elements that make up good
work will differ from one individual to the next but many of the aspects of work design,
opportunities for job-crafting, management and leadership dealt with elsewhere in this book
combine to determine the extent to which any specific job role provides a good basis for
positive psychological health.
Research in positive psychology has also made a major contribution to our understanding
of the benefits that higher levels of psychological well-being confer. Lyubomirsky et al. (2005)
and Ryff (2104) conclude those with higher levels of psychological well-being:
■
are more productive at work and tend to be engaged in healthy jobs;
■
more likely to participate in voluntary work;
246
Work Psychology
■
make more money;
■
are better leaders and negotiators;
■
are more likely to marry and to have fulfilling marriages, and less likely to divorce;
■
have more friends and social support;
■
are more helpful and philanthropic;
■
cope better with stress and trauma.
In the studies that underpin these conclusions, PWB is sometimes a direct predictor of
the outcomes listed, sometimes it is only associated with it. Therefore, the cause-and-effect
relationships are not yet clear. PWB may also act as a protective factor that strengthens or
weakens, or moderates, the effects of other environmental influences on these outcomes.
Some good quality longitudinal research also indicates that there may be significant
implications for physical health (Ryff, 2014). These include lower mortality rates, more
healthy behaviour and better cardiovascular health. Boehm and Kubzansky (2012) concluded
that, ‘Findings suggest that PPWB (positive psychological well-being) protects consistently
against CVD (cardiovascular disease), independently of traditional risk factors and ill-being’
(2012: 655). Chida and Steptoe (2008) reviewed prospective studies looking at mortality and
concluded that ‘positive psychological well-being has a favourable effect on survival in both
healthy and diseased populations’ (2008: 741).
Key learning point
There is evidence that the frequent and intense experience of positive emotions can have implications
for our physical health. This evidence is not yet as extensive as that linking stress and health.
Positive emotions are also linked with work engagement, organisational commitment
and citizenship. Most of the work on engagement emphasises its links with outcomes such
as performance and reluctance to leave (e.g. Halbesleben and Wheeler, 2008). The possible
downsides of engagement could be workaholic tendencies: engagement and workaholism are
correlated but not very strongly, with the latter associated with negative outcomes (Shimazu
et al., 2012). Engagement can also take on extreme, albeit short-lived, forms. Flow was first
described as a psychological concept by Csikszentmihalyi (1990, 1999). Nielsen and Cleal
(2010: 180) summarise it as ‘a state of consciousness where people become totally immersed
in a task and enjoy it immensely’. It tends to occur when people ‘lose themselves’ in highly
skilled tasks, e.g. a musician performing their best concert in front of an enthusiastic audience. Nielsen and Cleal (2010) argue that flow is more likely to occur when people are doing
things they prefer doing, that are well suited to their specialist capabilities and when there is
plentiful feedback and control. Philips (2013) points out that individuals seeking flow should
seek challenges that stretch their skills at least a little.
Key learning point
The concept of engagement provides some clues as to why positive psychological well-being could
influence work performance.
Chapter 8 Work-related stress and well-being
247
Ford et al.’s. (2011) meta-analysis of 111 independent samples showed a moderate
to strong relationship between PWB and independent supervisor / peer ratings of work
performance. Taris and Schreurs (2009) reviewed results from 66 different organisations
and found quite strong relationships between psychological health and client satisfaction. In a meta-analysis of data collected from nearly 8000 separate business units in 36
companies, Harter et al. (2002) found relationships between scores on an employee questionnaire (combining elements of self-reported engagement, satisfaction and PWB) and
business unit level outcomes, including customer satisfaction, productivity, profitability,
employee turnover and absence levels. Robertson et al. (2012) found that higher levels of
psychological well-being strengthened the relationship between employee engagement
and performance.
All this evidence about positive psychology sounds, well, very positive. Some have raised
concerns that new concepts that emerge from positive psychology have not been tested rigorously enough (Hackman, 2009). He also points out there is a danger excessive focusing on the
individual encourages people to try to find the positives in bad situations rather than taking
steps to address organisational-level problems (e.g. by changing the bad situation through
job redesign). Fineman (2006) pointed out that the value ascribed to positive traits can be
contested: he cites the example of how it may be seen to take courage to be a whistle-blower,
but that others may see the same behaviour as treacherous. He also points out that it is not
so easy to de-couple or separate positive and negative emotions (e.g. anger that can also be
energising and exciting). Many of the benefits of concepts may be culturally specific viewed
as positive from a North American point of view, yet different values may be placed on these
benefits in other cultures.
Factors linked to employee stress and well-being
In this section we examine the various working conditions and individual differences that
have been shown to be implicated in work stress and employee well-being. We describe the
nature of each factor and summarise research that provides solid evidence of its importance.
The factors that drive psychological well-being and stress may be very broadly grouped into
those factors that relate to individual differences and factors that relate to the environment.
Individual differences include the attitudes, personality, coping strategies, and skills and
abilities of the individual. Environmental factors include everything that is part of the work
situation. There are so many that we cannot cover them all in detail: we focus on those that
have been most extensively researched using strong research designs and commonly perceived to be stressful or positive. Excellent reviews can be found in several sources (e.g. Bonde,
2008; Bond et al., 2006; Ford et al., 2014; Kivimaki et al., 2006; Ng et al., 2006; O’Driscoll and
Brough, 2010; Podsakoff et al., 2007). It is important to remember that situational factors
do not always cause problems for each worker: much can depend on individual differences.
Therefore, we examine both the potential ‘upsides’ and the potential ‘downsides’ of each of
the factors.
Key learning point
Aspects of the situation and individual differences are both important in determining the impact of
work on individual employees. If there is agreement among a group of employees that an aspect of
work is a problem, then situational factors are more likely than individual differences to be an important source of stress.
248
Work Psychology
Situational factors
Demands
Job content and workload These include working at a fast pace, working intensively, having
to meet frequent tight deadlines under pressure, and extended periods of hard physical work
or concentration with very few breaks or respite. Demands can be cognitive, emotional and
physical. If demands are perceived as excessive by an individual then there is good evidence
that problems tend to follow (Bowling et al., 2015).
Two different types of work overload have been described by researchers. Quantitative
overload refers simply to having too much work to do. Qualitative overload often refers to
work that is perceived by the individual to be too difficult for them (French and Caplan,
1972). Perceptions of workload are used in research more often than objective measures.
High workload can be a symptom of other problems including lack of supervisory feedback and having unclear work roles (Bowling et al., 2015). This means that problems with
workload can sometimes be tackled by adjusting other aspects of work design. Repetitive
routines, boring and under-stimulating work can also be problems as the work environment
does not provide sufficient organisational resources for psychological growth (Bakker and
Sanz-Vergel, 2013).
Emotional labour is relevant in jobs (e.g. customer services) where an individual must
manage their emotions and their responses to the emotions of others (Morris and Feldman,
1996). Emotional dissonance has been found to have links with several measures of employee
well-being (Zapf, 2002). This is when an employee has to use effort to mask or hide their
emotions making their outward emotions insincere. For example, an airline worker may
need to appear sympathetic to the unreasonable demands of an angry passenger. These are
examples of surface acting: deep acting does not involve the faking of emotions (e.g. when
an employee feels sad and displays sadness when hearing about a client’s difficulties). In
their meta-analysis Kenworthy et al. (2014) found that the emotional exhaustion measure
of burnout was reliably associated with surface acting.
Working hours The opening case study illustrates the problems associated with long
working hours. Those working more than 48 hours per week are more likely than those
with lower working hours to report work-related tension, worry or stress (van Wanrooy
et al., 2014). Beyond 40 hours a week time spent working is increasingly unproductive and
increases the long-term risk of significant ill-health. Bannai and Tamakoshi (2014) argue
that those working long hours need more time to recover, but that they are unable to have
sufficient recovery time: a ‘vicious circle’ is established. Working over 40 hours per week is
associated with depression and anxiety, disturbed sleep patterns and coronary heart disease
(Bannai and Tamakoshi). Kivimaki et al. (2015) looked at health data from over 500,000
workers drawn from longitudinal studies across Europe. In comparison to a 35–40-hour
week, working up to 48 hours increased the risk of suffering a stroke by 10 per cent. Those
working over 55 hours had a 33 per cent increased risk. There could be several explanations
for such findings. People who choose to work long hours may be especially driven individuals who are prone to health problems. Their jobs may be demanding in different ways and
include features such as high workload and responsibility that lead to long hours (Burke
and Cooper, 2008).
Sonnentag and Fritz (2015) describe how important recovery processes might be impeded
by long work hours. Time away from work allows psychological strain and negative moods
to dissipate, particularly if a worker engages in pleasurable or relaxing experiences that they
are good at. These experiences support psychological detachment, that is not being involved
in work or work-related tasks and not thinking about the job. Sonnentag and Fritz summarise plenty of evidence that shows people in stressful jobs find it tougher to detach from
work (see Closing Case Study). Psychological detachment appears to be especially important
Chapter 8 Work-related stress and well-being
249
in determining employee fatigue but may be somewhat more difficult to achieve for those
whose jobs provide a high degree of challenge demands (Bennett et al., 2018; van Laethem
et al., 2019). Various researchers suggest that recovery can be improved by involving oneself
in non-work tasks that are very different to those carried out at work and by interacting with
people that one does not encounter at work.
Shift work can have quite a profound impact on physical functioning in addition to
having an impact on sleep patterns, family and social life, mental efficiency and work motivation. Vyas and Garg’s (2012) review found workers engaging in various types of shift work
were more likely to experience heart attacks and ischaemic stroke. Other studies focus on
the more immediate impacts of shift work. Demir et al. (2003) examined the effects on
nurses and found that working night shifts led to exhaustion, loss of energy and ‘detachment’ from patients. These problems were significantly greater than for nurses doing the
same amount and type of work on day shifts. However, research has indicated differences
in well-being outcomes depending on: evening versus day shifts; longer versus shorter shift
patterns; and fixed versus rotating shifts (Ball et al., 2014; Barnes-Farrell et al., 2008; Hoffman and Scott, 2003; Mitchell and Williamson, 2000; Sutherland and Cooper, 1987; Vyas
and Garg, 2012).
Work–life balance considers the interactions (or overlaps) between the work and
non-work domains. Kopelman et al. (1983) described this as ‘inter-role conflict’. This occurs
when the pressures and demands of one role result in the perception of a problematic conflict with the pressures and demands of the other (van Steenbergen et al., 2008). For example,
working late because of an urgent deadline might conflict with an arrangement to socialise
with friends.
Sometimes, however, positive spillover can occur. The skills or knowledge gained in one
domain can positively enhance performance in the other domain (e.g. Grzywacz and Marks,
2000). For example, someone who is required to be very organised at work might use similar
skills in planning their leisure activities. However, the evidence shows that poor work–life
integration appears to be associated with stress and ill-health (Geurts and Demerouti, 2003;
Hurst et al., 2009). Greenhaus and Beutell (1985) suggested that three types of conflict commonly occur:
1
Time-based conflict: Pressures from one domain (e.g. work) make it impossible to meet
demands from the other domain (e.g. home).
2
Strain-based conflict: The impact of engagement in one domain reduces the resources (e.g.
energy) available for the person to meet demands in the other domain. This can produce
exhaustion and irritability (see Sonnentag and Niessen, 2008).
3
Behaviour-based conflict: In this type of conflict the person finds themselves taking on a
different role at work from the one they take on at home (e.g. a hard negotiator at work,
and a sensitive friend or spouse outside of it). The kinds of values, attitudes and behaviours required in one domain may clash with those required in the other. This can lead
to a sense of alienation and/or loss of self-identity.
The impact of family life on work, such as sudden and unexpected family commitments
demanding an employee’s attention, can also be a source of stress (Gutek et al., 1991).
Westman et al. (2008) highlighted this important distinction between the impact of work
on the family (work–family conflict) and the impact of the family on work (family–work
conflict).
For lower income families the quality of work–life balance depends upon the availability
of good childcare support among family and friends (Kossek et al., 2008). For some working
away from home, the return home can be stressful: during a business trip, female workers’
psychological health improved, indicating that they were benefiting from some ‘respite’ from
non-work demands (Westman et al., 2008). It may also be that the role diversity offered by
engaging in a wide range of work and non-work commitments helps to build their personal,
250
Work Psychology
organisational and non-work resources. A person’s appraisal of their situation might be influenced by whether they are made aware that there are positive benefits to be gained from
balancing different work and non-work roles (van Steenbergen et al., 2008).
Key learning point
What happens at work can influence people’s lives outside of work and vice versa. This influence can
be both positive and negative.
The ability to work remotely can provide increased flexibility for workers and support
better efficiency and productivity (Yun et al., 2012). The possible costs include increasingly
blurred work–life boundaries. Kossek et al. (2012) found that some people identify very
closely with their work (‘work warriors’) and that some individuals (‘overwhelmed reactors’) value work and family equally, but have a tendency to react to demands to carry out
work tasks in non-work time and vice versa. Others take a more controlled and proactive
approach to managing the work–non-work boundary. These include ‘high dividers’ who
rarely allow a non-work task to interrupt work and vice versa and ‘fusion lovers’ who take
a proactive approach to managing the extent to which they choose to work in non-work
time. Those who were proactive in their control of the boundary tended to experience less
psychological distress.
Compressed hours/working weeks (e.g. working a small amount of extra time each day
so that employees could have one day off work every two weeks), on-site provision for
childcare, and maternity and paternity benefits have all been used to help employees.
Research suggests that the availability of informal flexible working arrangements is very
important. De Menezes and Kelliher (2011) found that these practices tend to have a reasonably strong positive impact on employee attitudes. Kröll et al. (2017) found moderate effects of flexitime and telecommuting interventions on psychological health. They
argue that such interventions provide employees with organisational resources linked to
autonomy.
Underload, underemployment and unemployment Certain workers, such as pilots, air traffic controllers and nuclear power workers, face a special aspect of work underload: threat-avoidant
vigilant work. They must deal with long periods of time in which they have little to do but
may suddenly be required to spring into action in a crisis. Where artificial intelligence is used
to complete some work tasks workers may become relatively unskilled machine-minders but
who nevertheless must respond quickly if the machine malfunctions (Chase and Karwowski,
2003). Gibbs (2017) points out that advances in technology mean that it can be difficult for
workers to adapt to change as computers quickly gain the ability to carry out more of the
cognitive elements of their role: as more tasks and skills are automated doing the job itself
may become less motivating and rewarding.
In order to secure an income, people may take jobs that provide few organisational
resources and that do not require them to work at full capacity. McKee-Ryan and Harvey
(2011) identify a host of factors that can contribute to a sense of underemployment and
Selenko et al. (2017) indicate that this situation can create a threat to the person’s identity.
Features of underemployment include: being paid less than one would like, working at what
are perceived to be low levels in the organisational hierarchy or working a limited and uncertain number of hours. Others reflect more psychological elements of person – job fit: being in
jobs where skills, education, qualifications or experience are under-utilised, a lack of prestige,
undemanding or monotonous work, or inadequate opportunities for development and progression. If people perceive themselves to be underemployed then various pieces of research
Chapter 8 Work-related stress and well-being
251
show that there is an increased risk of undesirable outcomes such as poor well-being, low job
satisfaction, poor job performance and a lack of discretionary effort.
It does seem that those on contingent contracts report, on average, lower job satisfaction
and lower overall life satisfaction than those on permanent contracts (Körner et al., 2012;
Wilkin, 2015). The differences are not especially large overall, and it is important not to see
such workers as a homogenous group. Wilkin suggests that contingent workers do not always
compare their circumstances to that of permanent workers: some interpret the nature of the
work they do and the rewards they get from it in different ways. Some of the negative health
effects of temporary work are also due to its precarious nature and associated job insecurity
(Benach et al., 2014).
People who experience unemployment tend to suffer from lower levels of personal happiness, life satisfaction, self-esteem and psychological well-being than (i) people who are
employed and (ii) when they themselves were in employment (e.g. Winefield, 1995; Fryer,
1998; Wanberg et al., 2001). Paul and Moser (2009) analysed 323 studies across a wide range
of different countries and concluded that 34 per cent of unemployed people reported psychological health problems of clinical severity, compared to 16 per cent in the employed group.
When people moved back into employment mental health improvements were bigger than
the deteriorations associated with job loss.
Key learning point
Becoming unemployed tends to have a negative impact on a person’s mental health, but the damage
is often repaired when the person finds another job that is at least as satisfying as the one they left
behind.
Studies indicate that long-term unemployment may adversely affect the longevity of the
unemployed by as much as two to three years. People unemployed for 90 days or more who
had experienced mental illness in the past were at significantly higher risk of death by suicide than those who had not experienced such illness (Milner et al., 2014). Of course, some
people experience more than average falls and rises in well-being when their employment
status changes while others experience less. High levels of social support and a high ability to
structure time tend to reduce the impact of unemployment on well-being. Paul and Moser’s
(2009) analysis also revealed important moderators of the relationship between unemployment and well-being:
■
The effects of unemployment were felt most by men, those in blue-collar jobs and by
people who had been unemployed for a long time.
■
Unemployment had more of an effect on mental health in less economically developed countries and those countries with less developed protection systems for the
unemployed.
Control
Control can apply to almost any aspect of work. In their widely used measure of control,
Dwyer and Ganster (1991) include 22 questions relating to different aspects of control. These
include: control over the way work is done and its pace and quality; influence over policies
and procedures; control over leave from work; and control over decision-making. As already
discussed, control appears to be a particularly powerful predictor of employee health. Workers
who experience increases in the opportunities to decide how to use their skills and to make
work-related decisions report significant improvements in their self-reported mental health
252
Work Psychology
(Bentley et al., 2015). This effect was found even after other factors associated with mental
health such as changing employment, age and gender were considered. High levels of control
are also associated with higher levels of job performance and lower levels of absence (Bond
and colleagues, 2006).
The effects of control may also depend on whether an individual prefers to be in control
or not and whether they have the personal psychological resources to utilise it. Meier et al.
(2008) found that high levels of job control helped to protect against the impact of work
stressors on well-being, but only for those people with an internal locus of control. Bond
et al. (2008) found that those with a high level of psychological flexibility benefited the most
from an intervention designed to enhance various aspects of control at work. These individual differences are discussed in more detail later.
Key learning point
Control at work has strong and significant links to individual and organisational health. This makes
changing control the focus of many intervention studies.
Participation in the decision-making process on the part of the individual may help with
the development of several personal and organisational resources. These include an increased
alignment between their own goals and those of the wider organisation, improved communication channels and greater access to social support from managers and colleagues (Nielsen
and Randall, 2012).
Managerial support
Reviews of numerous specific research studies testify to the importance of perceived supervisor support and its relationship with job satisfaction, organisational commitment and
turnover intention (e.g. Ng and Sorenson, 2008). Various authors argue that line manager
behaviour is linked to stress and well-being through two mechanisms (see Donaldson-Feilder
et al., 2008; Skakon et al., 2010). Pleasant or unpleasant manager behaviour can have a direct
effect and set the emotional tone of the workplace (Bakker et al., 2009) or an indirect effect
in determining the working conditions that employees experience. Skakon et al. (2010) argue
that when leaders experience stress they are more likely to behave in ways that result in a negative work environment for their subordinates by engaging in abusive behaviours, monitoring
employees very closely or offering little in the way of support. By contrast, Carlson et al. (2011)
found that managers’ experiences of positive emotions were linked to employees’ perceptions
of autonomy. Management training and performance appraisal can be used to help managers
recognise and manage better the impact of their behaviour on employees (Donaldson-Feilder
et al., 2008). It may also be that by using more participative management approaches, line
managers can increase subordinate workloads and change the nature of the demands placed
upon them. This can even lead to a spike in short-term absence, but benefits in the longer
term when things settle down (Böckerman et al., 2012).
Peer support
Most studies show that high levels of support help to maintain good employee well-being,
but that they are not always strong enough to buffer workers against the effects of other
stressors (Brough et al., 2009). Mistrust of fellow workers can trigger, or arise from, other
issues such as high role ambiguity and poor communications. Well-intentioned attempts to
Chapter 8 Work-related stress and well-being
253
support colleagues can have a negative impact (Beehr et al., 2010). These include: talking to
colleagues about the stressors they are facing as this can focus their attention on problems at
work; offers of help and support might make someone feel they are inadequate; and providing help when it is not wanted because the worker then has an additional difficult demand.
Peer support is particularly important in customer service work (Ng and Sorenson, 2008).
Employees in such roles share similar experiences of stressful interactions with customers
and understand each other’s sources of stress: this helps them to offer more effective practical
help and emotional support.
Key learning point
Lack of support from managers and/or colleagues can place employees at a significantly increased
risk of experiencing work-related stress. Unwanted support can also be a stressor.
Incivility, bullying and harassment
Friction, anger and disagreements between people at work are all well-documented sources
of work stress. Incivility usually refers to rudeness or lack of courtesy to others (Andersson
and Pearson, 1999).
Salin defines bullying as ‘repeated and persistent negative acts towards one or more
individual(s), which involve a perceived power imbalance and create a hostile work environment’
(2003: 1214, emphasis in original). Nielsen and Einarsen (2012) add to this that is difficult for
the target to defend themselves. It can come in the form of being ignored, humiliated or ridiculed, being subject to gossip or rumours, persistent criticism and a host of other behaviours.
Numerous studies have provided data on the prevalence of bullying, with rates between 3.5
per cent and 53 per cent reported (see Coyne, 2011 for a review). Meta-analytic data across 86
studies showed on average reported rates of bullying were 14.6 per cent (Nielsen et al., 2010) –
although differences emerged if studies used a behavioural rating scale versus the use of
self-labelling (with or without a definition of bullying provided).
Research methods in focus
Adapting an existing questionnaire to measure the online work environment
There is growing concern about the impact of online communication on individuals’ well-being. Researchers have
examined whether cyberbullying is different to face-to-face bullying and whether these differences help us to
understand its effects. Bullying and cyberbullying may be qualitatively different (i.e. cyberbullying may have unique
features that are linked to its effects). Interviews provide the flexibility of enquiry needed to identify such features.
Participants in interview studies mention the visibility, to a large group of people, of cyberbullying (West et al.,
2014). Cyberbullying can also reach beyond the workplace: for example, social media is accessed by employees outside of work hours, raising the possibility that cyberbullying can disrupt recovery from work (D’Cruz and
Noronha, 2014). Other features of cyberbullying are that the perpetrator can preserve their anonymity and retain
a physical distance from the victim. Coyne et al. (2017) hypothesised that some of these features could render
cyberbullying more harmful (i.e. it was more difficult to escape, that the bullying had a permanence and that the
physical separation might reduce even further the perpetrator’s empathy for the victim). Coyne and colleagues
254
Work Psychology
were able to adjust an existing quantitative measure of bullying (the Negative Acts Questionnaire: NAQ) to quantify
University workers’ exposure to cyberbullying. However, before doing this they checked that subject matter experts
agreed that the majority of the behaviours measured by the NAQ could be enacted online – but some could not
such as excessive teasing and withholding of information. Adapting existing questionnaires measures for use in
online settings may be challenging, but Coyne et al.’s work shows that it is important to check with experts how
much can be retained. Retaining elements of well-established measures can save a lot of work and help to ensure
that the new measure has good psychometric properties.
Acute bullying is associated with strong negative emotions and the associated consequences for employee health (Keashly, 2001). Naturally, the effects can be even more serious if it is prolonged. At the individual level, impacts are: physical, for example loss of
appetite, sleep problems (e.g. Einarsen, 1999); psychological, for example difficulty concentrating, anxiety, depression, suicidal thoughts (e.g. Mikkelsen and Einarsen, 2001); social,
for example relationships outside of work with friends and family (e.g. Keashly and Jagatic,
2003); and financial, for example loss of employment, inability to secure another position
(e.g. Einarsen and Mikkelsen, 2003). Longitudinal research appears to be especially important in the study of bullying as the impacts on absenteeism and mental health can take
time to emerge (Nielsen and Einarsen, 2012). This may be because the victims do not want
to give the perpetrators an opportunity to bully them about their poor work attendance
(Beale and Hoel, 2011).
Farley et al. (2018) argue that bullying can be the consequence of factors found in the
work environment such as poorly designed jobs, pressure on managers to implement change
quickly and role conflict. It may also be linked to the characteristics of the perpetrators or
targets of bullying behaviour. Various individual characteristics are associated with being
the target of bullying, for example: coping and conflict management skills (Einarsen et al.,
1994); low independence, extroversion and stability; high conscientiousness and achievement (Coyne et al., 2000); high anxiety and low social skills (Zapf, 1999). Perhaps unsurprisingly, few studies collect data from bullies but Zapf and Einarsen (2003) argue that bullying
can occur for three main reasons:
1
It acts as a self-regulatory process to protect or enhance self-esteem, for example when
threatened by a high-performing subordinate.
2
The perpetrator has poor social competencies, for example low emotional control.
3
It is political behaviour to protect the individual’s interests and improve their position
relative to others. There is some evidence that when employees feel low levels of job
security, they are more likely to engage in bullying behaviour (De Cuyper et al., 2009).
Fevre et al. (2011) found that managers and supervisors were most likely to be reported
as the sources of unreasonable treatment, incivility and disrespect at work, but also that
co-workers were often also the source. Overly competitive cultures with excessive workloads
and constant change can create a negative working environment. Autocratic management
styles may be more common when conditions in an organisation are difficult.
Few interventions for reducing bullying and harassment have been thoroughly tested
and this means that there is not sufficient evidence to draw strong conclusions about intervention outcomes (Gillen et al., 2017). The CREW (Civility, Respect and Engagement in the
Workforce) intervention (see Leiter et al., 2011; Osatuke et al., 2009) involves employees
using an educational toolkit and survey data within the context of weekly group meetings
where the group’s strengths and weaknesses regarding workplace civility are discussed.
A clear and consistent understanding of what does and what does not constitute bullying can
Chapter 8 Work-related stress and well-being
255
be communicated using a well-publicised organisational policy and awareness training that
set out the behaviour associated with bullying (Hoel and Beale, 2006). Support mechanisms
need to be available to employees experiencing bullying, whether they are victims, witnesses
or accused of bullying. As discussed earlier in this chapter, the behaviour of line managers
is a key element in tackling workplace bullying, and skills such as communication, conflict
resolution, stress management and team-building could all be associated with effective handling of bullying.
Key learning point
Workplace bullying, harassment and violence at work have particularly strong links to work stress and
negative outcomes for individuals and organisations.
Role at work
Three critical factors – role ambiguity, role conflict and the degree of responsibility for others –
are seen to be major sources of stress (O’Driscoll and Brough, 2010). Glazer and Beehr (2005)
found that across several nations and cultures, role-related stressors in nursing were consistently linked with psychological health and organisational outcomes.
Role ambiguity arises when individuals do not have a clear picture about their work objectives, their co-workers’ expectations of them, and the scope and responsibilities of their job. It
may arise because a supervisor does not clarify to the employee the exact nature of their role.
Other causes include: starting a new job, getting a new boss, the first supervisory responsibility, a poorly defined or unrealistic job description, a change in the structure of the existing
organisation or starting to work in a new team (Beehr, 1995). In their meta-analysis, Tubré
and Collins (2000) found that role ambiguity was significantly linked to low job performance.
When the role is ambiguous, employees are not sure what they need to do to get the job done
and are unable to recognise when they are doing well, thus creating additional workload
(Bowling et al., 2015).
Role conflict exists when an individual’s attention is torn by conflicting, or irreconcilable, job demands or by doing things that they do not really want to do. Workers may often
feel themselves torn between two groups of people who demand different types of behaviour or who believe the job entails different functions. In their meta-analysis, Schmidt et al.
(2014) found that there was a moderate, consistent and reliable link between role conflict
and depression.
Key learning point
If a job role is not clear or carries with it a high degree of responsibility it raises a potential for conflict
with others in the organisation.
Stress as offence to self Some tasks that are given to workers may be seen by workers as illegitimate: their perception is that they should not have to do them (Semmer et al., 2015). The
allocation of such tasks to a worker, or the felt expectation that they should perform them,
may be seen as a sign that they are being disrespected. These tasks can undermine a worker’s
pride in their job, professional identity and self-esteem. Stress as offence to self is experienced
256
Work Psychology
when we are asked to do tasks that are seen as unreasonable (because they fall outside of what
the worker sees as their remit, or as part of someone else’s job) or unnecessary (that it should
not have to be done at all). Workers may feel that such tasks have been unfairly distributed
i.e. being given them represents a distributive injustice. Semmer et al. (2015) studied a varied
working population and found that over a period of two months, exposure to such tasks was
linked to two negative outcomes: employees’ resentment towards their employer and their
self-reported irritability.
Sources of stress discussed in other chapters
Several chapters of this textbook cover in detail aspects of work that have implications for
employee health and well-being. In order to avoid duplication, these are not discussed in
detail here. Brief summaries are used to highlight the key issues.
Leadership (Chapter 10) Leader behaviour can do much to create a positive work climate by setting the ‘emotional thermostat’ at work (Bakker, 2009). Transformational
leadership can significantly reduce the amount of stress experienced by subordinates
because it includes giving priority to the development needs of specific individuals, setting a personal example and establishing a clear mission. Care needs to be taken not to
over-interpret the significance of leadership behaviour. In their study of public sector
employees, Mellor et al. (2009) found only modest or weak links between transformational
leadership behaviour and employee absence. Research shows that there is also a potential
‘dark side’ to inspiring leadership. Employees may be driven to work long hours and sacrifice non-work activities and recovery, coming in to work when they are unwell (Nielsen
and Daniels, 2016).
Organisational change (Chapter 12) Change can often make employees feel insecure about
their future in the organisation (Benach et al., 2014). It is very well established that the threat
of losing one’s job can have a significant impact on psychological well-being (Hellgren and
Sverke, 2003) and work-related attitudes (Sverke et al., 2002). Schreurs et al. (2010) found that
employees who had more control at work were less likely to experience the negative outcomes
of job insecurity. This finding lends support to the view that the way that change is handled
can make a significant difference to the impact of change: giving employees more autonomy
can help to buffer the effects of insecurity. When employees perceive that change is planned,
they have a more positive response to it, and when management offers support through the
change process, employees experience less uncertainty (Rafferty and Griffin, 2006). Similarly,
Swanson and Power (2001) showed that social support played a key role in helping employees
through a major restructuring. The negative effects of major sudden change can be reduced
through open and honest communications with employees, and the provision of ongoing
support (Kawakami et al., 1997).
Organisational culture (Chapter 12) An individual is likely to experience stress if they do
not share the values inherent in the employing organisation. For example, the mismatch
between individual and culture may lead the person to feel isolated and unable to communicate effectively with colleagues. It may mean that the person’s role includes activities that
they find distasteful and that conflict with personal preferences, a form of role conflict.
When employees see that their work activities are consistent with their own values and goals
employee well-being is enhanced (Flaxman and Bond, 2010). Perceived organisational support (see Stamper and Johlke, 2003) is defined as the extent to which employees perceive that
their contributions are valued by their organisation and that the company cares about their
well-being. This can be achieved through increasing the number of shared work activities
such as training, social activities and action planning related to improving things at work
(Daniels et al., 2017).
Chapter 8 Work-related stress and well-being
Exercise 8.2
257
A stressful job?
Look back over the situational factors described in this section. Now think of a job you know something about. It might be a job you have done yourself, or a job that is done by someone you know well.
Suggested exercises
What are the most likely sources of work stress in that job role? Set aside individual differences for a
moment and focus on the features of that job that might place someone doing that job ‘at risk’ of experiencing work-related stress. Then look at the same job in a different way. What aspects of the job are likely
to be sources of satisfaction or positive well-being?
Individual differences
Individual differences have one or more of the following roles in the processes that determine
the impact of work on employee health and well-being:
■
A direct effect on stress outcomes (e.g. anxious people may be more anxious across all
kinds of situations).
■
A moderating effect in the stressor strain relationship. In other words, some people are
more affected than others by an aspect of their work situation. For example, extroverted
people may find a socially isolated job more stressful than introverts.
■
A direct perceptual effect. Individual differences may have some impact on a person’s perceptions of what his or her job is like. For example, people with a high need
for control may be very aware of limitations on their autonomy and rate their work
autonomy low – for most people a similar level of autonomy at work would be seen
as sufficient.
There are many individual differences that are often investigated in stress research. Some
of the most prominent are discussed below.
Personality
The Big Five model of personality (see Chapter 2) has provided a useful summarising
framework for the key personality traits. One of the best-established is that linking neuroticism with poor psychological health. In an extensive meta-analysis, DeNeve and Cooper
(1998) found direct relationships between neuroticism and life satisfaction, happiness
and negative affect. Positive affect was predicted equally well by high extroversion and
agreeableness.
Watson and colleagues (1988) identified two elements of persistent and pervasive
mood that have been particularly important in stress research. High negative affect (NA)
refers to the propensity to experience affective states such as anger, guilt, fear and nervousness. High positive affect is the propensity to feel enthusiastic, energetic and alert.
Many researchers have found these variables to influence directly the perception of sources
and consequences of work stress. Bowling et al. (2015: 97) argue that those with high NA
may be pre-disposed to creating stressful work situations and to perceive as threatening
an innocuous work environment. Those high on conscientiousness may choose goals that
bring with them intense periods of high demands, those low in agreeableness may experience more conflict at work and those high in extraversion may perceive fewer daily hassles
(Vollrath, 2001).
Some people tend to set themselves very high performance standards. Perfectionism can
be driven by different underlying motives. It can be about taking pride in a job well done or
258
Work Psychology
come from a tendency to set excessively high standards and a very self-critical approach when
reviewing performance. Flaxman et al. (2012) looked at how academics with high and low
levels of this self-critical perfectionistic tendency differed in how they recovered from work
during and after an Easter break. Recovery was less evident for those reporting high levels
of self-critical perfectionistic tendencies: they reported more worry and rumination during
time away from work. A tendency to reflect positively on one’s work experiences during time
away from work appears far healthier than a tendency to ruminate about mistakes and errors
(Binnewies et al., 2009).
Psychological flexibility
A key implication of psychological flexibility – and hence its name – is that, in any given situation, people need to be flexible as to the degree to which they base their actions on their
internal events or the contingencies of reinforcement (or punishment) that are present in
that situation.
(Bond et al., 2013: 332)
To simplify somewhat, psychological flexibility is the extent to which a person recognises and
accepts in a non-judgemental way that some of the events going on in their mind – such as
fear, worry and anxiety – are unhelpful thoughts. The research base on psychological flexibility is rapidly expanding and is replete with good studies that show psychological flexibility is
associated with mental health and strong work performance (Bond et al., 2013). When workers are not expending resources wrestling with their emotions, they are in a better position
to pay more attention to the task at hand. This enables them to have a better chance of identifying the best ways of completing tasks, responding to demands as they arise, identifying
and using sources of social support and so on.
Cognitive styles
There are several elements of cognitive style that are important in the stress process. Locus
of control refers to the extent to which a person believes they have control over their life. In
many circumstances, having an internal locus of control (i.e. believing one is in control) is
helpful because it encourages a person to do something about their stressful situation. This
is fine unless there really is not anything a person can do; in which case an internal locus of
control will simply increase his or her frustration. Self-efficacy, a person’s belief that they
are capable (Bandura, 1982), and self-esteem (a person’s feeling of their own worth) are also
important.
Exercise 8.3
Your self-efficacy
Self-efficacy is a measure of a person’s own belief in their capacity to complete a task. It has been
shown to improve when we experience repeated successes and to diminish when we experience
repeated failures. Various pieces of research have shown that high levels of self-efficacy tend to be
healthy and can help people to deal with life’s difficulties.
One interesting feature of self-efficacy is that it can vary from one situation to another. So, after
reading this chapter if you were asked by a friend to describe what work stress is and what causes
it, hopefully your self-efficacy in relation to this task will be higher than it was before you read this
text. But what if you were asked to give a presentation about work stress to 100 senior executives?
Would your self-efficacy be as high in this situation?
Chapter 8 Work-related stress and well-being
259
Below is a measure of general self-efficacy. It is designed to measure a person’s belief in their capacity
to cope with what daily life throws at them. For each question responses are: 1 = not at all true;
2 = hardly true; 3 = moderately true; 4 = exactly true. High scores indicate higher self-efficacy.
In research this scale is often adapted a little by rewording the items to measure self-efficacy in
specified situations.
1
I can always manage to solve difficult problems if I try hard enough.
2
If someone opposes me, I can find the means and ways to get what I want.
3
It is easy for me to stick to my aims and accomplish my goals.
4
I am confident that I could deal efficiently with unexpected events.
5
Thanks to my resourcefulness, I know how to handle unforeseen situations.
6
I can solve most problems if I invest the necessary effort.
7
I can remain calm when facing difficulties because I can rely on my coping abilities.
8
When I am confronted with a problem, I can usually find several solutions.
9
If I am in trouble, I can usually think of a solution.
10
I can usually handle whatever comes my way.
Use the questionnaire to help you consider your own current level of general self-efficacy. Then
consider how your self-efficacy varies from one situation to another in your life. Is it high in situations you find that you tend to deal with effectively, but lower for those you find stressful? The use
of goal-setting theory (see Chapter 6) is one way in which self-efficacy can be developed.
Source: Schwarzer, R. and Jerusalem, M. (1995) ‘Generalised self-efficacy scale’, in J. Weinman, S. Wright, and
M. Johnston (eds), Measures in Health Psychology: A user’s portfolio. Causal and control beliefs (pp. 35–37).
Windsor, UK: NFER-NELSON; http://userpage.fu-berlin.de/~health/engscal.htm
Clusters of individual differences
In recent years researchers have started to become interested in how various individual differences might combine and interact to help individuals deal with stressors. Some researchers
have examined clusters of individual differences that help individuals to flourish and develop
in response to work demands.
The concept of psychological capital (PsyCap) has been developed to encapsulate what are
thought to be the underlying (or latent) human characteristics that explain many positive
outcomes for individuals. Most researchers in positive psychology propose that these can be
significantly developed or eroded over time.
The four different components of PsyCap are hope, optimism, efficacy and resilience
(see Luthans et al., 2008). Hope is will-power and determination combined with plans, or
‘pathways’, or ‘way-power’ for achieving goals. In other words, effort alone is not enough to
sustain hope without a clear mental model of how an outcome will be achieved. Optimism
is about the way people tend to attribute causes to events. A tendency to attribute successful
outcomes to one’s own efforts (internal attribution) linked to enduring capability (stable
attribution) that could help us to succeed in other tasks in the future (global attribution)
would indicate optimism. Negative events would be attributed to external, unstable and
specific causes. Some elements of PsyCap are influenced by traits (e.g. the links between conscientiousness and reported self-efficacy and between neuroticism and pessimistic attributional styles). Carefully structured relatively brief interventions and work activities that allow
people to plan for and to experience success (e.g. goal-setting activities) have been shown
260
Work Psychology
to develop PsyCap (Luthans and Youssef-Morgan, 2017). The results of a meta-analysis conducted by Avey et al. (2011) indicate that developing PsyCap may indeed be worthwhile. They
found links between PsyCap and job satisfaction, organisational commitment, psychological
well-being, organisational citizenship and various measures of performance. They also found
higher levels to be associated with lower stress, turnover intentions and anxiety: evidence of
cause-and-effect was a little more difficult to find.
Core self-evaluations (CSEs) are described by Judge (2009: 58) as ‘a broad, integrative trait
indicated by self-esteem, locus of control, generalised self-efficacy, and (low) neuroticism
(high emotional stability)’. Judge argues that there is enough evidence of strong correlations
between these components to conclude that these variables are all influenced by a single
underlying individual difference factor (CSEs). He cites numerous studies linking CSEs to
low stress and burnout. The jury is still out on whether CSE has the stability of a trait or the
malleability of a psychological state.
Resilience
There remains debate about the nature of resilience, but most researchers agree that in part it
refers to the individual’s capacity to respond positively to failure and set-backs. Some argue
that it extends to the positive appraisal of situations as challenging rather than as threatening. It is also strongly linked to the frequent experience of positive emotions as this helps
individuals to respond positively to failure when it occurs. Many theories of stress propose
that the experience of difficult conditions or circumstances are likely to be damaging, but
researchers have found that carefully managed exposure to adversity can help people draw
upon previously untapped psychological resources (Fletcher and Sarkar, 2013). The current
thinking is that regardless of how it is defined resilience has some connection to stable traits
but that it can be developed (Robertson et al., 2015).
In a wider systematic review of interventions designed to develop resilience, Robertson
et al. (2015) note that there is a wide diversity of intervention practices being followed. This
appears to be at least partly due to a wide range of different definitions of resilience. Intervention programmes include activities designed to develop self-efficacy, optimism, empathy,
ACT (see below), meditation practices designed to develop compassion for self and others,
relaxation, cognitive–behavioural training and biofeedback. The review found some evidence that resilience training did have a positive impact on mental health and self-reported
well-being, but that more research was needed before the most effective interventions strategies could be identified.
Coping strategies
How individuals cope with stress has been the focus of quite a lot of research and practice (see
Zeidner and Endler, 1996). Coping is usually defined as the efforts people make, through their
behaviour and thoughts, to alter their environment and/or manage their emotions. Coping
strategies include analysing the situation, planning a course of action, seeking information
from others, seeking support and comfort from others, relaxation techniques, counselling
and maladaptive coping strategies (e.g. using harmful drugs).
One general distinction is between problem-focused strategies (dealing with the original
cause of stress) and emotion-focused strategies (dealing with how one feels about the stressful situation). Many stress management training programmes are designed to have an effect
by extending an individual’s repertoire of healthy and productive coping strategies thus
helping them to avoid maladaptive coping and make better use of proven coping strategies.
Demographic factors
A wide range of demographic factors such as age, gender and tenure have been considered in
stress research. Relationships between these factors on the one hand, and the perception of
Chapter 8 Work-related stress and well-being
261
stressors and employee well-being on the other hand, tend to vary from one study to another.
Meta-analyses show that older workers tend to cope best with some sources of work stress
such as role ambiguity and that they tend to have lower turnover and absenteeism in the face
of sources of work stress than younger employees (Shirom et al., 2008). This may be because
older workers have had more time and experience to develop job-related skills. Thus, it could
be that tenure, or experience, may be more important than age per se.
Interventions to tackle work stress and promote well-being
There are different levels of interventions that have different objectives and target different
parts of the mechanisms that link work and well-being. Murphy (1988) identified three levels of stress management intervention (SMI): (i) primary, reducing exposure to the sources
of stress; (ii) secondary, the development of resources to mitigate the effects of stress; and
(iii) tertiary, supporting and helping those already significantly impact by their work. This
identifies the working mechanisms and objectives of a wide and varied range of intervention activities. All types of intervention can be delivered to individual employees, made
available to specific well-defined groups of employees, or made available across the whole
organisation. Tetrick and Winslow (2015) also point out that a useful way to think about
interventions is to link them to the JD-R model by identifying whether they are designed to
have an impact on personal, job or non-work demands or resources (or some combination
of these).
Key learning point
Primary intervention focuses on reducing the frequency, severity or intensity of stressors. Secondary
intervention focuses on changing employees’ personal resources and therefore their responses to
stressors. Tertiary interventions focus on rehabilitation. How effective do you think each of these
strategies might be? Which do you think would be a popular choice with (i) employees and (ii) line
managers and senior managers (and why)?
Primary interventions: changing the sources of workplace stress
Primary interventions tackle the sources of work stress or attempt to ‘design into the job’
the sources of positive well-being. Most often they are designed to deal with problems
identified by a significant proportion of employees, targeting a group rather than the
individual employee, i.e. organisational-level interventions. A list of examples of primary
interventions is shown in Table 8.1. In theory, these interventions should reduce the negative effects of stress in the long term as they tackle its source. As yet there is little long-term
evaluation work to either support or contradict this claim (LaMontagne and Keegel, 2012;
Montano et al., 2014; Richardson and Rothstein, 2008). We return to the issues of effectiveness later.
Participatory action research (PAR) is often used in the design of primary interventions. PAR involves employees working with researchers or consultants to identify problems, design solutions and implement solutions, or some combination of these activities
(Abildgaard et al., 2018). These processes seem to: help different stakeholders agree upon
the nature of the problems; provide pathways through which they can contribute to the
262
Work Psychology
Table 8.1
Examples of primary interventions
Tackling problems with job demands
• Job enrichment: the removal or automation of mundane cognitively taxing tasks with the introduction of more interesting
tasks that allow individuals to make better use of their skills and abilities (see Parker, 2014)
• Setting up quick informal meetings to provide timely feedback
• Adjusting staffing levels so that they reflect peaks and troughs in workload
• Increasing variety by, for example, allowing people to rotate around the different tasks carried out in their team
• Setting fixed and protected break times
Tackling problems with control
• Changing the guidance that staff are given about the completion of tasks so that they have more discretion and control
about how they are completed
• Giving staff the control and freedom to identify and rectify common problems quickly and without referring to management; allow teams discretion and control over how workload, roles and responsibilities are allocated (see Chapter 6)
Tackling problems with support
• Encouraging interaction between team members by providing them with tasks that require employees to work together
with shared objectives
• Providing new employees with a proper induction, and ensuring that existing employees have easy access to information
about the support available to them to help them do their job
Relationships
• Allowing staff to ‘rotate’ roles to experience different jobs or tasks within their team or the organisation
• Altering the physical layout of the workspace so that those who depend upon each other’s input can work more closely
together
Role
• Review, update and publicise job descriptions
• Include the active review and clarification of roles in the performance appraisal process
• Minimise the potential for competing or conflicting demands in roles when designing and revising job descriptions
• Reallocate tasks within a team to ensure that conflicting demands do not coincide for individual employees
Source: Includes examples of interventions from Randall, R. and Lewis, R. (2007) ‘Stress management interventions’,
in E. Donaldson-Feilder (ed.), Well-being and Performance. London: CIPD.
design of workable solutions; and can keep them involved in the sustained delivery of the
interventions (Nielsen and Randall, 2012; Nieslen and Noblet, 2018). Proper assessment of
the problems workers experience is a crucial precursor to primary intervention. Without
this assessment a suitable tailored intervention is difficult to design and there is a risk that
changes cause disruption for employees who were not facing significant problems to begin
with (Briner and Walshe, 2015).
Managers may feel that designing these interventions could be disruptive and
time-consuming. Many recent examples show this need not be the case (e.g. Holman and
Axtell, 2016; Busch et al., 2017). Elo et al. (2008) pointed out that in some instances, involving
employees in the design and implementation of primary interventions can lead to positive
Chapter 8 Work-related stress and well-being
263
outcomes because of its impact on perceived control, but it is better if participation leads to
tangible improvements (Holman and Axtell, 2016).
There are relatively few published studies of primary interventions (Kroll et al., 2017;
Tetrick and Winslow, 2015). This may be because it is not seen as the strategy of choice: it
remains a common view that stress is the result of individual weakness. It may also be that
many intervention studies do not meet the high standards of academic rigour required by
publishers of academic journals (Nielsen and Miraglia, 2017).
Research methods in focus
Quasi-experimental designs and primary interventions
In any intervention research it is desirable to compare groups receiving interventions with control groups. This
allows researchers to rule out some alternative explanations for change. For example, without the intervention
might things have improved anyway or might something other than the intervention have caused the effects?
However, randomly allocating workers to primary interventions would involve significantly disrupting established
working relationships and work practices. Therefore, most primary interventions are tested using quasi-experiments:
the intervention is delivered in some parts of the organisation while other parts form control groups. Below are
some examples of this approach.
Susan Jackson (1983) tested whether allowing hospital employees to have more input into decision-making helped
them to feel more involved at work and to develop a clearer understanding of the demands of their role. Regular staff
meetings were introduced. Researchers examined organisational records to check that meetings were happening
and were being used to discuss important and relevant issues. The study included several control groups to help rule
out various alternative explanations for change. For example, some participants completed questionnaires but did
not receive the intervention: this was to test whether being paid attention to as part of the research had an effect.
The meetings led to increased perceived participation and reduced problems with work role conflict and
ambiguity. These changes led to less emotional strain and higher job satisfaction among those involved in the
meetings. Susan Jackson had predicted that changes in communication and social support would be linked to
improvements in employee health. Stress theories do not always allow us to make precise predictions about how
the intervention will work in practice. The reductions in emotional strain were linked to lower levels of absence
and lower turnover intention. However, significant changes were only apparent when the intervention had been
in place for six months. The effects of primary interventions may take time to emerge because they require
employees to adapt to different ways of working. At first, they might even raise stress levels because they create
additional workload or uncertainty. Initial apprehension about change may need to dissipate.
Bond and Bunce (2001) looked at two groups (a control group and an intervention group) employed in a UK
government department. They established a series of problem-solving committee meetings in the intervention
group. These committees, with help from researchers, were tasked with using their expertise of the work setting
to identify interventions that would increase job control. The interventions that were introduced gave employees
more input into decision-making and control over their workload and put systems in place for getting quick advice
from managers about difficult or vague tasks. Employees from the intervention group reported less mental illhealth, lower sickness absence and higher job performance. Changes in perceived control in the intervention
group were linked to these outcomes.
Holman and Axtell (2016) implemented an intervention with several, randomly selected, work teams in a call
centre but were able to have an ‘active control’ group of other teams. Workers in these ‘control group’ teams
completed the various measures used in the study and received feedback on the results but did not take part in
the problem-solving element of the intervention. With the help of the researchers, employees in the intervention
group worked in small groups to devise changes that could be implemented to improve both performance
and employee well-being. These changes included: taking over administrative tasks from team leaders, being
264
Work Psychology
able to make more decisions about how complaints should be processed and taking over the running of team
meetings to increase control; and refining job performance criteria and taking responsibility for collating
performance data to improve feedback on performance. Proposed changes were presented to management,
agreed and implemented, with follow-up data collected one month after all the changes had been implemented.
Perceptions of job control improved in the intervention group but worsened in the control group. Perceptions
of feedback remained stable in the intervention group but decreased in the control group. These changes in
job characteristics accounted for the changes in well-being and performance. To be effective the intervention
needed to lead to changes in workers’ perceptions of features of their work.
A final example (Busch et al., 2017) demonstrates the importance of intervention delivery mechanisms. Their
intervention focused on low skilled immigrant workers, a group who found it difficult to identify and obtain support
at work. Peer mentors were trained to act as role models and to provide advice and support to other workers. They
were paid to attend three 4-hour workshops where they were trained in active listening techniques and emotional
support. Different divisions in the company were randomly allocated to a waiting-list control group or the
intervention group. To ensure that the intervention ‘reached’ employees, the researchers asked line managers and
co-workers to nominate peer mentors to attend the training, i.e. those that were well-known, trusted and whose
support was likely to be accessible and sought by the workers. The impact of the intervention was demonstrated
through a significant reduction in the mean diastolic blood pressure of the workers who had access to peermentor support. Interviews with participants were used in additional to quantitative data analysis to unpack the
mechanisms of change. These indicated that workers had benefitted because peer-mentors provided them with
support, helped them to improve their working conditions, and provided them with a role model.
Interventions can also be used to encourage employees to be proactive in designing
and implementing their own individual-level primary interventions. van den Heuvel et al.
(2015) used a one-day training intervention to provide participants with information about
job-crafting and the possibilities it might offer. Participants then examined their own work
situation and made plans for how they could craft their own environments over the coming
month to better manage demands and to identify, obtain and protect resources. Participants
then went away to implement their plans and returned to the training environment to
report on their progress. There was some initial evidence that this compact intervention
had good results.
Key learning point
Primary interventions could be effective in the long term – but more evidence is needed for these
effects. Employee involvement in the design and delivery of interventions can help to ensure that the
intervention addresses their problems.
Secondary interventions
Most workplace stress initiatives have been directed at changing employees’ psychological
resources (to change or build their capacity), thus enabling them to make different, more productive, responses to difficult or stressful elements of their work. Inherent in this secondary
intervention approach is the notion that the working environment may not be easily changed.
Therefore, the individual must adjust their capacity to meet the demands placed on them.
Through these changes the links between exposure to sources of stress and its negative
outcomes can become weaker or disappear. The objective is to reduce or eliminate the harm
Chapter 8 Work-related stress and well-being
265
that employees might experience. The way employees appraise their situation is crucial in
some stress theories: these interventions are often designed to help employees develop the
habit of appraising things more positively. A subtle potential risk of the widespread use of
secondary interventions is that employees may conclude that they are somehow to blame for
the problems they are experiencing. However, COR theory suggests that the development of
personal resources can be beneficial even in the absence of stress as employees are motivated
to ‘stockpile’ resources they can use later at work or elsewhere.
Specialist expertise is often needed for these interventions because the quality of the
delivery has been found to play an important part in determining their success (Hofmann
et al., 2012). Intervention is usually completed relatively quickly, with employees attending
several short training sessions and practising their new skills between these sessions (interventions often last a few weeks). Stress management training (SMT) is the most common
form of secondary intervention. Generic skills are often developed through these sessions so
that employees can use them to deal better with lots of different sources of stress. Cognitive
behavioural interventions directly target the way employees think about their work situation and the links between their perceptions, emotions and behaviour. For this training to
be effective, trainees need to work quite hard at developing competencies that allow them
to reap the hypothesized benefits, namely to ‘gain knowledge and control, and expend less
energy and time in their response to a potentially stressful situation’ (Kroll et al., 2017: 679).
Mindfulness training has received much recent attention. Mindfulness is about paying
attention to thoughts in a non-judgemental and compassionate way to facilitate keeping
one’s attention on the present moment. It is a meta-cognitive intervention in that individuals
are encouraged to notice their thinking without trying to control it. It has attracted attention
from work psychologists because there is some evidence that the development of mindfulness
has positive outcomes for employee performance, well-being and work relationships (Good
et al., 2016). These effects are thought to accrue because mindfulness helps individuals to have
richer experiences because they notice more of what is going on and because it breaks destructive thought patterns, disrupting the link between previous and present experiences. There is
some evidence that mindfulness practice is associated with structural changes in regions of
the brain associated with attention. Mindfulness-based stress reduction (MBSR) interventions
can be delivered over a relatively short time period (typically 8 weeks). Individuals practise
specific exercises such as focusing on their breathing while noticing, with curiosity, any intrusive thoughts before returning their focus to their breathing. While not all intervention studies allow us to draw strong conclusions about its effects, Lomas et al. (2017) and Virgili (2015)
have reviewed the evidence and found relatively consistent moderate effects on employee
well-being. A note of caution has been sounded by Moody et al. (2013). Some employees
who are highly stressed may not have the psychological resources to engage in demanding
mindfulness practices and the intervention may not be a good fit for them.
Acceptance and commitment therapy (ACT; see Flaxman and Bond, 2010) is designed
to enhance psychological flexibility. It provides trainees with techniques helping them to:
(i) be less focused on internal psychological experiences in order to ii) be more in contact
with the present moment and to place more of their resources and attention on actions that
are congruent with their values and goals. ACT uses a package of specific techniques to help
people direct their attention away from controlling thoughts and emotions (acceptance) to
make more personal resources available for committed action (commitment). This contrasts
with some other approaches to intervention that require participants to engage with their
negative emotions in order to manage them. Like mindfulness, these interventions can have
meditative elements to encourage contact with the present moment. Participants are also
provided with ways of reducing their tendency to allocate cognitive resources to ‘wrestling’
with their emotions. This can include practising noticing automatic thoughts (such as ‘I
can’t cope’), accepting them, but not seeing them as true and therefore as barriers to taking
action. Parts of the training can involve the identification of participants’ values: this is
thought to help participants develop increasing commitment to actions that are congruent
with their values.
266
Work Psychology
Cognitive behavioural training (CBT) is based on the concept that unhealthy human
behaviour and the experience of negative emotions can be reduced by changes in cognition/
appraisal. It involves training people to think differently about their experiences, or to give
different meanings to events. To illustrate this point, imagine you are awoken by a noise in
the middle of the night. If you think that noise is a burglar, then you would be understandably anxious. If you heard the same noise but thought it was the neighbour’s cat jumping from
a fence, then you might be somewhat less anxious. Workplace CBT focuses on changing the
way people perceive and attach meaning to experiences. Usually this training will start by
targeting situations that the employee has identified as being particularly stressful. In some
training the employee develops the skill of attaching more positive meanings to events. This
relabelling might involve training an employee to think of a difficult situation as a chance to
deal with a challenge rather than as a threat or problem to be coped with.
Stress Inoculation Training (Meichenbaum, 1996) involves focused practice of healthy
ways of thinking in response to stressful work events. Employees are encouraged to practise new strategies for thinking about and responding to stressors in response to carefully
controlled simulations that get progressively more difficult. They work towards using their
strategies in situations that resemble more closely those stressful situations faced in the work
environment. Fletcher and Sarkar (2013) suggest that these controlled experiences of adversity can help employees to develop psychological resources that make them more resilient.
Key learning point
There are several different approaches to secondary intervention. They work in different ways by
focusing on different psychological processes.
Relaxation training is designed to train the individual to recognise when the body is
becoming tense and then to think or behave in a way that relieves that tension. Physical
relaxation techniques include deep breathing exercises, muscle relaxation and stretching.
They are designed to help dissipate negative emotions and unpleasant sensations such as
muscle tension. These methods are relatively cheap and popular with organisations. Cognitively based relaxation techniques use imagery and meditation to clear the mind of external
thoughts relating to life events. Biofeedback is sometimes used in conjunction with relaxation training. This uses measurement devices (e.g. heart rate or blood pressure monitors) to
show people their physiological responses to stress. This allows people to see if they are using
the relaxation methods effectively enough to make a tangible difference.
Other possible stress management interventions include job-related skills training on
topics such as time management and assertiveness (see Speed et al., 2018). These can help
employees to deal with some of the most difficult and stressful aspects of their job role. Training to carry out core components of the job can have a significant and positive impact on
personal resources. For example, time management training can be used to help employees
make better decisions about how to organise their work activities, because time management
is ‘a form of decision-making used by individuals to structure, protect and adapt their time
to changing conditions’ (Aeon and Aguinis, 2017: 311). The working mechanism is similar
to that of a primary intervention: it works by helping employees to reduce their exposure to
intense periods of work or short deadlines. These interventions are often overlooked but there
are examples showing that some can be effective (Watson et al., 2017).
The packaging up of several secondary interventions is quite common in stress management training. Richardson and Rothstein’s (2008) review of intervention research indicated
this might not be a good idea: different intervention components might interfere with each
other or just give the employee too much to learn and practice. Tetrick and Winslow (2015)
and Monatno et al. (2014) highlight several examples of primary and secondary interventions
Chapter 8 Work-related stress and well-being
267
being combined with each other to form multimodal interventions. Simultaneously enhancing resources and tackling demands could be an effective intervention strategy. Secondary
interventions can be used to deal with the immediate risks to employee well-being while the
processes required to design and deliver primary intervention are enacted.
Key learning point
The various secondary interventions work in different ways. They are designed to weaken, or break,
the links between exposure to sources of stress and problems with well-being.
Tertiary interventions
The aim of tertiary interventions is to help to rehabilitate those who have already been
damaged by their work. Tertiary interventions are particularly important when primary
and secondary interventions are impractical or when they are unlikely to be totally effective for every employee. The original employee assistance programmes (EAPs) were designed
to help employees who were suffering from problems of alcohol dependency, but they are
now designed to help employees who are experiencing any one of several different problems.
Most EAPs offer psychological counselling through self and line manager referral. These referral routes may sometimes lead to individual-level primary or secondary interventions: an
employee may work with the line manager and their counsellor to redesign their workload
or may be referred to stress management or job skills training.
EAPs usually include a telephone-based helpline or Internet-based and information service.
This is often used for assessing individual employees’ needs, referring employees to sources of
help and sometimes for short-term counselling. EAPs also provide employees with resources
matched to non-work demands (e.g. a legal advice helpline), critical incident de-briefing and
coaching (Joseph et al., 2018). Health promotion activities such as on-site fitness facilities,
dietary control, cardiovascular fitness programmes, relaxation classes or stress and health
education are another common feature (Kinder et al., 2008).
Key learning point
Tertiary interventions are often designed to repair damage but contain some features that allow
employees to develop personal resources even if they are not currently experiencing problems. There
is widespread agreement that they should not be seen as a substitute for primary and/or secondary
interventions.
The popularity and effectiveness of interventions
Establishing and maintaining good research designs to evaluate interventions is challenging.
From a management perspective, exposing as many people as possible, as quickly as possible,
to an effective intervention is a likely priority. This often prevents researchers isolating the
effects of the intervention. For practical reasons discussed earlier, primary interventions have
been far less widely used than stress management training (LaMontagne et al., 2007; Tetrick and Winslow, 2015). In an extensive review Richardson and Rothstein found cognitive–
behavioural secondary interventions without other secondary interventions had the largest
effects and most consistent effect on employee well-being (relaxation interventions were the
268
Work Psychology
second most effective). In a more recent review, relaxation techniques were found to have the
largest effect size with the authors arguing that this may be because such techniques are the
easiest for employees to learn and to implement (Kroll et al., 2017).
Key learning point
Studies of intervention effectiveness indicate that, on average, secondary stress management interventions with a cognitive–behavioural element and relaxation interventions are the most effective.
Care should be taken when interpreting these findings because the paucity of research regarding other
types of intervention makes rigorous analysis difficult.
Montano et al. (2014) examined 39 good-quality primary intervention studies. The most
striking finding of their review was that when different primary interventions were delivered
in combinations the effects were larger than for single isolated interventions. They looked at
various combinations of ergonomic improvements, intervention to tackle work pressure and
intensity, and other improvements such as increasing levels of social support. Combining different primary interventions may increase their effects because different aspects of demands
and resource interact with each other in the way that models such as the JD-R describe.
Key debate
How do we know that a stress management intervention is effective?
The choice of criteria used to measure intervention outcomes might explain at least some of the differences in intervention effectiveness. Symptoms of anxiety and depression are often targeted directly in secondary interventions, and
it is the measures of these psychological variables that are also then used to determine intervention effectiveness.
Secondary and tertiary intervention effects may be short-lived if employees return to an unchanged and unsatisfactory
work situation. There is a small, but growing, number of studies showing primary interventions can have a significant
impact on outcomes such as absence and performance. These outcome measures can be influenced by many organisational factors unrelated to the intervention.
This line of reasoning is supported by evidence that changes in self-report measures are larger than changes in
other outcomes for most primary intervention studies. Just as secondary interventions tend to produce changes in
psychological symptoms, Nielsen and Miraglia (2017) have argued it is only reasonable to expect primary interventions
to have consistent impact on the problems they target, that is perceived working conditions. Changes in working conditions do not guarantee changes in well-being. Individual differences and coping can influence the way employees perceive change and determine the strength of the relationship between changes in working conditions and health. Those
already coping well may experience less benefit from a primary intervention than those who are currently struggling.
Some working conditions have a particularly strong link to organisational outcomes. Bond and colleagues (2006)
found that even modest interventions to improve control tend to have an appreciable impact on job performance and
that solving problems with role ambiguity and role conflict were more likely to impact on staff turnover. Therefore,
it is difficult to provide a simple answer to the question, ‘Are primary interventions effective?’; certainly they can be,
but much depends upon the nature of the intervention, the characteristics of those receiving it and the criteria used
to evaluate it.
Problems implementing interventions or disruptive events occurring in the organisation
can alter the effects of an intervention. Few studies use a rigorous approach to documenting
intervention processes and contexts to examine how these factors might have influenced
Chapter 8 Work-related stress and well-being
269
intervention outcomes (Egan et al., 2008; Havermans et al., 2016). Process evaluation is designed
to help the researcher answer questions such as ‘Why did the intervention work well?’ and ‘Why
did the intervention fail?’ (Nielsen and Randall, 2013). Research into secondary intervention
has shown that session impact factors such as the sense of comfort and belonging generated in
training sessions can significantly influence intervention outcomes (Bunce, 1997). For primary
interventions, process evaluation involves documenting the delivery of the intervention (implementation factors) and significant events taking place in the organisation, the contextual factors. Problems with maintaining staffing levels can make it difficult for people to find the time,
energy or resources for primary intervention activities (Nielsen et al., 2010; Nielsen and Randall,
2012). Process factors important in the implementation of primary interventions include:
■
good levels of employee involvement and participation in all aspects of the intervention
process, with senior management commitment to primary intervention and a culture
supportive of positive change (e.g. Abildgaard et al., 2018; Nielsen and Randall, 2013);
■
actions being taken to provide employees with the skills, resources and support they
need to extract the maximum possible benefit from the intervention (Nytrø et al., 2000;
Nielsen and Noblet, 2018);
■
employees being ready for change, line management support for the intervention, and
employee perceptions participation in intervention design; and exposure to the active
ingredients of the intervention, i.e. whether they are aware of, or exposed to, the elements of the intervention designed to improve their situation (Randall et al., 2009);
■
employees’ perceptions of the quality of the intervention and its likely success (Fridrich
et al., 2016) and the amount of information they are given about the intervention, such
as why it is being implemented and what it is intended to achieve (Nielsen et al., 2007);
■
shared common views among those involved about the nature of the problems and the
value of different intervention alternatives (Hasson et al., 2016).
Key learning point
The processes through which interventions are designed and implemented, and the context in which
they are delivered, play an important role in determining whether or not they are effective.
EAPs have proved more popular with organisations than other types of intervention. There
is some persuasive cost-benefit analysis of these interventions. Managers may consider it to be
easier, less disruptive to business, and less costly to change the individual than to change work
practices. The presence of an EAP indicates that ‘something is being done’ about stress. This
is important, given the increasing litigation fears that now exist among employers in many
countries. Increasingly, organisations are using EAPs as safety nets to deal with the minority
of problems where primary or secondary interventions do not deal with the problem.
Longitudinal evaluation data are not publicly available for all EAPs and the nature of the
intervention rarely facilitates the use of control groups in evaluation studies. Joseph et al.
(2018) reviewed the best evidence available and found that counselling services offered by
providers external to the organisation were linked to reduced presenteeism, with there being
mixed results for absenteeism. Elements of EAPs that resemble wellness programmes (i.e. the
provision of fitness facilities and health education) appear to significantly decrease absence
and increase job satisfaction (Parks and Steelman, 2008). Such benefits may be relatively short
term, particularly if individuals fail to maintain a long-term commitment to exercise habits
and are likely to revert to their previous lifestyle. Many EAPs also offer employees secondary interventions therefore similar concerns have been voiced that EAP interventions offer
short-term gains for the individual but have less impact on job satisfaction and performance.
270
Work Psychology
Drawing on reviews by Murphy (1988) and French et al. (1997), Randall and Lewis (2007)
identify that there are very significant variations in EAP provision that may impact on their
effectiveness. EAPs are often underused by employees who are not aware of the services on
offer, or who have concerns about confidentiality. It seems feasible that the following may
impact on EAP outcomes: visible commitment and support for the EAP from senior management; clear, well-publicised and written agreements on the purpose, policies and procedures
of the EAP training of line managers to identify problems; frequent communication and training for employees concerning EAP services; and programmes that offer long-term ongoing
support and referral to other agencies if needed.
Problem-solving approaches to intervention
A ‘one size fits all’ approach to intervention is not yet available nor might it ever be. What is
clear is that a process is needed to identify the problems employees are experiencing and to
design or choose appropriate interventions.
Stop to consider
Organisations may often be faced with a choice between primary, secondary and tertiary interventions.
What factors should be considered when they are making that choice? What, if anything, could prevent
them from using their preferred intervention strategy?
The complexity of employee well-being often leaves organisations confused about what to
do to tackle any problems they find. Many experts recommend that each organisation follows
a systematic problem-solving approach to the management of employee well-being. Nielsen
and Noblet (2018) identify five important stress management intervention phases that have
been used in a range of different countries and organisations. The specific phases are:
1 Initiation: setting up a steering group containing appropriately qualified, knowledgeable, well-known and
respected individuals to guide the project; preparing employees for change by developing their knowledge
and skills; and communicating the aims, objectives, timescales and activities associated with the process. It is
important to communicate to employees that the problem-solving process is an important part of the organisation’s agenda.
2 Screening: Systems should be in place to monitor trends in employee turnover and absence data, and more
subtle indices such as error and accident rates, tardiness, dips in job satisfaction and deteriorating industrial
relations. An employee survey can be used to assess and monitor employee health and well-being and to
point to the need for intervention. It is likely that different stressors operate in different parts of the organisation for different groups of employees. By using various data analysis methods, it is possible to establish
whether problems with working conditions are linked to poor employee well-being.
3 Action planning. From an ethical perspective, decision-makers should look carefully at the options for primary
interventions as the preferred option to deal with the sources of stress identified. Involving various stakeholders in the intervention decision-making process is crucial: this allows the practicalities of various intervention
options to be examined properly and for an intervention action plan to be devised (Abildgaard et al., 2018).
4 Implementation. Action plans are enacted by those tasked with responsibilities during action planning. Line
managers appear to be particularly crucial at this stage as they either deliver directly interventions components or provide resources and support for those tasked with implementation (Nielsen et al., 2017b).
5 Evaluation. Measuring again working conditions and employee well-being can show whether interventions
are having the desired effect. Evaluation should be used as the stimulus for new efforts to tackle any problems that remain, or any new problems that have emerged. This sets up a cycle of continuous improvement.
Process evaluation should also be carried out to ensure that the action plans are being implemented, experienced by employees and not disrupted (Havermans et al., 2016; Nielsen and Randall, 2013). Process evaluation can be carried out during implementation so that any problems with the intervention can be fixed.
Chapter 8 Work-related stress and well-being
271
Key learning point
Managing stress and well-being at work requires a stage-by-stage approach – identifying a problem,
intervening to change it or find ways of coping with it, and monitoring and reviewing progress.
Summary
Stress is unlikely to move off the agenda as international competition increases and organisations are faced with tougher market conditions. Work psychologists are in the fortunate
position that there is a lot of good theory available. The risks to employee well-being are
well-known, and research provides many good insights into how the positive effects of
work can be realised. An important point made throughout this chapter is that different
jobs have different sources of stress that impact on different people in different ways.
In fact, well-designed challenging work can do much to foster positive well-being.
Many primary interventions are manifestations of good management and good job
design. Involving employees in decision-making about interventions is important. Training employees to help them deal with unavoidable challenges at work also appears to have
good outcomes, but such training must be evidence-based and delivered by competent
practitioners. Some employees may still slip through the net and tertiary interventions
can help to repair the damage done.
The assessment and management of employee well-being are not simple tasks. There
are many factors to consider and these interact with each other in complex ways. However,
enough is already known for work psychologists to offer clear advice to organisations about
the processes that they should follow to help them develop a healthy workforce within
successful organisations.
Closing case study
Social networking and the increasing
burden on mental health
Barely a week passes without another study
raising concerns about our compulsive use
of smartphones and social media. Ofcom, the
UK communications regulator, found that
two-fifths of adults looked at their phones
within five minutes of waking up – rising to
almost two-thirds for those under 35. More
than half said their devices interrupted faceto-face conversations with friends and family.
The Pew Research Center, a US think-tank,
reported that a quarter of American adults
were ‘almost constantly’ online. About half
of respondents to the latest edition of a
long-running UK survey from the consultancy
Deloitte admitted they had a constant need
to check their phone. So how is this affecting business life? There is now widespread
acceptance that the burden of emails, which
continue to flood into inboxes even after
individuals have finished work, could have
negative effects on mental health. Indeed, it
was the mass adoption of email on mobiles
that began the modern era of ‘technostress,’
says Cary Cooper, professor of organisational
psychology and health at Manchester Business School. Cooper co-chairs the National
Forum for Health and Wellbeing at Work,
which brings together large employers such
as BT, BP and Shell with government, health
and social media representatives to ‘reduce
the stress levels and increase productivity of
272
Work Psychology
UK plc’. Constantly checking our devices is
‘causing people to be overloaded,’ he says. ‘It
diverts them away from their job and engages
them when they should be resting at night,
at weekends, and they’re not interacting
with their families as much.’ Paul McLaren,
a psychiatrist at the Priory Hospital in Kent,
concurs. ‘The ability to access email from
anywhere, from the toilet to the top of a
mountain, makes it hard for people who
struggle with boundaries. Weekend ‘down
time’ and holidays get invaded by the incoming flow. The flashing light on the phone or
incoming alert can become so integral to our
sensory experience that we can feel discomfort without them.’
Richard MacKinnon, an occupational psychologist and managing director of WorkLifePsych, which provides occupational
psychology support, takes a similar view.
‘Executives I work with list email as one of
their big stress points at work. They’re there
to create and implement strategy and yet here
they are, at eight, nine, 10 o’clock at night
emptying their inbox, only for it to refill the
next day.’ France has recognised the problem,
introducing a law guaranteeing employees the
right to ‘disconnect’ after they leave the workplace. And at corporate level, some big German companies have made attempts to tackle
the overload. Daimler automatically deletes
employees’ emails while they are on vacation
and Volkswagen blocks them out of hours –
although they are in workers’ inbox when
they return to work. Legislation is, however,
not the answer, argues MacKinnon. ‘As soon
as you create a law, the first thing people do is
look for a loophole. It’s like whack-a-mole,’ he
says. ‘Strict rules also prevent the flexibility
that some workers prefer’, says Cooper.
Cooper’s forum has instead produced guidelines to help companies instil a balance
between wellbeing and the benefits of being
connected. These include respecting colleagues’ rest-time, clear leadership from
management in setting behavioural norms,
discouraging ‘reply-all’ emails, and regular reminders of the power of switching off
through, for example, ‘no-email Fridays’.
Change must come from the top, says David
D’Souza, membership director for the Chartered Institute of Personnel and Development.
He says: ‘Leaders need to be aware of the fact
that responding to an email at 10 o’clock on a
Saturday night sends out a clear message to
the organisation: the most senior people here,
the people who are most successful, are the
ones that just don’t stop. That can have severe
health ramifications.’ The flashing light on the
phone can become so integral that we can feel
discomfort without it McLaren adds: ‘Most
employers will have policies which discourage the use of personal or company devices
for personal use during work hours but, if
the addiction is strong enough, then people
will still do it. Most people will glance at text
messages or Facebook during the working day,
and the line between personal communication
time and work time becomes blurred.’
These blurred lines may be one reason that
compulsive use of social media has yet to
show up as a big issue for employers. Business worries have until now focused on the
risks to a company’s reputation from a misjudged tweet or status update rather than an
employee’s mental health. Although not yet
considered to be as harmful as addictions to
online gambling, gaming or pornography,
concern is nevertheless growing about social
media’s effect on our working lives.
Susan Hepburn, a London-based therapist
and addiction expert, says there has been a
‘significant rise in individuals coming to me
regarding addictions to social media – especially over the past eight to 10 years. Some
have even lost their job as a result’. The
problem is almost certainly under-reported.
‘Social media is so ingrained now in most
people’s lives that they would probably not
consider mentioning it,’ says McLaren. But
while tools such as Apple’s Screen Time feature – which allows users to set limits on how
much time they spend on apps – acknowledge
the problem of excessive use, MacKinnon and
D’Souza think social media is unlikely to be
responsible for falling productivity.
‘People using Facebook at their workstation
is obviously a time-drainer but, historically,
we all found ways to waste time at work,’
says D’Souza. Hepburn, however, thinks that
because social media usage is still below most
companies’ radars, it could become a problem.
‘Many employers do not implement strict or
clear rules regarding the use of social media
at work and so with the absence of a threat
or a severe consequence, many [employees]
believe it is OK to have a quick glance at their
social media channels. Unfortunately, this
can then spiral into a more severe addiction.’
Chapter 8 Work-related stress and well-being
Historically, we all found ways to waste time
at work. Social media usage at work would be
very difficult to control. Companies increasingly use the same tools for business, such as
Twitter and Facebook, as their workers do for
social reasons, and those in sectors such as
marketing are actively encouraged to develop
their profiles. But employers need to be alert
to potential problems among their workers,
says McLaren. ‘Look at productivity. Is their
performance falling off? Are they doing what
they need to? Are they realising their potential? Managers can see phone use and texting
in open plan offices. Is someone distracted
because they are getting ‘buzzed’ or ‘bleeped’
throughout the working day? Are they paying more attention to their phone than the
work in front of them?’ Companies should
be explicit, he says. ‘Tell employees when to
and when not to access email. Tell employees
to protect their downtime. Make sure senior
managers set an example. As with other mental health issues, it’s about promoting awareness and making sure that if employees get
into difficulties then they can ask for help.’
Hepburn suggests simple solutions including
making sure employees have ‘offline periods
throughout the day, whether through organ-
273
ised team meetings, group lunches or providing lunchtime activities such as yoga – to
get employees to switch off completely and
take a step back from the continual stream
of online media.’ However, the bigger issue
is that work practices have yet to adjust to
a new age, says Lesley Giles, director of the
Work Foundation employment think-tank.
‘Technologies are such a disruptive force in
workplaces. They’re flipping on the head the
way we do certain things,’ she says. ‘The challenge for businesses is that they really have to
embrace that change and go with it because
the boundaries are blurring between what is
work and what isn’t work.’ Both MacKinnon
and Cooper think that if employers notice
staff are using social media excessively, then
work practices need to be examined. ‘It could
be boredom, it could be anxiety, it could be
uncertainty of what to do next, and it could
simply be a feeling of being overwhelmed by
workload,’ says MacKinnon. ‘What’s making
employees want to use social media during
work time?’ asks Cooper. ‘There is a reason for it. Maybe they’re not stretched, not
properly managed, not given adequate objectives. So, don’t blame the social media; try to
unwrap why they’re doing it’
Source: Darren Dodd (2019), Social networking and the increasing burden on mental health FT.com,
17 January © Financial Times Limited 2019. All rights reserved.
Suggested exercises
How do theories and models of work-related stress help us to understand the causes and consequences of ‘over-use’ of technology described in this case study? Several remedial interventions
are suggested by the experts interviewed in this article. Identify at least two and classify them as
primary, secondary or tertiary. How likely do you think it would be that senior managers would
support and implement these interventions? Why? Would all employees like these interventions? If
they were implemented, how would you measure intervention effects?
Individual and group discussion questions
1 What are the major sources of stress at work? Why are they stressful?
2 Compare and contrast transactional, structural and resources-based approaches to
work stress.
3 Is the concept of stress a useful one for those wishing to improve employee well-being?
4 Is it appropriate to describe stress as an illness? Explain your answer with reference to
theory and research on work stress.
5 Might individual-level stress management programmes be more or less effective than
organisation-orientated interventions? Explain your answer.
6 ‘The way a stress management intervention is implemented is just as important as the
content of that intervention.’ Discuss.
274
Work Psychology
Relevant websites
The work on line management behaviour and work stress that is part sponsored by the
CIPD can be found at http://www.cipd.co.uk/hr-resources/guides/line-management-behaviour-stress.aspx and https://www.cipd.co.uk/knowledge/culture/well-being
Several researchers in positive psychology provide extensive resources on their websites:
these include Martin Seligman https://www.authentichappiness.sas.upenn.edu/home,
Carol Ryff http://aging.wisc.edu/research/affil.php?Ident=55, Sonia Lyubomirsky http://
sonjalyubomirsky.com/ and Barbara Fredrickson http://peplab.web.unc.edu/
The European Agency for Safety and Health at Work provides many useful resources regarding the assessment and management of work-related stress at https://osha.europa.eu/en/
themes/psychosocial-risks-and-stress. Their e-guide to managing stress and psychosocial
risks is especially useful: https://osha.europa.eu/en/tools-and-publications/e-guide-managing-stress-and-psychosocial-risks
A host of resources, including the Management Standards Indicator Tool and case studies of
stress management interventions can be found at the UK HSE website: http://www.hse.gov.
uk/stress/. A full research report on an interesting case study from Somerset County Council is available to download in full from the HSE website: www.hse.gov.uk/research/rrhtm/
rr295.htm.
The British Association of Counselling and Psychotherapy (BACP) has a Workplace Division
that can be found at http://bacpworkplace.org.uk/. A selection of several interesting articles from their Counselling at Work journal are available from here free of charge. Many focus on the use of counselling to tackle contemporary work issues. They have also published
an in-depth review of counselling in the workplace: https://www.bacp.co.uk/media/1974/
bacp-counselling-in-workplace-systematic-review.pdf
Practical concern with how to manage (and/or reduce) stress at work is always high and
seems to be growing. One manifestation of that is the International Stress Management
Association, which can be found at https://isma.org.uk/. This site offers links to ideas and
resources for managing stress, and although it is UK-based, there are also links to equivalent
sites in some other countries.
Suggested further reading
Full details for all references are given in the list at the end of this book.
1 Organizational Stress Management: A Strategic Approach by Ashley Weinberg, Valerie Sutherland and Cary Cooper (Palgrave Macmillan, 2010) provides a good insight into how
organisations can take a pragmatic approach to the issue.
2 The Mindful and Effective Employee: An Acceptance and Commitment Therapy Training Manual for Improving Well-Being and Performance by Paul Flaxman, Frank Bond and Fredrik
Livheim (New Harbinger Publications, 2013) is an excellent, accessible and in-depth
account of how individual differences and their development through interventions
play a role in both well-being and performance.
3 Well-being: Productivity and Happiness at Work by Sheena Johnson, Ivan Robertson and Cary
Cooper (Palgrave MacMillan, 2017) provides a detailed account of how the findings from
contemporary research can be translated into practical interventions in organisational
settings.
Chapter 8 Work-related stress and well-being
275
4 Positive Psychology: The science of happiness and human strengths by Alan Carr (Routledge,
2011) provides a good introduction to a branch of psychology that is generating increasing levels of interest from academics and practitioners alike.
5 Contemporary Occupational Health Psychology: Global perspectives on research and practice
(Volume 3), published by Wiley-Blackwell in 2014, and edited by Stavroula Leka and
Robert Sinclair covers numerous work-related health topics in considerable depth and
detail, and provides some excellent case studies of the application of research findings
to pressing practical problems.
6 The analysis of the business case for managing work-related stress by Frank Bond and
colleagues in 2006 is an excellent and accessible review of the links between sources of
stress and important organisational outcomes. It can be downloaded free from the UK
HSE website: http://www.hse.gov.uk/research/rrpdf/rr431.pdf. Richardson and Rothstein’s (2008) review of stress management interventions is rather technical, but begins
with some very useful information about the various interventions that are available to
organisations (see references). Karina Nielsen and colleagues summarise a wide range
of interventions that can be integrated into organisational practices in their 2017 article on resource-based interventions that can be found at http://eprints.whiterose.
ac.uk/113229/7/Resourcemeta-analysis_paper_-_submittedRev3.pdf. David Holman and
Carolyn Axtell (2016) provide a detailed example of a primary intervention that can be
found at http://eprints.whiterose.ac.uk/98455/2/Holman-Axtell%202015.pdf.
CHAPTER 9
Groups, teams and
teamwork
LEARNING OBJECTIVES
After studying this chapter, you should be able to:
1 outline why human beings have a preference for living and working in groups;
2 summarise the trend towards team-based working in contemporary organisations and the
key dimensions upon which work teams differ;
3 using social psychological perspectives, critically explore how individual behaviour is influenced in the context of groups;
4 describe the fundamental differences between real and pseudo teams and their impact on
organisational performance;
5 discuss the benefits and challenges associated with team decision-making and team diversity;
6 understand the key propositions of the input-mediator-output-input model of team
effectiveness;
7 explain how team composition influences the way in which teams function and perform;
8 discuss the trend towards virtual team working, and the challenges faced by teams working
in highly virtual environments;
9 describe different types
of team processes and
emergent states and how
they influence team
outputs;
10 outline the future
challenges for the science
of teamwork, exploring
the notions of multi-team
systems and multiple team
membership.
Chapter 9 Groups, teams and teamwork
277
Opening case study
Multi-professional team working in mental health care
Multi-professional team working (MPTW), in which different professional group members work
together towards a common objective, has become a salient feature of organisational life in all
sectors, particularly that of healthcare. While the evidence for MPTW is growing, in practice organisations face significant difficulties in realising its benefits. In 2009 a group of researchers set out on a
three-year project to examine the factors that influence the effectiveness of MPTW. They focused on
a particular context in which MPTW is highly prevalent and has significant implications for patients;
adult mental health care in the community. The aim of the research was clear – what are the major
factors that ensure the effectiveness of the delivery and improvement of mental health care for
patients. This included both the facilitators of MPTW, as well as inhibitors that prevent teams from
being effective. Using this knowledge, the researchers aimed to develop a diagnostic tool that could
be used to measure MPTW effectiveness and to provide a clear set of practical guidelines that would
make a positive difference to experiences of patients.
The research itself involved three stages. Stage 1 engaged directly with team members, patients
and their carers to critically examine what exactly MPTW looks like in practice. Researchers used
qualitative techniques to uncover a set of themes that captured MPTW effectiveness from multiple
perspectives which could then be used as a basis for the development of a psychometric tool. These
themes were:
■
■
■
■
■
■
■
improved patient well-being
creative problem solving
continuous care
inter-team working
respect between professionals
responsiveness to carers
therapeutic relationships with patients.
Drawing upon these seven themes, Stage 2 involved developing and testing a diagnostic tool as part
of a large-scale survey with 111 community mental health teams (CMHTs) based in 11 National
Health Service (NHS) Trusts across the United Kingdom. As well as validating the newly developed
tool, this stage also compared CMHTs with other types of team working in the NHS. Compared to other NHS teams, CMHTs generally reported higher levels of: task and quality focus; constructive debate;
discussion of errors; and participation in decision-making. Levels of trust, support and psychological
safety were also relatively high. However, CMHTs generally reported lower levels of resources, organisational support, managerial praise for performance and achievement of goals, suggesting that these
are areas for improvement. Results also highlighted that the main drivers of MPTW in CMHTs included team leadership, a well-designed team task and team composition, regular team meetings, team
participation in decision-making, and trust, support and psychological safety among team members.
Finally, Stage 3 of the research used qualitative methods to uncover the detail and meaning behind
the findings of Stage 2. Observations of team meetings and interviews with team members revealed
more specific features of effective teams. For example, effective teams had the right mix of professionals and promoted positive attitudes towards others in different professions. They also demonstrated trusting relationships that extended beyond team members to patients and their carers. The
most effective teams were proactive in seeking out new and improved ways of delivering care and
were keen to work closely with other teams and agencies in order to do this. Leaders also appeared
278
Work Psychology
to play a crucial role in developing a supportive climate in teams, encouraging shared decisionmaking and sheltering the team from targets and bureaucracy and allow them to focus on their task.
Overall, this extensive multi-study investigation concluded with a clear set of practical guidelines for
NHS Mental Health Trusts. These include the need to:
■
■
■
■
■
■
clarify the purpose, objectives and functions of CMHTs so that team members can shape their
roles accordingly and ensure that they work interdependently;
provide effective leadership;
actively manage team composition and team processes;
promote inter-team working;
ensure teams have the opportunity for reflexivity and adaptation;
hold effective team meetings.
The following chapter will explore the research evidence that underpins many of these practical
recommendations, examining the meaning and implications behind them in more detail.
Adapted with permission from West, M., Alimo-Metcalfe, B., Dawson, J., El Ansari, W., Glasby, J.,
Hardy, G. et al. (2012). Effectiveness of Multi-Professional Team Working (MPTW) in
Mental Health Care. Final Report. NIHR Service Delivery Organisation Programme.
Introduction
It is the long history of humankind (and animal kind, too) that those who learned to collaborate
and improvise most effectively have prevailed.
Charles Darwin
Teamwork is at the very core of human behaviour and satisfies one of our most fundamental
needs: to belong. Living and working together in a collaborative way has ensured our survival
and evolution since the dawn of human kind (Despain, 2010). Since then, effective teamwork
has been key to the successful accomplishment of many of the most profound innovations of
our species such as the invention of the steam engine, pioneering the electric light, putting
astronauts on the moon, and in more recent times, the mapping of the human genome and
the development of the smart phone. All of these innovations have changed the course of
our history and are the product of collaboration and teamwork. It is no surprise, therefore,
that today’s organisations seek to capitalise on our innate preference to live and work in social
groups and benefit from the potential innovation that such teamwork offers.
Team-based working is now one of the most popular design choices of modern organisations. In a study published just after the turn of the century, 79 per cent of Fortune
1000 companies and 81 per cent of manufacturing organisations reported the widespread
use of self-managed teams (Thoms Pinto et al., 2002). More recently, research on 147 UK
NHS hospitals found that only 8 per cent of staff working in these hospitals reported that
they did not work in a team (Lyubovnikova et al., 2015), highlighting just how widespread
the practice of teamwork has become. Consequently, as this chapter will explore, research
into teamwork has rapidly increased over the past thirty years, making tremendous progress
in understanding the science and practice of team effectiveness (Mathieu et al., 2018). The
trend towards team-based organisation, both in science and practice is, therefore, clear. But
what exactly is team-based working?
According to West and Markiewicz (2004), team-based working reflects an organisational
design in which decision-making is decentralised, chains of command are eliminated, and the
Chapter 9 Groups, teams and teamwork
279
organisational hierarchy is flattened. The core building blocks of team-based organisations
are empowered teams: teams make decisions at the closest possible point to the service, product or customer. Teams also lead one another, and form the fundamental unit of accountability. Furthermore, as highlighted in the opening case study, teams are increasingly becoming
multidisciplinary and cross-hierarchical, meaning that they are often composed of members
from across an organisation (including senior executives, middle managers and operative
employees) who each perform very different functions and roles. The American printing giant
Xerox develops new products using multidisciplinary teams. Google encourages team collaboration and innovation through the way physical workspaces are designed to bring workers
together. Given the increasing internationalisation of marketplaces, teams are also becoming
more virtual in nature, with team members spanning across space and time zones, meaning
that team-based structures increasingly have the potential to break down geographical and
cultural barriers (see Chapter 13).
There are many reasons that can help explain why teamwork is now such a popular aspect
of organisational life. Team-based working allows organisations to learn more effectively,
retain this learning, better enact organisational strategy, and generally become more efficient, responsive and competitive (Cohen and Bailey, 1997). It also stimulates creativity and
innovation by mobilising groups of individuals to work together independently in order to
accomplish complex and often novel tasks that they simply could not achieve working in
isolation. Imagine the task of building a new house, for example. The physical requirements
alone would demand the efforts of a team. It is highly unlikely that any one individual would
posess all the knowledge, skills, experience, and resources necessary to plan and execute this
complex and difficult task entirely on their own. Input would be needed from a diverse team
of architects, structural engineers, builders, electricians, plumbers, building inspectors and
interior designers, working together towards a common goal to ensure the project was a success. The same is true of many tasks in modern organisations – the increasing complexity
of work requires us to collaborate more than ever before. However, effective teamwork certainly does not come easy to most teams. In this chapter we will explore the science behind
teamwork, and develop an understanding of how to build and manage effective teams in
increasingly dynamic and diverse work environments.
The chapter will start by summarising early research into group behaviour, considering
some of the seminal studies and research findings in social psychology that have shaped
our understanding of teamwork. Much of this early research was conducted in the laboratory using experimental designs to understand how individual behaviour is influenced
in the context of groups. These findings have important practical implications for leaders
and managers of teams in today’s organisations. The second part of the chapter will then
turn to more contemporary research in work psychology that has studied teams in real-life
organisational settings. Here we will consider exactly how work teams can be defined,
the key dimensions upon which they differ, and the major theoretical framework that
have sought to explain work team effectiveness, the input-mediator-output-input model.
We will also critically examine the current and future challenges for the science and practice
of teamwork.
Stop to consider
Think about your experiences of teamwork to date. This could include being part of a work team in an
organisation or a project team in an educational setting. Would you consider this team as successful?
If so, what made it effective? If not, where did the problems lie? Given the opportunity, would you be
happy to continue working in this team in the future? The answer to this last question is important
because team members’ preference to continue working together (referred to as team viability) is
argued to be a key indicator of team effectiveness.
280
Work Psychology
Foundations of group behaviour
Throughout history human beings have shown an unwavering preference for living and
working in groups (West, 2001). From an evolutionary perspective, group membership affords
us many advantages including close physical proximity to others, self-protection and reproductive opportunities, as well as mutual positive social support and pleasure of affiliation.
Beyond satisfying these basic innate needs, what else can explain our continued attraction
to social groups?
Social identity theory (Tajel and Turner, 1986) is a major theory in social psychology
which provides us with a crucial understanding about the function that group membership
serves in helping us to understand who we are. Social identity is defined as an individual’s self-concept which is derived from our understanding and knowledge about the social
group(s) to which we belong, and the emotional significance that we attach to these group
memberships. As individuals, we have a natural tendency to categorise ourselves and others
into groups. Being part of a group helps us to establish, to a greater or lesser extent, who we are
and the difference between ‘us’ and ‘them’ (Tajfel, 1978; van Knippenberg and Hogg, 2018).
We strive to belong to groups that have distinct and positive identities as this reflects on our
own social identity – boosting our self-esteem and giving us a sense of pride. Comparing our
own groups to others around us also helps to establish how positive we feel about our own
social identity (Kessler and Mummendey, 2008). We therefore show more favourable attitudes
to individuals who we perceive to be part of our group, referred to as in-group favouritism, as
well as discriminating against those we perceive to be a member of a different group, referred
to as out-group discrimination. In practice, this would mean ensuring that members of our
own work team receive more positive outcomes (such as resources or rewards), and members of
the out-group receive negative outcomes (such as higher workloads or cuts to their resources).
This is termed in-group bias and, according to social identity theory, serves a key function for
preserving the status of our own groups relative to others, thus enhancing our self-esteem.
The more strongly we identify with a particular social group, the more committed we tend
to be to the values and goals of the group and the more emotionally attached we are to fellow group members. For example, in a German study of 112 middle and top-level managers,
researchers found that leaders who strongly identify with their organisation were rated by
their followers as engaging in more transformational leadership activities (see Chapter 10)
such as displaying in-group behaviour, self-sacrificing behaviours or developing and communicating attractive visions that seek to enhance organisational performance (Effelsberg
and Solga, 2015). Employee identification with their team and organisation has been associated with many desirable work outcomes including increased performance, productivity,
job satisfaction, extra-role behaviours (see Chapter 4), and health and well-being, as well as
decreased intention to leave an organisation (see, for example, Riketta and van Dick, 2005;
Steffens et al., 2017). However, despite these potential benefits, researchers have also found
that a strong in-group orientation can trigger people’s willingness to engage in what have
been termed ‘unethical pro-organisational behaviours’, whereby individuals behave unethically (such as illegally withholding information or misreporting financial data, also known as
‘cooking the books’) in order to benefit the organisation (in-group) and harm those outside
of the organisation (out-group; Umphress et al., 2010).
Furthermore, according to a review by Hewstone et al. (2002), large groups with high status
generally show more in-group bias than low status low membership groups. Such dominant
groups show more generosity to members of out-group members when they perceive the
status gap as being very wide. In turn, low-status groups show higher in-group bias when they
perceive their low status as unfair, or see that there is an opportunity for closing the status gap.
However, it is important to note that not all individuals define their self-concept according to
group membership to the same extent. In some cultures, people tend to define themselves in
terms of their group memberships. In others they do so more in terms of their characteristics
as an individual (Ellemers et al., 2002). Therefore, the nature and extent of in-group bias is
somewhat contingent on wider contextual and cultural factors.
Chapter 9 Groups, teams and teamwork
281
Key learning point
Our self-concept is partly shaped by the groups to which we belong. The extent to which we define ourselves in terms of our group membership depends on individual, contextual and cultural differences.
Exercise 9.1
Think of the different groups that you currently belong to. These could be work teams in your organisation, a local sports team, a hobby or interest group, a religious group etc. Reflect on how you felt
when you joined these groups and how you feel being a member of them. Now imagine that you
were suddenly no longer part of one or more of these groups. What would your reaction be? To what
extent would it change how you see yourself?
While in-group bias serves as an important function for protecting an individual’s
self-concept and reducing uncertainty, it can also commonly lead to inter-team conflict
(i.e., conflict between teams). We shall be examining the concept of team conflict in more
detail later in this chapter. However, it is worth considering theories and concepts from
social psychology that can inform us about how to tackle in-group bias and reduce undesired
inter-group conflict in organisations. Early insight was provided over half a century ago when
Allport (1954) proposed the contact hypothesis, suggesting that, under certain conditions, the
mere contact of social groups can help to reduce prejudice and improve inter-group relations.
These conditions include ensuring equal status for members of both groups, the absence of
inter-group competition, establishing a superordinate goal that requires the cooperation
of team members, and ensuring a supportive wider context. Support for the positive effect of
contact on the reduction of inter-group prejudice was established in a meta-analysis of over
500 studies (Pettigrew and Tropp, 2006). The research concluded that while the original conditions outlined by Allport help to facilitate contact between conflicting groups, they are not
necessary for positive contact effects to occur. Pettigrew and Tropp (2006) suggest that the
contact hypothesis should be reformulated to consider negative contact conditions that should
be avoided. These include the extent to which there are frequent opportunities for contact
between groups (when contact is not frequent enough, negative contact can occur), as well
as reducing the extent to which a contact scenario may be considered as threatening (such
as provoking group member anxiety). Nevertheless, Allport’s (1954) original suggestion that
contact has alleviating effects on inter-group conflict appears to have sound empirical support.
Hewstone et al. (2002) summarised various other methods for reducing in-group bias, including teaching individuals about how to suppress their unconscious biases; encouraging them
to behave more positively towards out-group members; increasing people’s knowledge about
out-group members so that they are seen as unique individuals and not just group members;
and finally re-categorising out-group members to in-group members by emphasising or creating superordinate groups (for example, defining the group as everyone in an organisation).
Social norms
Social norms capture informal standards of behaviour that have been established and accepted
by a particular group and are shared by all group members. These belief systems serve as a
guide for expected standards of typical or desirable behaviour in a given group. For example,
does your department celebrate every colleague’s birthday with a card and gift, or is it the
282
Work Psychology
norm for birthdays to pass by unacknowledged? Work groups will have established social
norms for all sorts of behaviours, e.g. smart or casual clothing while in the office, responding
to work emails during non-working hours, and punctuality when attending meetings. When
an individual wishes to establish and maintain membership of a particular group, they will
observe the norms around them to understand what other group members do in certain
situations. These norms will be either descriptive in nature, informing us of how others are
likely to behave in similar situations (e.g. most conscientious people will be punctual) or
injunctive in nature, informing us about the type of behaviour that should be performed (e.g.
at a funeral, one should be quiet, respectful and dress in dark colours). The more consistently
a group behaves in a particular way, the more group members will deem that behaviour as
appropriate and will mimic the behaviour themselves (Thibaut and Kelley, 1959).
When agreed upon and internalised by the group, norms are incredibly powerful in shaping individual behaviour without any external pressure from leaders or managers (Ehrhart
and Naumann, 2004). Norms serve a critical function in ensuring that group members conform to expected standards of behaviour. They also help to coordinate and restrict individual
behaviour for the good of the wider group. In the UK even if you are in a rush to catch a train,
you must still comply with the social norm of queuing for a ticket: pushing to the front of the
queue or travelling without a ticket would violate such norms and would likely evoke negative responses from fellow passengers and station staff. Norms also help to reduce uncertainty
in unfamiliar situations. For example, if you change jobs, you can draw upon your knowledge
of what correct and derisible behaviour looked like in your old organisation and apply this
to your first day of work in your new organisation. Over time, you are likely to adapt your
behaviour and converge with new social norms that you observe and experience. Of course,
in reality, organisational life is made up of a complex network of multiple social norms, some
of which will reinforce one another, and others of which will significantly diverge and leave
us unsure of how to behave.
Early research examining the impact of social norms on individual behaviour stemmed
from the experimental research of Sherif (1936), who investigated how group norms are
formed using the ‘autokinetic effect’ – a perceptual illusion whereby a stationary light appears
to wobble in the absence of any referent point. In a darkened room, Sherif asked participants
to provide estimations regarding how much the light was ‘moving’ – of course, as it was an
optical illusion the task was highly ambiguous. When working alone, participants developed
stable personal norms regarding the extent of movement of the light, based on their own
internal frame of reference. When participants were sat together in groups and exposed each
other’s estimations of the light’s movement they quickly abandoned their personal norms
and their estimations converged to form a group norm: the assumption here being that ‘the
group must be right’ (Sherif, 1936: 111). Social norms can quickly emerge in groups and we
often look to others for how to behave in times of uncertainty.
The social influence that groups can have on individual decision-making was further
examined by Asch (1956), who examined the conditions under which people conform. Asch
argued that when tasks are clear, straightforward and low in uncertainty, individuals are perfectly capable of remaining independent of the influence of a group’s inaccurate judgement.
Asch wanted to refute the view that people are just ‘silly sheep’ willing to believe anything
that they are told (Cialdini and Trost, 1998).
Asch presented groups of individuals with two cards – one card depicting three lines of
differing lengths, and a second card depicting a single referent line. The task was simple –
individuals had to call out which of the three lines on the first card was the same length as the
referent line on the second card. Asch then introduced a set of conditions in which participants
sat next to each other and called out their answers publicly to the rest of the group. In fact, in
the majority of these conditions, most of the participants were actually confederates: Asch had
instructed them to deliberately give patterns of incorrect answers. Findings demonstrated that
naïve participants gave the same incorrect response as most confederates on 36.8 per cent
of trials. However, in conditions where naïve participants were told that they were late and
Chapter 9 Groups, teams and teamwork
283
should therefore write down their response rather than calling it out publicly, conformity
dropped to 12.5 per cent. In follow-up interviews, many reported that they thought the group
must be right and chose to deny the evidence of their own eyes. Others believed the group
was wrong but decided to go along with the incorrect answer to avoid feeling self-conscious,
ostracised or disapproved of by others.
Research on conformity has continued to grow, with many studies adopting similar tasks
to explore whether levels of conformity have changed over time, as well as the extent to
which conformity is contingent on factors such as culture or the size of the majority. For
example, in a meta-analysis of 133 studies from 17 countries, Bond and Smith (1996) found
that collectivist countries (where individual behaviour is typically focused on achieving collective goals, see also Chapter 12) tended to show higher levels of conformity than individualist countries (where individual behaviour is typically focused on achieving personal goals;
Triandis, 1990). Collectivist cultures place more value and emphasis on the group, and deviating from the group’s position in such contexts could severely jeopardise group membership.
In individualistic cultures, the ‘self’ is considered as more separate from the group, and social
identity tends to be more dependent on individual achievement than group achievement
(Hofstede, 1980). Bond and Smith (1996) also confirmed a number of other contingency
variables that are related to increased conformity. Conformity was significantly higher when:
■
the majority group was larger;
■
the majority group did not consist of out-group members;
■
there was a greater proportion of female respondents;
■
the task was more ambiguous.
Their analysis of US studies also found that conformity has appeared to decline since the
1950s, suggesting that gradual shifts in the cultural climate of Western societies may be making people less susceptible to majority influence.
Individual differences may also influence the extent to which individuals are likely to
conform. In an experimental study by Kosloff and colleagues (2017), researchers examined
the extent to which conformity was associated with two well-established meta-traits of personality referred to as the ‘Big Two’ – stability (a combination of low neuroticism, high agreeableness and high conscientiousness, a tendency to maintain stability in social relationships
and motivational and emotional domains) and plasticity (a combination of high extraversion and high openness to experience, a tendency towards exploratory behaviour related to
novelty seeking and voluntary engagement). Participants were required to rate the extent to
which they found unfunny cartoons humorous either while alone or surrounded by a group
of confederates. The tendency to rate unfunny cartoons as humorous when individuals faced
group pressure to do so was associated with stability. However, the effects of personality on
conformity were far more modest than the situational effect of the presence or absence of confederates. These findings suggest that more stable individuals may be more likely to adhere
to social norms, but they also re-emphasise the power of situational effects on conformity.
In other recent research, the role of emotions has been examined. Heerdink and colleagues (2013) were interested in how conformity versus deviance in groups can be shaped
by the group members’ emotional expressions. The researchers argued that happy and angry
responses to the deviating opinion of group members will provide signals of imminent
acceptance or rejection from the group. Across five experimental studies, this hypothesis
was confirmed with the majority group’s expression of happiness leading the deviant to feel
accepted, and expressions of anger making the deviant feel rejected. These findings indicate
that we are overly attuned to how our teammates respond to our behaviour and use these
emotional cues to make help make decisions about how to behave.
Despite the continued research into conformity, others have taken a more critical view of
Asch’s original studies. For example, Hodges and Geyer (2006) argue that results from conformity studies should be reinterpreted as providing a powerful insight into people’s concern
284
Work Psychology
for the views of others, and into our tendency to tell the truth when others do not. Indeed,
Hodges and Geyer (2006) proposed that by varying their patterns of dissent and providing
some wrong answers, participants produced a base layer of agreement among participants,
enabling groups to build up social solidarity.
The impact that group norms can have on the functioning and behaviour of work teams
was first recognised in a series of studies that took place at the Western Electric Company’s Hawthorn Works in Chicago between 1924 and 1932 – famously referred to as the
‘Hawthorne Studies’ (see Roethlisberger and Dickson, 1939). In a series of experiments on
groups of workers, the researchers found evidence for their assumption that increased lighting levels led to increased worker output. Intriguingly, however, worker output continued
to increase even when lighting levels were turned down, leading the researchers to question
whether group behaviour was actually more contingent on group dynamics. In follow-up
studies on a small group of women assembling telephones, it became apparent that the special attention that the group was receiving from management was having a positive impact
on group behaviour, including increased productivity and decreased absence – a response
that was later termed as the ‘Hawthorne Effect’. Furthermore, the studies highlighted how
groups can establish powerful norms for restricting individual productivity of group members. For example, when wage incentive plans were introduced in the bank wiring observation room, workers grew concerned that if their output was too high, expectations on daily
output might also increase which could lead to incentive cuts, layoffs and even punishments
for slower workers. In response the group established norms which restricted individual
productivity, thus ensuring that the group was operating well below its potential output.
Those who violated these norms were ridiculed, called names, and even punched on their
upper arms (Robbins and Judge, 2016). Clearly, groups hold significant power in implicitly or
explicitly enforcing such deviant behaviours. In a study of 61 work teams in a multinational
manufacturing company, Cole et al. (2008) found that dysfunctional team behaviours (i.e.
behaviours that violated norms for effective team functioning) had a negative effect on team
performance via the collective negative mood of the team. We revisit the roles that mood and
emotion appear to play in influencing team functioning later in this chapter.
Roles
Another fundamental feature in understanding group behaviour is that of group roles.
A role captures expected patterns of behavioural norms that should be performed by a person
who occupies a particular position in a group. We all have expectations about what a specific
role should look like, whether this be the role of a parent, a judge, a doctor or a waiter in a
restaurant. These expectations are derived from our experiences and perceptions of how an
individual is supposed to act in a certain situation. Early insight into how we comply with role
expectations was provided by Zimbardo (1989; Haney et al., 1973). Zimbardo was intrigued
as to whether the brutality that characterised many US prisons at the time was a product
of the individual who occupied and worked in prisons (i.e. a dispositional perspective), or
whether the institutional expectations relating to the roles of ‘prisoner’ and ‘guard’ could
make otherwise ‘normal’ compassionate individuals become bitter, malicious, or evil (i.e.
a situational perspective). To test this idea, Zimbardo recruited 24 male Stanford University
undergraduates to participate in a simulated prison experiment, in which the men were randomly assigned the role of either ‘prisoner’ or ‘guard’. Guards were provided with uniforms,
whistles and batons and were told to enforce basic prison rules. Prisoners were stripped of personal possessions, given a uniform and ID number and were locked up in cells, much like real
prisoners. After an uneventful first day, the guards and prisoners were observed to soon settle
into their new roles; guards began to harass and belittle prisoners, exercising their control by
introducing physical punishments and humiliating routines. While some of the prisoners
Chapter 9 Groups, teams and teamwork
285
rebelled, others broke down or became apathetic. After just six days, the emerging social
pathology left the researchers with no choice but to call off the experiment with Zimbardo
himself observing the increased confusion regarding what was role play and what was real.
In his later reflections he noted that around a third of the guards had become ‘tyrannical in
their arbitrary use of power, in enjoying their control over other people’ (Zimbardo et al.,
1982: 274). While others were considered as ‘good guards’ carrying out their role properly,
they never interfered with the comma