International Enterprises
and Global Information
Technologies:
Advancing Management
Practices
Felix B. Tan
Auckland University of Technology, New Zealand
Senior Editorial Director:
Director of Book Publications: Editorial Director:
Acquisitions Editor:
Development Editor:
Production Editor:
Typesetters: Print Coordinator:
Cover Design:
Kristin Klinger
Julia Mosemann
Lindsay Johnston
Erika Carter
Hannah Abelbeck
Sean Woznicki
Mike Brehm, Keith Glazewski, Natalie Pronio, Jen Romanchak, Milan Vracarich, Jr.
Jamie Snavely
Nick Newcomer
Published in the United States of America by
Information Science Reference (an imprint of IGI Global)
701 E. Chocolate Avenue
Hershey PA 17033
Tel: 717-533-8845
Fax: 717-533-8661
E-mail: cust@igi-global.com
Web site: http://www.igi-global.com
Copyright © 2011 by IGI Global. All rights reserved. No part of this publication may be reproduced, stored or distributed in
any form or by any means, electronic or mechanical, including photocopying, without written permission from the publisher.
Product or company names used in this set are for identification purposes only. Inclusion of the names of the products or
companies does not indicate a claim of ownership by IGI Global of the trademark or registered trademark.
Library of Congress Cataloging-in-Publication Data
International Enterprises and Global Information Technologies: Advancing Management Practices / Felix B. Tan, editor.
p. cm.
Includes bibliographical references and index.
Summary: “This book offers solutions, best practices, and case studies on information management on a global scale
among developed, developing, and emerging nations, including discussions of organizational and managerial implications”-Provided by publisher.
ISBN 978-1-60960-605-3 (hbk.) -- ISBN 978-1-60960-606-0 (ebook) 1. Information technology--Management. 2.
International business enterprises. 3. Management information systems. I. Tan, Felix B., 1959HD30.2.I556675 2011
658.4’038011--dc22
2011014967
British Cataloguing in Publication Data
A Cataloguing in Publication record for this book is available from the British Library.
All work contributed to this book is new, previously-unpublished material. The views expressed in this book are those of the
authors, but not necessarily of the publisher.
Table of Contents
Preface.................................................................................................................................................. xiv
Section 1
Technology Impact and Use
Chapter 1
Economic Freedom and the Impact of Technology on Productivity....................................................... 1
Ashok Robin, Rochester Institute of Technology, USA
Thomas Tribunella, State University of New York at Oswego, USA
Chapter 2
The Influence of National and Organizational Cultures on Technology Use:
An Exploratory Study Within a Multinational Organizational Setting.................................................. 19
Zixiu Guo, University of New South Wales, Australia
John D’Ambra, University of New South Wales, Australia
Chapter 3
Understanding IT Governance: A Case of Two Financial Mutuals....................................................... 41
Alastair Robb, University of Queensland, Australia
Michael Parent, Simon Fraser University, Canada
Section 2
Cases and International Perspectives
Chapter 4
User Evaluation of E-Government Systems: A Chinese Cultural Perspective...................................... 63
Nan Zhang, Tsinghua University, China
Xunhua Guo, Tsinghua University, China
Guoqing Chen, Tsinghua University, China
Patrick Y. K. Chau, University of Hong Kong, China
Chapter 5
Factors Influencing SME Compliance with Government Regulation on Use of IT:
The Case of South Africa....................................................................................................................... 85
Michael Kyobe, University of Cape Town, South Africa
Chapter 6
Knowledge Management Systems Diffusion in Chinese Enterprises:
A Multistage Approach Using the Technology-Organization-Environment Framework.................... 117
One-Ki (Daniel) Lee, University of Massachusetts Boston, USA
Mo (Winnie) Wang, City University of Hong Kong, Hong Kong
Kai H. Lim, City University of Hong Kong, Hong Kong
Zeyu (Jerry) Peng, University of Science and Technology of China, China
Chapter 7
Factors Affecting Information Communication Technologies Usage and Satisfaction:
Perspective from Instant Messaging in Kuwait................................................................................... 133
Kamel Rouibah, Kuwait University, Kuwait
Hosni Hamdy, Kuwait University, Kuwait
Chapter 8
The Information Overload Paradox: A Cross-Cultural Research Study.............................................. 162
Ned Kock, Texas A&M International University, USA
Ana Rosa Del Aguila-Obra, University of Malaga, Spain
Antonio Padilla-Meléndez, University of Malaga, Spain
Chapter 9
The Influence of Organizational Communication Openness on the Post-Adoption of Computers:
An Empirical Study in Saudi Arabia.................................................................................................... 179
Said S. Al-Gahtani, King Khalid University, Saudi Arabia
Hung-Pin Shih, Hsuan Chuang University, Taiwan
Chapter 10
Chronism Theory, Culture, and System Delay:
A Longitudinal Study of Post-Apartheid South Africa........................................................................ 203
Gregory M. Rose, Washington State University, USA
Carina DeVilliers, University of Pretoria, South Africa
Detmar W. Straub, Georgia State University, USA
Chapter 11
A Cross-National Study of Mobile Internet Services:
A Comparison of U.S. and Korean Mobile Internet Users.................................................................. 228
Dong Hee Shin, Sung Kyung Kwan University, South Korea
Section 3
Trends and Techniques
Chapter 12
Flexibility in Offshoring: Paradoxical Concepts of Control, Communication
and Facilitation..................................................................................................................................... 254
Vanita Yadav, Institute of Rural Management Anand (IRMA), India
Monica Adya, Marquette University, USA
Varadharajan Sridhar, Sasken Communication Technologies, India
Dhruv Nath, Management Development Institute, India
Chapter 13
Taking Information Systems Business Process Outsourcing Offshore:
The Conflict of Competition and Risk................................................................................................. 284
Georgia Beverakis, University of New South Wales, Australia
Geoffrey N. Dick, University of New South Wales, Australia
Dubravka Cecez-Kecmanovi, University of New South Wales, Australia
Chapter 14
Managing Knowledge Transfer in Offshore Software Development:
The Role of Copresent and ICT-Based Interaction.............................................................................. 302
Aini Aman, University Kebangsaan Malaysia, Malaysia
Brian Nicholson, Manchester Business School, UK
Chapter 15
IT Maturity and Strategic Alignment: Moderating Effect of Strategic
Organizational Contexts....................................................................................................................... 321
Leelien Ken Huang, Feng Chia University, Taiwan
Compilation of References................................................................................................................ 347
About the Contributors..................................................................................................................... 378
Index.................................................................................................................................................... 386
Detailed Table of Contents
Preface.................................................................................................................................................. xiv
Section 1
Technology Impact and Use
Chapter 1
Economic Freedom and the Impact of Technology on Productivity....................................................... 1
Ashok Robin, Rochester Institute of Technology, USA
Thomas Tribunella, State University of New York at Oswego, USA
A well-developed body of literature has detected positive effects of technology investments on economic growth. We contribute to this literature by studying the joint effects of technology and economic
freedom on economic growth. Using two different time points, 1990 and 2000, and a sample of over
100 countries, we find that economic freedom enhances the effect of technology on economic growth.
In fact, we find that the standalone effect of freedom is not as large as its interactive effect with technology.
Chapter 2
The Influence of National and Organizational Cultures on Technology Use:
An Exploratory Study Within a Multinational Organizational Setting.................................................. 19
Zixiu Guo, University of New South Wales, Australia
John D’Ambra, University of New South Wales, Australia
This article presents the results of an exploratory study examining the influence of national and organizational cultures on technology use in a multinational organizational (MNO) context. Data were
collected from 121 respondents of a multinational organization in the Asia Pacific with headquarters in
Australia and three subsidiaries in Asia: Korea, Malaysia, and Thailand. Even though significant differences were found between respondents from headquarters and subsidiaries in terms of perceived media
richness and preference for telephone and written documents, very similar media preference patterns
for face-to-face and e-mail between the headquarters and subsidiaries were identified. Furthermore,
face-to-face and e-mail were two primary media used for most communication activities in this MNO.
Follow-up interviews revealed that the universal organizational culture of this MNO may explain media
use consistency between the headquarters and subsidiaries. Implications of the findings are discussed
and future research considered.
Chapter 3
Understanding IT Governance: A Case of Two Financial Mutuals....................................................... 41
Alastair Robb, University of Queensland, Australia
Michael Parent, Simon Fraser University, Canada
Legislators, regulators, and shareholders increasingly demand good governance over all aspects of
their business. While much is made of financial governance, most legislation and regulation implicitly
recognizes the need for prudent governance of information technology (IT) functions. In this study we
conduct an exploratory collective case study of IT governance (ITG) in two financial mutuals - one in
Australia and one in Canada, using a contextual lens. In one case, the mutual governs its IT through
Board participation in a subsidiary. In the second, governance is delegated to management and a Lead
Director. Both of these mechanisms appear to minimize ITG risk, and are the result of their respective
regulatory environments. This research begins to lend some clarity regarding IT governance choices by
firms, and denotes important contextual differences between countries’ regulatory environments. This
will allow researchers, managers, and directors to better understand and discriminate between ITG
processes and structures.
Section 2
Cases and International Perspectives
Chapter 4
User Evaluation of E-Government Systems: A Chinese Cultural Perspective...................................... 63
Nan Zhang, Tsinghua University, China
Xunhua Guo, Tsinghua University, China
Guoqing Chen, Tsinghua University, China
Patrick Y. K. Chau, University of Hong Kong, China
Based on Information Technology adoption theories and considering Chinese cultural characteristics,
this chapter proposes a user centric IT/IS evaluation model composed of three determinants, namely
perceived usefulness, perceived ease of use, and perceived fit, for investigating the e-government systems application and management in China. By empirically validating the model with survey data, it is
demonstrated the perceived fit has significant impacts on the end users’ evaluation towards e-government systems, due to the special element of Hexie in the Chinese culture. The results also indicate that
the reasons for failures in e-government systems application in China largely lie in the lack of fit, which
may root in the long power distance characteristic of the Far Eastern culture.
Chapter 5
Factors Influencing SME Compliance with Government Regulation on Use of IT:
The Case of South Africa....................................................................................................................... 85
Michael Kyobe, University of Cape Town, South Africa
This study investigated the factors influencing SME compliance with regulation on use of IT in South
Africa. The researcher argues that these consist of a combination of business, industry, economic, technological, sociological, and psychological factors. The results show that cost of compliance was the
main influencing factor and that both rural and urban SMEs make limited effort to develop policies and
demonstrate compliance. Furthermore, a discriminant function analysis revealed that rural SMEs can
be distinguished from their urban counterparts in terms of their disposition towards IT regulation by
five factors. Perception of high compliance costs was the key distinguishing factor followed by lack of
awareness, lack of training on compliance and security, perception of unfair regulation and possession
of inadequate security controls. The author hopes that this research-based evidence will provide better
understanding of SME compliance behaviors and guide the development of appropriate solutions to
compliance challenges in these organizations.
Chapter 6
Knowledge Management Systems Diffusion in Chinese Enterprises:
A Multistage Approach Using the Technology-Organization-Environment Framework.................... 117
One-Ki (Daniel) Lee, University of Massachusetts Boston, USA
Mo (Winnie) Wang, City University of Hong Kong, Hong Kong
Kai H. Lim, City University of Hong Kong, Hong Kong
Zeyu (Jerry) Peng, University of Science and Technology of China, China
With the recognition of the importance of organizational knowledge management (KM), researchers
have paid increasing attention to knowledge management systems (KMS). However, since most prior
studies were conducted in the context of Western societies, we know little about KMS diffusion in other
regional contexts. Moreover, even with the increasing recognition of the influence of social factors in
KM practices, there is a dearth of studies that examine how unique social cultural factors affect KMS
diffusion in specific countries. To fill in this gap, this study develops an integrated framework, with
special consideration on the influence of social cultures, to understand KMS diffusion in Chinese enterprises. In our framework, we examine how specific technological, organizational, and social cultural
factors can influence the three-stage KMS diffusion process, that is, initiation, adoption, and routinization. This study provides a holistic view of the KMS diffusion in Chinese enterprises with practical
guidance for successful KMS implementation.
Chapter 7
Factors Affecting Information Communication Technologies Usage and Satisfaction:
Perspective from Instant Messaging in Kuwait................................................................................... 133
Kamel Rouibah, Kuwait University, Kuwait
Hosni Hamdy, Kuwait University, Kuwait
Instant messaging (IM) technology has received extensive focus in the West while there is lack of
knowledge of it in the Arab world. This study aims to shed light on factors affecting IM usage and user
satisfaction in an Arab country (Kuwait). To achieve this objective, this study develops a theoretical
model that is based on three well-known models. This model includes curiosity (from the theory of
flow), compatibility (from the innovation diffusion theory), perceived usefulness and perceived ease of
use (from the technology acceptance model), and individual characteristics in the form of prior similar
experience. The proposed model was tested using survey data from 609 students, with the results lending support for the proposed model. Importantly, results highlight the impact of social effect on curiosity as a new mediator of technology adoption and satisfaction. This study contributes to the literature
on technology adoption in the Arab word and aids educational institutions and companies to understand
the social and technical nature of users’ attitudes with regard to ICT adoption and satisfaction.
Chapter 8
The Information Overload Paradox: A Cross-Cultural Research Study.............................................. 162
Ned Kock, Texas A&M International University, USA
Ana Rosa Del Aguila-Obra, University of Malaga, Spain
Antonio Padilla-Meléndez, University of Malaga, Spain
Information overload is one of the major problems for individuals and organizations in modern urban
environments. This phenomenon has been studied for many years, and has proven to be more complex
than researchers believed it to be. It is important to better understand this “pathology of information”
for two reasons. First, it has a deleterious impact on work productivity and quality. Second, it has
traditionally been a driver of Information Technology developments aimed at helping individuals to
better cope with it. The study presented here aims at shedding light on the complexity of information
overload, by examining the relationship between perceived information overload intensity and three
traditional information overload predictors as well as one nontraditional predictor. The nontraditional
predictor is power distance, or the extent to which less powerful members of a national culture accept
that power is distributed unequally. Power distance was manipulated through the collection of data
from 184 local managers and professionals (in New Zealand, Spain and the U.S.A.). The data analyses
led to one surprising conclusion: perceived information overload intensity seems to be more strongly
related to power distance than to the volume of written information or number of information transactions processed by an individual.
Chapter 9
The Influence of Organizational Communication Openness on the Post-Adoption of Computers:
An Empirical Study in Saudi Arabia.................................................................................................... 179
Said S. Al-Gahtani, King Khalid University, Saudi Arabia
Hung-Pin Shih, Hsuan Chuang University, Taiwan
This study adopts social identity theory (SIT) to examine the post-adoption of computers using a research model that extends the theory of planned behavior (TPB) with two organizational cultural factors. Individual attitudes toward using computers and perceived behavioral control are TPB personal
factors, while subjective norms can be viewed as the social factor. Empirical findings from 400 Arab
end-users show that openness to superior-subordinate relationships (reducing organizational boundaries) significantly influences current computer usage only through personal factors. In contrast, openness
to superior-subordinate relationships significantly influences continued use of computers through personal and social factors. However, openness to work communication (reducing communicative boundaries) does not significantly influence either current computer usage or continued use of computers
through TPB beliefs. The implications for research and practice, and the limitations of this study, are
discussed accordingly.
Chapter 10
Chronism Theory, Culture, and System Delay:
A Longitudinal Study of Post-Apartheid South Africa........................................................................ 203
Gregory M. Rose, Washington State University, USA
Carina DeVilliers, University of Pretoria, South Africa
Detmar W. Straub, Georgia State University, USA
System response delay has been cited as the single most frustrating aspect of using the Internet and
the most worrisome aspect of Web application design. System response time (SRT) research generally
concludes that delay should be eliminated where possible to as little as a few seconds, even though
delay reduction is costly. Unfortunately, it is not clear if these conclusions are appropriate outside of
the developed world where nearly all of the SRT research has taken place. Cultural effects have been,
hence, generally missing from SRT research. The one SRT study to date outside of the developed world
did report differences using the theoretical construct of cultural chronism, and this finding could limit
the generalizability of SRT research findings from developed countries to many economically developing nations. However, limitations and potential confounds in this single study render those findings
tentative. The end of Apartheid in South Africa allowed an opportunity to conduct a longitudinal free
simulation experiment that overcomes the critical limitations of this previous research. Subjects were
members of historically polychronic and monochronic groups who had been segregated by Apartheid
and now live in an integrated society with shared infrastructure and computer access. Results find
that members of the historically polychronic group are more accepting of longer delays and are more
willing to trade longer delays for improved functionality than are their historically monochronic counterparts. Furthermore, tests find that members of the historically monochronic population that came of
age in a desegregated, majority-polychronic culture appear to be polychronic themselves and to differ
significantly from the older monochronic generation. Results from this study can be applied to design
culturally-sensitive applications for users in the developing economies of the world.
Chapter 11
A Cross-National Study of Mobile Internet Services:
A Comparison of U.S. and Korean Mobile Internet Users.................................................................. 228
Dong Hee Shin, Sung Kyung Kwan University, South Korea
This study surveyed mobile users in the United States and Korea to determine the key differences between the two countries. Survey questions, developed in two languages, were presented in each country
to explore the influences of informativeness, entertainment, interactivity, and availability on mobile
user dimensions. The study design methods were based on the revision of a uses and gratifications approach, and a relational model of antecedents and consequences was tested with a structural equation
modeling approach. Mobile Internet service uses and gratifications were analyzed cross-nationally in
a comparative fashion focusing on the differences in the composition of motives in the two countries.
Based on the results of this study, practical implications for marketing strategies in mobile service markets and theoretical implications for cross-country studies are recommended accordingly.
Section 3
Trends and Techniques
Chapter 12
Flexibility in Offshoring: Paradoxical Concepts of Control, Communication
and Facilitation..................................................................................................................................... 254
Vanita Yadav, Institute of Rural Management Anand (IRMA), India
Monica Adya, Marquette University, USA
Varadharajan Sridhar, Sasken Communication Technologies, India
Dhruv Nath, Management Development Institute, India
Globalization of software development has resulted in a rapid shift away from the traditional collocated, on-site development model, to the offshoring model. Emerging trends indicate an increasing interest
in offshoring, even in early phases like requirements analysis. Additionally, the flexibility offered by the
agile development approach makes it attractive for adaptation in globally distributed software work. A
question of significance then is what impacts the success of offshoring earlier phases, like requirements
analysis, in a flexible and globally distributed environment? This chapter incorporates the stance of
control theory to posit a research model that examines antecedent factors such as requirements change,
facilitation by vendor and client site-coordinators, control, and computer-mediated communication.
The impact of these factors on success of requirements analysis projects in a “flexible” global setting is
tested using two quasi-experiments involving students from Management Development Institute, India
and Marquette University, U.S.A. Results indicate that formal modes of control significantly influence
project success during requirements analysis. Further, facilitation by both client and vendor site coordinators positively impacts requirements analysis success.
Chapter 13
Taking Information Systems Business Process Outsourcing Offshore:
The Conflict of Competition and Risk................................................................................................. 284
Georgia Beverakis, University of New South Wales, Australia
Geoffrey N. Dick, University of New South Wales, Australia
Dubravka Cecez-Kecmanovi, University of New South Wales, Australia
As Business Process Outsourcing (BPO) moves offshore and now includes Information Systems (IS)
Processes, there is a need to consider a combination of the two. This article explores the factors that a
multinational organisation considered when it “offshored” its IS business processes to lower-cost destinations. It focuses on determining the driving factors and challenges faced during the offshore sourcing
project. A single, in-depth interpretive case study approach was used to explore this research topic. The
results of this study show that the organisation under investigation was primarily driven to offshore its
IS business processes in order to become more competitive in the marketplace. This was assisted by
the organisation reducing its operational costs, and establishing a global presence in many lower-cost
locations offshore. A model was developed, which illustrates the interrelationships that exist between
these concepts.
Chapter 14
Managing Knowledge Transfer in Offshore Software Development:
The Role of Copresent and ICT-Based Interaction.............................................................................. 302
Aini Aman, University Kebangsaan Malaysia, Malaysia
Brian Nicholson, Manchester Business School, UK
The aim of this article is to examine the role of copresent interaction and the extent to which this can be
supplanted by information and communication technology-based interaction for managing knowledge
transfer in distributed settings. This study draws on two case studies of small UK firms sourcing software development from India and Bangladesh. Using Nonaka and Konno’s knowledge creation theory,
the role of copresent and ICT-based interactions in managing knowledge transfer is explained. The
article contributes an extension of the concepts of knowledge creation theory by providing evidence
of the role of copresent and ICT-based interaction for knowledge transfer in the context of offshore
software development.
Chapter 15
IT Maturity and Strategic Alignment: Moderating Effect of Strategic
Organizational Contexts....................................................................................................................... 321
Leelien Ken Huang, Feng Chia University, Taiwan
Strategic alignment of IT/business remains one of the top IT management issues. We proposed, using
contingency theory, that the extent to which IT maturity can be evolved to support overall business
goals is moderated by strategic organizational contexts. Results suggested that companies could succeed when IT maturity is appropriate for a certain strategic organizational context. Our research model
was generic for foreign companies’ strategic behaviour because, based on contingency theory, these
companies make dynamic adaptations toward their particular external environment for a competitive
strategy. Implications of results are discussed.
Compilation of References................................................................................................................ 347
About the Contributors..................................................................................................................... 378
Index.................................................................................................................................................... 386
xiv
Preface
To keep pace with the rapidly expanding world of global business, it is vital to stay abreast of the latest
advances in management and technological strategies for a globalized market. This book of advances
in management practices offers just those set of tools to a manager, IT professional, business student,
or any member of the global business chain.
Narrowing the field of global business to give the most relevant and helpful practices exposure, “International Enterprises and Global Information Technologies: Advancing Management Practices” offers
fifteen chapters in three sections. The authors of these chapters hail from a dozen countries around the
world, offering insight in case studies and comparisons of international enterprises.
The first section, “Technology Impact and Use” introduces some of the latest in the information
technology, its governance, and the trends in terminology and semantics surrounding these advances.
Section one includes three chapters that introduce the book and lay a framework for implementing
practices discussed in further chapters.
Chapter one, “Economic Freedom and the Impact of Technology on Productivity,” by Ashok Robin
and Thomas Tribunella, introduces the book and gives the foundation for describing how productivity can
be affected by newly implemented technology. The authors contribute to this literature by studying the
joint effects of technology and economic freedom on economic growth. Using two different time points,
1990 and 2000, and a sample of over 100 countries, the chapter finds that economic freedom enhances
the effect of technology on economic growth. In fact, the chapter will explain that the standalone effect
of freedom is not as large as its interactive effect with technology.
Chapter two, “The Influence of National and Organizational Cultures on Technology Use: An Exploratory Study Within a Multinational Organizational Setting,” by Zixiu Guo and John D’Ambra, shows
how difference in culture and background is affected by technology use in in the international workplace. This chapter presents the results of an exploratory study examining the influence of national and
organizational cultures on technology use in a multinational organizational (MNO) context. Data were
collected from 121 respondents of a multinational organization in the Asia Pacific with headquarters in
Australia and three subsidiaries in Asia: Korea, Malaysia, and Thailand. Even though significant differences were found between respondents from headquarters and subsidiaries in terms of perceived media
richness and preference for telephone and written documents, very similar media preference patterns
for face-to-face and e-mail between the headquarters and subsidiaries were identified.
Chapter three, “Understanding IT Governance: A Case of Two Financial Mutuals,” by Alastair Robb
and Michael Parent, gives insight into IT governance by showing the link between effective management
of technology within two connected businesses at different sides of the globe. In this chapter, the authors
conduct an exploratory collective case study of IT governance (ITG) in two financial mutuals - one in
xv
Australia and one in Canada, using a contextual lens. In one case, the mutual governs its IT through
Board participation in a subsidiary. In the second, governance is delegated to management and a Lead
Director. Both of these mechanisms appear to minimize ITG risk, and are the result of their respective
regulatory environments. This research begins to lend some clarity regarding IT governance choices
by firms, and denotes important contextual differences between countries’ regulatory environments.
This will allow researchers, managers, and directors to better understand and discriminate between ITG
processes and structures.
The second section, “Cases and International Perspectives,” comprises the bulk of the book,
containing eight chapters that draw experiences from different sectors of global industry and enterprise.
Chapter four, “User Evaluation of E-Government Systems: A Chinese Cultural Perspective,” by Nan
Zhang, Xunhua Guo, Guoqing Chen, and Patrick Y. K. Chau, begins the second section of the book
with its first foray into e-government systems and their applications in China, specifically focusing on
the impact of cultural practices and how these considerations must be handled. Based on Information
Technology adoption theories and considering Chinese cultural characteristics, this chapter proposes
a user centric IT/IS evaluation model composed of three determinants, namely: perceived usefulness,
perceived ease of use, and perceived fit, for investigating the e-government systems application and
management in China. By empirically validating the model with survey data, it is demonstrated the
perceived fit has significant impacts on the end users’ evaluation towards e-government systems, due to
the special element of Hexie in the Chinese culture. The results also indicate that the reasons for failures
in e-government systems application in China largely lie in the lack of fit, which may take root in the
long power distance characteristic of the Far Eastern culture.
Chapter five, “Factors Influencing SME Compliance with Government Regulation on Use of IT: The
Case of South Africa,” by Michael Kyobe, continues the discussion of government role in international
e-business, specifically focusing on the impact of compliance within small and medium sized enterprises
(SMEs). The study within this chapter investigates the factors influencing SME compliance with regulation on use of IT in South Africa. The researcher argues that these consist of a combination of business,
industry, economic, technological, sociological, and psychological factors. The results show that cost of
compliance was the main influencing factor and that both rural and urban SMEs make limited effort to
develop policies and demonstrate compliance. Furthermore, a discriminant function analysis revealed
that rural SMEs can be distinguished from their urban counterparts in terms of their disposition towards
IT regulation by five factors. Perception of high compliance costs was the key distinguishing factor followed by lack of awareness, lack of training on compliance and security, perception of unfair regulation,
and possession of inadequate security controls.
Chapter six, “Knowledge Management Systems Diffusion in Chinese Enterprises: A Multistage Approach Using the Technology-Organization-Environment Framework,” by One-Ki (Daniel) Lee, Mo
(Winnie) Wang, Kai H. Lim, and Zeyu (Jerry) Peng, offers an architecture implemented across several
Chinese enterprises and the approaches behind its effects on knowledge management. With the recognition of the importance of organizational knowledge management (KM), researchers have paid increasing
attention to knowledge management systems (KMS). However, since most prior studies were conducted
in the context of Western societies, we know little about KMS diffusion in other regional contexts.
Moreover, even with the increasing recognition of the influence of social factors in KM practices, there
is a dearth of studies that examine how unique social cultural factors affect KMS diffusion in specific
countries. To fill in this gap, this study develops an integrated framework, with special consideration
on the influence of social cultures, to understand KMS diffusion in Chinese enterprises. In this frame-
xvi
work, the authors examine how specific technological, organizational, and social cultural factors can
influence the three-stage KMS diffusion process, that is, initiation, adoption, and routinization. This
study provides a holistic view of the KMS diffusion in Chinese enterprises with practical guidance for
successful KMS implementation.
Chapter seven, “Factors Affecting Information Communication Technologies Usage and Satisfaction: Perspective from Instant Messaging in Kuwait,” by Kamel Rouibah and Hosni Hamdy, goes to
the Middle East, and studies the effects of instant messaging software on communication within global
enterprise and the impact on overall ICT effectiveness. Instant messaging (IM) technology has received
extensive focus in the West while there is lack of knowledge of it in the Arab world. This study aims to
shed light on factors affecting IM usage and user satisfaction in an Arab country (Kuwait). To achieve
this objective, this study develops a theoretical model that is based on three well-known models. This
model includes curiosity (from the theory of flow), compatibility (from the innovation diffusion theory),
perceived usefulness and perceived ease of use (from the technology acceptance model), and individual
characteristics in the form of prior similar experience. The proposed model was tested using survey data
from 609 students, with the results lending support for the proposed model. Importantly, results highlight the impact of social effect on curiosity as a new mediator of technology adoption and satisfaction.
Chapter eight, “The Information Overload Paradox: A Cross-Cultural Research Study,” by Ned
Kock, Ana Rosa Del Aguila-Obra, and Antonio Padilla-Meléndez, details research done between two
universities in the United States and Spain on the topic of the information overload paradox, giving lessons learned from the study and offering best practices among globally relevant scenarios. Information
overload is one of the major problems for individuals and organizations in modern urban environments.
This phenomenon has been studied for many years, and has proven to be more complex than researchers
believed it to be. It is important to better understand this “pathology of information” for two reasons.
First, it has a deleterious impact on work productivity and quality. Second, it has traditionally been a
driver of Information Technology developments aimed at helping individuals to better cope with it. The
study presented here aims at shedding light on the complexity of information overload, by examining the
relationship between perceived information overload intensity and three traditional information overload
predictors, as well as one nontraditional predictor. The nontraditional predictor is power distance, or the
extent to which less powerful members of a national culture accept that power is distributed unequally.
Power distance was manipulated through the collection of data from 184 local managers and professionals
(in New Zealand, Spain and the U.S.A.). The data analyses led to one surprising conclusion: perceived
information overload intensity seems to be more strongly related to power distance than to the volume
of written information or number of information transactions processed by an individual.
Chapter nine, “The Influence of Organizational Communication Openness on the Post-Adoption of
Computers: An Empirical Study in Saudi Arabia,” by Said S. Al-Gahtani and Hung-Pin Shih, gives an
evidential account of the effective organization of communication technologies when granted openness
and acceptance within Saudi enterprise. The study in this chapter adopts social identity theory (SIT)
to examine the post-adoption of computers using a research model that extends the theory of planned
behavior (TPB) with two organizational cultural factors. Individual attitudes toward using computers
and perceived behavioral control are TPB personal factors, while subjective norms can be viewed as the
social factor. Empirical findings from 400 Arab end-users show that openness to superior-subordinate
relationships (reducing organizational boundaries) significantly influences current computer usage
only through personal factors. In contrast, openness to superior-subordinate relationships significantly
influences continued use of computers through personal and social factors. However, openness to work
xvii
communication (reducing communicative boundaries) does not significantly influence either current
computer usage or continued use of computers through TPB beliefs. The implications for research and
practice, and the limitations of this study, are discussed accordingly.
Chapter ten, “Chronism Theory, Culture, and System Delay: A Longitudinal Study of Post-Apartheid
South Africa,” by Gregory M. Rose, Carina DeVilliers, and Detmar W. Straub, returns to the question
of culture, this time within the specific setting of South Africa in a postmodern, post-Apartheid society,
giving critical commentary and a theory of chronism and system delay. The end of Apartheid in South
Africa allowed an opportunity to conduct a longitudinal free simulation experiment that overcomes the
critical limitations of this previous research. Subjects were members of historically polychronic and
monochronic groups who had been segregated by Apartheid and now live in an integrated society with
shared infrastructure and computer access. Results find that members of the historically polychronic
group are more accepting of longer delays and are more willing to trade longer delays for improved
functionality than are their historically monochronic counterparts. Furthermore, tests find that members
of the historically monochronic population that came of age in a desegregated, majority-polychronic
culture appear to be polychronic themselves and to differ significantly from the older monochronic
generation. Results from this study can be applied to design culturally-sensitive applications for users
in the developing economies of the world.
Chapter eleven, “A Cross-National Study of Mobile Internet Services: A Comparison of U.S. and
Korean Mobile Internet Users,” by Dong Hee Shin, elucidates a specific comparison between the mobile
infrastructures of two of the world’s leading technological countries and their m-Internet users. This
study surveyed mobile users in the United States and Korea to determine the key differences between
the two countries. Survey questions, developed in two languages, were presented in each country to
explore the influences of informativeness, entertainment, interactivity, and availability on mobile user
dimensions. The study design methods were based on the revision of a uses and gratifications approach,
and a relational model of antecedents and consequences was tested with a structural equation modeling
approach. Mobile Internet service uses and gratifications were analyzed cross-nationally in a comparative fashion focusing on the differences in the composition of motives in the two countries. Based on
the results of this study, practical implications for marketing strategies in mobile service markets and
theoretical implications for cross-country studies are recommended accordingly.
The third section, “Trends and Techniques,” concludes the book with four final chapters on topics
such as offshoring, knowledge transfer, IT maturity, and more from research and case studies across
the globe.
Chapter twelve, “Flexibility in Offshoring: Paradoxical Concepts of Control, Communication and
Facilitation,” by Vanita Yadav, Monica Adya, Varadharajan Sridhar, and Dhruv Nath, introduces the
concept of offshoring and the importance of its implementation and management therein. This chapter
incorporates the stance of control theory to posit a research model that examines antecedent factors
such as requirements change, facilitation by vendor and client site-coordinators, control, and computermediated communication. The impact of these factors on success of requirements analysis projects in
a ‘flexible’ global setting is tested using two quasi-experiments involving students from Management
Development Institute, India and Marquette University, U.S.A. Results indicate that formal modes of
control significantly influence project success during requirements analysis. Further, the chapter posits that
facilitation by both client and vendor site coordinators positively impacts requirements analysis success.
Chapter thirteen, “Taking Information Systems Business Process Outsourcing Offshore: The Conflict
of Competition and Risk,” by Georgia Beverakis, Geoffrey N. Dick, and Dubravka Cecez-Kecmanovic,
xviii
continues the discussion of offshore practices and focuses on honing economic theories towards the
best practices within information systems businesses. As Business Process Outsourcing (BPO) moves
offshore and now includes Information Systems (IS) processes, there is a need to consider a combination of the two. This chapter explores the factors that a multinational organisation considered when it
“offshored” its IS business processes to lower-cost destinations. It focuses on determining the driving
factors and challenges faced during the offshore sourcing project. A single, in-depth interpretive case
study approach was used to explore this research topic. The results of this study show that the organisation
under investigation was primarily driven to offshore its IS business processes in order to become more
competitive in the marketplace. This was assisted by the organisation reducing its operational costs and
establishing a global presence in many lower-cost locations offshore. A model was developed, which
illustrates the interrelationships that exist between these concepts.
Chapter fourteen, “Managing Knowledge Transfer in Offshore Software Development: The Role of
Copresent and ICT-Based Interaction,” by Aini Aman and Brian Nicholson, summarily bookends the
chapters on offshoring practices with a final, narrowed focus on knowledge transfer within software
development industries. The aim of this chapter is to examine the role of copresent interaction and the
extent to which this can be supplanted by information and communication technology-based interaction
for managing knowledge transfer in distributed settings. The study undertaken in the chapter draws on
two case studies of small UK firms sourcing software development from India and Bangladesh. Using Nonaka and Konno’s knowledge creation theory, the role of copresent and ICT-based interactions
in managing knowledge transfer is explained. The chapter contributes an extension of the concepts of
knowledge creation theory by providing evidence of the role of copresent and ICT-based interaction for
knowledge transfer in the context of offshore software development.
Chapter fifteen, “IT Maturity and Strategic Alignment: Moderating Effect of Strategic Organizational Contexts,” by Leelien Ken Huang, concludes the book with a general focus on the development
of Information Technology within a global enterprise, discussing the importance of strategic alignment
on the successful maturity therein. Strategic alignment of IT/business remains one of the top IT management issues. The authors propose, using contingency theory, that the extent to which IT maturity can
be evolved to support overall business goals is moderated by strategic organizational contexts. Results
suggested that companies could succeed when IT maturity is appropriate for a certain strategic organizational context. This research model was generic for foreign companies’ strategic behaviour, because,
based on contingency theory, these companies make dynamic adaptations toward their particular external
environment for a competitive strategy. Implications of results are discussed.
It is the sincere hope of the editors of this collection of case studies, research, and best practices that
it will serve as an essential reference tool for practitioners of global business. The international perspectives offered here give pointed insight into the latest word on Information Systems and Technology
within burgeoning global business.
Felix B. Tan
Auckland University of Technology, New Zealand
Section 1
Technology Impact and Use
1
Chapter 1
Economic Freedom and
the Impact of Technology
on Productivity
Ashok Robin
Rochester Institute of Technology, USA
Thomas Tribunella
State University of New York at Oswego, USA
ABSTRACT
A well-developed body of literature has detected positive effects of technology investments on economic
growth. We contribute to this literature by studying the joint effects of technology and economic freedom
on economic growth. Using two different time points, 1990 and 2000, and a sample of over 100 countries,
we find that economic freedom enhances the effect of technology on economic growth. In fact, we find
that the standalone effect of freedom is not as large as its interactive effect with technology.
INTRODUCTION
Economic growth is considered a key indicator
of national success. A country’s performance and
status is often determined by its level and growth
in economic income. Alternative measures such as
Gross National Product (GNP) or Gross Domestic
Product (GDP) (levels or percent changes) are used
as proxies to measure economic growth. Because
of its importance, considerable research has been
directed toward determining factors influencing
economic growth. This literature, inspired by
Solow (1956), spans half a century and hundreds
of publications. A recent offshoot, appearing in
DOI: 10.4018/978-1-60960-605-3.ch001
the Information Systems (IS) literature, seeks
to assess the effect of technology on growth.
No doubt, this research has been spurred by the
advent of the Internet and the digital economy.
Therefore, it is no surprise that a fairly narrow
definition of technology is used in most studies.
The independent variable (technology) typically
reflects the following three elements: computer
hardware, computer software, and communications equipment. The conclusion from these studies
is that technology has a positive impact on growth.
We study the impact of economic freedom
on the relationship between technology and economic growth. We use a sample of more than 100
countries and use two cross-sectional snapshots
during 1990 and 2000. We argue that a climate
Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
Economic Freedom and the Impact of Technology on Productivity
of economic freedom allows various entities
(individuals, teams, corporations, societies) the
flexibility to harness the positive effects of technology. Not only would greater investments be
made in technology but these investments would
also have a greater possibility of bearing fruit.
Thus, we expect technology to have a greater effect on economic growth when coupled with an
environment of economic freedom. We test this
proposition and find results consistent with our
hypothesis. We report robust results indicating a
significantly positive interaction between freedom
and technology. We note that it is not economic
freedom per-se that affects growth but technology
accompanied by economic freedom.
In the next section of the paper, we describe the
two streams of literature we draw on. We present
our research models along with a description of
our sample in the Data and Methodology section.
We report the outcomes of our statistical tests in
the Results section. Then we discuss the data and
results of our research findings. In the following section, we outline the contributions of our
study. Finally, the limitations of the research are
highlighted and we conclude in the last section.
BACKGROUND AND
LITERATURE REVIEW
Technology and Economic Growth
There is a large stream of literature relating technology and economic growth (for a comprehensive
review of this literature, see: Dedrick, Gurbaxani
and Kraemer, 2003; Indjikian and Siegel, 2005; and
Merville, Kraemer and Gurbaxani, 2004). The vast
majority of this literature focuses on a particular
aspect of technology, namely IT. This is a logical
focus in the last two decades because of the rapid
computerization of various business processes and
the advent of the Internet. Due to the focus on IT,
key independent variables have reflected investments in computer hardware, software, Internet
2
and communication technologies. The empirical
relationship between technology and growth is
studied using various specifications. Dedrick,
Gurbaxani and Kraemer (2003) categorize these
studies based on the aggregation level of data:
firm-level, industry-level, and country-level.
The main debate in the literature centered on
whether or not technology enhanced productivity (or economic growth). The evidence from the
1980s using data from the United States of America
(USA) was predominantly negative (e.g. Roach
1987; Strassman 1985). This is in contrast to the
evidence from the 1990s indicating a significant
and positive relationship between technology
and growth (e.g., Brynjolfsson and Hitt, 1996;
Jorgenson and Stiroh, 2000; Lichtenberg, 1995).
The so-called ‘productivity paradox’ (Solow,
1987) of the earlier time period has been attributed
to various reasons. Perhaps technology investments were too small to create a positive effect
(Oliner and Sichel, 1994), and these investments
needed to go beyond some minimum value before
they could affect growth (Osei-Bryson and Ko,
2004). Perhaps there was also a learning curve
associated with technology (Dedrick, Gurbaxani
and Kraemer, 2003). Additionally, the literature
has suggested that other factors contributing to
organizational performance may have been omitted in evaluating IT impacts (Devaraj and Kohli,
2000). For example, studies have suggested that
organizational factors (e.g. type of IT, management
and workplace practices, changes initiatives, the
organizational structure and culture, and financial
conditions), the competitive environment (e.g.
industry competitiveness and regulation), and
macro environment (e.g. level of development,
public policies, cultural factors, education, IT
infrastructure) are important factors influencing
the extent of IT business value (Merville, Kraemer
and Gurbaxani, 2004). Finally, the benefits of IT
may expand beyond the frontiers of the company
initiating the IT investments. Thus, part of these
benefits may be captured by business partners
Economic Freedom and the Impact of Technology on Productivity
or the end customers (Bresnahan, 1986; Hitt and
Brynjolfsson, 1996).
Since country level studies are especially
pertinent to our research, we list some of the key
studies using country level data in the table in
Appendix A. This table is obtained from Dedrick,
Gurbaxani, and Kraemer (2003) and has been
modified to include some more recent studies.
This literature can be characterized as follows:
Most studies are of recent vintage (mid-1990s
onward); most use univariate tests associating
technology investments with economic growth and
focus on developed countries. The link between
technology and economic growth is significant
and positive in developed countries and newly
developed economies (e.g., Daveri, 2000; Lee,
Gholami, and Tong, 2005; Oliner and Sichel,
2000; Pook and Pence, 2004). However, this link
is not significant in developing countries as shown
by most prior studies (e.g., Dewan and Kraemer,
2000; Lee, Gholami, and Tong, 2005).
The latter finding ties-in with the results from
the 1980s: it appears that a certain threshold of
investments is required or perhaps a certain level
of infrastructure is needed before a clear link
between technology and growth can be detected
in developing countries. Complementary factors
can modify the impact of technology on growth.
This is the main issue considered in our paper. At
the firm-level such crucial factors may include
complementary investments such as workforce
training (Dedrick, Gurbaxani and Kraemer, 2003).
At the country level, environmental factors may
play a big role. For instance, Shih, Kraemer and
Dedrick (2007) show that environmental factors (openness to trade) can affect the level of
technological investments. Other authors such as
Mbarika, Byrd, and Raymond (2002) have suggested that macro-level factors including policy,
economical, financial, technological, political,
and geographical factors are key determinants of
the level of IT/telecommunications infrastructure
growth, thus economic growth, in Least Developed
Countries (LDC). The current study builds on
this recent work. However, our focus is on how
economic freedom, as an environmental factor,
affects the impact of technology on growth.
Our study is related to Meso, Datta, and
Mbarika (2006) and Meso, Musa, Straub and
Mbarika (2009). These studies examined the link
between IT, governance and economic growth.
The Meso, Datta and Mbarika (2006) study
is especially relevant because it examines the
modifying effects of governance variables on
the relationship between economic growth and
IT and reports an interaction effect for certain
governance-related variables such as ‘Voice/Accountability’ and ‘Rule of Law’. The more recent
Meso, Musa, Straub and Mbarika (2009) study
uses structural equations modeling techniques to
examine similar issues and reports statistically
significant links between IT and governance and
between IT and socio-economic development.
However, there are important differences between
our research and these two studies. Our study uses
different dependent and independent variables as
well as data from different time periods. More
significantly, our focus is on economic freedom,
not governance. Thus, our approach is different
and complementary.
Finally, the literature on foreign direct investment (FDI) and its effects on economic growth
offers an interesting parallel to the literature on
technology and growth. The connection is evident when one realizes that FDI often involves
the transfer of technology and knowledge capital
by MNCs. Azman-Saini, Baharumshah and Law
(2010) study the nexus between FDI, freedom and
growth and find that the effect of FDI on growth
is contingent on the level of economic freedom in
host countries. The explanations they offer—that
an environment of freedom allows entrepreneurs/
firms to take risks and to try new ideas and to
deploy labor flexibly and generate positive spillovers—are easily adaptable to a discussion on
technology and growth.
3
Economic Freedom and the Impact of Technology on Productivity
Economic Freedom and
Economic Growth
While our study contributes to the literature linking
technology to growth, it is influenced by a long
literature stream linking freedom and growth.
This literature stream dates back to the ‘growth’
literature in mainstream economics (e.g., Solow,
1956). There are multiple approaches to explaining national economic growth. Perhaps the most
intuitive approach is the production function approach that relates output to inputs such as labor
and capital. Technology may be considered a
modifier of this relationship. Economic freedom
could be another modifying factor.
Economic freedom refers to the degree to
which a market economy exists. Components of
economic freedom are: an environment favoring
a voluntary exchange, free competition, protection of persons and property, and a limited degree
of interventionism in the form of government
ownership, regulations, and taxes (Gwartney
and Lawson, 2002; Berggren, 2003). One should
distinguish economic freedom from both civil and
political freedoms. Civil freedom includes such
elements as the freedom of the press, the freedom
of association, the freedom of religion, and the
freedom of speech whereas political freedom
involves the free and open participation in the
political process, and elections that are free, fair,
competitive and corruption-free (Gwartney and
Lawson, 2002). While civil and political freedoms
may allow societies to express themselves fully,
economic freedom has the potential to directly
affect economic activity. Historic events such
as the breakup of the former Soviet Union have
focused attention on this important issue. We
focus on economic freedom rather than political
freedom because we are more interested in the
policies that directly affect economic productivity.
Various studies have examined the crosssection of country level data to determine whether
economic growth is correlated with indicators of
economic freedom. Overall, studies have found
4
that the level of economic freedom exerts a positive
and significant effect on economic growth (Goldsmith, 1997; Ali and Crain, 2002; Vega-Gordillo
and Alvarez-Arce, 2003). A number of studies
that have used changes in economic freedom (as
opposed to the level of economic freedom) as
an independent variable have concluded that the
change in economic freedom is also positively and
significantly correlated to growth rate (Dawson,
1998; Gwartney, Holcombe, and Lawson, 2004).
It is conceivable that not all dimensions of
economic freedom impact economic growth.
The index of economic freedom used in most
studies (e.g., Gwartney and Lawson, 2002) covers various aspects of freedom such as size of
government, legal structure and property rights,
access to sound money, freedom of exchange and
regulation of business. Carlsson and Lundstrom
(2002) decompose the various aspects of economic
freedom and find that the most significant effects
are associated with legal structure and freedom of
exchange. These results, especially those pertaining to legal structure, are an excellent complement
to the well-known ‘law and finance’ literature
where a key result is that country level investor
protection enhances corporate value (La porta,
Lopez-De-Silanes, Shleifer, and Vishny, 2002).
More recently, studies have focused on the effect of government size (usually measured using
government expenditures or revenues) on growth
and carried out robustness tests: an example is
Bergh and Karlsson (2010) that reports a negative
relationship between government expenditure and
growth even after controlling for the overall level
of economic freedom and globalization.
While most of the studies on economic freedom
focus on its effects on economic growth (typically
measured using GDP), a small stream examines
the effects on freedom on entrepreneurial activity.
Studies in this genre report a negative relationship
between government size and entrepreneurial
activity (e.g., self-employment). Bjornskov and
Foss (2008) and Nystrom (2008) are examples of
studies reporting this negative relationship. The
Economic Freedom and the Impact of Technology on Productivity
Table 1. The ArCo Indicator
Creation of Technology
• Patents
• Scientific Articles
Technology Infrastructure
• Internet Penetration
• Telephone Penetration
• Electricity Consumption
first named study offers a systematic discussion
of why economic freedom can be expected to
affect entrepreneurial activity. If a government’s
size is particularly large, say it has nationalized
a certain industry, a consequence is that there is
no scope for entrepreneurial activity. More generally, the heavy hand of the government (high
taxes, regulations, incentive-distorting pricing/
subsidies) can deter entrepreneurial activity. Although we do not study entrepreneurship in our
paper, the preceding ideas have relevance: just as
the benefits of entrepreneurial activity are better
reaped in an environment of economic freedom,
the benefits of technology are better harvested in
an environment of economic freedom.
DATA AND METHODOLOGY
The dependent variable in our study is GDP per
worker adjusted for purchasing power parity and
expressed in the US currency (Prod2000 and
Prod1990). We use this as a proxy for economic
growth and refer to it as ‘productivity.’ We collected values for this variable for the years 1990
and 2000 from the Global Market Information
Database issued by Euromonitor International.
The database covers 205 nations from 1977 to
2006. We use GDP per worker while other similar
studies used GDP per capita. We believe that GDP
per capita can potentially be affected by factors
such as demographic shifts and employment levels.
For example, if a country has a large percent of
retired individuals, its GDP per capita will be low
because fewer people are working.
Our independent variables are technology and
economic freedom. Our measure of technology
Development of Human Skills
• Tertiary Science and Engineering Enrolment
• Mean Years of Schooling
• Literacy Rate
is the Indicator of Technological Capabilities for
Developed and Developing Countries (Tech2000
and Tech1990). The indicator was compiled by
Archibugi and Coco (2004). Values range from 0
(lowest capability) to 1 (highest capability). We
use this index because it comprehensively covers
most of the countries in the world with a ranking
of 162 countries in 1990 and 2000. It is one of the
most widely used and detailed indexes. Furthermore, the authors use publicly available data and
explain their methodology very clearly. The scale
has three major dimensions and eight sub-Indexes
of the Indicator of Technological Capabilities for
Developed and Developing Countries which are
listed in Table 1.
Our economic freedom variable is the Economic Freedom of the World (EFW) indicator by
Gwartney and Lawson (2002) from the Cato Institute (Free2000 and Free1990). The EFW index
has been compiled since 1970 and the data is
publicly available at www.freetheworld.com. The
EFW for a country is measured on a ten point
scale with 10 denoting the highest level of economic freedom. EFW contains five major areas:
Size of government, legal structure and property
rights, sound money policies, freedom to exchange, and business regulations. Each major area
has several components explained in Appendix
B. We use this index because it comprehensively
covers most of the countries in the world with
information on 123 countries in 2000 and 113 in
1990. It is also one of the most widely recognized
and detailed indexes of economic freedom. The
index has been stable over time and has been used
in several published papers (Cole, 2003; Heitger,
2004; Mbaku, 2003; Vega-Gordillo and AlvarezArce, 2003).
5
Economic Freedom and the Impact of Technology on Productivity
Table 2. Descriptive Statistics. This table reports descriptive statistics for the technology, freedom and
productivity variables. These variables are obtained from two different time points: 2000 and 1990
Variable
N
Min
Tech2000
162
.028
.867
.313
.329
.190
Free2000
123
3.500
8.700
6.500
6.399
1.066
Median
Mean
Sigma
Prod2000
166
183
105,064
3,978
13,273
19,377
Tech1990
162
.017
.735
.277
.278
.164
Free1990
114
1.300
9.300
5.350
5.632
1.743
Prod1990
156
193
67,100
3,138
10,902
15,664
Our main analysis uses data from the year
2000 and hence uses Tech2000 (Technology
2000), Free2000 (Economic Freedom 2000),
and Prod2000 (Productivity 2000) respectively.
Similarly, our robustness tests use data from 1990:
Tech1990, Free1990, and Prod1990.
We performed linear regression analyses on
the above data. Four models (A-D) are used as
follows.
Model A: Prod
Model B: Prod
Model C: Prod
+ β2Free2000
Model D: Prod
+ β2Free2000+
2000 = β0 + β1Tech2000
2000 = β0 + β1Free2000
2000 = β0 + β1Tech2000
2000 = β0 + β1Tech2000
β3Tech*Free
Model A is the basic model relating productivity to technology that has been extensively studied
in the literature. Model B is the basic model relating
productivity to freedom; this model is once again
fairly well researched. The focus of our paper is
on models C and D. Model C assesses the separate
effects of both technology and freedom on productivity. Model D adds further complexity by also
considering the interaction between technology
and freedom in their impact on productivity. Here,
the variable Tech*Free is defined as the product of
Tech2000 (or Tech1990) and a 0/1 dummy variable assuming the value of 1 when Free2000 (or
Free1990) is median or higher, and 0 otherwise.
6
Max
Models C and D, to the best of our knowledge,
are not found in the literature.
Similar models were applied to data from
1990. The interaction variable, although using the
same label as with data from 2000 in the interest
of shortening the label, is defined using data from
1990. The purpose of using data from 1990 is twofold: (a) to provide a test of robustness and (b)
to search for shifts in productivity relationships.
RESULTS
Descriptive Statistics
and Correlations
Table 2 reports descriptive statistics on the three
key variables: technology, freedom and productivity. We focus first on the most recent values,
from 2000. Tech2000 ranges from 0.028 to 0.867,
and has a mean and median of 0.329 and 0.313
respectively. Free2000 ranges from 3.5 to 8.7
with mean and median values of 6.399 and 6.5
respectively. Prod2000 ranges in value from 183
to 105,064 and has a mean and median of 13,273
and 3,978 respectively. We note the presence of
skewness in Prod2000: this is caused by a few
countries having extremely high values thereby
increasing the mean. Finally, we note that values
of all three variables are higher in the latter time
period (2000).
Economic Freedom and the Impact of Technology on Productivity
Table 3. Correlations. This table reports correlations and p-values
Tech2000
Free2000
Prod2000
Tech1990
Free2000
.691
<.0001
Prod2000
.772
<.0001
.694
<.0001
Tech1990
.982
<.0001
.650
<.0001
.727
<.0001
Free1990
.703
<.0001
.794
<.0001
.758
<.0001
.664
<.0001
Prod1990
.816
<.0001
.665
<.0001
.961
<.0001
.775
<.0001
Table 3 reports correlations between model
variables. We note significant correlations between
each of the independent variables (technology
and freedom) and the dependent variable (productivity). We note that the correlation between
Tech2000 and Prod2000 is 0.772, and that the
correlation between Free2000 and Prod2000 is
0.694. These values are significant at the 1%
level. We note a similar pattern with 1990 values
with analogous values of 0.775 and 0.740 respectively.
Table 3 also indicates a strong relationship
between technology and freedom. Higher technology investments appear to be associated with
higher levels of freedom. For instance, the correlation between Tech2000 and Free2000 is 0.691
and significant at the 1% level. A practical concern
is that the high correlation creates problems of
multi-collinearity in the regression tests noted
below.
Finally, in Table 3, we note that productivity in
the year 2000 is significantly related to technology
and freedom values in 1990.
Regression Analysis and
Models of Productivity
Table 4 reports the main results of the study. All
models are statistically significant: the F-statistics
for all models have p-values (not reported) of less
Free1990
.740
<.0001
than 1%. Model A is a regression of Prod2000 on
Tech2000. Consistent with results reported in the
literature, we find high R-squares (0.6416) as well
as a significant coefficient for Tech2000 (t-statistic
of 14.45). We also find strong results with Model
B that uses Free2000 as the independent variable:
the coefficient has a t-statistic of 10.44. However,
the R-square of Model B is lower. Model C uses
Tech2000 as well as Free2000 as independent
variables. Here both coefficients are significant
(t-statistics of 8.36 and 3.64 respectively) but the
R-square for model C is only slightly higher than
the R-square for model A.
Table 4 Model D indicates another interesting
result. Here we test for the explanatory value of
an interacting variable. Tech*Free is the product
of Tech2000 and dummy variable indicating
whether Free2000 is above its median value. The
coefficient of Tech*Free is significantly positive
with a t-statistic of 2.71. In contrast, the coefficient
for Free2000 is insignificant with a t-statistic of
1.12.
Tests of Robustness
We repeat tests using data from 1990. In addition to providing another set of data for verifying robustness of results, data from 1990, when
compared to 2000, have the potential to reveal
trends in how productivity is determined. Table 5
7
Economic Freedom and the Impact of Technology on Productivity
Table 4. Regression Analysis Using Data from 2000, All Countries.
Dependent Variable: Prod2000
(A)
(B)
Intercept
-15898
-6.37
<.0001
Tech2000
87530
14.45
<.0001
Free2000
(C)
-74165
-8.48
<.0001
14030
10.44
<.0001
(D)
-43146
-5.50
<.0001
-21675
-1.97
0.0512
67147
8.36
<.0001
48624
4.69
<.0001
5397
3.64
.0004
2113
1.12
0.2642
Tech*Free
26254
2.71
0.0077
N
117
117
117
117
F
208.66
108.97
122.10
88.39
Adj. R2
.6416
.4821
.6762
.6933
For each regressor identified in the first column, we report the coefficient, t-statistic and p-value.
is analogous to Table 4 but uses data from 1990.
The results for the overall sample found in Table
5 confirm the results of Table 4. Overall, we find
a great deal of consistency between the results
using 1990 and 2000.
DISCUSSION OF DATA AND
RESULTS
Before turning to our main results, we note a few
patterns in the key independent variables. First,
values of technology as well as economic freedom
are higher in 2000 compared to 1990. The rapid
evolution of electronic commerce in the second
half of the 90’s resulted in its adoption by organizations and the general population worldwide. As
a consequence, various investments were made in
IT infrastructure in general and Internet technologies in particular, mainly in developed countries
but also in LDC. These investments may explain
why the technology variable has a higher score
in 2000. Other explanations for the high value
of technology in 2000 include Y2K related IT
8
investments and deregulation. As with technology, economic freedom is also higher in 2000. A
plausible explanation may be the sweeping political changes across the world (e.g., dismantling of
the Soviet Union); the resulting political freedom
may have led to economic freedom. An alternative explanation for higher values of economic
freedom is the success of global institutions such
as the WTO in instilling values of economic freedom and prosperity. Second, we find a significant
correlation between the technology and freedom
variables. Although possible, we do not assume
or infer a causal relationship between these two
variables. It is possible that the high correlation
between technology and freedom is caused by the
effect of other variables (e.g., education levels)
not studied in this research. Third, consistent
with expectations conditional on higher values of
technology and freedom, the value for productivity
is higher in 2000 compared to 1990. The higher
value of productivity supports suggestions in the
literature (Dedrick, Gurbaxani and Kraemer, 2003)
concerning (a) lagged effects of IT and (b) the IT
learning curve. Our data suggest that the gain in
Economic Freedom and the Impact of Technology on Productivity
Table 5. Regression Analysis Using Data from 1990
Dependent Variable: Prod1990
(A)
(B)
Intercept
-12088
-6.37
<.0001
Tech1990
85604
15.49
<.0001
Free1990
(C)
-29126
-7.42
<.0001
7547
11.33
<.0001
(D)
-24387
-8.27
<.0001
-11918
-2.41
.0179
62804
9.41
<.0001
47992
5.98
<.0001
3386
5.11
<.0001
943
0.93
.3566
Tech*Free
29520
3.07
.0027
N
108
108
108
108
F
239.97
128.46
161.55
119.50
Adj. R2
.6907
.5436
.7501
.7687
For each regressor identified in the first column, we report the coefficient, t-statistic and p-value.
productivity observed in 2000 may be attributable
in part to IT investments made by corporations
as well as governments in the 90s to embrace ecommerce and to face the Y2K issue.
In this part of the paper, we discuss the results
of our research models. We replicate and confirm
fundamental results relating productivity to technology and economic freedom. We find significant
relationships between technology and productivity as well as between freedom and productivity.
These results arise from long literature streams
in the IS and economics fields respectively. We
report correlations as well as regression coefficients consistent with these prior results in Tables
3 and 4 respectively. What is perhaps novel about
our results is the use of a more comprehensive
variable for technology, the ArCo variable. Also,
we are able to compare the relative importance of
technology and economic freedom as influencers
of productivity. We find that technology is a more
important variable. In Table 4, model A (using
technology as the independent variable) has a
higher R-squares than model B (using economic
freedom). We confirm this result with 1990 as
well as 2000 data. The relative importance of
technology is also confirmed with model C where
both technology and economic freedom are used
as independent variables: we note a marginally
higher R-squares compared to model A. Thus, the
addition of freedom in model C as an explanatory
variable does not appear to add much beyond the
use of technology.
A key result of our study is that freedom
modifies the effect of technology on productivity.
Model D adds the interaction variable indicating
how freedom modifies the effect of technology
on productivity. We note that this variable has a
significantly positive coefficient of 26,254 with
a t-statistic of 2.71; this indicates that countries
with higher levels of freedom have a greater link
between technology and productivity. Since the
coefficient of Tech2000 is 48,624 (Table 4), we
note that the response coefficient (productivity
related to technology) for high freedom countries
is 54% (54% = 26,254/48,624) greater than the
figure for low freedom countries. Our robust-
9
Economic Freedom and the Impact of Technology on Productivity
ness test using data from 1990 displays an even
greater response coefficient for high freedom
countries. This is consistent with Gompert (1998)
who argues that there is synergy between IT and
economic freedom. Arguably, the marginal effect
of IT investments is greater when companies in
a country have the freedom to creatively obtain
value from it. Freedom appears to influence
productivity through technology. Interestingly,
when the interactive variable is added in model
D, economic freedom as a stand-alone variable is
no longer significant. We know of no other study
indicating these results.
CONTRIBUTIONS OF THE STUDY
Our study contributes to the IS literature in two
ways. First, we build on recent efforts by Archibugi and Coco (2004; 2005) to measure the
technological capability of a country. We use the
ArCo index compiled by these authors to more
comprehensively measure the technological prowess of a country. Although the ArCo index and the
more traditional IT measurements are highly correlated, we feel that our study allows us to make
broader inferences about technology. Second, and
perhaps more importantly, we assess the impact
of economic freedom, particularly as a variable
that modifies the effect of technology on growth.
We draw on and complement the research stream
(predominantly in economics) that uses country
level data to examine the link between economic
freedom and economic growth.
Our results can be understood at two levels:
country level and firm level. Our tests use country
level data, so the main inferences are at the country
level. However, with appropriate assumptions,
we can also obtain inferences at the firm level.
Below, we explore these two issues. At the country
level, the main implication is that policy makers
should pay attention to the economic freedom
infrastructure. As our empirical variable indicates,
the extent of economic freedom in a country
10
depends on: (a) size of government (b) property
rights (c) sound money (d) trade barriers and (e)
regulatory barriers. By taking the proper steps to
create and shore up the institutions supporting
these variables (e.g., an independent central bank
is necessary for sound money), the conditions can
be created not only for technology investments but
also for bearing the fruits of such investment. We
realize that politics, tradition and culture have a
large role to play in this regard. It might be argued
that historical factors concerning its societal and
political landscape may predestine a country.
Nevertheless, most governments, especially in
democratic countries, are held accountable for
economic growth. Therefore, government officials
should know that it is in their own best interests
to build up the economic freedom infrastructure.
Our results do not just speak to government officials. Concerned and influential citizens of a
country can use our results to lobby for greater
economic freedom by articulating why it is useful. For instance, entities such as Chambers of
Commerce could be interested in our results.
Such activism may be especially valuable in
developing countries where it may not just be an
issue of shoring up institutions: the relevant task
is the much more difficult one of actually creating these institutions. Our results have also some
implications for international organizations such
as the United Nations agencies that are promoting IT investments in developing countries. In
order to make their actions more effective, these
organizations should collaborate with governments in developing countries to put into place
and strengthen institutions supporting economic
freedom. Such an environment will create favorable conditions to attract more IT investments as
well to maximize economic growth resulting from
these investments.
We now turn to firm level implications. This
is the age of global corporations. A typical large
corporation headquartered in the USA or in an
OECD country has operations in many foreign
countries. Our results suggest that a firm’s
Economic Freedom and the Impact of Technology on Productivity
technology investment strategy should take into
consideration the location of the investment.
Specifically, technological investments should
flow toward locations (countries) with greater
economic freedom. It is possible that this strategy
would produce better financial results (profits)
for the corporation. We realize that our data are
not firm-level and that our independent variable
is growth and not profit. This calls for caution in
making firm-level inferences.
RESEARCH LIMITATIONS
The current research has the following limitations.
Because we rely on cross-sectional associations,
we cannot make strong statements about causality.
Furthermore, our research design could be subject
to the problem of correlated omitted variables.
Nevertheless, our results are reasonable and
consistent with expectations. Another limitation
of this study concerns the data analysis technique.
Instead of using regression analysis, we could have
adopted different data analysis techniques such as
structural equations modeling. Additionally, this
study relies on data collected in 1990 and 2000.
Using data from different periods may have yielded
different results. Finally, our focus in the current
study is economic freedom only. We realize that
political freedom and economic freedom may be
synergistic, especially in developing countries.
Nobel laureate Becker (2007) notes that “private
property and open markets help economies grow,
which gives the political process a strong shove
toward democracy.” Other potential explanatory
variables are geography and culture.
CONCLUSION
Computerization and the Internet have changed the
world dramatically during the last few decades.
This has spurred interest in determining whether
IT in particular and technology in general has a
positive economic impact. A major concern has
been whether nations experience economic growth
as a consequence of investments in technology. A
well-developed stream of research has addressed
this issue. We contribute to this stream by exploring the role played by economic freedom in the
relationship between technology and growth.
Our principal finding is that technology creates
economic growth to a greater extent in countries
with higher levels of economic freedom.
We also validate the important role played by
technology in creating economic growth. While
both technology and economic freedom produce
economic growth, we find that technology is the
more dominant factor. Economic freedom does
affect growth, but this effect is primarily through
its interaction with technology.
ACKNOWLEDGMENT
An earlier version of this paper was presented at
the 18th Annual IRMA International Conference
(2007, at Vancouver, Canada) and later published
in the Journal of Global Information Management
(volume 17.3, 2009). In addition to the two authors
listed above, Koffi N’Da was also an author in
this earlier version. Ashok Robin is grateful for
the financial support of the Madelon and Richard
Rosett Chair for Research at the Saunders College
of Business, RIT.
REFERENCES
Ali, A. M., & Crain, W. M. (2002). Institutional
distortions, economic freedom, and growth. The
Cato Journal, 21(3), 415–426.
Archibugi, D., & Coco, A. (2004). A new indicator of technological capabilities for developed
and developing countries (ArCo). World Development, 32(4), 629–654. doi:10.1016/j.worlddev.2003.10.008
11
Economic Freedom and the Impact of Technology on Productivity
Archibugi, D., & Coco, A. (2005). Measuring
technological capabilities at the country level: A
survey and a menu for choice. Research Policy,
34, 175–194. doi:10.1016/j.respol.2004.12.002
Azman-Saini, W. N. W., Baharumshah, A. Z.,
& Law, S. H. (2010). Foreign direct investment,
economic freedom and economic growth: International evidence. Economic Modelling, 27,
1079–1089. doi:10.1016/j.econmod.2010.04.001
Becker, G. (2007, March 6). Economic and political freedom: Does one lead to the other? The
Becker-Posner Blog. Retrieved August 12, 2007,
from www.becker-posner-blog.com
Berggren, N. (2003). The benefits of freedom.
The Independent Review, 8(2), 193–211.
Bergh, A., & Karlsson, M. (2010). Government
size and growth: Accounting for economic
freedom and globalization. Public Choice, 142,
195–213. doi:10.1007/s11127-009-9484-1
Bjornskov, C., & Foss, N. J. (2008). Economic
freedom and entrepreneurial activity: Some crosscountry evidence. Public Choice, 134, 307–328.
doi:10.1007/s11127-007-9229-y
Bresnahan, T. F. (1986). Measuring the spillovers
from technical advance: Mainframe computers
in financial services. The American Economic
Review, 76(34), 742–755.
Brynjolfsson, E., & Hitt, L. M. (1996). Paradox
lost? Firm-level Evidence on the Returns to Information Systems Spending. Management Science,
42(4), 541–559. doi:10.1287/mnsc.42.4.541
Carlson, F., & Lundstrom, S. (2002). Economic freedom and growth: Decomposing the
effects. Public Choice, 112(3-4), 335–344.
doi:10.1023/A:1019968525415
Cole, J. H. (2003). The contribution of economic
freedom to world economic growth, 1980-99. The
Cato Journal, 23(2), 189–198.
12
Daveri, F. (2000). Is growth an Information
Technology story in Europe too? Working paper,
University di Parma, Parma, Italy.
Dawson, J. W. (1998). Institutions, investments,
and growth: New cross-country and panel data
evidence. Economic Inquiry, 36(4), 603–619.
doi:10.1111/j.1465-7295.1998.tb01739.x
Dedrick, J., Gurbaxani, V., & Kraemer, K. L.
(2003). Information Technology and economic
performance: A critical review of the empirical
evidence. ACM Computing Surveys, 35(1), 1–28.
doi:10.1145/641865.641866
Devaraj, S., & Kohli, R. (2000). Information
Technology payoff in the health-care industry:
A longitudinal study. Journal of Management
Information Systems, 16(4), 41–68.
Dewan, S., & Kraemer, K. L. (2000). Information Technology and productivity: Preliminary
evidence from country-level data. Management Science, 46(4), 548–562. doi:10.1287/
mnsc.46.4.548.12057
Goldsmith, A. A. (1997). Economic rights and government in developing countries: Cross-national
evidence on growth and development. Studies in
Contemporary International Development, 32(2),
29–44. doi:10.1007/BF02687323
Gompert, D. C. (1998). Right makes might: Freedom and power in the Information age. McNair
Paper 59. Defense Dept., National Defense University, Institute for National Strategic Studies.
Gwartney, J. D., Holcombe, R. G., & Lawson,
R. A. (2004). Economic freedom, institutional
quality, and cross-country differences in income
and growth. The Cato Journal, 24(3), 205–233.
Gwartney, J. D., & Lawson, R. A. (2002). Economic freedom of the world - 2002 annual report.
Washington, DC: Cato Institute.
Economic Freedom and the Impact of Technology on Productivity
Heitger, B. (2004). Property rights and the wealth
of nations: A cross country study. The Cato Journal, 23(3), 381–402.
Hitt, L. M., & Brynjolfsson, E. (1996). Productivity, business profitability, and consumer surplus:
Three different measures of information technology value. Management Information Systems
Quarterly, 20(2), 121–142. doi:10.2307/249475
Indjikian, R., & Siegel, D. S. (2005). The impact
of investment in IT on economic performance:
Implications for developing countries. World
Development, 33(5), 681–700. doi:10.1016/j.
worlddev.2005.01.004
Jorgenson, D. W. (2001). Information Technology
and the U.S. economy. The American Economic
Review, 91(1), 1–32. doi:10.1257/aer.91.1.1
Jorgenson, D. W., & Stiroh, K. J. (2000). Raising the speed limit: U.S. economic growth in the
information age. Brookings Papers on Economic
Activity, 1, 125–211. doi:10.1353/eca.2000.0008
La Porta, R., Lopez-De-Silanes, F., Shleifer, A.,
& Vishny, R. (2002). Investor protection and corporate valuation. The Journal of Finance, 57(3),
1147–1170. doi:10.1111/1540-6261.00457
Lee, S. T., Gholami, R., & Tong, T. Y. (2005). Time
series analysis in the assessment of ICT impact at
the aggregate level – Lessons and implications for
the new economy. Information & Management,
42(7), 1009–1022. doi:10.1016/j.im.2004.11.005
Lichtenberg, F. R. (1995). The output contributions of computer equipment and personnel:
A firm level analysis. Economics of Innovation and New Technology, 3(4), 201–217.
doi:10.1080/10438599500000003
Mbaku, J. M. (2003). Entrenching economic freedom in Africa. The Cato Journal, 23(2), 217–225.
Mbarika, V. W., Byrd, T. A., & Raymond, J.
(2002). Growth of teledensity in least developed
countries: Need for a mitigated euphoria. Journal
of Global Information Management, 10(2), 14–27.
doi:10.4018/jgim.2002040102
Melville, N., Kraemer, K., & Gurbaxani, V.
(2004). Information Technology and organizational performance: An integrative model of IT
business value. Management Information Systems
Quarterly, 28(2), 283–322.
Meso, P., Datta, P., & Mbarika, V. W. (2006).
Moderating Information and Communication
Technologies’ influences on socio-economic development with good governance: A study of the
developing countries. Journal of the American
Society for Information Science and Technology,
57(2), 186–197. doi:10.1002/asi.20263
Meso, P., Musa, P., Straub, D., & Mbarika, V.
(2009). Information infrastructure, governance
and socio-economic development in developing
countries. European Journal of Information Systems, 18, 52–65. doi:10.1057/ejis.2008.56
Nystrom, K. (2008). The institutions of economic freedom and entrepreneurship: Evidence
from panel data. Public Choice, 136, 269–282.
doi:10.1007/s11127-008-9295-9
Oliner, S. D., & Sichel, D. E. (1994). Computers and output growth revisited: How big is the
puzzle? Brookings Papers on Economic Activity,
2(2), 273–317. doi:10.2307/2534658
Oliner, S. D., & Sichel, D. E. (2000). The resurgence of growth in the late 1990s: Is Information
Technology the story? The Journal of Economic
Perspectives, 14(4), 3–22. doi:10.1257/jep.14.4.3
Osei-Bryson, K. M., & Ko, M. (2004). Exploring
the relationship between information technology
investments and firm performance using regression splines analysis. Information & Management,
42(1), 1–13.
13
Economic Freedom and the Impact of Technology on Productivity
Pook, L. A., & Pence, N. E. (2004). Evaluation of information infrastructures and social
development among the Visegrad-four countries
of Central Europe. Journal of Global Information Management, 12(2), 63–83. doi:10.4018/
jgim.2004040104
Roach, S. S. (1987). America’s technology dilemma: A profile of the information economy.
Morgan Stanley Special Economic Study.
Shih, E., Kraemer, K. L., & Dedrick, J. (2007).
Determinants of country-level investment in
Information Technology. Management Science,
53(3), 521–528. doi:10.1287/mnsc.1060.0670
Solow, R. (1987). We’d better watch out. New
York Times Book Review, (July 12).
Solow, R. M. (1956). A contribution to the theory
of economic growth. The Quarterly Journal of
Economics, 70(1), 65–94. doi:10.2307/1884513
14
Strassman, P. A. (1985). Information payoff: The
transformation of work in the electronic age. New
York, NY: Free Press.
Vega-Gordillo, M., & Alvarez-Arce, J. L. (2003).
Economic growth and freedom: A causality study.
The Cato Journal, 23(2), 199–215.
Yoo, S. H. (2003). Does Information Technology contribute to economic growth in developing countries? A cross-country analysis.
Applied Economics Letters, 10(11), 679–682.
doi:10.1080/1350485032000138971
ENDNOTE
GCR = Global Competitiveness Report;
ICRG = International Country Risk Guide
Economic Freedom and the Impact of Technology on Productivity
APPENDIX A
Table 5. Studies on IT Investment Implications at the Country Level
Study
Daveri (2000)
Dewan & Kraemer (2000)
Data sample
Independent Variables &
Operationalization
18 OECD and
European Union
(EU) countries
•
36 Countries
1985-1993
•
•
•
•
•
•
Jorgenson
(2001)
USA
Jorgenson and
Stiroh (2000)
USA
•
•
•
•
•
•
•
•
•
Lee, Gholami, and Tong
(2005)
20 developed
and developing
countries 19802000
•
•
•
Dependent variables &
Operationalization
IT Investment and Economic
Growth
IT: Hardware, Software, Communications equipment
Other Capital
Labor hours
Multifactor productivity
GDP growth
IT added to GDP growth in the
1990s for all countries studied, but
the contribution in EU countries
was smaller in 1992–1997 than
in other industrialized countries.
Within the European Union,
differences in IT contribution to
growth were also due to lower IT
investment.
IT capital stock including
Computer Hardware asset, data
communication asset, software
asset, and services asset
Non-IT capital stock
Annual labor hours employed
GDP Growth
A positive and significant relationship between IT capital investments and labor productivity was
found in developed countries but
not in developing countries.
IT: Hardware, Software, Communication equipment
Non-IT capital
Labor hours
Labor quality
GDP Growth
IT investment contributed more
than one half of the 1 percent
increase in economic growth
since 1995. About one half the
productivity growth since 1995
has occurred in the IT-producing
sector but growth has occurred in
IT-using industries as well.
IT Investments: Computer
hardware, Computer software,
Communication equipment
Other Capital
Labor hours
Labor quality
Multifactor productivity
Productivity growth
IT investment contributed one
half of GDP and labor productivity growth between 1995–
1999 and contributed moderately during earlier periods.
IT contributes to productivity in
the IT-using and -producing sector.
Labor: As measured by the
World Development Indicators (WDI)
Capital: As measured by the
World Development Indicators (WDI)
ICT investments: The International Telecomunication
Union’s (ITU) annual telecommunications investments
were used as a proxy for ICT
investment.
Economic growth as measured
by the GDP.
ICT contributes to economic
growth in developed countries and
newly industrialized economies,
but not in developing countries.
Developed countries are able to
gain positive and significant returns from ICT because they have
invested in these technologies over
a long period, and they have accumulated a substantial installed base
and complementary investments in
telecommunications. This is not
the case in developing countries.
continued on following page
15
Economic Freedom and the Impact of Technology on Productivity
Table 5. continued
Study
Meso, Datta,
and Mbarika
(2005)
Data sample
104 Countries
Independent Variables &
Operationalization
•
•
Dependent variables &
Operationalization
ICT: Operationalized as a
combination of broadcasting ICT and interactive ICT:
Broadcasting ICT measured
as the summated effects of the
average newspaper circulation density, radio ownership
density, and television ownership density (per 1000 people)
between 1997 and 2001;
Interactive ICT infrastructure
measured as the summation of
average computer ownership,
telephone density, and Internet
acesss (per 1000 people) between 1997 and 2001.
Governance: Measured with
six aggregated indices: Voice
ad accountability; Political
stability and absence of violence; Government effectiveness; Regulatory quality; Rule
of law; Control of corruption.
•
•
Social Development: Operationalized using a
summation of two indices:
Average rate of change in
life expectancy Literacy
between 1997 and 2001;
Literacy between 1997
and 2001.
Economic Development:
Measured as the average
change in GDP between
1997 and 2001
IT Investment and Economic
Growth
Contributions of ICTs to social
and economic development in developing countries are influenced
by socio-political governance.
Governance is perceived as exerting a contingent (moderating) role
on ICTs and national development
in developing countries.
Oliner and Sichel (1994)
USA
•
•
•
•
IT: Computer equipment
Other Capital
Labor hours
Multifactor productivity
Productivity growth
IT investment too small to have
substantial economic growth.
IT associated with 0.16–0.28 percent additional effects.
Oliner and Sichel (2000)
USA
•
IT capital: Computer hardware, Computer software,
Communication equipment
Other Capital
Labor hours
Labor quality
Multifactor productivity
Productivity growth
The contribution to productivity
growth from the use of information
technology — including computer
hardware, software, and communication equipment — surged in the
second half of the 1990s. In addition, technological advance in the
production of computers appears
to have contributed importantly to
the speed-up in productivity growth
between the first and second halves
of the 90’s.
Physical capital: Average of
annual ratios of real domestic
investment to real GDP as
measured by the World Bank
Human capital: Average of
percentage of the working-age
population that is in secondary school as measured by
UNESCO
IT capital: Average of annual
ratios of gross domestic IT
investment to nominal GDP as
measured by the International
Telecommunications Union
GDP per person of working age as measured by the
World Bank
IT investment has a significant
impact on the level of GDP per
capita in developing countries.
•
•
•
•
Yoo (2003)
56 developing countries
1970-1998
•
•
•
This table is obtained from Dedrick, Gurbaxani, and Kraemer (2003) and has been modified toinclude some more recent studies.
16
Economic Freedom and the Impact of Technology on Productivity
APPENDIX B
Areas and Components of the Economic Freedom of the World Index
1. Size of Government: Expenditures, Taxes, and Enterprises
A.
B.
C.
D.
General government consumption spending as a percentage of total consumption
Transfers and subsidies as a percentage of GDP
Government enterprises and investment as a percentage of GDP
Top marginal tax rate (and income threshold to which it applies)
2. Legal Structure and Security of Property Rights
A.
B.
C.
D.
E.
Judicial independence: The judiciary is independent and not subject to interference by the government or parties in disputes (GCR)
Impartial courts: A trusted legal framework exists for private businesses to challenge the legality
of government actions or regulation (GCR)
Protection of intellectual property (GCR)
Military interference in rule of law and the political process (ICRG)
Integrity of the legal system (ICRG)
3. Access to Sound Money
A.
B.
C.
D.
Average annual growth of the money supply in the last five years minus average annual growth of
real GDP in the last ten years
Standard inflation variability in the last five years
Recent inflation rate
Freedom to own foreign currency bank accounts domestically and abroad
4. Freedom to Exchange with Foreigners
A.
B.
C.
D.
Taxes on international trade
◦⊦
Revenue from taxes on international trade as a percentage of exports plus imports
◦⊦
Mean tariff rate
◦⊦
Standard deviation of tariff rates
Regulatory trade barriers.
◦⊦
Hidden import barriers: No barriers other than published tariffs and quotas (GCR)
◦⊦
Costs of importing: The combined effect of import tariffs, license fees, bank fees, and the
time required for administrative red-tape raises costs of importing equipment by (10 = 10%
or less; 0 = more than 50%) (GCR)
Actual size of trade sector compared to expected size.
Difference between official exchange rate and black market rate
17
Economic Freedom and the Impact of Technology on Productivity
E.
International capital market controls
◦⊦
i Access of citizens to foreign capital markets and foreign access to domestic capital markets
(GCR)
◦⊦
ii Restrictions on the freedom of citizens to engage in capital market exchange with foreigners - index of capital controls among 13 IMF categories.
5. Regulation of Credit, Labor, and Business
A.
B.
C.
18
Credit Market Regulations
◦⊦
Ownership of banks: Percentage of deposits held in privately owned banks
◦⊦
Competition: Domestic banks face competition from foreign banks (GCR)
◦⊦
Extension of credit: Percentage of credit extended to private sector
◦⊦
Avoidance of interest rate controls and regulations that lead to negative real interest rates
◦⊦
Interest rate controls: Interest rate controls on bank deposits and/or loans are freely determined by the market (GCR)
Labor Market Regulations
◦⊦
Impact of minimum wage: The minimum wage, set by law, has little impact on wages because it is too low or not obeyed (GCR)
◦⊦
Hiring and firing practices: Hiring and firing practices of companies are determined by private contract (GCR)
◦⊦
Share of labor force whose wages are set by centralized collective bargaining (GCR)
◦⊦
Unemployment Benefits: The unemployment benefits system preserves the incentive to
work (GCR)
◦⊦
Use of conscripts to obtain military personnel
Business Regulations
◦⊦
Price controls: Extent to which businesses are free to set their own prices
◦⊦
Administrative conditions and new businesses: Administrative procedures are an important
obstacle to starting a new business (GCR)
◦⊦
Time with government bureaucracy: Senior management spends a substantial amount of
time dealing with government bureaucracy (GCR)
◦⊦
Starting a new business: Starting a new business is generally easy (GCR)
◦⊦
Irregular payments: Irregular, additional payments connected with import and export permits, business licenses, exchange controls, tax assessments, police protection, or loan applications are very rare (GCR)
19
Chapter 2
The Influence of National
and Organizational Cultures
on Technology Use:
An Exploratory Study Within a
Multinational Organizational Setting
Zixiu Guo
University of New South Wales, Australia
John D’Ambra
University of New South Wales, Australia
ABSTRACT
This article presents the results of an exploratory study examining the influence of national and organizational cultures on technology use in a multinational organizational (MNO) context. Data were
collected from 121 respondents of a multinational organization in the Asia Pacific with headquarters in
Australia and three subsidiaries in Asia: Korea, Malaysia, and Thailand. Even though significant differences were found between respondents from headquarters and subsidiaries in terms of perceived media
richness and preference for telephone and written documents, very similar media preference patterns
for face-to-face and e-mail between the headquarters and subsidiaries were identified. Furthermore,
face-to-face and e-mail were two primary media used for most communication activities in this MNO.
Follow-up interviews revealed that the universal organizational culture of this MNO may explain media
use consistency between the headquarters and subsidiaries. Implications of the findings are discussed
and future research considered.
DOI: 10.4018/978-1-60960-605-3.ch002
Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
The Influence of National and Organizational Cultures on Technology Use
INTRODUCTION
Multinational organizations (MNOs) are characterized by their engagement in global markets
(Umanath & Campbell, 1994). In order to remain
competitive in today’s global marketplace, MNOs
have adopted advanced information and communication technologies (ICTs) as part of their
global communication networks to facilitate the
coordination of far-flung subsidiaries, suppliers,
and customers (Kogut, 2003; Sambharya, Kumaraswamy, & Banerjee, 2005). ICTs facilitate
the ability of the individual or the organization to
communicate easily and inexpensively across time
and geographic locations, rapidly and with greater
precision to targeted groups, and to selectively
control access and participation in communication
networks (Huber, 1990).
In spite of the enabling and facilitation of
globalization (Ford, Connelly, & Meister, 2003),
ICTs have also presented challenges to MNOs in
their implementation and use in subsidiaries due
to technical issues and social contexts (Robey &
Rodriguez-Diaz, 1989). There is accumulating
evidence that people of different national origins
have different perceptions of, and preferences
for, communication technologies (see Leidner &
Kayworth, 2006, for a recent review of cultural
impact on information systems (IS) research).
Scholars who study cross-cultural factors in ICT
use agree that understanding national culture (NC)
is critical (Gallivan & Srite, 2005). IS literature
focused on national cultural influence on the
adoption of IS has grown substantially over the
past decade (Niederman, 2005). However, little
research has been conducted into the effect of
culture on ICT use within MNOs, and its implications for individuals as well as organizational
performance (Gallivan & Srite, 2005). An MNO
is composed of a set of subsidiaries that operate
in distinct national contexts. Due to potential tension between employees’ host cultural values of
communication and the values of the MNO home
culture, national cultural influence on employees’
20
communication behavior will become an issue
when a subsidiary’s host culture differs from the
MNO’s home culture (Naisbitt, 1994). To understand why such a tension exists and how managers
of MNOs face this challenge, we must learn more
about the role that culture plays in our theories
of work behavior (Karahanna, Evaristo, & Srite,
2005). A better understanding of how employees
working for MNOs perceive and prefer communication technology for communication can
mean the difference between success and failure
in implementing new IT within MNOs (Ford et
al., 2003; Straub, 1994), as well as strengthening
the opportunity of improving organizational communication performance.
Understanding the relationship between ICT
and organizational culture (OC) has challenged
IS researchers for nearly three decades (Gallivan
& Srite, 2005). In their recent comprehensive
review of culture and IS research, both Gallivan
and Srite (2005) and Leidner and Kayworth (2006)
highlighted the key issues in the relationship between ICT and culture in order to lend insights
into our understanding of the linkage between ICT
and OC. However, as Gallivan and Srite (2005)
indicate, studies examining issues related to ICT/
NC fit and ICT/OC fit are separated into distinct
“stovepipes.” Based on social identity theory
(SIT), Straub, Loch, Karahanna, Evaristo, and
Srite (2002) proposed that NC and OC interact.
Similarly, Robey and Rodriguex-Diaz (1989)
identified that the social context of implementation of information systems, which is important
for a successful implementation process, included
both organizational and national cultural contexts.
They further noted that: “in the case of MNOs,
both of these contexts may become influential”
(Robey & Rodriguez-Diaz, 1989, p. 230) Given the
increasing dominance of MNOs and globalization
of world markets, where cultural fit is critical to
achieving the benefits of ICT (Gallivan & Srite,
2005), studying the interaction between NC and
OC in terms of technology use within MNO
contexts is a timely pursuit.
The Influence of National and Organizational Cultures on Technology Use
Furthermore, most cultural-related IS studies
usually focus on a selection of one communication medium, primarily e-mail, or on comparisons
between two media, often face-to-face vs. some
electronic medium (Watson-Manheim & Belanger,
2007), with one exception (Rice, D’Ambra, &
More, 1998). As O’Sullivan (2000) noted, use
of any one technology should be considered in
light of the repertoire of other media available to
fully understand when, why, and how any single
medium is used. Communication technology is
often used in conjunction with other media, and
this unusual relationship with other media is certainly deserving of further investigation (Hung,
Kong, Chua, & Hull, 2006).
This study was designed to explore cultural
effects on one of the core precepts of IS research―
communication media choice―examining how
employees from different subsidiaries of one MNO
perceived and preferred e-mail, along with other
available media, such as face-to-face, telephone,
and written documents for their daily communication, and to address two understudied areas in
global information systems: the lack of focus on
MNO contexts and the lack of focus on the interaction between NC and OC within the MNO contexts.
The lack of attention paid to the MNO context is
significant. On the theoretical side, examining
cultural impact on technology use within MNOs
can bridge the gap between the literature on global
information systems and international business, a
very important, but neglected research area (Ives
& Jarvenpaa, 1991; Lehmann & Gallupe, 2005).
On the practical side, the results of this study
can help managers of MNOs understand cultural
impact on their employees’ communication media use. Thus, MNOs can make better decisions
regarding adapting communication media to the
local environment or implementing a similar communication management strategy globally. The
lack of attention paid to the interaction between
NC and OC on ICT use within MNO contexts is
also significant because globalization and ICT
adoption and use have gone hand in hand with
the growing market power and geographic reach
of MNOs (Sambharya et al., 2005). Managers of
MNOs who are aware of the cultural implication of
ICT within a specific context can better use those
technologies to their firm’s advantage (Gallivan
& Srite, 2005). Such awareness is important to
shape the manager’s approach to planning, design,
and implementation of new forms of ICT globally
(Gallivan & Srite, 2005).
Our exploration was guided at a general level
by the current literature on media selection theories
and cultural differences. Given the limited research
examining the interaction between NC and OC in
media use, we conducted our exploratory research
by collecting data through a mailed survey, plus
a small-size field research design built around
mostly open-ended, in-depth interviews. This
method would give an opportunity to follow up
on unexpected findings, thus enhancing opportunities for refining and developing new theories.
The sample consisted of four subsidiaries of one
American-based MNO located in Australia, Korea,
Malaysia, and Thailand, in which the Australian
office is the company’s Asian-Pacific headquarters
and the other three offices are under the management of the Australian office.
The remainder of this article is organized
as follows. In the next section, we first briefly
explore media selection theories since they may
help us understand how and why people choose
different communication media. Then the concept
of culture, the cultural components and characteristics which distinguish the four countries, and
the literature suggesting links between aspects of
culture and media use are considered. Following
this section, we discuss the differences between
NC and OC, as well as their potential impact
on employees’ behavior. The research method
employed and the results are then presented.
The article concludes with a discussion of the
implications of the findings in terms of using
communication technologies in MNO contexts.
21
The Influence of National and Organizational Cultures on Technology Use
MEDIA SELECTION THEORIES
According to media richness theory (MRT) (Daft
& Lengel, 1986), media vary in their capacity to
transmit rich information, which refers to the ability of information to change understanding within
a time interval. Communication media are ranked
along a richness hierarchy based on criteria such as
speed of feedback, the form of language employed
(body, natural, and/or numeric), language variety,
and personal focus (Daft & Lengel, 1986; Daft,
Lengel, & Trevino, 1987). Face-to-face is considered the richest medium, followed in decreasing
order of richness by telephone, e-mail, written
personal, written formal, and the numeric formal
media. MRT proposes that efficient communication takes place when the match is perfect: the
medium has neither more nor less communication
capability than the task requires.
However, research shows that media attributes
(in this case, media richness) is not the only concern
in making sense of communication media selection (Fulk, Schmitz, & Steinfield, 1990; Yates &
Orlikowski, 1990). MRT, claimed as a rational
model of media choice, has led to inadequate
attention to the individual differences and social
contexts in which media choice and usage decisions are made. Decisions about media choice do
not occur in a vacuum; both decision makers and
media are socially embedded within organizational
settings, thus media perceptions and choices are
subjective and socially constructed (Fulk et al.,
1990). Studies suggest that examination of media
attributes independent of the broader communication contexts in which they are used may not be
enough (Flanagin & Metzger, 2001). As suggested
by some researchers, individual differences, participants’ media experience, and social context
surrounding them are also important in the assessment and selection of media, especially for new
media (e.g., Carlson & Zmud, 1999; Fulk, 1993;
Johnson & Kaye, 2003; King & Xia, 1997; Schmitz
& Fulk, 1991; Straub, 1994; Watson-Manheim
& Belanger, 2007). This study will focus on the
22
social context by firstly examining the national
cultural impact on media selection after controlling
for demographic variables, media use experience,
and media richness attribute.
NATIONAL CULTURE
AND COMPONENTS
Hofstede’s (1980, 1991) original taxonomy
describing culture along the dimensions of individualism/collectivism (I/C), uncertainty avoidance, power distance, masculinity/femininity, and
long-/short-term orientation has been recognized
as the most popular conceptualization of national
culture (Leidner & Kayworth, 2006). Some researchers criticize Hofstede’s work on the basis of
methodologies used and the validity of the data.
The central critique of his work is that it relies
on interviews with IBM employees conducted
between 1968 and 1973, thus raising serious questions about extending any of Hofstede’s findings
to national cultures (e.g., McCoy, Galletta, &
King, 2005). Yet, Hofstede’s framework has been
adopted and widely validated by more than 140
studies (Usunier, 1998) and forms the basis for a
significant proportion of the cross-cultural studies undertaken in various disciplines (Callahan,
2004; Yeniyurt & Townsend, 2003). In a recent
cross-cultural study of information systems which
examined how national culture, social presence,
and group diversity may affect majority influence
in group decision making, Zhang, Lowry, Zhou,
and Fu (2007) show that Hofstede’s construct of
the I/C appears to still hold. By adopting Hofstede’s original scales, authors found that Chinese
participants were significantly lower on the index
of individualism than U.S. participants. Zhang et
al.’s (2007) study supports Hofstede’s claim in his
updated work (Hofstede, 2001) that his cultural
model is still valid in the digital age. Hofstede’s
model thus is adopted as a theoretical framework
for this current study.
The Influence of National and Organizational Cultures on Technology Use
Table 1. Cultural dimension scores for four countries (based on Hofstede, 2001; Hofstede, Bond, &
Luk, 1993)
Dimensions/
Countries
ID
UA
PD
MA
LT
Korea
18L
85H
60M
39L
75H
Malaysia
26L
36L
104H
50M
Not available
Thailand
20L
64M
64M
34L
56H
Australia
90H
51L
36L
60H
31L
ID: individualism; UA: uncertainty avoidance; PD: power distance; MA: masculinity; LT: long-term orientation
H: top third, M: medium third, L: bottom third (among 53 countries and regions for the first four dimensions, among 23 countries for the fifth)
Hofstede (1980) defines culture as: “collective
programming of the mind which distinguishes the
members of one human group from another.” Numerous cross-cultural social psychology studies
have demonstrated that culture and communication technology are closely related (e.g., Earley,
1993; Hofstede, 1980).
Among cultural dimensions identified by
Hofstede, the cultural I/C dimension may explain
best how individuals might perceive and choose
different media for communication (Leidner &
Kayworth, 2006; Rice et al., 1998). Rice et al.
(1998) argued that the I/C dimension should be
most related to individuals’ different media choice
behavior. The I/C dimension, a basic distinction
among cultures (Singelis & Brown, 1995), has
been researched extensively in communication
studies (e.g., Gudykunst, 1997; Rice et al., 1998;
Singelis & Brown, 1995) and is the one dimension that reflects the fundamental contrast in
cultural orientations between Australian and the
three Asian countries (Ho, 1979). (See Table 1 for
Hofstede’s dimensions for the four countries of
interest in this study.) While there are differences
in the other four dimensions between Australia
and the other three countries, the contrast is most
striking on I/C. This study groups the three Asian
countries as a single Asian group. If cultural groups
do not have significantly different ranks on the
dimension, there will be no cultural variance in
the independent variables, and therefore it would
not be possible to test whether culture is a causal
explanation for differences in media choice behavior (Tinsley, 1994). In addition, although the
power distance dimension is included in several
cross-cultural media use studies (e.g., Mejias,
Shepherd, Vogel, & Lazaneo, 1996-1997; Tan,
Wei, Watson, Clapper, & McKean, 1998a; Tan,
Wei, Watson, & Walczuch, 1998b; Watson, Ho, &
Raman, 1994), power distance was mainly found
to play a significant moderating role to influence
individuals’ behavior in group decision making
through CMC, not in media choice. Hofstede’s
other cultural dimension, uncertainty avoidance,
was also found to play a significant role in communication technology diffusion and adoption
stages (Ford et al., 2003; Leidner & Kayworth,
2006; Thatcher, Srite, Stepina, & Liu, 2003;
Yeniyurt & Townsend, 2003). Since the main
focus of this study was to examine the technology use pattern after the technology was adopted,
uncertainty avoidance is not being considered as
part of this study. This article treats the cultural
dimension of I/C as an explanatory mechanism
for particular media choice behavior within
organizations. Hence, as a starting point, it will
be useful to explore the impact of this important
empirically established cultural dimension on the
differential media perception and choice behavior
between Australia and the three Asian countries.
The cultural I/C dimension is described by
Hofstede (1980) as a conglomeration of values
concerning the relation of an individual to his or
her collectivity in society. An individualistic so-
23
The Influence of National and Organizational Cultures on Technology Use
ciety is one in which self-concept is defined in
individual or trait terms, whereas a collectivistic
society is one in which an individual is defined
with reference to a societal and cultural context.
Collectivistic societies strive to maintain harmony and avoid confrontation or disagreement
among group members (Watson et al., 1994).
The need to preserve group harmony in a collectivistic society can be seen in the communication style. People in a collectivistic society favor
a high-context communication style (Gudykunst,
Matsumoto, & Ting-Toomey, 1996; Singelis
& Brown, 1995), in which most of the content
of messages is either in the physical context or
internalized in the person; very little is in the
coded, explicit part of the message (Hall, 1976).
In contrast, a low-context communication style is
predominant in an individualistic society (Singelis & Brown, 1995). It occurs when the greatest
amount of information is vested in the explicit
communication code (Hall, 1976).
The differences between individualism and
collectivism assist in anticipating different communication media choice behaviors between
Australian and Asian respondents. Australian
culture, a typical western culture, is characterized with individualistic values and low-context
communication orientation. Australians can be
said to prefer an explicit communication style,
be more rational than emotional, emphasize individual decision making, prefer fewer meetings,
and be less reliant on social cues in ambiguous
situations. They are expected to communicate in
ways that are consistent with their feelings (Hall,
1976) and prefer to be precise in communication
(Grice, 1975). They see e-mail as an opportunity
to share opinions frankly (Tan et al., 1998a) and
seek technology as a means for self-betterment
(Umanath & Campbell, 1994).
In contrast, Korea, Malaysia, and Thailand,
belonging to Asian cultures, are characterized by
high collectivism and high-context communication style. Such cultural characteristics promote
an implicit and ambiguous communication style,
24
more emotional than rational, with an emphasis
on group decision making, frequent meetings, and
relying on more social cues in ambiguous situations. Asian cultural communicators might prefer
synchronous media more, such as face-to-face,
because of their greater emphasis on traditional
use of time, and the increased value they place
on accessing and evaluating the contexts underlying communication (Rice et al., 1998). They are
expected to communicate in ways that camouflage and conceal the speakers’ true intentions
(Gudykunst & Ting-Toomey, 1988) to maintain
harmony in their in-groups. They may try to avoid
conflict through increased vagueness, rather than
by increased explicit communication (Rice et al.,
1998). They may see e-mail as a threat to group
harmony because e-mail allows loyalty and obligation to be challenged (Tan et al., 1998a).
INTERACTION BETWEEN NATIONAL
AND ORGANIZATIONAL CULTURES
The concept of culture could be studied at different
levels of analysis. In addition to conceptualizing
culture as a societal level (national level) phenomenon, as discussed above, the field of organizational behavior has addressed the construct
of culture at the organizational level, namely OC.
Hofstede (2001) defines OC as the manifestation of practices/behaviors evolving from the
shared values in the organization. In other words,
OC differs from NC with great emphasis placed
on the role of practices across organizations
such as rituals, heroes, and symbols. This view
is consistent with much of the popular literature
on OC that suggests OC is the pattern of basic assumptions that underlie the values and determine
not only behavior patterns, but also such visible
artifacts as architecture, office layout, dress codes,
and so forth (Schein, 1983). Similarly, Pratt and
Beaultieu (1992, p. 667) define OC in terms of
values shared by members of an organization unit
The Influence of National and Organizational Cultures on Technology Use
Figure 1. The balance of values vs. practices at the national, occupational, and organizational levels
(adapted from Hofstede, 2001)
or subunits, which manifests themselves in the
practices of that organization.
Recently, Karahanna et al. (2005) suggested
that values and practices are important elements
of culture, with values being closer to the core of
how individuals think, and shared practices being
more peripheral. Values are acquired early on in
life through childhood socialization and education,
whereas practices develop later in the life through
socialization at the workplace. A significant difference between values and practices is that values
are more stable and fairly hard to change, whereas
practices can be altered (Karahanna et al., 2005).
Similarly, researchers in cross-cultural studies argue that in any given culture values can
be categorized as either “core” or “periphery”
according to the characteristics of whether they
are more important, more enduring and resistant
to change, are highly accepted and agreed upon,
and hence more involved in social control or not
(Lachman, Nedd, & Hinings, 1994; Triandis,
1995). Core values are important in regulating
social behavior and tend to be enduring. In contrast, periphery values are relatively susceptible
to change. Early socialization affects core values,
and late socialization affects only periphery values
(Lachman et al., 1994).
These views are consistent with Hofstede’s
writings on values and practices of culture. Values
are invisible until they become evident in behavior,
whereas practices, which include symbols, heroes,
and rituals, are visible to an outside observer
(Hofstede, 2001). Figure 1, adapted from Hofstede
(2001), illustrates the relative importance of values
and practices at various levels of cultures. Values
are more important than practices at the national
level and practices for the organizational level.
Figure 1 also illustrates that values are acquired
mainly in the family and in the neighborhood, and
later at schools. On the other hand, organizational
practices are learned through socialization at the
workplace. In his IBM cross-cultural project,
Hofstede (2001) found considerable differences in
values, in spite of similarities in practices among
IBM employees in similar jobs but in different
national subsidiaries. While comparing organizational cultures across 10 organizations within one
country, Hofstede, Neuijen, Ohayv, and Sanders
(1990) found considerable differences in practices
but much smaller differences in values.
Collectively, literature suggests that values
and practices (or core values and peripheral values) are intertwined (Karahanna et al., 2005). The
relative balance of values and practices across
different levels of culture is different. And most
importantly, an individual’s practices would be
influenced primarily by OC (Karahanna et al.,
2005; Hofstede, 2001).
NC influences organizations through societal
structures such as laws and political systems
25
The Influence of National and Organizational Cultures on Technology Use
and also through the values, attitudes, behavior,
goals, and preferences of participants (clients,
employees, and managers) (Adler 1986). Because organizations interact continuously with
their environment, they cannot avoid taking on
aspects of the culture. Hence, organizations have
characteristics in common with their NC. In other
words, the values of founders and key leaders of
the organization appear to be reflected in the values
of their organization, even if the organizations
spread globally (Hofstede, 2001).
Under the assumption that OC is highly related
to NC, and given the fact that MNOs are trying to
develop a common set of management practices
across all subsidiaries in line with those of the
headquarters, the issues of conflicting cultures is
most relevant (O’Connor, 1995). Thus, it would
be useful if we can examine the relative strength
of each NC at the organizational level in terms
of their influences on management practice in
general (O’Connor, 1995; Pratt & Beaulieu, 1992)
and communication technology use practice in
particular. We might expect that the MNO’s OC in
terms of communication technology use practice
might reflect the norms and values associated
with the NC of the mother organization. Thus,
employees from different subsidiaries would
have similar communication and technology use
practice. Unfortunately, limited attention has been
paid to this issue. As a consequence, the following
research questions guided our study:
Research Question 1: Which media do employees
at both headquarters and subsidiaries prefer for
communication within an MNO context?
Research Question 2: How do national and
organizational cultures impact on employees’
media choice behavior within an MNO context?
METHODS
We chose to explore our research questions with
a mailed survey plus a field study built around a
series of interviews. Data collected from the survey
were used to examine how employees adopted
different media for communication and whether
national culture had any impact on media preferences. The interview results allowed for detailed
exploration of the important question of how NC
and OC interact on technology use.
The Investigation Site
This study was conducted in an American-based
multinational company, CEC (a pseudonym).
CEC operates in the food industry with offices
distributed around the world. The Australian office, the headquarters of CEC in the Asia Pacific
area, and its three Asian subsidiaries in Korea,
Malaysia, and Thailand were selected as the study
sites to ensure the comparability and homogeneity
of the sample. These four countries provided two
different types of cultural background in terms of
the cultural dimension considered in the present
study: namely Australia as a Western cultural
country characterized by individualism, and the
other three countries as belonging to the Eastern,
in particular, Asian traditional value system with
a collectivistic orientation.
The CEC Australian office has 150 staff, the
Korean, Malaysian, and Thai offices have 40, 40,
and 18 respectively. Most of the employees are
white-collar workers ranging from senior management to administrative workers. Computers and email are essential work and communication tools.
Data Collection
Survey
Quantitative data were collected by means of a
questionnaire sent by internal mail to all employees in all the Asia Pacific regional offices. Since
26
The Influence of National and Organizational Cultures on Technology Use
Table 2. Descriptive statistics of demographic variables and e-mail usage
Variables
Korea
Malaysia
39.00
34.06
Thailand
30.28
Asia*
34.56
Australia
33.47
Overall
Age (years)
Mean
34.12
S.D.
4.81
6.18
5.18
6.38
7.94
7.04
Work Tenure (years)
Mean
4.68
1.60
2.29
2.67
4.27
3.32
S.D.
4.94
1.74
1.17
3.21
4.28
3.75
Computer Experience
(years)
Mean
5.48
4.91
4.50
4.97
5.04
5.00
S.D.
1.21
.88
.786
1.02
.91
0.98
E-Mail Use (years)
Mean
5.12
5.14
3.97
4.84
5.69
5.19
S.D.
2.57
3.09
2.48
2.81
3.08
2.94
E-Mail Received (per week)
Mean
23.67
48.27
39.56
45.86
66.53
50.10
S.D.
13.40
29.39
49.31
84.58
51.57
43.80
E-Mail Sent (per week)
Mean
52.38
73.79
48.94
68.28
103.90
78.57
S.D.
35.31
41.57
55.34
90.15
18.99
64.17
N 18 33 21 72 49 121
Note: Asia includes Korea, Malaysia, and Thailand. Here is the mean of the three countries.
English is the work language throughout the organization, an English version of the questionnaire
was used throughout, thus minimizing translation
problems related to this cross-cultural survey. The
questionnaire was pre-tested on a small sample of
Australian subjects to determine the clarity and
relevance of the instrument. After two follow-up
letters, a total of 121 usable questionnaires (73%
response rate) were returned: 49 from Australia
(33%), 21 from Korea (52.5% response rate),
33 from Malaysia (82.5% response rate), and 18
from Thailand (100% response rate). Tables 2 and
3 show descriptive statistics and frequencies of
the samples, as well as the media usage variables.
Face-to-Face Interviews
In order to answer how and why questions, a series
of follow-up face-to-face interviews was conducted with eight employees from the Malaysian office
and five employees from the Australian office. The
interviewees were chosen to maximize variation in
hierarchical level, job type, and attitudes toward
communication media. Each of the interviews
was of 30 minutes duration. Using a structured
protocol, interviewees were asked about: (a) their
perceptions of different communication media,
such as e-mail, face-to-face, and telephone; (b)
their internal criteria used to judge which medium
should be used for an effective communication;
(c) written or unwritten rules about how people
should communicate within the organization;
(d) whether the individual’s supervisor or peers
influence the use of media for communication;
(e) whether the individual’s national culture influences the use of media for communication; and
(f) which culture, national or organizational, is
the dominant factor influencing the individual’s
communication behavior. The interviews were
conducted in English and recorded.
Measures
The previously used multi-item richness scale
(D’Ambra & Rice, 1994) was used to rate each
of the selected media’s perceived richness: faceto-face communication, the telephone, e-mail, and
written document. As a guide, Nunnally (1967)
advises that a modest reliability in the range of
0.5 to 0.6 will suffice. The alpha reliability for
27
The Influence of National and Organizational Cultures on Technology Use
Table 3. Frequencies of demographic variables and media accessibility
Korea
Malaysia
Thailand
Asian
Australia
Overall
Freq
Freq
Freq
Freq
Freq
Freq
Male
19
13
5
37
24
61
Female
2
20
13
35
25
60
3
1
4
4
8
Variable
Gender
Education
Organization
Level
Ethnicity
Category
High school or less
College Certificate/
diploma
2
11
1
14
11
25
University degree
13
17
15
45
26
71
Master’s degree
6
2
1
9
8
17
Senior executive
4
3
7
5
12
Manager
12
18
9
39
15
54
Professional
3
4
1
8
19
27
Admin
2
8
8
18
10
18
18
18
12
Thai
Malaysian
12
12
Chinese
16
16
4
20
Indian
5
5
1
6
Korean
21
21
Australian or Western Countries*
Telephone
Accessibility
E-Mail
Accessibility
28
21
44
44
1
4
Available with much
inconvenience
1
1
1
3
Available with some
inconvenience
1
2
7
10
Available with slight
inconvenience
2
2
8
12
2
14
Readily available
17
28
2
47
46
93
3
3
Available with much
inconvenience
10
3
Available with some
inconvenience
1
1
7
9
2
11
Available with slight
inconvenience
2
2
4
8
2
10
Readily available
18
30
4
52
45
97
*Note: Western countries include New Zealand, the UK, and the United States; Freq: Frequency
the four-item media richness scale was generally
satisfactory, ranging from .67 to .80.
Nine communication tasks developed by
D’Ambra and Rice (1994) were used to measure
media preference. Media preference was measured
by directly asking the respondents to specify their
first three preferences of media when they per-
28
form an intracultural communication with their
colleagues for each of these nine communication
tasks. For each task, for each medium, these rankings were scaled as 0=not chosen, 1=chosen 3rd,
2=chosen 2nd, and 3=chosen 1st (Rice et al., 1998;
Straub, 1994).
The Influence of National and Organizational Cultures on Technology Use
A three-item task equivocality scale, developed
by Goodhue (1995), was adopted to measure the
corresponding communication task equivocality.
The Cronbach’s alpha reliability for the three-item
equivocality mean scale was generally satisfactory,
ranging from .71 to .86.
The cultural I/C variable was measured at the
individual level using a nine-item scale derived
from Earley’s (1993) work. Two items were eliminated from subsequent analysis since they loaded
highly on two factors. The reliability (Cronbach’s
alpha) of the resulting scale was 0.71.
dimension of interest, rather than assuming that
our subjects were representative of their own
society based on Hofstede’s dimensions. The
results of such examination will be introduced
in the next section. Furthermore, in order to rule
out plausible rival hypotheses that may account
for the observed differences, all factors that may
have potential impact on technology use were
included in our statistical analysis.
Homogeneity of Each Cultural Group
Manipulation Checking of Group
Differences on Cultural Dimension
To make a valid comparison between two cultural
groups, it is necessary to establish the homogeneity
of each cultural group of respondents in terms of
their perceptions of and preferences for the technologies. At the same time, it is also necessary to
ensure that samples from two different cultures
are matched so that the samples are similar in all
respects except culture (Adler, 1984).
Majority participants from the Australian
sample were ethnic Australians, New Zealanders, English, and American, with only four of
them being of Chinese origin and one from India.
However these five of non-Western ethnicity have
resided in Australia for more than 20 years. Thus
it is reasonable to treat them as a homogonous
group. The Korean and Thai samples were all
ethnic Korean and Thai. Although two-thirds of the
Malaysian respondents were not ethically Malaysians, they have resided in Malaysia for more than
20 years. Thus, it is reasonable to treat them as a
homogonous group of Malays. Although Korea,
Malaysia, and Thailand are different in terms of
other factors, the explainable variable considered
in this study, cultural I/C dimension, was similar,
at least from Hofstede’s perspective. However,
in line with Myers and Tan (2002) and Gallivan
and Srite (2005), we explicitly determined the
culture values of our samples by measuring the
RESULTS AND DISCUSSION
A t-test analysis revealed a significant difference
in cultural I/C dimension between respondents
in the headquarters and those in the subsidiaries,
t (119) =-5.36, ρ<0.001. The Asian respondents
were more collectivistic (M=4.92) than those in
the Australian headquarters (M=4.31). An ANOVA
test also indicated that Australian respondents
were significantly different from all three Asian
countries (F (3,117) =11.65, p<0.001). Follow-up
analysis has shown that there were no differences
among the three Asian countries. Therefore, the
planned comparison could be made.
Media Preference Patterns Between
Headquarters and Subsidiaries
In order to answer research question one, we
first compared perceived media richness in each
culture and across cultures. The results are provided in Table 4. Four media were found to be
significantly different from each other, except in
Australia where face-to-face and telephone were
perceived to be similar in terms of media richness.
As found in prior studies, face-to-face was rated
as having the highest media richness, followed by
telephone, e-mail, and written document. The t-test
results demonstrated that there were significant
29
The Influence of National and Organizational Cultures on Technology Use
Table 4. Comparison of media richness across media and across cultures
Cultural Group
E-Mail_R Mean
Face_R Mean
Tel_R Mean
WD_R Mean
F-Value
Overall
4.04
6.17
5.33
3.36
170.07***
Asian
4.33
6.28
5.14
3.62
82.25***
Australian
3.61
6.02a
5.62a
2.97
111.23***
t value#
3.27**
3.26**
-2.41*
2.46*
E-Mail_R: e-mail richness; Face_R: face-to-face richness; Tel_R: telephone richness; WD_R: written document richness; N=72 for Asian
group and N=49 for Australian group
:
Indicates number is not significantly different from other numbers with the same subscript letter in the same row.
a
* p<.05, ** p<.01, *** p<.001
#
: t value shows the significance test after controlling demographics, media experience, and media accessibility variables.
differences between these two cultural groups in
terms of perceived media richness. Although the
ranking of media richness was in the same decreasing order of face-to-face, telephone, e-mail, and
written document for each cultural group, Asian
respondents rated face-to-face, e-mail, and written document as significantly richer, compared
to their Australian counterparts, while Australian
respondents rated telephone as significantly richer,
compared to their Asian respondents.
Table 5 shows the mean rating of media-task
preference in two different cultures. In general,
e-mail and face-to-face were the two most popular media and were rated highest for fulfilling
most communication activities, whereas telephone
was less popular than face-to-face and e-mail. In
any situation, written document was always the
least preferred medium for communication. In
other words, employees in this particular MNO
would always have a preference for face-to-face
or e-mail regardless of the nature of the communication task.
Influence of National Culture on
Media Preferences
In order to examine whether NC has any impact
on employees’ media preferences after controlling
for demographic variables, media use experience,
and media richness attribute, a hierarchical regression analysis technique was adopted. In using
30
hierarchical regression analysis, the demographic
variables, media use experience, and media richness perception were added as the first block, and
cultural group―dummy coded by choosing the
Australian sample as a reference group (Asian
group=1, Australian group=0)―was added as
the second block, thus statistically controlling for
the effects of control variables on the dependent
variable to determine the proportion of variance
attributable to culture (Pallant, 2001). For demographics and media-related variables, we included
age, gender, education background, work tenure,
organizational level, computer experience, media
availability, and media experience in our analysis
since they were significant factors identified in the
literature to influence media use. To save space,
only the final regression runs for each medium are
reported in Table 6, which presents standardized
coefficients that are significant.
For e-mail, accessibility was significantly
related with its preference. People’s perception
of e-mail also had significant impact on its preference: the higher the perceived richness, the more
it is used. Trevino, Webster, and Stein (2000) also
found a positive relationship between e-mail richness and its use. For face-to-face communication
and telephone, two traditional media, we did not
find any demographic and media-related factors
having significant impacts on their preferences.
Educational level was found to have significant
impact on written document preference: the
The Influence of National and Organizational Cultures on Technology Use
Table 5. Mean media preference rating for nine communication activities across cultures
Cultural
Group
T-E
E-Mail
FtF
Tel
WD
Request funding for sundry
expenses
Asian
4.34
1.82
1.51
.99
1.42
Aus
4.23
1.64
2.00
1.21
.95
Need approval for allocation of
funding
Asian
4.30
1.74
1.49
.87
1.51
Aus
3.98
1.86
1.84
1.05
.93
Discuss a departmental performance issue with your colleague
Asian
4.18
1.27
2.69
.96
.54
Aus
3.71
1.22
2.89
1.41
.30
Clarify with your manager on a
critical issue
Asian
4.03
1.49
2.61
1.11
.52
Aus
3.51
1.35
2.59
1.78
.16
Clarify a procedural matter with
your colleagues
Asian
4.02
1.50
2.18
1.47
.54
Aus
3.23
1.98
1.94
1.59
.29
Respond to a colleague’s urgent
request
Asian
3.85
1.44
2.26
1.89
.22
Aus
2.9
1.17
2.27
2.42
.04
Discuss a problem with your
manager
Asian
3.82
1.15
2.99
1.06
.44
Aus
3.01
1.08
2.92
1.71
.12
Respond to an urgent request by
your manager
Asian
3.81
1.58
2.14
1.42
.58
Aus
3.03
1.49
2.16
2.08
.12
Need some important figures from
another department
Asian
3.76
2.07
1.41
1.42
.76
Aus
3.12
1.89
1.50
2.26
.20
Communication Task
Aus: Australian; T-E: task equivocality; FtF: face-to-face; Tel: telephone; WD: written document
Table 6. Hierarchical regression results of four media on demographics, media experience, media richness perception, and national culture
Predictor
E-Mail
FtF
Telephone
WD
Age
--
--
--
--
Gender
Education
Work Tenure
Organizational Level
Computer Experience
------
------
------
-.18*
----
E-Mail Experience
E-Mail Accessibility
-.18*
n/a
n/a
n/a
n/a
n/a
n/a
Medium Richness
.23**
--
--
--
Block 1: Control Variables
Block 2: Culture
Culture
--
--
-.41***
.34***
Incremental R2
.08
.05
.24***
.18***
Change in R
.01
.001
.13***
.10***
2
Gender: 1=male, 2=female; Culture: Australian=0, Asian=1; N=121; *ρ< .1, **ρ< .05, ***ρ< .01
31
The Influence of National and Organizational Cultures on Technology Use
higher the degree, the more use of this medium
for communication.
After controlling for demographics, media
experience, and media richness perception, we
found significant cultural-level differences in
terms of telephone and written document preferences. Asian respondents preferred the use of
written documents, while Australian participants
favored the telephone. However, there were no
cultural-level differences between face-to-face
and e-mail preferences.
In summary, we did find significant culturallevel differences in terms of perceived media
richness. We also found cultural-level differences
for telephone and written document preferences.
However, we did not find any significant differences between these two cultural groups in terms
of face-to-face and e-mail preferences. In particular, we found that these two media were the two
primary media being preferred in most situations.
Face-to-face communication is a traditional
and quite stable element in organizational infrastructure, and a cultural effect, if it ever existed,
may have ceased to be meaningful over time
(Rice et al., 1998; Straub, 1994). The result here
provides empirical evidence to support this argument. Straub (1994) also found no significant difference in face-to-face communication preference
between Japanese and U.S. knowledge workers. In
addition, Chidambaram, Moe, and Olsen (1998)
found that face-to-face communication was used
to a greater extent than other modern media (such
as e-mail and fax) across various tasks. D’Ambra
(1995) found that face-to-face communication
was the most preferred medium over the range
of equivocality within an organization. Thus, it
is reasonable to say that face-to-face communication may be one of the most important media for
organizational communication, especially for this
particular MNO.
With regard to the first preference of e-mail
across various tasks, one explanation may be that
e-mail is of value to all respondents regardless
of the level of task equivocality. An additional
32
explanation may be found in social influence
theory (Fulk et al., 1990), which suggests that
highly educated and technically oriented people
perceive channel attributes (such as richness of
the media) based on social norms as opposed
to rational construction. Fulk and Boyd (1991)
reasoned that if social influences on media use
were more important than task influences, one
should observe similar patterns of individual
media use within work groups regardless of the
communication task’s characteristics. Critical
mass theory can also provide another possible
explanation for extensive e-mail use across an
organization. The critical mass perspective claims
that individuals who have access to multiple communication media will generally use the medium
most widely available within their communication
community, even when it is not the medium they
prefer. This practice enables them to communicate
with the largest number with the least amount of
effort or personal cost (Markus, 1987; Rogers,
1983; Zmud, Lind, & Young, 1990). E-mail is
the essential communication tool for the MNO
examined; virtually all company sites worldwide
can be reached through the e-mail system.
The above analyses, while helping to explain
the particular results obtained here, do not provide
clues in answering the overarching question: Why
was either face-to-face or e-mail always chosen
as the first preference for communication within
both headquarters and subsidiaries?
Influence of Organizational
Culture on Media Preferences
As we discussed earlier, the conflict between OC
and the individual’s own NC may become an issue
when the subsidiaries with an OC are influenced
by the headquarters located in a country with
a different NC (O’Connor, 1995). As literature
suggests, OC has predominant influence on
individuals’ practice development through socialization at the workplace. Such socialization
in Asian subsidiary workplaces may internalize
The Influence of National and Organizational Cultures on Technology Use
some Western influences, which in turn alter their
attitudes toward work and ICT to be assimilated
with their Western headquarters (Gallivan & Srite,
2005). People learn to accept some organizational
shared values which may not be congruent with
their own value systems. Even though people may
not change themselves internally (especially for
core values), there is some adaptation through
interaction between the individuals’ OC and their
NC. Ultimately, organizational socialization can
effectively override the influence of members’
previously internalized values, and change them
to conform to those required by the organization
(Lachman et al., 1994). For survival, employees
adjust themselves to suit the OC.
A comment made during the interviews by one
human resources manager (of Chinese cultural
origin) supports the role of socialization:
Basically, to me, it is very fine to accept American
culture…when I came here, I already expected
that sorts of attitudes, you know, being perceived
by the American. You know, I may work here in
my whole life, I am getting used to American
culture…being a person working here, you have
to, sometimes, compromise on something which
this broad party wants, you know, going to be
understanding each other. Organizational work
is teamwork. It is very important to put emphasis
on the team, not on yourself.
In their study of organizational and cultural
context of systems implementation, Robey and
Rodriguez-Diaz (1989) suggest that OC is most
relevant to management issues that focus on organizational tasks and technologies. Karahanna et al.
(2005) also argue that a strong OC might impact
the adoption, acceptance, and use of a mandated
new technology, such as e-mail systems.
Regarding the relationship between national
and organizational culture, another young Chinese
supply chain administrator said:
It is more organizational culture than national
culture. Everyone, I think, will follow organizational culture because when you get here, you
have to accept everything it has, and if you want
to work here, you have to adapt.
One sales administration manager complained,
saying that although there were too many e-mail
messages to reply to, he still used e-mail because,
overall, others used it to respond to him no matter
what medium he used to contact them. So, even
though employees have negative attitudes toward
e-mail, they may feel compelled to use it because
others do, or because it is expected in the organization. This sales administration manager said:
I know e-mail is good. But it can be abused. You
may not believe every morning I will see at least
100 new mails in my inbox. You know, some
people send a message to everyone no matter if
it is related to you or not. If you check and reply
all mails, it will take you whole morning to finish.
Not save time, it wastes my time. But if you don’t
do it, you may miss some important messages.
Because I know this is our policy to use e-mail as
basic communication tool, I use it. But, most of the
time, I will also use face-to-face or telephone to
contact people for double confirmation if possible.
In this MNO, e-mail is predominantly used.
However, for critical tasks (e.g. clarifying issues,
discussing issues, or equivocal tasks), face-to-face
communication is preferred. The organization
has an e-mail policy to minimize e-mail abuse.
When asked whether the organization has any
procedures or policy concerning media use, the
IT director said:
We actually, to be frank, did get what not to do
over e-mail, what to do over e-mail. We had this
policy established…most of time, we use e-mail
because you know, when you look at organizational
situation, it obviously has less critical situations,
compared with normal communication tasks.
33
The Influence of National and Organizational Cultures on Technology Use
Things should be done in that way, smoothly. It
happens to the critical situations, you know, it is
not effective to do by e-mail. Group meetings, or
face-to-face discussion can be used.
With regard to cultural influence on communication behavior, only one interviewee indicated
that NC did not influence his behavior. While the
quantitative data did not fully suggest the presence
of cultural effects as anticipated, the interview
data revealed that cultural influences were indeed
operating, but in ways not anticipated and overridden by the OC. One information technology
manager’s comment supports this finding:
Yes, national culture does play a role in guiding
your behavior. You know, certain culture may
have certain ways of communication, shy, not
question your boss, and so on…if you work in a
company, however, you have to accept that company’s culture because you have to be under that
roof. If you go to another company, you have to
adapt that environment concrete culture…to me,
the way I look at is, we are one person, okay, but
you have to behave ‘when’. You have to behave
differently when you are at home or when you are
at work...no matter what happens, if you want to
stay under this roof, you have to accept all cultures
in this environment…of course, you don’t throw
away your culture. You are not supposed to do
so, that is your identity. What you do is to adapt
it, not change yourself inside. But once again, if
you want to work under that roof, you have to do
so. Otherwise, you can’t stay.
In short, the data from the face-to-face interviews lent support to organizational cultural
influencing organizational communication within
this MNO. This influence is reflected in the similarities of preference for face-to-face and e-mail
media between headquarters and its subsidiaries
in countries with different cultures. These findings
are consistent with much of the popular literature
on OC that suggests that behaviors that involve
34
practices will more likely be influenced by organizational cultures (Hofstede, 2001; Karahanna et
al., 2005; Robey & Rodriguez-Diaz, 1989). The
interaction between national and organizational
culture is possible when the actual interaction
structures (OC) show effects of the NC on the home
organization (Soeters & Schreuder, 1988, p. 76).
LIMITATIONS AND FUTURE
RESEARCH DIRECTION
Several limitations and opportunities for future
research are noteworthy. The sample size of the
study is an issue. Indeed, a small sample size
reduces the external validity of this study and
provides less statistical power. Thus, one fruitful
direction for future research would be to replicate
this study with a larger sample. The sampling
design of this study restricts generalizability inasmuch as one cannot be sure that samples obtained
were representative of the culture. However, this
sample is deemed to be representative of the
cultural population of interest: English-speaking,
educated individuals from specific national cultures who have experienced more contact within
a multinational environment. It is to this group
that the findings of the study can be addressed.
It should be noted that the emphasis of this
research was on central tendencies. Cultural groups
are of course heterogeneous; there is within-culture
variation on any measure of culture or behavior,
even when sampling from just one stratum of
society. This research, however, did not focus on
the variations within one culture. Rather, the goal
was to test certain causal influences of cultural
membership, and with this goal it was necessary
to focus on central tendencies rather than on the
variation surrounding these tendencies.
This study examined media preference behavior by only considering intracultural, rather
than intercultural communication within MNO
settings (Adler & Doktor, 1986). Intercultural
communication might be especially difficult for
The Influence of National and Organizational Cultures on Technology Use
employees from cultures that have different media
richness perceptions or media preferences, such
as organizational employees from Australia and
Korea interacting through telephone or e-mail
(Rice et al., 1998). However, Ryan’s (1993) online
survey of 1,000 frequent e-mail users from 24
countries found that, although 40% of respondents
noticed some differences in how they interacted
online with people from other cultures, 75%
indicated that the cultural identity of the other
participants was of minimal or no importance.
The primary distinction was due to the medium,
as 95% reported differences in ICT compared to
face-to-face interaction.
Finally, the research has been limited to one
Western-cultural-based multinational organization
together with its Asian-cultural subsidiaries. There
is no evidence that an Asian-cultural-based multinational organization with Western-cultural subsidiaries would have identical behaviors. Hence,
the generalizability of this study’s findings beyond
Western-cultural-based multinational organizations must be viewed with caution. However, as
more and more Western-based MNOs shift their
business to Asia (Adler & Doktor, 1986), this
study’s results would help those MNOs’ managers
by providing a preliminary guideline for effective
communication within organizations.
Based on the experience with this study, some
directions for future research are proposed. Although organizational cultural influence was a
possible explanation for similar media preference
within this MNO, this evidence was obtained
within only one MNO through the interview
results. Further research is required in at least
two MNO settings to empirically examine the
interaction between NC and OC. Do organizations
need to modify their management strategy and
practice to suit the national cultural dictates of
their overseas host countries, where the national
culture of the host country differs from that of the
home country? The answer to this question could
be further achieved through examining the interaction between NC and OC within MNO settings.
In addition, this study provides a knowledge
base for a future research domain, which has yet
to be explored: how intercultural communication
differs from intracultural communication within
MNO settings. The cross-cultural communication
and psychology literature suggests that people behave differently with members of their own culture
than they do with members of foreign cultures
(Adler & Doktor, 1986). In addition, intercultural
communication, not merely intracultural, is the
essence of most international business managerial activity (Adler & Doktor, 1986). By knowing
how a particular cultural group communicates
with its own cultural members, an organization
is ready to confront the next step: to explore the
generalizability of these intracultural behaviors
and outcomes to the intercultural context.
IMPLICATIONS AND CONCLUSION
Drawing largely on media selection, NC and OC
literature in this article demonstrates the important potential influence of OC over NC in media
preferences, as well as illustrates empirically the
role of culture in individuals’ views of communication media. Thus, this article has pragmatic
importance for managers in organizations that
operate their businesses across Western and Asian
cultural contexts. There is a practical relevance
to managers in Western cultural multinational
organizations interested in understanding the differences they see in approaches to communication
media perceptions and choice behavior in Asian
countries. Although NC has been recognized
to influence individuals’ norms and behavior,
the development of a common set of norms and
values across all subsidiaries in line with those
of the headquarters is possible. Employees’ work
value can be influenced by the OC of the MNO
headquarters. In other words, engaging a universal
culture within one MNO can be achieved when
MNO managers can minimize the incongruence
between the national culture of subsidiaries and
35
The Influence of National and Organizational Cultures on Technology Use
the national culture of the “parent” organization,
which can be achieved through employing socialization (Soeters et al., 1988). Although these
findings have been obtained in one MNO, they
provide a clearer insight into the possibility of a
global form of organization where a seamless/borderless corporate culture can be achieved. These
research findings provide encouragement for the
practice of integrating multi-location operations,
whose employees may hold diametrically opposed
values, into a single corporate culture. As a result,
a similar information technology strategy could be
implemented throughout the organization.
D’Ambra, J. (1995). A field study of information
technology, task equivocality, media richness and
media preference. Unpublished doctoral dissertation, University of New South Wales, Australia.
REFERENCES
Daft, R. L., Lengel, R. H., & Trevino, L. K. (1987).
Message equivocality, media selection, and
manager performance: Implications for information systems. Management Information Systems
Quarterly, 11(3), 355–366. doi:10.2307/248682
Adler, N. J. (1984). Understanding the ways
of understanding: Cross-cultural management
methodology reviewed. In Farmer, R. N. (Ed.),
Advances in comparative management (pp.
31–67). Greenwich, CT: JAI.
Adler, N. J., & Doktor, R. (1986). From the Atlantic
to the Pacific century: Cross-cultural management reviewed. Journal of Management, 12(2),
295–318. doi:10.1177/014920638601200210
Callahan, E. (2004). Interface design and culture.
Annual Review of Information Science & Technology, 39, 257–310.
Carlson, J. R., & Zmud, R. W. (1999). Channel
expansion theory and the experiential nature of media richness perceptions. Academy of Management
Journal, 42(2), 153–170. doi:10.2307/257090
Chidambaram, L., Moe, C. E., & Olsen, D. H.
(1998). A study of factors influencing media
choice in Norwegian organizations. Proceedings
of the 31st Hawaii International Conference on
System Science.
36
D’Ambra, J., & Rice, R. E. (1994). Multimethod
approaches for the study of computer-mediated
communication equivocality, and media selection.
IEEE Transactions on Professional Communication, 37(4), 231–239. doi:10.1109/47.365170
Daft, R. L., & Lengel, R. H. (1986). Organizational
information requirements, media richness and
structural design. Management Science, 32(5),
554–571. doi:10.1287/mnsc.32.5.554
Earley, P. C. (1993). East meets west meets Mideast: Further explorations of collectivistic and individualistic work groups. Academy of Management
Journal, 36(2), 319–348. doi:10.2307/256525
Flanagin, A. J., & Metzger, M. J. (2001). Internet
use in the contemporary media environment. Human Communication Research, 27(1), 153–181.
doi:10.1093/hcr/27.1.153
Ford, D. P., Connelly, C. E., & Meister, D. B.
(2003). Information systems research and Hofstede’s culture’s consequences: An uneasy and
incomplete partnership. IEEE Transactions on Engineering Management, 50(1), 8–25. doi:10.1109/
TEM.2002.808265
Fulk, J. (1993). Social construction of communication technology. Academy of Management
Journal, 36(5), 921–950. doi:10.2307/256641
Fulk, J., & Boyd, B. (1991). Emerging theories of communication in organizations.
Journal of Management, 17(2), 407–446.
doi:10.1177/014920639101700207
The Influence of National and Organizational Cultures on Technology Use
Fulk, J., Schmitz, J., & Steinfield, C. W. (1990).
A social influence model of technology use. In
Fulk, J., & Steinfield, C. (Eds.), Organizational
and communication technology (pp. 117–142).
Newbury Park, CA: Sage.
Gallivan, M., & Srite, M. (2005). Information
technology and culture: Identifying fragmentary
and holistic perspectives of culture. Information
and Organization, 15(4), 295–338. doi:10.1016/j.
infoandorg.2005.02.005
Goodhue, D. L. (1995). Understanding user
evaluations of information systems. Management Science, 41(12), 1827–1844. doi:10.1287/
mnsc.41.12.1827
Grice, H. (1975). Logic and conversation. In Cole,
P., & Morgan, J. (Eds.), Syntax and semantics
(Vol. 3, pp. 107–142). New York: Academic Press.
Gudykunst, W. B. (1997). Cultural variability in
communication. Communication Research, 24(4),
327–348. doi:10.1177/009365097024004001
Gudykunst, W. B., Matsumoto, Y., & TingToomey, S. (1996). The influence of cultural
individualism-collectivism, self construals, and
individual values on communication styles across
cultures. Human Communication Research,
22(4), 510–543. doi:10.1111/j.1468-2958.1996.
tb00377.x
Gudykunst, W. B., & Ting-Toomey, S. (1988).
Culture and interpersonal communication. Newbury Park, CA: Sage.
Hall, E. T. (1976). Beyond culture. New York:
Doubleday.
Ho, D. (1979). Psychological implications of collectivism: With special reference to the Chinese
case and Maoist dialectics. In Eckensberger, L.,
Lonner, W., & Poortinga, Y. (Eds.), Cross-cultural
contributions to psychology. Amsterdam: Swets
& Zeitlinger.
Hofstede, G. (1980). Culture’s consequence:
International differences in work-related values.
Beverly Hill, CA: Sage.
Hofstede, G. (1991). Culture and organizations:
Software of the mind. London: McGraw-Hill.
Hofstede, G. (2001). Culture’s consequences:
Comparing values, behaviors, institutions, and
organizations across nations (2nd ed.). Thousand
Oaks, CA: Sage.
Hofstede, G., Bond, M. H., & Luk, C. L. (1993).
Individual perceptions of organizational cultures: A methodological treatise on levels of
analysis. Organization Studies, 14(4), 483–503.
doi:10.1177/017084069301400402
Hofstede, G., Neuijen, B., Ohayv, D. F., & Sanders,
G. (1990). Measuring organizational cultures: A
qualitative and quantitative study across twenty
cases. Administrative Science Quarterly, 35,
286–316. doi:10.2307/2393392
Huber, G. P. (1990). A theory of the effects of
advanced information technologies on organizational design, intelligence, and decision making.
Academy of Management Review, 15(1), 47–71.
doi:.doi:10.2307/258105
Hung, Y. T. C., Kong, W. C., Chua, A. L., & Hull,
C. E. (2006). Reexamining media capacity theories
using workplace instant messaging. Proceedings
of the 39th Hawaii International Conference on
System Sciences, Kauai, Hawaii.
Ives, B., & Jarvenpaa, S. L. (1991). Applications
of global information technology: Key issues for
management. Management Information Systems
Quarterly, 15(1), 33–49. doi:10.2307/249433
Johnson, T. J., & Kaye, B. K. (2003). Around
the World Wide Web in 80 ways. Social Science Computer Review, 21(3), 304–325.
doi:10.1177/0894439303253976
37
The Influence of National and Organizational Cultures on Technology Use
Karahanna, E., Evaristo, J. R., & Srite, M. (2005).
Levels of culture and individual behavior: An
integrative perspective. Journal of Global Information Management, 13(2), 1–20. doi:10.4018/
jgim.2005040101
King, R. C., & Xia, W. D. (1997). Media appropriateness—effects of experience on communication
media choice. Decision Sciences, 28(4), 877–910.
doi:10.1111/j.1540-5915.1997.tb01335.x
Kogut, B. (2003). The global Internet economy.
Cambridge, MA: MIT Press.
Lachman, R., Nedd, A., & Hinings, B. (1994).
Analyzing cross-national management and organizations—a theoretical framework. Management
Science, 40(1), 40–55. doi:10.1287/mnsc.40.1.40
Mejias, R. J., Shepherd, M. M., Vogel, D. R., &
Lazaneo, L. (1996-97). Consensus and perceived
satisfaction levels: A cross-cultural comparison of
GSS and non-GSS outcomes within and between
the United States and Mexico. Journal of Management Information Systems, 13(3), 137–161.
Myers, M., & Tan, F. B. (2002). Beyond models of
national culture in information systems research.
Journal of Global Information Management,
10(1), 24–32. doi:10.4018/jgim.2002010103
Naisbitt, J. (1994). Global paradox. New York:
Avon.
Niederman, F. (2005). International business and
MIS approaches to multinational organizational
research: The cases of knowledge transfer and IT
workforce outsourcing. Journal of International
Management, 11(2), 187–200. doi:10.1016/j.intman.2005.03.004
Lehmann, H., & Gallupe, B. (2005). Information
systems for multinational enterprises—some factors at work in their design and implementation.
Journal of International Management, 11(2),
163–186. doi:10.1016/j.intman.2005.03.003
Nunnally, J. C. (1967). Psychometric theory. New
York: McGraw-Hill.
Leidner, D. E., & Kayworth, T. (2006). Review: A
review of culture in information systems research:
Toward a theory of information technology culture conflict. Management Information Systems
Quarterly, 30(2), 357–399.
O’Connor, N. G. (1995). The influence of organizational culture on the usefulness of budget
participation by Singaporean-Chinese managers.
Accounting, Organizations and Society, 20(5),
383–403. doi:10.1016/0361-3682(94)00034-S
Markus, M. L. (1987). Toward a “critical
mass” theory of interactive media: Universal access, interdependence, and diffusion.
Communication Research, 14(5), 491–511.
doi:10.1177/009365087014005003
O’Sullivan, P. B. (2000). What you don’t know
won’t hurt me: Impression management functions
of communication channels in relationships. Human Communication Research, 26(3), 403–431.
doi:10.1093/hcr/26.3.403
McCoy, S., Galletta, D.F., & King, W.R. (2005).
Integrating national culture into IS research: The
need for current individual-level measures. Communications of the AIS, (15), 211-224.
Pallant, J. F. (2001). SPSS survival manual: A
step-by-step guide to data analysis using SPSS
for Windows (Version 10). St. Leonards, N.S.W.
Allen & Unwin.
Pratt, J., & Beaulieu, P. (1992). Organizational culture in public accounting: Size, technology, rank,
and functional area. Accounting, Organizations
and Society, 17(7), 667–684. doi:10.1016/03613682(92)90018-N
38
The Influence of National and Organizational Cultures on Technology Use
Rice, R. E., D’Ambra, J., & More, E. (1998).
Cross-cultural comparison of organizational media
evaluation and choice. The Journal of Communication, 48(3), 3–26. doi:10.1111/j.1460-2466.1998.
tb02757.x
Straub, D., Loch, K., Karahanna, E., Evaristo, J.
R., & Srite, M. (2002). Toward a theory based
measurement of culture. Journal of Global Information Management, 10(1), 13–23. doi:10.4018/
jgim.2002010102
Robey, D., & Rodriguez-Diaz, A. (1989). The
organizational and cultural context of systems
implementation: Case experience from Latin
America. Information & Management, 17(4),
229–239. doi:10.1016/0378-7206(89)90046-3
Straub, W. (1994). The effects of culture on IT
diffusion: E-mail and fax in Japan and the USA.
Information Systems Research, 5(1), 23–47.
doi:10.1287/isre.5.1.23
Rogers, E. M. (1983). Diffusion of innovations.
New York: The Free Press.
Ryan, R. (1993). Brief communications. The
Electronic Library, 11(3), 193–195. doi:10.1108/
eb045230
Sambharya, R. B., Kumaraswamy, A., & Banerjee, S. (2005). Information technologies and the
future of the multinational enterprise. Journal
of International Management, 11(2), 143–161.
doi:10.1016/j.intman.2005.03.005
Schein, E. H. (1983). The role of the founder in
creating organizational culture. In Staw, B. M.
(Ed.), Psychological dimensions of organizational
behavior. New York: MacMillan.
Schmitz, J., & Fulk, J. (1991). Organizational
colleagues, media richness and electronic mail:
A test of the social influence model of technology
use. Communication Research, 18(4), 487–523.
doi:10.1177/009365091018004003
Singelis, T. M., & Brown, W. J. (1995). Culture, self, and collectivist communication:
Linking culture to individual behavior. Human
Communication Research, 21(3), 354–389.
doi:10.1111/j.1468-2958.1995.tb00351.x
Soeters, J., & Schreuder, H. (1988). The interaction between national and organizational cultures.
Accounting, Organizations and Society, 13(1),
75–85. doi:10.1016/0361-3682(88)90026-8
Tan, B. C. Y., Wei, K. K., Watson, R. T., Clapper, D.
L., & McKean, E. R. (1998a). Computer-mediated
communication and majority influence: Assessing
the impact in an individualistic and a collectivistic
culture. Management Science, 44(9), 1263–1278.
doi:10.1287/mnsc.44.9.1263
Tan, B. C. Y., Wei, K. K., Watson, R. T., &
Walczuch, R. M. (1998b). Reducing status effects with computer-mediated communication:
Evidence from two distinct national cultures.
Journal of Management Information Systems,
15(1), 119–141.
Thatcher, J. B., Srite, M., Stepina, L. P., & Liu, Y.
(2003). Culture, overload and personal innovativeness with information technology: Extending the
nomological net. Journal of Computer Information
Systems, 44(1), 74–81.
Tinsley, C. (1994). The influence of culture on the
management of workplace conflict. Unpublished
doctoral dissertation, Northwestern University,
USA.
Trevino, L. K., Webster, J., & Stein, E. W. (2000).
Making connections: Complementary influences
on communication media choices, attitudes, and
use. Organization Science, 11(2), 163–182.
doi:10.1287/orsc.11.2.163.12510
Triandis, H. C. (1995). Individualism and collectivism. Boulder, CO: Westview Press.
39
The Influence of National and Organizational Cultures on Technology Use
Umanath, N. S., & Campbell, T. L. (1994). Differential diffusion of information systems technology
in multinational enterprises: A research model.
Information Resources Management Journal,
7(1), 6–18.
Yeniyurt, S., & Townsend, J. D. (2003). Does
culture explain acceptance of new products in
a country? An empirical investigation. International Marketing Review, 20(4), 377–396.
doi:10.1108/02651330310485153
Usunier, J. C. (1998). International and crosscultural management research. London: Sage.
Zhang, D., Lowry, P. B., Zhou, L., & Fu, X. (2007).
The impact of individualism—collectivism, social
presence, and group diversity on group decision
making under majority influence. Journal of
Management Information Systems, 23(4), 53–80.
.doi:10.2753/MIS0742-1222230404
Watson, R. T., Ho, T. H., & Raman, K. S. (1994).
A fourth dimension of group support systems.
Communications of the ACM, 37(10), 45–55.
doi:10.1145/194313.194320
Watson-Manheim, M. B., & Belanger, F. (2007).
Communication media repertoires: Dealing with
the multiplicity of media choices. Management
Information Systems Quarterly, 31(2), 267.
Zmud, R. W., Lind, M. R., & Young, F. W. (1990).
An attribute space for organizational communication channels. Information Systems Research,
1(4), 440–457. .doi:10.1287/isre.1.4.440
Yates, J., & Orlikowski, W. J. (1990). Mediated
communication in organizations: A theoretical
framework and research agenda. Proceedings of
the Conference of Sloan Working Papers, Cambridge, MA.
This work was previously published in International Journal of Global Information Management, Volume 17, Issue 4, edited
by Felix B. Tan, pp. 74-94, copyright 2009 by IGI Publishing (an imprint of IGI Global).
40
41
Chapter 3
Understanding IT Governance:
A Case of Two Financial Mutuals
Alastair Robb
University of Queensland, Australia
Michael Parent
Simon Fraser University, Canada
ABSTRACT
Legislators, regulators, and shareholders increasingly demand good governance over all aspects of
their business. While much is made of financial governance, most legislation and regulation implicitly
recognizes the need for prudent governance of information technology (IT) functions. In this study we
conduct an exploratory collective case study of IT governance (ITG) in two financial mutuals - one in
Australia and one in Canada, using a contextual lens. In one case, the mutual governs its IT through
Board participation in a subsidiary. In the second, governance is delegated to management and a Lead
Director. Both of these mechanisms appear to minimize ITG risk, and are the result of their respective
regulatory environments. This research begins to lend some clarity regarding IT governance choices by
firms, and denotes important contextual differences between countries’ regulatory environments. This
will allow researchers, managers, and directors to better understand and discriminate between ITG
processes and structures.
INTRODUCTION AND RESEARCH
OBJECTIVES
Legislators, regulators, and shareholders increasingly demand that directors and managers exercise
good governance over all aspects of their business.
Indeed, since the release of the Cadbury Report
(1992) into corporate governance, much effort has
DOI: 10.4018/978-1-60960-605-3.ch003
been put into ensuring governance remains high
on the agenda of directors and managers. Following such failures as Enron, Parmalat, One-Tel,
WorldCom, and HIH regulators, too, have made no
secret of their willingness to pursue directors and
managers who do not exercise good governance
over their organizations. As the inimitable Donald Trump stated (regarding the Sarbanes-Oxley
Act (SOX)): “They [regulators] make you sign
documents that basically say every transaction
Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
Understanding IT Governance
made by everybody in your company is 100 per
cent honest. If it isn’t, you go to jail for the rest of
your life” (Pitts, 2007). Further, in the wake of the
Global Financial Crisis, there is a real likelihood
that governance will, once again, resume centre
stage. In the latter part of 2010, The Basel Committee is expected to announce a series of new
initiatives designed to help avoid a repeat of the
Global Financial Crisis. Referred to as Basel III,
the initiatives focus on banks, their liquidity, and
capital (The Economist, 2010).
While much is made of financial governance,
implicit in most governance legislation and
regulation is the need for prudent governance of
organizations’ IT functions. Further, regulators
fail to distinguish between organization types in
mandating good governance.
In this study, we seek to better understand the
effect of different regulatory environments on IT
governance (ITG) frameworks by conducting an
exploratory collective case study of IT governance in two financial mutuals - one Australian
Credit Union and one Canadian Credit Union.
Both outsource their IT functions but use different approaches in doing so. The Australian
regulatory environment differs significantly from
the Canadian regulatory environment, largely
due to Canada’s proximity to the US. Unlike
Australia, the closeness of Canada and the US,
both geographically and commercially, tends to
result in US regulation more directly affecting
Canadian business interests. For example, the
Canada-US Free Trade Agreement has been in
place for over twenty years. In its preamble, one
of its objectives is “to ensure a predictable commercial environment for business planning and
investment” (Foreign Affairs and International
Trade Canada, 1987, p.5) The intention of the FTA
is that both the US and Canada pursue stable and
open markets, but it serves to underline that Canadian and Australian mutuals operate in structures
and environments that are contextually unique to
each country (Markus & Soh, 2002; Weisinger &
Trauth, 2002). Using a cross-national approach
42
to investigate the contextual differences between
Australian and Canadian ITG frameworks is warranted, particularly given the relatively similar
market sizes, but markedly different context in
which the two mutuals operate. Furthermore,
mutual organizations tend to fall under the same
regulatory framework as mainstream financial
institutions. Unlike mainstream financial institutions, however, mutual organizations have some
unique characteristics, largely arising from their
raison d’être. For example, their responsibilities
typically do not include maximizing shareholder
dividends, nor do mutual organizations have the
luxury of targeted recruiting for Board members.
In addition, as their boards are usually elected from
their membership, mutual organizations typically
experience elevated levels of director turnover. As
a result, mutual organizations represent unique,
if under-studied cases in ITG. Insights gained
from this study may help other non-mainstream
businesses.
The main objective of this research is to seek a
deeper understanding of the drivers behind adoption of different IT governance (ITG) mechanisms
for mutual organizations. While a substantial
number of prior studies have focused on ITG in
mainstream organizations (for example, Willcocks
et al., 2006; Weill & Ross, 2004), we are not aware
of any other studies that expressly focus on mutual
organizations operating in markedly different
national regulatory contexts. Weisinger & Trauth
(2002), in fact, note the lack of guidance available
to aid our understanding of context in the global
IT setting. More specifically, we investigate how
the different legislative and regulatory operating
environments of the focal organizations affect the
ITG mechanisms they adopt. This study seeks to
address the gap that exists between research on
ITG relative to mutual organizations in different
operational contexts. We expect this study will
help mutual organization directors in at least
three ways. First, it will reinforce the message
that directors of mutual organizations of the need
to ensure their ITG mechanisms remain strong,
Understanding IT Governance
particularly when IT functions are outsourced.
Second, it may help directors assess their current
ITG mechanisms and develop plans to address
shortcomings they may uncover. Third, it can help
directors understand the need to consider mutuals’
operating environments and contexts when planning and implementing ITG frameworks.
THEORETICAL FOUNDATIONS
In this section we develop the theoretical bases
for our investigation. First, we explain the importance of sound IT governance and pose our
research question. Next, we describe the theory
underpinning our study.
Directors and IT Governance
Addressing the Australian Institute of Company
Directors in August 2006, Chairman of the Australian Securities and Investments Commission
(ASIC), Jeffrey Lucy, stated that “Directors are
expected to meet an increasing range of obligations amidst commercial and legal expectations
of their actions.” (Lucy, 2006) He also indicated
that directors no longer are merely expected to
comply with the legalities of their position, but that
there are other expectations such as the efficient
conduct of their business, setting strategy for their
management to work towards, safeguarding the
assets of their business and, fostering an environment where instances of material fraud and error
are not present. Reinforcing Lucy’s (2006) address
and also speaking to the Australian Institute of
Company Directors, ASIC Commissioner Belinda
Gibson, outlining where global regulators stand
post-Global Financial Crisis indicated that “we
must promote market integrity – the attributes
of transparency and fairness…” (Gibson, 2009).
Information Technology (IT) has the potential
to be one of the most significant drivers of economic wealth. In many organizations IT is a critical
asset, not simply for organizational success, but
to provide opportunities to obtain competitive advantage (IT Governance Institute, 2003). Further, a
large portion of the market value of organizations
has transitioned from the tangible, (e.g., facilities,
inventory, etc.) to the intangible (e.g., information,
knowledge, expertise, reputation, etc.). However,
despite the large investments and potentially huge
risks associated with IT, boards typically focus on
business strategy and strategic risks, perhaps at
the cost of less effective IT Governance (ITG) in
the hope that nothing goes wrong (IT Governance
Institute, 2003).
While sound economic and technical reasons
may exist to pursue the outsourcing of IT functions,
the issue of governance of those outsourced IT
functions remains significant. ASIC’s message is
clear - outsourcing of IT functions will not absolve
directors of their IT governance responsibilities.
Shailer (2004) has defined governance as “…
decision-making in the exercise of authority for
direction and control” (p.11). This definition implies four interrelated principles: first, directors
should know the strategic direction the company
is pursuing. Second, they act, or make decisions.
Third, directors hold ultimate authority over the
affairs of the organization. Fourth, directors’ dutyof-care centers on oversight and control. Gillan
(2006) suggests that corporate governance consists
of a set of mechanisms internal and external to
the firm. The former includes control systems
(like information systems) that enable regulatory compliance. The Australian Stock Exchange
(ASX) uses Justice Owen’s interpretation of corporate governance, viz, “the framework of rules,
relationships, systems and processes within and
by which authority is exercised and controlled
in corporations.” (ASX, 2010). The ASX further mentions setting and achieving objectives,
assessing and monitoring risk, and optimising
performance. In sum, directors need to exercise
a duty of care, using both internal and external
mechanisms. Boards generally get themselves into
trouble, either through incompetence (which leads
to civil penalties) and/or dishonesty, which can
43
Understanding IT Governance
have criminal consequences (Colley et al., 2005;
Waring & Pierce, 2005).
Earlier research (Parent & Reich, 2009) has
noted that a plethora of possible ITG frameworks
exist – over 12 at last count, with more evolving.
These frameworks differ somewhat in their approach, for example, CoBIT, COSO, and ITIL
provide comprehensive guidance from the micro
level upwards. Given their focus, they also tend to
be fairly prescriptive. In contrast, ISO/IEC 38500,
the International Standard for ICT governance
(ISO/IEC 38500:2008 was prepared by Standards Australia as AS 8015:2005, the Australian
Standard for ICT governance, hence, the two
standards are, with minor wording differences,
identical) is targeted at the strategic level. Its focus
is more macro level and discretionary, offering
principles rather than prescription (see Appendix
A for an overview of current ITG frameworks).
While these frameworks differ relative to their
focus, they still have a single common goal: the
good governance of organizational IT. Given the
number of alternative ITG frameworks, however,
it is fair to conclude that no single approach to IT
governance has emerged. Rather, recent research
has conceived of IT Governance as having two
distinct modes: defensive and strategic (Parent &
Reich, 2009). Defensive ITG seeks to fire-proof
the organization by preventing or mitigating the
consequences of disasters. Strategic ITG, on the
other hand, aims to create sustainable shareholder
value by either reducing costs (such as the cost
of capital, or of IT projects) or creating a sustainable competitive advantage. We contend that
governance legislation and regulations help shape
organizational responses to ITG, as do the particular organizations’ approach to their IT functions.
That is, organizations’ ITG mechanisms evolve
so they align with the legislation and regulation
prevailing in their particular jurisdiction and their
methodology relative to sourcing IT resources.
Continuing this line of reasoning we propose the
following principal research question: “To what
extent does regulatory context affect the ITG of
44
mutual organizations when their IT resources are
outsourced?”
Outsourcing of IT Functions
In common with others in the same business
environment, outsourcing of IT functions has
penetrated the finance sector. While popular
opinion suggests IT outsourcing is advantageous,
the evidence on outsourcing remains contentious
(e.g., Berg & Stylianou, 2009; Caldwell, 1997;
Clemons et al., 1993; Gwebu et al., 2010; Kobelsky & Robinson, 2010; Lacity & Willcocks,
1998; Subramani & Walden, 2001). Much of the
literature on outsourcing centers on transaction
cost economics (Walden & Hoffman, 2007).
Transaction cost economics’ premise is that
firms engaging in outsourcing are attempting to
leverage economies of scale and thus cut down
on transaction costs (Walden & Hoffman, 2007).
The transaction-based view of the firm has
been augmented by resource-based views that
emphasis the importance of resources in guiding
firm activity and the management of a firm’s
various capabilities when seeking competitive
advantage (Mata el al., 1995; Holcomb & Hitt,
2007). The resource-based view focuses on the
importance of the relative capabilities of focal
firms and their outsource partners (Holcomb &
Hitt, 2007). The resource-based view posits that
firms are largely heterogeneous relative to their
resources and capabilities and hence carry out
the same activity with different efficiencies and
costs (e.g., Barney, 1991; Wernerfelt, 1984). Accordingly, individual firms that can accomplish
the same activity in different ways can achieve
different efficiencies and costs (Holcomb & Hitt,
2007). Given the potential to reduce costs and
leverage service providers’ expertise, organizations might appropriately consider outsourcing
their IT functions to improve their competitive
position. Organizations in the finance sector
may find outsourcing IT functions particularly
Understanding IT Governance
appealing especially where cost and expertise
considerations dominate.
The simplest form of outsourcing involves the
decision to outsource existing production, with
capabilities provided by intermediate markets
(Gilley & Rasheed, 2000). By contrast, organizations may decide to outsource ex ante, acquiring
capabilities from intermediate markets rather than
incurring the expenses to internalize production
(Gilley & Rasheed, 2000). Walden & Hoffman
(2007) make the point that one of the primary
steps in the outsourcing decision is evaluation.
Organizations considering outsourcing must
understand the circumstances of the decision to
outsource. That is, unless prospective outsourcing
partners can provide a product that is superior,
outsourcing is not a viable option.
Relative to outsourcing IT functions, by implication then, ensuring strong ITG should form
part of any outsourcing decision. To reinforce
this point, Australia’s corporate regulators, ASIC
and APRA, have clear messages to organizations
- outsourcing of IT functions will not absolve
directors of their ITG responsibilities.
IT Governance
As mentioned above, there is a plethora of IT
governance frameworks, which tends to suggest
that there is no one solution to good IT governance.
Further, defining what IT governance is, in an
agreed and clear fashion has proved elusive (Van
Grembergen, 2004). What appears to be generally
agreed is that implementing and using ITG minimizes risk to the organization associated with IT
investments. ISO/IEC 38500 (2008) suggests that
implementing sound ITG frameworks minimizes
risks to directors and management of organizations in two ways. First, sound ITG frameworks
reduce the risk of director and management of not
fulfilling their responsibilities to conform to the
laws in force. Second, sound ITG frameworks help
minimize the risk of not obtaining the best return
on the organizations’ investment in IT. While ISO/
IEC 38500 (ISO/IEC, 2008) provides no specific
procedure for ITG, it does suggest a framework
of ITG. That framework includes three main elements. First, the evaluation of the pressures on
the business, and, taking those into account, the
current and future business needs relative to IT.
Second, the preparation and implementation of
plans and policies for the implementation of the
required IT investments. Third, the monitoring
of the conformance and performance of the IT
implementation relative to the plans and policies. In other words, ITG may be conceived as
consisting of structure, process, and measurement
(Bowen et al., 2007). Such a framework: structure,
process, and measurement, provides an appropriate and meaningful means of analyzing the ITG
mechanisms of our focal organizations.
IT Governance Decisions
As Agarwal & Sambamurthy (2002) and Zain et
al. (2005) note, IT plays a more prominent role in
corporate agility, helping enable rapid and continual business innovation. Implicit in Agarwal
& Sambamurthy’s (2002), and ISO/IEC’s (2008)
view, is the notion that there can be no one solution to ITG, particularly in light of organizations’
need to respond rapidly to their unique operating
environments.
Like Agarwal & Sambamurthy (2002), Brown
(1997) also contends that there is no universal
“best” solution to ITG. Analogous with this notion,
we contend that different legislative jurisdictions,
likewise, oblige organizations to pursue ITG governance mechanisms that best suit their particular
business environments. Brown (1997) asserts that
the best ITG solution for a firm is contingent on
its organizational context.
Brown’s (1997) work on predictors of systems
development governance at the business unit level
is informative and compelling. By inference, and
intuitively, it provides guidance for predicting
ITG mechanisms in different legislative jurisdictions. Brown (1997) compares centralized versus
45
Understanding IT Governance
decentralized systems development governance
in organizations. In her study, centralized refers
to uniform, enterprise-wide governance solutions,
i.e., locus of decision-making authority belongs
to a central unit, while decentralized refers to
governance solutions for a specific business unit,
i.e., the locus of decision-making authority belongs
to the particular business unit.
A pertinent comparison between centralized
and decentralized governance in legislative
jurisdictions could be made between the US
and Australia. The US Securities and Exchange
Commission’s mission is to “protect investors,
maintain fair, orderly, and efficient markets, and
facilitate capital formation” (SEC, 2010). Of
the laws within the SEC’s remit, the SarbanesOxley (SOX) legislation is probably the most
recognized. In particular, Section 404 mandates
organizations’ annual reports contain an “internal
control report”. Further, SOX requires identification of the systems which will have a material
effect on financial statements (Pinder, 2006).
Typically, organizations affected by SOX make
such assessments against the COBIT or COSO
frameworks (Pinder, 2006). Both frameworks
are highly prescriptive with much emphasis on
micro-level controls. McFarlane (2005) notes that
COBIT involves 34 processes, it also details each
process’s control objectives and critical quality
criteria. Parent and Reich (2009) observe that
from a Board perspective, many IT governance
frameworks are too simplistic and high-level or,
so detailed and technically intricate, that Boards
fail to implement them due to their complexity.
Given the nature of SOX, and as put forward
by Pinder (2006), ITG frameworks like CoBIT and
COSO are more likely than others to be selected
as the ITG framework of choice for organizations
affected by SOX. Accordingly, such organizations
are likely to have little discretion with regard to
implementing their ITG mechanisms. In brief,
organizations operating in jurisdictions affected
by SOX (including SOX by proxy) are analogous
46
to Brown’s (1997) firms that maintain centralized
systems of control.
Like the SEC in the United States, ASIC,
Australia’s corporate regulator is charged with
protecting consumers, investors, and creditors
by enforcing and regulating company and financial services laws (ASIC, 2010). ASIC regulates
“Australian companies, financial markets, financial services organisations and professionals who
deal and advise in investments, superannuation,
insurance, deposit taking and credit”. APRA,
Australia’s prudential regulator’s, mission is to
“establish and enforce prudential standards and
practices designed to ensure that, under all reasonable circumstances, financial promises made
by institutions we supervise are met within a
stable, efficient and competitive financial system”
(APRA, 2010).
While both ASIC and APRA have the task of
ensuring good governance in Australian companies, the (then) Treasurer of Australia in his
Statement of Expectations to APRA states, “The
Government’s preference is for the avoidance
of unnecessarily prescriptive regulation and for
APRA’s prudential regulation to identify the
outcomes that are desired from financial institutions rather than prescribe how those outcomes
should be achieved” (Costello, 2007). Clearly, the
Treasurer’s expectations are for compliance and
good governance, but the governance frameworks
and mechanisms used are at the organizations’
discretion.
This discretionary theme is reflected in ISO/
IEC38500, the International ICT Standard (formerly AS8015). ISO/IEC38500 focuses on Boardlevel governance mechanisms, consequently ISO/
IEC38500 focuses on strategy and outcomes rather
than prescribing operational-level processes to
accomplish sound governance (ISO/IEC, 2008).
Governance in Australia and by inference, ITG,
both by expectation and legislation, falls within
the remit of the individual organization. While the
regulators make clear their desire for good governance, its achievement is left to the individual
Understanding IT Governance
firms. The Australian approach thus, appears to
be substantially consistent with Brown’s (1997)
description of decentralized governance solutions,
i.e., decision-making authority residing locally.
Harking back to our primary research question,
viz., “how the different legislative and regulatory
operating environments of the focal organizations
affect the ITG mechanisms they adopt” we posit
that there should be substantial differences in ITG
mechanisms between different business environments. That is, where the jurisdiction’s locus of
decision-making control is more centralized and
prescriptive, we expect implementations of ITG
mechanisms that closely conform to ‘black letter’
laws, e.g., SOX or SOX by proxy. In contrast,
where the locus of decision-making resides locally,
we expect implementations of ITG mechanisms
that reflect the ‘spirit’ of the prevailing laws, e.g.,
Australian Prudential Standard APS 510.
To investigate “…how the different legislative
and regulatory operating environments of the focal organizations affect the ITG mechanisms they
adopt” we undertook a cross-national multi-case
study. The methodology is discussed in the following section.
RESEARCH METHODOLOGY
Our investigation focuses on two credit unions in
different countries. These credit unions have many
similar characteristics, in that they both outsource
their IT operations, however, both have pursued
different models in doing so. While there are many
similarities between the two credit unions in this
study, their respective operating environments
differ significantly. The Canadian credit union,
because of its geographical and commercial
proximity to the US finds itself being influenced
by the US SOX governance legislation. On the
other hand, the Australian credit union’s primary
influence is the governance legislation enforced
by APRA. While both the SOX and APRA seek
the same goal, that is, good governance, as men-
tioned previously, the SOX legislation tends to be
more prescriptive than the legislation enforced by
APRA. We would expect to observe noticeable
differences between the ITG approaches adopted
by the Canadian and Australian credit unions.
The primary research question of this study
is to discover why different legislative and
regulatory operating environments affect the ITG
mechanisms an organization adopts. We draw on
an array of documentary evidence, as suggested
by Yin (2003). For example, current ITG frameworks, published governance guidelines, and prior
research on ITG. We contend, however, that to
find out why different legislative and regulatory
operating environments affect ITG we needed
to get data from people directly involved in the
execution of ITG. We posit, therefore, that a case
study methodology is appropriate to this research.
Yin (2003) notes that case study methodologies
are useful when trying to answer ‘who’, ‘what’,
‘where’, ‘how’, and ‘why’ type questions. Further,
we adopted a case study approach as case studies
permit the investigation of real-life events such
as organizational and managerial processes (Yin,
2003). Case studies are appropriate where the
boundaries between context and phenomenon are
not clearly evident (Yin, 2003). Case studies are
generalizable to theoretical propositions, if not
to populations (Yin, 2003). While we study only
one credit union in each country, the legislation
affecting each credit union is largely uniform
within the sector. Accordingly, we believe the
consequences of the prevailing regulation and
legislation is generally applicable within each
jurisdiction. Lastly, by engaging in a collective
case study, we will be able to potentially discern
a richer description and understanding of the
phenomenon (Stake, 1995).
Specifically, we used a cross-national approach
that investigated the cases of two mutuals. Regulatory environment aside, as outlined in Table 1, the
Canadian and Australian financial sectors share
many traits and similarities. The countries have
comparable ownership concentrations, market
47
Understanding IT Governance
Table 1. Economic Comparison between Australia and Canada
Characteristic
Australia
Canada
Ownership concentration (mean % held by top 3 shareholders for top 10 companies)
28
40
Market capitalization (mean USD million for top 10 companies.)
5,943
3,015
GNP per capita (USD)
17,500
19,970
Table 2. Key characteristics
Characteristic
Canadian
Australian
Location
British Columbia
Queensland
Number of Directors
12
7
Term of directorship
Varied (staggered appointments)
3 years before standing for election
Number of branches
17
7
Asset base
CAN3.4B
AUD135M
Primary Market
Savings, mortgages
Secured housing loans
Committees
Executive, Audit
Executive, Audit, Risk Management, Director Nomination
Regulator: Risk
Canadian Public Accountancy Board
APRA for Prudential Risk Management
Regulator: Compliance
Province of BC
ASIC for adherence to Corporations Act, Accounting
Standards disclosures in the financial report, and Financial Services Reform requirements
Auditors report to:
Members, CPAB
APRA, ASIC, Audit Committee
capitalizations, and GNP per capita (La Porta et
al., 1998). They are also notably culturally similar (Hofstede, 2010). Further, as noted in Table
2, there are also many similarities between the
two mutuals that are the focus of this study. We
posit that the two mutuals are sufficiently similar
to permit meaningful comparison of the effects
of their respective operational contexts on their
ITG frameworks. The access to the organizations became available to the authors through
acquaintance with the managers and boards of
the organizations. The data were collected at
the board and corporate level from the two focal organizations. Open-ended interviews were
conducted with the CIO and board members of
the Canadian mutual. The GM and Chairman of
the Australian mutual were interviewed using
the same open-ended interview technique. The
interviews lasted some 90 minutes, and covered
48
topics including regulation, outsourcing, and IT
governance. To ensure the information offered by
the interviewees’ was adequately reflected, we also
drew on observations, field notes, and secondarysource documents provided by the organizations.
Our aim was to reveal the interviewees’ rationales
for adopting their respective ITG frameworks in
all its nuances.
Analysis of the interviews was undertaken
using grounded techniques. A grounded theory
approach allows researchers to focus on elements
of context and process and on the actions of the
main players in the organization (Orlikowski,
1993; Fernandez, 2005). Our approach to analysis was consistent with Glaser’s (2001) and Yin’s
(2003) approaches to case study research (cf.
Lehmann, 2001). In this approach, ‘slices’ of data
are collected, codified and categorized in an evernarrowing spiral that leads to a deeper understand-
Understanding IT Governance
ing of the phenomenon. In our case, we first independently open-coded the interviews (that is,
coded for as many categories as possible, with no
pre-existing codes in mind). Repeated engagement
with the data led to emergence of patterns which
formed the basis for our findings regarding ITG
decisions in the two organizations.
In Canada, proximity to the United States and
to contagion by SOX has led Canadian regulators
to adopt SOX-like (SOX by proxy) measures. That
is, the tendency is to be punitive rather than preventative, confrontational (especially when breaches
occur) rather than collaborative. In Australia, the
approach has been to have an industry-led standard
(AS8015, now ISO/IEC38500:2008) developed.
As a result, Australian companies have willingly
and widely adopted the practices embodied in
this standard, and treated IT Governance and
compliance as business processes, not as punitive
proscriptions. We posit therefore, that, a priori, we
would see different forms of organization for ITG
in organizations in each country. Those forms of
ITG adopted such that they reflect the prevailing
legislation of the particular jurisdiction.
ORGANIZATIONAL STRUCTURE
Mutuals
Both organizations are mutuals, that is, they operate for the benefit of their members. As is typical
for mutuals, each member has one share in the
organization, with that share conferring certain
rights. Of those rights, probably the most exercised
are the right to stand for election as a director of
the mutual, and the right to vote for a member
standing for election to the board. For mutuals a
major effect of such a structure is the inability to
target recruiting for board members. Hence, the
skill set of the board may vary substantially from
term to term.
Like most organizations operating in the finance sector, there are significant pressures on
our two credit unions. For example, domestic and
international competition is driving down margins,
there is a surfeit of financial products, and ever
increasing demands for compliance with legislation have demanded that our credit unions respond
rapidly to a volatile operating environment. To
ensure survival in their respective operating environments, our credit unions have had to make
sure they are both lean and agile.
A BRIEF HISTORY OF
THE CREDIT UNIONS
Friendly and Amity
Friendly Credit Union was created in the Province
of British Columbia (BC), Canada, when two
long-standing credit unions merged. Friendly
has a number of subsidiary businesses including
Friendly Insurance Services, Friendly Investment
Services, Friendly Leasing and iSupport Technology Services. Amity Savings was formed in
Alberta (AB) when eight credit-unions merged,
and two additional credit unions were purchased.
Its subsidiary businesses include iSupport Technology Services.
In 2004, the two companies created a strategic
partnership. Under Canadian laws, credit unions
are not allowed to join forces, or to operate interprovincially, at least not by full-scale mergers.
The joint strategy, then, centered on creation of
a jointly-owned strategic plan and formation of
jointly-owned companies for IT-centered business
operations, insurance services, wealth management, and auto dealer services. Each credit union
retained its own members, separate boards, and
existing brands, branding, name and products.
The partnership explicitly recognized that to
avoid unnecessary duplication of functions and
to maximize cost efficiencies, there had to be a
single point of leadership for back-office activities.
49
Understanding IT Governance
Responsive
Responsive credit union originated in Brisbane,
Australia. Responsive was one of the major providers of financial services to a vertical industry
sector in Queensland. The mutual had a board of
directors, a general manager, functional managers
and line staff.
As indicated in Table 2, aside from the asset
base, there are substantial similarities between the
two mutuals. Both outsource their IT functions;
however, each has taken a very different approach
to their IT governance responsibilities.
IT FUNCTION
Friendly and Amity
In recognition of the need for a single point of
leadership on back-office activities, Friendly and
Amity’s strategic IT plan called for a joint core
banking system. It adopted Fincentric’s Wealthview Banking System as the common platform for
both organizations. One CIO was appointed and
charged with its implementation in both organizations. The end result of strategic IT plan was the
creation of iSupport Technology Services Ltd, a
wholly-owned subsidiary company responsible
for all IT services in both credit unions.
iSupport is responsible for maintaining and
managing the core banking system in both credit
unions; all information processing operations; and
all desktops. The company also has an outward
mandate to act as a credit union outsourcer, and is
looking to sell its services to other credit unions
in Canada.
Responsive
Responsive had historically outsourced its IT function and continued with that strategy. Responsive
had just completed a major update of its core
banking system. The Data Action Phoenix system
50
provided core banking, loans origination, collections, internet banking, and archiving facilities.
Data Action was selected by management and
the directors who believed Phoenix offered more
functionality than the previous system and better aligned with Responsive’s corporate strategy
of high-level personalized banking. Responsive
outsourced its IT function for two primary reasons.
First, Data Action was able to provide services to
Responsive that fulfilled current and future product requirements, met customers’ expectations of
high-level personalized banking services, and,
importantly, satisfied cost control objectives i.e.,
Responsive leverages Data Action’s expertise.
Second, Responsive relied on Data Action to
adhere to all appropriate operations-level regulations (e.g., privacy, security). Responsive, via Data
Action, thus ensured its own compliance with
operational regulations relative to its IT function.
In the GM’s words, using Data Action was “safer”.
In summary, having assessed the pressures in
their respective operating environments Friendly,
Amity, and Responsive have outsourced their IT
functions for at least the following three reasons.
First, the organizations were able to leverage
the expertise of their IT providers, e.g., security
and privacy management. Second, in response
to customer demands, the credit unions used the
capabilities of their IT providers’ platforms to
offer new or extended products. Third, outsourcing offers economies of scale and, hence, cost
efficiencies that otherwise would not be available
to the credit unions.
IT GOVERNANCE PROCESSES
iSupport
iSupport is a privately-owned company, and as
such, not subject to the usual listing and other
requirements of public organizations. It is a small
organization with less than 25 employees. iSupport’s structure is shown in Figure 1.
Understanding IT Governance
Figure 1. iSupport Structure
In addition to ensuring high levels of IT expertise, customer satisfaction, and cost efficiencies, Friendly and Amity had a further compelling
consideration in their decision to outsource their
IT function. Friendly and Amity had to make
certain liability issues between the two credit
unions were avoided. For example, if a problem
occurred with data processing, then the members
of one credit union could not sue the other
credit union for negligence. As a separate company, iSupport, an outsourcer, is at arm’s length
from the credit unions, and assumes all liability.
Canadian credit unions are governed federally by the Canadian Cooperatives Act and the
Canadian Banking Act as well as their respective
provincial statutes. BC and AB have also signed
a Free-Trade act, which allows for, amongst
other things, unfettered sales of some financial
products like insurance. The creation of iSupport
was reviewed by, and approved, a priori, by the
relevant provincial regulators.
Prior to the creation of iSupport, the CIO
implemented the CoBIT governance framework in
both Friendly and Amity. He did so for a number
of reasons, foremost among them that it was the
framework used by, and recommended by their
auditors along with a number of other auditing
firms – in effect the standard for the industry.
Furthermore, the CIO was very mindful of the
need to implement an ITG framework that was
acceptable to both the national and provincial
regulators. The CIO also felt that CoBIT was more
focused on process management than competing
frameworks like ITIL. In addition, the CIO and
his staff also conduct annual SAS 70 audits on
their operations.
iSupport’s external auditor was assigned the
task of completing an IT Risk audit and letter as
well as a full governance audit. Both the external and internal auditors report directly to the
Board’s Audit Committee. The CIO describes the
relationship with auditors as a partnership – one
in which they both work together to improve the
organization. The CIO also works with the external
and internal auditors to develop a 3-year audit
plan. In his words, the auditors are “protagonists,
not antagonists, who help us to manage risk.”
He sees them acting as a check-and-balance on
management.
According to the CIO, only one director was
IT-savvy. As a result, the Board tends to rely on
the CIO for advice and for his expertise. In his
words, the Board “just trusts the CIO”. He asserts
that directors confuse the size of a project and its
risk, often causally linking the latter to the former.
He disagrees, stating that the board should not be
governing “IT-guys doing ERP but the real risk like
evergreening, data management and storage, and
other functions with organization-wide impacts.”
The CIO had prepared a comprehensive IT
Governance model that was presented to the board
at a regular meeting for approval. It is reproduced
in Figure 2. The CIO believes that IT and business strategies across all 3 companies are strongly
aligned. This is due, he states, to the single, common point of leadership in all 3 organizations,
Friendly’s dominant role in the alliance, and in the
focus on creating and sustaining business value
in all 3 organizations.
As mutuals, credit unions have lay boards, and
often cannot recruit with specific skills, like IT,
in mind. For example, Annual General Meetings
typically only attract a very small percentage of
members. Further, only a small number of members cast absentee ballots. Resulting from such
low turnout, and low response rates, directors may
51
Understanding IT Governance
Figure 2. iSupport’s IT Governance Model
not reflect the interests of all members, or even a
majority of them. The CIO believes that it is incumbent on senior staff in the credit unions, and
now him at iSupport, to educate directors on the
scope and nature of their responsibilities. As a
result, and as the above framework attests, IT
governance is very much viewed as an operational item rather than a board responsibility
under these circumstances. As the CIO states,
“We assist the board in that the IT audit is presented
to them, the IT strategy is shared, the subsidiary
is reported on and we have the skills-based recruiting in the works. However, at the high level,
the operations of IT would not be the normal
reporting practice, more about efficiency, forward
planning for expansion and scale as well as some
performance items that are customer-impact and
member-experience related.”
52
Responsive
As noted in Table 2, Responsive operated in a
regulatory framework overseen by ASIC and
APRA. ASIC ensures that organizations such as
credit unions adhere to the Corporations Act, for
example, complying with such matters as Accounting Standards disclosures in financial reports.
Responsive’s structure is shown in Figure 3.
The regulatory framework for Authorised
Deposit-taking Institutions (ADIs), such as
credit unions, in Australia is comprehensive,
binding, and actively enforced. Whether outsourced or not, organizational IT functions come
under the scrutiny of APRA. Complying with IT
governance requirements necessitates knowledge
of the applicable prudential standards. Responsive’s Board has delegated responsibility for IT
governance to the General Manager. Delegating
governance responsibility to the General Manager and functional responsibility to Data Action
does not, as noted in APRA’s (2002) APS231,
Understanding IT Governance
Figure 3. Responsive Credit Union Structure
absolve the Directors of any of their governance
obligations and responsibilities.
A persistent and pervasive thread of Board discussions was the need to ensure all strategies and
decisions were aligned with APRA’s expectations
and requirements. With regard to outsourcing its
IT function, the Board was guided by APS 231:
Outsourcing. For example, APS231 states that
directors must assess and understand the business
case for outsourcing, must assess the selection
process, undertake a due diligence review of the
provider, and ensure the agreement contains all
appropriate matters. Furthermore, the Board has
to ensure any outsourcing agreements are entered
into with providers that have the ability to conduct
their business on an ongoing basis, that agreements
are subject to review, and that contingency plans
are put in place should the need arise to bring the
activity back in-house. By working within regulatory frameworks, and regularly engaging with
auditors and regulators, Responsive, its managers and directors, had fulfilled its IT governance
obligations to the satisfaction of its regulators.
Both the internal and external auditors reported
directly to the Audit Committee. The internal and
external auditors had a cordial relationship with
Responsive. Points of contention were generally
resolved by way of exit interviews. Responsive’s
directors and management felt that maintaining a
cordial, rather than adversarial, relationship with
its auditors resulted in better business processes,
greater awareness of regulatory requirements, and
better risk management. In brief, both directors
and management of Responsive viewed the audit
function as a benefit rather than an impost.
Like Amity and Friendly, Responsive typically
had very low member turnouts at Annual General
Meetings. Further, only a handful of proxy votes
were exercised. Like the Canadian mutuals,
Responsive’s board composition came about,
perhaps, more by chance than design. Of the 7
board members, two had IT expertise. Certainly,
all board members were interested in IT issues,
however, most board members lacked the specific
knowledge and skills to critically evaluate current
and planned IT implementations. Consequently,
they relied heavily on the advice and judgment of
the GM, at least, where IT matters were concerned.
53
Understanding IT Governance
IT GOVERNANCE MEASUREMENT
Since entering into its agreement with its new IT
provider, Responsive has undertaken measurements of its IT governance from a number of
perspectives. The following four examples are
indicative of Responsive’s means of measuring
ITG success. First, Responsive reviewed the core
banking service provided by Data Action relative
to key business practices. Second, to ensure risk
arising from use of Data Action is minimized,
i.e., Data Action is a dependable provider of IT
functions, Responsive required that Data Action
provide its Business Continuity Plan for scrutiny
both by their auditors and by APRA. Third, actual
financial costs and financial benefits are regularly
compared to the budgeted costs and benefits for
the new agreement. Fourth, Responsive reviews
intangible elements such as staff satisfaction
with the Data Action product, as well as seeking
to ascertain customer satisfaction with the Data
Action web interface.
In contrast, ITG measurement at Friendly and
Amity is more institutional than process-based. By
creating iSupport, its wholly-owned subsidiary,
Friendly and Amity have effectively adopted a
series of governance and control mechanisms
embodied in the structure of the new organization.
They have decided to use the CoBIT and SAS70
frameworks, both of which contain specific, measurable ITG metrics, and to have their external
auditor conduct an IT risk audit annually. Any
deficiencies arising from the audit then become
action items for the board and iSupport’s management. As such, ITG mechanisms in this case, are
the result of imposition of a formal organizational
context, and thus are embedded in the firm’s
operations and subject to fiduciary verification.
FINDINGS & IMPLICATIONS
Friendly, Amity, and Responsive have many common points. Both are mutuals, both had recently
54
upgraded their core banking system with off-theshelf software, both outsource their IT function,
and they have lay Boards. However, the regulatory environment of the Canadian and Australian
mutuals is substantially different. In Australia, the
regulatory and legislative framework is national,
industry focused, clear, and actively enforced.
How the individual organization implements
governance mechanisms within the legislative
framework, however, is left to the discretion of
the organization. John Trowbridge, an Executive
member of APRA addressing The Institute of
Actuaries on 29 September 2009 reinforces this
approach,
“An important part of our oversight is our reliance on the board to take full responsibility for
the company. i.e., APRA holds the board directly
accountable for the company meeting APRA’s
prudential standards.”
Indeed, some would argue that the reasons for
Australia’s better performance than most during
the Global Financial Crisis were APRA’s conservative approach to financial regulation and in the
words of John Laker, APRA’s Chairman (2009)
that “the buck stops with the board!”
The thrust of the legislation remains similar
between Australia and Canada, that is, to ensure
good governance and hence protection of stakeholders, however, the execution is different. The
legislation under which Canadian institutions
operate tends to be more punitive than in the
Australian environment, most likely as a result of
Canada’s proximity to the US, and the attendant
contagion of SOX. Canadian legislation is more
concerned with punishment after the fact (in
the case of egregious breaches) than in prevention. In the previous section, we mentioned that
Friendly and Amity have implemented CoBIT
ITG frameworks. As noted by Pinder (2006),
organizations affected by SOX tend to assess
their ITG frameworks against frameworks like
CoBIT. SOX appears to have, de facto, promoted
Understanding IT Governance
a centralized, uniform ITG framework, that, by
way of threat of punitive action appears to have
elevated frameworks like CoBIT to the position of
ITG frameworks of choice. Organizations affected
by SOX appear to be driven by compliance with
“black letter” law and avoidance of penalty. That
is, the locus of control for ITG, in effect, resides
centrally with the SEC.
In contrast, the Australian approach tends to
be more cooperative as it arose from industry-led
initiatives and its focus is on ongoing compliance. The legislation and, in consequence, the
regulators’ focus centers on good governance
but without as the (then) Australian Treasurer
put it “unnecessarily prescriptive regulation” and
“to identify the outcomes that are desired from
financial institutions rather than prescribe how
those outcomes should be achieved” (Costello,
2007). In brief, Australian ITG frameworks appear driven by compliance with the “spirit” of the
law, whereby the locus of control for ITG resides
with the particular organization. That is, control
is decentralized.
Returning to our research question: “How
the different legislative and regulatory operating
environments of the focal organizations affect
the ITG mechanisms they adopt”? The Australian
model with its clear regulatory and legislative
framework and focus on the actions of directors
and managers provides a well-defined operational
environment for mutuals. The expectations of
those operating within the industry are made abundantly clear. Consequently, the mutuals, directors
and managers, auditors, and service providers are
all marching to the beat of the same drum. If all of
the stakeholders adhere to the framework, good
governance of the IT function should take place.
In the Canadian context, governance of the IT
function of mutuals seems to rest more heavily on
the CIO and his or her ability to, as Willcocks et
al. (2006, p. 30) state ”… devise organizational
arrangements – structures, processes and staffing –
to successfully manage the interdependencies…”
In brief, both the Canadian and Australian models
seek to ensure good ITG. In the Canadian model
it is more dependent, however, on the individual,
which is why it might make sense to curtail this
individual’s influence by creating a hierarchical
structure (in this case, a wholly-owned subsidiary)
to minimize potential ITG shortcomings.
The consequences for directors of these two
disparate IT governance models are somewhat different. The Australian model does not permit directors to side-step their responsibilities. Regardless
of whether IT, or any other function, is outsourced,
directors remain personally responsible for their
actions and the effects of those actions on owners. As such, it makes more sense for them to opt
for a multi-facetted (e.g., outsourcing operational
processing, delegating managerial authority, while
retaining strategic control) form of ITG to help
minimize potentially negative outcomes. In the
Canadian case, directors are no less responsible.
However, the existence of the subsidiary limits
that liability to only that part of the organization.
The creation of a separate entity such as iSupport
would be unlikely reduce the liability of directors of organizations were a similar undertaking
to happen in Australia. However, unlike the
Australian experience, creation of an entity like
iSupport mitigates the power of the individual CIO
by having them directly accountable to a Board
composed of executives/members of its owners.
Therefore, in this case, where regulations are
punitive, potential penalties are minimized using
hierarchical mechanisms.
IMPLICATIONS FOR PRACTICE
There are two key implications of this study for
practice. First, is the necessity of mutuals to use
ITG mechanisms that reflect and compliment
the regulations and legislation in force in their
particular jurisdiction. In Canada where proximity to the US has lead to alignment with the
SOX legislation, adoption of ITG frameworks
such as COSO and CoBIT offer ITG processes
55
Understanding IT Governance
that have regulators’ imprimatur. While it may be
possible to adopt other ITG frameworks, COSO
and CoBIT are widely known and their adoption
is likely to garner support from regulators keen to
see implementation of proven ITG frameworks.
In Australia, there appears to be more flexibility in choosing appropriate ITG frameworks,
as implied by the (then) Treasurer Peter Costello’s
(2007) Statement of Expectations. In the case of
the Australian credit union, it was a small organization, offering straightforward products hence,
its ITG framework is appropriate to its situation.
That said, its ITG framework, at all times, was
subject to scrutiny by APRA during its regular
inspections. On all occasions the ITG framework
had been given APRA’s imprimatur.
Second, while directors may not need to understand every technical aspect of their respective
ITG frameworks, it is imperative that they understand their rationale and operation. Furthermore
they need to have a clear understanding of their
responsibilities relative to good ITG. During
APRA’s scrutiny of Responsive’s ITG framework
(and other aspects of governance) its inspectors
interviewed Responsive’s directors to ensure they
understood the rationale and the operation of the
ITG framework. Directors were also questioned
in relation to their responsibilities should failures
of governance, both within Responsive and Data
Action, arise. Directors of Friendly and Amity,
likewise, have had to demonstrate their understanding of the ITG framework through board
meetings and management briefings.
IMPLICATIONS FOR RESEARCH
Mutual organizations offer an interesting, but
seemingly understudied area of research for information systems researchers. While many are
individually small, as an industry their financial
worth is substantial. For example, Australian credit
unions account for some $73 billion in assets and
have some 4.5 million members (Abacus 2010).
56
Mutuals have at least three features that make them
unique from other mainstream financial organizations. First, their Boards are typically elected
from the membership. An obvious consequence
of this is their restricted ability to target specific
skill sets for the Board. This restricted ability to
target specific skill sets has a number of implications. For example, it is quite conceivable that a
member with limited or no skills in business or
commerce is elected to the Board (Australian
Prudential Standard 520: Fit and Proper Standard
does limit the likelihood of such a member being
elected to the Board).
Second, they do not have to pay dividends in
the same way as mainstream financial organizations. The dividend is typically returned by way
of reduced fees, lower cost loans, etc., in consequence, maximizing profit may not be mutuals’
primary goal. Third, mutuals, as financial institutions, are often subject to the same regulatory
and legislative controls as mainstream financial
organizations, however, many are smaller, have
fewer resources, and often have a different raison
d’être to mainstream financial institutions. They
often deal with their regulatory responsibilities
in unique and innovative ways.
Weisinger & Trauth’s (2002), comments relative to the lack of guidance to help understanding
of context in the global IT setting offer great scope
for further research. As regulation becomes more
harmonized, particularly post-GFC, understanding
how different national contexts affect the implementation of regulations offers fruitful research
opportunities. Our work on ITG frameworks is
but one aspect of the global IT sphere.
REFERENCES
Abacus. (2010). Factsheets. Retrieved November
10, 2010, from http://www.abacus.org.au/images/
stories/factsheets/2010/CUBS_Factsheet_October_2010.pdf
Understanding IT Governance
Agarwal, R., & Sambamurthy, V. (2002). Principles and models for organizing the IT Function.
MIS Quarterly Executive, 1(1), 1–16.
APRA. (2010). About APRA. Retrieved on November 8, 2010, from http://www.apra.gov.au/
aboutApra/
ASIC. (2010). Our role? Retrieved on October 5,
2010, from http://www.asic.gov.au/asic/asic.nsf/
byheadline/Our+role?openDocument
Australia, S. (2005). Corporate governance of
Information and Communication Technology.
Sydney, Australia: Standards Australia.
Australian Prudential Regulatory Authority.
(2002). Prudential standard APS231 – Outsourcing. Retrieved on October 5, 2010, from http://
www.apra.gov.au/policy/final_adi_standards/
APS231.pdf
Australian Securities Exchange. (2010). Corporate governance principles and recommendations
with 2010 amendments (2nd ed.). Sydney, Australia: Australian Securities Exchange.
Barney, J. (1991). Firm resources and sustained
competitive advantage. Journal of Management,
17(1), 99–120. doi:10.1177/014920639101700108
Berg, B., & Stylianou, A. (2009). Factors considered when outsourcing an IS system: An empirical
examination of the impacts of organization size,
strategy and the object of a decision. European
Journal of Information Systems, 18(3), 235–248.
doi:10.1057/ejis.2009.18
Bowen, P., Cheung, M., & Rohde, F. (2007). Enhancing IT governance practices: A model and case
study of an organization’s efforts. International
Journal of Accounting Information Systems, 8,
191–221. doi:10.1016/j.accinf.2007.07.002
Brown, C. (1997). Examining the emergence of
hybrid IS governance solutions: Evidence from
a single case site. Information Systems Research,
8(1), 69–94. doi:10.1287/isre.8.1.69
Caldwell, B. (1997, September 29). Outsourcing
backlash. Information Week, 650, 14–16.
Clemons, E., Reddi, S., & Row, M. (1993). The
impact of information technology on the organization of economic activity: the “move to the middle”
hypothesis. Journal of Management Information
Systems, 10(2), 9–29.
Colley, J., Doyle, J., Logan, G., & Stettinius, W.
(2005). What is corporate governance?New York,
NY: McGraw-Hill.
Costello, P. (2007). Statement of expectations from
the treasurer. Retrieved on October 5, 2010, from
http://www.apra.gov.au/AboutAPRA/upload/
Statement-of-Expectations-from-Treasurer-20Feb-07.pdf
Fernandez, W. D. (2005). The grounded theory
method and case study data in IS research: Issues and design. In D. Hart & S. Gregor (Eds.),
Information Systems foundations: Constructing
and criticising, (pp. 43-59). Canberra, AU: ANU
E Press.
Foreign Affairs and International Trade Canada.
(1987). The Canada – US free trade agreement.
Toronto, Canada: Foreign Affairs and International
Trade Canada. Retrieved on October 6, 2010, from
http://www.international.gc.ca/trade-agreementsaccords-commerciaux/assets/pdfs/cusfta-e.pdf
Gibson, B. (2009). Regulators at the forefront of
change. Australian Securities and Investments
Commission, 11 June 2009. Retrieved on October
5, 2010, from http://www.asic.gov.au/asic/pdflib.
nsf/LookupByFileName/Gibson-ASIC-Company-directors-speech-regulators-11-June-2009.
pdf/$file/Gibson-ASIC-Company-directorsspeech-regulators-11-June-2009.pdf
Gillan, S. L. (2006). Recent developments in corporate governance: An overview. Journal of Corporate Finance, 12(3), 381–402. doi:10.1016/j.
jcorpfin.2005.11.002
57
Understanding IT Governance
Gilley, K., & Rasheed, A. (2000). Making more
by doing less: An analysis of outsourcing and its
effects on firm performance. Journal of Management, 26(4), 763–790. doi:10.1016/S01492063(00)00055-6
Lacity, M., & Willcocks, L. (1998). An empirical
investigation of Information Technology sourcing
practices: Lessons from experience. Management
Information Systems Quarterly, 22(3), 363–408.
doi:10.2307/249670
Glaser, B. G. (2001). The grounded theory
perspective: Conceptualisation contrasted with
description. Mill Valley, CA: Sociology Press.
Laker, J. (2009). Mutuals after turbulent times.
Australian Prudential Regulation Authority, 9 November 2009. Retrieved on October 6, 2010, from
http://www.apra.gov.au/Speeches/upload/09Abacus-Convention-9-Nov-09.pdf
Gwebu, K., Wang, J., & Wang, L. (2010). Does
IT outsourcing deliver economic value to firms?
The Journal of Strategic Information Systems,
19(2), 109–123. doi:10.1016/j.jsis.2010.05.003
Hofstede, G. (2010). Cultural dimensions. Retrieved October 5, 2010, from http://www.geerthofstede.com/hofstede_australia.shtml and http://
www.geert-hofstede.com/hofstede_canada.shtml
Holcomb, T., & Hitt, M. (2007). Toward a model
of strategic outsourcing. Journal of Operations
Management, 25(2), 464–481. doi:10.1016/j.
jom.2006.05.003
ISO/IEC. (2008). Corporate governance of
Information Technology. Geneva, Switzerland:
ISO/IEC.
IT Governance Institute. (2003). Board briefing
on IT governance (2nd ed.). Rolling Meadows,
IL: IT Governance Institute.
Kobelsky, K., & Robinson, M. (2010). The impact
of outsourcing on information technology spending. International Journal of Accounting Information Systems, 11(2), 105–119. doi:10.1016/j.
accinf.2009.12.002
La Porta, R., Lopez-de-Silanes, F., Shleifer, A.,
& Vishny, R. (1998). Law and finance. The Journal of Political Economy, 106(6), 1113–1155.
doi:10.1086/250042
Lehmann, H. (2001). A grounded theory of international Information Systems. Unpublished
doctoral dissertation, The University of Auckland.
Lucy, J. (2006). Directors’ responsibilities: The
reality vs. the myths. Australian Securities and
Investments Commission, 17 August 2006. Retrieved on October 5, 2010, from http://www.
asic.gov.au/asic/pdflib.nsf/LookupByFileName/
Directors_responsibilities_August2006.pdf/$file/
Directors_responsibilities_August2006.pdf
Marcus, M., & Soh, C. (2002). Structural influences on global e-commerce activity. Journal of
Global Information Management, 10(1), 5–12.
doi:10.4018/jgim.2002010101
Mata, F., Fuerst, W., & Barney, J. (1995). Information Technology and sustained competitive advantage: A resource-based analysis. Management
Information Systems Quarterly, 19(4), 487–505.
doi:10.2307/249630
McFarlane, G. (2005, 1 July). A model for CIO
success. CIO Canada.
Orlikowski, W. (1993). CASE tools as organizational change: Investigating incremental and radical changes in systems development. Management
Information Systems Quarterly, 17(3), 309–340.
doi:10.2307/249774
Parent, M., & Reich, B. H. (2009). Governing
IT risk. California Management Review, 51(3),
134–152.
58
Understanding IT Governance
Pinder, P. (2006). Preparing information security
for legal and regulatory compliance (Sarbanes-Oxley and Basel II). Information Security Technical
Report, 11, 32–38. doi:10.1016/j.istr.2005.12.003
Pitts, G. (2007, March 20,). Black shouldn’t
have gone public, Trump says. The Globe and
Mail online. Retrieved on March 21, 2007, from
http://www.theglobeandmail.com/servlet/story/
RTGAM.20070320.WxBlackTrump20/BNStory/
ConradBlack
Walden, E., & Hoffman, J. (2007). Organizational form, incentives and the management
of information technology: Opening the black
box of outsourcing. Computers & Operations
Research, 34(12), 3575–3591. doi:10.1016/j.
cor.2006.01.002
Waring, K., & Pierce, C. (Eds.). (2005). The
handbook of international corporate governance.
London, UK: Institute of Directors (IoD) &
Kogan-Page.
Report, C. (1992). The report of the Committee on
the Financial Aspects of Corporate Governance.
Retrieved on October 6, 2006, from http://rru.
worldbank.org/Documents/PapersLinks/1253.pdf
Weill, P., & Ross, J. (2004). IT governance: How
top performers manage IT decision rights for
superior results. Boston, MA: Harvard Business
School Press.
SEC. (2010). What we do. Retrieved on October
5, 2010, from http://www.sec.gov/about/whatwedo.shtml
Weisinger, J., & Trauth, E. (2002). Situating
culture in the global information sector. Information Technology & People, 15(4), 306–320.
doi:10.1108/09593840210453106
Shailer, G. (2004). An introduction to corporate
governance in Australia. Frenchs Forest, NSW:
Pearson Education Australia.
Stake, R. (1995). The art of case study research.
Thousand Oaks, CA: Sage Publications.
Subramani, M., & Walden, E. (2001). The impact
of e-commerce announcements on the market
value of firms. Information Systems Research,
12(2), 135–154. doi:10.1287/isre.12.2.135.9698
The Economist. (2010). The banks battle back:
Basel 3. The Economist, 395(8684), 75. London.
Trowbridge, J. (2009). Beyond the global financial crisis. A Prudential perspective. Australian
Prudential Regulation Authority, 29 September
2009. Retrieved on October 6, 2010, from http://
www.apra.gov.au/Speeches/upload/IAA-Beyondthe-GFC-Sep-09.pdf
Wernerfelt, B. (1984). A resource-based view
of the firm. Strategic Management Journal, 5,
171–180. doi:10.1002/smj.4250050207
Willcocks, L., Feeny, D., & Olson, N. (2006).
Implementing core IS capabilities: Feeny-Willcocks IT governance and management framework
revisited. European Management Journal, 24(1),
28–37. doi:10.1016/j.emj.2005.12.005
Yin, R. (2003). Case study research design and
methods (3rd ed.). Thousand Oaks, CA: Sage
Publications.
Zain, M., Rose, R., Abdullah, I., & Masrom, W.
(2005). The relationship between information
technology and organizational agility in Malaysia.
Information & Management, 42(6), 829–839.
doi:10.1016/j.im.2004.09.001
Van Grembergen, W. (2004). Strategies for Information Technology governance. Hershey, PA:
Idea Group Publishing.
59
Understanding IT Governance
APPENDIX
Table 3. IT Governance Frameworks. Adapted from citation omitted
Framework
Origin / Definition
Description
CoBIT
Control Objectives for Information and related Technologies – developed by the IT Governance Institute (ITGI – www.itgi.org) for
the Information Systems Audit and Control
Association (ISACA – www.isaca.org)
First published in 1996, and now in its fourth version,
this is a best practices framework that lists 34 high-level
objectives covering 215 control objectives grouped into 4
domains: Plan & Organize; Acquire & Implement; Deliver
& Support; and Monitor & Evaluate.
COSO
Committee of Sponsoring Organizations of the Treadway
Commission (www.coso.org)
A private-sector initiative formed in 1985 to deal with
fraudulent financial reporting. Published its 8-component
Enterprise Risk Management Integrated Framework in
2004 as a result of the passage of SOX in 2002. This framework has been endorsed by the SEC as SOX-compliant.
ITGAP
IT Governance Assessment Process – Peter
Weill & Jeanne W. Ross (2004). IT Governance: How
Top Performers Manage IT Decisions Right for Superior
Results. Cambridge: Harvard Business School Press.
A 4-stage, matrix process in which organizations and
managers map 5 IT Decision Domains against one or
more of 5 Governance Archetypes.
ITGDF / GAM
IT Governance Design Framework / Governance Arrangements Matrix (Weill & Ross, 2004)
Part of the ITGAP process, in which managers first map
IT-related processes in their organizations before changing
them to conform to one or more governance archetype.
CMM / CMMI
Capability Maturity Model / Capability Maturity Model Integration (Software Engineering Institute at Carnegie-Mellon
University – www.sei.cmu.edu/cmmi/)
CMM was originally developed in the mid-1980s to aid in
managing large software development projects. It posits
5 maturity levels. CMM was replaced by CMMI which
is broader in scope to include most business processes.
ITServiceCMM
Information Technology Service CMM (Consortium of 3
companies and 3 universities in the Netherlands, c1998 –
the Kwintes Project (www.itservicecmm.org)
CMM for providers of IT services, with corresponding
levels of service maturity.
ITIL
IT Infrastructure Library (www.itil-itsm-world.com/ or
www.itil.co.uk)
A set of 8 reference books developed by the UK Office of
Government Commerce (OGC), dealing with the management of IT services and asset management.
ASL
Application Services Library (www.aslfoundation.org)
Provides a framework for IT applications development that
centers on 3 areas: functional management, application
management, and technical management.
SDM / SDM2
Software Design and Management (Cap Gemini, www.
capgemini.com)
Developed in 1970 (by Pandata-since bought by Cap
Gemini) and revised in 1987, it presents a 7-stage waterfall model.
SAS 70
Statement on Auditing Standards no. 70 (American Institute
of Certified Public Accountants – AICPA, www.sas70.com)
Developed in 1992 as a follow-on to SAS 55 (internal
controls), this is an auditing standard to assess the internal
controls on transaction processing. Became the basis of the
standard eventually advocated by COSO.
ISO / IEC 17799
(soon to be 27002)
International Organization for Standardization / International Electrotechnical Commission Standard 17799,www.
standardsdirect.org
An information security standard first published in 2005.
Provides best practice recommendations in 12 key information security areas. To be renumbered in 2007.
6-Sigma
Improving business processes to within 6 standard deviations (+/-) from perfection (www.motorola.com)
Developed by Motorola in 1986, and widely-used and
adapted since, 6-sigma is a business improvement methodology aimed at measuring and correcting process and
product defects. This may include IT services.
Prince2
PRojects IN Controlled Environments(www.prince2.
org.uk)
A project management and control methodology developed
in 1989 by the UK government for project managers. It
defines project management along 8 broad processes,
with 45 subprocesses.
continued on following page
60
Understanding IT Governance
Table 3. continued
Framework
Origin / Definition
Description
AS 8015 (2005)
(now ISO /
ICE38500: 2008)
Australian Standard for Corporate Governance of Information & Communication Technology (ICT) (www.
standards.org.au)
The world’s first national standard for governing ICTs, introduced in March, 2003 and subsequently revised in 2005.
It prescribes a 3-sequential-step approach to governing ICT
projects and ICT operations: EVALUATE (uses of ICT) –
DIRECT (preparation & implementation of ICT plans and
policies to best support the organization) – MONITOR (conformance to policies and performance against the plans).
61
Section 2
Cases and International
Perspectives
63
Chapter 4
User Evaluation of
E-Government Systems:
A Chinese Cultural Perspective
Nan Zhang
Tsinghua University, China
Xunhua Guo
Tsinghua University, China
Guoqing Chen
Tsinghua University, China
Patrick Y. K. Chau
University of Hong Kong, China
ABSTRACT
Based on Information Technology adoption theories and considering Chinese cultural characteristics,
this chapter proposes a user centric IT/IS evaluation model composed of three determinants, namely
perceived usefulness, perceived ease of use, and perceived fit, for investigating the e-government systems
application and management in China. By empirically validating the model with survey data, it is demonstrated the perceived fit has significant impacts on the end users’ evaluation towards e-government
systems, due to the special element of Hexie in the Chinese culture. The results also indicate that the
reasons for failures in e-government systems application in China largely lie in the lack of fit, which
may root in the long power distance characteristic of the Far Eastern culture.
INTRODUCTION
In recent years, e-government has created a lot
of interest among researchers in the IS field. The
adoption of IT/IS leads to dramatic changes in the
DOI: 10.4018/978-1-60960-605-3.ch004
relationship between governments and citizens
(G2C), governments and businesses (G2B), and
governments and governments (G2G). Policy
makers and practitioners are facing additional
challenges arising out of the application of egovernment information systems. Issues regarding
IT/IS evolution, adoption, and diffusion in govern-
Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
User Evaluation of E-Government Systems
mental organizations have been widely addressed
at international academic conferences, as well as
major MIS journals (Grant 2005; Hackney et al.
2005; Heeks et al. 2006; Irani et al. 2007; Srivastava et al. 2006). To the extent of our knowledge,
however, the studies on e-government in China
from either theoretical or empirical perspectives
are all still scarce. Furthermore, in the existing
studies on e-government adoption and diffusion,
cultural factors have been widely noticed as
important but not yet thoroughly examined (Gallivan and Srite 2005). Accordingly, this chapter
is aimed at probing into the e-government adoption or evaluation mechanism from the cultural
perspective, in the light of data collected from a
Chinese context.
On the other hand, as the economy of China
keeps growing rapidly, IT penetration brings dramatic changes to organizations and individuals on
various respects. According to some observations,
the impact of IT/IS application in China might
be even more profound than that in any other
country, as China is changing from an isolated
centrally-controlled economy to a market that
opens to the global economy (Martinsons 2005).
On such a background, it is worthwhile to probe
into the research issues regarding IT/IS adoption
and diffusion in Chinese governments.
As one of the driving forces of IT/IS applications, Chinese governmental organizations have
long been striving to promote the use of information systems and are achieving elementary
progress in their own IT/IS implementation and
use. Government websites have been developed
continuously. The ranking of China in the chart of
e-government application (or readiness) has been
rising as well (Liu 2006; UN. 2005; West 2006).
Despite the rising in the ranking systems which
are mostly based on evaluating the functions of
government websites, it is believed that IT/IS
application in Chinese governments still lacks
the abilities to support the goal of “providing
more services and more convenience to citizens”,
64
which is generally considered as the essential
target for e-government applications (Holliday
et al. 2005). According to a recent investigation,
most of the e-government projects implemented in
Chinese governments on various levels were not
planned and designed for the citizens (Lu 2006).
Moreover, Chinese government leaders are using
information technology to drive efforts both to
accelerate decentralized public administration
and at the same time to enhance government’s
ability to oversee key activities (Ma et al. 2005).
The concurrent pursuit of these two seemingly
paradoxical objectives leads to a highly demanding context for e-government application and
management, in which administrative reform and
systems implementation are tightly linked and
frequently interacting.
A recent study examined the status the IT/IS
diffusion in Chinese governmental organizations
using a two-dimension framework, in which
diffusion width (the extent to which IT/IS has
reached in the organization) and depth (the extent
to which an organization depends on IT/IS) were
considered, and concluded that the e-government
application in China has expanded to a large extend
but not yet reach effective depth of use (Zhang et
al. 2007b). While evolving from an “expanding
stage” (during which the diffusion width expands)
to a “penetrating stage” (during which the diffusion
depth increases), user evaluation on the systems
becomes a critical force that might accelerate or
constraint the process (Zhang et al. 2007b). Consequently, a user-centric IT/IS evaluation model
would be particularly helpful for understanding
the development of e-government application in
China in the current period.
As a typical Eastern civilization, Chinese
culture has always been deemed as very much
different from the West in many aspects (Hofstede
2001). Some studies have examined the characteristics of IT/IS management in Chinese companies
(Guo et al. 2005; Mao et al. 2006; Martinsons
2005). While a lot of Chinese companies have
User Evaluation of E-Government Systems
already transformed into international enterprises,
in which different cultures are fusing gradually,
governments in China remain as traditional Chinese organizations where all the characteristics
of Chinese culture are preserved (Lu et al. 2005).
Therefore, we feel that it is worthwhile to take the
culture issues into account while studying the IT/
IS diffusion in Chinese governments.
In this chapter, a user-centric IT/IS evaluation
model is proposed, based on a comprehensive
review on related literature and incorporating the
concept of perceived fit, a factor that typically
reflects some distinguishing characteristics of the
Chinese culture which holds a high esteem for
Hexie, a Chinese word that can be approximately
interpreted as interactive cooperation and synergy
in a complex system that includes a large amount
of subsystems or elements (Xi et al. 2006). As a
cultural concept derived from traditional Chinese
philosophy, Hexie is particularly considered as
not equal to “Harmony” in English. It refers to
not only the cooperation among the organization
members’ expectation, but also the coordination
in member’s tasks or resource allocation in the
organization (Xi et al. 2006). Furthermore, the
proposed model is validated in the light of survey
data. The survey was conducted in the local government of an administrative district in Beijing, the
capital of China. The district is one of the largest
in Beijing, which accounts for a big portion of
the capital’s economy. It is also where the Central
Business District (CBD) of the city and the main
venues of the 2008 Olympic Games are located.
There are over 3,000 foreign capital enterprises
in the district, and 130 of them are among the 500
world largest companies. Through the empirical
study, we expect that the presented model and
the validation results may provide scholars and
practitioners with better understanding of the user
evaluation regarding e-government systems in a
Chinese cultural context.
THEORETICAL CONSTRUCTION
Review of Related Research
As mentioned above, our research was aimed at
understanding the IT/IS user evaluation regarding
e-government systems from a Chinese cultural
perspective. Therefore, the research is based on
three major areas of efforts in IS discipline, namely
IT/IS adoption, e-government, and culture. In the
following part, we will review the related literature
in these three areas, respectively.
IT/IS Adoption and Evaluation
Since late 1980s, information technology adoption and use have remained a central concern in
the field of information systems research. Related
efforts have produced a number of theoretical
models, such as Technology Acceptance Model
(TAM) (Davis 1989; Davis et al. 1989; Venkatesh
et al. 2000a), Task-Technology Fit Model (TTF)
(Goodhue et al. 1995), and Unified Theory of
Acceptance and Use of Technology (UTAUT)
(Venkatesh et al. 2003), to interpret the psychological and social mechanisms that potentially
determine the behavior of technology adoption. As
well, influential theories in other areas, including
Theory of Planned Behavior (TPB) (Mathieson
1991; Taylor et al. 1995), Social Cognitive Theory
(SCT) (Compeau et al. 1999), and Innovation Diffusion Theory (IDT) (Moore et al. 1991; Plouffe
et al. 2001; Rogers 1995), were borrowed into the
studies of technology adoption. Meanwhile, large
amounts of empirical and field studies were conducted in various regions of the world to validate
the theories. Results from these studies illustrated
that the models generally explain a substantial
proportion (about 40% in average) of the variance
in usage intention and behavior (Legris et al. 2003)
and it is believed that the achievements in technology adoption research could essentially help IT
65
User Evaluation of E-Government Systems
manufacturers and managers to better handle the
development, application, and management of IT.
During the past two decades, TAM has been
prevailing in the area of IT adoption research.
The model was firstly proposed by Davis in 1989
(Davis 1989; Davis et al. 1989) and later gained
substantial theoretical and empirical support
(Chau 1996; Chau et al. 2002; Legris et al. 2003;
Venkatesh et al. 2000a; Venkatesh et al. 2007).
Recently, however, the over-intensive focus on
expanding the TAM is being criticized for the
lack of innovation in the theoretical nature and
the departure from the current important IT/IS
management problems (Benbasat et al. 2007; Venkatesh et al. 2007). It is suggested that the future
research on IT/IS adoption should be redirected
toward examining different antecedents (e.g., IT
artifact and design) and different consequences
(e.g., adaptation and learning behaviors) in order
to reach a more comprehensive understanding
of what influences adoption and acceptance in
different IT use contexts and to provide more
useful recommendations for practice. (Benbasat
et al. 2007). In other words, it would be more
promising to take the conceptual models as tools
for examining the implementation and management of specific applications in various contexts.
Evidentially, in recent years, TAM and related
theories have been introduced to discussing the
issues regarding e-government systems application (Carter et al. 2005; Fu et al. 2006; Phang et
al. 2006), as well as analyses from cultural perspectives (Lippert et al. 2007; Mao et al. 2006).
Conceptually, IT/IS evaluation is closely
related to adoption. In early 1980s, motivated
by management’s desire to improve productivity
of IS, researchers started to measure and analyze user satisfaction. As the first variable that
described individual’s feeling, satisfaction was
an inclusive concept that covered many aspects
of one’s reactions to the target IT/IS (Bailey et
al. 1983). Dividing satisfaction into usefulness
and ease of use, the most two important dimensions to examine systems, TAM has been used
66
in both adoption studies and evaluation research
extensively (Keil et al. 1995; Legris et al. 2003;
Venkatesh et al. 2000b). Around the Millennium,
the third generation of IT/IS evaluation models
that added some factors from usability design
and human-computer interaction (HCI) research,
was developed. Among the efforts Microsoft usability guideline (MUG) is the most influential
one, which comprises five specific categories:
content (similar as usefulness, focused on some
specific kinds of systems, e.g. web sites), ease of
use, promotion, made-for-the-medium (MFTM),
and emotion (Agarwal et al. 2002).
Although MUG provides a powerful model
that has been used to evaluate some on-line
e-commerce systems (Agarwal et al. 2002;
Venkatesh et al. 2006a; Venkatesh et al. 2006b),
the factors in the model may not be suitable for
studying e-government evaluation. For instance,
promotion, which represents the advertisement
investment and performance, is critical to web
sites but may not have strong impact on the user
evaluation of e-government system because of
the partly mandatory adoption settings. Similarly,
the discussion on emotion is more meaningful for
a hedonic website than a system in workplace.
Generally, IT/IS adoption and evaluation issues
have been extensively discussed in the IS area
for the past twenty years. However, the adoption/
evaluation of e-government systems has not yet
been sufficiently addressed. Furthermore, to the
extent of our knowledge, few of the existing studies
have taken into account the cultural perspective,
which is gaining more and more attention in related
research efforts (Gallivan and Srite 2005). In other
words, it would be worthwhile to probe into the
e-government adoption or evaluation mechanism
from the cultural perspective.
E-Government
In recent years, the application of IT/IS in governmental organizations has developed rapidly
and created a lot of interest among researchers.
User Evaluation of E-Government Systems
As reported, the number of papers regarding egovernment systems published on journals has
also increased dramatically in the past three years
(Zhang et al. 2007a).
Interestingly, although e-government has
been discussed extensive, there has not yet been
a commonly accepted definition for the concept.
In some studies in that e-government is defined
by the objectives of the activities while in others
by the specific technology used (Yildiz 2007).
Considering e-government from a perspective
of the use of IT/IS to enhance the access to and
delivery of government information and service
to citizens, Layne and Lee provided a four step
framework for understanding e-government development process, which includes the catalogue
stage, the transaction stage, the vertical integration
(local systems are linked to higher levels) stage,
and the horizontal integration (systems are integrated across different functions) stage (Layne
et al. 2001). From the application view, Siau and
Long provided another framework, describing
four types of e-government systems, namely
government-to-customer (G2C), governmentto-business (G2B), government-to-government
(G2G), and Government-to-Employee (G2E)
(Siau et al. 2006). According to this framework,
our empirical study in this chapter can be regarded
as conducted in G2E settings.
IT/IS and Culture
The past decade has witnessed growing interests
in cultural factors, particularly national culture
dimensions, in the areas of IT/IS research (Davison et al. 2003; Evaristo 2003; Weisinger et al.
2002), as the trends towards globalization have
been accelerating and IT/IS is generally considered
to be one of the most significant driving forces
(Friedman 2005).
When cultural issues are taken into account,
Hofstede’s well-known five dimensions model
provides the most influential theoretical framework, which has been widely cited for cross-
cultural studies in various areas (Davison et al.
2003; Ford et. al. 2003; Gallivan et. al. 2005).
The original version of Hofstede’s model includes 4 dimensions, namely power distance, uncertainty avoidance, individualism/collectivism,
and masculinity/femininity. It was later found that
the above four dimensions provide limited ability
to explain the characteristics of Eastern cultures,
so the fifth dimension, Long-term Orientation, or
Confucian Dynamism, was added, referring to the
extent to which a culture programs its members
to accept delayed gratification of their material,
social, and emotional need (Hofstede 2001; Hofstede et al. 1988). Although this framework has
sometimes suffered from criticism for the lack
of practical feasibility and incompleteness of the
dimensions, most of the cross-cultural researches
in the past two decades still adopt one or more
dimensions from the model (Ford et al. 2003;
Lim 2004; Lippert et al. 2007; Mao et al. 2006;
Straub et al. 1997).
A recent literature review of culture in IS
research categories the accumulated efforts into
four types, namely IS development, IT adoption
and diffusion, IT use and outcomes, and IT management and strategy (Leidner et. al. 2006). The
IT/IS researches that considered Chinese culture
can also be found in these four directions (Chen
et. al. 2007; Guo et. al. 2005; Hempel et al. 2001;
Huang et. al. 2003; Martinson 2005; Mao et. al.
2006). Although these studies were conducted on
different perspectives, they were commonly based
the fundamental judgment that at least some of
the special characteristics in IT/IS application
and management in China can be attributed to
China’s Eastern cultural nature and, consequently,
the Chinese culture and Confucians values may
essentially affect IT/IS development strategy and
management practice in China.
After a critical review on the related literature
of culture in IT/IS research in China and other
Eastern countries, we found that the existing studies, especially those on IT/IS adoption by Chinese
users (Huang et. al. 2003; Mao et al 2006), were
67
User Evaluation of E-Government Systems
mostly designed to analyze the special phenomena
in China with Western theoretical frameworks, or
to explain some unexpected empirical results with
some Chinese cultural considerations. In comparison with this, our study in this chapter attempts
to extend the classical IT/IS adoption/evaluation
model by introducing new factors originated from
Chinese cultural values, so as to better reflect the
IT/IS practice, especially e-government systems
implementation and diffusion, in China.
HEXIE AND THE PERCEIVED FIT
Hexie in Chinese Culture
As mentioned above, Hofstede’s original four dimensions model was developed based on the West
cultural philosophy. Since it was later discovered
that the model lacks the capability to explain the
characteristics of Eastern cultures, Hoftede conducted a “Chinese Value Survey” (CVS) in 1985
and proposed the well known fifth dimension,
long term orientation (LTO), which is also called
Confucius values (Hofstede et al. 1988). However,
this new dimension has not been really accepted
as the other dimensions were, as it not seems unfamiliar to Western people, but also tends to get
lost in confusion for Chinese minds. Hofstede’s
divides the Confucian values into the “positive”
pole and the “negative” pole, in a similar way as the
other four dimensions (Hofstede 2001). However,
in the traditional philosophy of China, the positive
and negative aspects are often considered to be
indivisible and depend on the specific contexts
in which they are discussed. Especially, if the
prefixes tai (too much) or guofen (excessive) are
added to these values, the positive one would be
likely to become negative (Fang 2003). According to an ancient Chinese understanding of how
things work, the Yin-Yang principle (Wang 2005),
the positive and negative aspects both produce one
another and overcome one another. Therefore the
ideal status of everything is to achieve a balance
68
between positive and negative, and moreover, to
achieve Hexie, the interactive cooperation and
synergy in the whole system.
Many Chinese researchers believe that Hexie
is the most principal rule of Confucian values (Ye
2006). Since the ancient times, Hexie has always
been one of the basic elements of the Chinese
culture, as indicated with an old proverb: “Hexie is
priceless”. However, it is not easy to clearly define.
As mentioned above, Hexie describes a balance
status of complementary equals in the Yin-Yang
theory (Fang 2003), and can also be approximately
interpreted as interactive cooperation and synergy
in a complex system that includes a large amount
of subsystems or elements from systems theories
(Xi et al. 2006). The importance of Hexie shows
that the Chinese culture puts emphasis on the
smooth relationships among all aspects of all the
involved persons and organizations, and usually
prefers evolution rather than revolution when innovations are introduced. In such situations, gradual
changes are much easier to be accepted. This
basic characteristic is reflected in various areas.
For instance, Chinese medicine prefers “therapy”
than “treatment”. Although frequent revolutions
in the economy and society had taken place in
China in the 20th century, in the current period,
however, the country is consciously rebuilding
the traditional cultural values, as significantly
symbolized with the concept of “Hexie Society”
proposed by Hu Jintao, the leader of the central
government of China.
Recognizing the important role of Hexie in
Chinese culture, some IT/IS researchers have
started to introduce this concept the related studies
in China or some other Eastern environment. For
instance, researchers have analyzed the unique key
factors for e-commerce success or information
sharing in China through exploring the importance
of Guanxi network (Martinson 2002, Shin et al.
2007), where Guanxi refers to “the existence of
direct particularistic ties between people” (Shin
et al. 2007) and in considered as a feasible way to
preserve Hexie. Another study on Bunkhun in Thai
User Evaluation of E-Government Systems
culture analyzed the inner relationship network
in a firm during the process of ERP implementation, where Bunkhun refers to “the psychological
bond between two persons: one who renders the
needy help and favors out of kindness and other’s
remembering of the goodness done and his everreadiness to reciprocate the kindness, not bound
by time or distance” (Walsham 2001). In our
study, we will discuss Hexie in a comparatively
narrow sense and focus on individuals’ behavior
and related perceptions in the workplace.
Perceived Fit: The Equivalent of
Hexie Value in the Workplace
With the development of globalization and the
fusion of Eastern and Western culture, the traditional Confucian values and implications of Hexie
have also be evolving in latest years. A recent
empirical research found through principal factor
analysis that the values of current Chinese people
have been further dispersed, comparing to their
father-generation (Hwang 2006). Although the
rules such as Zhongyong (moderation, following
the middle way) and Jinshen (prudence) were the
basic behavior pattern to practice Rendao (kindness) in traditional Confucian values, modern
Chinese people tend to regard them as a series
of instrumental values that can be applied in the
workplace, namely as work values (Hwang 2006).
With an overview on some key factors and
principles of Confucian work values, Zhongyong,
Jinshen, Guoyoubuji (excessive maybe worse than
scarce), and Shiying (Adaptability), we found that
the common ground of those values is opposed to
Jigongjinli (eager for quick success and instant
benefit). Accordingly, the Hexie value in the
workplace can be defined as to practice with fit
to current status and environment. Based on this
consideration, we introduce the new concept of
perceived fit, which is regarded as equivalent of
Hexie in the Chinese work values, referring to
“the degree in user perception to which the target
innovation (e-government applications for this
study) fits to all aspects of the user’s work style
and environment”.
As described in a study from a historical view,
the development process of IT/IS application
in China has experienced the “contagion” and
“cooling” stages and is currently progressing in
the “permeation” stage, during which the growth
patterns are usually steady and smooth (Guo et al.
2005; Guo et al. 2003). It can be inferred that in
such a stage, the cultural value of Hexie would play
a more important role the evaluation and adoption
of IT/IS in Chinese organizations, especially the
governmental organizations that are generally seen
as more “traditionally” structured. Consequently,
it is reasonable to predict that the perceived fit,
which represents the degree of Hexie between the
IT/IS and the other aspects of the users’ work and
life style, would have a strong impact on the end
user evaluation towards e-government systems.
There are some similar concepts in former
IT/IS literature, such as the “fit” in TTF and the
“compatibility” in IDT (and its derived models).
It needs to be clarified that the perceived fit in our
model is not equal to any of them. The “fit” in
TTF focuses on the relationship between technology and individuals’ performance (Goodhue et al.
1995), while the “compatibility” in the classical
IDT theory is defined as the degree to which an
innovation is perceived as consistent with the
existing values, past experiences, and client needs
(Rogers 1995). The concept of compatibility was
later derived into a new construct of perceived
characteristics of innovating (PCI) (Moore et al.
1991). Although PCI is actually a narrower concept
than the compatibility in IDT, as it is limited to
the matching of the tasks, it has been so widely
accepted that the original compatibility factor has
almost been abandoned. A recent research note
even proposed that the content of compatibility
should be disaggregated into four separable constructs and modeled as a multivariate structure
(Karahanna et al. 2006). Their suggestions provide
more in-depth insights on the compatibility con-
69
User Evaluation of E-Government Systems
Figure 1. E-Government Evaluation Model
struct and imply that it is meaningful to consider
the factor more comprehensively.
Conceptually, the perceived fit in our model is
more close to the original compatibility by Rogers,
but the new concept has its particular implications
at least on the following aspects: First, perceived
fit as a related dimension to Hexie in Confucian
values, would easier to be accepted than compatibility by practitioners and managers in Chinese
and other Eastern culture; Second, since Moore
and Benbasat’s compatibility has become too pervasive to be adjusted, a new concept can provide
researchers with a new perspective on this issue.
Therefore, we introduced the perceived fit in our
research model.
E-GOVERNMENT
EVALUATION MODEL
Based on TAM and considering the characteristic
of Hexie in the Chinese culture, we developed
our research model to explain the determinants of
individuals’evaluation towards new e-government
systems, as illustrated in Figure 1.
As a recent extension to TAM, UTAUT has
integrated the outcomes of a large number of
related studies. This comprehensive model, how-
70
ever, has sometimes been criticized for having
incorporated too many inter-related concepts and
even challenged as “deceptively parsimonious”
(Straub et al. 2007). Moreover, the results from
empirical test on hypotheses in UTAUT are mixed
(Al-Gahtani et al. 2007, Carlsson et al. 2006). As
the research of this chapter was focused on the
cultural factors, our model has been developed
based on the traditionally TAM. The model inherits TAM’s two core variables, perceived usefulness (PU), defined as “the prospective user’s
subjective probability that using a specific application system will increase his or her job
performance within an organizational context”
(Davis et al. 1989), and perceived ease of use
(PEOU), defined as “the degree to which the
prospective user expects the target system to be
free of effort” (Davis et al. 1989). Traditionally
TAM and other former behavioral intention models (e.g. TPB, TAM2) all follow the logical path
of “Determinant → Attitude → Intention → Actual Use”, which has been widely validated in the
voluntary environment (Davis et. al. 1989, Taylor
et al. 1995, Venkatesh et. al. 2000). However, in
the mandatory environment, it is frequently assumed that there is little variance in usage, making it difficult to be predicted empirically (Hartwick et. al. 1994, Venkatesh et. al. 2003). For this
User Evaluation of E-Government Systems
reason, some scholars introduced the concept of
Symbolic Adoption in mandatory environment to
replace the actual use for exploring user’s mental
acceptance (Nah et. al. 2004). In our study, although the individual users are required to use
the system by their superiors, the extent of use
may vary, and in turn, the users’ performance may
also vary more or less. Therefore, the influence
path in the mandatory environment is considered
as “Determinant → Attitude → User Performance”.
In governmental organizations, individuals
usually do not have the right to decide to adopt
the system or not. Employees in the government
often have to abide to their superiors’ intention in
information systems adoption. In other words, the
adoptive environment for e-government systems
is usually mandatory. For the case in our study,
the situation is more complex. The voluntary and
mandatory environments both exist in different
sub-units of the local government due to the variance of the leaders’ perceptions and determinations. Therefore, our model is focused on the determinants of end users’ attitude on e-government
systems, rather than the linkage between attitude
and behavioral intention or actual use, because
it would be meaningful to measure the users’ attitude towards the systems in both voluntary and
mandatory environment. Since attitude has been
reckoned as a medium factor in most previous
behavioral models, we replace it by a closely
related concept, namely evaluation, refers to
“the degree of a user’s overall perceptions on the
power and performance of the target system”. The
instruments of evaluation were designed according to the items of attitude in previous research.
Therefore, following Davis’ original model (Davis
1989), the three basic hypotheses are as follows:
•
Hypothesis 1: Perceived usefulness will
have a positive effect on general evaluation
toward the target system.
•
•
Hypothesis 2: Perceived ease of use will
have a positive effect on general evaluation
toward the target system.
Hypothesis 3: Perceived ease of use
will have a positive effect on perceived
usefulness.
As mention above, when considering the IT/
IS evaluation issues in the Chinese governments,
it is worthwhile to take into account the characteristics of Chinese culture. Consequently, we
introduced the factor of perceived fit, which is
developed from the Chinese cultural concept of
Hexie, as complementary to the two TAM factors,
perceived usefulness and perceived ease of use.
Previous research has discussed the different effect paths of perceived ease of use (Davis et al.
1989, Szajina 1996). Accordingly, it is reasonable
to hypothesize that the effect paths of perceived
fit will show similar differences. There are evidences that the users’ perception may change along
with the technology diffusion process (Szajina
1996). Besides directly impacting the evaluation
perception, perceived fit may also influence that
perceived usefulness, as a user that feels that the
system does not fit with his or her work and life
style is likely to consider the system as not useful to him or her. Consequently, in the proposed
model, we suppose that perceived fit would have
significant influence on IT/IS evaluation both
directly and indirectly through perceived usefulness, as does perceived ease of use. Therefore,
we have the below hypotheses:
•
•
Hypothesis 4: Perceived fit will have a
positive effect on general evaluation toward the target system.
Hypothesis 5: Perceived fit will have a
positive effect on perceived usefulness.
71
User Evaluation of E-Government Systems
EMPIRICAL STUDY
To validate our model, we conducted a survey to
investigate users’ evaluation and perceptions of
a new public service management system implemented in the local government of the Chaoyang
District, Beijing China, in 2006.
Research Site
More than 10,000 employees work for the local
government of the studied district, including all
of its departments and other subsidiaries. Partly
motivated by China’s accession to the World
Trade Organization (WTO) in 2001, the district
government accelerated its adoption of advanced
information technologies to improve international communication. Especially since 2003,
the executive officers have paid more and more
attention to e-government services and electronic
data storage/exchange in the local government,
and consequently lots of applications have been
implemented, such as the “One-Stop Service
System”, the “E-Document Exchange System”,
and the “Public Service Management System”.
Currently, the government allocates 4 million
dollars in its annual budget to the development
of e-government. Combined with the IT budgets
of all subsidiaries, the annual IT/IS spending of
the government has exceeded 20 million dollars
and the number is increasing continuously (CINC.
2006). While the large amount of IT/IS investment
has contributed to the fast and extensive network
connections in the district, some serious problems
are increasingly concerning the officials, such as
the lack of utilization of established systems, the
conflicts between different systems, the “information island” problems, and so on. More than
30 e-Government application systems have been
implemented and promoted among the government’s subsidiaries and citizens in the district
since 2001. Each time when a new system was
introduced, the government placed weighty
expectation on it and promoted it strongly, and
72
the reward was always an exciting beginning of
usage. About three months later, however, users
would gradually give up using the system. By June
2006, none of the systems has been effectively
utilized, making the officers of the government
disappointed and depressed. As such failures accumulated, the most critical view looks upon IT/
IS implementation as a “money burning” behavior
of the government with high expenditure and low
performance. Facing such difficulties, most of the
government officials feel that it would be helpful to develop efficient evaluation instruments in
the local government in search of enlightenment
regarding e-government practice.
Methodology
Instrument Development and
Questionnaire Design
While inheriting the measurement scales of
usefulness and ease of use from Davis’ original
work (Davis et al. 1989), we developed the scale
items for the new construct, perceived fit (PF),
through a pilot study among ten employees of
the IT/IS management office of the focus district
government, who are the first group of people
that experience the target system and were very
familiar with it. The 16 candidate items drawn
from former studies regarding related factors such
as compatibility, job fit, job relevance and some
other similar concepts (Chau et. al. 2002; Hu et.
al. 2003; Moore et al. 1991; Taylor et. al. 1995;
Thompson et al. 1991; Venkatesh et al. 2000b) were
all focused on the relations between the target IT/IS
and the all aspects of workplace and environment
(See Table 1). Through a rigorous categorizationprioritization method (Davis 1989), three items,
numbered 01, 04, and 16 respectively, were finally
selected for our survey. The two items of the
construct of user evaluation were also inherited
from some former studies (Davis et al. 1989;
Venkatesh et al. 2006a). All the 13 items for the
three independent constructs and one dependent
User Evaluation of E-Government Systems
Table 1. Candidate Items of Perceived Fit (PF)
Item
No.
Content
Original Article
1
Using … is compatible with all aspects of my (day-to-day) work.
2
Using … is completely compatible with my current situation.
3
I think that using … fits well with the way I like to work.
4
Using … fits into my work style.
5
Use of … will have no effect on the performance of my job (reverse score).
6
Use of … can decrease the time needed for my important job responsibilities.
7
Use of … can significantly increase the quality of output of my job.
8
Use of … can increase the effectiveness of performing job tasks.
9
… can increase the quantity of output for same amount of effort.
10
Moore & Benbasat 1989
Thompson et al. 1991
Considering all tasks, the general extent to which use of … could assist on job.
11
The setup of … will be compatible with the way I work.
12
In my job, usage of the system is important.
13
In my job, usage of the system is relevant.
Taylor & Todd 1995
Venkatesh & Davis 2000
14
Using … does not fit with my practice preferences.
15
Using … fits with my services need.
16
I would find the system fits well with the other IT/IS applications (including hardware and software)
used in my work.
variable were listed in the Appendix 1. Although
the original TAM study utilized a seven point
Likert scale, both 5 and 7 point scales are deemed
suitable for related studies (DeVellis 2003).In the
our questionnaire, we measured all the 13 factors
with a five points Likert-type scale, ranging from
“strongly disagree” (1) to “strongly agree” (5).
Data Collection
The system we studied was designed as an integrated workflow management system to support
and routine procedures of public service in subdistricts. Approximately 200 service-agents in
42 sub-districts are supposed to be the end-users
of the system. Three month later than the system
was put online, the IT/IS management office of
the district government arranged a distance video
training for employees in 20 subsidiaries that were
intended to use the system.
Chau & Hu 2002
Hu et. al. 2003
After the training, we uploaded our questionnaire to the intranet of the government and encouraged the users of the systems to participate.
Two weeks later, we collected in total 97 valid
responses for analysis (including 35 male and
62 female employees), which accounts for about
50% of the number of users that attended the
training. Respondents’ age are distributed from
23 to 55, while 31 of them are elder than 40. The
demography information of the survey is shown
as Table 2. Chaoyang is a “mixed” district consisting of urban and suburban sub-districts. In our
survey, 54 percent of the respondents were from
suburban areas.
Reliability and Validity
The internal consistency reliability (ICR) was
assessed by computing composite reliability coefficients (Fornell coefficients), of which values
higher than 0.8 generally indicate acceptable
73
User Evaluation of E-Government Systems
Table 2. Demography Information
Measure
Items
Frequency
Percent
Gender
Female
62
63.9
Male
35
36.1
Age
23-29
35
36.1
Mean = 35.0
30-39
31
31.9
St. Dev = 8.6
40-49
25
25.8
50-55
6
6.2
IT experience (year)
1-5
10
10.3
Mean = 9.9
6-10
49
50.5
11-15
29
29.9
15-21
9
9.3
St. Dev = 4.1
reliability (Gefen et al. 2005). In our study, the
ICR values range from 0.89 (for PEOU) to 0.93
(for evaluation and perceived fit). With none of
the values for all 4 constructs less than 0.8 (see
Table 3), the reliability of the scales could be accepted. Although most quantitative studies using
PLS-Graph reported Fornell coefficients because
the software provides the values of Fornell with
SEM analysis results, we agree that the Cronbach
Alpha is a more rigorous index for reliability.
Generally, Cronbach Alpha values higher than
0.7 indicate acceptable reliability. The Cnronbach
Alpha values of the four constructs calculated by
SAS 9.1 software are 0.86 (evaluation), 0.87 (PU),
083 (PEOU), and 0.88 (PF), respectively, which
also demonstrate significant reliability.
Convergent validity was evaluated by the
average variance extracted (AVE). According to
related studies, AVE values higher than 0.5 are
acceptable. For a satisfactory degree of discriminant validity, the square root of AVE of a construct
should exceed inter-construct correlations that
reflect the variance shared between the construct
and the other ones in the model (Gefen et al. 2005).
In our research, although some of the variables’
inter-correlations were relatively high, convergent
and discriminant validities of the model both at-
74
tained a satisfying level, with all the AVE square
root values above 0.7.
Analysis Results
The research model was tested using partial least
squares (PLS), a structural equation modeling
(SEM) technique suitable for highly complex
predictive models. Whereas covariance-based
SEM techniques such as LISREL and EQS use a
maximum likelihood function to obtain the estimated values in the models, the component-based
PLS uses a least-squares estimation procedure. By
this means, PLS avoids many of the restrictive
assumptions underlying covariance-based SEM
techniques, such as multivariate normality and
large sample size (Chin 1998; Chin et al. 2003;
Gefen et al. 2005). For the model provided in the
chapter, a minimum sample size of 84 is needed
with a medium effect to obtain a power of 0.8.
With a large effect size, the sample requirement
drops to 39 (Chin et al. 1999). Therefore, the 97
responses in our survey are sufficient for validating the model.
We used the PLS-Graph software (version 3.0)
for the analysis, utilizing the bootstrap resampling
method (200 resamples) to determine the significance of the paths within the structural model
(Chin et al. 2003; Gefen et al. 2005). Validating
results for the classical TAM (Davis 1989) and
our “perceived fit” model are shown in Figure 2
and Figure 3, respectively.
The results reveal that the new factor, perceived
fit, has significantly positive effects on both
evaluation and usefulness, demonstrating its
importance in this user-centric evaluation framework in the current Chinese governmental settings.
Comparing the two figures, it can be concluded
that by adding the perceived fit construct into
TAM, our evaluation model explains higher variance in the two endogenetic variables than the
original TAM. The variance explained increases
from 28% to 41% for PU and from 49% to 59%
for Evaluation. Therefore, it is reasonable to ac-
User Evaluation of E-Government Systems
Table 3. Measurement Properties and Correlations of Latent Variables
Evaluation(2)
ICR
Mean
St. Dev
Evaluation
0.93
3.49
0.84
0.94
PU
PEOU
PU(4)
0.92
3.76
0.76
0.67
0.86
PEOU(4)
0.89
3.59
0.70
0.55
0.52
0.82
PF(3)
0.93
3.34
0.80
0.69
0.66
0.57
PF
0.90
Figure 2. The PLS validating results of the TAM
Figure 3. The PLS validating results of our model
75
User Evaluation of E-Government Systems
cept that the IT/IS evaluation model proposed in
this chapter could help us better understand and
describe individuals’ attitudes and perceptions
towards e-government systems in China.
Meanwhile, while four of the five hypotheses
in the model are supported by the statistical results, the impact of ease of use on evaluation was
not validated in our study. Additional interviews
with some service agents in the government who
gave low scores to perceived ease of use reveal
that most of them tend to appeal for more training
and will still keep using the system. The reason
for this unexpected result may also lie in some
characteristics of Chinese culture. In fact, the
lack of direct relationship between ease of use
and evaluation in users’ perceptions has already
been observed in other studies in China and its
neighboring countries (Mao et al. 2006; Straub
et al. 1997). Generally, ease of use is one of the
most important characteristics of a good system.
When a user feels that a system is difficult to use,
the reason may lie in either the system or the user.
The responses in additional interviews showed
that most service agents in the government were
often inclined to learn more about the system
rather than to give the system bad evaluation
when they felt that the systems was not easy to
use. This also reflects the tendency of introspection, or Xiushen (improving oneself is the most
important) (Hofstede et al. 1988; Hwang 2006), a
typical behavior principle of Confucian traditions.
Moreover, the significance of the path from PEOU
to evaluation declines remarkably after adding
the perceived fit to the TAM. The phenomenon
may indicate that ease of use is not as dominant
as predicted by TAM or other derived models. At
least a notable portion of the reasons that drive
the users from accepting the system lies in fit,
instead of ease of use.
We also categorized our samples along three dimensions: sub-unit locations (urban or suburban),
gender, and age. The results are shown in Table 4.
Consistent with previous studies, men are usually
more willing to accept new information technolo-
76
gies than women (Venkatesh et al. 2000b), and
young people are usually more willing to accept
information technology than elders (Lam et al.
2006). The practice of Chaoyang District shows
that the working staffs in urban areas are usually
more willing to accept information technology
than those in suburban areas. In Table 4, along all
the three dimensions, the model provides better
results in the groups of higher evaluation levels.
Another notable phenomenon is that the impacts
of perceived fit are stronger in the groups of higher
evaluation levels.
DISCUSSION
It can be inferred from the results of our study that
regarding e-government application in China, end
users care more about fit than ease of use when
they are evaluating an information system. It is
not easy to explain this phenomenon with classical theories such as IDT and TAM. As mentioned
above, we suggest to probe into this result from
a cultural perspective.
In Hofstede’s famous UAI-PDI matrix
(Hofstede 2001), China has been described as a
culture with long power distance (the extent to
which the less powerful members of organizations accept and expect that power is distributed
unequally) and weak uncertainty avoidance the
extent to which a culture programs its members
to feel either uncomfortable or comfortable in
unstructured situations. Moreover, there is a long
existing principle in Chinese administration which
is called “government by man”, in contrast to the
Western idea of “government by law” (Hofstede
2001). From the Confucian values and the Hexie
perspective, it is the principle of Zunbeiyouxu
(Ordering relations by status and observing this
order) that obliges Chinese to follow the superior
as long as possible (Fang 2003; Ye 2006). In this
case, when the head of government recognizes
the importance of IT/IS application, implementation will be carried out fleetly under the direct
User Evaluation of E-Government Systems
Table 4. The Results of Cross-Group Test
Eval.
PU
Sub-unit
Eval.
R =0.82
R =0.74
Age
0.15
0.61***
R2=0.49
0.36**
0.44***
R =0.45
2
0.49**
0.09
0.39*
R2=0.53
0.28*
0.56***
0.11
0.56***
R =0.45
0.24*
0.50***
Overall
PF
0.39*
0.16
0.24
R2=0.38
0.17
0.49***
R =0.54
0.32*
0.18
0.36*
R2=0.38
0.28
0.40**
Elder than 40 (N = 31)
0.21*
2
PEOU
Female (N = 63)
2
Younger than 40 (N = 66)
R2=0.62
PU
PU
PU
Male (N = 34)
2
PU
Eval.
Eval.
Village (N = 51)
0.27**
Gender
Eval.
PF
Block (N = 46)
2
PU
Eval.
PEOU
R2=0.65
0.56***
0.23
0.14
R =0.35
0.23
0.42*
2
N = 97
R2=0.59
0.36***
0.14
0.40**
R =0.41
0.26*
0.46***
2
Notes: * p < 0.05, ** p < 0.01, *** p < 0.001
supervision of top administrators. Such strong
support enables Chinese governments to do well
in IT/IS development during the expanding stage
characterized by high diffusion width (Zhang et
al. 2007b) and also led to the success of governmental portals (UN. 2005).
Unfortunately, the effects of such characteristics of Chinese culture may change when IT/IS application in governmental organizations evolves to
the penetrating stage. With the centralized decision
making mechanism and the top-down implementation patterns, many systems are implemented in
hurries and superficially. They facially fit with
the bosses’ expectations but eventually conflict
with the real needs of end users and requirements
of tasks. Obviously, such systems would never
be used effectively. As far as the system studied
in this chapter is concerned, service agents were
the last one to know about the system. They were
expected to accept the system without any participation in the process of design and development
before the implementation. This phenomenon was
observed not only in the cases of small systems
for several hundred users, but also in the cases of
large public e-service systems facing millions of
citizens. While a lot of MIS studies emphasized
top-leader supports in IT/IS implementation (Chen
et al. 2007), our study provides evidence to show
that top-leaders’ determination can often lead to
harm in the e-government practice in a Chinese
cultural context.
In our IT/IS evaluation model, the significant
impact of perceived fit has been validated by
empirical tests among users. As discussed above,
this factor is closely related to Hexie, which has
played a continuously important role in Confucian values since 500 B. C. and roots in Chinese
society in all ages (Hofstede 2001; Hofstede et al.
1988). In other words, the fit between the technology and the other aspects of work are considered
more important in the Chinese culture.
Furthermore, the above results also imply that
the reason for the low success rate of e-government
application in China may lie in the low level of
fit. Interestingly, the lack of fit, again, is due to
the Far Eastern cultural characteristics in Chinese
governmental organizations. Because of the nature
of long power distance in the Far Eastern culture,
the decision making mechanism in Chinese government is often highly centralized, making the
degree of fit highly dependent on the vision the top
administrator. In such a situation, it is extremely
77
User Evaluation of E-Government Systems
difficult, if not impossible, to increase the fit of
newly implemented e-government systems. This
problem is less perplexing in enterprises, where
CIO and IT/IS managerial teams play more effective roles in planning and coordination. As
reported, in 2003, it is no longer a top urgent
mission to endow senior IT/IS managers with
a proper position and sufficient power, because
this has already been done in many Chinese
companies, where the top persons in charge of IT/
IS departments are now CIO or vice-presidents
(Chen et al. 2007). It is critical not to treat the IT/
IS team purely as a technical service department
(McFarlan et al. 2007). In Chinese governmental
organizations, however, the IT/IS departments are
still often regarded as technical support forces. In
the Chaoyang district, 87% of the IT/IS managers
reported that the main task of the IT/IS departments
is “maintenance and technical support for current
systems”, and their departments have little chance
to participate in the planning and design for new
systems. In many sub-units, there are only network
administrators, instead of IT/IS managers. There
would be little chance for increasing the degree
of fit, if the situation cannot be changed.
These findings may at least convey managerial implication for e-government application
and management on two respects. First, we have
demonstrated that it is a feasible way to analyze
the users’ evaluation by using our extended model
to examine the influence of determinants in Chinese government settings. Moreover, we believe
that the model can be applied in other situations
in Eastern cultural contexts for understanding IT/
IS adoption issues deeply. For instance, a former
cross-culture study found that TAM has not been
successful in explaining user’s behavior in Japan
and the authors speculated that it may be the high
uncertainty avoidance that has led to this unexpected result (Straub et. al. 1997). Based on the
findings of this chapter, it is also possible that the
concept of perceived fit and the extended model
may help better interpret this phenomenon, since
78
Japan is also an Eastern country deeply influenced
by the Confucian values.
Second, at least some of the failures in systems
application were rooted in the planning and design
of the software or systems themselves because
the fit issue, which reflects the level of harmony
or Hexie between systems and other aspects of
the users’ work and life, had not been considered
thoroughly. It would be helpful, if not necessary,
for government leaders and IT/IS managers to
evaluate the users’ recognition before and after
the implementation of new systems, so as to
understand whether the systems bring positive
impressions to the users’ perceived fit.
CONCLUDING REMARKS
Based on information technology adoption theories and considering Chinese cultural characteristics, this chapter proposed a user centric IT/IS
evaluation model composed of three determinants,
namely perceived usefulness, perceived ease of
use, and perceived fit, for investigating the e-government systems application and management in
China. After validating four of the five hypotheses
through empirical tests, we analyzed the results
from a cultural perspective and illustrated that the
perceived fit is an important antecedent to end users’ evaluation towards e-government application
in China, because of the special element of Hexie
in the Chinese culture. The results also indicate
that the reasons for failures in e-government systems application in China largely lie in the lack
of fit, which may root in the long power distance
characteristic of the Far Eastern culture. These
findings may help better understand the IT/IS
implementation and adoption characteristics in
Chinese governmental organizations and evaluate
e-government applications in a more effective way.
As an empirical study from the cultural
perspective, a major limitation of our work is
that we only validate the extended model in one
country, without actual cross culture comparison.
User Evaluation of E-Government Systems
Therefore, on-going research will attempt to
conduct more in-depth studies on the perceived
fit, especially cross-culture comparisons, so as to
further test and develop the research model, as
well as to further investigate the cultural driving
factors of IT/IS management in China and other
Eastern countries.
REFERENCES
Agarwal, R., & Venkatesh, V. (2002). Assessing
the firm’s Web presence: A heuristic evaluation
procedure for the measurement of usability.
Information Systems Research, 13(2), 168–186.
doi:10.1287/isre.13.2.168.84
Al-Gahtani, S. S., Hubona, G. S., & Wang, J.
(2007). Information Technology in Saudi Arabia: Culture and the acceptance and use of IT.
Information & Management, 44(8), 681–691.
doi:10.1016/j.im.2007.09.002
Bailey, J. E., & Pearson, S. W. (1983). Development of a tool for measuring and analyzing
computer user satisfaction. Management Science,
29(5), 530–545. doi:10.1287/mnsc.29.5.530
Benbasat, I., & Barki, H. (2007). Quo vadis, TAM?
Journal of AIS, 8(4), 211–218.
Carlsson, C., Carlsson, J., Hyvonen, K., Puhakainen, J., & Walden, P. (2006). Adoption of
mobile devices/services: Searching for answers
with UTAUT. In Proceedings of the 39th Hawaii
International Conference on System Sciences,
Hawaii, USA.
Carter, L., & Belanger, F. (2005). The utilization of e-government services: Citizen trust,
innovation and acceptance factors. Information
Systems Journal, 15(1), 5–25. doi:10.1111/j.13652575.2005.00183.x
Chau, P. Y. K. (1996). An empirical investigation
on factors affecting the acceptance of CASE by
systems developers. Information & Management, 30(6), 269–280. doi:10.1016/S03787206(96)01074-9
Chau, P. Y. K., & Hu, P. J. (2002). Examining a
model of information technology acceptance by
individual professionals: An exploratory study.
Journal of Management Information Systems,
18(4), 191–229.
Chen, G., Wu, R., & Guo, X. (2007). Key issues in
information systems management in China. Journal of Enterprise Information Management, 20(2),
198–208. doi:10.1108/17410390710725779
Chin, W. W. (1998). Issues and opinion on structural equation modeling. Management Information
Systems Quarterly, 22(1), vii–xvi.
Chin, W. W., Marcolin, B. L., & Newsted, P. R.
(2003). Partial least squares latent variables modeling approach for measuring interaction effects:
Results from a Monte Carlo simulation study
and an electronic-mail emotion/adoption study.
Information Systems Research, 14(2), 189–217.
doi:10.1287/isre.14.2.189.16018
Chin, W. W., & Newsted, P. R. (1999). Structural
equation modeling analysis with small samples
using partial least squares. In Hoyle, R. (Ed.),
Statistical strategies for small sample research
(pp. 307–341). Sage Publications.
CINC. (2006). The blueprint of informationization
development in Chaoyang district: 2006~2010.
Chaoyang Info-Net Center. (in Chinese). Retrieved
July 29, 2008, from www.chy.egov.cn
Compeau, D., Higgins, C. A., & Huff, S. L. (1999).
Social cognitive theory and individual reactions
to computing technology: A longitudinal study.
Management Information Systems Quarterly,
23(2), 145–158. doi:10.2307/249749
79
User Evaluation of E-Government Systems
Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. Management Information Systems
Quarterly, 13(3), 319–340. doi:10.2307/249008
Davis, F. D., Bagozzi, R. P., & Warshaw, P. R.
(1989). User acceptance of computer technology:
A comparison of two theoretical models. Management Science, 35(8), 982–1003. doi:10.1287/
mnsc.35.8.982
Gallivan, M., & Srite, M. (2005). Information
technology and culture: identifying fragmentary
and holistic perspectives of culture. Information
and Organization, 15(2), 295–338. doi:10.1016/j.
infoandorg.2005.02.005
Gefen, D., & Straub, D. W. (2005). A practical
guide to factorial validity using PLS-Graph: Tutorial and annotated example. Communications of
AIS, 16(1), 91–109.
Davison, R. M., & Martinsons, M. G. (2003). Cultural issues and IT management: Past and present.
IEEE Transactions on Engineering Management,
50(1), 3–7. doi:10.1109/TEM.2003.808249
Goodhue, D. L., & Thompson, R. L. (1995).
Task-technology fit and individual performance.
Management Information Systems Quarterly,
19(2), 213–226. doi:10.2307/249689
DeVellis, R. F. (2003). Scale development: Theory
and applications. Sage Publications.
Grant, G. (2005). Realizing the promise of
electronic government. Journal of Global Information Management, 13(1), 1–4. doi:10.4018/
jgim.2005010101
Evaristo, R. (2003). Cross-cultural research in
IS. Journal of Global Information Management,
11(4), i–iii.
Fang, T. (2003). A critique of Hofstede’s fifth
national culture dimension. International Journal
of Cross Cultural Management, 3(3), 347–368.
doi:10.1177/1470595803003003006
Ford, D. P., Connelly, C. E., & Meister, D. B.
(2003). Information systems research and Hofstede’s culture’s consequences: An uneasy and
incomplete partnership. IEEE Transactions on Engineering Management, 50(1), 8–25. doi:10.1109/
TEM.2002.808265
Friedman, T. L. (2005). The world is flat: A brief
history of the twenty-first century. Farrar, Straus
and Giroux.
Fu, J., Farn, C., & Chao, W. (2006). Acceptance of
electronic tax filing: A study of taxpayer intentions.
Information & Management, 43(1), 109–126.
doi:10.1016/j.im.2005.04.001
80
Guo, X., & Chen, G. (2005). Internet diffusion in
Chinese companies. Communications of the ACM,
48(4), 54–58. doi:10.1145/1053291.1053318
Guo, X., Chen, G., & Reimers, K. (2003). IT diffusion stages in the network era of Chinese companies. In Proceedings of the Third International
Conference on Electronic Business, Singapore.
Hackney, R., Kim, S., & Wastell, D. (2005). Editorial. Information Systems Journal, 15(1), 1–3.
doi:10.1111/j.1365-2575.2005.00182.x
Hartwick, J., & Barki, H. (1994). Explaining the
role of user participation in information system
use. Management Science, 40(4), 440–465.
doi:10.1287/mnsc.40.4.440
Heeks, R., & Bailur, S. (2006). Analyzing egovernment research: Perspectives, philosophies,
theories, methods, and practice. Government Information Quarterly, 24(2), 243–265. doi:10.1016/j.
giq.2006.06.005
User Evaluation of E-Government Systems
Hempel, P. S., & Kwong, Y. K. (2001). B2B ecommerce in emerging economies: I-metal.com’s
nonferrous metals exchange in China. The Journal
of Strategic Information Systems, 10(3), 335–355.
doi:10.1016/S0963-8687(01)00058-0
Keil, M., Beranek, P. M., & Konsynski, B. R.
(1995). Usefulness and ease of use: Field study
evidence regarding task considerations. Decision
Support Systems, 13(1), 75–91. doi:10.1016/01679236(94)E0032-M
Hofstede, G. (2001). Culture’s consequences:
Comparing values, behaviors, institutions, and
organizations across nations (2nd ed.). Sage
Publications.
Lam, J. C. Y., & Lee, M. K. O. (2006). Digital
inclusiveness: Longitudinal study of Internet adoption by older adults. Journal of Management Information Systems, 22(4), 177–206. doi:10.2753/
MIS0742-1222220407
Hofstede, G., & Bond, M. H. (1988). The Confucius connection: From cultural roots to economic
growth. Organizational Dynamics, 16(4), 5–21.
doi:10.1016/0090-2616(88)90009-5
Holliday, I., & Yep, R. (2005). E-government in
China. Public Administration and Development,
25(3), 239–249. doi:10.1002/pad.361
Hu, P. J., Clark, T. H. K., & Ma, W. W. (2003).
Examining technology acceptance by school
teachers: A longitudinal study. Information &
Management, 41(2), 227–241. doi:10.1016/
S0378-7206(03)00050-8
Huang, L., Lu, M., & Wong, B. K. (2003). The
impact of power distance on e-mail acceptance:
Evidence from the PRC. Journal of Computer
Information Systems, 44(1), 93–101.
Hwang, K. K. (2006). Confucian relationalism:
Cultural reflection and theoretical construction.
Peking University Press. (in Chinese)
Irani, Z., Love, P. E. D., & Montazemi, A. R.
(2007). E-government: Past, present and future.
European Journal of Information Systems, 16(2),
103–105. doi:10.1057/palgrave.ejis.3000678
Karahanna, E., Agarwal, R., & Angst, C. M.
(2006). Reconceptualizing compatibility beliefs
in technology acceptance research. Management
Information Systems Quarterly, 30(4), 781–804.
Layne, K., & Lee, J. (2001). Developing fully
functional e-government: A four stage model. Government Information Quarterly, 18(2), 122–136.
doi:10.1016/S0740-624X(01)00066-1
Legris, P., Ingham, J., & Collerette, P. (2003). Why
do people use information technology? A critical review of the technology acceptance model.
Information & Management, 40(3), 191–204.
doi:10.1016/S0378-7206(01)00143-4
Leidner, D. E., & Kayworth, T. (2006). A review of
culture in information systems research: Toward a
theory of information technology culture conflict.
Management Information Systems Quarterly,
30(2), 357–399.
Lim, J. (2004). The role of power distance and
explanation facility in online bargaining utilizing software agents. Journal of Global Information Management, 12(2), 27–43. doi:10.4018/
jgim.2004040102
Lippert, S. K., & Volkmar, J. A. (2007). Cultural
effects on technology performance and utilization:
A comparison of U.S. and Canadian users. Journal
of Global Information Management, 15(2), 56–90.
doi:10.4018/jgim.2007040103
Liu, L. (2006). Evaluations on China e-government
from international organizations. In Wang, C.
(Ed.), China e-government development report
(pp. 19–30). Social Sciences Academic Press.
(in Chinese)
81
User Evaluation of E-Government Systems
Lu, Y. (2006). E-government and the enhancement
of government efficiency. In Wang, C. (Ed.), China
e-government development report (pp. 31–46).
Social Sciences Academic Press. (in Chinese)
Lu, Y., & Zhou, T. (2005). An empirical analysis of
factors influencing consumers’ initial trust under
B2C environment. Nankai Business Review, 8(6),
2005, 96-101 (in Chinese).
Ma, L., Chung, J., & Thorson, S. (2005). E-government in China: Bringing economic development
through administrative reform. Government Information Quarterly, 22(1), 20–37. doi:10.1016/j.
giq.2004.10.001
Mao, E., & Palvia, P. C. (2006). Testing an extended model of IT acceptance in the Chinese
cultural context. The Data Base for Advances in
Information Systems, 37(2/3), 20–32.
Martinsons, M. G. (2002). Electronic commerce
in China: Emerging success stories. Information
& Management, 39(7), 571–579. doi:10.1016/
S0378-7206(02)00009-5
Martinsons, M. G. (2005). Transforming China.
Communications of the ACM, 48(4), 44–48.
doi:10.1145/1053291.1053316
Mathieson, K. (1991). Predicting user intentions:
Comparing the technology acceptance model
with the theory of planned behavior. Information
Systems Research, 2(3), 173–191. doi:10.1287/
isre.2.3.173
McFarlan, F. W., Chen, G., Reimers, K., &
Guo, X. (2007). Managing the transition from
a product-oriented towards a service-oriented
company. (HBS Case #9-307-093). Digital China
Holdings Ltd.
Moore, G. C., & Benbasat, I. (1991). Development
of an instrument to measure the perceptions of
adopting an information technology innovation.
Information Systems Research, 2(3), 192–222.
doi:10.1287/isre.2.3.192
82
Nah, F. F., Tan, X., & Teh, S. H. (2004). An empirical investigation on end-users’ acceptance
of enterprise systems. Information Resources
Management Journal, 17(3), 32–53. doi:10.4018/
irmj.2004070103
Phang, C. W., Sutanto, J., Kankanhalli, A., Li,
Y., Tan, B. C. Y., & Teo, H. H. (2006). Senior
citizens’ acceptance of information systems: A
study in the context of e-government services.
IEEE Transactions on Engineering Management,
53(4), 555–569. doi:10.1109/TEM.2006.883710
Plouffe, C. R., Hulland, J. S., & Vandenbosch,
M. (2001). Richness vs. parsimony in modeling
technology adoption decisions: Understanding
merchant adoption of a smart card-based payment
system. Information Systems Research, 12(2),
208–222. doi:10.1287/isre.12.2.208.9697
Rogers, E. M. (1995). Diffusion of innovations
(4th ed.). The Free Press.
Shin, S. K., Ishman, K., & Sanders, L. (2007). An
empirical investigation of socio-cultural factors
of information sharing in China. Information &
Management, 44(2), 165–174. doi:10.1016/j.
im.2006.11.004
Siau, K., & Long, Y. (2006). Using social development lenses to understand e-government development. Journal of Global Information Management,
14(1), 47–62. doi:10.4018/jgim.2006010103
Srivastava, S. C., & Teo, T. S. H. (2006). Performance impact of e-government: An international
perspective. In Proceedings of the Tenth Pacific
Asia Conference on Information Systems, Kuala
Lumpur, Malaysia.
Straub, D. W., & Burton-Jones, A. (2007). Veni,
vidi, vici: Breaking the TAM logjam. Journal of
AIS, 8(4), 223–229.
User Evaluation of E-Government Systems
Straub, D. W., Keil, M., & Brenner, W. (1997).
Testing the technology acceptance model across
cultures: A three country study. Information &
Management, 33(1), 1–11. doi:10.1016/S03787206(97)00026-8
Szajna, B. (1996). Empirical evaluation of the revised technology acceptance model. Management
Science, 42(1), 85–92. doi:10.1287/mnsc.42.1.85
Taylor, S., & Todd, P. A. (1995). Understanding
information technology usage: A test of competing models. Information Systems Research, 6(2),
144–176. doi:10.1287/isre.6.2.144
Thompson, R. L., & Higgins, C. A. (1991). Personal computing: Toward a conceptual model of
utilization. Management Information Systems
Quarterly, 15(1), 125–143. doi:10.2307/249443
Venkatesh, V., & Morris, M. G. (2000). Why
don’t men ever stop to ask for directions? Gender, social influence, and their role in technology
acceptance and usage behavior. Management
Information Systems Quarterly, 24(1), 115–239.
doi:10.2307/3250981
Venkatesh, V., Morris, M. G., Davis, G. B., & Davis, F. D. (2003). User acceptance of information
technology: Toward a unified view. Management
Information Systems Quarterly, 27(3), 425–478.
Venkatesh, V., & Ramesh, V. (2006). Web and
wireless site usability: Understanding differences
and modeling Use. Management Information
Systems Quarterly, 30(1), 181–205.
Walsham, G. (2001). Making a world of difference: IT in global context. John Wiley & Sons.
UN. (2005). Global e-government readiness
report 2005: From e-government to e-inclusion.
United Nations. Retrieved July 29, 2008, from
www.unpan.org
Wang, R. R. (2005). Dong Zhounshu’s transformation of Yin-Yang theory and contesting of gender
identity. Philosophy East & West, 55(2), 209–231.
doi:10.1353/pew.2005.0013
Venkatesh, V., & Agarwal, R. (2006). Turning visitors into customers: A usability-centric
perspective on purchase behavior in electronic
channels. Management Science, 52(3), 367–382.
doi:10.1287/mnsc.1050.0442
Weisinger, J. Y., & Trauth, E. M. (2002). Situating
culture in the global information sector. Information Technology & People, 14(5), 306–320.
doi:10.1108/09593840210453106
Venkatesh, V., & Davis, F. D. (2000). A theoretical extension of the technology acceptance
model: Four longitudinal field studies. Management Science, 46(2), 186–204. doi:10.1287/
mnsc.46.2.186.11926
Venkatesh, V., Davis, F. D., & Morris, M. G.
(2007). Dead or alive? The development, trajectory and future of technology adoption research.
Journal of AIS, 8(4), 267–286.
West, D. M. (2006). Global e-government. Center
for Public Policy, Brown University.
Xi, Y., & Wang, D. (2006). Harmony configuration and Hexie management: An empirical
examination of Chinese. In Proceedings of the
13th International Conference on Management
Science and Engineering.
Ye, J. (2006). The modern value of Confucianism
and Hexie mentality. Guangdong People’s Press.
(in Chinese)
Yildiz, M. (2007). E-government research:
Reviewing the literature, limitations, and ways
forward. Government Information Quarterly,
24(3), 646–665. doi:10.1016/j.giq.2007.01.002
83
User Evaluation of E-Government Systems
Zhang, N., & Guo, X. (2007). Review and trends
analysis on e-government research. In Proceedings of the Second CNAIS National Congress,
Kunming, China (in Chinese).
84
Zhang, N., Guo, X., & Chen, G. (2007). Diffusion and evaluation of e-government systems: A
field study in China. In Proceedings of the 11th
Pacific Asia Conference on Information Systems,
Auckland, New Zealand.
85
Chapter 5
Factors Influencing SME
Compliance with Government
Regulation on Use of IT:
The Case of South Africa
Michael Kyobe
University of Cape Town, South Africa
ABSTRACT
This study investigated the factors influencing SME compliance with regulation on use of IT in South
Africa. The researcher argues that these consist of a combination of business, industry, economic, technological, sociological, and psychological factors. The results show that cost of compliance was the
main influencing factor and that both rural and urban SMEs make limited effort to develop policies and
demonstrate compliance. Furthermore, a discriminant function analysis revealed that rural SMEs can
be distinguished from their urban counterparts in terms of their disposition towards IT regulation by
five factors. Perception of high compliance costs was the key distinguishing factor followed by lack of
awareness, lack of training on compliance and security, perception of unfair regulation and possession
of inadequate security controls. The author hopes that this research-based evidence will provide better
understanding of SME compliance behaviors and guide the development of appropriate solutions to
compliance challenges in these organizations.
INTRODUCTION
Information technologies (IT) play a key role in
the development of nations. Governments are encouraging IT adoption by small and medium-sized
organizations (SME) to address poverty, facilitate
networking and alliances between buyers and
sellers, improve market intelligence, and increase
productivity (Nicol, 2003). While IT can provide
such benefits, the rise in the level of electronic
Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
Factors Influencing SME Compliance with Government Regulation on Use of IT
abuse has led to the establishment of regulatory
measures by governments and several SMEs are
struggling to comply with them (SMME Review
2003, 2004; SBP, 2003). In addition, there have
been very few studies investigating compliance
with legislations in SMEs (Massey, 2003; SBP,
2003; Washkush, 2006) and in particular compliance with regulation on use of IT. Consequently,
little assistance or guidance is available for these
organizations on how to address these issues
(Upfold & Sewry, 2005). The present study examines this problem in one developing country,
South Africa.
South Africa has recently enacted the Electronic Communications and Transactions Act
(ECT Act, 2002) and compliance with this law
has however created many challenges for business
organizations. Computer Business Review (2004)
shows that non-compliance with regulations
governing Web sites and online sales of goods
have been on the rise in South Africa. This was
confirmed by Buys Inc., who reported that several
organizations had not implemented notices, terms
and conditions and opt-out options on their Web
sites and that many did not even know how to go
about handling compliance issues (Elc, 2005).
SMEs in South Africa operate in a unique
regulatory environment characterized by overregulations, poor resource allocation, high illiteracy levels and lack of capital. The factors
influencing SME compliance with regulations on
use of IT have not been investigated. Furthermore,
given the fact that the profile of SMEs in urban
areas differs from that of rural SMEs in South
Africa (see Orford, Herrington & Wood, 2004),
it is possible that factors constraining compliance in these groups may also be different. These
factors need to be understood clearly if effective
solutions to SME compliance challenges are to
be provided and appropriate enforcement policies developed. SBP (2003) argues however that
there is limited research-based evidence to create this level of understanding and to guide the
86
development of useful solutions to compliance
challenges in South Africa. They maintain that
available information is patchy, contradictory,
and over-reliant on anecdote.
The contribution of this article is in providing
empirical evidence about the factors influencing
SME compliance with the ECT Act of South
Africa. Literature on compliance across various
fields was synthesized to create a comprehensive
framework for examining compliance behaviors
in SMEs. The study confirms that several factors
(e.g., business, industry, economic, technological, sociological and psychological) combine to
influence compliance behaviors in SMEs. The
study also shows that compliance levels are much
lower in rural than urban SMEs. This difference
in behavior can be explained by the fact that rural
SMEs are more likely to incur high compliance
costs; less likely to be aware of regulations; less
likely to be educated on compliance and security;
more likely to perceive regulations to be unfair
and less likely to have adequate system control
measures. SMEs in the marketing and finance
sectors however tend to respond positively to the
Act because they establish formal procedures and
adopt good practices.
In the following sections the researcher begins
with a review of the South African regulatory
environment and its impact on the SME sector.
The ECT Act (2002) and the key compliance
elements examined in the present study are introduced and compared with e-commerce laws
of other countries. This is followed by a review
of the theories of compliance with regulations
which leads to the development of a framework
for evaluating compliance with legislation on use
of IT in SMEs. The research methodology is then
presented followed by the results and analysis
of findings. Finally, conclusions are drawn and
useful recommendations are made at the end of
this paper.
Factors Influencing SME Compliance with Government Regulation on Use of IT
LITERATURE REVIEW
South Africa’s Regulatory
Environment & SMEs
South Africa has both First World and Third World
characteristics in one country. It is faced with a
double challenge of establishing its competitive
position in the global market while simultaneously addressing internal problems created by
the legacy of apartheid. This has resulted in the
enactment of many pieces of legislation (SBP,
2005) and consequently, a high regulatory burden
for business organizations.
While it is acknowledged that a vibrant SME
sector holds the potential to drive national economic growth, and indeed much in form of finance
and other resources have been invested in this
sector by the government, over-regulation and
poor allocation of resources continue to threaten
the success of the small business sector. Some of
the main areas of concern include legislations on
taxation, tourism, labor and IT. According to the
DPRU (Development Policy Research Unit, 2007),
South Africa ranks 84th worldwide in terms of ease
of paying taxes compared to Botswana (59th) and
New Zealand (16th). Obstacles identified include
VAT requirements, South African Revenue Service (SARS) administration, penalties and interest
charges. Rankin (2005) also argues that labor
market regulations (e.g., LRA, BCEA, EEA, and
Skills Act) constrain growth in smaller firms by
limiting the hiring of new employees (both skilled
and unskilled). Some firms have reacted by subcontracting and outsourcing labor while others
have replaced labor with machines. Firms in the
tourism industry have also been constrained by
the regulatory requirements of the Tourism Act of
1993 (amended in 2000), municipal regulations,
foreign exchange regulations, and excessive visa
requirements. In the IT industry, main challenges
relate to IT security, costly bandwidth problems,
complexity of the ICT-sector regulations, lack
of resources and skills to meet the requirements
of recent IT regulations (SBP, 2005; Technews,
2004; Upfold et al., 2005).
Lack of capital, IT and legal skills, poor
planning and high illiteracy levels are common
problems facing the SME sector in South Africa
(Orford et al., 2004). The combined impact of
resource limitations and the implications of numerous regulations make it difficult for SMEs to
perform effectively and respond positively to legal
demands. Hudson (2003) argues that in an environment characterized by such resource limitations,
it is imperative to understand the implications
regulations impose on regulated entities in order
to develop appropriate solutions to these problems.
If existing regulatory burdens are not addressed
soon enough, the economic contribution of the
SME sector and South Africa’s role in the global
arena may be affected. In the following sections
the author presents the ECT Act of 2002 and the
elements that constitute compliance requirements
for organizations. These are then compared with
e-commerce laws enacted in other countries.
The Electronic Communications and
Transactions Act (ECT Act, 2002)
The ECT Act of 2002 is a wide South African
legislation dealing with any form of electronic
communication (e.g., by e-mail, the Internet, and
SMS) and other issues relating to cyber inspectors, service provider liability and domain names.
Its main purpose is to provide for the facilitation
and regulation of electronic communications and
transactions; provide for the development of a
national e-strategy; promote universal access to
electronic communications and transactions and
the use of electronic transactions by SMMEs;
provide for human resource development in
electronic transactions; prevent abuse of information systems; encourage the use of e-government
services and other related matters.
According to Michalson and Hughes (2005,
pp. 2-6), one does not have to comply with the
entire act but with sections relating to incor-
87
Factors Influencing SME Compliance with Government Regulation on Use of IT
poration by reference; electronic signatures;
electronic evidence; production of information;
record retention; automated transactions; Web
site architecture and content; contract formation;
cryptography service providers; secure payment
systems; spam and protection of critical data
(i.e., chapters III, V, VI, VII, IX, XII, and XIII).
A person who contravenes these sections would
be guilty of an offence and liable on conviction
to a fine or imprisonment.
In the present study, the author will focus on
the following sections:
i.
ii.
iii.
iv.
Electronic Evidence & Record retention
(Part 1, Chapter III);
Consumer Protection - Web site content &
Spam (Chapter VII);
Personal Information and Privacy Protection
(Chapter VIII);
Illegal activities and offenses (Cyber Crime
- Chapter XIII).
These are chosen because of the growing
concern over non-compliance reported in these
areas (Burrows, 2004; Elc, 2005; Morgan, 2004;
Mullon, 2006; Upfold & Sewry, 2005).
Electronic Evidence and Record
Retention (Part 1, Chapter III)
According to Michalson and Hughes (2005, p.
2), this section aims to remove legal barriers to
electronic transactions by providing for the legal
recognition of electronic contracts and signatures
and facilitating record retention, electronic evidence and automated transactions. They explain
that electronic documents and signatures (e.g., a
typed name or a scanned image of hand written
signature) can serve as the electronic equivalent
of the paper based counterpart. Special types of
electronic signature would however be required
in situations where the law requires a signature.
Minimum requirements for transactions on the
Web where signatures are not required by parties
88
may be an expression of intent (e.g., clicking on
an “I agree” button). The act also permits keeping
records in electronic form and makes provision
for integrity as key to ensuring proper evidentiary weight of electronic evidence. Contracts
can now be performed by machines functioning
as electronic agents for parties to an electronic
transaction. This section also makes presumptions
as to the time and place of receipt of information.
This section therefore makes management of
electronic documents a critical compliance issue
because such documents may be required in legal
proceedings. The importance of proper e-mail
record keeping has been shown in recent high
profile corporate cases where firms like Morgan
Stanley, a giant U.S. financial institution, was fined
15 million U.S. dollars in March, 2006, for failing
to manage its e-mails (Image & Data Manager,
2006). Even where back-ups are regularly done,
the discovery costs could turn out to be extremely
high if this process was not conducted properly.
Consumer Protection: Web Site
Content and Spam (Chapter VII)
This section stipulates certain information to be
provided on the Web site. According to Michalson
and Hughes (2005, p. 2), merchants (foreign and
local) are obliged to provide consumers with a
minimum set of information such as the price of
products, contact details, and the right to withdraw
from an electronic transaction before completion.
In certain circumstances, consumers are also
entitled to a “cooling off” period within which
they may correct mistakes or withdraw from the
transaction without incurring any penalty. Any
provision in the agreement that excludes any
rights guaranteed under this chapter is null and
void. In addition, the act also places on businesses
trading online the responsibility of ensuring that
sufficiently secure payment systems are used. If
a payment system is breached, the supplier must
reimburse the consumer for any loss suffered.
Factors Influencing SME Compliance with Government Regulation on Use of IT
Consumers are further protected from unsolicited communications (spam). A merchant who
sends unsolicited electronic communication must
provide the consumer with the option to cancel its
subscription to the mailing list and must identify
the source from which that merchant obtained such
customer’s personal information. The consumer
has the right not to be bound to unsolicited communication. It is an offense to continue to send
unsolicited communications after having been
advised that they are not welcome. No agreement may be concluded if the consumer has not
responded to the spam.
There are already many indications that South
African firms are not compliant with these sections
of the act. Technews (2004) shows that compliance levels dropped by more than 30% between
2002 and 2004. They also show that 81% of the
Web sites did not even have legal notices. Some
had simply adopted notices and disclaimers from
foreign Web sites which never addressed South
African legal requirements (Technews, 2004).
These have been attributed to high Web site
development expenses, limited understanding of
legal requirements, misinformation about legal
compliance and poor coordination of Web projects
(Technews, 2004).
users in South Africa have limited knowledge of
who else shares their personal information on the
Web. They found that the top 100 South African
economic sites had not implemented appropriate
information protection principles.
Personal Information and Privacy
Protection (Chapter VIII)
E-Commerce Regulation in Other
Countries
This chapter establishes a voluntary regime for
protection of personal information (Michalson
& Hughes, 2005, p.2). Personal information
includes any information capable of identifying
an individual. Collectors of such information
must have the express written permission of the
consumer for the collection, collation, processing
or disclosure of any personal information on that
consumer. Information obtained must be used for
the lawful purpose for which it was obtained and
the collectors are expected to subscribe to a set of
universally accepted data protection principles.
Joubert and Van Belle (2004) allege that Internet
Most countries have e-commerce laws similar to
the South African ECT Act since these have been
modeled after the UNCITRAL Model laws on
electronic commerce and e-signature (1996, 2001).
UNCITRAL is the United Nations Commission
on International Trade Law established by United
Nations in 1996 to harmonize International Trade
law (UNCITRAL, 2004). This organization also
focuses on issues relating to contracting, online
dispute resolution and removal of legal barriers to
e-commerce. Most countries have the following
issues addressed by their e-commerce laws: the
assimilation of electronic forms of transaction to
Illegal Activities and Offenses
(Cyber Crime - Chapter XIII)
This chapter addresses the problems relating to
cyber crime in South Africa. According to Michalson and Hughes (2005), it introduces statutory
criminal offenses relating to unauthorized access
to data (e.g., through hacking), interception of data
(e.g., tapping into data flows or denial of service
attacks), interference with data (e.g., viruses) and
computer related extortions, fraud and forgery.
They also state that a person aiding those involved
in these crimes will be guilty as an accessory. The
Act prescribes the penalties for these offences.
Cyber-crime and other factors identified above
impact on the development of e-commerce globally. Consequently, many countries have found
it necessary to implement e-commerce laws to
address these issues. Some of these national ecommerce laws are discussed below and compared
with the South African ECT Act.
89
Factors Influencing SME Compliance with Government Regulation on Use of IT
conventional ones; enforceability and formation
of online contracts; use of economic agents; data
management and consumer protection. Differences however exist where national and cultural
concerns have been taken into consideration. For
instance, Quebec province in Canada has its own
e-commerce regulation; the South African ECT
Act addresses the imbalances created by apartheid
regime; while the development of local languages
has been an issue in South East Asian nations.
Different approaches have also been adopted in
regulating spam and resolving online disputes.
The adoption process has been generally slow
in developing than developed countries. Mauritius,
Tunisia and Cape Verde were among the first countries in Africa to pass e-commerce laws in 2000
(Ogunyemi, 2006). The Electronic Transactions Act
of Mauritius was modeled after the UNCITRAL
Model E-Commerce law of 1996. Key provisions
in the Tunisian Electronic Exchange and Electronic
Act include use of electronic contracts, signatures
and establishment of a certification authority while
the Cape Verde law covers the admissibility of
electronic evidence (Ogunyemi, 2006). Egypt implemented its E-Signature Law in 2004, two years
after the enactment of the South African ECT Act.
Countries like Kenya, Uganda and Tanzania are
in the process of finalizing their cyber crime laws
which will cover data security, network security,
cyber crime, information systems and electronic
transactions (Ruiz, 2006).
Developing countries in South East Asia (e.g.,
Malaysia and Singapore) adopted the e-commerce
legislation in 1998, much earlier than South Africa.
Their laws are also based on the UNCITRAL
model law and security. Other countries like
Columbia approved similar laws in 1999 and
Argentina in 2001. Most developed countries (e.g.,
United States of America, EU countries, Canada
and Australia) have often led the development of
e-commerce regulatory activity. The Australian
E-Commerce Privacy and Online Gambling legislation is remarkable for its controls over online
content and spam.
90
While national regulations and legislations
have had significant impact on e-commerce and
have provided certainty in respect of a number
of previously confusing areas, compliance with
such laws continue to be a major challenge for
many organizations (SBP, 2003; Massey, 2003,
Upfold et al, 2005). In the following sections, the
researcher reviews some fundamental theories on
compliance and the strategies adopted to address
compliance challenges.
Compliance Theories and Strategies
Compliance is a state in which someone or something is in accordance with established guidelines,
specifications, or legislation. Theories about
compliance provide accounts of why institutions
and individuals comply with or do not comply
with international and domestic laws (Bentham,
1789; Becker, 1968). In the current study we will
focus on compliance with domestic laws. According to Grossman and Zaelke (2005), rationalist
and normative models have been widely used to
explain compliance. Rationalist models focus on
deterrence and enforcement as a means to prevent
and punish non-compliance while normative
models encourage cooperation and compliance
assistance. They contend that compliance has to
do with behavioral motivation and categorized
the basic logic of human action into the logic of
consequences and logic of appropriateness. In the
first category, the actors are perceived to choose
rationally among alternatives based on their calculations of expected consequences while in the
second category human action is considered to be
based on identities, obligations, and conceptions
of appropriate action.
Rationalist Theories
The rationalist model of domestic compliance
follows the logic of consequences and posits that
regulated individuals act rationally to maximize
their economic self-interest. This emphasizes en-
Factors Influencing SME Compliance with Government Regulation on Use of IT
forcement and deterrence as measures to change
the individual’s economic perceptions. However,
this model has several limitations. For instance, it
does not consider social action based on values,
beliefs, or emotions. Value is restricted to utility
measured in economic terms. The assumption
that self-interested actors calculate the costs and
benefits of alternatives does not consider peoples’
inabilities to calculate probabilities of compound
and conditional events (Coleman, Fararo & Zey,
1993).
Furthermore, the model assumes certainty in
present and future evaluations of consequences.
This may not always be true especially in the present environment characterized by rapid changes in
demand, technology and legislations. This model
also fails to explain the issue of conflict. While
compliance is conceptualized as an economic
transaction in which both parties benefit from the
relationship, in reality there may be conflicts and
resistance when one party gains from a transaction and the other does not (Coleman et al., 1993).
In addition, many examples show that deterrence is difficult and costly to enforce. During
apartheid, the laws discouraged the development of black entrepreneurial activities in South
Africa and it was impossible for blacks to get
financial assistance from banks and government institutions. Despite this discouragement
and discrimination, successful taxi businesses,
Spaza shops and personal services developed
in the homelands (SMME Review 2003, 2004).
Schronen and Van der Merwe (2002) also argue
that the ineffectiveness of the enforcement system
may facilitate non-compliance. They blame the
escalating e-crime problem on the South African
court system and on the detectives’ inability to
handle e-crime cases. They state that 20% of the
reported white-collar crimes reach the courts and
of these only 8.3% result in convictions. Gordon
(n.d.) argues that since computer crime is a new
field in South Africa, the police service may be
hesitant to get involved in certain computer crime
investigations. He adds that the decision to prosecute is usually difficult to make since such cases
take a long time and they are expensive. This is
strongly supported by recent findings showing
that it takes 206 days to enforce legal contracts in
South Africa compared to 39 days in Netherlands
(SMME Review 2003, 2004).
Studies conducted elsewhere also report incidents where the law has been found to be lacking
in its enforcement mechanisms and penalties
(Furlong, 1991). In Australia, for instance, there
has been concern over the fact that the types of
penalties invoked on spammers did not address
the severity of the crime (Vircom.com, 2004).
More recently in United States, there has also
been concern over the CAN-SPAM Act of 2003.
Critics argue that this Act may potentially increase the amount of spam by providing a set of
guidelines for spammers on how to spam legally
(Zhang, 2005).There is also fear that this act will
lead more advertisers to rely on spam as their
preferred method of advertising.
Sometimes an organization may deliberately
fail to comply with aspects of the legislation if
the cost of compliance far exceeds the potential
liabilities (Mayne, 2006). Such organizations may
set aside funds at the beginning of their projects
to cover potential liabilities. Lynch (2005) discusses the difficulties in deterring and catching
phishers. She argues that computer crime’s effective anonymity and the fact that cybercrimes
are much less expensive to commit limits the
chances of catching and punishing perpetrators.
Zhang (2005) also contends that unlike traditional
telemarketers and junk mailers, spammers spend
significantly less costs in advertising their stuff
and often shift the costs to recipients. Cheating
in online auctions (e.g., shill bidding) also leaves
no direct evidence of its occurrence to validate
the facts and as such it is difficult to punish the
cheaters via law enforcement (Jenamani, Zhong
& Bhargava, 2007).
91
Factors Influencing SME Compliance with Government Regulation on Use of IT
Normative Theories
The assumption that people obey the law because
it is in their interest to do so has been criticized for
failing to account for factors such as social influence, moral values and perceived legitimacy of the
regulations (Grossman et al., 2005; Tyler,1990).
The normative perspective on compliance held by
the sociology literature is that laws that are developed and implemented fairly should be followed
and that individuals and firms will comply if they
believe the rules are legitimate and fairly applied
(Tyler, 1990). Similar views have been expressed
in the public choice literature and psychological
theories of leadership (Tyler, 1990). In his study
of the challenges SMMEs faced in reconstructing South Africa, Visagie (1997) reports that as
many as 25.9% of the entrepreneurs surveyed
considered the affirmative action policy to be
reversed discrimination, 28.1% refused to apply
this policy, and 39.1% believed it would lead to
lowering of standards.
Research in psychology and sociology emphasizes the importance of cognitive and social learning processes (Kohlberg, 1984). Here, compliance
with rules is perceived to depend very much on
the individual’s capabilities and ability to address
environmental influences. They maintain that
non-compliance will occur mainly in situations
where the regulated entity lacks capacity (e.g.,
knowledge of the rules, financial abilities and
technological abilities) and commitment. Many
small businesses are complacent about cyberattacks and often shun good security systems
(Zorz, 2002). Upfold and Sewry (2005) found
that most SME leaders in South Africa had not
adopted any security standards and saw information security as a technical intervention to address
basic security issues such as virus threats and
data backup. This false sense of security is partly
attributed to the lack of awareness of the security
problem. Orford et al. (2004) attributes the high
rate of small business failure to poor education
and business skills. Lack of compliance with the
92
taxation and computer security regulations in the
South African SME sector has also been attributed
to factors such as lack of planning, lack of skills,
limited resources, and poor communication. Such
factors impede the implementation of legislative
reforms and upgrade of technologies.
Psychological and sociological theories
therefore call for a more cooperative approach,
increased dissemination of information, technological assistance and adequate inspections.
The South African Revenue services adopted
this approach by providing incentives to SMEs
to encourage compliance with taxation regulations (South African Revenue Services, 2006).
Thatcher, Foster and Ling (2005) argue that the
adoption of Electronic Data Interchange (EDI)
in Singapore was successful because the government took a leading role in IT development and
launched educational programs which exposed
the communities to this technology.
However, normative models also have some
limitations. According to Grossman et al. (2005)
compliance norms become salient when there are
indications of the relevance of the norm. Actors
therefore need to be conversant with the standards
and have authority to initiate actions. However,
in an environment where rules and regulations
are changing rapidly and increasing in numbers
and complexity, the actors may lack a clear understanding of what is required of them or the
capacity to comply (Brown, 1994). Furthermore,
this model implicates an environment where
there is adequate information flow and interactions between stakeholders (e.g., owner, manager,
employees, etc.). This may not always be the case
in smaller organizations.
Other Models
There are other models that can be used to explain
compliance. In his sense-making model, Weick
(1995, 2001) argues that where there is information
overload, complexity, and ambiguity (e.g., lack of
clarity and consistency) people will ignore what-
Factors Influencing SME Compliance with Government Regulation on Use of IT
ever does not make sense to them. Game theory
has also been used to explain compliance issues.
For instance, Rabin (1993) introduced the concept
of fairness equilibrium whereby it is argued that
the belief about the attitude of a counterpart in a
game may change behavior of people and that of
the organization. Compliance is usually achieved
when the beliefs of players are consistent.
Trust also influences compliance to a great
extent. This consists of correct expectations
about actions of others that have an influence on
one’s own choices (Gambetta, 1988). Gupta and
Zhdanor (2006) contend that lack of trust among
individuals involved in a transaction may result
in non-compliance. In early 2000, SMEs in South
Africa lost their confidence in the Khula support
program for failing to deliver on its mandate due
to poor management, poor corporate governance
and greed by its members (Hudson, 2003). Jensen
(2004) found that one of the reasons inhibiting
SMEs from adopting e-commerce effectively was
the fact that the legal system did not provide them
with proper protection especially when transactions occur internationally.
Research in the business and strategic management fields also shows that structural and
organizational components of the firm influence
compliance behavior (Henriques & Sadorsky,
1996; Malloy, 2003). Large firms have many
characteristics that encourage compliance e.g.,
size, resources, skills, public image, and so forth.
However greater coordination, monitoring, or inspection costs may affect their ability to comply.
Small-sized firms, on the other hand, face poor
allocation of resources, limited capital to invest in
technology or hire appropriate resources, limited
cash flow to pay for penalties and lack of planning.
Compliance Strategies
The theories introduced above provide useful explanatory frameworks for examining compliance
issues in the SMEs. Researchers and practitioners
have also developed various compliance strategies
based on the above assumptions and some of these
are discussed in this section. Kalember (2004) has
suggested that enforcement of compliance with
e.g., SOX, HIPAA & GLBA, can be achieved by
adopting good practices. For instance, he states
that organizations can successfully comply with
SOX by following these steps:
•
•
•
•
•
Determine how the system will be operated
and configured once it is in compliance,
including processes and controls.
Assess the current state of the system,
performing a gap analysis relative to the
compliant state.
Implement any process improvements or
new controls, and remedy identified vulnerabilities.
Monitor the system to ensure that it is in line
with the compliance requirements (i.e., with
vulnerability scanning, intrusion detection,
or log monitoring).
Report on the compliance status in a format
that is intelligible to the audit staff or other
management.
Since it is now important for organizations
to know which documents to save and what can
be safely deleted, Mullon (2006) and Elc (2005)
also recommend the adoption of professional and
international standards like ISO 17999 and SAN
15801 (South African National standard) to effectively manage electronic documents (Mostert,
2005). They argue that these can help organizations understand their information assets, the
asset risks involved and also provide a basis for
development of good document retention policies.
Organizations are also realizing the importance of enforcing compliance by empowering
the regulated entities through capacity building
and use of technology. A general view held by
many is that organizations fail to comply not
because they do not want to, but because they
lack awareness of the law and its implications,
lack necessary skills and resources and techno-
93
Factors Influencing SME Compliance with Government Regulation on Use of IT
logical abilities (Allan, Annear, Beck & Beveren,
2003; Upfold & Sewry, 2005). The government of
South Africa has implemented capacity building
programs and support programs such as Ntsika,
Khula and SETAs. Use of technologies can also
enable compliance by facilitating planning and
analysis, and by automating the compliance and
monitoring processes. ISIZA developed tools
to assist SMEs evaluate their compliance with
IT security policies. Modern applications have
tools and templates that could be used to draw
contracts, archive e-mails, update documents
and notices required for legal compliance. These
cost less than the legal fees. Microsoft Exchange
2007 is another useful tool for e-mail archiving
and many audit applications can also be used to
demonstrate compliance.
Towards a Framework for Measuring
Compliance in SMEs
The literature suggests that compliance with
legislation is influenced by various factors. For
instance, economic factors (e.g., potential legal/
illegal gains, cost of compliance); psychological
factors (e.g., cognitive capabilities, learning, previous interactions, individual capabilities, ability to
address environmental influences); sociological
factors (e.g., what is perceived to be moral, fair or
equitable, norms of the groups involved and trust);
the industry (professional standards, ethics, good
practice); technological factors (e.g., IT skills, IT
resources, content and complexity of electronic
information) and business related factors (such
as type, size and location).
This suggests therefore that no single model
(e.g., rationalist, normative or business management models) can comprehensively explain
compliance behaviors (Black, 2001; Brown,
1994). One framework that captures most of these
aspects is the Australian Taxation Office compliance model (Murphy, 2004). This framework
blends various compliance aspects to provide
insight into compliance behaviors of regulated
94
entities. It has assisted the Australian Taxation
Office (ATO) in understanding short comings in
their practice, helped them acknowledge relationships with their clients and also in addressing
various compliance issues (Murphy, 2004). This
framework has therefore been adapted by adding
the technological factor which is crucial to the
present study. Figure 1 represents the adapted
framework. The researcher contends that the six
factors used in this framework mold the perceptions and reactions of SMEs to the regulation on
use of IT. By using this framework to examine
these compliance factors, useful insight into the
problems experienced by SMEs will be provided
and organizations may use it to make appropriate
predictions about compliance behaviors.
Literature suggests that organizations that
comply with legislations usually possess or have
positive perceptions of certain drivers of compliance. Compliance was therefore determined by
evaluating the possession and the perceptions of
SMEs of certain compliance drivers. First the researcher identified relevant drivers of compliance
as indicated in Figure 1. Respondents were then
requested to indicate their opinion about statements that evaluated these drivers. For instance,
technological drivers such as possession of adequate security controls and effective management
of electronic records indicate compliance with
requirements in chapters III, VIII, and XIII of
the ECT Act. If the respondents possessed these
drivers, this would then indicate a possibility of
complying with the act. On the other hand, if
respondents perceived the act to be unfair or immoral (sociological factor), chances of compliance
with the act would be slim. A questionnaire was
developed based on this framework and the details
are provided in the following section.
METHODOLOGY
The researcher employed three techniques (observations, survey and interviews) to collect and
Factors Influencing SME Compliance with Government Regulation on Use of IT
Figure 1. Factors influencing SME compliance with regulation on use of IT (Adapted from the Australian
Taxation Office -ATO)
Business
e.g. Appropriate business
exposure, proximity to
support facilities (location),
approprite allocation of
resouces, availability of
Capital
Sociological
e.g. perception of
fair/moral legislation
Technological
e.g. possession of IT skills,
up-to-date IT systems,
appropriate electronic
record management
capabilities, quality
systems controls/security
SME
compliance
with regulation
on use of IT
Industry
e.g. Adoption/use of good
practices, demonstration of
compliance
Psychological
e.g. possession of cognitive
capabilities, awareness of the
legislation, ability to
understand environ. issues
validate the accuracy of responses in the present
study (Yin, 1994). A search on Proudly South African Web site and also in Braby business directory
provided several SMEs engaged in e-commerce.
A random sample from different business sectors,
age, size, and locations was extracted.
A questionnaire was developed to collect further
information from SME respondents (see Appendix
1). The framing of the questions was guided by similar previous studies (e.g., Evans & Shirley, 2005;
Hurley, 2006; SBP, 2003), key issues discussed in
literature, and input from academics and business
managers. This questionnaire was piloted with
selected business managers, IT service providers,
students and academics in order to get feedback on
its structure, content and other possible problems.
The questions used cover most of the compliance
elements outlined in the sections above.
Section A collected general information about
the characteristics of the organization and in particular, years in business, the business sector, size,
number of employees, and experience in doing
business on the Internet. Section B measured the
extent to which respondents possessed or used the
drivers and their perceptions of some of these drivers. The questions asked in this section evaluated
the elements in the compliance model. For instance,
these determined:
Economic
e.g. perception of benefits,
perception of lower costs of
compliance
1.
Whether they are aware of the electronic
communication and transaction act.
2. Whether they adopted or possessed professional standards such as 1SO 17799, SANS
15801.
3. Whether they perceived the costs of compliance to be high.
4. Whether they demonstrated compliance, for
example through security audits.
5. Whether there was a specific person responsible for ensuring compliance in the
organization.
6. Whether they possessed policies on compliance and record retention.
7. Whether the compliance process was automated, for example, through the use of
application tools.
8. Whether adequate system security measures
or controls were implemented.
9. Whether backups and e-mail archiving were
done promptly.
10. Whether they participated in compliance
and security education or training.
11. Whether sufficient IT resources were allocated for ensuring compliance.
12. Whether they perceived the act to be just and
fair and trusted its implementation process.
95
Factors Influencing SME Compliance with Government Regulation on Use of IT
In the last part of section B, the researcher
collected information about the opinions of the
respondents regarding the fairness of the ECT Act.
In April, 2007, a letter explaining the purpose
of the study together with the questionnaire were
sent to the organizations in four provinces of
South Africa. Two hundred questionnaires were
administered in total and only 50 were returned
resulting in a response rate of only 25%. Further
effort was made to increase the number of responses by contacting non-respondents. Fifteen
additional responses were received in July while
others declined to respond because of company
policy restrictions, lack of time and lack of interest.
In September, 2007, a second round of surveys was
conducted. An additional 120 questionnaires were
administered to respondents in different organizations located in Cape Town, Limpopo, Eastern
Cape, and Johannesburg. Fifty-nine responses
were received thereby providing 124 responses
in total and a response rate of 39 percent.
Follow-up interviews with respondents were
then organized by the researcher. Twenty six interviews were conducted altogether (16 with urban
managers and 10 with rural managers). Only ten
of these interviews were held at the respondents’
premises. The rest had to be conducted by telephone because respondents had limited time. The
interviews were semi-structured, guided by earlier
observations on Web sites, the questionnaire and
key responses or comments made by managers.
These interviews allowed the researcher an opportunity to verify and validate several survey
responses and earlier Web site observations.
Test for Non-Response Bias
In order to confirm that the responses came from
the same population, a test was conducted whereby
the responses obtained in the earlier survey were
compared with those obtained in the second (including late responses). The t test results show that
no significant difference between the two groups
existed (t = 1.45; p = 0.161 at 5% significant level).
96
RESULTS AND ANALYSIS
In this section, the researcher presents the findings and analysis of the results. Both qualitative
techniques and quantitative methods were used
to analyze the data.
General Information about
Respondents
The characteristics of the respondents are summarized in Table 1. Most respondents were
business managers or owners and majority came
from the service industries (e.g., retail, finance
and marketing). Those involved in agricultural
activities and property management possess
more years of business experience (18.81, 17.17
respectively). While managers in property management, tourism, marketing, finance and retail
appear to have much broader Internet experience,
those in the agricultural sector are least involved
in online trading. The researcher found that the
Internet is mainly used for online buying and
selling, provision of services, marketing and
advertising. Further analysis shows that 31% of
the respondents were female managers mainly
in the retail and finance sectors. Furthermore,
only 37% of the respondents are located in urban
areas and majority (63%) operate in rural regions
of the Free State, Kwazulu-Natal, Eastern Cape
and Limpopo provinces. The breakdown of the
sample by industry corresponds to some extent
to the industry’s e-commerce contribution to the
national product as reflected by SAITIS Baseline
(2000). For instance, of the 16 billion Rand generated from e-commerce in 1998, Manufacturing
contributed (38%), Retail (16%), Finance (13%),
Construction (3%), Agriculture (3%) and other
services (19%).
Since there are major differences in the profile of rural and urban managers in South Africa
(Orford et al, 2004), and it is alleged that these
differences impact on SME potential, planning and
performance, the researcher decided to compare
Factors Influencing SME Compliance with Government Regulation on Use of IT
Table 1. Sample characteristics of SMEs
Industry
No
Business experience
(Yrs) Mean (SD)
No. of employees
Mean (SD)
Internet Trading Experience (Yrs) Mean
(SD)
Gender of Managers
M
F
Manufacturing and Production
10
11.29 (4.79)
10-50
4.72 (2.24)
10
0
Retail and Wholesale
28
14.52 (5.16)
0-15
5.19 (3.15)
20
8
Finance
21
12.52 (5.19)
4-10
5.75 (3.85)
14
7
Construction
Marketing
Education and Training
Agriculture
Tourism, Hotel and Tours
Property Management
Total
8
13.87 (5.84)
15-40
2.63 (1.36)
8
0
22
12.41 (5.37)
3-12
7.86 (3.44)
17
5
5
11.25 (3.14)
1-9
2.75 (0.96)
2
3
9
18.81 (8.64)
5-24
2.22 (1.48)
7
2
10
14.36 (8.72)
15-44
9.64 (3.93)
5
5
11
17.17 (6.91)
0-15
10.53 (1.22)
124
the compliance results in these two groups. First,
the mean score for each group on the driver was
determined by assigning values to the responses.
The range of values used corresponded to the
number of choices provided in each questions
(excluding the N/A option). For example, values
from 1 to 5 (where 1 = lowest and 5 = highest)
were used in the case of five choices, and 1 to 2
(where 1 = No, 2 = Yes) in the case of two choices.
The results are presented in Table 2 and discussed
in subsequent sections.
Awareness and Training
Awareness of the Act
Table 2 indicate that awareness of the ECT Act had
the lowest mean score in the rural region and fairly
high score in urban areas. This means that most
rural SMEs were not certain about the requirements
and liabilities of the act (mean score = 2.32) where
as their urban counterparts knew some of these
requirements (mean score = 3.41). This difference in
mean scores suggests that disparities in awareness
exist between these two groups. Further analysis
indicates that in total, 32% of the respondents were
not aware of the requirements and liabilities of the
act, 16% were fully aware and 34% only knew some
3
86 (69%)
8
38 (31%)
of the requirements. The rest of the respondents
indicated that the question on awareness was not
applicable to them. The researcher also conducted
t-tests to determine the significance of the differences due to location and gender. Results confirm
that such differences in awareness exist between
rural and urban SMEs (t = 1.68; p = 0.001) and also
between male and female respondents (t = 1.48; p
= 0.005) at 5% significant level.
These results support earlier claims that lack of
awareness is a major factor affecting compliance
with regulations in South Africa (Elc, 2005; Orford
et al., 2004; Technews, 2004; Upfold & Sewry,
2005). Discussions with four female respondents
in rural areas revealed that attention to compliance issues placed additional time constraints on
their domestic and business responsibilities. They
attributed the lack of awareness of the regulations to the high degree of illiteracy and lack of
support for female entrepreneurs in their region.
This confirms earlier observations by Rogerson
(2004) that existing national support programs
respond better to the needs of urban than rural
and emerging SMEs.
The qualitative analysis of responses also
provides further evidence to support some of the
mentioned findings. Some of the remarks made
by managers were:
97
Factors Influencing SME Compliance with Government Regulation on Use of IT
Table 2. Mean scores and ranks
Rural: N = 78 (63%)
Urban: N= 46 (37%)
Mean Score
Mean Score
Awareness of the legislation
2.07(1.17)
3.39(1.29)
2
Adoption of professional standards
2.62(1.12)
3.41(1.38)
3
IT budget allocation
2.32(1.22)
3.01(1.19)
4
Cost of compliance
4.53 (1.58)
4. 01(1.43)
5
Demonstration of compliance
2.44(1.25)
2.91(1.18)
6
Backup, restore, electronic document management
3.01(1.10)
3.59(1.01)
7
Compliance and security training
2.53(0.96)
2.87(0.90)
8
Implementation of security controls
2.81(1.12)
3.81(1.29)
9**
Perception of penalties and liabilities
1.33(0.49)
1.64(0.55)
10
Possession of record retention policy
1.29(0.47)
1.44(0.51)
11
Possession of compliance policy
1.11(0.21)
1.14(0.22)
12
Presence of personnel responsible for compliance
1.15(0.31)
1.55(0.46)
13
Presence of qualified IT person
1.55(0.42)
1.60(0.51)
Ref
Compliance drivers
1*
* Items 1 – 8 were measured on a scale of 1 to 5 (N/A responses were excluded)
** Items 9 – 13 were measured on a scale of 1 to 2 (N/A responses were excluded
This act is never intended for small firms like mine
…. Besides no one has bothered to tell us more
about it …. [Urban manager]
What is this ECT Act?” …… “Which one is this
now? Don’t we have enough regulations already?
[Rural manager].
Three urban managers had this to say when
asked why they had not implemented legal notices,
opt-out options and clear terms on their Web sites:
We are not sure of all these requirements….”
[manager 1]. “When we implemented our Web
site, our main focus was mainly on meeting business needs not the compliance you are asking me
about…..” [manager 2]. “You ask me about legal
notices……I do not even know what format these
should take but I’m sure this should be handled
by our staff…. [manager 3]
98
Lack of awareness and misconceptions of regulations in SMEs seem to be common problems not
only in South Africa but also in other parts of the
world. Mayne (2006) argues that small firms in the
developed countries also lack awareness and clear
understanding of the legal requirements governing
trade across international borders. Schmidt, Bennison, Bainbridge and Hallsworth (2007) report
that information overload and lack of awareness
of where to get information on regulations affect
the ability of UK SME retailers to comply with
regulations. Unum (2004) also found that almost
75% of small business owners surveyed in UK
were entirely ignorant of the changes to the Disability Discrimination Act.
Respondents in the present study who were
fully aware of the act were motivated by factors
such as participation in the consultative process
that led to the drafting of the act. Others were
forced to work towards compliance by external
consultants from banks, auditing firms and insur-
Factors Influencing SME Compliance with Government Regulation on Use of IT
ance companies. There were also those who simply
feared the consequences of non-compliance. These
findings indicate therefore that factors such as
pressure from business partners, fear of loss of
reputation and punishment, and moral reasons
can impact on SME compliance.
Compliance and Security Training
Table 2 indicates that rural and urban respondents
were not involved in regular compliance and security training programs (2.53, 2.87 respectively).
The researcher found that 48% participated in
these programs once a year and none conducted
these programs weekly. It was also revealed that
regular training could not be conducted due to lack
of sufficient financial resources to hire consultants.
Respondents also alleged that where training
assistance was provided, this was limited to certain individuals or regions and in some cases the
mentors assigned to them inspired limited trust.
Training on security and compliance are vital for
SMEs which possess limited skills and resources
to meet compliance requirements.
Good Practice
Michalson and Hughes (2005) stated that in the case
of controversial issues such as those relating to the
authenticity and admissibility of electronic imaging, an organization would be in a better position
to counter technical challenges to evidential weight
if it consistently adopted good practices. These
may include use of professional standards; proper
policies, processes and procedures; making use of
expert skills; and use of appropriate data management principles. The researcher examined the extent
to which these practices were used by SMEs.
Adoption of Professional Standards
ISO 17799 and SANS 15801
Since the law does not often provide sufficient
guidelines on how to achieve compliance, profes-
sional standards can act as useful frameworks for
developing good practice necessary for ensuring
compliance. Table 2 suggests however that several
SMEs have not adopted these standards. The urban
group only implemented some of these standards
(3.41) while most rural SMEs were just planning
to do so (2.62). Further analysis indicates that only
26% adopted most of the professional standards,
12% adopted a few and 44% had not adopted any
standards. The rest of the respondents indicated
that the question was not applicable to them. Most
of those who implemented standards operate in
the urban areas in industries such as finance,
marketing and production. The researcher found
that some urban SMEs possessed ITIL, SANS
17799-2, SAN 15489 and HIPAA compliance
certifications while others had implemented procedures which were in line with King II report
and JSE (Johannesburg Stock Exchange) listing
requirements. There were concerns however that
the process of adoption was generally difficult
and assistance with the implementation had to be
provided. According to one manager:
We struggled at first with the issue of document
retention. We did not clearly know what exactly
this meant, what needed to be retained, how and
when. Fortunately, with the assistance of our
consultants, we have adopted some guidelines
in SANS 15801 and SANS 15489. I can now confidently say that our documents are reliable to a
reasonable extent.
Similar difficulties in adopting standards and
good practices are also experienced by firms in the
developed countries. Patton (2006) reports that in
the USA, small firms still have trouble complying
with HIPAA requirements and identified cost and
limited resources as the major constraints. She
cites the AHIMA survey in which 55% of the
respondents identified resources as their most
significant barrier to full privacy compliance.
Taylor, McWilliam, Gresty and Hanneghan (2005)
also found that many SMEs in UK struggle to
99
Factors Influencing SME Compliance with Government Regulation on Use of IT
achieve accreditation and it is not easy for them
to keep up-to-date with relevant and changing
EU legislations
Possession of Policies (Record Retention Policy and ECT Act Compliance
Policy)
Table 2 also indicates that most rural and urban
respondents did not possess a record retention
policy (1.29, 1.44 respectively) and an ECT
compliance policy (1.11, 1.14 respectively).These
organizations did not seem to appreciate the fact
that with proper policies in place, compliance effort can be focused and planning conducted more
proactively. Planning and policy development are
still problems in many SMEs in the Southern
African region. In their study of e-readiness of
SMEs in Botswana, Mutula and Van Brakel (2006)
found that a number of organizations could not
develop e-readiness strategies. Such problems
have been attributed to lack of skills and planning
information (Macleod, 1995). Entrepreneurs in
many countries are also known to have strong
desire for autonomy and control. They tend to be
rigid and traditional in their business practices
(Curtis, 1983) which affects the development of
policies and specialized skills in areas like IT.
Existence of Support Personnel
(Person Responsible for Compliance,
Qualified IT Person)
Twenty-three percent of the respondents had qualified IT personnel assisting with compliance and
only 8% had a specific person fully responsible
for compliance. Table 2 shows that while both
rural and urban SMEs had someone to assist with
IT needs (1.55, 1.60 respectively); only the urban
SMEs had specific people dedicated to compliance
activities (1.15, 1.55). This is not surprising given
the level of awareness disparities in the regions
and resource constraints. Majority of the SMEs
100
are still struggling to acquire affordable and reliable support services in technical and legal fields.
Backup and E-Mail-Archival
Table 2 suggests that most rural and urban SMEs
appear to backup and archive records once a year
(3.01, 3.59 respectively). Forty-eight percent of
the respondents indicated that they back-up and
archive twice a year, 12% quarterly, 10% weekly
and 16% never do so. Experts suggest that an
effective backup strategy should enable the recovery of electronic data whenever necessary. It
is usually recommended that a planned schedule
of backup cycles should be implemented. For
instance, an end of month full backup should be
done, followed by daily or weekly incremental
backups for the remainder of the month. Monthly
backups should be retained for at least a year and
copies of which kept off-site.
Demonstration of Effort Made Towards
Compliance
The results in Table 2 indicate that while the urban score on this driver is slightly higher (2.91)
compared to that of the rural respondents (2.44),
SMEs in this study do not generally demonstrate
their effort towards compliance. Majority (54%) of
all respondents never reported their effort towards
compliance to auditors, 4% reported twice a year,
23% reported once a year, 6% reported quarterly
and 13% indicated that this question was not
applicable to them. This problem could perhaps
be explained by the fact that small organizations
do not keep records and may not engage the services of auditors (Rwigema & Karungu, 1999).
Interviews held with some managers confirm that
many are not involved in rigorous audits and find
the service fees of auditors and lawyers unaffordable. While some have occasionally hired the
services of accountants, these are usually limited
to preparations of end of year financial statements
and tax returns. Rigorous examination of systems
Factors Influencing SME Compliance with Government Regulation on Use of IT
functionality and control is hardly done. The
researcher also found that the approaches used
to evaluate compliance are still predominantly
manual. These organizations have not adopted
automated techniques to measure compliance on
a continuous basis.
These findings contrasted with those obtained
by Evans et al. (2005) in Australia. According to
Evans et al., over 72% of the respondents perceived
the quality of their business records to be high and
65% used computerized record-keeping systems.
The problem of failure to demonstrate compliance regularly seems to be world-wide however.
A world-wide study by the Security Compliance
Council (2005), which involved over 200 IT securities and compliance professionals shows that
95% of the respondents still depend on manual
procedures in IT for demonstrating compliance.
This limits the ability of the organizations to
identify problems before hand and makes it difficult to evaluate the effectiveness or necessity of
proposed changes to business procedures.
However, there were a few firms in the financial and marketing sectors that benefited from the
automation of the audit processes in the present
study. In order to establish whether those reporting to auditors had implemented better system
controls, participated in training and adopted
professional standards than those that failed to
do so, a chi-square test was conducted. This
confirmed that the difference was statistically
significant (p = 0.007).
Perceptions of Cost of Compliance
and Budget Allocation for ECT Act
Compliance
Table 2 also shows that rural SMEs perceive the
cost of compliance to be very high (4.53 = rural,
4.01 = urban). Overall, 72% indicated that the
costs were high or very high, 18% found them to
be moderate, 8% indicated that they were low and
2% stated that they were very low. The researcher
verified some of these responses by inspecting
financial records in a few organizations. These
revealed that fairly high expenses were incurred
not only for maintenance or upgrade of hardware
and software but also for consultation, access to
information, documentation of controls, training
and correction of errors. These findings confirm
earlier observations by the Security Compliance
Council (2005), that IT functions in many parts
of the world spent significant amount of staff time
documenting controls, prioritizing deficiencies for
remediation and making changes to procedures.
Schmidt, Bennison and Bainbridge (2007) also
report that over 63% of the SME retailers surveyed
in UK spent much time every week on compliance related work. Other factors raised by SMEs
which made compliance costly in the present
study include the burden of complying with other
regulations and the fact that compliance impact
on ability to perform other business functions.
One rural female respondent complained about
having to spend her private time in the evening
working on compliance issues.
The analysis of budget allocations shows that
rural SMEs allocate not more than 10% of their
IT budget on ensuring compliance (2.32) while
that of urban SMEs does not exceed 30% (3.01).
This indicates that not many SMEs allocate sufficient resources for this purpose. Since the cost of
compliance was generally high, it is possible that
SMEs had to divert funds or resources from other
commitments to meet compliance obligations. It
is important therefore that SMEs understand the
cost implications of compliance and also plan
properly for this activity to minimize costs. In his
study of the legal and regulatory environment for
SMEs in Kazakhstan and Uzbekistan, Klawonn
(2001) reports that an overly burdensome legal
and regulatory environment made it difficult and
costly for SMEs to compliance. Similar concerns
over costs of compliance have also been echoed
in studies conducted in other areas in South Africa, New Zealand and Australia. For instance,
The Strategic Business Partnership for growth
in Africa - SBP (2005) reports that firms in the
101
Factors Influencing SME Compliance with Government Regulation on Use of IT
tourism industry are burdened with high regulatory costs compared to firms in other sectors.
Goldsworth (2002) also reports that compliance
costs are an issue for every business in New
Zealand and in particular administrative costs are
quite substantial for SMEs. This is confirmed by
Massey (2003) who identified factors such as cost
and time needed to represent the firm overseas,
lack of technical skills and difficulties in firing
employees who do not perform. In Massey’s study,
57.8% of the respondents felt that compliance
distracts and diverts senior staff time, 36.3% were
concerned about the ambiguity and misinterpretation of regulations and 36.1% felt that it dampens
manager’s enthusiasm for innovation.
Possession of Adequate Systems
Controls
This was determined by asking respondents to
indicate the quality of measures implemented
to protect their information resources. Possession of only basic measures such as passwords
and anti-virus software would indicate that the
organizations had inadequate controls while the
presence of advanced security mechanisms certified by recognized authorities or professional
bodies would indicate existence of tougher and
adequate controls. Table 2 suggests that controls
implemented by rural SMEs (2.81) were relatively
inadequate compared with those implemented by
urban SMEs (3.81). Very few SMEs (18%) had
implemented advanced security measures. Only
4% of the organizations reported possession of
ITIL, ISIZA, Microsoft and other proprietary
certifications. It was difficult to prove the actual
existence of security measures because such
permission was declined in most cases. With
the escalating levels of e-crime, it is surprising
to find that few organizations have implemented
advanced security measures. Technologies can
easily become outdated as new and dynamic
attacks take place. Certification provides some
level of assurance that the controls implemented
102
are up to standard and meet general compliance
requirements. The above problems have also been
identified in many other places. Mutula and Brakel
(2006) found that SMEs in Botswana were not
exploiting the potential of technologies due to lack
of skills. They found that organizations were still
using obsolete technologies and only employed
basic security controls such as anti-virus software
and regular backups.
Perceptions of the Penalties and
Liabilities Imposed by the ECT Act
of 2002
Table 2 also shows that most respondents in the
urban areas agree that the penalties or liabilities
imposed by the act are just and fair (rural = 1.33,
urban = 1.64). Those who agreed were 34%,
28% did not agree and 38% indicated that they
did not know or the question was not applicable
to them. One respondent could not imagine that
small organizations could ever run into trouble
with the electronic law. Another respondent was
not bothered about the ECT Act requirements
because he felt such issues should be handled
by his service providers and the insurance firm.
Those that disagreed were concerned about lack
of clarity on issues relating to handling scanned
documents and some felt that penalties were not
commensurate with the crime committed in some
cases. Four of these respondents indicated that
legal provisions such as the right to withdrawal
from transactions limited them from selling certain goods online.
Respondents also felt that the consultation
process leading to the drafting of the act was not
all that inclusive. Consequently, their views were
excluded and as such the act was perceived to
be unfair. The researcher could not verify these
claims but in another study reported by ITWeb
(2002), it is stated that only 11% of the respondents were consulted during the drafting of the
ECT legislation.
Factors Influencing SME Compliance with Government Regulation on Use of IT
Similar views have been expressed in other
parts of the world. In their report on corruption in
SMEs, Tag (2006) states that SMEs in countries
like Slovakia engage in corruption because they
do not have strong ties or connections to influence
policy decisions.
Discriminant Function Analysis
Table 2 indicates that in many instances rural and
urban managers had different opinions concerning
compliance with the IT regulation. For instance,
the mean scores for awareness were (2.07, 3.39)
respectively, implementation of controls (2.81,
3.81), cost of compliance (4.53, 4.21), and perception of penalties and liabilities (1.59, 1.64). In
order to understand what factors distinguish rural
SMEs from their urban counterparts in terms of
their psychological, sociological, technological,
business, industry, and economic disposition to IT
regulation, a discriminant function analysis was
conducted and the results are presented in Table 3.
A stepwise Wilk’s Lambda was used and one
discriminant function emerged. This accounted
for 20% of the total variance. Respondents were
also correctly placed in their corresponding groups
(e.g., 82% – rural, 54% – urban). As indicated
in Table 3, correlation coefficients between the
variables and the discriminant function indicate
that five variables contribute to the discriminant
function. Where the rural group scored highly on
the variable, a negative coefficient is indicated.
Five distinguishing factors are revealed in Table
Status of Existing South African
Web Sites
The researcher also examined 30 of the Web site
of organizations involved in this study. About
40% of these had not implemented necessary
legal requirements. Organizations that implemented most requirements were mainly urban
firms in the marketing, finance and property
management industries. Perhaps this is due to
awareness of the requirements, availability of
resources, good practice and regular exposure
to Internet-trading. The researcher also observed
that while most sites had information about their
products and services, few provided information
on legal notices, opt-out options and clear terms
and conditions. These organizations could be in
breach of certain provisions of the ECT Act.
Table 3. Discriminant function analysis results
Ref
Predictor variables
Standardized Canonical Discriminant function coefficients
Wilk’s
Lambda
F-Value
1
Perception of the cost of compliance
-0.74
0.990
111.36**
2
Awareness of the Act
0.66
0.992
41.47*
3
Compliance and security training
0.31
0.942
31.6*
4
Perception of penalties and liabilities
0.21
0.913
10.12*
5
Implementation of adequate security controls
0.20
0.911
9.33*
Canonical R
0.44
Eigen value
0.20
Chi-Square
122.29
Significance
p< 0.001
** p < 0.001, * p < 0.01
103
Factors Influencing SME Compliance with Government Regulation on Use of IT
3. This suggests that rural SMEs are more likely
to incur high compliance costs (-0.74), less likely
to be aware of regulations (0.66), less likely to
be educated on compliance and security (0.31),
more likely to perceive the regulation to be unfair
(0.20) and less likely to have adequate security
controls (0.20). The Eigen value, chi-squared and
Wilk’s Lambda suggest that the variable utilized
give moderate discriminatory power in separating
rural from urban SMEs.
The discriminant function analysis shows that
the cost of compliance (economic factor) was the
key discriminant of rural SMEs from their urban
counterparts in terms of their disposition to the
regulation on use of IT. As revealed in Table 2,
rural SMEs perceived the costs of compliance
to be very high. Further analysis of the results
showed that indirect costs such as difficulties
involved in comprehending the regulations, time
spent looking for information and assistance, overreliance on consultants and the failure to attend
to other important issues affected compliance
in these SMEs. Apparently indirect costs have
not been given much consideration by policymakers in South Africa (SBP, 2003) and in some
developed countries like New Zealand (Massey,
2003). The compliance burdens relating to other
legislations (e.g., penalties, interest charged and
complicated tax and labor registration processes)
also complicate SME compliance problem. The
study therefore supports calls for a review of regulations to take into account the practical realties of
SMEs and to reduce regulatory costs (SBP, 2005;
World Bank, 2004). Regulators should ensure
that complexity in the information presented is
avoided and regular consultations with businesses
on technical and legal issues done. Assistance
through provision of financial support for training
and awareness campaigns will also be necessary.
Other factors distinguishing rural from urban
SMEs were psychological, sociological and technological. Awareness of the Act was the second
highest discriminant variable. Table 2 confirms
that there exists a great disparity in resources
104
and awareness between these two groups. Many
regulations are made on the incorrect assumptions
that all citizens (including those in rural areas)
are equally literate and understand English or
Afrikaans. In their study of issues preventing
compliance with regulatory requirements in the
mining, agri-processing and tourism sectors in
South Africa, SBP (2005) found that many SME
respondents did not know the regulations governing their sectors nor did they know where to find
assistance. SMEs in rural areas are most affected
because of lack of support and resources in their
areas. Similar problems relating to such disparities
have also been reported in the developed world.
For instance, in the UK, the geographic location
of rural SMEs presents major challenges to the
provision of e-government products and services
(Choudrie, Weerakkody & Jones, 2005).
Policy-makers and support institutions should
be more active in assisting SMEs understand regulations and obligations. This could be achieved
through effective communication mechanisms
such as radio programs, advertisements, regional
seminars and affordable online access to materials.
Much on the ECT Act regulation and discussion
forums can be found on the Internet. However,
not many small organizations can afford regular
access to the Internet. SBP (2003) suggests that
basic techniques such as visits, telephone helplines and leaflets could also be very effective.
These measures would also be appropriate in
addressing the growing misconceptions about
the new regulations.
The third factor distinguishing rural SMEs
relates to their perception of the fairness of the
act (sociological factor). The efficacy of the
ECT Act appears to impact on the compliance
of rural SMEs. Some respondents felt that the
‘opt-out’ provision limits what could be sold
online and some alleged that they were not
consulted during the drafting of the Act. While
this lack of involvement could not be verified,
evidence suggests that minority groups often
fail to have their views considered because they
Factors Influencing SME Compliance with Government Regulation on Use of IT
can not lobby policy-makers (Common Wealth
Secretariat, 2002; ITWeb, 2002). The responses
obtained from the female managers in the present study indicate that while the government has
made much effort to address the inequalities of
the past, more work is still needed to reduce the
legislative burdens of female entrepreneurs. Very
few female respondents in this study knew much
about the regulation. Many possess limited skills
and often struggle to manage business, compliance and domestic responsibilities. Extensive
consultations on legal issues affecting different
SMEs and managers should be done if the desired
compliance results are to be achieved.
Technological factors (e.g., implementation
of security controls) surprisingly had the lowest
influence as reflected in Table 3. Table 2 shows
earlier that most SMEs (mainly from rural areas)
had not implemented adequate system controls.
Basic security controls were in use and just a
few possessed controls certified by recognized
professional bodies. One would have expected
this variable to have greater discriminant effect.
Closer examination of results in Table 3 suggests
a shift in the perspective of managers regarding
security. Apparently, concern over compliance
issues such as cost (0.74) and awareness (0.66)
are overtaking the traditional concerns about
security, virus and worms (0.20).The implication
of this would be that attention may be diverted
from the escalating security crisis which SMEs
desperately need to address in order to survive.
This problem should be investigated further by
researchers in future studies.
The present study shows that SMEs did not
make substantial effort towards ensuring compliance and not many have implemented measures to
demonstrate such effort. Comments such as “…
no one has bothered to tell us about it ….” (p.15)
clearly suggest that someone else should take that
responsibility and perhaps should carry the blame
for non-compliance. However, much is revealed
in this study that suggests that SMEs have not
done much on their part. We found, for instance
that more than half of the respondents were still
using basic security measures. Situations where
SMEs have been reluctant to abandon obsolete
and risk prone technologies have been reported in
many parts of the world (Jensen, 2004; Mutula &
Brakel, 2006). Enforcement and deterrence may
be necessary in such situations to alter the calculation of those organizations less inclined to comply
voluntarily. SMEs need to take a more proactive
stance in terms of compliance rather than wait
for the government to do everything for them.
For instance, visits to information centers where
relevant materials are freely provided, upgrading
their systems and taking basic measures to secure
them would reduce potential risks and liabilities.
The industry factor (e.g., good practice) was
excluded from the discriminant function. This
confirms earlier results that both rural and urban
SMEs are not performing very well in areas like
policy development, adoption of professional
standards and demonstration of compliance. This
suggests therefore that South African SMEs in
general need to be encouraged to adopt good
practices in order to minimize potential risks and
liabilities as suggested earlier by (Michaelson and
Hughes, 2005). This study also confirms that business related factors such as structure and location
influence compliance with the law. Providing such
assistance to organizations especially in rural areas
which have limited support and resources will be
necessary if compliance goals are to be achieved.
CONCLUSION, LIMITATIONS, AND
IMPLICATIONS
This study has some limitations. It was quite
difficult getting many SMEs to respond to this
survey. A response rate of 39% was disappointing
given the importance of this issue to the survival
of many SMEs. Many showed limited interest in
this area and this seems to be a major impediment
to the effort directed at ensuring better compliance
with regulations in this sector.
105
Factors Influencing SME Compliance with Government Regulation on Use of IT
Not all regions of South Africa were represented in the sample therefore caution need to be
taken in generalizing the findings. Furthermore,
while some effort was made to gather and verify
responses from multiple sources, few observations
and interviews were conducted to draw strong
conclusions. To gain a more understanding of the
compliance problem in SMEs, further in-depth
investigations involving managers, technical
staff and other stakeholders (e.g., policy-makers)
should be conducted by researchers. In addition,
the variables identified in the model as compliance drivers are not necessarily causal factors,
therefore the causality of these factors need to
be tested in future studies.
However, this study achieved its main objective
of investigating and providing empirical evidence
about the factors influencing SME compliance
with the ECT Act. Such research–based information is necessary to create better understanding and
also enable the development of effective solutions
to the compliance problem in SMEs. By blending
rationalist, normative and business management
models, the researcher provides more comprehensive insight into the effect of compliance drivers
on SME behaviors. The findings indicate that
SME decision to comply or violate the IT law may
be influenced by business, industry, economic,
technological, sociological and psychological
factors. This therefore implicates a combined
strategy that encourages cooperation, assistance
and incentives while maintaining some degree of
enforcement and deterrence where compliance
can not be achieved voluntarily.
Economical factors such as indirect compliance costs were found to be the main barriers to
SME compliance in the present study. Unfortunately, this continues to receive limited attention
from policy-makers. Regulatory impact assessments conducted regularly will be necessary to
identify ways to minimize these costs. The study
also revealed differences in compliance behaviors
of rural and urban SMEs. Policy-makers need to
be aware of the fact that rural SMEs are more
106
likely to incur high compliance costs; are less
likely to be aware of regulations; are less likely
to be educated on compliance and security; are
more likely to perceive regulations to be unfair
and are less likely to have adequate system control
measures. Such differences in behaviors demand
specific rather than blanket solutions to compliance challenges in SMEs.
Majority of the firms surveyed were not proactive in their approaches to compliance. Few
firms planned their security systems, allocated
sufficient resources for compliance, trained their
staff regularly and developed policies. SMEs
need to be encouraged to adopt good practices,
demonstrate compliance effort regularly, and
establish appropriate procedures that will ensure
better and continuous compliance in this sector.
The findings of this study have implications
for other developing countries. For instance Kenya, Uganda, and Tanzania, which are currently
developing similar legislations to enable the establishment of e-government and e-commerce
programs have much to learn from South African
findings (Ruiz, 2006). According to Macculloch
(2001), the regulatory burden of SMEs in East
Africa is very high, the private sector is overregulated, requirements are often confusing and
contradictory and costly administrative delays are
common. Businesses operate in an environment
where there is lack of basic services, facilities,
infrastructure and unreliable power supply. There
is little doubt that given such circumstances,
there will be major barriers to compliance with
the proposed new regulations as revealed in the
present study. It is, therefore, imperative that East
African policy-makers take cognizance of these
potential problems, minimize the complexities of
their regulations, involve many stakeholders in
policy development and address potential disparities between rural and urban SMEs.
This study has also implications for foreign
firms doing (or intending to do) business with
South African SMEs. Understanding of the provisions of the ECT Act, particularly how agreements
Factors Influencing SME Compliance with Government Regulation on Use of IT
are formed using online transactions, rules on
consumer protection and online dispute resolution, will be important. Precautions need to be
taken however when dealing with firms that do
not meet legal requirements. It is imperative that
careful assessment of relationships with business
partners, terms and conditions of agreement and
accuracy of information on product or services
are done. Firms can also seek assistance from
the department of trade and industry or other
support organizations. Foreign firms should also
be aware that agreements concluded on ‘foreign’
Web sites that do not comply with the ECT Act
requirements may not be enforceable in South
Africa, if challenged.
REFERENCES
Allan, C., Annear, J., Beck, E., & Beveren, J.
(2003). A framework for the adoption of ICT
and security technologies by SMEs. Retrieved
January 11, 2005, from http://www.cecc.com.au/
programs/resource_manager/accounts/seaanz_
papers/62AnnearetalFinal.pdf
Becker, G.S. (1968). Crime and punishment: An
economic approach. Journal of Political Economy,
76(2), 169-217.
Bentham, I. (1789). An introduction to the principles of morals and legislation. Oxford: Basil
Blackwell.
Black, J. (2001). Managing discretion. Unpublished manuscript, London School of Economics,
UK.
Brown, R. (1994). Theory and practice of regulatory enforcement: Occupational health and safety
regulation in British Colombia. Law and Policy,
16, 63-71.
Burrows, T. (2004). Making web site compliance
easy. ITWeb. Retrieved September, 04 2005,
from http://www.itWeb.co.za/sections/Internet/2004/0409 220806.asp?A=COV&S=Cover
Business.iafrica (2006). BEE “an opportunity”
for small businesses. Business.iafrica. Retrieved
February 20, 2007, from http://business.iafrica.
com/hidden/542064.htm
CATA – Common wealth Association of Tax Administrators, 2006. Taxation of self employment
sector. Cata Newsletter. Retrieved March 10, 2007,
from http://www.cata-tax.org/upiloads/051006/
NL.spt%202006.pfd
Choudrie, J., Weerakkody, V., Jones, S. (2005).
Realising e-government in the UK: Rural and
urban challenges. Journal of Enterprise Information Management, 18(5), 568-585.
Coleman, S., Fararo, T., & Zey, M. (1993). Rational
Choice Theory: Advocacy and critique. Social
Forces, 72(1), 273-275.
Common wealth Secretariat. (2002). Common
wealth business women: Trade matters, best
practices and success stories. London.
Computer Business Review. (2004). Web site
compliance survey. Computer business review.
Retrieved September 20, 2006, from http://www.
cbr.co.za/article.aspx?pklArticleId=3246&pklCa
tegoryId=384
Curtis, A. (1983). Strategic planning for smaller
businesses. Toronto: Lexington Books.
Dans, E. (2003) IT investment in Small and
Medium Enterprises: Paradoxically Productive?
Electronic Journal of Information Systems Evaluation, 4(1), Retrieved September 20, 2006, from
http://www.ejise.com/volume-4/volume4-issue1/
vol4-issue1-papers.htm
DPRU – Development Policy Research Unit.
(2007). Recent findings on tax-regulated regulatory burden on SMMEs in South Africa. DPRU
Policy Brief series, School of Economics, University of Cape Town.
ECT Act. (2002). Electronic communications and
transactions Act. 2002. Acts Online. Retrieved
March 30, 2005, from http://www.acts.co.za/
ect_act/
107
Factors Influencing SME Compliance with Government Regulation on Use of IT
Elc (2005). 17799 security standard and document
retention. Electronic Consultancy. Retrieved September 13, 2005, from http://elc.co.za/article.php
Evans, C., Carlon, S., & Massey, D. (2005). Record
keeping practices and tax compliance of SMEs.
E-Journal of Tax Research, 3(2), 288-334.
Furlong, W.J. (1991). The deterrent effect of regulatory enforcement in the fishery. Land Economics,
67(1), 116-29.
Gambetta, D.G. (1988). Can we trust trust? In
D.G. Gambetta (Ed.), Trust, (pp. 213-237). New
York: Basil Blackwell.
Goldsworthy, B. (2002). Address by bruce goldsworthy ema (n) to HSNO compliance costs
workshop. ERMA New Zealand. Retrieved
January 10, 2007, from www.ermanz.govt.nz/
news-events/archives/events/erma-conf2002/
bruce-goldsworthy.pdf
Gordon, B.J. (n.d). Internet Criminal Law. Cyberlaw@SA. Retrieved June 22, 2006, from
http://www.cyberlawsa.co.za/cyberlaw/cybertext/
chapter15.htm
Grossman, D., & Zaelke, D. 2005. An introduction to theories of why states and firms do (and
do not) comply with law. INECE Conference
Proceedings. INECE. Retrieved September 10,
2006, from www.inece.org/conference/7/vol1/
index.html
Gupta, A., & Zhdanov, D. (2006). Trust and Fairness as incentives for compliance with Information
security policies. WITS-2006, Milwaukee, WI,
December 9-10, 2006.Retrieved February 12,
2007, from www.misrc.umn.edu/workingpapers/
fullpapers/2006/0607_102306.pdf
Henriques I., & Sadorsky, P. (1996). The determinants of an environmentally responsive firm:
An empirical approach. Journal of Environmental
Economics and Management, 30, 381-384.
108
Hudson, J. (2003). RIAS and private sector development: Some thoughts from the South African
context. Centre on Regulation and Competition.
Retrieved January 4, 2007, from http://www.
competition-regulation.org.uk/conferences/
mcrria03/conf8.pdf
Hurley, J. (2006). The struggle to manage security compliance for multiple regulations. White
paper: Enterprise security. Symantic. Retrieved
February, 15, 2007, from http://searchsecurity.
techtarget.com/whitepaperPage/0,293857,sid14_
gci1169051,00.html
Image & Data Manager. (2006). Poor e-mail
management cost US dollar 20 million. Image
& Data Manager, Retrieved November 20, 2006
from http://www.idm.net.au/story.asp?id =7016
ISIZA- Information Security Institute of South
Africa. (2004). Resources and self assessment
tool. ISIZA. Retrieved September 5, 2004, from
http://www.isi-za.org
ITWeb, 2002. Online industry ignorant of
new laws. ITWeb. Retrieved February 20,
2006, from www.itWeb.co.za/sections/Internet/2002/0210141156.asp?O=TE
Jenamani, M., Zhong, Y., & Bhargava, B. (2007).
Cheating in online auction – Towards explaining the popularity of English auction. Electronic
commerce research and applications. 6, 53-62.
Jensen, J. (2004). Issues facing SMEs in their
adoption of electronic commerce. Retrieved February 24, 2005, from http://www.crf.dcita.gov.au/
papres03/ jensenpaper12final.pdf
Joubert, J., & Van Belle, J.P. (2004). Compliance
of South African E-Commerce Web sites with
Legislation to protect consumer rights. Retrieved
March 14, 2007, from http://www.commerce.uct.
ac.za/InformationSystems/Staff/PersonalPages/
jvbelle/pubs/IBIMA04.doc
Factors Influencing SME Compliance with Government Regulation on Use of IT
Kalember, R. (2004). Compliance Essentials:
standard method of fulfilling requirements, storage
& security journal. Retrieved September 20, 2006
from http://issj.sys-con.com/47516.htm
Klawonn W. (2001). Some observations on the
legal and regulatory environment for SMEs
in Central Asia. Retrieved February 20, 2007,
from http://rru.worldbank.org/paperslinks/open.
aspx?id=3529.
Kohlberg, L. (1969). Stage and sequence: the
cognitive-development approach to socialization.
In Goslin, DA. (Ed.), Handbook of socialisation
theory and research. New York:Rand McNally.
Lombard, E. (2003). Hacker cleans out bank
accounts. Sunday Times News. Retrieved October 22, 2004, from http://www.sundaytimes.
co.za/2003/07/20/news01.asp
Lynch, J. (2005). Identity theft in cyberspace:
Crime control methods and their effectiveness in
combating phishing attacks. Berkeley Technology
Law Journal, 20, 259-300.
Macculloel, F. (2001). Government administrative burdens on SMEs in East Africa: Reviewing
issues and Action. Retrieved May, 20, 2006,
from http://www.blackwell-synergy.com/ doi/
pdf/10.1111/1468-0270.00287
Macleod, G. (1995). Starting your own business
in South Africa (8th ed.). South Africa: Oxford
University Press
Malloy, T.F. (2003). Compliance and the firm.
Temple Law Review, 76, 451-457.
Massey, C. (2003, June 15-18). Business compliance and SMEs: Too much a cost? In Proceedings of the 48th ICSB World conference.,Dublin,
Ireland: ICSB.
Mayne, M. (2006). How are businesses facing up to
the compliance challenge? SC Magazine Australia.
Retrieved November 30, 2006, from http://www.
scmagazine.com.au/feature/how-are-businessesfacing-up-to-the-compliance-challenge.aspx
Michalson, L., & Hughes, B. (2005). Guide to the
ECT Act. Michalsons Attorneys. Retrieved September 20, 2006, from http://www.michalson.com
Mostert. (2005). Imaging Law and SANS 15801.
Mostert Opperman Goodburn Incorporated.
Retrieved January 25, 2007, from World Wide
Web: http://www.mostert.co.za/ Publications.htm
Morgan, G. (2004). Managing e-mail to avoid liability and ensure compliance. Computer Business
Review. Retrieved June 20, 2005, from http://cbr.
co.za/news.aspx
Mutula, S., & Van Brakel, P. (2006). E-readiness
of SMEs in the ICT sector in Botswana with
respect to information access. Emerald. Retrieved January 15, 2006, from www.emeraldinsight.com/.../published/emeraldfulltextarticle/
pdf/2630240310_ref.html
Mullon, P. (2006). The six elements of legally
acceptable electronic documents. Computer Business Review. Technews. Retrieved December 05,
2006, from http://cbr.co.za/article.aspx?pklArticl
eId=3972&pklIssueId=447&pklCategoryId=384
Murphy, K. (2004). Moving towards a more effective model of regulatory enforcement in the
Australian taxation office. Working paper No.
45. Retrieved November 30, 2006, from http://
ctsi.anu.edu.au/publications/WP/45.pdf
Nicol, C. (2003). ICT Policy: A beginner’s Handbook. The Association for Progressive Communications (APC). South Africa. Internet and ICT
for Social Justice and Development.
Ogunyemi, D. (2006). Key challenges to implementing e-commerce in ghana. Economic Commission for Africa. Retrieved October 20, 2007,
from http://www.uneca.org/disd/events/2006/
ecowas-uemoa/content/Key%20challenges%20
in%20implementing%20e-commerce%20in%20
Ghana%20-%20Dayo%20Ogunye mi%20-%20
EN.pdf
109
Factors Influencing SME Compliance with Government Regulation on Use of IT
Orford, J., Herrington, M., & Wood, E. (2004).
South African Report Global Entrepreneurship
Monitor. Retrieved January 22, 2005, from www.
gsb.uct.ac.za/gsbWebb/userfiles/GEM_ 2004.pdf
Patton, S. (2006). Small firms still having trouble
complying with HIPAA. CIO, Retrieved February
20, 2007, from http://www.cio.com/article/25787/
Small_Firms_Still_Having_ Trouble_Complying_with_HIPAA
Rabin, M. (1993). Incorporating fairness into game
theory and economics. The American Economic
Review, 83(5), 1281-1302.
Rankin, N. (2005). The regulatory environment
and SMMEs. Evidence from south african level
data. Retrieved January 24, 2007, from http://
uamp.wits.ac.za/sebs/downloads/2006/the_
regulatory_environment_and_smmes_final.pdf
Rogerson, C. (2004). The impact of the South
African government’s SMME programmes: a tenyear review (1994-2003). Development Southern
Africa, 21(5), 765-773.
Ruiz, M. (2006). Internet Law – Kenya, Uganda
and Tanzania adopt cyber laws. Ibls.com. Retrieved February 20, 2007, from http://www.ibls.
com/Internet_law_news_portal_region.aspx?s
=Africa&id=7
Rwigema, H., & Karungu, P. (1999). SMME development in Johannesburg’s Southern Metropolitan
Local Council: An assessment. Development
Southern Africa, 16(1), 107-124.
SAITIS. (2000). A survey of the IT industry and
related jobs and skills in south africa. SAITIS
BASELINE STUDIES. Retrieved October 14,
2004, from http://www.saitis.co.za
SBP – Strategic business partnership for growth
in Africa (2003). Is south africa a good place
to do business? SME Alert. Retrieved January,
15, 2006, from http://www-za.sbp.org.za/docs/
SME_Alert_Nov_2003.pdf
110
SBP - Strategic business partnership for growth
in Africa (2005). The impact of sector-specific
policies and regulations on the growth of SMEs in
8 sectors of the SA economy. Retrieved February
10, 2007, http://www.thepresidency.gov.za/docs/
pcsa/economic/sbp.pdf
Schmidt, R., Bennison, D., Bainbridge, S., &
Hallsworth, A. (2007). Legislation and SME
retailers—compliance costs and consequences.
International Journal of Retail and Distribution
Management, 35(4), 256-270.
Schronen, J., & Van der Merwe, J. (2002).
White-collar crime cleans up R40bn a year. IOL
South Africa. Retrieved October 15, 2004, from
http://www.iol.co.za/index.php?set_id=1& click_
id=13&art_id=ct20010319091018396L300510
Schwella, E. (2002). Regulation and competition in south africa. Centre on Regulation and
Competition, Institute for Development Policy
and Management, University of Manchester, UK.
Security Compliance Council. (2005). The CSO’S
security compliance Agenda, Benchmark Research Report. Retrieved February 17, 2007, from
http://www.theiia.org/download. cfm?file=94803
SMME Review 2003. (2004). Annual Review
of Small Business in South Africa – 2003. Retrieved January 18, 2007, from http://www.tips.
org.za/2003/SMMEReview SCREENFINAL.pdf
Stern, E.R., & Barbour, P.A. (2005). Designing a small business tax system that enhances
growth: lessons from Africa. Retrieved February, 10 2006, fromhttp://www.fias.net/
ifcext/fias.nsf/AttachmentsByTitle/Zambiataxconference_PapersRichardandPaul/$FILE/
Ta x i ng +SM E s+ i n+A f r ic a _
Richard+Stern+and+Paul+Barbour.pdf
South African Revenue Services. (2006). Introduction of the Small Business Tax Amnesty. SARS.
Retrieved January 22, 2007, from http://www.
sars.gov.za/Small%20Business%20 Amnesty/
default.htm
Factors Influencing SME Compliance with Government Regulation on Use of IT
Sunday Times. (2002). Dozens of threats beset
your data. Sunday Times Surveys. Retrieved
October 5, 2004, from http://www.sundaytimes.
co.za/business/surveys/Internet/survey10.asp
Tag (2006). Expert group meeting on small business development and corruption. Discussion
Paper, 6-7 March. Retrieved January 15, 2006,
from http://www.tag.org.ph/pdf/SmallBussDevt_Corruption.pdf
Taylor, M., McWilliam, J., & Hannegham, M.
(2005). Cyber law: Case studies in the SME
environment. System Research and Behavioral
science, 22(3), 261-267.
Technews. (2004). 2004 Web site compliancy
survey, Computer Business Review. Retrieved
January 15, 2005, from World Wide Web: http://
cbr.co.za/article.aspx
Thatcher, M. Foster, W., & Zhu, L. (2006). B2B
e-commerce adoption decisions in Taiwan: The
interaction of cultural and other institutional
factors. Electronic Commerce Research and Application, 5, 92-104.
Tyler, T. R. (1990). Why people obey the law. New
Haven: Yale University Press.
UNCITRAL (2004). Raising confidence in
e-commerce: The legal framework. Retrieved
September 20, 2006, from http://www.unescap.
org/tid/projects/ecom04_s4sorieul.pdf
Unum (2004). SME’s lack awareness of disability law changes. Retrieved September 20, 2006,
from http://www.unum.co.uk/Home/Corporate_Information/Press_ Releases/2004/SMEs_
lack_awareness_of_Disability_Law_changes.
htm?UPCC=True
Upfold, C.T., and Sewry, D.A. (2005). An investigation of Information security in small and medium
enterprises (SMEs) in Eastern Cape. Retrieved
September 20, 2006, from http://icsa.cs.up.ac.za/
issa/2005/Proceedings/Research/082_Article.pdf
Vircom.com (2004). Can laws block spam? An
analysis of the effect of international spam legislation. White paper, Vircom.
Visagie, J.C. (1997). SMMEs’ challenges in reconstructing South Africa. Management Decision,
35(9), 660-667.
Washkuch, F. (2006). Compliance still a struggle
for most corporations. News, SC Magazine Australia, Retrieved November, 30, 2006, from http://
www.scmagazine.com.au/news/66100
Weick, K.E. (1995). Sense making in organizations. Thousand Oaks, CA: Sage.
Weick, K.E. (2001). Making sense of the organisation. Oxford: Blackwell Publishers.
World Bank. (2004). Doing business 2004: Understanding regulation. A co-publication of the World
Bank, the International Finance Corporation and
Oxford University Press. The International Bank
for Reconstruction and Development/World Bank.
Retrieved June 10, 2006, from http://www1.worldbank.org/publications/pdfs/ 15341frontmat.pdf
Yin, R. (1994). Case study research: Design and
methods. London: Sage Publications.
Zhang, L. (2005). The CAN-spam Act: An insufficient response to the growing spam problem.
Berkeley Technology Law Journal, 20, 301-332.
Zorz. (2003). Small firms ‘shun’ PC security, BBC
NEWS, Retrieved October 15, 2004, from http://
www.net_security.org/news:php?id=2650
111
Factors Influencing SME Compliance with Government Regulation on Use of IT
APPENDIX
Cover Letter
Dear Sir/Madam
RE: Compliance with the Electronic Communications & Transactions - ECT Act (2002)
The Electronic Communications & Transactions Act (ECT Act, 2002) is a South African law that
facilitates regular electronic communication and transactions, promotes universal access to electronic
communications and prevents the abuse of information systems. All organizations in South Africa are
expected to comply with this Act in order to avoid serious risks and legal liabilities. However, compliance with this Act appears to be difficult for many organizations due to many different factors.
We would like to investigate some of these problems, gather your views and some information about the
steps your organization has taken towards ensuring compliance. Participation in this study is voluntary
and at the end of the study useful recommendations will be made known to the participants. Please be
assured that the information provided will be handled in a confidential manner and only used for the
purpose of this research
We request that you complete and submit this short questionnaire to us before 15th September 2007. If
you have any questions about the survey or the questionnaire please contact:
Michael Kyobe
Dept. of Information Systems, University of Cape Town
P.O.Box 34240, Rhodes Gift, Cape Town, 7707.
Tel: 021 6502597 Fax: 021 650 2280
Email: Michael.kyobe@uct.ac.za
Thank you very much for your time and cooperation
Yours truly,
Michael Kyobe
112
Factors Influencing SME Compliance with Government Regulation on Use of IT
SMEs Survey
1. Section A
Please provide the following general information about your organization
1. Name of organization and establishment date
2. Email
3. Telephone
4. Business sector (e.g., Retail, manufacturing)
5. Number of employees
6. How long has your organization been trading on the Internet? (state your answer in Years)
2. Section B
Please answer the following questions about compliance with the ECT Act of 2002, (section 25) by selecting only one number representing your choice.
Not at all
Have heard about it but not certain about its requirements
1. Is your organization aware of the ECT Act (2002)?
Aware of some of its requirements and liabilities
Aware of most of its requirements and liabilities
Fully aware of all its requirements and liabilities
Not Applicable
Not at all
2. Has your organization adopted any professional standards
(e.g., ISO 17799 or SANS 15801) to guide you towards ensuring compliance?
We plan to adopt some of the guidelines
Have adopted a few of the guidelines
Have adopted most of the guidelines
Have fully adopted all the guidelines
113
Factors Influencing SME Compliance with Government Regulation on Use of IT
3. What percentage of your information technology budget is
spent on ECT Act compliance?
Not Applicable
0%
1% – 10%
11% - 30%
31% - 50%
above 50%
4. Which of the following best describes the cost to your
organization of compliance with the ECT Act? (Tick one appropriately number)
1(Very Low); 2(Low); 3(Moderate); 4(High); 5(Very High)
Not applicable
Never
5. How often do you report the steps taken towards ensuring
compliance with the Act to auditors?
Once a year
Twice a year
Four times a year
Every month
6. Do you have a specific person fully responsible for ensuring
compliance with the act in our organization?
Not Applicable
No
Yes
Not Applicable
7. Do you have a record retention and disposition policy?
No
Yes
Not Applicable
8. Do you have an ECT Act compliance policy?
No
Yes
9. Do you have a qualified information technology person assisting your organizations with ECT Act compliance?
Not Applicable
No
Yes
114
Factors Influencing SME Compliance with Government Regulation on Use of IT
Not applicable
Have not yet implemented any measures
10. Have you implemented adequate systems controls to
prevent viruses, spam and other attacks?
Have only basic measures e.g., password & anti-virus. Have no
intentions to upgrade
Have basic measures but intend to implement stronger measures
in future
Have implemented reasonably strong physical and electronic
measures
Have implemented advanced and certified control measures
Not Applicable
Never done so
11. How often do you backup and archive electronic mail &
records?
Once a year
Twice a year
Quarterly (4 times a year)
Weekly
Not Applicable
Never
12. How often do you conduct or participate in compliance
and security training programs?
Once a year
Twice a year
Quarterly (4 times a year)
Weekly
13. Do you think the penalties or liabilities imposed by the
ECT Act of 2002 (section 25) are just and fair?
Not Applicable
No
Yes
115
Factors Influencing SME Compliance with Government Regulation on Use of IT
14. If you have any additional comments please write them
here
Thank you very much for your assistance.
This work was previously published in International Journal of Global Information Management, Volume 17, Issue 2, edited
by Felix B. Tan, pp. 30-59, copyright 2009 by IGI Publishing (an imprint of IGI Global).
116
117
Chapter 6
Knowledge Management
Systems Diffusion in
Chinese Enterprises:
A Multistage Approach Using
the Technology-OrganizationEnvironment Framework
One-Ki (Daniel) Lee
University of Massachusetts Boston, USA
Mo (Winnie) Wang
City University of Hong Kong, Hong Kong
Kai H. Lim
City University of Hong Kong, Hong Kong
Zeyu (Jerry) Peng
University of Science and Technology of China, China
ABSTRACT
With the recognition of the importance of organizational knowledge management (KM), researchers
have paid increasing attention to knowledge management systems (KMS). However, since most prior
studies were conducted in the context of Western societies, we know little about KMS diffusion in other
regional contexts. Moreover, even with the increasing recognition of the influence of social factors in KM
practices, there is a dearth of studies that examine how unique social cultural factors affect KMS diffusion in specific countries. To fill in this gap, this study develops an integrated framework, with special
consideration on the influence of social cultures, to understand KMS diffusion in Chinese enterprises.
In our framework, we examine how specific technological, organizational, and social cultural factors
can influence the three-stage KMS diffusion process, that is, initiation, adoption, and routinization. This
study provides a holistic view of the KMS diffusion in Chinese enterprises with practical guidance for
successful KMS implementation.
DOI: 10.4018/978-1-60960-605-3.ch006
Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
Knowledge Management Systems Diffusion in Chinese Enterprises
INTRODUCTION
Organizational knowledge has been recognized
as the primary driver of an enterprise’s long-term
growth and its sustained competitiveness (Bock,
Zmud, Kim, & Lee, 2005; Miller & Shamsie,
1996; Wasko & Faraj, 2005). Hence, enterprises
are becoming more interested in taking up knowledge management (KM) initiatives to manage
their knowledge assets (Kankanhalli, Tan, & Wei,
2005). A common practice is adopting knowledge
management systems (KMS) to “support and
enhance the processes of knowledge creation,
storage/retrieval, transfer, and application” (Alavi
and Leidner, 2001, p. 114). A research from KPMG
(2000) reports that a considerable number of
enterprises in Europe and the U.S. have initiated
KM with adoption of the relevant technologies.
Like their Western counterparts, more and more
Chinese enterprises recognize the importance of
KM and show great interest in KMS implementation (Burrows, Drummond, & Martinsons, 2005;
Xiao, 2005; Zhou, Hu, & Pang, 2005). However,
the overall KM maturity of Chinese enterprises
may still be in its infancy. For example, Mainland
Chinese enterprises have seldom been listed in
the Asian Most Admired Knowledge Enterprises
(Teleos, 2003, 2004, 2006, 2007; Voelpel & Han,
2005). Failing to take advantages of KM due to
technical limitations is one of the main difficulties
they face (Zhou et al., 2005). In particular, many
enterprises fail to maximize employees’ effective usage beyond the initial adoption of KMS
(Burrows et al., 2005). Therefore, many Chinese
enterprises simply postpone their decision on
KMS adoption, pondering whether their current
organizational conditions are ready for KMS
implementation. Thus, it will be both interesting and useful to examine the entire process of
KMS diffusion, from initiation to routinization,
in Chinese enterprises so as to provide practical
guidance to Chinese managers for their KMS
implementation.
118
Our literature review substantiates that most
of the existing research in the KM field has been
conducted in Western counties or highly developed
countries, such as the U.S., Western Europe, and
Japan (Voelpel & Han, 2005). Since “national
environments (e.g., economic development, cultural expectations, organizational structure) shape
the practice of KM” (Geng, Twonley, Huang, &
Zhang, 2005, p. 1031), key findings from the
context of these countries may only partially or
even inaccurately apply to the context in China
(Chow, Deng, & Ho, 2000; Geng et al., 2005;
Michailova & Hutchings, 2006; Ramasamy, Goh,
& Yeung, 2006). According to Selmer (2005), due
to the large cultural distance between China and
Western countries, China is frequently regarded
by Westerners as “the most foreign of all foreign
places” (p. 78). Moreover, Chinese enterprises are
highly embedded in the Chinese traditional cultural
context, such as Confucian cultural values, with
regard to their management practice, business
process, and organizational structures (Ramasamy
et al., 2006). As Burrows et al. (2005) pointed
out, KM practice within a Chinese enterprise
will also be strongly influenced by such cultural
values. Therefore, we aim to fill this cultural gap
of the literature in investigating organizational
KM initiative and providing practical guidance
for KMS diffusion within Chinese enterprises.
To achieve this, we set up the following research
objectives: (1) To develop an integrated framework
that enables a holistic perspective to investigate
potential factors that can affect the entire KMS
diffusion process within Chinese enterprises;
(2) To generate high-level propositions from the
framework developed so that future studies can
develop specific hypotheses from the propositions
for their empirical tests.
Drawing from Cooper and Zmud’s (1990)
technology diffusion process model and Zhu,
Kraemer, and Xu’s (2006) e-business diffusion
model, we propose a three-stage KMS diffusion
model in this study. We also adopt the TechnologyOrganization-Environment (TOE) framework
Knowledge Management Systems Diffusion in Chinese Enterprises
(Tornatzky & Fleischer, 1990) to capture factors
and theorize their effect on relevant stages of
KMS diffusion process. We organize our article
as follows: the next section provides an introduction to our theoretical base. The following section
comprises the conceptual development where
we proposes eight propositions to illustrate how
factors from the three aspects, that is, technology
aspect, organization aspect, and environment aspect, affect each of the stages of KMS diffusion
process. We then discuss both the theoretical and
practical implications of our study. Conclusions
are offered in the final section with the directions
of future research.
THEORETICAL BACKGROUND
This study draws from three streams of research:
(1) technology diffusion process, (2) technologyorganization-environment (TOE) framework, and
(3) Chinese Confucian culture, each of which is
discussed in greater detail.
Technology Diffusion Process
Innovation diffusion is recognized as a stage-based
process of “spreading a new technology within
a universe of potential adopters” (Nilakanta &
Scamell, 1990, p. 25). From this technological
diffusion perspective, Cooper and Zmud (1990)
suggest a six-stage model of diffusion process,
consisting of (1) initiation: scanning organizational problems, collecting, and evaluating the
information for IT solutions, finding the “right”
IT application for the organization, (2) adoption:
getting organizational support and resource commitment for IT implementation by negotiation, (3)
adaptation: installing the IT application, adjusting both the IT and organizational procedures to
achieve a good fit, and preparing employees to
use the IT, (4) acceptance: encouraging employees to commit to using the IT application in their
work, (5) routinization: using the IT application
to become a part of the working procedures and
employees’ habit, and (6) infusion: using the IT
application in a more integrated manner to obtain
its full potential in supporting the organization’s
work. On the other hand, Zhu, Kraemer, and Xu
(2006) suggested a simpler version of this multistage diffusion process model to develop their
e-business diffusion model, in which the above
six stages were simplified into three stages of
initiation, adoption, and routinization.
KMS diffusion, like the diffusion of other
kinds of technology, is also believed to be a
process consisting of multiple stages (Xu &
Quaddus, 2005c). For parsimony, following Zhu
et al.’s (2006) approach, we propose a threestage model of KMS diffusion. The three stages
are KMS initiation, adoption, and routinization.
First, the KMS initiation stage incorporates the
activities of information gathering, synthesizing,
and KMS evaluating. Second, the KMS adoption
stage includes decision making on KMS adoption,
allocation of the necessary resources for KMS,
KMS development, adjustment of organizational
procedures, and employee training. This stage corresponds to the adoption and adaptation stages in
the original technology diffusion process model.
Third, the KMS routinization stage, which corresponds to the acceptance, routinization, and
infusion stages in the original technology diffusion
process, involves operating KMS and inducing
employees to use the system as a part of their
work process.
Technology–Organization–
Environment (TOE) Framework
To define the factors that may affect the three stages
of KMS diffusion, we rely on the TOE framework
proposed by Tornatzky and Fleischer (1990). The
TOE framework suggests investigating the effects
of technology itself, organizational characteristics, and the external environment on adoption
and implementation of technological innovation
(Tornatzky & Fleischer, 1990). The TOE frame-
119
Knowledge Management Systems Diffusion in Chinese Enterprises
work has been thought to be a good theoretical
angle for defining IT deployment factors which
are rooted in technological, organizational, or
environment contexts and thus are widely adopted
by IS researchers (e.g., Charalambos, Benbasat,
& Albert, 1995; Chau & Tam, 1997; Cooper &
Zmud, 1990; Kuan & Chau, 2001; Premkumar
& Ramamurthy, 1995; Thong, 1999; Zhu et al.,
2003). Therefore, we believe the TOE framework
is useful in understanding KMS diffusion process.
KMS diffusion is a complex social technical
process (Alavi & Leidner, 1999; Chauvel & Despres, 2002; Davenport, De Long, & Beers, 1998;
Davenport & Prusak, 1998; Kamla & Lorne, 2005;
Xu & Quaddus, 2005a). In the literature, several
researchers have pointed out that technical factors
are not the only factors that cultivate KMS success
and that organizational and cultural factors also
make considerable contributions to KMS success
(Al-Busaidi & Olfman, 2005; Kankanhalli et al.,
2005; McDermott, 1999; Waston & Hewett, 2006).
In particular, we focus on the cultural context
for the organizational and environmental aspects.
In the literature, several researchers (Deans &
Ricks, 1991; Raman & Watson, 1994; Watson,
Kelly, Galliers, & Brancheau, 1997) have suggested that cultural differences at national level
may explain differences in IT value. They usually adopt Hofstede’s (1991) framework which
include five cultural dimensions, that is, distance,
uncertainty avoidance, individualism, masculinity,
and time orientation. This framework has been
widely adopted to explain the impacts of cultural
differences on organizational behaviors, including IT behaviors. In this study, our aim is not to
compare the diffusion process under different
cultural backgrounds, but rather to investigate the
role of specific cultural influence on KMS diffusion process in a specific cultural context (namely
China). Therefore, Hofstede’s framework, which
is useful for cross-cultural comparison, does not
meet the aim of our study. Rather, we focus on
Chinese-specific cultural factors with regard to
KMS diffusion.
120
Chinese Society and
Confucian Culture
Prior studies on Chinese culture revealed several
cultural values which are unique to Chinese society. Examples include harmony, interpersonal
relationships, reciprocal benefits, and impression
management (Cheung et al., 2001; Hwang, 1987;
Liu, Friedman, & Chi, 2005; Liu, 2003). One of
the most fundamental bases of these values is Confucianism, a Chinese philosophy which regards
appropriate human relationships as the basis of
society (Yun, 2001). Although new to IS research,
it has been widely studied in the organizational
behavior domain to explain various behavioral
phenomena in Chinese society (Jaw, Ling, Wang,
& Chang, 2007; Shin, Ishman, & Sanders, 2007;
Wong, Leung, Hung, & Ngai, 2007). Therefore,
investigating this cultural background will also be
vital to understand social interaction processes of
the KMS diffusion in Chinese enterprises.
Confucianism regulates human relationships
by the Five Code of Ethics, namely (1) affection
between father and son, (2) righteousness between
rulers and subjects, (3) distinction between husband and wife, (4) order between older and younger
brothers, and (5) faithfulness between friends
(Chen and Chung, 1994; Liu, 2003). Influenced
by these culture values, Chinese enterprises are
known for the status-based hierarchy which is
demonstrated in their organizational structure and
behaviors (Lockett, 1988; Sheh, 2001; Tjosvold,
Yu, & Liu, 2004). They are also known for the
reciprocal obligation embedded in interpersonal
relationships (Chen & Chung, 1994; Liu, 2003).
The former can be characterized as hierarchical
relationships and the latter as social reciprocal
relationships (Chen & Chung, 1994; Liu, 2003;
Yum, 1988). In particular, this study focuses on
guanxi and renqing, which are perceived as the
interlocking components of such reciprocal relationships (Yun, 2001). Since the KMS diffusion
process in Chinese enterprises involves diverse
interpersonal and interorganizational relation-
Knowledge Management Systems Diffusion in Chinese Enterprises
Figure 1. A KMS diffusion model for Chinese enterprises
ships, we believe that the cultural value of guanxi
and renqing, that is, reciprocity, will affect various
stages of the diffusion process. Moreover, since
these cultural factors influence both social and
public life in China more than in most Western
countries (Pye, 1988), examining these cultural
factors would provide us a deeper understanding
of the context in which KMS diffusion proceeds.
This also allows us to explore the specificity of the
context and to look for better methods to facilitate
the diffusion process.
our research model. Each of the variables and
proposed relationships is discussed in this section.
Conceptual Development
Organizational IT competence refers to “a firm’s
capacity for IT-based innovation” (Sambamurthy,
Bharadwaj, & Grover, 2003, p. 244). Prior studies have identified several important elements of
IT competence, which include IT infrastructure,
human IT resource (both technical and managerial
skills), and IT/business partnerships (Bharadwaj,
2000; Feeny & Wilcocks, 1998; Henderson, 1990;
Ross, Beath, & Goodhue, 1996; Sambamurthy et
Chau and Tam (1997) emphasize the importance
of tailoring frameworks borrowed to cater for the
context of a specific application. Likewise, when
we adopt the TOE framework to study KMS diffusion in Chinese enterprises, we pay attention to the
specific context of China and tailor the variables
according to China’s specificity. Figure 1 shows
Technological Aspect
We first examine the effect of the factors from
technological aspect on KMS diffusion process in
terms of organizational IT competence and KMS
characteristics.
Organizational IT Competence
121
Knowledge Management Systems Diffusion in Chinese Enterprises
al., 2003; Weill & Broadbent, 1998). These elements have been illustrated as playing important
roles in organizational processes to launch, deploy,
and manage technologies and IT-based systems
(Bharadwaj, 2000; Ross et al., 1996; Wade &
Hulland, 2004). For example, organizations with
a flexible IT infrastructure can rapidly adopt new
technologies at lower costs and risks (Bharadwaj,
2000; Premkumar & Ramamurthy, 1995). Organizations with strong human IT resources may
have an advantage in terms of integrating IT and
business resources more effectively (Bharadwaj,
2000; Piccoli & Ives, 2005). Moreover, organizations with a strong partnership between IS and
business units can smooth IT project diffusion by
reducing resistance from business units (Feeny &
Wilcocks, 1998; Ross et al., 1996). In line with
these arguments, we believe that IT competence
plays an important role during the KMS diffusion process in Chinese enterprise, especially
when considering that technical deficiency or
immaturity has been one of the biggest barriers
for Chinese enterprises to adopt new technology
resources in general (Harrison & Farn, 1990) and
to conduct KM activities in specific (Burrows et
al., 2005; Hu, 2006; Zhou et al., 2005).
The influence of IT competence, however, may
vary throughout the entire diffusion process. In the
first two stages, that is, KMS initiation and adoption stages, an organization’s IT competence may
have a direct effect on how the organization evaluates certain KMS. A high level of IT competence
with a flexible IT infrastructure, strong IT human
resources, and a strong IT-business partnership,
may allow resources to be more effectively utilized
for KMS initiation and implementation. In particular, IT competence may influence whether there
are available or sufficient resources for a KMS
implementation project and how these resources
are allocated for the project and its related activities, such as procedural adjustment and employee
training. On the other hand, since organizational
IT competence may not have a direct bearing on
employees’ personal motivation to accept and
122
continuously use the KMS, it is unlikely to have
a direct impact on the last stage of KMS diffusion process, that is, KMS routinization. Based
on these arguments, we propose that:
P1: Organizational IT competence will have
positive impacts on KMS initiation and
adoption stages.
KMS Characteristics
In the literature, the influence of the perceived
characteristics of a specific innovation on the
adoption has been widely studied (Rogers, 1995).
According to Tornatzky and Klein (1982), three
characteristics of an innovation, that is, compatibility, relative advantage, and complexity, are significantly associated with its adoption. Following
this perspective, we posit that the compatibility,
relative advantage, and complexity of a certain
KMS influence its diffusion process. Drawing
from the literature (Kuan & Chau, 2001; Rogers,
1995; Thong, 1999), we define KMS compatibility
as the extent to which KMS is consistent with the
existing values, needs, and work practices of an
enterprise. KMS relative advantage refers to the
degree to which KMS can provide benefits to the
organization (Kuan & Chau, 2001; Rogers, 1995;
Thong, 1999). Particularly, this relative advantage
refers to the potential new benefit which KMS can
bring to the organization and/or employees when
compared to the traditional knowledge management practices without such systemic support.
On the other hand, KMS complexity refers to the
extent to which KMS is difficult to use.
The first two characteristics of compatibility
and relative advantage are believed to be positively
related to KMS diffusion while the last characteristic of complexity may be negatively related to
KMS diffusion. These technology characteristics
will become the critical concerns particularly for
the post-initiation stages, that is, adoption and
routinization of a specific KMS. Furthermore,
the three characteristics are believed to be impor-
Knowledge Management Systems Diffusion in Chinese Enterprises
tant particularly in the China context for several
reasons. First, several China-based studies (e.g.,
Hu, 2006; Zhou et al., 2005) have documented
that some Chinese enterprises considered only the
benefits that the system could bring but neglected
whether or not the system was actually compatible
with their Chinese-specific business practices. As
a result, their KMS could not achieve continuous
usage by employees. On the other hand, by carefully taking this compatibility issue into consideration, other enterprises could achieve successful
adoption and continuous use of KMS. Based on
these arguments, we propose that:
P2: KMS compatibility will have positive impacts on KMS adoption and routinization
stages.
Second, KMS relative advantage is the potential benefits derived from using a specific KMS.
The benefits can be viewed from two distinctive
levels, that is, organizational level and individual
level. Moreover, the perceived benefits by organization and individual employees may not be the
same. Hence, we propose their different impacts
on KMS diffusion process. First, organizationallevel relative advantage may have a positive relationship with the adoption stage. This is mainly
because the benefits perceived at organizational
level can lead to an organization’s decision on
adopting a specific KMS and also motivate the
organization to adjust for this new IT-based initiative. On the other hand, individual-level relative
advantage may have a positive relationship with
the routinization stage. This is because individual
employees should be more motivated to use the
KMS when they recognize the usefulness of the
system to their work (Davis, 1989). Based on
these arguments, we propose that:
P3a: KMS relative advantage at the organizational
level will have positive impacts on KMS
adoption stage.
P3b: KMS relative advantage at an individual
level will have positive impacts on KMS
routinization stage.
Third, the complexity of KMS is also another
major concern for potential adopters because it
affects how much resources should be invested and
how much effort should be made for adaptation of
both the IT and operational procedures. Moreover,
since the complexity can make employees’ use of
the proposed system difficult (Davis, 1989), it will
eventually influence whether or not employees
can achieve a rountinized usage. Based on these
arguments, we propose that:
P4: KMS complexity will have negative impacts
on KMS adoption and routinization stages.
Organizational Aspect
Top Management Commitment
Incorporated in the traditional respect for status
which originates from the Confucian cultural
values, a status-based behavioral tendency exists
between top management and employees within
Chinese enterprises. Specifically, top management
behaves as the authority and employees show
obedience and respect for their superiors (Li &
Scullion, 2006; Lockett, 1988). Many prior studies
pointed out that Chinese particularly follow the
principle of respecting and obeying superiors (e.g.,
Han, Li, & Kwang-Kuo, 2006; Hofstede, 1991;
Wei & Hwang, 1998). Therefore, the opinions and
behaviors of top management may have a strong
effect throughout the entire diffusion process. In
particular, top management’s commitment toward
organizational KM and a specific KMS may
directly dictate how the organization perceives
organizational opportunities and problems with
regard to this technology-based innovation and
how the organization evaluates KMS. Moreover,
top management’s strong commitment toward an
IT project has shown to be a critical success fac-
123
Knowledge Management Systems Diffusion in Chinese Enterprises
tor of successful implementation of the project
(Jarvenpaa & Ives, 1991; Keil, Cule, Lyytinen,
& Schmidt, 1998). This will be especially true
for Chinese enterprises because top management’s opinion usually plays an important role in
allocating organizational resources and making
adjustments of business procedures. Furthermore,
top management commitment may be a significant driving force of the last stage, that is, KMS
rountinization. This is because the opinion of top
management greatly influences how the organization introduces KMS to its employees and further
employees have a high obedience and respect for
their top management. Based on these arguments,
we propose that:
P5: Top management commitment toward KM
initiative and a specific KMS will have positive impacts on KMS initiation, adoption,
and routinization stages.
Hierarchical Organizational Structure
Chinese enterprises usually have a strong and
deep hierarchical structure in management and
communication (Jiang & Liang, 2005; Sheh,
2001). Under such a hierarchical environment,
upper-level managers may restrict lower-level
employees’ access to crucial information by exerting control over the vertical and horizontal flow
of information (Ardichvili, Maurer, Li, Wentling,
& Stuedemann, 2006; Hofstede, 2001). Moreover,
individual employees tend to easily accept the
delivered decisions and the information given
from their supervisors (Jiang & Liang, 2005).
Under this situation, the common practice of
organizational knowledge transfer and sharing
in Western organizations having less hierarchical
structure may not apply to Chinese organizations.
Furthermore, as Ardichvili et al. (2006) pointed
out, higher-level managers who emphasize their
authority and image do not want to participate in
online-based communications. Under a strong and
deep hierarchical structure, employees in differ-
124
ent departments also may not realize the value of
knowledge sharing across departments. Hence,
even if a KMS has been well implemented in an
organization, the KMS may not actually facilitate
sharing of knowledge if the organization has a
strong and deep hierarchical structure. Based on
these arguments, we propose that:
P6: A strong and deep hierarchical organizational
structure will have a negative impact on
KMS routinization stage.
Social Cultural Aspect:
Environmental Aspect
Confucian ethics places high values on interpersonal harmony and views social interaction as a
reciprocally obligatory, giving-and-taking process
(Chen & Chung, 1994; Liu, 2003; Yum, 1988).
Influenced by such a cultural value, China has
been thought to be a relation-centered society, in
which guanxi and renqing dominate interpersonal
relationship (Fan, 2002; Farh, Tsui, Xin, & Cheng,
1998; Hwang, 1987; Ramasamy et al., 2006).
Guanxi can be understood as “drawing on connections to secure favors in personal relationships”
(Luo, 1997, p. 44). Guanxi emphasizes the value
of interdependence, mutuality, and reciprocity. It
has been recognized as the lifeblood of personal
and business relationships (Hutchings, 2005; Ramasamy et al., 2006; Shin et al., 2007; Su, Sirgy, &
Littlefield, 2003; Yeung & Tung, 1996). The other
concept tightly coupled with guanxi is renqing,
which is perceived as the interlocking component
and a prerequisite of guanxi (Yun, 2001). According to Hwang (1987), renqing implies both a
normative standard for regulating social exchange
and a social mechanism for individuals to obtain
desirable resources. The rule of renqing regulates
that in a mutual relationship, people who receive
help are obligated to repay (Hwang, 1987). People
who refuse to return favors are thought to be immoral and untrustworthy (Shin et al., 2007). Hence,
it is the social norm that people need to follow in
Knowledge Management Systems Diffusion in Chinese Enterprises
order to maintain harmonious guanxi with each
other (Yun, 2001). Furthermore, people can use
renqing to build and improve guanxi, which is
neatly captured by the Chinese phrase “sending
renqing.” For example, when a person has happy
occasions or meets difficulties, the other person
can use resources to give a gift or provide help
(Hwang, 1987). This promotes the indebtedness
feeling of the receiver with the need to look for
opportunities to repay (Hoivik, 2007; Zhang &
Zhang, 2006). Through such repeated sending and
repaying, people build and maintain their guanxi
in order to make more use of their connections
for the future occasions (Chan, Denton, & Tsang,
2003). Thus, skillful practitioners of guanxi may
use the way of renqing to cultivate relationships
with particular individuals or groups of people
who might provide help in the future (Yun, 2001).
Based on these understandings, we believe that
the cultural value of guanxi and renqing affect the
KMS diffusion process in Chinese enterprises.
This is because the diffusion process involves
diverse interpersonal and interorganizational
relationships.
In particular, the cultural value of guanxi and
renqing in the relationship with external vendors
may facilitate the KMS initiation and adoption
stages. This is because Chinese enterprises may
tend to initiate and evaluate system vendors based
on the closeness of the guanxi rather than on an
objective cost-benefit justification. Moreover,
the decision on the selection of a specific vendor
could also be treated as an opportunity to repay
the renqing debt to the vendor. Based on these
arguments, we propose that:
P7: A strong cultural value of “guanxi” and
“renqing” in managing external vendors will
have positive impacts on KMS initiation and
adoption stages.
On the other hand, in order to proactively
build guanxi, internal employees are likely to
intentionally provide their valuable knowledge
to colleagues who are perceived as potential
resources of future help. Hence, they prefer a
face-to-face knowledge sharing to cultivate a
direct indebtedness feeling of the receiver. Since
knowledge can be viewed as a significant resource
in building interpersonal relationships, employees
may have a tendency to keep it privately and not
to publicize their knowledge through KMS. Based
on these arguments, we propose that:
P8: A strong cultural value of “guanxi” and
“renqing” among internal employees will
have a negative impact on KMS routinization stage.
DISCUSSION AND IMPLICATIONS
This study proposes an integrated framework to
examine how the technology, organization, and
social-cultural (environment) factors affect KMS
diffusion in the context of Chinese enterprises. In
particular, the study highlights the roles of cultural specificities of Chinese enterprises, such as
Confucianism, guanxi, and renqing, in explaining
the impacts of those factors on each of the stages
of KMS diffusion process. With regard to their
impacts, there should be a variation among the
sub-cultural groups within China, such as Center
China, Northern China, Southern China, Hong
Kong, and Taiwan. Moreover, one could recognize
the cultural difference among the different types
of Chinese enterprise, such as government-owned,
private, foreign-invested, and public enterprises.
Regardless of such variances among the subcultural groups, however, the nation-wide cultural
specificities play an important role in organizational behavior (Hofstede, 1991). Thus, it will
be both interesting and important to understand
how these cultural specificities define specific
behavioral pattern during KMS diffusion process.
The study has several significant implications.
First, the study suggests a holistic view of KMS
diffusion process stages and the factors that may
125
Knowledge Management Systems Diffusion in Chinese Enterprises
affect the stages involved in the process. Thus
the study enriches practitioners with a comprehensive understanding of critical areas of concern
in implementing KMS in Chinese enterprises.
Through this study, we have conceptualized KMS
diffusion process as a parsimonious three-stage
model. We also adopt the TOE framework to
frame the factors that influence the three stages
of KMS diffusion within Chinese enterprises.
The proposed framework will benefit Chinese
enterprises who are planning to initiate a KMS and
who are confronting difficulties in certain stages of
their KMS diffusion. In particular, the framework
highlights areas for management attention in each
of the stages of the KMS diffusion process. By
providing a deeper understanding of their potential
customers, our research framework can also help
foreign system vendors, who would like to take
an edge over their competitors in China market.
Second, in our conceptualization, we place emphasis on the influence of social culture factors,
that is, the cultural value of guanxi and renqing.
Our approach is consistent with the increasing
recognition that the success of KMS can be determined by social issues to a large extent (Kamla
& Lorne, 2005; Kankanhalli et al., 2005; Ruppel
& Harrington, 2001; Scott, 1998; Voelpel & Han,
2005). This also implies that practitioners need
to take into consideration the specific context in
which a KMS is embedded when implementing
the KMS. Third, this study will serve as a base
for further empirical study with conceptual extension and testable hypotheses. Fourth, our research
framework, which adopts TOE framework, can
be extended by exploring the potential influence
of other additional factors. For example, another
salient Chinese cultural value that has been recognized as exerting significant influence on human behaviors in Chinese society is face concern
(Chow et al., 2000; Hwang, Francesco, & Kessler,
2003; Kim & Nam, 1998). Taking additional factors into consideration will make the proposed
framework more comprehensive and with higher
prediction power.
126
CONCLUSION
Future research can extend our conceptualization
of KMS diffusion. In particular, a multicase study
might be appropriate for this type of study. Case
study has been known to be especially useful when
the phenomenon under investigation is contemporary and when the study addresses the “how” and
“why” questions (Benbasat, Goldstein, & Mead,
1987; Zhang, Lee, Huang, Zhang, & Huang, 2005).
Moreover, case study has been used successfully
by previous IS implementation research (e.g., Xu
& Quaddus, 2005b; Zhang et al., 2005). Since the
unit of analysis of this study is KMS diffusion
project, future research can select enterprises
which are involved in certain stages of the KMS
diffusion. Data can be gathered through observations and interviews. Active involvement in a
company’s decision making and manipulation on
certain factors will be another possible approach,
that is, action research (Myers, 1997). In addition
to such qualitative methods, the framework can
be validated using quantitative methods, such as
a survey and statistical analyses.
In conclusion, even with the increasing recognition of the influence of social cultural factors
in KM practices, there is a dearth of studies that
specifically examine how unique social cultural
factors affect KMS diffusion in specific countries.
This research is an attempt to capture the social
cultural factors, in combination with other factors,
to explore the intricate influence on the KMS
diffusion process. We provide a starting point for
future researchers to investigate the phenomenon
further based on the logic and knowledge that this
framework provides.
ACKNOWLEDGMENT
The work described in this article was╯supported
by a grant from the Research Grants Council of
the Hong Kong Special Administrative Region,
China (Project No. CityU 1497/06H)
Knowledge Management Systems Diffusion in Chinese Enterprises
REFERENCES
Al-Busaidi, K. A., & Olfman, L. (2005). An
investigation of the determinants of knowledge
management systems success in Omani organizations. Journal of Global Information Technology
Management, 8(3), 6–27.
Alavi, M., & Leidner, D. E. (1999). Knowledge
management systems: Issues, challenges and
benefits. Communications of AIS, 1(7), 2–37.
Alavi, M., & Leidner, D. E. (2001). Review:
Knowledge management and knowledge management systems: Conceptual foundations and
research Issues. Management Information Systems
Quarterly, 25(1), 107–136. doi:10.2307/3250961
Ardichvili, A., Maurer, M., Li, W., Wentling, T.,
& Stuedemann, R. (2006). Cultural influences on
knowledge sharing through online communities
of practice. Journal of Knowledge Management,
10(1), 94–107. doi:10.1108/13673270610650139
Benbasat, I., Goldstein, D. K., & Mead, M. (1987).
The case research strategy in studies of information systems. Management Information Systems
Quarterly, 11(3), 369–386. doi:10.2307/248684
Bharadwaj, A. S. (2000). A resource-based perspective on information technology capacity and
firm performance: An empirical investigation.
Management Information Systems Quarterly,
24(1), 169–196. doi:10.2307/3250983
Bock, G.-W., Zmud, R. W., Kim, Y.-G., & Lee,
J.-N. (2005). Behavior intention formation in
knowledge sharing: Examining the roles of extrinsic motivators, social-psychological forces, and
organizational climate. Management Information
Systems Quarterly, 29(1), 87–111.
Burrows, G. R., Drummond, D. L., & Martinsons,
M. G. (2005). Knowledge management in China.
Communications of the ACM, 48(4), 73–76.
doi:10.1145/1053291.1053322
Chan, A. K. K., Denton, L. T., & Tsang, A. S.
L. (2003). The art of gift giving in China. Business Horizons, 46(4), 47. doi:10.1016/S00076813(03)00051-X
Charalambos, L. I., Benbasat, I., & Albert, S. D.
(1995). Electronic data interchange and small
organizations: Adoption and impact of technology. Management Information Systems Quarterly,
19(4), 465–485. doi:10.2307/249629
Chau, P. Y. K., & Tam, K. Y. (1997). Factors affecting the adoption of open systems: An exploratory
study. Management Information Systems Quarterly, 21(1), 1–24. doi:10.2307/249740
Chauvel, D., & Despres, C. (2002). A review
of survey research in knowledge management:1997-2001. Journal of Knowledge Management, 6(3), 207. doi:10.1108/13673270210434322
Chen, G.-M., & Chung, J. (1994). The impact
of Confucianism on organizational communication. Communication Quarterly, 42(2), 93–105.
doi:10.1080/01463379409369919
Cheung, F. M., Leung, K., Zhang, J.-X., Sun,
H.-F., Gan, Y.-Q., & Song, W.-Z. (2001). Indigenous Chinese personality constructs: Is
the five-factor model complete? Journal of
Cross-Cultural Psychology, 32(4), 407–434.
doi:10.1177/0022022101032004003
Chow, C. W., Deng, F. J., & Ho, J. L. (2000). The
openness of knowledge sharing within organizations: A comparative study in the United States
and the People’s Republic of China. Journal of
Management Accounting Research, 12, 65–95.
doi:10.2308/jmar.2000.12.1.65
Cooper, R. B., & Zmud, R. W. (1990). Information
technology implementation research: A technology diffusion approach. Management Science,
36(2), 123–139. doi:10.1287/mnsc.36.2.123
127
Knowledge Management Systems Diffusion in Chinese Enterprises
Davenport, T. H., De Long, D. W., & Beers, M. C.
(1998). Successful knowledge management projects. Sloan Management Review, 39(2), 43–57.
Davenport, T. H., & Prusak, L. (1998). Working
knowledge: How organizations manage what they
know. Boston, MA: Harvard Business School
Press.
Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. Management Information Systems
Quarterly, 13(3), 319–340. doi:10.2307/249008
Deans, P. C., & Ricks, D. A. (1991). MIS research:
A model for incorporating the international dimension. The Journal of High Technology Management Research, 2(1), 57–81. doi:10.1016/10478310(91)90014-F
Fan, Y. (2002). Questioning guanxi: Definition,
classification and implications. International
Business Review, 11, 543–561. doi:10.1016/
S0969-5931(02)00036-7
Farh, J.-L., Tsui, A. S., Xin, K., & Cheng, B.-S.
(1998). The influence of relational demography
and guanxi: The Chinese case. Organization Science, 9(4), 471–488. doi:10.1287/orsc.9.4.471
Harrison, W. L., & Farn, C.-K. (1990). A comparison of information management issues in the
United States of America and the Republic of
China. Information & Management, 18, 177–188.
doi:10.1016/0378-7206(90)90038-J
Henderson, J. C. (1990). Plugging into strategic
partnerships: The critical IS connection. Sloan
Management Review, 31(3), 7–18.
Hofstede, G. (1991). Cultures and organizations:
Software of the mind. New York: McGraw-Hill.
Hofstede, G. (2001). Culture’s consequences:
Comparing values, behaviors, institutions and
organizations across nations (2nd ed.). Thousand
Oaks, CA: Sage Publications.
Hoivik, H. W. (2007). East meets West: Tacit
messages about business ethics in stories told by
Chinese manager. Journal of Business Ethics,
74(4), 457–469. doi:10.1007/s10551-007-9508-z
Hu, X. (2006). The current situation and existing
problems in Chinese enterprises’ knowledge management implementation. Technology Economic
Market, 11, 202.
Feeny, D. F., & Wilcocks, L. P. (1998). Core IS
capabilities for exploiting information technology.
Sloan Management Review, 39(3), 33–50.
Hutchings, K. (2005). Examining the impacts of
institutional change on knowledge sharing and
management learning in the People’s Republic
of China. Thunderbird International Business
Review, 47(4), 447–468. doi:10.1002/tie.20062
Geng, Q., Twonley, C., Huang, K., & Zhang, J.
(2005). Comparative knowledge management: A
pilot study of Chinese and American universities.
Journal of the American Society for Information
Science and Technology, 56(10), 1031–1044.
doi:10.1002/asi.20194
Hwang, A., Francesco, A. M., & Kessler, E.
(2003). The relationship between individualismcollectivism, face, and feedback and learning
process in Hong Kong, Singapore, and the United
States. Journal of Cross-Cultural Psychology,
34(1), 72–91. doi:10.1177/0022022102239156
Han, K.-H., Li, M.-C., & Kwang-Kuo, H. (2006).
Cognitive response to favor requests from different social targets in a Confucian society. Journal
of Social and Personal Relationships, 22(2),
283–294. doi:10.1177/0265407505050952
Hwang, K.-k. (1987). Face and favor: The Chinese
power game. American Journal of Sociology,
92(4), 944–974. doi:10.1086/228588
128
Knowledge Management Systems Diffusion in Chinese Enterprises
Jarvenpaa, S. L., & Ives, B. (1991). Executive
involvement and participation in the management of information technology. Management
Information Systems Quarterly, 15(2), 205–227.
doi:10.2307/249382
Jaw, B.-S., Ling, Y.-H., Wang, C. Y.-P., & Chang,
W.-C. (2007). The impact of culture on Chinese
employees’ work values. Personnel Review, 36(5),
763. doi:10.1108/00483480710774034
Jiang, Y., & Liang, L. (2005). The study on the
relationship between Chinese traditional culture
and knowledge management. Science Research
Management, 26(3), 116–120.
Kamla, A. A.-B., & Lorne, O. (2005). An investigation of the determinants of knowledge
management systems success in Omani organizations. Journal of Global Information Technology
Management, 8(3), 8–27.
Kankanhalli, A., Tan, B. C. Y., & Wei, K.-K.
(2005). Contributing knowledge to electronic
knowledge repositories: An empirical investigation. Management Information Systems Quarterly,
29(1), 113–143.
Keil, M., Cule, P. E., Lyytinen, K., & Schmidt, R.
C. (1998). A framework for identifying software
project risks. Communications of the ACM, 41(11),
76–83. doi:10.1145/287831.287843
Kim, J. Y., & Nam, S. H. (1998). The concept and
dynamics of face: Implications for organizational
behavior in Asia. Organization Science, 9(4),
522–534. doi:10.1287/orsc.9.4.522
KPMG. (2000). Knowledge management research
report 2000. KPMG Consulting.
Kuan, K. K. Y., & Chau, P. Y. K. (2001). A
perception-based model for EDI adoption in
small businesses using a technology-organization-environment framework. Information
& Management, 38(8), 507–521. doi:10.1016/
S0378-7206(01)00073-8
Li, S., & Scullion, H. (2006). Bridging the distance: Managing cross-border knowledge holders.
Asia Pacific Journal of Management, 23, 71–92.
doi:10.1007/s10490-006-6116-x
Liu, L. A., Friedman, R. A., & Chi, S.-C. (2005).
Ren Qing vs. big five: The role of culturally sensitive measures of individual difference in distributive negotiations. Management and Organization
Review, 1(2), 225–247. doi:10.1111/j.17408784.2005.00010.x
Liu, S. (2003). Cultures within culture: Unity
and diversity of two generations of employees in
state-owned enterprises. Human Relations, 56(4),
387–417. doi:10.1177/0018726703056004001
Lockett, M. (1988). Culture and the problems of
Chinese management. Organization Studies, 9(4),
475–496. doi:10.1177/017084068800900402
Luo, Y. (1997). Guanxi: Principles, philosophies,
and implications. Human Systems Management,
16(1), 43–51.
McDermott, R. (1999). Why information technology inspired but cannot deliver knowledge
management. California Management Review,
41(4), 103–117.
Michailova, S., & Hutchings, K. (2006). National
cultural influences on knowledge sharing: A comparison of China and Russia. Journal of Management Studies, 43(3), 384–405. doi:10.1111/j.14676486.2006.00595.x
Miller, D. T., & Shamsie, J. (1996). The resourcebased view of the firm in two environments:
The Hollywood film studio from 1936 to 1965.
Academy of Management Journal, 39(3), 519–543.
doi:10.2307/256654
Myers, M. D. (1997). Qualitative research in
information systems. Retrieved August 1, 2008,
from http://www.qual.auckland.ac.nz/
129
Knowledge Management Systems Diffusion in Chinese Enterprises
Nilakanta, S., & Scamell, R. W. (1990). The effect of information sources and communication
channels on the diffusion of innovation in a data
base development environment. Management
Science, 36(1), 24–40. doi:10.1287/mnsc.36.1.24
Sambamurthy, V., Bharadwaj, A., & Grover,
V. (2003). Shaping agility through digital options: reconceptualizing the role of information
technology in contemporary firms. Management
Information Systems Quarterly, 27(2), 237–263.
Piccoli, G., & Ives, B. (2005). Review: ITdependent strategic initiatives and sustained
competitive advantage: A review and synthesis of
the literature. Management Information Systems
Quarterly, 29(4), 747–776.
Scott, J. E. (1998). Organizational knowledge
and Internet. Decision Support Systems, 23(4),
347–357.
Premkumar, G., & Ramamurthy, K. (1995). The
role of interorganizational and organizational
factors on the decision mode for adoption of
interorganizational systems. Decision Sciences,
26(3), 303–336. doi:10.1111/j.1540-5915.1995.
tb01431.x
Pye, L. W. (1988). The Mandarin and the Cadre:
China’s political cultures. Ann Arbor, MI: The
University of Michigan Press.
Raman, K. S., & Watson, R. T. (1994). National
culture, IS, and organizational implications. In
Deans, P. C., & Karwan, K. R. (Eds.), Global
information systems and technology: Focus on
the organization and its functional areas (pp.
493–513). Harrisburg, PA: Idea Group.
Ramasamy, B., Goh, K. W., & Yeung, M. C. H.
(2006). Is guanxi a bridge to knowledge transfer.
Journal of Business Research, 59(1), 130–139.
doi:10.1016/j.jbusres.2005.04.001
Rogers, E. M. (1995). Diffusion of innovations
(4th ed.). New York: Free Press.
Ross, J. W., Beath, C. M., & Goodhue, D. L. (1996).
Develop long-term competitiveness through IT
assets. Sloan Management Review, 31–45.
Ruppel, C. P., & Harrington, S. J. (2001). Sharing
knowledge through intranets: A study of organizational culture and intranet implementation. IEEE
Transactions on Professional Communication,
44(1), 37–52. doi:10.1109/47.911131
130
Selmer, J. (2005). Western expatriates in China:
Adjusties to the most foreign of all foreign places.
In Scullion, H., & Linehan, M. (Eds.), International firms: A critical text. London: Palgrave
Macmillan.
Sheh, S. (2001). Chinese cultural values and their
implications to Chinese management. Singapore
Management Review, 23(2), 309–342.
Shin, S. K., Ishman, M., & Sanders, G. L. (2007).
An empirical investigation of socio-cultural factors of information sharing in China. Information
& Management, 44(2), 165–174. doi:10.1016/j.
im.2006.11.004
Su, C., Sirgy, M. J., & Littlefield, J. E. (2003). Is
guanxi orientation bad, ethically speaking? A study
of Chinese enterprises. Journal of Business Ethics, 44, 303–312. doi:10.1023/A:1023696619286
Teleos. (2003). 2003 Asian most admired knowledge enterprises study: Executive summary.
Retrieved August 1, 2008, from http://unpan1.
un.org/intradoc/groups/public/documents/APCITY/UNPAN020331.pdf
Teleos. (2004). 2004 Asian MAKE finalists named.
Retrieved August 1, 2008, from http://findarticles.
com/p/articles/mi_m0EIN/is_2004_Oct_13/
ai_n6233691
Teleos. (2006). 2006 Asian most admired knowledge enterprises (MAKE) winners. Retrieved August 1, 2008, from http://knowledgemanagement.
wordpress.com/2007/01/05/2006-make-winners/
Knowledge Management Systems Diffusion in Chinese Enterprises
Teleos. (2007). 2007 Asian most admired
knowledge enterprises study: Executive summary. Teleos. Retrieved August 1, 2008, from
http://www.knowledgebusiness.com/knowledgebusiness/templates/TextAndLinksList.
aspx?siteId=1&menuItemId=133
Waston, S., & Hewett, K. (2006). A multi-theoretical model of knowledge transfer in organizations: Determinants of knowledge contribution
and knowledge reuse. Journal of Management
Studies, 43(2), 142–173.
Thong, J. Y. L. (1999). An integrated model of
information systems adoption in small businesses.
Journal of Management Information Systems,
15(4), 187–214.
Watson, R. T., Kelly, G. G., Galliers, R. D., &
Brancheau, J. C. (1997). Key issues in information
systems management: An international perspective. Journal of Management Information Systems,
13(4), 91–115.
Tjosvold, D., Yu, Z.-y., & Liu, H. (2004). Traditional values for applying abilities and leader
effectiveness in China. Leadership and Organization Development Journal, 25(3/4), 318–331.
doi:10.1108/01437730410538653
Wei, C. F., & Hwang, K. K. (1998). Positive duties
and negative duties: A cross-cultural comparison of moral judgment between Taiwanese and
American College Students. Chinese Journal of
Psychology, 40, 137–153.
Tornatzky, L. G., & Fleischer, M. (1990). The
processes of technological innovation. Lexington,
MA: Lexington Books.
Weill, P., & Broadbent, M. (1998). Leveraging the
new infrastructure: How market leaders capitalize on information technology. Cambridge, MA:
Harvard Business School Press.
Tornatzky, L. G., & Klein, L. (1982). Innovation
characteristics and innovation-implementation: A
meta-analysis of findings. IEEE Transactions on
Engineering Management, 29(11), 28–45.
Voelpel, S. C., & Han, Z. (2005). Managing
knowledge sharing in China: The case of siemens
ShareNet. Journal of Knowledge Management,
9(3), 51–63. doi:10.1108/13673270510602764
Wade, M., & Hulland, J. (2004). Review: The
resource-based view and information systems
research: Review, extension, and suggestions for
future research. Management Information Systems
Quarterly, 23(1), 107–142.
Wasko, M. M., & Faraj, S. (2005). Why should I
share? Examining social capital and knowledge
contribution in electronic networks of practice.
Management Information Systems Quarterly,
29(1), 35–57.
Wong, Y. H., Leung, T. K. P., Hung, H., & Ngai,
E. W. T. (2007). A model of guanxi development:
flexibility, commitment and capital exchange. Total Quality Management & Business Excellence,
18(8), 875. doi:10.1080/14783360701350607
Xiao, X.-f. (2005). Knowledge management in
China under the knowledge economic. Technology
and Innovation Management, 26(4).
Xu, J., & Quaddus, M. (2005a). From rhetoric towards a model of practical knowledge management
systems. Journal of Management Development,
24(4), 291–319. doi:10.1108/02621710510591325
Xu, J., & Quaddus, M. (2005b). Exploring the
perceptions of knowledge management systems.
Journal of Management Development, 24(4),
374–389. doi:10.1108/02621710510591361
Xu, J., & Quaddus, M. (2005c). A sixstage model for the effective diffusion of
knowledge management systems. Journal of
Management Development, 24(4), 362–373.
doi:10.1108/02621710510591352
131
Knowledge Management Systems Diffusion in Chinese Enterprises
Yeung, I. Y. M., & Tung, R. L. (1996). Achieving
business success in Confucian societies: The importance of guanxi (connections). Organizational
Dynamics, 25(2), 54–65. doi:10.1016/S00902616(96)90025-X
Zhang, Z., Lee, M. K. O., Huang, P., Zhang, L., &
Huang, X. (2005). A framework of ERP systems
implementation success in China: An empirical
study. International Journal of Production Economics, 98, 56–80. doi:10.1016/j.ijpe.2004.09.004
Yum, J. O. (1988). The impact of Confucianism
on interpersonal relationships and communication
patterns in EastAsia. Communication Monographs,
55, 374–388. doi:10.1080/03637758809376178
Zhou, Y., Hu, H.-h., & Pang, A.-c. (2005). Investigation and analysis of current states of the knowledge management activities in Chinese enterprises.
Science Research Management, 26(4), 80.
Yun, G. (2001). Personal relations in Chinese
enterprise reforms: The interaction between formal-impersonal structure and informal-personal
power in the process of transforming state-owned
enterprise property rights. Unpublished doctoral
dissertation, Temple University.
Zhu, K., Kraemer, K. L., & Xu, S. (2003). Electronic business adoption by European firms: A
cross-country assessment of the facilitators and
inhibitors. European Journal of Information
Systems, 12(4), 251–268. doi:10.1057/palgrave.
ejis.3000475
Zhang, Y., & Zhang, Z. (2006). Guanxi and organizational dynamics in China: A link between
individual and organizational levels. Journal
of Business Ethics, 67, 375–392. doi:10.1007/
s10551-006-9031-7
Zhu, K., Kraemer, K. L., & Xu, S. (2006). The
process of innovation assimilation by firms in
different countries: A technology diffusion perspective on e-business. Management Science,
52(10), 1557–1576. doi:10.1287/mnsc.1050.0487
This work was previously published in International Journal of Global Information Management, Volume 17, Issue 1, edited
by Felix B. Tan, pp. 70-84, copyright 2009 by IGI Publishing (an imprint of IGI Global).
132
133
Chapter 7
Factors Affecting Information
Communication Technologies
Usage and Satisfaction:
Perspective from Instant
Messaging in Kuwait
Kamel Rouibah
Kuwait University, Kuwait
Hosni Hamdy
Kuwait University, Kuwait
ABSTRACT
Instant messaging (IM) technology has received extensive focus in the West while there is lack of knowledge
of it in the Arab world. This study aims to shed light on factors affecting IM usage and user satisfaction
in an Arab country (Kuwait). To achieve this objective, this study develops a theoretical model that is
based on three well-known models. This model includes curiosity (from the theory of flow), compatibility (from the innovation diffusion theory), perceived usefulness and perceived ease of use (from the
technology acceptance model), and individual characteristics in the form of prior similar experience.
The proposed model was tested using survey data from 609 students, with the results lending support
for the proposed model. Importantly, results highlight the impact of social effect on curiosity as a new
mediator of technology adoption and satisfaction. This study contributes to the literature on technology
adoption in the Arab word and aids educational institutions and companies to understand the social and
technical nature of users’ attitudes with regard to ICT adoption and satisfaction.
DOI: 10.4018/978-1-60960-605-3.ch007
Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
Factors Affecting Information Communication Technologies Usage and Satisfaction
INTRODUCTION
The 21st century has witnessed a high growth in
the use of the Internet in the Middle East. This
growth has been substantial in the Gulf Cooperation Council countries and will likely continue to
increase. For example, the Internet penetration rate
has increased between 2000 and 2005 by 300% in
Kuwait, 443% in Qatar, 257% in Bahrain, 201%
in Oman, 207% in Saudi Arabia, and 310% in the
UAE (Internet World Stats 2008).
Because of this trend, Li, Chau, and Lou
(2005) call for studies to examine the value of
several information and communication technologies (ICT) available for use. It is important
to note, however, that although the Arab world
has increased its Internet usage and connectivity
(Loch, Straub, & Kamel, 2003); little is known
on how these ICTs are used and the motivation
behind their acceptance.
Among these ICTs, IM technology is one of
the most widespread in the world.
We define IM as the ability to see if a chosen
friend, co-worker, or associate is connected to the
Internet and, if they are, to exchange real time
messages with them. Two important components
define IM capabilities: synchronicity and presence
awareness. Synchronicity relates to the transaction
speed of the technology while presence awareness
lets someone knows whether his correspondent
is online or not. IM is used instead of e-mail if
an immediate response is desired, such as when
clarifying questions need to be answered before
the conversation can proceed or when a matter
must be settled quickly.
Related IM Past Studies
and Their Limits
With the widespread use of IM by individuals,
a recent focus in the MIS field started to focus
on why and how IM is used as well as what the
consequences of its usage are.
134
Analysis of previous IM studies reveals three
observations (see literature review in Rouibah,
2008, in appendix 1). First, since 2002, the number
of studies on IM has increased. Second, despite
an increased focus on IM, there has only been
three studies (Chen, Yen, & Huang, 2004; Li et
al., 2005; Rouibah, 2008) about IM in the leading,
international IS journals (e.g., MIS Quarterly, and
Journal of Global Information Management).
Third, while past studies have focused on IM,
both in the workplace and outside the workplace,
there has been a lack of studies that focused solely
on social usage in developing countries. With the
increasing role of social influences on IM use, Shen
& Gallivan (2006) suggested conducting further
research in this perspective. Fourth, this study
also notes that few studies focus on IM satisfaction as compared to current and intention-to-use
IM. Fifth, while few studies have examined IM
adoption in the workplace (Cameron & Webster,
2005), little is known about how IM is actually
used outside the workplace and for social usage
in an international context (except Rouibah 2008;
Simon, 2006). There is also a lack of research
that addresses IM consequence as observed by
Shen and Gallivan (2006). Sixth, there has been
an increase in research dedicated to IM, but these
studies were not based on well known technology acceptance theories (e.g., Nicholson 2002),
except De Vos, Hofte, and de Poot (2004), Lin,
Chan, and Jin (2004), Wang, Hsu, and Fang (2005)
Li et al. (2005), and Rouibah (2008). However,
these studies did not investigate the relationship
between IM usage and users’ satisfaction.
Moreover, with regard to IM usage in the Arab
world, only two studies focused on IM (Rouibah
2008; Rouibah & Ould-Ali, 2005). Rouibah and
Ould-Ali, (2005) described the factors that lead
211 Kuwaiti students to use IM and what were the
consequences of their IM usage. Results of their
study revealed that respondents use IM with people
of the same gender (64%) and opposite gender
(46%). As for motives of using IM, results revealed
two main motives: for curiosity satisfaction (81%)
Factors Affecting Information Communication Technologies Usage and Satisfaction
and for friendship (39%). As for impact of IM,
results revealed 90% have succeeded in creating
a wide social network, while 65% admitted that
IM has a negative impact since they lose a lot of
time, and 35% realized that it decreased their study
performance. The study also revealed a variety
of discussed topics such as social, entertainment,
cultural, political, love, study, family, scientific,
religion, medical, personal, unethical, and sports.
With regard to the overall appreciation of IM,
statistics revealed that 50% perceived it as good
and 50% as bad technology.
Rouibah (2008) studied motives for IM usage by 191 employees for social use outside the
workplace in Kuwait. Results indicate that social
norm, ease of use, and enjoyment affect the level
of IM usage by Kuwaiti employees.
Background on IM Acceptance
IM acceptance is the potential user’s predisposition toward personally using a specific system. In
order to adequately evaluate acceptance, multiple
constructs should be utilized. The constructs of
usage and user satisfaction are commonly studied
dimensions (DeLone & McLean, 1992). These two
constructs are utilized to form the dimensions of
IM acceptance in this study.
There are several technology adoption models,
such as the technology acceptance model (TAM).
Past studies have shown the limits of TAM when
applied outside the workplace and across cultures.
In particular the literature review done by Legris
et al., (2003) revealed the need to include other
components in order to gain a broader view and
a better explanation of IT/ICT adoption. Other
studies on intrinsic motivation (Anandarajan,
Igbaria, & Anakwe, 2002; Li et al., 2005) have
shown that adoption of information systems that
provide self-fulfillment, rather than being instrumental, are strongly connected to home and leisure
activities. Therefore, a new theory seems desirable
and timely to explain IM adoption.
Specifically, there is a need to include emotion and motivation in ICT adoption. Although
these affects are touched upon in some studies,
the affective aspects are incorporated in some
exceptional studies such as computer playfulness (Webster & Martocchio, 1992), flow (Ghani
& Deshpande, 1994), and cognitive absorption
(Agarwal & Karahanna, 2000). Even when some
affective constructs are studied, other important
constructs escape scrutiny, and researchers fail to
agree on definitions.
In order to advance the knowledge about ICT
adoption, this study includes curiosity as an effective/emotion variable besides IM.
THE ROLE OF CURIOSITY
IN IM ACCEPTANCE
Curiosity has its origin in the psychology theory of
flow (Csikszentmihalyi, 1990). flow offers a basis
to understand the important structural features of
intrinsically motivating activities and, therefore,
explains individual attitudes and behaviors towards a target IT/ICT. Flow is the state in which
people are so involved in an activity that nothing
else seems to matter (Csikszentmihalyi, 1990;
Ghani & Deshpande, 1994). Flow, according to
Agarwal & Karahanna, (2000), includes intense
concentration, a sense of being in control, and a
transformation of time. When people are in flow,
they shift into a common mode of experience when
they become absorbed in their activity.
In the last few years, flow has been studied
in the context of computer mediated environments and IT including the curiosity dimension
(Agarwal & Karahanna, 2000; Webster, Trevino,
& Ryan, 1993).
Malone (1981) reviewed a number of intrinsic
motivation theories and suggested that challenge,
fantasy, and curiosity are components of intrinsically motivating computer-based instruction.
However, he saw curiosity as the most important
feature of intrinsic motivation and stated that one
135
Factors Affecting Information Communication Technologies Usage and Satisfaction
of the most important features of an intrinsically
motivating environment is the degree to which
individuals can continue to arouse and then satisfy
their curiosity. Examples may include interacting
with the opposite gender or discovering other
cultures most individuals ignore.
Lieberman (1977) sees playfulness as a multifaceted construct encompassing five distinct
factors: cognitive spontaneity, social spontaneity,
physical spontaneity, manifest joy, and sense of
humor. He also states that the overt manifestations of cognitive spontaneity are curiosity and
inventiveness.
Berlyne (1965) goes further and claims that
the principal factor producing curiosity is what he
calls conceptual conflict. By this, he means conflict
between incompatible attitudes or ideas evoked
by a stimulus situation which may explain the
opposition between attitude in the real life world
and the other in a virtual world (e.g., the ability
to flirt online that is not possible in face-to-face
high context societies).
Since the flow construct has been operationalized, tested, and applied in different ways,
additional research is needed to shed light on its
meaning to further our knowledge about its positive influence on human computer interaction, as
suggested by previous researchers in playfulness
(Malone, 1981; Webster & Martocchio, 1992).
The next section explains the main characteristics of the Arab culture and argues why ICT is
highly affected by the national culture.
IM, ICT AND THE
CHARACTERISTICS OF
ARAB CULTURE
The Arab world includes 22 countries which share
similar culture, values, language, history and geographic location. It spans from the west coast of
Africa through the northern part of Africa to the
Arabian Gulf, and from Sudan to the Middle East
Gulf Cooperation Council states which include
136
Kuwait, Bahrain, Oman, Qatar, the United Arab
Emirates, and Saudi Arabia (see appendix 2).
While the Arab world is large in terms of
population (335,827,330), very few studies have
focused on IT/ICT in the Arab world compared
to those in North America and Europe and none
of them focused on IM (Rouibah, 2008). Rose
& Straub (1998) tested and validated TAM application in five Middle East countries and called
for further applications of TAM in other Arab
countries, because it was their view that there
are many cultural differences between Arab and
non Arab countries. Loch et al. (2003) studied the
role of social norms in the diffusion of Internet
in Egypt. Al-Gahtani (2004) explored the impact
of individual, technological, and organizational
factors on PC usage and satisfaction.
These studies made systematic efforts to ascertain that culture affects IT/ICT transfer to the
Arab world. This is probably why researchers
propose integrating culture in IS research (Myers
& Tan, 2002) as well as studying the compatibility
of the target technology with the national culture.
Loch et al., (2003) focused on factors affecting e-mail acceptance in Egypt. They indicated
that culture indirectly affects technology usage.
As with the aforementioned studies, this study
focuses in particular on the IM acceptance in a
high context society.
Kuwait is part of the Arab world, whose culture
has several characteristics.
First, Arab culture is highly social and family
oriented. It is high context where Arab values put
more emphasis on the group than on the individual.
the social values and patterns of relationship all
center on the primary group (family, tribes, and
friends) toward whom primary loyalty and responsibilities of the individual are essentially directed.
Second, Arab language emphasizes indirectness, politeness and ambiguity in communication.
At the same time, love discussions are not viewed
favorable.
Third, people in the Arab world like maintain
high social interactions and personalization of
Factors Affecting Information Communication Technologies Usage and Satisfaction
relationships. In line with this idea, Al-Shareef
(1995) found that Saudi female office workers
in an educational institution commonly make
frequent personal phone calls at work and visit
socially in colleagues’ offices.
Fourth, Arab culture is a high masculinity
culture that is directly influenced by Islam, which
significantly impacts Arab society as a whole. The
Muslim is guided by his Islamic beliefs to maintain
high ethical standards during his life. All the acts
that non-married couple may do, such as being
alone in secluded areas, loving, touching, kissing, and so forth, are prohibited and condemned
in Islam. In their lives, Arab people show a high
commitment to the Islamic work ethic and comply with its principles. Arab culture is masculine,
where men dominate in most settings, clear gender
roles are the norm, and social interactions with
the opposite gender are not tolerated.
In summary, Arab culture puts an emphasis
on social relationships, is based on oral communication, and has its priority on face-to-face
interactions. These characteristics may explain (in
a certain context) why Arabs like to use IM. With
the emergence of new ICT, Arab people found
new channels to disguise their real identity, to talk
free of any social norm and to use direct language
with the opposite sex. Such an opportunity makes
IM usage a controversial issue in Arab countries,
including Kuwait.
With globalization, the widespread use of the
Internet and users experience with e-mail and the
Web, people may use IM either to satisfy their curiosity about the opposite sex or to enjoy knowing
other people and cultures. It is not surprisingly
that Arab people may use IM to socialize since it
is compatible with the way they live.
GENERAL RESEARCH QUESTION
Therefore, the deployment of new ICT (including
IM) in Arab societies requires understanding the
role played by IM curiosity, IM compatibility
with tasks and values of users, as well as the role
of prior similar experience of related technologies. However, evidence documenting the causal
relationship between these variables is lacking.
To fill in this gap, this paper seeks to shed light
on the issue using a sample of Kuwaiti students,
through the following question:
How does compatibility and similar prior experience relate to perceived ease of use, curiosity,
perceived usefulness, level of usage IM and user
satisfaction?
In order to shed light on this research question,
this article is organized as follows: section 2 discusses the theory and the proposed research model;
the methodology used in this study is described
in section 3; results are reported in section 4; and
section 5 concludes and presents future research
directions and points to managerial implications.
THEORY AND BACKGROUND
The Research Model
While the study claims that several factors affect
IM usage and user satisfaction, it limits its focus
to two external variables and their effect on dependent variables. Figure 1 depicts the research
model that includes two independent variables:
(compatibility and prior similar experience), and
five dependent variables (PU, PEOU, curiosity,
usage, and user’ satisfaction).
This model is based on the technology acceptance model-tam (Davis, 1989), which advocates that perceived ease of use and perceived
usefulness are always the primary determinants
of behavior. Since this model has limited capabilities to predict technology adoption outside the
workplace, it was enriched with additional constructs borrowed from other three models. Usage
and user satisfaction are commonly studied dimensions of technology success (Al-Gahtani & King,
1999; Delone & McLean, 1992). Curiosity is
derived from the theory of flow (Csikszentmihalyi,
137
Factors Affecting Information Communication Technologies Usage and Satisfaction
Figure 1. Research model
1990). Compatibility is derived from the diffusion
of innovation theory (IDT) of Rogers (1983). Only
compatibility was added for the following reasons:
Moore and Benbasat (1991) analyzed previous literature on IDT and found that: only three
innovation characteristics (relative advantage,
complexity, and compatibility) are being consistently related to adoption behavior (triability and
demonstrability are excluded); relative advantage
is similar to the notion of PU; and complexity is
similar to PEOU, so the integration of the IDT
and TAM requires only adding compatibility to
TAM, besides PU and PEOU. Trialability and
demonstrability (which are not issues here) were
not included because the values are too low and
in essence a constant given that this study focuses
only at one technology (IM).
Research Hypotheses
The study hypotheses are outlined in Table 1. The
expected type of association is stated for each
of the 19 hypotheses. These associations were
hypothesized based on prior supporting studies
listed in Table 1.
IM Usage and User Satisfaction
IS usage and satisfaction are two common measure
of IS success. With regard to type of relationships
138
between the two construct Baroudi, Olson, and
Ives, (1986) suggested three models: the dominant model is that satisfaction and usage are not
related; usage influences satisfaction; and satisfaction influences usage. Furthermore, Delone and
McLean (1992) argued that system usage and
user’s satisfaction affect each other simultaneously and that the type of relationship between
them is reciprocal. Torkzadeh and Dwyer (1994)
reported that this issue is still debated among MIS
researchers and practitioners, and further study
is required to investigate the causality direction
between the two constructs. While Al-Gahtani and
King (1999) examined if satisfaction influences
usage and failed to find any significant relationship, this study hypothesizes that usage influences
satisfaction. Empirical studies have found that
IS usage is directly and positively related to user
satisfaction (Anandarajan et al., 2002; Baroudi
et al., 1986). In the Arab world, few empirical
studies which focused on the computer in the
workplace, found a correlation between IS usage
and user satisfaction (Al-Gahtani, 2004). As Arab
culture strongly influences social interaction and
personalization of relationships, it is likely that
the use of IM will increase users’ satisfaction.
Factors Affecting Information Communication Technologies Usage and Satisfaction
Table 1. Hypotheses to be tested and selected supporting references with findings between hypotheses
Hypotheses
Supporting references outside the Arab
countries
Supporting references from Arab
countries
H1a. There is a positive effect between
compatibility and PU.
There is a positive relationship between compatibility and PU in the workplace office automation system
(Al-Gahtani & King, 1999; Moore & Benbasat, 1991)
None
H1b. There is a positive effect between
compatibility and PEOU.
There a positive effect between compatibility and
PEOU of spreadsheet (Al-Gahtani & King, 1999).
None
H1c. There is a positive effect between
compatibility and curiosity.
Tested for the 1st time
Tested for the 1st time
H1d. There is a positive effect between
compatibility and IM usage.
There is a positive effect between compatibility and
usage of computer resource center (Taylor & Todd,
1995a), and the Web (Agarwal & Prasad, 1997).
There is a positive relationship between
compatibility and PC usage in Saudi Arabia
(Al-Gahtani, 2004).
H1e. There is a positive effect between
compatibility and user’ satisfaction.
There is a positive relationship between compatibility spreadsheet and user’ satisfaction in UK
(Al-Gahtani & King, 1999).
There is a positive relationship between
compatibility and PC user’ satisfaction in
Saudi Arabia (Al-Gahtani, 2004).
H2a. There is a positive effect between
PSE and PU.
There is a positive relationship between PSE and
PU of PCs (Igbaria, Giumaraes, & Davis, 1995)
None
H2b. There is a positive effect between
PSE and PEOU.
There is a positive relationship between PSE and
PEOU of PCs (Igbaria et al., 1995)
None
H2c. There is a positive relationship
between PSE and curiosity.
Tested for the 1st time
Tested for the 1st time
H2d. There is a positive effect between
PSE and IM usage.
There is a positive relationship between PSE and
usage of PCs (Igbaria et al., 1995), usage of Spreadsheet (Al-Gahtani & King, 1999)
None
H2e. There is a positive effect between
PSE and user’ satisfaction of IM.
Webster and Mortocchio (1992) found a positive
relationship between PSE and PC user’ satisfaction.
No study integrates PSE in technology
adoption.
H3a. There is a positive effect between
PEOU and PU.
Past studies on TAM found a positive relationship
between PEOU and PU on office automation systems
(Davis, 1989).
Past studies on TAM found a positive relationship between PEOU and PU (Rose &
Straub, 1998).
H3b. There is a positive effect between
PEOU and curiosity.
Tested for the 1st time
Tested for the 1st time
H3c. There is a positive effect between
PEOU and usage.
There is a positive relationship between PEOU and
PC usage (Davis, 1989; Anandarajan et al., 2000).
There is a positive relationship between PEOU and
social use of IM (Chen et al., 2004; Lancaster & Yen
2007; Lin et al., 2004; Wang et al., 2005).
There is a positive relationship between PEOU and PC usage (AlGahtani, 2004; Rose & Straub, 1998).
There is a positive relationship between
PEOU and the level of IM usage by employees (Rouibah, 2008).
H3d. There is a positive effect between
PEOU and IM satisfaction.
There is a positive relationship between PEOU and
user’ satisfaction of PCs (Anandarajan et al., 2002).
There is a positive relationship between
PEOU and PC user satisfaction in Saudi
Arabia (Al-Gahtani, 2004).
H4a. There is a positive effect between
curiosity and Usage.
Tested for the 1st time
Tested for the 1st time
H4b. There is a positive effect between
curiosity and IM satisfaction.
Tested for the 1st time
Tested for the 1st time
H5a: There is no positive effect between
PU and usage.
Past studies (Al-Gahtani & King, 1999; Davis,
1989;) found a positive relationship between
PU and usage of office automation systems.
There is a positive relationship between PU and the
level of IM usage in the workplace (De Vos et al.,
2004); and for IM social use (Wang et al., 2005)
Past studies found a positive relationship between PU and usage of office automation systems (Al-Gahtani,
2004; Rose & Straub, 1998;).
There is no significant relationship between
PU and the level of IM usage by employees
(Rouibah, 2008).
continued on following page
139
Factors Affecting Information Communication Technologies Usage and Satisfaction
Table 1. continued
Hypotheses
Supporting references outside the Arab
countries
Supporting references from Arab
countries
H5b. There is a positive effect between
PU and user’ satisfaction.
There is a positive relationship between PU and user’
satisfaction (Anandarajan et al., 2002).
Al-Gahtani (2004) found a positive relationship between PU and PCs user’ satisfaction.
H6. Current usage of IM will positively
affect user satisfaction.
There is positive relationship between Usage and
user’ satisfaction when using office automation systems (Baroudi et al., 1986; Anandarajan et al., 2002).
There is positive relationship between Usage and user’ satisfaction when using PCs
(Al-Gahtani, 2004).
Antecedents of Curiosity
•
Compatibility: When a technology is used,
it is important that it fits and is consistent
with the task it will be used to accomplish
and with the culture where it will be used.
Such a fit calls for the compatibility of the
culture and the technology. The concept of
compatibility stems from the IDT (Moore
& Benbasat, 1991) who defined it as the
degree to which an innovation is perceived
as consistent with the existing values, past
experiences, and needs of potential adopters. Al-Gahtani and King (1999) viewed
compatibility as “the adequacy of the system to the task performed.”
Most previous studies investigated the compatibility of a technology with regard to its adequacy
and fit with the individual’s task or style of work
(Agarwal & Prasad, 1997;-Al-Gahtani & King,
1999; Moore & Benbasat, 1991; Taylor & Todd,
1995a). However, there is very little conceptual
and empirical work done on the effects of compatibility in user acceptance theories in high context
societies.
Any technology that has a chance to be used on
a large scale should be compatible with people’s
value and culture. If a technology increases the
social presence between people, it is likely that
it will be adopted and used; therefore, a person is
unlikely to accept a technology that is perceived
incompatible with his or her life and practice.
Several past studies integrate compatibility in
the following cases: workstations (More & Ben-
140
basat, 1991), the Web (Agarwal & Prasad, 1997),
spreadsheet (Al-Gahtani & King, 1999). Their
findings are reported in Table 1. However, no
study integrated compatibility on IM adoption.
In the Arab world, Al-Gahtani (2004) is the
only study that included compatibility to test the
success factor affecting computer acceptance in
the workplace in Saudi Arabia. The author found
that relative advantage (PU) and compatibility
have both positive and significant correlations
with computer usage and user satisfaction.
Since IM is the ICT used, compatibility can be
defined rather loosely in this article as the adequacy
with the social setting in which a user is able to
use a technology. This view is compatible with
Moore and Benbasat’s (1991) view.
In a high context culture, this research posits
compatibility to affect PU, PEOU, current usage,
and user satisfaction (see Table 1). Understandably,
users are likely to “accept” or “feel comfortable”
with IM when it is perceived to be compatible
with the way they live and socialize. At the same
time, users are likely to consider a technology
easy to use when its adoption does not require
practice or major changes in their life. Significant incompatibility necessitates major changes
that require a considerable shift in their life and
learning on the part of users who, as a result, are
likely to perceive the technology to be not easy
to use. In addition, users who feel that a technology is compatible with the way they live (e.g.,
increasing social interaction) are likely to use it
and be satisfied about its usage.
Moreover, this article posits for the first time
that compatibility is linked to curiosity (see pre-
Factors Affecting Information Communication Technologies Usage and Satisfaction
vious discussion on the characteristics of Arab
culture). We argue that the more a user feels that
an ICT is compatible with his way of life and
enables him to satisfy his own curiosity, the more
he or she is likely to use it. In turn, the curiosity
derived from the IM with other people (such as
the opposite gender) will positively affect an individual’s perception of the usefulness of the IM.
•
Prior similar experience: The second
determinant of IM acceptance focuses on
past and similar experiences (PSE) of users. This article argues that the degree to
which a user adopts a previous, similar
system will influence his acceptance process of another system. A user who has
past experience with technologies, such
as the computer, e-mail and the Web, and
has accumulated many skills in using these
technologies has less anxiety to approach
new technologies such as IM. Individuals
who are not computer literate usually need
more training than individuals who have
substantial similar experience in similar
technologies. Experience with technology
was found to be an important factor that influences IT acceptance (Agarwal & Prasad,
1997; Taylor & Todd, 1995b).
A number of studies have integrated prior
experiences in the adoption of different technologies including computers (Igbaria et al., 1995;
Webster & Martocchio, 1992), Web (Agarwal &
Prasad, 1997), spreadsheet (Al-Gahtani & King,
1999), and the Internet (Loch et al., 2003). These
findings are reported in Table 1.
In the Arab world, Loch et al., (2003) focused
on the Internet’s usage and found that a lack of
education was the second largest obstacle to Internet usage in Egypt, after training availability.
We define PSE as the extend of an end-user’s
characteristics that encompasses both, and an end
user’s past exposure to a particular technology and
the end user’s subsequent level of expertise with
the technology. PSE is viewed as an individual
characteristic which could affect how easily and
well a user utilizes a particular technology. Based
on previous studies, we posit for the first time,
that past experience with similar technologies that
makes knowledge more accessible in memory,
will affect PEOU, curiosity, PU, usage and user’
satisfaction (see Table 1). People with high skills
and experience in similar technologies are likely
to derive more usefulness from the IM usage,
feel the technology is easy the use, and can use
more the IM and to satisfy their curiosity about
things they have partial or incomplete knowledge
of (e.g., talking to opposite sex). Thus, the more
a user has skills about a similar technology, the
more likely he perceives the new technology as
effective, productive and efficient.
Mediating Variables
•
Curiosity: In the information system field,
focus was on two types of motivations: extrinsic and intrinsic. Curiosity has been the
focus of few studies and was recognized as
an important intrinsic motivator (Agarwal
& Karahanna, 2000; Malone, 1981; Webster
et al., 1993).
Several researchers, who concentrated on the
effect of culture, found that Arab people have a
tendency to perform tasks, not only for economic
reward, but also for intrinsic motivation (Rouibah,
2008; Rouibah & Ould-Ali, 2005). Considering
that the Arab culture is masculine, clear gender
roles are the norm, and social interactions with
the opposite sex are not tolerated, and where most
overseas films are censored for emotional content,
we expect that curiosity, as intrinsic motivation,
will play an important role in IM acceptance.
Research indicated that individuals who experience immediate pleasure and joy from using the
technology and perceive any activity involving use
of a computer as inherently enjoyable are likely
141
Factors Affecting Information Communication Technologies Usage and Satisfaction
to use it more extensively than others (Malone,
1981; Webster & Mortocchio, 1992).
It is well known that, when people are intrinsically motivated (possess a high degree of
curiosity), they become absorbed with technology
(Csikszentmihalyi, 1990). The study of Webster
and Martocchio (1992) about playfulness found
that this concept is positively related to positive
mood and user satisfaction.
Using a similar argument, it can be inferred
that an individual with high curiosity dispositions
can be described as being guided by internal
motivation and orientation toward process with
self-imposed goals, and a tendency to satisfy his
or her curiosity through knowing new things about
which he or she has incomplete knowledge of.
Curiosity is defined as the extent the experience
arouses an individual’s sensory and cognitive
curiosity (Malone, 1981).
For the first time, this study posits that PEOU
affects curiosity, which in turn affects both IM
usage and user’ satisfaction. The rational behind
is that it is unlikely that the more people feel a
system is easy to use, and does not require mental
effort to use, the more it helps them to arouse their
curiosity. Moreover, people with high curiosity
feel more inclined to interact and to use IM programs as well as to be satisfied with the intrinsic
motivation they derive from its usage. It is also
likely that those individuals who interact more
curiously with IM programs may use IM as well
to be satisfied. Therefore, the hypotheses related
to curiosity were posited (see Table 1).
•
142
PU and PEOU: These are recognized as
two important mediating variables in technology adoption. PU refers to the extent
to which a person believes that using a
system would enhance his or her job performance, productivity and effectiveness
(Davis, 1989). PEOU refers to the degree
to which the technology is perceived as
relatively easy to understand and to use
(Davis, 1989).
Past studies have shown that PEOU is correlated with PU (Davis, 1989; Wang et al., 2005),
PEOU and PU are also related to current usage
(Davis, 1989).
With regard to IM technology, and opposite to
Li et al., (2005) who excluded PEOU, this study
included it. Other studies have shown the motives
of using IM included getting information, to hang
out with a friend, to flirt, and to avoid being alone
(Rubin, Perse, & Barbato, 1988). Other studies on
IM found that PEOU and PU are positively correlated to intention to use IM (Chen et al., 2004;
Wang et al., 2005).
METHOD
To provide insights into the research question, a
field study was investigated that consisted of a
survey questionnaire.
Subjects and Procedure
Data was collected from student subjects enrolled
at Kuwait University. Besides being widely used
at the university, this technology is appropriate for
at least two other reasons: first, it is a voluntary
technology in the sense that students use it of their
own accord and not from any mandate; and second,
the technology exemplifies the characteristics of
contemporary ICT that Arab people embrace on
a large scale. Moreover, the technology is widely
available from a number of wireless locations on
campus, franchised cafés, and libraries; thus, access is not an inhibitor to technology usage. Target
students are those enrolled in social sciences and
who use IM more than any other students. The
number of questionnaires needed was estimated
before distribution. It was calculated as follows:
W2= [(Z2) * P(1-P)]/ N. W2 refers to the precision
of the study. P refers to the proportion of students
in social sciences compared to the total students
at Kuwait University, which accounts for 58%. Z
(1.96) refers to the critical value from the normal
Factors Affecting Information Communication Technologies Usage and Satisfaction
table which corresponds to 95% confidence, that
is, answers from the sample are 95% close to the
reality at Kuwait University. N refers to the total
number of students at Kuwait University (17,000
in 2006). Therefore, the precision is W2= 0.0379,
which requires a sample size of 650.
Since all students were from social sciences,
we applied the stratified sample according to
their gender which is 70% female and 30% male.
Therefore the sample consisted of 455 female
and 195 male students. Furthermore, the pretest
sample requires an additional of 10% of the total
sample (65), 46 female and 19 male students. From
among the male students at social sciences, 214
were selected applying a simple random sample,
and the same was applied to the female students.
The specification of the above sample size was
done in line with recommendations of past studies
which require a minimum of 150 (Hair, Anderson,
Tatham, & Black, 1998). Such a sample generates
a parameter estimate with standard errors small
enough to be of practical usefulness and adheres
to structural equation modeling (Hair et al., 1998).
Data was collected during the 2005 spring
semester. 715 questionnaires were distributed
randomly to students in five social science colleges, but only 609 were completed and returned.
This study offered no incentives for completing
the questionnaire.
to confirmatory analysis. In a second phase, an
instrument was designed. Two faculty members
checked the validity of the adapted measures and
made changes concerning the format and wordings of the questions.
In a third phase, a pilot phase was conducted.
For this purpose, the questionnaire was distributed
to 75 students from one college at Kuwait University. Reliability and validity of the instrument
were checked based on the return responses. The
results of the pre-tests suggest that the instrument
possessed adequate reliability and validity.
Constructs Measurement
Most of the constructs in the research model were
taken from past and well known studies (see
appendix 3) with the exception of curiosity and
user’s satisfaction. All the constructs were subject
to convergence and discriminant validity. The reuse of well established constructs re-ensures the
content validity of the current study. All measures
were translated into Arabic, then translated back
into English by an independent translator, and
verified by an independent native speaker.
•
Pre-Tests
Before questionnaire design, the authors conducted a pre-test in three phases in order to examine and
validate the survey instrument ensuring validity
and reliability. In the first phase and before data
collection, personal interviews were conducted
with 30 students in order to identity the basic
motives to use instant messaging. The answers
were gathered, compiled and summarized (see
Rouibah & Ould-Ali, 2005), and revealed the
importance of curiosity as a separate construct
which was included in the research model (Figure 1). The curiosity construct was also subject
•
Curiosity: Initially, curiosity was measured by three items from Webster et
al. (1993) and reused by Agarwal and
Karahanna (2000): “using the system excites my curiosity,” “interacting with the
system makes me curious,” and “using the
Web arouses my imagination.” However,
during the pilot phase, we found some difficulties by our respondents to comprehend
the third item. Moreover, our respondents
found similarities between the first and
second item. With feedback received from
the respondents and two academic professionals, the two first items were adapted
and two new items were added (see appendix 3)
Prior similar experience (PSE): was
measured using three similar indicators of
143
Factors Affecting Information Communication Technologies Usage and Satisfaction
Igbaria et al., (1995). Compatibility was
measured by three indicators adapted from
Moore and Benbasat (1991) to fit the IM.
PU and PEOU were measured with indicators of Davis (1989)
Items of all variables, except current usage and
user’s satisfaction, were evaluated on five-point
semantic differential scales, ranging from strongly
disagree (1) to strongly agree (5). Students were
asked to rate each item according to how they
feel about using IM.
•
144
Current Usage: Review of past literature
showed that usage was measured through
self-reporting and five commonly used
measures were used: perceived frequency,
perceived daily usage, number of used
application, perceived usage level, and
sophistication level of application used.
This study re-used and customized four
of these measurements. Usage volume or
daily usage (UV) refers to time spent on
the use of IM per day. Respondents were
asked to indicate the amount of time using a six point scale ranging from (1) “almost never” to (6) “more than three hours
per day.” Frequency of Use (FU) refers to
how many times per week a respondent
reports using the IM. It was measured using a six point scale ranging from (1) don’t
use it at all to (6) use it several times each
day. With regard to variety of use (VU),
respondents were given a list of topics
they may use during IM and were asked
to select those that often fit their interest
including: social, cultural, political, love,
study, family, scientific, religion, medical,
personal, unethical, sport and entertainment. Type of use and synchronicity (TUS)
refers to whom the user frequently communicates with and exchanges information in real-time. Respondents were given
a list of six choices that they were asked
•
to select from: Family members; friends
with the same gender; friends with the opposite gender; unknown person with the
same gender; and unknown person with
the opposite gender. Social presence (SP),
or perceived information richness (Daft &
Lendgel, 1986), refers to the preference of
respondents to an existing communication
channel including: typing, talking, visualizing pictures, or watching each other using Web camera.
Users’ satisfaction: It refers to the extent
to which the usage of IM impacts the performance of students either positively or
negatively. Since this is the first study that
investigates the impact of IM on students,
a scale was proposed that is related to impact of IM on their study performance. The
proposed scale includes eight items (see
appendix 3), and it should be noted that
the indicators entail both positive (USIM2,
USIM6, USIM7, and USIM8) and negative
impacts, that is, dissatisfaction, (USIM1,
USIM3, USIM4, and USIM5).
Sample Characteristics
77.1% of the respondents were female students.
This is a little higher than the percentage at
Kuwait University, which is 70%. The majority
of respondents (96.2%) are young and their age
range from 18 to 25. 82% of the respondents are
single, while 17.7% are married. 26.1% are from
the college of business administration, 21.2%
from the college of social sciences, 33.1% from
the college of education, 35% from the college of
law, and 13.7% are from the college of art. 28.7%
have spent one to two years at their college and
23.4% from 3-to-4 years. 55.2% of the respondents have a GPA comprised between 2 and 3. In
addition, MSN software was found to be the most
frequently used IM program (78.5%), followed
by Yahoo (21.4%) and IRC (21.3%), which are
similar to the study of Wang et al., (2005). 92%
Factors Affecting Information Communication Technologies Usage and Satisfaction
of the respondents use typing, 18.7% report talking instead of typing, 11% send their pictures,
while a small percentage, (9.8%), report using
web camera during IM usage. Furthermore, the
majority of respondents (79.3%) use a nickname
to disguise their real identity. With regard to the
location where students often use IM, the majority
reported they use IM from their home (83.2%),
and (25.6%) from their college.
RESULTS AND DISCUSSION
The research model represents a structural equation modeling. We used SPSS to achieve data
reduction and LISREL to build the path model as
well as to simultaneously examine the potential
relationships between variables. Before applying the LISREL, the researchers first assessed
measurement model (in term of convergent and
discriminant validity) and then tested for significant relationships in the structural model.
The Measurement Model
Convergent and discriminant validity was
achieved based on factor analysis. Factors were
extracted using covariance matrices and the method of principal components. The factor loadings
(see appendix 4) show eight distinct constructs
instead of seven as well as their item-indicators.
Usage was separated into two distinct constructs:
The first one is labeled “usage magnitude” and
includes daily usage and frequency of use, and the
second one is labeled “socialization” and is related
to variety of use, type of use and synchronicity,
and social presence. The variance extracted by the
eight factors accounts for 73.7%, which depicts
a good model. Five item-indicators loaded very
low. One of these items is related to curiosity
(CU4) and four are related to the positive impact
of IM(USIM2, USIM6, USIM7, USIM8). Those
which loaded very well refer to negative perception (negative impact) of IM which we refer to as
dissatisfaction. The four indicators that were not
loaded were excluded since their factor loadings
are less than the cutoff value (0.50) recommended
by Hair et al. (1998).
Convergent validity can be assessed by three
measures: reliability (Hair, et al., 1998), composite
reliability (Fornell & Larcker, 1981; Hair et al.,
1998), and average variance extracted (Fornell
& Larcker, 1981; Hair et al., 1998). Reliability
measures the internal consistency of the questionnaire. Composite reliability is the extent to which
a variable or a set of variables is consistent with
what it is intended to measure. Average variance
extracted (AVE) measures the amount of variance
captured by the construct in relation to the amount
of variance due to indicator error.
Table 2 shows reliability, composite reliability
and AVE. Consistent with Hair et al., (1998), all the
Cronbach’s alpha values are above 0.80. Therefore,
the scales demonstrate good distribution properties. Composite reliabilities were found to range
from 0.800 to 0.924, which are consistent with the
recommendations of Fornell and Larcker (1981)
and Hair et al. (1998) since all values are above
0.80. AVEs were found to range from 0.620 to
0.844, which are consistent with the recommendations of Fornell and Larcker (1981) and Hair
et al., (1998). These three indicators reflect high
levels of convergent validity.
Discriminant validity refers to the degree to
which indicators differentiate among construct
measured based upon the correlations between
the indicators of each pair of constructs. Table 2
presents the inter-correlation (r) and the shared
variance among variables. The values “b” in the
diagonal represent the AVE. As suggested by Hair
et al., (1998), all r2 are found not to exceed the
AVE. This suggests that the model constructs are
distinct and unidirectional and ensure discriminant
validity.
To test the goodness of fit of the structural
model, several criteria were computed (Hair et
al., 1998; Taylor & Todd 1995a,b). The model
parameters estimate show the following: χ2=785;
145
Factors Affecting Information Communication Technologies Usage and Satisfaction
Table 2. LISREL Inter-correlation Matrix and shared variance among variables of the model
CU: curiosity; CP: compatibility; PSE: Prior similar experience; PU: perceived usefulness; PEOU: perceived ease of use
1
2
3
4
5
6
7
1-CU
0.820
0.620b
0.800c
2-CP
0.450d
0.202e
0.892
0.736
0.858
3-PSE
0.260a
0.067b
0.390
0.152
0.924
0.803
0.878
4-PU
0.420 a
0.176b
0.490
0.240
0.400
0.160
0.910
0.788
0.890
5-PEOU
0.610a
0.372b
0.510
0.260
0.540
0.291
0.650
0.422
0.842
0.574
0.821
6-Usage magnitude
0.340a
0.115b
0.220
0.048
0.190
0.036
0.280
0.078
0.370
0.137
0.800
0.665
0.890
7-Socialization
0.340a
0.115b
0.250
0.062
0.250
0.062
0.280
0.078
0.480
0.230
0.190
0.006
0.840
0.639
0.900
8-Disatisfaction
0.080a
0.006b
0.060
0.003
0.050
0.002
0.070
0.004
0.10
0.010
0.160
0.025
0.130
0.016
8
a
0.913
0.844
0.837
a
Composite reliability, bAverage variance extracted, cCronbach alpha, dCorrelation; eShared variance among variables of the model (Square
correlation).
degree of freedom (DF)= 226; χ 2/DF=3.47, GFI
= 0.90; AGFI = 0.86; NFI= 0.95; NNFI = 0.96;
CFI = 0.96; RMSR = 0.056; RMSEA=0.05 and
SRMR = 0.056. This study’s results provide
evidence of the relatively strong model fit to the
observed data.
Test of the Structural Model
Figure 2 shows the validated model for IM usage
and students’ satisfaction in Kuwait. Parameters
whose t-values are greater than or equal to +- 1.96
are considered to be significantly different from
0 (Hair et al., 1998).
•
146
Predicting PEOU: Compatibility has a direct effect on PEOU (γ=0.36; P<0.05) and
•
PSE also has a direct effect on PEOU
(γ=0.40; P<0.05). Hypotheses (H1b and
H2b) are supported. Among these two variables, PSE plays the strongest effect on
PEOU.
Predicting curiosity: There is a direct effect between compatibility and curiosity
(γ=0.21; P<0.05), PEOU and curiosity (β
=0.57; P<0.05). However, PSE has a direct
and negative effect on curiosity (γ=-0.12;
P<0.05). Hypotheses H1c, H2c and H3b
are, therefore, valid. The negative effect
between PSE and curiosity means the lower the level of experience with computers,
web and e-mail, the more she or he is curious to use IM. The strongest direct and total effect on curiosity comes from PEOU,
Factors Affecting Information Communication Technologies Usage and Satisfaction
Figure 2. The LISREL model solution with significant direct effects of IM acceptance in Kuwait
•
while PSE exerts the strongest indirect effect (γ=0.23; P<0.05). In addition, PEOU
indirectly mediates the effect of compatibility and PSE on curiosity. However, the
indirect effect of PSE on curiosity is lower
than the direct effect of PEOU since the total effect of PSE (γ=0.10; P<0.05) is less
than that of PEOU (β =0.52; P<0.05).
Predicting PU: There is a direct effect
between compatibility and PU (γ=0.21;
P<0.05) and between PEOU and PU
(γ=0.52; P<0.05), but PSE did not effect PU. Therefore, hypothesis H1a, not
H2a, is supported. The strongest direct
and total effect on PU comes from PEOU.
Compatibility exerts the second largest direct and total effect, while PSE exerts the
strongest indirect effect (γ=0.21; P<0.05).
Similar to curiosity, PEOU mediates indirectly the effect of compatibility and
PSE on PU. However, the indirect effect
of PSE on PU is lower than the direct effect of PEOU since the total effect of PSE
(γ=0.24; P<0.05) is less than that of PEOU
•
(β =0.52; P<0.05). This implies that external variables play less an affect on PU than
does PEOU. Such a result reveals the importance of PEOU on IM acceptance in the
Arab world.
Predicting IM usage: Compatibility exerts only an indirect effect on usage magnitude (γ=0.13; P<0.05) and socialization
(γ=0.12; P<0.05); therefore, hypothesis
H1d is partially supported. Similar to compatibility, PSE exerts an indirect effect on
usage magnitude (γ=0.08; P<0.05), but it
also exerts a direct (γ=0.20; P<0.05) and
indirect effect (γ=0.11; P<0.05) on socialization. Hypothesis H2d is, therefore,
supported. While PEOU exerts a direct
(β =0.13; P<0.05) and indirect effect (β
=0.12; P<0.05) on usage magnitude, it
exerts only a direct effect on socialization
(γ=0.31; P<0.05). PEOU exerts a direct effect on usage magnitude (β =0.13; P<0.05)
and on socialization (β =0.31; P<0.05) supporting hypothesis H3c. Curiosity exerts a
direct effect on usage magnitude (γ=0.18;
147
Factors Affecting Information Communication Technologies Usage and Satisfaction
P<0.05) and no effect on socialization.
Hypothesis H4a is, therefore, partially
supported.
The strongest direct effect on usage magnitude comes from curiosity followed by PEOU.
Compatibility exerts the strongest indirect effect
followed by PEOU. Surprisingly, the total effect
of curiosity is less than that of PEOU.
PSE exerts the strongest direct effect on socialization followed by prior similar experience.
Surprisingly, the strongest indirect effect on socialization comes from compatibility followed by
prior similar experience. Similar to the total effect
on usage magnitude, PEOU exerts the strongest
total effect on socialization followed by prior
similar experience.
As hypothesized, PU has no effect on current usage (neither on usage magnitude nor on
socialization) since IM is mainly used for social
purpose which confirms hypothesis H5a.
•
148
Predicting
user
satisfaction:
Compatibility has no direct effect but exerts an indirect (γ =0.18; P<0.05) and total
effect (γ =0.27; P<0.05) on user dissatisfaction. However, PSE exerts a negative
direct effect (γ =-0.19; P<0.05) and positive indirect effect (γ =0.14; P<0.05) on
user dissatisfaction. The negative effect
means the more a user has previous experience with related technologies, the more he
is satisfied with IM. Thus, hypothesis H1e
is partially supported while H2e is fully
supported. Furthermore, the results indicate that PEOU exerts an indirect effect on
user dissatisfaction (β =0.20; P<0.05) and
PU exerts a direct effect (β =0.22; P<0.05).
Therefore, hypothesis H3d is partially validated while H5b is validated. Curiosity
exerts a direct positive effect (β =0.13;
P<0.05) on user dissatisfaction supporting
hypothesis H4b. Similarly, usage magnitude (γ =0.14; P<0.05) and socialization
(β =0.11; P<0.05) exert a direct effect on
user dissatisfaction leading to validate hypothesis H6. Results also indicate that PU
exerts the strongest direct effect on user
dissatisfaction followed by prior similar
experience. However, results highlight the
importance of PEOU as it exerts the strongest indirect effects on user dissatisfaction
followed by the compatibility. When computing the total effect, results reveal the
predominance of PEOU and compatibility
since they exert the strongest same effect,
followed by the curiosity construct.
Furthermore, results indicate that the model
explains substantial variance in PEOU (R2=40%),
curiosity (R2=40%), PU (R2=46%), usage magnitude (R2=17%), socialization (R2=25%), and
users’ dissatisfaction (R2=14%).
Table 3 summarizes the supported and nonsupported hypotheses.
In order to test the effect of gender on IM
acceptance, factor scores were first computed,
saved and then tested against gender using the
non-parametric Mann-Whitney test because the
scores were not normally distributed. Results
indicated that no significant differences (at 5%)
were found between male and female students
in their perception of factors scores. The sample
was, therefore, considered as one group.
Discussion
This study focused on factors affecting IM usage and user satisfaction in Kuwait, in an Arab
country. Importantly, it shows why, how and what
consequences of IM usage on a sample of students.
This study reveals two types of usage: People
use IM to socialize (socialization) in their daily
life with a high frequency (usage magnitude).
The more frequently people use IM, the more
they engage to know new people with whom
they spend much of their time. The long time
they spent enables users to satisfy their curios-
Factors Affecting Information Communication Technologies Usage and Satisfaction
Table 3. Summary of hypotheses validation
Hypotheses
Validation
H1a: Compatibility-PU
Yes
H1b: Compatibility-PEOU
Yes
H1c: Compatibility-Curiosity
Yes
H1d: Compatibility-Usage
Partially
H1e: Compatibility-User’s satisfaction
Partially
H2a: PSE-PU
No
H2b: PSE-PEOU
Yes
H2c: PSE-Curiosity
Yes
H2d: PSE-Usage
Yes
H2e: PSE- User’s satisfaction
Yes
H3a. PEOU-PU
Yes
H3b: PEOU-Curiosity
Yes
H3c: PEOU-Usage
Yes
H3d: PEOU- User’s satisfaction
Partially
H4a: Curiosity-Usage
Partially
H5a: PU-Usage
Yes
H4b: Curiosity- User’s satisfaction
Yes
H5b: PU-User’s satisfaction
Yes
H6: Usage – User’s satisfaction
Yes
ity from different angles including knowledge
about new subjects, habits, and cultures as well
as knowing with the opposite gender. This is an
expected result since the culture is characterized
by high social interaction and personalization of
relationships and gender segregation.
With regard to socialization, the study reveals
that sampled students use IM to socialize with
people they know (such as friends and family)
and those they do not know from both genders
based on IM synchronicity. Also, students use
IM for the social presence features since they use
several ways to communicate including typing,
talking, and seeing each other via pictures and
web camera. Sampled students also use IM for
several subjects including those related to their
studies, loves, politics, and so forth. In contrast,
user’s satisfaction is essentially associated with the
negative impact of IM on students’ performance,
which we labeled dissatisfaction.
The study findings support, conflict and extend previous findings, in particular those using
strong theory models outside the workplace and
limits the comparison to three (Chen et al., 2004;
Li et al., 2005; Wang et al., 2005). Therefore, this
study excludes from comparisons past studies
that focused on IM in the workplace (Cameron
& Webster, 2005; De Vos et al., 2004; Huang et
al., 2007; Huang & Yen, 2003; Leung 2006; Shen
& Gallivan, 2006).
While this study focuses on the social use of
IM, as the case of Li et al.’s (2005) study, it extends
their findings and highlights the current practices
of ICT for social use and outside the workplace
in an Arab country. In particular, the extensions
include the role of compatibility, prior similar
experience, and curiosity and their potential effect
on IM acceptance. Li et al. (2005) developed a
TAM model to explain an individual’s use of IM
in keeping and sustaining interpersonal relationships. Their model was tested using a sample 273
students and revealed that intention to use IM is
driven by perceived usefulness and perceived enjoyment since they influence the intention to use
with the same effect. Opposing Li et al. (2005),
our study focuses on IM usage and user’s satisfaction and reveals that social usage is ease of use
driven, usage magnitude is curiosity driven, while
user satisfaction is compatibility and ease of use
driven. In addition, the current study is different
from Li et al. (2005) since it includes ease of use
and excluded perceived enjoyment.
Results of this study support and extend the
results of Chen et al.’s (2004) who found that IM
users tend to use IM for both social and work/
course reasons because of its ease of use. Chen et
al.’s study found also IM usage for social interaction is higher, which also supports findings of our
current study since the variance of socialization is
the largest one 25%. Indeed, Chen et al.’s (2004)
compared the relation of eight motivation variables
(including: ease of use, social interactions, enjoy-
149
Factors Affecting Information Communication Technologies Usage and Satisfaction
ment, and other variables) with past experience
(desktop and wireless) and intention to use IM
in the future. The model was tested on a sample
similar to the current study (239 male and 219
female students in the USA). Chen et al. (2004)
revealed that IM users found IM more enjoyable,
easier to use, and better than e-mail for social
interactions. Chen et al.’s study also revealed that
certain factors (including enjoyment, ease of use
and social interaction) positively correlated with
the intention to use IM in the future. In addition,
our findings extend Chen et al.’s results who found
that previous experience (in the forms of years of
experience and frequency of use) has an effect on
intention to use IM in the future while our results
reveal that past similar experience (in the form of
Web, computer and e-mail) has an effect not only
on current usage (i.e., socialization) but also on
user’ satisfaction.
Our findings disagree with those of Wang et
al. (2005) in Taiwan who found that IM intention
to use is perceived as usefulness driven.
With regard to the new added construct, previous studies did integrate curiosity as a dimension
of the flow construct (Webster et al., 1993) or
cognitive absorption (Agarwal & Karahanna,
2000); however, this is one of the rare studies that
operationalizes curiosity as a separate construct
and provides evidence of its importance on ICT
adoption and satisfaction. This study found a
direct positive affect of curiosity on both usage
behavior and users’ dissatisfaction. Curiosity was
found to be essentially associated only with usage magnitude, and not with socialization, which
reveals that users spend a lot time to satisfy their
curiosity about other people they do not know or
the opposite gender.
Another finding of the article is the effect
played by prior similar experiences on current IM
usage since it plays the strongest indirect effect on
socialization. While PEOU exerts the most direct
influence on usage and users’ dissatisfaction, PSE
exerts the strongest direct effect on PEOU, PU and
curiosity, thus confirming the role of knowledge
150
gained from prior similar technologies to ease
the acceptance of new ones. This is in line with
previous studies conducted outside the Arab world
which have demonstrated the direct and indirect
effect of prior experience on either PEOU, PU
(Al-Gahtani & King, 1999; Igbaria et al., 1995;
Taylor & Todd, 1995b; Webster & Martocchio,
1992). This study has proved that prior similar
experience with computers, the Web and e-mail
has a direct affect on PEOU, curiosity, current
usage and users’ dissatisfaction. This study is also
distinct from previous studies since it contains one
of the rare studies that provides evidence of the
existence of a direct positive effect between prior
similar experience and users’ dissatisfaction which
call for caution when using IM in the workplace.
Another strong finding of the study is related
to the role of compatibility of ICT with the private
life of users (family, friends) and their workplace
(study). This study reveals that compatibility is
positively related to PU, which is in line with studies outside the Arab world (Al-Gahtani & King
1999; Moore & Benbasat, 1991). More valuable,
this study is the first that integrates compatibility
in the Arab world.
Another strong contribution of this article
outlines the role of compatibility on user’s dissatisfaction. Our study found that compatibility exerts
the strongest direct effect on user dissatisfaction
and the second strongest total effect on usage
magnitude after PEOU. These findings conflict
with those of Moore and Benbasat (1991) who
found a strong correlation between compatibility
and PU, and Al-Gahtani and King (1999) who
found compatibility as the strongest determinant
of PU, the current study reveals less an effect of
compatibility compared to PEOU.
This study conflicts and supports the findings
of previous studies (see Table 1) and extends
previous ones. In particular, the study tested and
validated for the first time five hypotheses (one is
partially validated) related to antecedents (compatibility, PSE, and PEOU) and consequences (use
and user satisfaction) of curiosity. The effect of
Factors Affecting Information Communication Technologies Usage and Satisfaction
curiosity supports previous results on the effect
of culture on ICT acceptance in the Arab world
(Loch et al., 2003).
With regard to the difference between the
study’s results and previous ones, four plausible
explanations can be advanced. First, the used technology is an ICT that has enjoyment components.
Second, the technology usage is voluntary at an
individual level, while most past studies referred
to mandatory office systems in workplace settings
(Al-Gahtani & King, 1999; Moore & Benbasat,
1991). Third, what could be valid in an individual
and high feminine culture may not be necessarily
valid in a collective and high masculine one (as
is the Arab culture). Fourth, the subjects were
young college students (between 17 and 25).
Older persons might use IM for both work and
non-work activities.
In summary with the findings discussed previously, it can be concluded that PEOU is the
strongest determinant of IM acceptance in the
Arab world, which, therefore, indicates that ICT
adoption is ease of used driven.
Another contribution of this article is related
to the potential effect between usage and user’s
dissatisfaction. With regard to the three models
proposed by Torkzadeh and Dwyer (1994), this
study succeeds in showing that current usage
directly and positively affects users’ satisfaction.
CONCLUSION AND IMPLICATIONS
This study develops a new model for ICT (i.e.,
instant messaging technology) usage and satisfaction valid in an Arab country. The study reveals
two types of use: Usage magnitude and socialization. Major findings indicate that socialization is
perceived ease of use driven; usage magnitude is
curiosity driven; and user satisfaction is ease of
use and compatibility driven.
This article makes four contributions to the
field of information system and information technology. First, the study tests an IM model that is
comprised of a widely held set of relationships
in an Arab country (Kuwait) where they have not
been tested before. Second, the study adds a new
driver, “curiosity,” for technology adoption. This
is a new mediating factor between external factors
and user satisfaction, since, to our knowledge,
no study has investigated curiosity and succeeds
to provide evidence about its value to IM circumstances. Third, it shows a strong distinction
between usage quality and quantity which may
be local or a missed factor across all studies on
technology acceptance. Fourth, it reveals that
some relationships are not as strong (e.g., the
relationship between curiosity and satisfaction)
which suggests room for further investigation.
The findings of this study contribute to a
better theoretical understanding of the factors
that promote ICT in the Arab world. It should be
noted that the variance (R2) explained by each
variable in the model is successively: 40% for
curiosity, 46% for PU, 40% for PEOU, 17% for
usage magnitude, 25% for socialization, and 14%
for user dissatisfaction. These values are acceptable and are quite similar to previous studies. To
limit the comparison to one study that focused on
the effect of PSE on intention to use, Taylor and
Todd (1995b) found: R2= 43% (intention to use)
and R2= 21% (usage) for experienced users; R2=
60% (intention to use); and R2= 17% (usage) for
inexperienced users.
The results of this study suffer from certain
limitations which must be considered in interpreting the results. First, the results of the study are
valid for students using IM for social purposes.
Thus, the generalizability of the respondents’
behaviors to a more general workforce may be
somewhat limited. However, for this particular age
group (18-23), one that likely is a primary user of
IM and that will move into the workforce in the
future, the study should be reasonably applicable.
Second, the age of the majority of the respondents
is between 18-23 years, and, therefore, study’s
results cannot be generalized to people of other
ages. Third, conclusions drawn in this study are
151
Factors Affecting Information Communication Technologies Usage and Satisfaction
based on IM, which is a single communication
technology, and some of the questions, particularly
regarding curiosity, were created specifically for
this technology. As a communication technology,
curiosity may play a larger role than in other categories of IT, which represents another potential
contingency.
Even though it has its limits, this study has
implications for both practice and research.
From a managerial perspective, this study has
several implications. First, this study suggests
that curiosity is a determinant of IM acceptance.
Thus, curiosity results in immediate subjective
experience such as people’s involvement in
their activities and, consequently, people will be
more motivated to engage in ICT adoption in the
future. Users with high curiosity will exercise
and develop skills through exploratory behavior
and, thus, will be better able to react to situations
requiring these skills in the future. As students
are potential employees, the usage of IM in the
workplace is questioned. One possible caution
of using IM in the workplace is that students
(potential employees) may not draw a clear line
between professional and personal use. This study
has shown that students (aged between 18-23),
will move into the workforce, have a tendency
to spent lot of time socializing with people and,
therefore, personal use of IM may distract them
from their business activities. Employees may
give priority to personal socializations, and not
business objectives, and thus, lead to lower employee productivity. Second, the study’s results
have far reaching implications for developed
countries as well. With the negotiation of World
Trade Organization, many global companies
are expected to penetrate the Gulf region in the
service (including banking1, investment, and real
state) and other business sectors. A recent study
has shown that the IT/ICT investment growth
rate in the Arabian Gulf countries during the last
three years represented 9.4% and is expected to
reach $7.9 billion (Research Institute for Digital
Economy, 2006). Another study has shown that IT
152
consulting services accounted for $700 million in
the Arabian Gulf countries in 2005, which is increasing annually (Gulf Business, 2006). Qatar has
invested $10.4 billions in the gas industry, while
Saudi Arabia is planning to invest $350 billion on
infrastructure projects, including IT/ICT. Other
reports showed that the ICT market in the region
is estimated at $46 billion and increases annually
by 11% (Delta Capital, 2006). The amount of 2,000
projects2 in the GCC is evaluated at $1 trillion.
Therefore, understanding what factors contribute
to successful IT/ICT adoption is crucial for global
companies wishing to tap into these investments
in potential sale. This study provides to a certain
extend some guidelines on how to understand the
Arab culture and people’s behavior with regard
to ICT adoption.
From a practical, educational and scholarly
perspective, this study suggests three perspectives.
First, it encourages researchers to consider the
whole world as potential locations for research
to test and enrich standard technology adoption
theories. In testing these theories, this study
suggests to consider both quantity and quality
measure such as the usage in this study. One
benefit of testing theories in varied settings is
that differences in culture may reveal differences
that exist everywhere but in different measures.
A potential region that deserves more focus is the
GCC3 countries since their total GDP is high (see
appendix 2). A longitudinal cross cultural research
among GCC, which share same culture in term of
language, history, values, wealth, and have enormous investment on IT/ICT about IM acceptance
may shed more light on ICT acceptance.
Second, while the study provides evidence that
curiosity contributes to an individual’s willingness
to try something new, the question of whether
curiosity continues to play the same effect in the
intention to continue using the technology and be
satisfied remains unanswered and needs further
research.
Third, the low variance explained by current
usage (usage magnitude with 17% and socializa-
Factors Affecting Information Communication Technologies Usage and Satisfaction
tion with 25%) and users’ dissatisfaction (14%)
indicate that a large percentage variance remains
unexplained and suggests the need for additional
research incorporating potential unmeasured
variables such as the inclusion of social norm.
This study reveals that Kuwaiti people use IM
to socialize with a high frequency and 79.3% of
respondents use a nickname which reveals that
they use IM to discuss subjects not allowed in
face-to-face discussions or those with the opposite
gender because of high social norm factors that
prevent such an attitude. This article encourages
investigating social norms as additional constructs.
ACKNOWLEDGMENT
This research was funded by Kuwait University,
Research Grant IQ 01/05. The author acknowledges the Research Administration Project for
its support.
REFERENCES
Abla, G., Flanagan, S. M., Peng, Q., Burruss, J.
R., & Schissel, D. P. (2006). Advanced tools for
enhancing control room collaborations. Fusion
Engineering and Design, 81(15-17), 2039–2044.
doi:10.1016/j.fusengdes.2006.04.059
Agarwal, R., & Karahanna, E. (2000). Time flies
when you’re having fun: Cognitive absorption
and beliefs about information technology usage.
Management Information Systems Quarterly,
24(4), 665–694. doi:10.2307/3250951
Agarwal, R., & Prasad, J. (1997). The role of
innovation characteristics and perceived voluntariness in the acceptance of information technologies. Decision Sciences, 28(3), 557–5582.
doi:10.1111/j.1540-5915.1997.tb01322.x
Al-Gahtani, S. S. (2004). Computer technology
acceptance success factors in Saudi Arabia: an
exploratory study. Journal of Global Information
Technology Management, 7(1), 5–29.
Al-Gahtani, S. S., & King, M. (1999). Attitudes,
satisfaction and usage: Factors contributing to
each in the acceptance of information technology. Behaviour & Information Technology, 18(4),
277–297. doi:10.1080/014492999119020
Al-Shareef, B. (1995). Positive and negative behavior at the workplace. Public Administration,
35(1), 129–170.
Anandarajan, M., Igbaria, M., & Anakwe, U. P.
(2002). IT acceptance in a less-developed country:
a motivational factor perspective. International
Journal of Information Management, 22, 47–65.
doi:10.1016/S0268-4012(01)00040-8
Baroudi, J. J., Olson, M. H., & Ives, B. (1986). An
empirical study of the impact of user involvement
on system usage and information satisfaction.
Communications of the ACM, 29(3), 232–239.
doi:10.1145/5666.5669
Berlyne, D. E. (1965). Structure and direction in
thinking. New York: Wiley.
Cameron, F. A., & Webster, J. (2005). Unintended
consequences of emerging communication technologies: Instant Messaging in the workplace.
Computers in Human Behavior, 21, 85–103.
doi:10.1016/j.chb.2003.12.001
Chen, K., Yen, D. C., & Huang, A. H. (2004).
Media selection to meet communications contexts: Comparing e-mail and instant messaging
in an undergraduate population. Communications
of the Association for Information Systems, 14,
387–4051.
Csikszentmihalyi, M. (1990). Flow: The psychology of optimal experience. New York: Harper
and Row.
153
Factors Affecting Information Communication Technologies Usage and Satisfaction
Daft, R. L., & Lengel, R. H. (1986). Organizational
information requirements, media richness, and
structural determinants. Management Science,
32, 554–571. doi:10.1287/mnsc.32.5.554
Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. Management Information Systems
Quarterly, 13(3), 319–340. doi:10.2307/249008
De Vos, H., Hofte, H., & de Poot, H. (2004).
IM at work: Adoption of instant messaging in a
knowledge organization. In proceedings of the
37th Annual Hawaii International Conference on
System Sciences (HICSS), (pp. 1-10).
Delone, W. H., & Mclean, E. R. (1992). Information systems success: the quest for dependent
variable. Information Systems Research, 3(1),
60–95. doi:10.1287/isre.3.1.60
Delta Capital. (2006). Al-Qabas Newspaper 20
December 2006
Fornell, C. R., & Larcker, D. F. (1981). Structural
equation models with unobservable variables and
measurement error. JMR, Journal of Marketing
Research, 18, 39–50. doi:10.2307/3151312
Huang, A. H., Hung, S. Y., & Yen, D. C. (2007).
An exploratory investigation of two internet-based
communication modes. Computer Standards
& Interfaces, 29(2), 238–243. doi:10.1016/j.
csi.2006.03.011
Huang, A. H., & Yen, D. (2003). Usefulness of
instant messaging among young users: social vs.
work perspective. Human Systems Management,
22(2), 63–72.
Igbaria, M., Guimaraes, T., & Davis, G. B. (1995).
Testing the determinants of microcomputer usage via a structural equation model. Journal of
Management Information Systems, 11(4), 87–114.
Internet World Stats. (2008), web: www.internetworldstats.com
Isaacs, E., Walendowski, A., & Ranganathan, D.
(2002). Mobile Instant Messaging through Hubbub. Communications of the ACM, 45(9), 68–72.
doi:10.1145/567498.567501
Lancaster, S., & Yen, D. C. (2007). The selection
of instant messaging or e-mail college students’
perspective for computer communication. Information Management & Computer Security, 15(1),
5–22. doi:10.1108/09685220710738750
Ghani, J. A., & Deshpande, S. P. (1994). Task
characteristics and the experience of optimal flow
in human-computer interaction. The Journal of
Psychology, 128(4), 381–391. doi:10.1080/002
23980.1994.9712742
Leung, L. (2006). Unwillingness-to-communicate
and college students’ motives in SMS mobile
messaging. Telematics and Informatics, 24(2),
115–129. doi:10.1016/j.tele.2006.01.002
Gross, E. F. (2004). Adolescent internet use: What
we expect, what teens report. Applied Developmental Psychology, 25, 633–649. doi:10.1016/j.
appdev.2004.09.005
Li, D., Chau, P. Y., & Lou, H. (2005). Understanding individual adoption of instant messaging: An
empirical investigation. Journal of the Association
for Information Systems, 6(4), 102–126.
Hair, J. F., Anderson, R. E., Tatham, R. L., &
Black, W. C. (1998). Multivariate data analysis
(5th ed.). New York: Prentice Hall.
Lieberman, J. N. (1977). Playfulness. New York:
Academy Press.
154
Lin, J., Chan, H. C., & Jin, Y. (2004). Instant
messaging acceptance and use among college
students. In Proceedings of the Pacific Conference
on Information Systems (PACIS), (pp. 181-194).
Factors Affecting Information Communication Technologies Usage and Satisfaction
Loch, K. D., Straub, D. W., & Kamel, S. (2003).
Diffusing the internet in the arab world: The role
of social norms and technological culturation.
IEEE Transactions on Engineering Management,
50(1), 45–63. doi:10.1109/TEM.2002.808257
Malone, T. W. (1981). Toward a theory of intrinsically motivating instructions. Cognitive Science,
4, 333–369. doi:10.1207/s15516709cog0504_2
McClea, M., Yen, D., & Huang, A. H. (2004).
An analytical study towards the development
of a standardized IM application. International
Journal of Computer Standards and Interfaces,
26(4), 343–355. doi:10.1016/j.csi.2003.11.002
Moore, G. C., & Benbasat, I. (1991). Development of an instrument to measure the perception
of adopting an information technology innovation.
Information Systems Research, 2(3), 192–222.
doi:10.1287/isre.2.3.192
Myers, M. D., & Tan, F. B. (2002). Beyond models of national culture in informational system
research. Journal of Global Information Management, 24–32. doi:10.4018/jgim.2002010103
Nicholson, S. (2002). Socialization in the “virtual
hallway;” Instant messaging in the asynchronous
web-based distance education classroom. The
Internet and Higher Education, 5, 363–372.
doi:10.1016/S1096-7516(02)00127-6
Peter, J., Valkenburg, P. M., & Schouten, A. P.
(2006). Precursors of adolescents’ use of visual
and audio devices during online communication.
Computers in Human Behavior, 23(5), 2473–2487.
doi:10.1016/j.chb.2006.04.002
Rau, P., Gao, Q., & Wu, L. (2006). Using mobile communication technology in high school
education: Motivation, pressure, and learning
performance. Computers & Education, 50(1),
1–22. doi:10.1016/j.compedu.2006.03.008
Rogers, E. M. (1983). Diffusion of innovations
(3rd ed.). New York: The Free Press.
Rose, G., & Straub, D. (1998). Predicting IT use:
applying TAM to the arab world. Journal of Global
Information Management, 6, 39–46.
Rouibah, K. (2008). (Forthcoming). Social usage
of instant messaging by employees outside the
workplace in Kuwait: A structural equation model.
IT and People. doi:10.1108/09593840810860324
Rouibah, K., & Ould-Ali, S. (2005). Investigation of certain factors associated with the chat
technology acceptance and its impact on CBA
students. Arab Journal of Administrative Sciences,
12(2), 1–56.
Rubin, R. R., Perse, E. M., & Barbato, C. A.
(1988). Conceptualization and measurement
of interpersonal communication motives. Human Communication Research, 14, 602–628.
doi:10.1111/j.1468-2958.1988.tb00169.x
Shen, Y., & Gallivan, M. (2006, August 4-6). Understanding workplace instant messaging adoption, use and consequences via social perspectives.
In Proceedings of the 12th Americas Conference
on Information Systems, Acapulco, Mexico.
Simon, A. F. (2006). Computer-mediated communication: Task performance and satisfaction. The
Journal of Social Psychology, 143(3), 349–379.
doi:10.3200/SOCP.146.3.349-379
Taylor, S., & Todd, P. A. (1995a). Understanding
information technology usage: a test of competing models. Information Systems Research, 6(2),
144–174. doi:10.1287/isre.6.2.144
Taylor, S., & Todd, P. A. (1995b). Assessing IT
usage: The role of prior experience. Management
Information Systems Quarterly, 19(4), 561–570.
doi:10.2307/249633
Research Institute for Digital Economy. (2006),
Al- Watan 15 January 2006
155
Factors Affecting Information Communication Technologies Usage and Satisfaction
Torkzadeh, G., & Dwyer, D. J. (1994). A path analytic study of determinants of information system
usage. OMEGA International Journal of Management Science, 22(4), 339–148. doi:10.1016/03050483(94)90060-4
Wang, C. C., Hsu, Y. H., & Fang, W. C. (2005). Acceptance to technology with network externalities:
An empirical study of internet instant messaging
services. Journal of information technology theory
& application, 6(4), 15-29.
Webster, J., & Martocchio, J. J. (1992). Microcomputers playfulness: Development of a measure with workplace implications. Management
Information Systems Quarterly, 16(1), 201–224.
doi:10.2307/249576
Webster, J., Trevino, L. K., & Ryan, L. (1993).
The dimensionality and correlates of flow in
human-computer interactions. Computers in Human Behavior, (9): 411–426. doi:10.1016/07475632(93)90032-N
156
(2008). World Bank Development Indicator.
Washington, DC: World Bank.
ENDNOTES
1
2
3
Several foreign banks have already joined
the Kuwaiti market including: Citibank, BNP
Paribas, and HSBC.
This is a quotation of Hashem Al-Refaee,
CEO of Project Management Association,
Al-Qabas 30th May 2007.
GCC is short for Gulf Cooperative Council.
These countries include Kuwait, Saudi Arabia, Bahrain, Qatar, Oman, and the United
Arab Emirates.
Factors Affecting Information Communication Technologies Usage and Satisfaction
APPENDIX 1
Table A1. Past studies on IM usage (Rouibah 2008)
Study
Authors
Objectives
Journal/ conference
Sample
Usage vs. satisfaction
Isaacs et al., (2002)
Describe usage of an IM system
“Hubbub”
Communication of the
ACM
Observation and interviews with 25 users
Current and continuous
usage
Nicholson (2002)
Investigate why students use IM
Internet and Higher
Education
30 students at Syracuse University
Type of use
Huang and Yen, (2003)
Compare usefulness of IM among young
users for social and work perspective.
Identify factors that are important to
the usefulness of IM
Human Systems Management
500 college students
N/A
Gross (2004)
Social activities of online adolescents
on the Internet including IM
Applied Developmental Psychology
100 male and 161
female adolescents
Current usage
McClea et al., (2004)
IM development
Computer Standards &
Interfaces
N/A
N/A
Chen et al., (2004)
Compare intention and future use of IM
with email in the work and coursework
CAIS
239 male and 219 female students in USA
Intention to use
Lin et al., (2004)
Test and validate the motivational factors of using IM in Singapore
PACIS Conference
300 students at eight
faculties
Current usage
De Vos et al., (2004)
Investigate the motivational factors behind IM adoption in the workplace from
four months before to three months after
HICSS Conference
104 employees in
a Dutch knowledge
work organization
(28% female and 72%
male)
Current usage
Wang et al., (2005)
Propose a modified version TAM for
IM (MSN, Yahoo messenger and ICQ)
Journal of Information
Technology Theory
and Application
437 students (283 females and 154 males)
from several Taiwanese colleges
Current usage
Li et al., (2005)
Determinants of IM intention to use
(based on TAM)
Journal of Association for Information
Systems
273 undergraduate
students in two American universities
Intention to use
Cameron & Webster
(2005)
Investigation of IM usage in the workplace (why and type of use)
Computers in Human
Behavior
Interviews with 19
employees from four
organizations
Type of use
Huang et al., (2007)
Compare the use of IM versus e-mail
as the communication tool to facilitate
group problem-solving tasks for geographically dispersed teams
Computer Standards &
Interface
44 teams in the United
States
N/A
Peter et al., (2006)
Characteristics of individual adolescents who occasionally used webcams
and a microphone during IM for social
use.
Computers in Human
Behavior
1060 adolescents
between 12 and 17
years old
N/A
Leung (2006)
IM use and its impact on interpersonal
mediated communication (social use)
among teenagers and college students.
Telematics and Informatics
532 college students
in Hong Kong (China)
Current usage
Abla et al., (2006)
Description of software IM service for
collaboration
Fusion Engineering
and Design
Case study
N/A
Rau et al., (2006)
Impacts of using a combination of
mobile and Internet communication
technology (SMS, email and IM) in
education
Computers & Education
176 students
N/A
continued on following page
157
Factors Affecting Information Communication Technologies Usage and Satisfaction
Table A1. continued
Study
Authors
Objectives
Journal/ conference
Sample
Usage vs. satisfaction
Simon (2006)
Compare task performance and satisfaction among three communication
media: Face-to-face, IM and video
conferencing
Journal of Social Psychology
150 undergraduate
students (70 men and
80 women)
Current usage
Shen and Gallivan
(2006)
State why, how and what the consequences of IM use are
Americas Conference
on Information Systems
N.A
Current and satisfaction
Lancaster and Yen
(2007)
Investigate and compare the students’
perceptions of instant messaging and
e-mail by a sample of college students
based on mass richness theory
Information Management & Computer
Security
545 US college students
Perception about actual
usage
Rouibah (2008)
Investigate why Kuwaiti adults (employees) use IM in Kuwait
IT & People
191 adult employees
in Kuwait
Current usage
APPENDIX 2
Table a2. Arab countries: population size and number of Internet users (Internet World Stats 2008;
World Bank Development Indicator 2008)
Country
Country area (sq
km)
Population
Number of Internet users
Internet
penetration (%)
GDP (in 2007 in $
Billion)
1. Algeria
2,381,741
33,333,216
2,460,000 (Aug. 2007)
7.4
125.866
2. Bahrain
694
738,874
157,300 (Sept. 2007)
21.3
16.893
3. Comoros
1,862
711,417
21,000 (Sept. 2007)
3.0
0.436
4. Djibouti
23,200
496,374
11,000 (Sept. 2007)
2.2
0.834
5. Egypt
1,001,450
80,335,036
6,000,000 (Aug. 2007)
7.5
127.930
6. Iraq
434,128
27,162,627
6,000 (Oct. 2005)
0.1
45.239
7. Jordan
89,342
5,375,307
796,900 (Aug. 2007),
14.8
15.697
8. Kuwait
17,818
2,730,603
816,700 (Aug. 2007)
29.9
103.367
9. Lebanon
10,201
4,556,561
950,000 (Aug. 2007)
20.8
23.995
10. Libya
1,777,060
6,036,914
232,000 (Sept. 2007)
3.8
66.005
11. Mauritania
1,035,000
3,270,065
30,000 (Sept. 2007)
0.9
2.747
12. Morocco
458,730
33,757,175
6,100,000 (Sept. 2007)
18.1
72.764
13. Oman
309,500
2,452,234
319,200 (Aug. 2007)
13.0
40.523
14. Palestine
(West Bank and
Gaza)
6,242
3,070,228
266,000 (Sept. 2007),
8.7
5.320
15. Qatar
11,521
824,355
289,900 (Aug. 2007)
35.2
65.814
2,149,690
24,069,943
4,700,000 (Aug. 2007)
19.5
374.457
637,657
12,448,179
94,000 (Sept. 2007)
0.8
4.388
16. Saudi Arabia
17. Somalia
18. Sudan
2,505,810
39,379,358
3,500,000 (Sept. 2007)
8.9
46.708
19. Syria
185,180
19,514,386
1,500,000 (Aug. 2007),
7.7
37.414
continued on following page
158
Factors Affecting Information Communication Technologies Usage and Satisfaction
Table A2. continued
Country area (sq
km)
Population
Number of Internet users
Internet
penetration (%)
GDP (in 2007 in $
Billion)
20. Tunisia
163,610
10,276,158
1,618,440 (June 2007)
15.7
34.142
21. United Arab
Emirates
77,700
3,981,978
1,708,500 (Aug. 2007),
42.9
189.644
22. Yemen
528,076
21,306,342
270,000 (Aug. 2007)
1.3
22.738
13,806,212
335,827,330
31,846,940
-
1422.921
Country
Total
APPENDIX 3. MODEL MEASUREMENTS
Curiosity
CU1: Using IM excites my curiosity to get to know new friends.
CU2: Using IM programs excites my curiosity to get to know new people (from both genders).
CU3: IM programs help me to know new subjects and gain new knowledge.
CU4: IM programs help me to know new habits and cultures of other countries.
Prior Similar Experience (PSE)
PSE1: I am familiar with using computers.
PSE2: I am familiar with using the web.
PSE3: I am a familiar with using e-mail.
Compatibility
CP1: Using IM would be compatible with all aspect of my life with friends.
CP2: Using IM would be compatible with all aspect of my life with my family.
CP3: Using IM would be compatible with all aspect of my studies.
Perceived Usefulness
PU1: Using IM improves my performance.
PU2: Using IM increases my productivity.
PU3: Using IM would enhance my effectiveness.
Perceived Ease of Use
PEOU1: Interacting with IM does not require a lot of my mental effort.
PEOU2: I find IM flexible to interact with.
PEOU3: My interactions with IM are clear.
PEOU4: I find IM easy to use.
159
Factors Affecting Information Communication Technologies Usage and Satisfaction
Current Usage
UV (Usage Volume or daily usage): Time spent on the use of IM per day.
FU (Frequency of Use): Time per week a respondent reports using the IM.
VU (Variety of Use): Respondents were given a list of topics they may use during IM, and were asked
to select those they often fit their interest, including: social, cultural, political, love, study, family,
scientific, religion, medical, personal, unethical, sport and entertainment.
TUS (Type of Use & Synchronicity): to whom the user frequently communicates with and exchange
information in real-time (family members; friends with the same sex; friends with the opposite sex;
unknown person with the same sex; and unknown person with the opposite sex).
SP (Social Presence): Preference of respondents to an existing communication channel including: typing,
talking, visualizing pictures, or watching each other using digital camera.
Users’ Satisfaction of Instant Messaging
USIM1: My GPA has decreased since I started to use IM.
USIM2: IM usage has a positive impact on my study performance.
USIM3: IM usage makes me miss my classes.
USIM4: IM usage makes me overdue my homework’s and assignments.
USIM5: IM usage distracts my attention during class time.
USIM6: Use of IM helps me to finish my assignment on time.
USIM7: IM usage has enabled me to gain knowledge useful to my study.
USIM8: IM usage has enabled me to improve my GPA.
APPENDIX 4
Table A3. Results of factors analysis of loadings and cross loadings
PU╯
USIM
CU╯
PSE╯
PEOU╯
CP╯
╯Usage
magnitude
Socialization╯
CP1
0.13
0.10
0.23
0.22
0.16
0.76
0.14
0.06
CP2
0.15
0.07
0.07
0.14
0.14
0.84
0.05
0.14
Item
Compatibility
Curiosity
CP3
0.32
0.05
0.13
-0.02
0.07
0.70
0.02
0.01
CU1
0.16
0.15
0.83
0.05
0.18
0.09
0.11
0.11
CU2
0.13
0.15
0.84
0.07
0.21
0.12
0.07
0.09
CU3
0.33
0.13
0.70
0.07
0.15
0.20
0.16
0.06
CU4
0.37
0.01
0.31
0.09
0.16
0.09
0.04
-0.05
continued on following page
160
Factors Affecting Information Communication Technologies Usage and Satisfaction
Table A3. continued
Item
Prior Similar Experience
Usefulness
Ease of Use
PSE1
PU╯
USIM
CU╯
PSE╯
PEOU╯
CP╯
╯Usage
magnitude
Socialization╯
0.10
0.01
0.05
0.86
0.15
0.12
0.07
0.02
PSE2
0.08
-0.03
0.09
0.89
0.15
0.09
0.09
0.09
PSE3
0.12
0.01
0.08
0.82
0.21
0.07
0.17
0.05
PU1
0.64
0.03
0.35
0.25
0.17
0.09
0.12
0.10
PU2
0.78
0.09
0.22
0.15
0.22
0.13
0.01
0.11
PU3
0.73
0.12
0.25
0.14
0.24
0.06
0.05
0.14
PEOU1
0.15
0.12
0.07
0.20
0.70
0.01
0.08
0.04
PEOU2
0.21
0.05
0.23
0.12
0.79
0.13
0.06
0.05
PEOU3
0.28
0.12
0.20
0.08
0.74
0.16
0.12
0.07
PEOU4
0.17
0.02
0.24
0.24
0.75
0.15
0.17
0.03
Usage
magnitude
UV
0.08
0.05
0.16
0.06
0.11
0.06
0.13
0.81
FU
0.08
0.06
0.00
0.07
0.02
0.10
0.05
0.85
VU
0.17
0.04
0.04
0.14
0.11
-0.06
0.81
0.07
Socialization
TUS
0.03
0.01
0.14
0.15
0.16
0.00
0.77
0.10
SP
-0.01
-0.07
0.09
0.01
0.04
0.28
0.72
0.03
Users’ satisfaction
USIM1
0.03
0.77
0.20
-0.02
0.10
0.11
0.02
0.02
USIM2
0.46
0.48
0.04
0.08
0.11
0.24
0.10
-0.06
USIM3
0.15
0.86
0.07
-0.03
0.01
0.02
-0.01
0.03
USIM4
0.12
0.89
0.05
0.07
0.06
0.03
0.02
0.04
USIM5
0.14
0.83
0.03
-0.04
0.07
-0.03
-0.07
0.07
USIM6
0.29
0.29
0.04
-0.07
0.08
0.18
-0.01
-0.03
USIM 7
0.25
0.10
0.18
0.03
0.18
0.16
0.09
0.03
USIM 8
0.03
0.09
0.09
0.11
0.14
0.14
0.10
0.07
This work was previously published in International Journal of Global Information Management, Volume 17, Issue 2, edited
by Felix B. Tan, pp. 1-29, copyright 2009 by IGI Publishing (an imprint of IGI Global).
161
162
Chapter 8
The Information
Overload Paradox:
A Cross-Cultural Research Study
Ned Kock
Texas A&M International University, USA
Ana Rosa Del Aguila-Obra
University of Malaga, Spain
Antonio Padilla-Meléndez
University of Malaga, Spain
ABSTRACT
Information overload is one of the major problems for individuals and organizations in modern urban
environments. This phenomenon has been studied for many years, and has proven to be more complex
than researchers believed it to be. It is important to better understand this “pathology of information”
for two reasons. First, it has a deleterious impact on work productivity and quality. Second, it has traditionally been a driver of Information Technology developments aimed at helping individuals to better
cope with it. The study presented here aims at shedding light on the complexity of information overload,
by examining the relationship between perceived information overload intensity and three traditional
information overload predictors as well as one nontraditional predictor. The nontraditional predictor
is power distance, or the extent to which less powerful members of a national culture accept that power
is distributed unequally. Power distance was manipulated through the collection of data from 184 local managers and professionals (in New Zealand, Spain and the U.S.A.). The data analyses led to one
surprising conclusion: perceived information overload intensity seems to be more strongly related to
power distance than to the volume of written information or number of information transactions processed by an individual.
DOI: 10.4018/978-1-60960-605-3.ch008
Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
The Information Overload Paradox
INTRODUCTION
Typically information overload has been viewed
as problematic and as leading to losses in work
productivity and quality of work outcomes. Information overload has also traditionally been a
driver of information technology developments;
information technology solutions have often
been developed specifically to help individuals
cope with information overload (Turetken &
Sharda, 2004). As early as 1970, the award winning writer and futurist Alvin Toffler argued that
information overload would become one of the
main problems facing modern society and organizations (Toffler, 1970). The emergence of the
Internet and its increasing use by individuals and
organizations (Chung & Tan, 2004; Teo, 2007) has
been presented as contributing to increasing levels
of information overload (Kiley, 1995; Berghel,
1997), apparently lending support to Toffler’s
(1970) prediction.
Generally speaking, information overload
can be defined as a state in which the amount of
information an individual must process exceeds
the individual’s information processing resources
(O’Reilly, 1980).
The study presented here aims at shedding light
on the complexity of the information overload
phenomenon, by looking at it from a different
and arguably novel lens. The study examined
the relationship between perceived information
overload intensity and one nontraditional and
three traditional information overload predictors.
One of the goals of the study was to compare the
influence, if any, of the nontraditional predictor
against the more traditional ones.
Power distance was the nontraditional predictor. Power distance is part of Geert Hofstede’s
model of cultural dimensions (Hofstede, 2001;
Lippert & Volkmar, 2007), and is defined as the
extent to which less (and more) powerful members
of organizations (e.g., employees and their supervisors) accept that power is distributed unequally.
Hofstede’s model has been widely applied in
various disciplines (Sivakumar & Nakata, 2001),
particularly in psychology, sociology, marketing,
and management (Søndergaard, 1994). Hofstede’s
model is considered by many to be the most
comprehensive model of the influence of national
culture on organizational behavior.
The traditional information overload predictors
were the volume of written information processed
by individuals, in terms of pages read and written on a daily basis; the number of information
transactions, or the average number of information
giving and information receiving interactions per
working day; and business process knowledge,
assessed as the number of months of formal education and hands-on practice needed to perform
work-related activities well.
This is a cross-cultural research study (Hunter,
2006) in which variations in the power distance
predictor were incorporated into the study’s design
through the collection of data from three different countries, which also differ significantly in
terms power distance. Data were obtained from
184 local managers and professionals in New
Zealand, Spain and the U.S.A. The data were
analyzed employing partial least squares-based
structural equation modeling. The analysis led to
one surprising conclusion, which is that perceived
information overload intensity is more strongly
related to power distance than to the volume of
written information or the number of information
transactions processed by an individual. This
surprising conclusion is referred to here as the
information overload paradox.
RESEARCH BACKGROUND
AND HYPOTHESES
Much research on information overload has taken
the view that the phenomenon has both organizational and societal implications (Edmunds &
Morris, 2000; Jones et al., 2004), and that those
implications are more negative than positive.
More frequently than not, information overload is
163
The Information Overload Paradox
perceived as an undesirable occurrence that must
be either minimized or completely eliminated
(Losee, 1989; Glazer et al., 1992; Meyer et al.,
1997). In many cases, new technology-based
information visualization and summarization
solutions are seen as among the key solutions
to the problems posed by information overload
(Johnsson, 1991; Turetken & Sharda, 2004).
However, many of those information processing
aids, such as electronic collaboration technologies,
can indirectly lead to an overall increase in the
amount of information that an individual has to
process (Schultze & Vandenbosch, 1998; Allen
& Shoard, 2005). For example, Mano and Mesch
(2010) report undesirable effects on performance
related with an excessive e-mail use.
Complicating the picture further is the fact
that rarely information overload is measured
directly and objectively. Usually information
overload can only be measured indirectly and
subjectively, through perceptions of information
overload. Many attempts to measure information
overload directly and objectively fall into the trap
of measuring it through effects that are caused
by information overload. Decision quality, for
example, is generally expected to go down as
information overload goes up (Eppler & Mengis, 2004). Yet, using decision quality as a proxy
measure for information overload would conflate
cause and effect.
When one looks at communication at work on
a daily basis, written information exchanges such
as those through electronic messages and printed
documents seem to be the ones used to convey high
volumes of information; with lower volumes transferred through unwritten information exchanges,
such as oral and telephone exchanges (Wainfan
& Davis, 2004). These patterns are probably motivated by choices that have a cognitive basis, as
argued by Robert and Dennis’s (2005) cognitive
model of media choice. For example, unwritten
information exchanges pose difficulties for the
time-disconnected review of complex information, which may be necessary for the effective
164
understanding of work-related activities. Hypothesis H1 below incorporates the notions that high
information volumes more often than not cause
information overload, and that high information
volumes are more likely to be exchanged through
written than unwritten media.
H1: The volume of written information processed
at work on a daily basis is significantly and
positively related to perceived information
overload intensity.
As can be inferred from hypothesis H1, this
study is particularly concerned with information
overload as perceived by workers. Arguably
perceived information overload is not the same
as actual information overload. Nevertheless, it
is reasonable to conclude based on past research
that the two variables are high correlated, and
that negative effects such as the deterioration of
decision making abilities are often associated
with perceptions of elevated information overload
(Edmunds & Morris, 2000; Eppler & Mengis,
2004; Evaristo, 1993).
While written information exchanges at work
are associated with high volumes of information
transfer, both written and unwritten information
exchanges seem to be associated with a high
number of information transfer transactions
(Chervany & Dickson, 1974; Evaristo, 1993). A
simple scenario can help illustrate this point. Let
us consider a lawnmower sales representative
who has received a manual about lawnmowers,
containing complex information that he must
understand in order to do an effective job as a
salesperson. He then sends 3 brief e-mails and
places 5 short telephone calls over a week to his
manager, with the goal of clarifying parts of the
lawnmowers’ manual. In this example, a large
written information exchange (the receipt of the
lawnmowers’ manual) leads to several short and
fragmented written and unwritten information
exchanges. This discussion highlights the difference between volume of information processed,
The Information Overload Paradox
and number of information transactions. Both are
likely to be associated with information overload
(not only the first), because they are both associated
with information exchanges that use up workers’
cognitive and time resources (Speier et al., 1997).
This is incorporated into hypothesis H2, which
is stated below.
H2: The number of information transactions taking
place at work on a daily basis is significantly
and positively related to perceived information overload intensity.
The information overload literature suggests
that individuals employ coping mechanisms
when facing information overload, and that those
mechanisms present a certain degree of uniformity.
Jones et al.’s (2004) study shows that individuals
experiencing information overload in online interaction environments are more likely to respond
to and generate simple messages, as opposed to
complex messages. This could potentially have a
negative overall effect on the outcomes of tasks
performed by the individuals, and illustrates the
fact that not all coping mechanisms are likely to
be effective ones in the long run.
There is also evidence that certain business
process characteristics are associated with reasonably effective information overload coping mechanisms, especially when one looks at those types
of mechanisms from an individual competence
perspective (Lindgren, Henfridsson & Schultze,
2004). One example of this is the amount of
knowledge viewed as necessary to perform work
well in connection with a business process (Iselin,
1989). The rationale here is that properly trained
and more experienced individuals are better able
to cope with information overload (Camerer &
Johnson, 1991; Chervany & Dickson, 1974), and
that the explicit recognition that individuals must
be appropriately trained to perform work well
contributes to that proper training. This leads to
the conclusion that the amount of business process
knowledge necessary to perform work well is
inversely correlated with perceived information
overload. The conclusion is incorporated into
hypothesis H3, formally stated below.
H3: The amount of business process knowledge
necessary to perform work well is significantly and negatively related to perceived
information overload intensity.
Kock’s (2000) research suggests one additional
predictor of perceived information overload that
is arguably associated with what Hofstede (2001)
refers to as power distance, or the extent to which
less (and more) powerful members of a national
culture accept that power is distributed unequally.
The majority of respondents in Kock’s (2000)
study, when asked for the main reasons why
they experienced information overload at work,
indicated that pressure placed on them to perform
tasks efficiently and effectively was the main reason. This finding is consistent with other studies
that linked information overload perceptions with
time pressure experienced by individuals (see,
e.g., Evaristo, 1993; Schick et al., 1990; Paul &
Nazareth, 2010). Geert Hofstede has conducted
extensive studies of organizational and national
cultures (Hofstede, 1980; 1991; 2001), and has
identified power distance as being likely associated with stress at work. The reason seems to be
that individuals who are more accepting of power
differentials in organizations are also likely to feel
more time pressure to perform work well. (Nearly
all individuals in organizations have to report to
other individuals, which makes power differentials
ubiquitous.) This conjecture, combined with the
notion that information overload is associated with
time pressure, leads to hypothesis H4.
H4: Power distance is significantly and positively
related to perceived information overload
intensity.
The link predicted through hypothesis H4 is
arguably more counterintuitive than the links
165
The Information Overload Paradox
predicted through hypotheses H1, H2, and H3.
It also contradicts some evidence suggesting that
societies that present high power distance are also
polychromic (Hall, 1976; Waarts & van Everdingen, 2006). In polychromic societies individuals
pay less attention to time constraints in businessrelated activities than in monochromic societies.
It should be noted, however, that the evidence
in question has not suggested the existence of a
causal relationship between power distance and
the extent to which a society is polychromic.
These two constructs may be independent from
each other.
A simple work-related example can be used to
illustrate the link predicted through hypothesis H4.
Let us assume a plant manager who is overloaded
with information about the plant’s operation; information that she needs to make various decisions
on a daily basis. The manager may experience a
reduction in the amount of information overload
by employing a decision support system that
automatically summarizes that information for
her based on a pooling of information from other
systems. Yet, the benefits of that reduction may
be lost due to an increase in the time pressures
placed on her by other managers to whom she
reports. She may be asked to provide additional
information to those managers, who are higher
up in the organization’s hierarchy, and who may
be more demanding because of their knowledge
of the existence of the decision support system.
Those extra information processing demands
may be amplified by a significant power distance,
causing an overall increase in perceived information overload intensity experienced by the plant
manager.
RESEARCH METHOD
Data from 184 local managers and professionals
were collected in New Zealand, Spain and the
U.S.A. In terms of gender distribution, 70.7%
of the respondents were males. The respondents’
166
distribution in terms of economic sector was approximately as follows: service (87%), manufacturing (12.5%), and agriculture (0.5%). In terms
of distribution by country, 26.6% of the respondents were from New Zealand, 41.3% were from
Spain, and 32.1% were from the U.S.A. Generally
the respondents were evenly distributed across
countries in terms of gender and economic sector.
The managers and professionals were participating in executive education programs in their
respective countries at the time the data were collected. The data were collected through questionnaires distributed in class by research collaborators
who taught in the executive education programs.
The questionnaires used in the three countries
had the same questions. The questions make up
a measurement instrument that was previously
used and validated in the single-country study
reported by Kock (2000). Those used in Spain
were first translated to Spanish. In order to check
the correctness of the translation, the questionnaire in Spanish was back-translated to English
by a second translator, and then compared with
the original questionnaire in English by a third
translator. Questions included both perceptual
and demographic items, and referred primarily
to work activities conducted on a regular basis
by those individuals.
The analysis of the data was conducted through
variance-based structural equation modeling employing the partial least squares method (Chin,
1998; 2001; Chin et al., 1996). It included independent variables, of which three were defined
as latent variables and measured through two
indicators. Unlike traditional reflective measures
(Chin, 1998; 2001), each indicator refers to a
different facet of the same underlying construct,
which justifies the use of only two (as opposed
to many) indicators (see, also, Kline, 1998). The
analysis also included control variables. The main
dependent variable of the structural model was
information overload intensity. These variables as
well as brief related definitions and measurement
descriptions are shown in Table 1.
The Information Overload Paradox
Table 1. Main variables of this study
Independent variables
Definition / measurement
Volume of information
processed
Average number of pages read and written per working day
Number of information
transactions
Average number of information giving and information receiving interactions per working day
Business process
knowledge
Number of months of formal education and hands-on practice needed to perform work-related activities well
Power distance
The extent to which less powerful members of a national culture accept that power is distributed
unequally
Control variables
Definition / measurement
Gender
Gender (male or female) of the individual participating in the study
Sector
Economic sector (service, manufacturing, or agriculture) of the organization to which the individual
belongs
Business process
complexity
Average number of relevant work-related decisions made and different work activities performed
made on a working day
Decision making
style
Decision making style in terms of scope (focus on details or big picture) and rationality (intuitive or
rational)
Dependent variable
Definition / measurement
Information overload
intensity
Perceived information overload intensity experienced at work
The three independent latent variables were
volume of information processed, number of
information transactions, and business process
knowledge. One of the independent variables,
namely power distance, was assigned a specific
score for each country. The scores were taken
from Hofstede’s (2001) cultural dimensions
model, and were the following: 22 for New Zealand, 57 for Spain, and 40 for the U.S.A. That is,
among the three countries New Zealand was assigned the lowest power distance score, followed
by the U.S.A., and then Spain with the highest
power distance score. Two of the control variables
were demographic variables: gender of the respondent and economic sector. The other two
control variables were perception-based variables:
business process complexity and decision making
style.
Hofstede’s (2001) cultural dimensions model
comprises power distance scores that have been
obtained primarily based on data collected from
employees of a single organization (i.e., IBM)
with offices in various countries. This may be
seen as detrimental to the external validity of those
scores, and their use as country scores. Nevertheless, one could also argue that the differences in
power distance proposed by Hofstede among
various countries provide a conservative measure
of the actual differences in power distance among
those countries. The reason is that being part of
the same multinational organization (i.e., IBM)
would tend to reduce variation of power distance
measures obtained from different countries (Hofstede, 2001; Mani, 2007); that is, the multinational
organization’s management-employee interaction
rules and procedures could be expected to have
a uniformity-inducing effect on power distance
perceptions, even if a small one. This rationale
suggests that the power distance scores proposed
by Hofstede can be used in studies that hypothesize effects of score differences on endogenous
variables, as this study does, as opposed to studies
that hypothesize null effects based on similarities
in scores (Bing, 2004; Chiang, 2005).
167
The Information Overload Paradox
Figure 1. Hypotheses shown as paths in a structural equation model
Figure 1 shows the hypotheses as part of a
structural equation model. Each hypothesis is
shown next to a path in the model, together with
the expected sign of the relationship represented
through the path. That is, a “+” sign indicates
a positive predicted relationship between two
variables in the model, and a “-” sign indicates a
negative predicted relationship. The control variables are not represented because no predictions
were made regarding their relationship with other
variables; that is, their inclusion in the analysis
was exploratory. In the structural equation model,
the control variables were added as variables that
actually pointed to the main dependent variable of
the model, but with no a-priori predicted relationships with that variable.
In Figure 1, three variables are indicated as
intuitive predictors of the variable information
overload intensity. Those variables are volume of
information processed, number of information
transactions, and business process knowledge.
They are indicated as such because it is intuitive
to think of them as contributing to increase (the
first two) or decrease (the third) information
overload intensity. Power distance, however, is
indicated as a counterintuitive predictor because
168
its relationship to information overload intensity
is less direct and obvious.
VALIDATION OF THE
MEASUREMENT MODEL
Structural equation modeling results can generally be considered trustworthy if the underlying
measurement model passes some validation tests.
Those tests normally address reliability and validity concerns (Kline, 1998; Rosenthal & Rosnow,
1991). Validity tests come in many forms, and the
most widely used are convergent and discriminant
validity tests. When used in combination and in
the context of questionnaire-based studies, reliability and validity tests allow for the conclusion
that respondents generally: (a) answered questions
in a careful way, as opposed to answering in a
hurried and mindless manner; (b) understood the
questions in the way that they were intended to be
understood by the researcher who designed the
questions; and (c) understood the questions in the
same way as the other respondents.
Reliability tests are conducted through the
calculation of reliability coefficients, such as composite reliability coefficients, and their comparison
The Information Overload Paradox
Table 2. Factor loadings and composite reliabilities
Infvol1
Infvol
Inftran
Knowle
CR
.78
.26
-.03
.712
Infvol2
.85
.12
.07
Inftran1
.14
.84
.14
Inftran2
.19
.82
.13
Knowle1
.03
.10
.91
Knowle2
.02
.20
.90
.940
.953
Notes:
Infvol = Volume of information processed
Inftran = Number of information transactions
Knowle = Business process knowledge
CR = Composite reliability
with a certain threshold coefficient (Nunnally,
1978; Nunnally & Bernstein, 1994). Similarly,
convergent validity tests rely on the calculation
of factor loadings, and the comparison of those
loadings against a threshold loading (Hair et al.,
1987). Finally, discriminant validity tests rely
on the calculation of correlations between latent
variables and the comparison of those correlations
with average variances extracted for each latent
variable (Fornell & Larcker, 1981).
Table 2 displays factor loadings, cross-loadings, and composite reliability coefficients for the
three latent variables in the structural equation
model. The loadings and cross-loadings are shown
under the columns labeled “Infvol” (volume of
information processed), “Inftran” (number of information transactions), and “Knowle” (business
process knowledge). The extraction method used
in the calculation of the loadings and cross-loadings was principal components, and the rotation
method was varimax (Ehremberg, & Goodhart,
1976; Thompson, 2004). Those loadings that
were expected to be conceptually associated
with the respective latent variables are shown in
shaded cells. In the column indicated as “CR”
the composite reliability coefficients calculated
for each of the latent constructs are shown also
in shaded cells.
For a measurement model to be considered as
presenting an acceptable level of convergent
validity certain factor loadings must be .5 or above;
namely the loadings of indicators expected to be
conceptually associated with their respective latent
variables (Hair et al., 1987). Those are the loadings shown in shaded cells in Table 2, which range
from .78 to .91, and therefore meet the criterion
for acceptable convergent validity.
Similarly to the convergent validity criterion,
a measurement model is generally considered to
present an acceptable level of reliability if the
composite reliability coefficients are equal to or
greater than a certain threshold. Normally, that
threshold is expected to be .7 (Fornell & Larcker,
1981; Nunnally, 1978). As can be seen from the
“CR” column in Table 2, the composite reliability
coefficients ranged from .712 to .953, which allows
for the conclusion that the reliability of the measurement model used in this study is acceptable.
With the model validated in terms of convergent
validity and reliability, the next step is to assess
the model’s discriminant validity. Table 3 displays
several figures that can be used in this assessment.
The correlation coefficient calculated for each pair
of latent variables is shown in the cell where each
pair of variables cross each other, and the square
root of the average variance extracted (also known
169
The Information Overload Paradox
Table 3. Correlations and square roots of AVEs
€€€€€€Infvol
€€€€€€Inftran
€€€€€€Infvol
€€€€€€(.741)
€€€€€€Inftran
€€€€€€.314
€€€€€€(.941)
€€€€€€Knowle
€€€€€€.045
€€€€€€.278
€€€€€€Knowle
€€€€€€(.954)
Notes:
Infvol = Volume of information processed
Inftran = Number of information transactions
Knowle = Business process knowledge
Coefficients shown in cross-cells are correlations
Square roots of average variances extracted (AVEs) are shown in diagonal
as AVE) for each of the latent variables is shown
in the diagonal within parentheses.
One can consider a measurement model to
present an acceptable level of discriminant validity if the square root of the average variance extracted for each latent variable is above all values
belonging to a certain set of values. Those values
are the correlations between the latent variable in
question and all of the other latent variables in
the measurement model (Fornell & Larcker, 1981).
As can be inferred from Table 3, all of the square
roots of the average variances extracted are
higher than the correlations shown below them
or to their left. (Those are all of the correlations
involving each latent variable.) Therefore, one
can conclude that the discriminant validity of the
measurement model is acceptable.
The above discussion suggests that the measurement model employed in this study meets the
general criteria that allow for the conclusion that
it is both reliable and valid. This is a precondition for the interpretation of the results of the
structural equation modeling analysis, discussed
below. Based on the results of the validity and
reliability tests, one can conclude that the level
of measurement errors found in the data is likely
to be low enough that the results of the analysis
can be used for a credible assessment of the hypotheses guiding the study.
170
DATA ANALYSIS RESULTS
Figure 2 summarizes the results of the structural
equation modeling analysis conducted to test the
hypothesized relationships proposed earlier. Arrows represent relationships between variables,
and β coefficients associated with each relationship
are shown near the arrows. Also near the arrows,
and below the β coefficients, are the hypotheses
associated with the relationships.
The β coefficients are the estimated path coefficients associated with each of the hypothesized
relationships. The symbol “*” follows the β coefficients associated with relationships significant
at the .05 level; and the symbol “**” at the lower
and more stringent .01 level. The letters “NS”
stand for “not significant” and are shown in place
of the β coefficient for one statistically insignificant relationship. The actual sigma associated
with that relationship was .16, which is much
higher than the .05 threshold generally used for
hypothesis testing.
Figure 2 suggests, somewhat surprisingly, that
the volume of information processed by individuals, in terms of pages read and written on a daily
basis, is unrelated to the perceived information
overload intensity experienced at work by those
individuals. The number of information transactions, or the average number of information giving
and information receiving interactions per working
day, was found to be significantly and positively
The Information Overload Paradox
Figure 2. Path coefficients in the structural equation model and related hypotheses
related to perceived information overload intensity
(β=.137, P<.05).
Business process knowledge, assessed as
the number of months of formal education and
hands-on practice needed to perform work-related
activities well, was found to be significantly and
negatively related to perceived information overload intensity (β=-.114, P<.05). That is, business
processes that require more knowledge acquisition
time and effort to be executed effectively seemed
to be associated with lower levels of perceived
information overload intensity.
Finally, power distance, or the extent to which
less powerful members of a national culture accept
that power is distributed unequally, was found to
be significantly and positively related to perceived
information overload intensity (β=.192, P<.01). In
terms of the strength of the hypothesized relationships, this relationship between power distance
and perceived information overload intensity was a
great deal stronger than the relationships between:
number of information transactions and perceived
information overload intensity (approximately
40% stronger), and business process knowledge
and perceived information overload intensity
(approximately 69% stronger).
As discussed earlier, the following control
variables were included in the structural equation modeling analysis as possible predictors of
perceived information overload intensity: gender
of the respondent, economic sector (service,
manufacturing, or agriculture) of the organization
to which the individual belongs, business process
complexity (measured through the number of
work-related decisions and different work activities on a working day), and decision making style
in terms of scope (focus on details or big picture)
and rationality (intuitive or rational). None of these
control variables were found to have a significant
relationship with perceived information overload
intensity.
DISCUSSION
Not only is perceived information overload
strongly related to power distance, according to
the structural equation modeling analysis results,
but that relationship appears to be stronger than
the relationship between perceived information
overload and other more intuitive predictors.
Among those more intuitive predictors are the
171
The Information Overload Paradox
volume of written information and number of information transactions processed on a daily basis.
The volume of written information processed on
a daily basis seems to be unrelated to perceived
information overload. This is rather surprising,
since it is reasonable to assume that most of the
information flow in organizations takes place
through written media.
Moreover, the relationship between power distance and perceived information overload intensity
appears to be 40% stronger than the relationship
between number of information transactions and
perceived information overload intensity. This
observation comes from the comparison of the
absolute path coefficients in the structural equation model, assuming that their absolute values
reflect the strength of their associated two-variable
relationships.
The above findings constitute the essence
of what is referred to here as the information
overload paradox, which is also reflected in the
summary of the hypothesized predictions provided
in Table 4. The fact that hypothesis H1 was not
supported by the analysis results indicates that the
associated prediction was incorrect. It nevertheless lends additional support for the information
overload paradox idea. That is, volume of written
information processed on a daily basis, which
likely correlates highly with the total amount of
information processed daily, does not matter much
in terms of information overload perceived by the
individuals processing the information.
Put simply, the paradox entails the conclusion
that perceived information overload is more
strongly related to power distance than to the
volume of written information or number of information transactions processed by an individual. This paradoxical conclusion follows from the
above findings, and is likely to serve the basis for
further research, either with the goal of disproving
it, or to further our understanding of the information overload phenomenon. The conclusion suggests that our understanding of the phenomenon
is rather limited, and in dire need of expansion
and refinement.
The importance assigned to the above conclusion assumes that perceived information overload
is a good proxy for actual information overload,
given the data collection and analysis methods
employed in this study. While it is very likely that
perceived information overload is not the same
as actual information overload, it is reasonable to
conclude that the two variables are highly correlated, and that the negative effects of information
overload are often associated with perceptions of
information overload (Edmunds & Morris, 2000;
Eppler & Mengis, 2004; Evaristo, 1993). Actual,
as opposed to perceived, information overload is
difficult to measure objectively (Evaristo, 1993;
Kock, 2000).
While the information overload paradox does
go against much of the literature on information
overload, it is consistent with the belief that perceived pressure placed on an individual to perform
work-related tasks efficiently and effectively is
Table 4. Support for the hypotheses based on the results
Hypothesis
Supported?
H1: The volume of written information processed at work on a daily basis is significantly and positively related to
perceived information overload intensity.
No
H2: The number of information transactions taking place at work on a daily basis is significantly and positively
related to perceived information overload intensity.
Yes
H3: The amount of business process knowledge necessary to perform work well is significantly and negatively
related to perceived information overload intensity.
Yes
H4: Power distance is significantly and positively related to perceived information overload intensity.
Yes
172
The Information Overload Paradox
one of the main reasons why an individual experiences information overload (Kock, 2000; Schick
et al., 1990). Given this, one can reasonably infer
that perceived time pressure to accomplish workrelated tasks has a significant and negative impact
on the cognitive abilities of an individual, and
thus a similar impact on decision making quality.
As such, work-related pressure may be viewed
as having a relationship with decision making
quality that is similar to that described by the
information overload inverted U-curve (Eppler
& Mengis, 2004), but with the magnitude of the
effect amplified by power distance.
As far as the control variables are concerned,
one noteworthy result is that both males and females seem to experience information overload
equally. Also, the insignificant relationship between business process complexity and perceived
information overload is consistent with the significant and negative relationship between business
process knowledge and perceived information
overload. That is, given the latter relationship, one
would expect business process complexity to be
met with better preparation by workers in terms
of knowledge acquisition, and thus no related
increase in information overload.
Several of the results in connection with the
control variables go against much of the information overload literature. The insignificant
relationship between decision making style and
perceived information overload intensity goes
against Kock’s (2000) suggestion that decisional
scope (i.e., whether one focuses on details or the
big picture when making decisions) is significantly
associated with information overload. The insignificant relationship between economic sector and
perceived information overload intensity goes
against the notion that volume of information
processed affects information overload (Eppler &
Mengis, 2004), although in a somewhat indirect
way. The reason is that the amount of information processed in service organizations is generally considered to be significantly higher than
in manufacturing organizations (Davenport &
Prusak, 2000; Drucker, 1995). This insignificant
relationship is nevertheless consistent with the
other findings of this study.
MAIN LIMITATIONS
As mentioned earlier, the managers and professionals who provided the data for this study were
participating in executive education programs
in their respective countries at the time the data
were collected. Because of the nature of those
programs, as well as the recruitment approaches
employed, the vast majority of the participants
were probably long-term residents of and in many
cases native to the countries in which the data were
collected. However, this cannot be ascertained
for all of the participants because data about their
nationality were not collected. Arguably a variation in within-country nationality could have led
to corresponding within-country variability in
the answers provided by the participants. Given
that lower within-country variability in answers
would have generally led to stronger effects in
our model, it is safe to conclude that our main
findings do not have to be revised. Nevertheless,
future studies aimed at replicating the findings
of this investigation should include one or more
questions related to the respondent’s nationality. A
related limitation of this study is that the data does
not differentiate between bosses and subordinates,
nor contains specific details about each group.
CONCLUSION
The phenomena addressed through information
systems research often appear simple at first, only
to reveal inner complexities upon closer examination. The identification of paradoxical results is one
of the many ways in which research progress can
be achieved, and the complexity of information
systems phenomena unraveled (Pearlson & Saunders, 2001; Whitley & Darking, 2006). Underlying
173
The Information Overload Paradox
this study is the belief that information overload
is a complex and misunderstood phenomenon.
As such it needs to be investigated from different
perspectives, such as in investigations involving
different organizational cultures, and the various
results triangulated so that the multiple facets
of the phenomenon can be properly grasped. It
was assumed in this study that the nature of the
interaction between individuals in organizations
may significantly influence information overload.
Predictions were made here about information
overload that incorporate traditional as well as
nontraditional views of the phenomenon. Among
the nontraditional views is the one that certain
organizational cultures may incorporate unstated
rules of interaction that may lead to perceived pressures on individuals, which are in turn reflected
in information overload perceptions.
Contrary to predictions, the volume of written
information processed by the individuals was found
to be unrelated to perceived information overload
intensity experienced at work. The number of
information transactions and business process
knowledge were found to be significantly related
to perceived information overload intensity; the
former positively and the latter negatively related.
Power distance was found to be significantly
and positively related to perceived information
overload intensity. Surprisingly, this relationship
was found to be much stronger than the other
relationships, and its contribution to explaining
the variance of the variable perceived information overload was found to be greater than the
combined contribution of the other two significant
relationships. As for the control variables, none of
them was found to have a significant relationship
with perceived information overload intensity.
The findings of this study are paradoxical
in light of the existing research literature on
information overload, as well as in light of commonsense expectations. It is intuitive to assume
that the volume of information and number of
information transactions processed should be more
strongly related to information overload than any
174
other possible variable; this assumption follows
directly from the definition of information overload (Eppler & Mengis, 2004; O’Reilly, 1980).
Yet, this study found support for the notion that
perceived information overload is more strongly
related to power distance than to the volume of
written information and number of information
transactions processed by an individual. This
conclusion notion is referred to here as the information overload paradox.
The results of this study are counterintuitive
enough to serve as the basis for future research
that will not only replicate but expand on the current research. Hopefully that will lead to a more
comprehensive theoretical understanding of the
information overload phenomenon. Such understanding is very much needed, given information
overload’s increasing importance as a societal and
organizational phenomenon. The negative effects
of information overload on work productivity and
quality are primarily what make it an increasingly
important phenomenon. Nevertheless, a better
understanding of its underlying causes holds an
additional potential promise. That understanding
can also shed light on important mechanisms of
the human mind, which is still one of the most
mysterious and complex research realms.
ACKNOWLEDGMENT
The authors would like to thank the Spanish, Kiwi
and American managers and professionals who
participated in this study. Thanks are also due to
John Buchanan for his help in the collection of part
of the data used in this analysis, and for previous
discussions in connection with the information
overload phenomenon from a decision making
perspective.
The Information Overload Paradox
REFERENCES
Allen, D. K., & Shoard, M. (2005). Spreading
the load: Mobile information and communications technologies and their effect on information
overload. Information Research, 10(2) paper 227.
Retrieved from http://InformationR.net/ir/10-2/
paper227.html
Bawden, D., & Robinson, L. (2009). The
dark side of information: Overload, anxiety
and other paradoxes and pathologies. Journal of Information Science, 35(2), 180–191.
doi:10.1177/0165551508095781
Berghel, H. (1997). Cyberspace 2000: Dealing
with information overload. Communications of the
ACM, 40(2), 19–24. doi:10.1145/253671.253680
Bing, J. W. (2004). Hofstede’s consequences: The
impact of his work on consulting and business
practices. The Academy of Management Executive,
18(1), 80–88. doi:10.5465/AME.2004.12689609
Chin, W. W., Marcolin, B. L., & Newsted, P. R.
(1996). A partial least squares latent variable
modeling approach for measuring interaction
effects: Results from a Monte Carlo simulation
study and voice mail emotion/adoption study. In J.
I. DeGross, S. Jarvenpaa, & A. Srinivasan (Eds.),
Proceedings of the 17th International Conference
on Information Systems (pp. 21-41). New York,
NY: The Association for Computing Machinery.
Chung, J., & Tan, F. B. (2004). Antecedents of
perceived playfulness: An exploratory study on
user acceptance of general information-searching
websites. Information & Management, 41(7),
869–882. doi:10.1016/j.im.2003.08.016
Davenport, T., & Prusak, L. (2000). Working
knowledge. Cambridge, MA: Harvard Business
School Press.
Drucker, P. F. (1995). The information executives
truly need. Harvard Business Review, 73(1),
54–62.
Camerer, C. F., & Johnson, E. J. (1991). The
process-performance paradox in expert judgment.
In Ericsson, K. A., & Smith, J. (Eds.), Toward a
general theory of expertise (pp. 195–217). Cambridge, MA: Cambridge University Press.
Edmunds, A., & Morris, A. (2000). The problem
of information overload in business organisations:
A review of the literature. International Journal of Information Management, 20(1), 17–28.
doi:10.1016/S0268-4012(99)00051-1
Chervany, N., & Dickson, G. (1974). An experimental evaluation of information overload in a
production environment. Management Science,
20(10), 1335–1344. doi:10.1287/mnsc.20.10.1335
Ehremberg, A. S. C., & Goodhart, G. J. (1976).
Factor analysis: Limitations and alternatives.
Cambridge, MA: Marketing Science Institute.
Chiang, F. (2005). A critical examination of Hofstede’s thesis and its application to international
reward management. International Journal of
Human Resource Management, 16(9), 1545-1563.
Chin, W. W. (1998). Issues and opinion on structural equation modeling. Management Information
Systems Quarterly, 22(1), vii–xvi.
Chin, W. W. (2001). PLS-Graph user’s guide.
Houston, TX: Soft Modeling Inc.
Eppler, M., & Mengis, J. (2004). The concept of
information overload: A review of literature from
organization science, accounting, marketing, MIS,
and related disciplines. The Information Society,
20(5), 325–344. doi:10.1080/01972240490507974
Evaristo, R. (1993). An empirical investigation
of the impact of information characteristics and
information technology on individual information load. Ph.D. Dissertation. Minnesota, MN:
University of Minnesota.
175
The Information Overload Paradox
Foley, J. (1995, October 30). Managing information - infoglut - new tools can help tame an ocean
of data. InformationWeek, pp. 30-33.
Johnsson, J. (1991). Information overload: CEOs
seek new tools for effective decision making.
Hospitals, 65(20), 24–28.
Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. JMR,
Journal of Marketing Research, 18(1), 39–50.
doi:10.2307/3151312
Jones, Q., Ravid, G., & Rafaeli, S. (2004). Information overload and the message dynamics of online
interaction spaces: A theoretical model and empirical exploration. Information Systems Research,
15(2), 194–211. doi:10.1287/isre.1040.0023
Glazer, R., Steckel, J. H., & Winer, R. S. (1992).
Locally rational decision making: The distracting effect of information on managerial performance. Management Science, 38(2), 212–227.
doi:10.1287/mnsc.38.2.212
Kiley, K. (1995). The cyberspace database information overload. Catalog Age, 12(9), 56–59.
Hair, J. F., Anderson, R. E., & Tatham, R. L.
(1987). Multivariate data analysis. New York,
NY: Macmillan.
Hall, E. T. (1976). Beyond culture. Garden City,
NY: Anchor Press.
Hofstede, G. (1980). Culture’s consequences:
International differences in work-related values.
Beverly Hills, CA: Sage.
Hofstede, G. (1991). Cultures and organizations: Software of the mind. London, England:
McGraw-Hill.
Hofstede, G. (2001). Culture’s consequences:
Comparing values, behaviors, institutions, and
organizations across nations. Thousand Oaks,
CA: Sage.
Hunter, M. G. (2006). Experiences conducting
cross-cultural research. Journal of Global Information Management, 14(2), 75–89. doi:10.4018/
jgim.2006040104
Iselin, E. (1989). The impact of information
diversity on information overload effects in
unstructured managerial decision making. Journal of Information Science, 15(3), 163–174.
doi:10.1177/016555158901500304
176
Kline, R. B. (1998). Principles and practice of
structural equation modeling. New York, NY:
The Guilford Press.
Kock, N. (2000). Information overload: A
process-centered view. Knowledge and Process
Management, 7(4), 256–264. doi:10.1002/10991441(200010/12)7:4<256::AIDKPM79>3.0.CO;2-U
Kock, N., Del Águila-Obra, A. R., & PadillaMeléndez, A. (2009). The information overload
paradox: A structural equation modeling analysis
of data from New Zealand, Spain and the U.S.A.
Journal of Global Information Management,
17(3). doi:10.4018/jgim.2009070101
Lindgren, R., Henfridsson, O., & Schultze, U.
(2004). Design principles for competence management systems: A synthesis of an action research
study. Management Information Systems Quarterly, 28(3), 435–473.
Lippert, S. K., & Volkmar, J. A. (2007). Cultural
effects on technology performance and utilization:
A comparison of U.S. and Canadian users. Journal
of Global Information Management, 15(2), 56–90.
doi:10.4018/jgim.2007040103
Losee, R. M. Jr. (1989). Minimizing information
overload: The ranking of electronic messages.
Journal of Information Science, 15(3), 179–189.
doi:10.1177/016555158901500306
The Information Overload Paradox
Mani, R. (2007). Global mindset, Indian roots: The
experience of a small Indian company--application
of Hofstede scores. ICFAI Journal of Management
Research, 6(11), 20–31.
Mano, R., & Mesch, G. (2010). E-mail characteristics, work performance and distress. Computers
in Human Behavior, 26, 61–69. doi:10.1016/j.
chb.2009.08.005
Meyer, M. E., Sonoda, K. T., & Gudykunst, W.
B. (1997). The effect of time pressure and type
of information on decision quality. The Southern
Communication Journal, 62(4), 280–292.
Nunnally, J. C. (1978). Psychometric theory. New
York, NY: McGraw Hill.
Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory. New York, NY: McGraw-Hill.
O’Reilly, C. A. (1980). Individuals and information overload in organizations: Is more necessarily
better? Academy of Management Journal, 23(4),
684–696. doi:10.2307/255556
Pearlson, K. E., & Saunders, C. S. (2001). There’s
no place like home: Managing telecommuting paradoxes. The Academy of Management Executive,
15(2), 117–129. doi:10.5465/AME.2001.4615008
Robert, L. P., & Dennis, A. R. (2005). Paradox
of richness: A cognitive model of media choice.
IEEE Transactions on Professional Communication, 48(1), 10–21. doi:10.1109/TPC.2004.843292
Rosenthal, R., & Rosnow, R. L. (1991). Essentials of behavioral research: Methods and data
analysis. Boston, MA: McGraw Hill.
Schick, A. G., Gordon, L. A., & Haka, S. (1990).
Information overload: A temporal approach.
Accounting, Organizations and Society, 15(3),
199–220. doi:10.1016/0361-3682(90)90005-F
Schultze, U., & Vandenbosch, B. (1998). Information overload in a groupware environment: Now
you see it, now you don’t. Journal of Organizational Computing and Electronic Commerce, 8(2),
127–148. doi:10.1207/s15327744joce0802_3
Sivakumar, K., & Nakata, C. (2001). The stampede toward Hofstede’s framework: Avoiding
the sample design pit in cross-cultural research.
Journal of International Business Studies, 32(3),
555–574. doi:10.1057/palgrave.jibs.8490984
Søndergaard, M. (1994). Research note-Hofstede’s
consequences: A study of reviews, citations, and
replications. Organization Studies, 15, 447–456.
doi:10.1177/017084069401500307
Speier, C., Valacich, J. S., & Vessey, I. (1997).
The effects of task interruption and information presentation on individual decision
making. Decision Sciences, 30(2), 337–360.
doi:10.1111/j.1540-5915.1999.tb01613.x
Teo, T. S. H. (2007). Organizational characteristics, modes of Internet adoption and their impact:
A Singapore perspective. Journal of Global Information Management, 15(2), 91–117. doi:10.4018/
jgim.2007040104
Thompson, B. (2004). Exploratory and confirmatory factor analysis: Understanding concepts and
applications. Washington, DC: American Psychological Association. doi:10.1037/10694-000
Toffler, A. (1970). Future shock. New York, NY:
Bantam Books.
Turetken, O., & Sharda, R. (2004). Development
of a fisheye-based information search processing
aid (FISPA) for managing information overload
in the Web environment. Decision Support
Systems, 37(3), 415–434. doi:10.1016/S01679236(03)00047-2
177
The Information Overload Paradox
Waarts, E., & van Everdingen, Y. M. (2006).
Fashion retailers rolling out across multicultural Europe. International Journal of Retail
& Distribution Management, 34(8), 645–657.
doi:10.1108/09590550610675958
Wainfan, L., & Davis, P. K. (2004). Challenges in
virtual collaboration: Videoconferencing, audioconferencing and computer-mediated communications. Santa Monica, CA: RAND Corporation.
Whitley, E. A., & Darking, M. (2006). Object lessons and invisible technologies. Journal of Information Technology, 21(3), 176–184. doi:10.1057/
palgrave.jit.2000065
178
ENDNOTE
This chapter builds upon the ideas first
presented in Kock, N., Del Águila-Obra,
A.R., Padilla-Meléndez, A. (2009). The
Information Overload Paradox: A Structural
Equation Modeling Analysis of Data from
New Zealand, Spain and the U.S.A., Journal
of Global Information Management, 17(3),
1-17.
179
Chapter 9
The Influence of Organizational
Communication Openness
on the Post-Adoption
of Computers:
An Empirical Study in Saudi Arabia
Said S. Al-Gahtani
King Khalid University, Saudi Arabia
Hung-Pin Shih
Hsuan Chuang University, Taiwan
ABSTRACT
This study adopts social identity theory (SIT) to examine the post-adoption of computers using a research model that extends the theory of planned behavior (TPB) with two organizational cultural factors. Individual attitudes toward using computers and perceived behavioral control are TPB personal
factors, while subjective norms can be viewed as the social factor. Empirical findings from 400 Arab
end-users show that openness to superior-subordinate relationships (reducing organizational boundaries)
significantly influences current computer usage only through personal factors. In contrast, openness to
superior-subordinate relationships significantly influences continued use of computers through personal
and social factors. However, openness to work communication (reducing communicative boundaries)
does not significantly influence either current computer usage or continued use of computers through
TPB beliefs. The implications for research and practice, and the limitations of this study, are discussed
accordingly.
DOI: 10.4018/978-1-60960-605-3.ch009
Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
The Influence of Organizational Communication Openness on the Post-Adoption of Computers
INTRODUCTION
An important issue in innovation diffusion is
whether the usage behavior of information technologies (IT) in less developed countries is similar to that in industrialized countries (Berkhart,
Goodman, Mehta, & Press, 1998; Brown, Hoppe,
Mugera, Newman, & Stander, 2004; Goodman &
Green, 1992; Loch, Straub, & Kamel, 2003; Zhao,
Kim, Suh, & Du, 2007). The diffusion (adoption
and use) of IT in Arab society is a good example
for addressing this question. According to Rogers
(1995), the diffusion of IT generally starts from
industrialized countries such as those of North
America or Western Europe, and then expands
across less developed countries such as Arab nations. Therefore, this study addresses whether a
well-developed theory based on Western contexts
can explain the adoption and use of IT in Arab
society. Cultural differences or cultural boundaries
might cause inconsistent findings of IT diffusion
between industrialized and less developed countries. Rose and Straub (1998) applied rationalization processes to explain the adoption and use
of IT in the Arabic context. However, Elbeltagi,
McBride, and Hardaker (2005), in a survey of Arab
senior managers, found that although some organizational factors (e.g., top management support
and organizational characteristics) may facilitate
IT adoption, others (e.g., task and environmental
characteristics) may inhibit user acceptance of IT,
even where the IT was considered user-friendly
in the adoption process. Rationalization processes
cannot completely explain these findings. In contrast, an emerging cultural influence school adopts
social processes and social influence to explain
IT diffusion and transfer in the Arabic context
(Loch, Straub, & Kamel, 2003; Straub, Loch, &
Hill, 2001). Although the cultural influence on IT
diffusion in Arab nations is now being increasingly
addressed, no generalized theory yet exists to
explain user behavior in organizational contexts.
The impact of cultural factors, such as socialpolitical control, on the common bond of computer
180
use in Arab society cannot be ignored (Goodman & Green, 1992). A cultural boundary (bias)
is likely to create cultural and social obstacles
for less developed countries in transferring IT
into practice. As the cultural boundary between
Western and non-Western countries affects the
success of IT transfer, the perceived boundary
within groups or among users may drive personal
computer usage for coordination and communication at work. Socio-cultural factors, such as social
class and degree, create cultural boundaries that
may support or impede the transfer of IT from
non-Arab to Arab cultures (Hill, Loch, Straub, &
El-Sheshai, 1998). In particular, the hierarchical
structure of Arab society is likely to strengthen
the strong “identification” perception among Arab
people. Cultural influence implies that people
often perceive social identification in terms of
their interaction with others. The social identity
theory (SIT) is a well-developed theory of the
influence of culture on IT usage (Straub, Loch,
Evaristo, Karahanna, & Strite, 2002), employed
in both Western (Gefen & Ridings, 2003) and
non-Western (Gefen, Rose, Warkentin, & Pavlou,
2005) contexts.
Kluckhohn and Strodbteck (1961) defined
culture as the shared beliefs and values that
simultaneously drive desirable and undesirable
behaviors in a society. Hofstede (1980) defined
culture as “the collective programming of the
mind which distinguishes the members of one
human group from another” (p. 21). Tichy (1983)
argued that culture is often implicit, implying
that individuals are not aware of its influences
as they are of political and technical dynamics.
In summary, culture is a multi-faceted concept
connecting nations, societies, and organizations
(Gallivan & Srite, 2005). Whereas Hasan and Ditsa
(1999) addressed the influence of cultural identity
and values on IT adoption, this study finds that
organizational culture is more likely than national
or social cultures to influence computer adoption
and use by office workers in the workplace. Organizational culture refers to the common norms
The Influence of Organizational Communication Openness on the Post-Adoption of Computers
and values that are accepted and followed by the
majority of members of an organization (Kats &
Kahn, 1978). Moreover, organizational culture
acts as a “filter” that embeds shared history and
experience to distinguish between members and
non-members of an organization (Weber & Pliskin,
1996). Organizational culture is conceptualized
in terms of specific norms, values, assumptions,
and social structures to shape members’ beliefs
and behaviors in organizations (Gallivan & Srite,
2005). Thus, organizational culture often drives
the formation of social beliefs and norms regarding individual behaviors, such as IT adoption and
use in a specific organizational context (Davison
& Martinsons, 2003).
The globalization of business has highlighted
the need to understand computer applications that
span different nations (Niederman, Boggs, & Kundu, 2002; Shin, Ishman, & Sanders, 2007). Multinational and trans-cultural organizations have a
growing need to apply IT to achieve economies of
scale, coordinate global operations, and facilitate
collaborative work across distributed locations and
diverse cultures (Ives & Jarvenpaa, 1991). The
globalization of IT means that cultural differences
in user evaluations of information systems significantly influence the post-adoption of computers.
Cultural differences, rather than technological
benefits, determine the success of adoption and
implementation of information systems (Png, Tan,
& Wee, 2001; Straub, Keil, & Brenner, 1997).
For research communities, this study extends an
existing well-developed theory based on Western
cases by incorporating organizational culture to
assess the post-adoption of computers by Arab endusers in the workplace. Both user acceptance and
continuance must be considered, because they are
both important measures for assessing the success
of IT implementation (DeLone & McLean, 2003).
For practitioners, multinational and trans-cultural
enterprises could apply the findings of this study to
help them understand how organizational cultures
can motivate/inhibit Arab end-users to adopt and
continue using computers at work.
TECHNOLOGICAL VERSUS
SOCIAL VIEWS
Many previous studies of IT tend to assess end-user
behavior in the technology-orienting countries
(Ein-Dor, Segev, & Orgad, 1992). For instance,
drawing upon the “theory of reasoned action
(TRA)” (Fishbein & Ajzen, 1975), Davis (1989)
proposed a technology acceptance model (TAM)
to predict the adoption of new IT. TAM defines
perceived usefulness (PU) and perceived ease
of use (PEOU) as the two major rational beliefs
that influence the adoption of IT. To examine the
influence of culture on IT adoption, a series of
TAM-related studies based on Hofstede’s work
was conducted across industrialized and less developed countries (Elbeltagi, McBride, & Hardaker,
2005; Gefen, Rose, Warkentin, & Pavlou, 2005;
Straub, 1994; Straub, Keil, & Brenner, 1997;
Veiga, Floyd, & Dechant, 2001). For instance,
Straub (1994) adopted Hofstede’s uncertainty
avoidance and TAM beliefs to examine the diffusion process of e-mail and fax in the U.S. and
Japan. Straub, Keil, and Brenner (1997) found that
culture affects e-mail use differently in U.S. and
Japanese companies. Japanese office workers view
e-mail as a poor communication medium, because
of the high uncertainty of carrying messages in
their social interactions. In contrast, American
workers perceive e-mail as a useful medium for
improving work productivity. Rose and Straub
(1998) examined IT use in the Arabic world using
TAM beliefs, noted that over-emphasis on rational
factors based on technological viewpoints, such
as TAM beliefs, may lead to a cultural backlash
rather than cultural acceptance. Although social
influences (e.g., social presence and social identification) on IT adoption are important, few studies
have examined the social process of IT diffusion in
less developed countries. As noted by Mathieson
(1991), TAM does not explicitly include social
factors for predicting IT usage in organizations.
Thus, a new direction to address the influence of
“social norms” on Internet diffusion in the Arab
181
The Influence of Organizational Communication Openness on the Post-Adoption of Computers
world has been widely discussed recently (Loch,
Straub, & Kamel, 2003). TAM was originally
developed to assess the adoption of computers
underlying an individual-oriented context, but
not the diffusion of computers in a social-oriented
context, such as the Arab society. In contrast to
previous studies of IT adoption based on rational
beliefs (e.g., PU and PEOU), this study focuses
simultaneously on assessing personal and social
influences on the post-adoption of computers by
Arab end-users.
CULTURE AND IT DIFFUSION
IN THE ARAB SOCIETY
Arabs are heavily attached to each other by family
bonds and obligations, limiting their geographic
and occupational mobility (Palmer, Al-Ghofaily,
& Alnimir, 1984). Hill et al. (1998) contend that
Arab culture is not a monolithic structure or a
mosaic of peoples, but a complex system that is
always changing. They suggest that Arab culture
and history, while selectively borrowing ideas
and materials from the west, is an independent
complex system infused with contradictions and
opposing forces. In summary, Arab culture seems
to exert a stronger social influence than Western
culture on its society through the development of
social norms and beliefs.
Hofstede (1980, 1991) initially defined national culture in four dimensions, namely uncertainty avoidance, individualism, masculinity, and
power distance. Uncertainty avoidance denotes the
level of tolerance for uncertainty and ambiguity
within the society. Individualism is the degree to
which the society reinforces individual or collective achievement and interpersonal relationships.
Masculinity is the degree to which “masculine”
values, like assertiveness, performance, success,
and competition, prevail over “feminine” values,
including quality of life, maintaining warm personal relationships, service, caring, and solidarity.
Power distance refers to the degree of equality or
182
inequality among people in the society. As noted
by Hofstede (1991), Arab culture is characterized
by more uncertainty avoidance and power distance
than American culture, but less individualism and
masculine values. In comparison, Arab society is
a social-oriented culture, unlike the individualoriented culture prominent in the USA and other
Western countries. In addressing the interaction
between organizational culture and IT from the
social identity perspective (Tajfel, 1978), Gallivan
and Srite (2005) argue that in-group relationships
among members would initially cause fragmented
and differentiated interpretations of their identities, which would finally shape their beliefs and
behaviors of IT in the cognitive process.
Although Saudi organizational leaders regard
computers as signs of modernization and automation compared with manual practice, many of their
systems are incompatible with each other (Abdul
Ghani & Al-Sakran, 1988; Atiyyah, 1989). Most
users have not been aware of this incompatibility,
or are unwilling to accept it. Such problems are
compounded by fear of and alienation from computers, especially among some Saudi executives
(Yavas, Liqmani, & Quraeshi, 1992). In facing
the impact of new IT from Western countries, the
top executives may worry about lack of control
over their political and economic benefits, and
the threat from Western values to the existing
Arab society. Saudi Arabia is a rather culturally
conservative society, where Islamic teachings
and Arab cultural values are dominant. The
country plays vital political and economic roles
for Arabs and Muslims all over the world. Saudi
Arabia has tried to balance modernization with
its Islamic identity. Al-Farsy (1996) assessed the
progress of technological development including
IT in Saudi Arabia, and provided an overview of
the most ambitious development program in the
history of the country, thus setting Saudi Arabia
in a global political and economic context. He
described how rapidly progress has been achieved
without diminishing the distinctive Islamic identity of a country that employs the Quran as its
The Influence of Organizational Communication Openness on the Post-Adoption of Computers
constitution. In summary, Arab social and cultural
characteristics define organizational culture, and
call for a survey of their roles as driving forces
in IT adoption.
IT adoption in Saudi Arabia is influenced by explicit government policy in the attempt to enhance
national and organizational productivity. Saudi
Arabia occupies an area of 2.15 million square
kilometers, with a population estimate of 24.07
million, and Saudi Internet users were estimated to
number 4.7 million in 2007 (Internet World Stats,
2007). The ratio of PCs per 1,000 persons in 2005
was increased to 354 (World Bank Group, 2007).
Furthermore, Saudi Arabia ranks third among the
22 Arab Nations in total IT market penetration.
Thus, some observers are expecting strong growth
in the Saudi IT market for the foreseeable future.
To enhance computer acceptance and diffusion, the
government has implemented a motivation policy
offering a bonus of 25% of basic salary to Saudi
nationals who are specialists in IT or have careers
in computers. The advent of the Internet, and the
impetus from high authority toward e-commerce
and e-government, are essential catalysts for individuals and organizations toward IT adoption for
their competitive advantage. However, religious
values and political environments still implicitly
control Arab society, and govern the development
of organizational culture.
RESEARCH MODEL
TPB Beliefs
Based on TRA (Fishbein & Ajzen, 1975), Ajzen
(1985) proposed the theory of planned behavior
(TPB) to predict goal-directed human behavior.
According to TPB (Ajzen, 1985; Ajzen & Madden,
1986; Ajzen, 1991), an individual’s behavior is
determined by his or her behavioral intention (BI)
to perform it. The BI is determined by individual
attitudes (A) toward behavior, subjective norms
(SN) and perceived behavioral control (PBC).
Additionally, PBC is another determinant of
behavior. From TPB, BI refers to the subjective
probability that a person will perform a particular
behavior in the near future (Fishbein & Ajzen,
1975, p. 12). Attitude is considered as a personal
(or “attitudinal”) factor, and denotes “a person’s
general feeling of favorableness or unfavorableness toward some stimulus object” (Fishbein &
Ajzen, 1975, p. 216). Subjective norms, that is,
the social (or “normative”) factor, refers to “the
person’s perception that most people who are
important to him think [that] he should or should
not perform the behavior in question” (Fishbein
& Ajzen, 1975, p. 302). Perceived behavioral
control, which is a personal factor, refers to “[a]
person’s belief as to how easy or difficult [the]
performance of the behavior is likely to be” (Ajzen
& Madden, 1986, p. 457). TPB encompasses those
personal and social factors that are particularly
suited to assessing the adoption and use behavior
of computers underlying a social-oriented culture.
Thompson, Higgins, and Howell (1991) argued
that BI should be excluded if actual behavior (system usage) has taken place. Thus, this study applies
TPB beliefs to predict end-users’ current usage
(i.e., adoption) and continued use (i.e., adoption
confirmation) of computers at work by omitting
the BI construct in accordance with our study
and survey context (Figure 1). In explaining the
association between attitude toward a particular
behavior and the behavior, this study focuses on
the attitudinal process in human behavior. In the
course of a person’s attitude formation, attitude is
determined by salient beliefs about the behavior
of interest (e.g., behavioral beliefs) and outcome
evaluations (Ajzen, 1985; Ajzen & Madden, 1986).
When a person believes that his or her behavior
leads to a particular outcome of interest, he or she
wishes to take real action and continue for it. This
attitude-usage link has been found by Thompson,
Higgins, and Howell (1991) and Davis (1993),
while the link between attitude and continued
use can be deduced similar to the link of attitude
with current use behavior. Regarding the adoption
183
The Influence of Organizational Communication Openness on the Post-Adoption of Computers
Figure 1. Research model
and use of computers by Arabic end-users, this
study posits that:
H1a: An individual’s attitude toward computer
use positively affects his or her current use
of computers.
H1b: An individual’s attitude toward computer
use positively affects his or her continued
use of computers
In addressing the association between subjective norms and behavior, this study focuses on the
normative formation underlies human behavior.
The social factor (i.e., subjective norms) is determined by the normative expectations of relevant
referents (e.g., normative beliefs) and the motivation to comply with the referent (Ajzen, 1985;
Ajzen & Madden, 1986). People who believe that
most important referents think they should or
should not perform a particular behavior would
feel pressure to comply with these referents. Arab
office workers under social-oriented contexts are
easily influenced by social norms or values from
colleagues in the adoption and use of computers.
Thus, TPB beliefs, such as subjective norms, are
more suitable than TAM beliefs (e.g., PU and
PEOU) for assessing the post-adoption of comput-
184
ers by Arabic end-users. That is, Arabic end-users
who tolerate the influence of subjective norms
would use computers and continue the behavior
at work. An association between subjective norms
and usage has been found (Thompson, Higgins, &
Howell, 1991; Thompson & Rose, 1994), while
the link between subjective norms and continued
use is logically reasonable and innovatively hypothesized in this study for examination. Thus,
this study posits that:
H2a: Subjective norms positively affect an individual’s current use of computers.
H2b: Subjective norms positively affect an individual’s continued use of computers.
PBC refers to a person’s perception of the
presence or absence of requisite resources and
opportunities to perform the behavior of interest
(Ajzen & Madden, 1986). This behavioral control
belief is determined by internal factors, such as
personal skills, abilities, and knowledge, as well
as external factors, such as resources and opportunities required in performing goal-directed
behaviors. In summary, PBC is a function of
the products of control beliefs about available
resources and opportunities, and the perceived
The Influence of Organizational Communication Openness on the Post-Adoption of Computers
facilitation of resources and opportunities for
achieving outcomes. Thus, TPB argues that PBC
is a key determinant of actual behavior (Ajzen
& Madden, 1986; Ajzen, 1991). People who are
confident that they can control computer applications at work are likely to use the computer, and
continue to use them. The PBC-usage link has
been identified from empirical support (Taylor &
Todd, 1995), while the PBC-continued use link
tends to be valid for experienced end-users. Two
hypotheses based on TPB have been proposed for
examining computer adoption and use behavior
in an Arab context:
H3a: Perceived behavioral control positively
affects an individual’s current use of computers.
H3b: Perceived behavioral control positively
affects an individual’s continued use of
computers.
SIT and Organizational
Communication
SIT explains how people perceive their belonging
to groups based on their recognition of self-identity
(Tajfel, 1978). Self-identity refers to a person’s
perception of self-identification as part of the same
group as others (in-group), or not as a member of
the group (out-group). SIT adopts personal and
social identity to address inter-group relationships
in social categorizations (Tajfel & Turner, 1979;
Turner, 1982). Personal identity tells people their
self-identity within groups, while social identity
defines their memberships of certain groups.
According to SIT, in-group members are more
likely to share values, and to have similar beliefs
and consistent behaviors. In contrast, out-group
members are less likely to generate close perceptions of thinking and acting.
Social categorization is a self-identification
process to create inter-group boundaries, and thus
shape members’ values, attitudes, and behaviors.
Inter-group boundaries often exist among IT us-
ers from different departments, who have distinct
views about the application of IT at work. Intergroup boundaries often increase the time and
costs of work, impeding cross-group members
from cooperating and communicating with others. Therefore, this study extends the concept of
inter-group boundaries to address how to reduce
organizational and communicative boundaries in
social interactions (Tushman & Scanlan, 1981).
Organizational boundaries often exist in the
categorization process, but can be reduced when
individuals identify their “in-group” relationships
with others. In contrast, organizational boundaries
can be increased when individuals perceive their
“out-group” relationships with others. People
tend to have inconsistent or conflict rationality
boundary owing to gaps in their knowledge and
competency. The categorization process thus involves two opposite scenarios. From the positive
viewpoint, people can reduce their communicative
boundaries to achieve effective coordination and
communication by expanding their in-group memberships with others. From the negative viewpoint,
people can increase their communicative boundaries to impede valuable information from crossing
benefit-conflict departments by expanding their
out-group memberships with certain groups. In
social identification process, re-categorization and
de-categorization are two effective methods for
reducing boundaries (Brown, 1996). Re-categorization places out-group members into an extended
in-group. Thus, all members have common goals,
reducing organizational boundaries across groups.
De-categorization emphasizes the interaction of
members in joint activity, so that they can easily
share values and perceptions with others, which
reduces communicative boundaries across groups.
Organizational culture has been found to influence IT adoption and use behaviors via personal
or social factors (Kedia & Bhagat, 1988). Thus,
this study applies SIT to explain the potential
influence of organizational culture, and how it
is related to individuals’ beliefs and behaviors
regarding computers in the workplace. Arab
185
The Influence of Organizational Communication Openness on the Post-Adoption of Computers
office workers under social-oriented contexts
would have an organizational culture different
from American office workers, whose beliefs and
behaviors are formed under individual-oriented
contexts, particularly in the context of organizational communication (Goodman & Green, 1992).
Regarding the use of computers for office information or communication, this study focuses on
using openness in organizational communication
to approach organizational culture. Beyond the
concept of organizational proximity in addressing
the effective use of communication technologies
(Monge, Rothman, Eisenberg, Miller, & Kirste,
1985), openness is an appropriate measure for
assessing organizational structures and communication networks at virtual offices, particularly
in a social-oriented context.
Eisenberg and Witten (1987) described
openness in three different ways. First, openness is treated as personal disclosure to support
superior-subordinate relationships. Second,
openness is considered as the free flow of taskrelated, non-personal information for effective
organizational communication. Third, openness
refers to clear and unambiguous communication
in language. In approaching the culture around
organizational communication, this study assumes
that all members under a social-oriented context
would use and can use the same language at work,
thus significantly alleviating the concern of communication ambiguity. This study thus defines
the openness of organizational communication
in terms of (1) openness to superior-subordinate
relationships (OSSR) and (2) openness to work
communication (OWC). Openness to superiorsubordinate relationships can be employed in
the re-categorization to assess the intensity of
interactions among members across organizational
boundaries. The influence of an open organization has been addressed in the literature (Steele,
1975). An open organizational structure with
mutual respect and trust relationships could help
members reduce organizational boundaries, and
easily strengthen their relationships with others.
186
Openness to superior-subordinate relationships
encourages in-group members, such as managers
and subordinates, to strengthen their interactions
by creating an integral family-like bond in the recategorization process, leading them to have favorable self-assessment on common goals. In-group
members who work under an open organization
with mutual respect and trust relationships tend
to have positive attitudes toward using computers
to achieve common goals. Thus, the following
link between openness to superior-subordinate
relationships and attitude toward using computers is posited:
H4a: Openness to superior-subordinate relationships positively affects individual attitudes
toward using computers.
In the de-categorization, openness to work
communication can be employed to assess information flows across communicative boundaries.
Openness to work communication helps in-group
members to create an appropriate work context
to ‘fit’ the de-categorization process and support
their self-enhancement of thoughts and actions
in social processes. According to SIT (Tajfel &
Turner, 1979; Turner, 1982), openness to work
communication expands in-group members’ social identities, thus reducing their communicative
boundaries through interaction with others. An
open communicative context with well-defined
rules, cooperation, and shared values not only
motivates in-group members to develop cohesive
memberships, but also encourages them to form
positive attitudes toward using computers for
communication. The positive association between
openness to work communication and attitude is
thus hypothesized as follows:
H4b: Openness to work communication positively
affects individual attitudes toward using
computers.
The Influence of Organizational Communication Openness on the Post-Adoption of Computers
Although unable to see culture, individuals can
strongly feel cultural constraints on their thoughts
and actions in daily life. In a social-oriented context
such as Arab society, organizational culture often
guides individual behavior via shared values or
beliefs (Eisenberg & Riley, 1991; Geertz, 1973).
Islam is the dominant religion among all social
groupings, and sets constraints on beliefs and
actions in Arab society. The use of Islam as the
major organizing principle for Arab thoughts and
actions has been criticized. For instance, Hill et
al. (1998) concluded that the influence of Islam is
always changing as people “practice” their daily
life. This can be a creative force that negotiates
the meaning of Arab’s actions, while simultaneously forming the basis for transformation of their
beliefs. Owing to the differences between Arab
culture and Western culture, human behavior in
Arab organizations is more suited to be explained
through cultural characteristics and cultural constraints. Generally, similar cultural characteristics
lead individuals to identify their in-group memberships; while different cultural constraints lead
them to perceive their out-group memberships in
the categorization process. Sub-cultural factors,
such as openness to superior-subordinate relationships, are often infused with major cultural
characteristics, such as religious values. These
factors influence individuals’ thoughts and actions, particularly in a social-oriented context
like Arab society. Straub et al. (2002) argue that
cultural similarity enables in-group members to
have proactive psychological process in social
categorization, enabling them to rely on “referent norms” to guide their behaviors. An open
organization can weaken the impact from cultural
constraints to reduce inter-group boundaries in
the re-categorization process. In-group members
with high similarity of cultural characteristics and
low difference of cultural constraints would easily
develop consistent perception to group norms. An
open organization facilitates the development of
group norms, such as consistent normative ex-
pectations and motivations when using computer
applications. Thus, this study posits that:
H5a: Openness to superior-subordinate relationships positively affects an individual’s
subjective norms regarding computer use.
Culture represents the actions, practices, stories, and artifacts in human behaviors (Eisenberg
& Goodall, 1998). The culture of an organization
can be developed by constructing it symbolically
around human interactions with the environment.
Organizational culture thus evokes its social
influence on the beliefs and actions of members
when they try to interpret the meanings of symbolic constructions in communication processes.
Cultural constraints often become barriers when
in-group members try to reduce communicative
boundaries by touting out-group members to join
the same group. According to an extended view
of Triandis’ (1972) subjective culture, openness
to work communication facilitates Arab end-users
to create similar patterns of thinking and understanding about the role of IT in organizations, as
well as to approach a consensus on well-defined
rules, cooperation and shared values for interpreting information in computer applications. Thus,
in-group members under an open communication
context are most likely to achieve consensus on
group norms, and to take it as the “referent” for
guiding their behaviors. This study extends the
subjective culture concept to predict the adoption and use of computers by Arab end-users by
positing that:
H5b: Openness to work communication positively
affects an individual’s subjective norms
regarding computer use.
In contrast to the pre-Internet era, the diffusion
of the Internet enables Arab culture to become
more “open” in a rapidly growing information
society than in the past, because Arab end-users
have more opportunities to access a wealth of
187
The Influence of Organizational Communication Openness on the Post-Adoption of Computers
information in the internalization environment.
Thus, the increasing openness in current IT context reduces the formalization and centralization
of work, thus increasing the positive personal
and social beliefs of Arab end-users concerning the adoption and use of computers for work
productivity. Considering cultural constraints on
control beliefs and perceived facilitation, an open
organization would enable in-group members
to develop flexible institutional-based rules to
reduce their conflicts in values, beliefs, and behaviors (Eisenberg & Goodall, 1998). Openness
to superior-subordinate relationships drives ingroup members to build flexible institution-based
memberships in group interactions, so that they
can easily adjust themselves to fit the environment.
The congruence between the expected values
of an in-group and the organizational culture
creates a “person-culture fit” in the cooperation
and coordination process with others (O’Reilly,
Chatman, & Caldwell, 1991). In-group members
can control their actions for particular outcomes,
provided that their interactions are constituted
under an open organization with high levels of
“person-culture fit.” Thus, this study posits that:
H6a: Openness to superior-subordinate relationships positively affects an individual’s
perceived behavioral control regarding
computer use.
Openness to work communication could help
in-group members to build interpersonal trust
around them. Beyond the parallel argument of
trust-control (Das & Teng, 1998), this study considers confidence-control instead of trust-control
as the causal association for two reasons. First,
personal confidence enables in-group members
to trust the abilities of others (Mayer, Davis, &
Schoorman, 1995). Second, personal confidence
leads in-group members to trust strong support
from others (Mayer, Davis, & Schoorman, 1995).
The development of confidence relies on an open
communicative context that facilitates individuals
188
to believe in their personal control over computer
applications in the workplace. The causal link
between openness to work communication and
PBC can thus be hypothesized as follows:
H6b: Openness to work communication positively
affects an individual’s perceived behavioral
control regarding computer use.
METHODOLOGY
Data Collection
A list of governmental ministries with their
branches and major companies in the main four
provinces of Saudi Arabia was collected with the
aid of the Chamber of Commerce in each region.
A letter was sent to each of the 136 private and
public organizations across the country seeking
their participation in the study. Those organizations that accepted the invitation to participate
were asked to nominate a contact person to liaise
in distributing and collecting the paper-and-pencil
questionnaires. Through that procedure, 1,900
end-users were surveyed at offices, and 1,190 usable responses were obtained, achieving a response
rate of 62.63%. Participants in this study were
end-users from 56 private and public organizations in Saudi Arabia. The participating organizations were distributed throughout the country
and represented various types of institutions such
as banking, merchandizing, manufacturing, and
petroleum industries, as well as educational,
health, and public services. Based on one rule of
thumb, a sample size with at least ten times (the
number of measurement items + the number of
hypothesized links) is required to analyze both
measurement and structural models simultaneously (Muthén & Muthén, 2002). Table 1 shows
400 participants were randomly selected out of
the 1190 responses for examining the research
model in order to relax some concerns on statistical outputs from a large sample.
The Influence of Organizational Communication Openness on the Post-Adoption of Computers
Table 1. Profile of the participants (N=400)
Characteristics
Statistics
Gender
Male: 316 (79%)
Female: 84 (21%)
Age
18-30
31-40
41-50
51 or more
Mean = 38.2 Years, Range 18-58 Years
155 (38.75%)
171 (42.75%)
67 (16.75%)
7 (1.75%)
Saudi/Non-Saudi
Saudi: 366 (91.5%)
Non-Saudi: 34 (8.5%)
Education levels
€€€Less than High School
€€€High School
€€€Some College
€€€College Graduates
22 (5.5%)
64 (16%)
100 (25%)
214 (53.5%)
Instrument Development
As shown in Appendix A, openness to superiorsubordinate relationships and openness to work
communication were measured based on Eisenberg and Witten (1987). Attitude, subjective norms
and perceived behavioral control were measured
using Mathieson’s (1991) scales. Current and
continued usages are two good measures used to
assess the success of implementing an innovation,
particularly in the process from adoption decision to adoption confirmation (Rogers, 1995).
Current usage was measured using the four-item
scale proposed by Igbaria, Guimaraes, and Davis
(1995) to meet computer use context, covering (1)
perceived daily use of computers, (2) perceived
frequency of computer use, (3) the number of
computer applications used by participants and
(4) the number of business tasks supported by the
computer. Continued use was measured in terms
of the willingness and the possibility of using
computers for work in the near future.
Data Analysis
A structural equation modeling tool, partial least
squares (PLS), was used to examine the proposed
research model (Chin, 2001). PLS-Graph was em-
ployed to simultaneously assess the psychometric
properties of the measurement model, including
the reliability and validity of all constructs, and to
estimate the parameters of the structural model,
including the strength of the path relationships
among research constructs (Fornell, 1982; Lohmoller, 1989).
RESULTS
Measurement Model
This study assessed measurement validity in terms
of unidimensionality, reliability, and construct
validity. Table 2 shows that all observed indicators intended to measure the same latent construct
exhibited higher factor loadings (i.e., more than
0.7) on the posited correspondent construct than
on other constructs (Yoo & Alavi, 2001), indicating that the unidimensionality of each construct
was achieved. Table 3 shows that the composite
reliability of the seven constructs ranged between
0.78 and 0.96, exceeding the threshold of 0.7
recommended by Nunnally (1978). The AVEs of
all constructs exceeded 0.5 (Fornell & Larcker,
1981), indicating that the measurement model
exhibited satisfactory convergent validity. From
189
The Influence of Organizational Communication Openness on the Post-Adoption of Computers
Table 2. Analysis of unidimensionality (factor loadings (bolded) and cross loadings)
Items/Factor
Current
Use
Continued
Use
Attitude
SN
PBC
OSSR
OWC
CURNTUSE1
0.80
0.44
0.26
0.12
0.21
0.12
0.10
CURNTUSE2
0.78
0.35
0.32
0.06
0.23
0.17
0.11
CURNTUSE3
0.78
0.45
0.24
0.08
0.27
0.19
0.17
CURNTUSE4
0.70
0.36
0.26
0.07
0.26
0.26
0.14
CONTUSE1
0.27
0.85
0.28
0.28
0.41
0.28
0.21
CONTUSE2
0.61
0.79
0.31
0.18
0.34
0.26
0.21
ATT1
0.31
0.32
0.90
0.13
0.27
0.28
0.23
ATT2
0.33
0.34
0.91
0.12
0.34
0.27
0.17
ATT3
0.33
0.32
0.91
0.10
0.33
0.29
0.22
ATT4
0.33
0.33
0.89
0.09
0.28
0.23
0.23
ATT5
0.29
0.29
0.89
0.07
0.30
0.26
0.23
SN1
0.12
0.28
0.11
0.95
0.24
0.19
0.17
SN2
0.12
0.26
0.08
0.96
0.22
0.18
0.15
SN3
0.07
0.25
0.12
0.91
0.21
0.24
0.21
PBC2
0.11
0.30
0.24
0.12
0.74
0.18
0.19
PBC3
0.36
0.41
0.25
0.21
0.80
0.11
0.07
PBC4
0.17
0.27
0.26
0.18
0.70
0.25
0.20
OSSR1
0.22
0.30
0.28
0.20
0.17
0.81
0.54
OSSR2
0.22
0.28
0.25
0.14
0.19
0.85
0.53
OSSR3
0.16
0.23
0.18
0.18
0.21
0.78
0.50
OWC1
0.21
0.24
0.23
0.18
0.24
0.57
0.87
OWC2
0.12
0.21
0.21
0.14
0.12
0.58
0.90
OWC3
0.11
0.22
0.20
0.18
0.13
0.56
0.89
Note: PBC1 loaded weakly and was discarded from final analysis
Table 4, the square root of the AVE for each
construct exceeded the correlation between every
pair of constructs, ensuring adequate discriminant validity (Fornell, Tellis, & Zinkhan, 1982).
Consequently, the measurement model exhibited
acceptable construct validity.
Structural Model
The PLS results from the structural model in Table
5 show that attitudes toward using computers and
perceived behavioral control have significant
and positive effects on current use of computers
(direct effects = 0.281 and 0.223, respectively),
190
supporting hypotheses H1a and H3a, respectively.
In contrast, the effect of subjective norms on current computer usage is insignificant, meaning that
H2a is not supported. However, attitudes toward
using computers, subjective norms, and perceived
behavioral control all have significant and positive effects on continued use of computers (direct
effects = 0.223, 0.177 and 0.344, respectively),
supporting hypotheses H1b, H2b and H3b, respectively. In comparison, perceived behavioral
control is the largest determinant and predictor
of continued use of computers. Moreover, the
explained variance of continued use of computers (28.24%) by the three TPB beliefs is greater
The Influence of Organizational Communication Openness on the Post-Adoption of Computers
Table 3. Analysis of reliability and convergent validity
Latent Variables
Composite Reliability
(Internal Consistency Reliability)
Average Variance
Extracted/Explained
(AVE)
OSSR
0.85
0.659
OWC
0.92
0.784
ATT
0.96
0.810
SN
0.96
0.886
PBC
0.78
0.544
Continued Use
0.81
0.678
Current Use
0.85
0.584
than that of current computer usage (17.03%).
From Table 5, the PLS results also support the
three hypotheses (H4a, H5a, and H6a) for the
significant and positive effects of openness to
superior-subordinate relationships with attitudes
toward using computers, subjective norms, and
perceived behavioral control (direct effects =
0.246, 0.155 and 0.185, respectively). In contrast,
openness to work communication does not significantly influence individual attitudes toward
using computers, subjective norms, and perceived
behavioral control, showing hypotheses H4b, H5b
and H6b are not supported in the current empirical
survey. Overall, organizational communication
openness accounted for 9.24%, 5.14%, and 5.79%
of variance in attitude, subjective norms, and
perceived behavioral control, respectively. Path
analysis was also used to examine the indirect
effects of OSSR and OWC on the adoption and
adoption confirmation of computers. The indirect
effects of OSSR on current use and continued
use of computers were (0.114, 0.146), whereas
the indirect effects of OWC on current use and
continued use of computers were (0.023, 0.062),
respectively. Overall, the tests of the structural
model show that openness to superior-subordinate
relationships is more influential than openness
to work communication in driving the adoption
and adoption confirmation of computers by Arab
end-users via TPB beliefs.
Table 4. Analysis of discriminant validity and intercorrelations
Latent Variables
Mean
SD
1
OSSR
5.90
1.23
.81
2
3
4
5
6
OWC
6.43
1.02
.65
.89
ATT
6.12
1.00
.30
.24
.90
SN
5.54
1.46
.22
.19
.11
.94
PBC
5.29
0.87
.23
.20
.34
.23
.74
Continued Use
4.43
1.22
.33
.25
.36
.28
.46
.82
Current Use
6.18
0.96
.25
.17
.35
.11
.32
.52
7
.76
Note: The bolded diagonals represent the square root of the AVE from each construct, while the off-diagonal elements represent the correlations between different constructs
191
The Influence of Organizational Communication Openness on the Post-Adoption of Computers
Table 5. Results of PLS analysis
Causal Paths
Path Coefficient
T-Statistic
Current Use ←
ATT
SN
PBC
0.281**
0.026
0.223**
4.82
0.46
4.01
Continued Use ←
ATT
SN
PBC
0.223**
0.177**
0.344**
3.79
3.59
6.89
ATT ←
OSSR
OWC
0.246**
0.082
3.72
1.07
SN ←
OSSR
OWC
0.155*
0.091
2.12
1.19
PBC ←
OSSR
OWC
0.185**
0.080
2.74
0.95
Explained Variance (R2)
17.03%
28.24%
9.24%
5.14%
5.79%
Hypothesis Testing
H1a is supported
H2a is not supported
H3a is supported
H1b is supported
H2b is supported
H3b is supported
H4a is supported
H4b is not supported
H5a is supported
H5b is not supported
H6a is supported
H6b is not supported
* Significant at (p<0.05) ** Significant at (p<0.01)
DISCUSSION AND IMPLICATIONS
This study integrates organizational communication openness with TPB beliefs (personal and
social factors) to assess current computer usage
and predict continued use of computers. Consistent with TPB, empirical findings show that Arab
end-users with positive attitudes toward using
computers perceive strong social norms from
colleagues about the use of computers, and feel
strong control when using computers are likely
to continue to use computers for future work.
That is, Arab end-users would continue to use
computers if they feel that their current use of
computers is interesting, they believe that they
must achieve the expectations of computer use
from colleagues, and they are confident in their
skills in computer use. Partly consistent with
TPB, Arab end-users’ current usage of computers
can be explained in terms of their interests and
personal control over computers, but not social
pressures. Our empirical findings contribute different insights on social pressure from two previous
studies that were conducted in western contexts.
Thompson, Higgins, and Howell (1991) identi-
192
fied a positive association between social norms
and utilization from a survey of computer use
by knowledge workers in a large multinational
firm. Taylor and Todd (1995) found that all TPB
beliefs have significant influence on behavioral
intention from the usage survey of a computing
resource center by business school students in
Canada. This study shows that attitude is the greatest factor influencing current computer usage. In
particular, the insignificant effect of subjective
norms on current computer usage indicates that
the social factor does not significantly change
the decision to use computers by Arab end-users.
This phenomenon can be explained based on Hill
et al.’s (1998) observations that many Arab endusers tend to ignore social norms initially in the
current use of computers at work when they fail
to observe unanimous recognition among their
colleagues. Another reason may be attributed to
fear among some executives that computers might
replace their existing jobs, meaning that they did
not strongly push employees to currently adopt
and use computers in the workplace. As noted by
Loch, Straub, and Kamel (2003), technology is not
force-fed through social mechanism, but depends
The Influence of Organizational Communication Openness on the Post-Adoption of Computers
on whether managerial levels recognize they can
receive the benefits of adoption. Moreover, the
findings indicate that the influence of subjective
norms increases in the process from adoption to
adoption confirmation of computers, revealing that
the influence of social factors becomes significant
in predicting Arab end-users’ post-adoption of
computers, rather than the adoption decision. Thus,
social influence begins to work when most Arab
office workers continue to spend time running
computer applications, thus producing a “critical
mass” of end-users in the process from adoption to
adoption confirmation (Markus, 1990). Additionally, perceived behavioral control is the greatest
determinant and predictor of continued use of
computers and achieved higher effect compared
with its effect on current computer usage. That
is, Arab end-users are more concerned with their
control of resources and opportunities, and the
governance of knowledge and ability, than with
other factors when they decide to continue to use
computers for future work. In summary, our empirical findings show that personal factors (e.g.,
attitude and perceived behavioral control) influence the adoption and post adoption of computers,
while the social factor (i.e., subjective norms)
only influences the post adoption of computers.
Our empirical findings show that openness
to superior-subordinate relationships drives significant personal and social influences on Arab
end-users’ beliefs about computers than openness
to work communication does. Thus, reducing
organizational boundaries is more effective than
reducing communicative boundaries to develop
personal and social beliefs of computer adoption
and use. The identified associations between
openness to superior-subordinate relationships and
TPB beliefs confirm the arguments in the research
model. That is, an open organizational structure
is likely to create mutual respect and trust relationships around Arab end-users, which motivate
them to form favorable attitudes, comply with the
expectations from colleagues, and believe of their
control abilities regarding computer adoption and
use in the workplace. In contrast, the insignificant
associations between openness to work communication and TPB beliefs reveal the ignorance
of an open communicative context within Arab
organizations. Such a context with well-defined
rules, cooperation, and shared values around Arab
end-users did not encourage their interests and
give them greater control in using computers, or
even force them to use computers at work. This
phenomenon might be attributed to the lack of
social processes in achieving organizational consensus and confidence regarding computer use for
communication by Arab end-users. Inconsistent
with SIT (Tajfel & Turner, 1979; Turner, 1982),
Arab end-users in an open communicative context
cannot necessarily reduce their communicative
boundaries through interaction with others. The
survey of Arab end-users indicates that openness
to superior-subordinate relationships has a stronger effect than openness to work communication
in social categorization, because it can reduce
perceived boundaries among in-group members
in social processes. A prior survey of media use
for work communication among technical and
administrative employees in a large U.S. telecommunications firm by Hinds and Kiesler (1995)
concluded that organizational boundaries can be
reduced by strengthening horizontal communication, particularly in the communicative structure
of technical workers. In contrast, this study suggests that Arab end-users are more inclined to
vertical communication (open organization) than
horizontal communication (open communication)
when they form their personal and social beliefs
of computer adoption and use. In other words, an
organizational structure is more influential than
a communication network to expand “boundarycrossing” by Arab end-users in the formation of
favorable attitudes, subjective norms, and perceived behavioral control regarding the adoption
and use of computers in the workplace.
193
The Influence of Organizational Communication Openness on the Post-Adoption of Computers
Implications for Academic Research
Implications for Practice
Computer use behavior in Saudi Arabia based
on social factors has seldom been addressed
(Al-Khaldi & Wallace, 1999; Loch, Straub, &
Kamel, 2003). Moreover, the impact of cultural
boundaries on IT diffusion in Arab contexts has
seldom been explored as well. Therefore, this
empirical survey in Saudi Arabia helps improve
our understanding of the impact of organizational
and communicative boundaries on computer use
behavior in less developed countries. The proposed
extended TPB model can provide a theoretical
basis to assess the effects of organizational culture
on personal and social beliefs regarding IT adoption, implementation, and confirmation. Social
influence is often produced and expanded from the
interaction and communication among in-group
members (Gefen & Ridings, 2003). Therefore,
the hierarchical structure of Arab organizations
should be reshaped to meet the social process (e.g.,
formal and informal interactions) in IT diffusion
and transfer. Cultural values are implicit beliefs
(Tichy, 1983). Therefore, the influence of organizational culture on computer use behavior can be
posited and identified through particular personal
and social factors (i.e., explicit beliefs). Culturespecific beliefs, such as the Arab sense of time,
can be used to predict the outcomes of IT transfer
(Straub, Loch, & Hill, 2001). Thus, we believe
that a small subset of organizational culture can
explain computer adoption and use in the complicated Arab context. A wider set of organizational
culture can be considered to improve understanding of computer adoption and use in this scope.
In contrast to TAM, which predicts pre-adoption
decisions under individual-oriented contexts, the
proposed extended TPB is a comprehensive model
for assessing the post-adoption behavior of IT
in social-oriented contexts. In addition to TAM
beliefs (PU and PEOU), our empirical findings
suggest that further cultural factors pertaining to
TPB beliefs should be assessed to explain the IT
diffusion processes.
This study will hopefully help managers to develop
on-the-job training plans for the adoption and use
of new computer applications in less developed
countries. The empirical survey reveals that onthe-job training program should be designed by
motivating particular personal and social factors
pertaining to the diffusion of computers in socialoriented contexts, such as the Arabic world. Social
influence may not push knowledge workers to
use IT at Western contexts (Lewis, Agarwal, &
Sambamurthy, 2003), but our empirical findings
of computer usage from Arab end-users imply that
the social influence can spread across the interaction of in-group members (social processes). Most
Arab office workers are willing to use computers if managers can help them understand how
computers can satisfy their special interests, and
if managers can develop group norms for encouraging them to try new applications. Managers
should also recognize that Arab office workers
are likely to adopt computers if they are given
the requisite resources and opportunities. Creating a social-oriented context to fit the organizational culture around Arab end-users can reduce
organizational and communicative boundaries
over the adoption and adoption confirmation of
computers. Unlike Western end-users, who are
trained to act independently, Arab end-users do
not easily form open communicative contexts,
because they often consider an integral group as
a strong in-group boundary (Palmer, Al-Ghofaily,
& Alnimir, 1984). According to Straub’s (1994)
diffusion phase theory, media users with cultural
differences are likely to have different perceptions of social presence and information richness
of electronic media. The empirical findings of
this study imply that social presence (reducing
in-group boundaries) is more influential than
information richness (reducing communicative
boundaries) if top management wants to motivate
Arab end-users’ personal and social beliefs of
computer adoption and use.
194
The Influence of Organizational Communication Openness on the Post-Adoption of Computers
LIMITATIONS
FUTURE RESEARCH
This study has four notable limitations. First, organizational culture is defined in terms of openness
to superior-subordinate relationships and openness
to work communication. Some researchers may
be concerned that this study only examines two
dimensions of organizational culture, thus possibly restricting the understanding of computer
use behavior. However, this study suggests an
extensive direction to address social influence
on IT diffusion by incorporating SIT and TPB to
explain computer usage behavior of Arab endusers. Second, careful application of this study is
necessary when generalizing the empirical findings to assess or predict computer use behavior in
other Arab countries. In particular, the R2 values
of current and continued use of computers are sufficient, which can be enhanced by incorporating
some rational factors such as TAM beliefs (PU
and PEOU) into this research model. Moreover,
additional dimensions of organizational culture,
such as autonomy, work climate, and social
memberships (Gallivan & Srite, 2005; O’Reilly,
Chatman, & Caldwell, 1991), should be considered in future research. Third, Arab and Western
computer end-users were not compared because
of research scopes in this study. Cross-cultural
studies can be considered as an extension of this
study. Fourth, organizational culture is assumed
to affect computer user behavior via personal and
social beliefs, but not vice versa. This assumption is widely accepted in previous literature
(Elbeltagi, McBride, & Hardaker, 2005; Gefen,
Rose, Warkentin, & Pavlou, 2005; Straub, 1994;
Straub, Keil, & Brenner, 1997; Straub, Loch, &
Hill, 2001; Veiga, Floyd, & Dechant, 2001). The
potential influence of computer usage behavior on
organizational culture is not examined because of
the focus of particular research scopes in this study.
Arab end-users may understand the usefulness
and ease-of-use of computers, but they still have
other concerns about personal and social influences on computer use. Putting it into context, this
study adopts an extended TPB model to explain
computer use behavior in Arab organizations.
Computer usage in the workplace is not only a
personal-purpose behavior, but is also an organizational-purpose behavior. The mix-purpose
user behavior can be driven by some particular
cultural factors. The influence of organizational
culture, as a multi-dimensional concept, on computer use behavior needs to be further studied.
Other researchers are encouraged to generalize
this model in other social-oriented cultures, and
examine its influence on different diffusion processes. For instance, the proposed research model
can be combined with Straub’s (1994) diffusion
phase theory to assess the adoption of new IT in
less developed countries. Future research can also
expand this model to assess the social process for
transferring new IT from industrialized countries
to less developed ones (Kedia & Bhagat, 1988).
ACKNOWLEDGMENT
The authors would like to thank three anonymous
reviewers and the editor-in-chief (Professor Tan)
for their invaluable comments and suggestions on
earlier version of this paper. Extended thanks go to
King Abdul-Aziz City for Science and Technology
(KACST), Saudi Arabia for financially supporting this project and the Taiwan National Science
Council (NSC 93-2416-H-364-004) for partially
supporting this cooperative research.
195
The Influence of Organizational Communication Openness on the Post-Adoption of Computers
REFERENCES
Abdul Ghani, J., & Al-Sakran, S. (1988). The
changing data processing environment in Saudi
Arabia. Information & Management, 14, 61–66.
doi:10.1016/0378-7206(88)90049-3
Ajzen, I. (1985). From intentions to actions: A
theory of planned behavior. In Kuhl, J., & Beckman, J. (Eds.), Action-control: From cognition
to behavior (pp. 11–39). Heidelberg: Springer.
Ajzen, I. (1991). The theory of planned behavior.
Organizational Behavior and Human Decision
Processes, 50(2), 179–211. doi:10.1016/07495978(91)90020-T
Ajzen, I., & Madden, T. J. (1986). Prediction of
goal-directed behavior: Attributes, intentions,
and perceived behavioral control. Journal of
Experimental Social Psychology, 22, 453–474.
doi:10.1016/0022-1031(86)90045-4
Al-Farsy, F. (1996). Modernity & tradition: The
Saudi equation. London: Knight Communications Ltd.
Al-Khaldi, M. A., & Wallace, R. S. O. (1999).
The influence of attitudes on personal computer
utilization among knowledge workers: The case of
Saudi Arabia. Information & Management, 36(4),
185–204. doi:10.1016/S0378-7206(99)00017-8
Atiyyah, H. S. (1989). Determinants of computer
system effectiveness in Saudi public organizations. International Studies of Management and
Organization, 19(2), 85–103.
Berkhart, G. E., Goodman, S. E., Mehta, A., &
Press, L. (1998). The Internet in India: Better
times ahead? Communications of the ACM, 41(11),
21–26. doi:10.1145/287831.287835
196
Brown, I., Hoppe, R., Mugera, P., Newman, P.,
& Stander, A. (2004). The impact of national
environment on the adoption of Internet banking:
Comparing Singapore and South Africa. Journal
of Global Information Management, 12(2), 1–26.
doi:10.4018/jgim.2004040101
Brown, R. (1996). Tajfel’s contribution to the
reduction of intergroup conflict. In Robinson, W.
P. (Ed.), Social groups and identities: Developing the legacy of Henri Tajfel (pp. 69–190). UK:
Butterworth-Heinemann.
Chin, W. W. (2001). PLS-Graph User’s Guide,
v. 3.0.
Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. Management Information Systems
Quarterly, 13(3), 319–339. doi:10.2307/249008
Davis, F. D. (1993). User acceptance of information technology: System characteristics, user
perceptions and behavioral impacts. International
Journal of Man-Machine Studies, 38(3), 475–487.
doi:10.1006/imms.1993.1022
Davison, R., & Martinsons, M. (2003). Cultural
issues and IT management: Past and present. IEEE
Transactions on Engineering Management, 50(1),
3–7. doi:10.1109/TEM.2003.808249
DeLone, W. H., & McLean, E. R. (2003). The
DeLone and McLean model of information
systems success: A ten-year update. Journal of
Management Information Systems, 19(4), 9–30.
Ein-Dor, P., Segev, E., & Orgad, M. (1992). The
effect of national culture on IS: Implications for
international information systems. Journal of
Global Information Management, 1(1), 33–44.
Eisenberg, E. M., & Goodall, H. L. (1998). Organizational communication: Balancing creativity
and constraints. New York: St. Martin’s Press.
The Influence of Organizational Communication Openness on the Post-Adoption of Computers
Eisenberg, E. M., & Riley, P. (1991). A communication approach to organizational culture. In
Putnam, L., & Jablin, F. (Eds.), New handbook
of organizational communication. Newbury Park,
CA: Sage Publications.
Gallivan, M., & Srite, M. (2005). Information
technology and culture: Identifying fragmentary
and holistic perspectives of culture. Information
and Organization, 15(4), 295–338. doi:10.1016/j.
infoandorg.2005.02.005
Eisenberg, E. M., & Witten, M. G. (1987).
Reconsidering openness in organizational communication. Academy of Management Review,
12(3), 418–426.
Geertz, C. (1973). The interpretation of cultures.
New York: Basic Books.
Elbeltagi, I., McBride, N., & Hardaker, G. (2005).
Evaluating the factors affecting DSS usage by
senior managers in local authorities in Egypt.
Journal of Global Information Management,
13(2), 42–65. doi:10.4018/jgim.2005040103
Fishbein, M., & Ajzen, I. (1975). Belief, attitude,
intention, and behavior. Reading, MA: AddisonWesley.
Ford, D. P., Connelly, C. E., & Meister, D. B.
(2003). Hofstede’s culture’s consequences and
MIS research. IEEE Transactions on Engineering Management, 50(1), 8–25. doi:10.1109/
TEM.2002.808265
Fornell, C. R. (1982). A second generation of
multivariate analysis. New York: Praeger Special
Studies.
Fornell, C. R., & Larcker, D. F. (1981). Structural
equation models with unobservable variables and
measurement error. JMR, Journal of Marketing
Research, 18(1), 39–50. doi:10.2307/3151312
Fornell, C. R., Tellis, G. L., & Zinkhan, G. M.
(1982). Validity assessment: A structural equation
approach using partial least squares. In Walker,
B. J. (Eds.), An assessment of marketing thought
and practice (pp. 405–409). Chicago: American
Marketing Association.
Gefen, D., Rose, G. M., Warkentin, M., & Pavlou,
P. A. (2005). Cultural diversity and trust in IT
adoption: A comparison of potential e-voters in the
USA and South Africa. Journal of Global Information Management, 13(1), 54–78. doi:10.4018/
jgim.2005010103
Goodman, S. E., & Green, J. D. (1992). Computing in the Middle East. Communications of the
ACM, 35(8), 21–25. doi:10.1145/135226.135236
Hasan, H., & Ditsa, G. (1999). The impact of
culture on the adoption of IT: An interpretive
study. Journal of Global Information Management, 7(1), 5–15.
Hill, C., Loch, K., & Straub, D. W., & EI-Sheshai,
K. (1998). A qualitative assessment of Arab culture
and information technology transfer. Journal of
Global Information Management, 6(3), 29–38.
Hinds, P., & Kiesler, S. (1995). Communication
across boundaries: Work, structure, and use of
communication technologies in a large organization. Organization Science, 6(4), 373–393.
doi:10.1287/orsc.6.4.373
Hofstede, G. (1980). Culture’s consequences:
International differences in work-related values.
Newbury Park, CA: Sage.
Hofstede, G. (1991). Cultures and Organizations.
London: McGraw-Hill.
Hofstede, G. (1993). Cultural constraints in management theories. The Academy of Management
Executive, 7(1), 81–94.
197
The Influence of Organizational Communication Openness on the Post-Adoption of Computers
Igbaria, M., Guimaraes, T., & Davis, G. B. (1995).
Testing the determinants of microcomputer usage
via structural equation model. Journal of Management Information Systems, 11(4), 87–114.
Internet World Stats. (2007). Retrieved from http://
www.internetworldstats.com.
Ives, B., & Jarvenpaa, S. (1991). Wiring the stateless corporation: Empowering the drivers and
overcoming the barriers. SIM Network, 6, 5–8.
Kats, D., & Kahn, R. (1978). The social psychology of organizations. New York: Wiley.
Kedia, B. L., & Bhagat, R. S. (1988). Cultural
constraints on transfer of technology across nations: Implications for research in international and
comparative management. Academy of Management Review, 13(4), 559–571.
Kluckhohn, R. R., & Strodtbeck, F. (1961).
Variation in value orientations. Westport, CT:
Greenwood.
Lewis, W., Agarwal, R., & Sambamurthy, V.
(2003). Sources of influence on beliefs about
information technology use: An empirical study
of knowledge workers. Management Information
Systems Quarterly, 27(4), 657–678.
Loch, K. D., Straub, D. W., & Kamel, S. (2003).
Diffusing the Internet in the Arab world: The role
of social norms and technological culturation.
IEEE Transactions on Engineering Management,
50(1), 45–63. doi:10.1109/TEM.2002.808257
Lohmoller, J. B. (1989). Latent variable path
modeling with partial least squares. Heidelberg:
Physica-Verlag.
Markus, M. L. (1990). Toward a critical mass
theory of interactive media. In Fulk, J., & Steinfield, C. (Eds.), Organization and communication
technology (pp. 194–218). Newbury Park, CA:
Sage.
198
Mathieson, K. (1991). Predicting user intentions:
Comparing the technology acceptance model
with the theory of planned behavior. Information
Systems Research, 2(3), 173–191. doi:10.1287/
isre.2.3.173
Monge, P. R., Rothman, L. W., Eisenberg, E. M.,
Miller, K. I., & Kirste, K. K. (1985). The dynamics
of organizational proximity. Management Science,
31(9), 1129–1141. doi:10.1287/mnsc.31.9.1129
Muthén, L. K., & Muthén, B. O. (2002). How
to use a Monte Carlo study to decide sample
size and determine power. Structural Equation Modeling, 9(4), 599–620. doi:10.1207/
S15328007SEM0904_8
Niederman, F., Boggs, D., & Kundu, S. (2002).
International business and global information
management research: Toward a cumulative tradition. Journal of Global Information Management,
10(1), 33–47. doi:10.4018/jgim.2002010104
Nunnally, J. C. (1978). Psychometric theory (2nd
ed.). New York: McGraw-Hill.
O’Reilly, A. A. III, Chatman, J., & Caldwell, D.
F. (1991). People and organizational culture: A
profile comparison approach to assessing personorganization fit. Academy of Management Journal,
34(3), 497–516.
Palmer, M., Al-Ghofaily, I., & Alnimir, S. (1984).
The behavioral correlates of rentier economies: A
case study of Saudi Arabia. In Stookey, R. (Ed.),
Arabian Peninsula: Zone of ferment. Stanford,
CT: Hoover Institute Press.
Png, I. P., Tan, B. C., & Wee, K. L. (2001).
Dimensions of national culture and corporate
adoption of IT infrastructure. IEEE Transactions
on Engineering Management, 48(1), 36–45.
doi:10.1109/17.913164
Rogers, E. M. (1995). The diffusion of innovation.
New York: Free Press.
The Influence of Organizational Communication Openness on the Post-Adoption of Computers
Rose, G., & Straub, D. (1998). Predicting general
IT use: Applying TAM to the Arabic world. Journal
of Global Information Management, 6(3), 39–46.
Shin, S. K., Ishman, M., & Sanders, G. L. (2007).
An empirical investigation of socio-cultural factors of information sharing in China. Information
& Management, 44(2), 165–174. doi:10.1016/j.
im.2006.11.004
Steele, F. (1975). The open organization: The
impact of secrecy and disclosure on people and
organizations. Reading, MA: Addison-Wesley.
Straub, D., Loch, K., Evaristo, R., Karahanna,
E., & Strite, M. (2002). Toward a theory-based
measurement of culture. Journal of Global Information Management, 10(1), 13–23. doi:10.4018/
jgim.2002010102
Straub, D., Loch, K. D., & Hill, C. E. (2001).
Transfer of information technology to the Arab
World: A test of cultural influence modeling.
Journal of Global Information Management, 9(4),
6–28. doi:10.4018/jgim.2001100101
Straub, D. W. (1994). The effect of culture on IT
diffusion: E-mail and fax in Japan and the U.S.
Information Systems Research, 5(1), 23–47.
doi:10.1287/isre.5.1.23
Straub, D. W., Keil, M., & Brenner, W. H. (1997).
Testing the technology acceptance model across
cultures: A three country study. Information &
Management, 33(1), 1–11. doi:10.1016/S03787206(97)00026-8
Tajfel, H. (1978). Differentiation between social
groups. Cambridge, UK: Cambridge University
Press.
Tajfel, H., & Turner, J. (1979). An integrative
theory of intergroup conflict. In Austin, W. G.,
& Worchel, S. (Eds.), The social psychology of
intergroup relations. Monterey, CA: Brooks/Cole.
Taylor, S., & Todd, P. A. (1995). Understanding
information technology usage: A test of competing models. Information Systems Research, 6(2),
144–176. doi:10.1287/isre.6.2.144
Thompson, R. L., Higgins, C. A., & Howell,
J. M. (1991). Personal computing: Toward a
conceptual model of utilization. Management
Information Systems Quarterly, 15(1), 125–143.
doi:10.2307/249443
Thompson, R. L., & Rose, G. (1994). Information
technology adoption and use. In Proceedings of
the Administrative Sciences Association of Canada
(ASAC) Annual Conference (pp. 16-26).
Tichy, N. (1983). Managing strategic change:
Technical, political, and cultural dynamics. New
York: Wiley.
Triandis, H. C. (1972). The analysis of subjective
culture. New York: Wiley.
Turner, J. C. (1982). Toward a cognitive redefinition of the social group. In Tajfel, H. (Ed.), Social
identity and intergroup relations (pp. 1–40). UK:
Cambridge University Press.
Tushman, M. L., & Scanlan, T. J. (1981).
Boundary spanning individuals: Their role in
information transfer and their antecedents. Academy of Management Journal, 24(2), 289–305.
doi:10.2307/255842
Veiga, J. F., Floyd, S., & Dechant, K. (2001).
Towards modeling the effects of national culture
on IT implementation and acceptance. Journal
of Information Technology, 16(3), 145–158.
doi:10.1080/02683960110063654
Weber, Y., & Pliskin, N. (1996). The effects of information systems integration and organizational
culture on a firm’s effectiveness. Information &
Management, 30(2), 81–90. doi:10.1016/03787206(95)00046-1
World Bank Group. (2007). Saudi Arabia data
profile. World Development Indicators Database.
199
The Influence of Organizational Communication Openness on the Post-Adoption of Computers
Yavas, U., Liqmani, M., & Quraeshi, Z. (1992).
Facilitating the adoption of information technology in a developing country. Information &
Management, 23(2), 75–82. doi:10.1016/03787206(92)90010-D
Yoo, Y., & Alavi, M. (2001). Media and group
cohesion: Relative influences on social presence,
task participation, and group consensus. Management Information Systems Quarterly, 25(3),
371–390. doi:10.2307/3250922
200
Zhao, H., Kim, S., Suh, T., & Du, J. (2007). Social institutional explanation of global Internet
diffusion: A cross-country analysis. Journal of
Global Information Management, 15(2), 28–55.
doi:10.4018/jgim.2007040102
The Influence of Organizational Communication Openness on the Post-Adoption of Computers
APPENDIX: THE MEASUREMENT SCALES
Table A1. Post-Adoption Behavior of Computers
Continued Use (CONTUSE)
CONTUSE1: I predict I will use computers on a regular basis in the future (seven point Likert scale anchored with strongly disagree to
strongly agree).
CONTUSE2: What are the chances in 100 that you will continue as a computer user in the future? (1) Zero; (2) 1-10%; (3) 11-30%; (4)
31-50%; (5) 51-70%; (6) 71-90%; or (7) More than 90%.
Current Use (CURNTUSE)
CURNTUSE1: On an average working day, how much time do you spend using computers (choose one)?: (1) Almost never; (2) Less than
½ hour; (3) From ½ to 1 hour; (4) From 1 to 2 hours; (5) From 2 to 3 hours; (6) More than 3 hours.
CURNTUSE2: On average, how frequently do you use computers?: (1) Less than once a month; (2) Once a month; (3) A few times a month;
(4) A few times a week; (5) About once a day; (6) Several times a day.
CURNTUSE3: How many different computer applications have you worked with or used in your job (choose one)?: (1) None; (2) One;
(3) Two; (4) Three to five applications; (5) Six to ten applications; (6) More than ten applications.
CURNTUSE4: According to your job requirements, please indicate each task you use computers to perform (count of all that apply)?: (1)
Letters and memos; (2) Producing reports; (3) Data storage and retrieval; (4) Making decisions; (5) Analyzing trends; (6) Planning and
forecasting; (7) Analyzing problems and alternatives; (8) Budgeting; (9) Controlling and guiding activities; (10) Electronic communications
with others; (11) Others (please indicate. . .).
201
The Influence of Organizational Communication Openness on the Post-Adoption of Computers
Table A2. Organizational Communication Openness and TPB Beliefs
Please declare how important would it be to you:
Of very little or no importance
……….…
€€€€€€€€€€€€€
€1
2
3
4
5
6
Of utmost important
7
Openness to superior-subordinate relationships (OSSR)
OSSR1: Have a good working relationship with your direct superior.
OSSR2: Be consulted by your direct superior in his/her decision.
OSSR3: Feel free to express disagreement with your superiors.
Openness to work communication (OWC)
OWC1: It is very important to work in a well-defined job situation where the requirements are clear.
OWC2: Work with people who cooperate well with one another.
OWC3: Work in an environment with values like assertiveness, performance, success, and competition prevail among people.
Attitude (ATT): All things considered, my using computers is:
(On a Semantic Differential Scale)
Extremely
Quite To a certain Extent
Neither To a certain Extent
Anchored with 5 attributes:
Quite
Extremely
ATT1: (Bad/Good).
ATT2: (Foolish/Wise).
ATT3: (Unfavorable/Favorable).
ATT4: (Harmful/Beneficial).
ATT5: (Negative/Positive).
Subjective Norms (SN)
SN1: Most people who are important to me think I should use computers.
SN2: Most people who are important to me would want me to use computers.
SN3: People whose opinions I value would prefer me to use computers.
Perceived Behavioral Control (PBC)
PBC1: Using the computer or other alternatives is entirely within my control. “Control” refers to the freedom you feel you have to make
a choice
PBC2: I would be able to use the computer.
PBC3: I have the resources, the knowledge, and the ability to make use of the computer.
PBC4: Given the resources, opportunities, and knowledge it takes you to use computers; it would be easier for me to choose computers
rather than any other means available.
This work was previously published in International Journal of Global Information Management, Volume 17, Issue 3, edited
by Felix B. Tan, pp. 20-41, copyright 2009 by IGI Publishing (an imprint of IGI Global).
202
203
Chapter 10
Chronism Theory, Culture,
and System Delay:
A Longitudinal Study of PostApartheid South Africa
Gregory M. Rose
Washington State University, USA
Carina DeVilliers
University of Pretoria, South Africa
Detmar W. Straub
Georgia State University, USA
ABSTRACT
System response delay has been cited as the single most frustrating aspect of using the Internet and
the most worrisome aspect of Web application design. System response time (SRT) research generally
concludes that delay should be eliminated where possible to as little as a few seconds, even though
delay reduction is costly. Unfortunately, it is not clear if these conclusions are appropriate outside of
the developed world where nearly all of the SRT research has taken place. Cultural effects have been,
hence, generally missing from SRT research. The one SRT study to date outside of the developed world
did report differences using the theoretical construct of cultural chronism, and this finding could limit
the generalizability of SRT research findings from developed countries to many economically developing
nations. However, limitations and potential confounds in this single study render those findings tentative.
The end of Apartheid in South Africa allowed an opportunity to conduct a longitudinal free simulation
experiment that overcomes the critical limitations of this previous research. Subjects were members of
historically polychronic and monochronic groups who had been segregated by Apartheid and now live
in an integrated society with shared infrastructure and computer access. Results find that members of the
historically polychronic group are more accepting of longer delays and are more willing to trade longer
delays for improved functionality than are their historically monochronic counterparts. Furthermore,
tests find that members of the historically monochronic population that came of age in a desegregated,
DOI: 10.4018/978-1-60960-605-3.ch010
Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
Chronism Theory, Culture, and System Delay
majority-polychronic culture appear to be polychronic themselves and to differ significantly from the
older monochronic generation. Results from this study can be applied to design culturally-sensitive applications for users in the developing economies of the world.
INTRODUCTION AND BACKGROUND
A generally accepted rule in software design
is that users do not want to wait for systems to
respond. System response delay has been cited
as the most irritating aspect of using the Internet
(Ewalt, 2002) and the most worrisome aspect of
Web application design (Khosrowpour, 2000).
System Response Time (SRT) research has repeatedly found that users have negative attitudes and
reactions toward waiting as little as 2-4 seconds
for systems to respond (Galletta, Henry, McCoy,
& Polak, 2004; Nah, 2004)
Delay is a byproduct of network-based,
database-intensive, and/or multimedia-intensive
software. Many application designs are impossible
without long delays and can only be mitigated via
some combination of: (a) expensive hardware; (b)
expensive network connections; or (c) the elimination of preferred content and services. None of
these alternatives are attractive to software owners
/ service providers (henceforth referred to simply
as service providers) and each option is costly.
The vexing question for service providers is
how much delay is tolerable in which situations,
how much should be spent on reducing delay,
and how much content should be eliminated in
order to limit delay. Because solutions for reducing delay are costly, understanding when to do
so is most important for those with the fewest
resources. As a result, understanding how and
when to manage delay is critical for the developing world where computing resources such as
network infrastructures are less robust and capital
is scarce. For example, in Brazil, the government
is rolling out heavily subsidized computing and
Internet connectivity for its poor (Benson, 2005).
204
For these millions of Brazilians, low bandwidth
dialup computing will be the standard means of
connectivity for the foreseeable future (Benson,
2005) and long download delays will be commonplace.
What makes this question of knowing when
to eliminate delay so difficult is that delay is still
not a well-understood phenomenon. While it is
generally accepted that delay is to be avoided, it
is unclear that this is true in all cases. Research
to date has concluded that while attitudes toward
delay are predominantly negative, the negative
consequences of delay are not well delineated
(Otto, Najdawi, & Caron, 2000; Rose, Meuter, &
Curran, 2005; Rose & Straub, 2001) and many
scientific findings appear to be counterintuitive
(Davis & Hantula, 2001; Dellaert & Kahn, 1999;
Rajala & Hantula, 2000; Rose, Meuter & Curran
2005).
Among the plethora of such SRT studies, one
study questioned whether SRT was as important
a factor in the developing world as in developed
economies where nearly all SRT research has taken
place thus far. Specifically, Rose et al. (2003)
identified a possible link between the theoretical
construct of cultural chronism and SRT delay
attitudes. Chronism is a cultural characteristic
describing a group’s relationship to time (Hall,
1989, 1990). Monochronism is a cultural trait
where members of a culture conceive of time as
linear and tangible. Where this cultural trait prevails, time moves in a set sequence (A leads to B
leads to C) and time is an asset that can be spent
or wasted. Polychronism, by way of contrast, is a
cultural trait where time is malleable; life is felt to
be less linear than it is a movement of many things
in parallel. Members of polychronic cultures, liv-
Chronism Theory, Culture, and System Delay
ing in predominantly developing economies (Hall,
1989; van Everdingen & Waarts, 2003), appear to
be more tolerant of waiting longer than members
of monochronic cultures, living predominantly
in economically developed nations. However,
potential confounds in Rose et al. (2003) leave it
unclear as to whether cultural differences related
to SRT are scientifically valid or artifactural.
Furthermore, there is no evidence that chronism
differences translate into differences in outcomes
such as design preferences.
The IT market in South Africa is the largest in
Africa and the 20th in the world in overall market
size. It is considered to be 8th world wide in terms
of IT spending as a proportion of Gross Domestic
Product (Bayo, Ekene, Kenneth, & Idowu, 2007).
According to Mohammed and Miller (Mohammed & Miller, 2002) “ … the South African ICT
market was valued at US$4.1 billion, establishing
the country’s position as the ICT leader on the
continent…..For a middle-income, developing
country with unreliable mail and an expensive
courier service, it is noteworthy that 33 percent
of South Africa’s online population has used the
Internet to purchase goods. As for locally relevant
content, at least twenty-seven newspapers have
an online presence; they also serve the South
African expatriate community. The diffusion of
ICT into major industries can be seen primarily
in retailing and financial services;…..”. This indicates that South Africa has a strong ICT industry
and response time is therefore an issue for South
Africans.
In this context, the end of Apartheid in South
Africa offered a unique opportunity to study chronism in a manner that overcomes the limitations
of Rose et al. (2003). We chose to conduct a longitudinal free simulation experiment that confirms
that members of a historically polychronic group
are more accepting of longer delays and are willing
to trade longer delays for improved functionality
than are their historically monochronic counterparts. Furthermore, tests reveal that members of
a historically monochronic population that came
of age in a desegregated, majority-polychronic
culture appear to display characteristics of polychronism and to differ significantly from the older
monochronic generation. Implications are that
chronistic differences related to SRT are indeed
cultural and that software designs for polychronic
populations in much of the developing world
should differ from those traditionally prescribed
for monochronic-dominated cultures.
Most of the research on system response time
has been conducted in Western societies. Culture
may diminish the undesirable effects of system
delay, but it is unlikely that it will eliminate it
entirely. The research question addressed in this
paper is intended to reify the scientific research
to date and, at the same time, explore the dividing line for when cultural chronism matters. It
is: Irrespective of culture, do users greatly prefer
faster system response times?
In this paper, a literature review focusing specifically on SRT and cultural chronism theory is
given. The research methodology is discussed, after which the results of the research are explained.
The contribution of this research, limitations and
future research are presented in the last paragraphs.
LITERATURE REVIEW
Research has found that attitudes toward delay
may be related to cultural perceptions of time.
In monochronic cultures such as the USA, Scandinavia, Germany, etc., lateness is considered to
be slightly offensive and the difference in a few
seconds can be seen as a loss of meaning and
value. Monochronism is primarily found in developed economies of the world (van Everdingen &
Waarts, 2003). In the polychronic cultures of Latin
America, the Arab world, and Sub-Saharan Africa,
the passage of time is not seen as important in and
of itself. Attitudes toward passage of an hour are
not dissimilar to the passage of a day, month or
year for members of these groups (Hall, 1989).
Large portions of the less developed countries of
205
Chronism Theory, Culture, and System Delay
the world are also polychronic (van Everdingen
& Waarts, 2003).
Because these cultural groups relate to the
passage of time in such different ways, Rose et
al. (2003) posited that members of these groups
would respond differently to the download delays they experienced through Web applications.
This lab experiment found evidence that, indeed,
polychronic Peruvians and Egyptians were less
troubled by increases in Web delay than monochronic US and Finnish subjects. Their conclusions
suggest that developers should tailor content to
groups based on readily identifiable traits correlating with cultural chronistic characteristics
such as language and nationality.
While Rose, et al. (2003) does break new
ground and hints at profound ramifications if
its conclusions are substantiated, there are two
shortcomings or potential confounds noted by the
authors. First, whereas subjects were all university
students with access to university laboratories,
the monochronic countries in the study enjoyed
superior infrastructures to the polychronic countries studied. As a result, researchers could not rule
out the possibility that polychronic subjects were
simply accustomed to slow throughput rates due to
weak backbones and poor final mile connections.
This shortcoming begs an important question
about research findings of general culture differences in preferences. Namely, are these differences
in responses to software design cultural or circumstantial? Would Peruvian software users tolerate
delay less if they had been living in Finland for
several years? Would American software users
become more patient with delay if they lived in
Egypt over an extended period of time and had to
share the same wait times as Egyptians? This is a
critical question for researchers since, if preferences and attitudinal differences are merely the
result of individual differences, then prescriptions
for software design need not take culture into
account at all.
With system response time (SRT), this is
clearly an issue. Rose et al. (2003) assumed that
206
the differences were cultural and suggest that
software designers for traditionally-recognized
polychronic populations “will have a freer rein
to include rich content for polychronic cultures”
(p. 40). Basically, they advise designers for polychronic audiences to diverge from commonly
accepted “Western” standards with respect to
delay speeds because “polychronics [are] more
willingly sacrifice speed for other functionality
when compared with monochronic users” (p. 39).
However, if the effects they found from chronism
were not truly cultural, but rather artifactual, the
Peruvian and Egyptian subjects might have been
willing to tolerate longer delays only because
they were accustomed to longer online waits. As a
result, their actual preferences may be no different
from monochronic users in that they would prefer
not to wait any longer than accustomed. Thus by
designing systems that allowed delays exceeding such norms, developers would be increasing
them beyond what those who were theorized to be
polychronic would actually tolerate. This advice,
therefore, would not just be incorrect, but would
likely be punitive to the users and damaging to
the service provider. Lee et al. (2007) found in
their study on mobile Internet users, that “a tendency to monochronic time perception led users
to attribute greater usefulness, greater enjoyment,
greater ease of use, and greater monetary value
to the mobile Internet” (p. 38).
This problem highlights the second acknowledged shortcoming in Rose et al. (2003). Rose et al.
(2003) only tested differences in attitudes toward
SRT. The study did not test for whether there are
differences in preferences for trading longer delays for better functionality between the cultural
groupings. As a result, even if differences in delay
attitudes exist and are cultural, we still do not know
where there are prescriptive differences for building software for polychronic versus monochronic
populations. While these posited differences in
preferences would seem logical, SRT research
has shown that predicted outcomes of delay are
often not supported by empirical research (Otto,
Chronism Theory, Culture, and System Delay
Najdawi & Caron, 2000; Rose, Meuter & Curran
2005; Rose & Straub, 2001). Therefore, without
this empirical support, it is perhaps not prudent
for software designers to presume any differences
in preferences for tradeoffs.
The purpose of the current research is to examine a new setting where these two limitations can
be ruled out in the research design. In the process,
the present study validates and extends the work of
Rose et al. (2003), advancing research on culture
and SRT. Specifically, data was collected over a
three-year period from members of historically
polychronic and monochronic subgroups in South
Africa in order to test for differences in download
delay attitudes and preferences. Additionally,
culture effects were compared among members
of the same ethnicity raised after a transitional
period where their national culture differed from
their predecessors.
During Apartheid (1948-1994) “whites” and
“blacks” were separated by law (apartheid translates into “apartness” in English) and “forbade
most social contacts between races, authorized
segregated public facilities, established separate school systems with lower standards for
non-whites, and restricted each race to certain
jobs” (Dowling, 2002). The cultural setting of
blacks was a segregated environment generally
recognized as polychronic (Prime, 1999; van
Everdingen & Waarts, 2003) while the setting
for whites was monochronic (Morrison, Conway,
& Douress, 1999; Prime, 1999; van Everdingen
& Waarts, 2003). After 1994, in what is known
as “the Transition” (M. Wines, 2003), historically monochronic whites were integrated into
a majority-black nation whose population was
historically polychronic (Wines & LaFraniere,
2004). Because of these historical developments
in South Africa, individuals who grew up prior to
1994 in their separate ethnic groups experienced
different cultural chronisms. Since 1994, they now
share exactly the same access to infrastructure in
their daily lives by law and have so for years now
(Wines & LaFraniere, 2004).
South Africans who grew up after 1994 share
the benefits of that common information and communications infrastructure as well. Unlike those
older South Africans, however, they spent their
formative years in an integrated society. As a result,
younger whites and younger blacks came of age in
a majority-black culture with integrated schools,
neighborhoods, and shopping areas (Wines &
LaFraniere, 2004). Thus, culturally-chronistic
segregation was likely not a phenomenon experienced by these younger South Africans.
All South Africans are now free to attend
integrated Universities and work in integrated
workplaces with shared computer access and
network infrastructure. Furthermore, South Africa
has slow and prohibitively expensive “pay per
minute” home Internet connectivity that dissuades
the use of home network connections even by the
professional class (Goble, 2002) and encourages
the use of fast and free broadband connections at
work and school. Major university network connections in particular are world-class broadband
connections with modern computing equipment
(as indicated by the system configurations in our
methods description below). As a result, members
of South Africa attending a University are likely
to share identical network access. Tests among
this population would eliminate the circumstantial
differences in bandwidth that could not be ruled
out in Rose et al. (2003).
There are two relevant literature streams to
examine cultural chronism and download delay.
These streams support a theoretical distinction
that separates cultural chronism from circumstances. The first of these is the system response
time literature and the second is the literature on
chronism.
System Response Time Literature
System response time (SRT) research has a long
tradition in the Information Systems and Computer
Science communities. A reference list of 108 SRT
articles had been cataloged at the “Time Design”
207
Chronism Theory, Culture, and System Delay
research portal (www.timedsn.net) ranging from
1966 through 2004. SRT research to date is rich
and varied, and the effects of SRT delay are “more
complicated than originally thought” (Davis &
Hantula, 2001) (p. 260).
The literature is diverse with regards to metrics
and outcomes. Measures of delay can be classified
into four categories. The first is perceived wait
time (Hui & Tse, 1996; Ranganathan & Ganapathy,
2002; Straub, Hoffman, Weber & Steinfield, 2002;
Tom, Burns, & Zeng, 1997; Weinberg, 2000),
which is a subject’s perception of how much time
has elapsed. The second is actual or controlled
delay (Davis & Hantula, 2001; Galletta, Henry,
McCoy & Polak, P., 2004; Nah, 2004; Rajala &
Hantula, 2000; Rose, Evaristo & Straub 2003;
Rose, Meuter & Curran 2005), where delay is
either controlled by the researcher or the passage
of actual time is measured and collected. Third is
attitude toward delay (Palmer, 2002; Rose, Meuter
& Curran 2005), where some attitudinal measure
in response to the length of the delay is captured.
Finally, attributable delay (Jacko, Sears, & Borella,
2000; Rose, Meuter & Curran 2005) is a measure
of a subject’s perception of the cause of the delay.
In this last case, where delay is attributed to the
object of study, long delays are theorized to cause
more negative attitudes or intentions toward that
object than if the delay is attributed to something
beyond that object’s control. Outcomes of delay are
diverse as well and include: attitude toward a Web
page (Dellaert & Kahn, 1999; Ramsay, Barbesi,
& Preece, 1998; Rose, Meuter & Curran 2005);
aborting of a page load (Nah, 2004; Rose & Straub,
2001); system satisfaction (Rushinek & Rushinek,
1986; Wirtz & Bateson, 1995); frustration (Barber
& Lucas, 1983); anxiety (Guynes, 1988; Schleifer
& Amick, 1989); effectiveness of and time spent
on material (Davis & Hantula, 2001); patronage
(Rajala & Hantula, 2000); online purchase intent
(Ranganathan & Ganapathy, 2002); and number
of tasks completed (Galletta et al., 2004).
While the measures of download delay and
impacts are various and sundry, the general
208
thrust of these studies is the same. Namely, as
delay increases, there are predominantly negative user responses. This was also illustrated in
research done on waiting for online services,
where it was found that waiting has a significant
impact on the commercial development of the
web (Ryan & Valverde, 2006). Similarly, Devaraj
et al. (Devaraj, Fan, & Kohli, 2006) found that
time responsiveness of the online transaction is
important for customer satisfaction. While not all
studies have found significance (Otto, Najdawi
& Caron, 2000; Rose & Straub, 2001), and some
of the effects are non-linear (Davis & Hantula,
2001; Galletta et al., 2004; Rajala & Hantula,
2000), these studies show consistently that increased delay has negative impacts. Since Rose
et al. (2003) found this relationship was true for
both theorized monochronics and polychronics,
the following hypothesis is proposed:
H1: Irrespective of culture, attitudes toward longer
system response time will be more negative
than toward shorter system response time.
Cultural Chronism Theory
In the search for literature on cultural differences
and evaluation of processing speed, only one article deals with cultural differences in the developing world (2003). Thus existing system response
time (SRT) and Web download time research has
been almost exclusively set in the economicallydeveloped monochronic world (e.g., U.S.A.,
Northern Europe). But in many lesser-developed
countries, polychronism is prevalent (e.g., South
America, Africa) and, given the dearth of research
into the reactions to download delay in these
countries, it is fair to say that the needs of these
system users are not currently well understood.
Unless their unique needs are taken into account,
these users will be subjected to assumed design
preferences based on the science conducted in
monochronic cultures.
Chronism Theory, Culture, and System Delay
This lack of investigations in polychronic
populations should be worrisome to those trying
to adapt conventional wisdom from monochroniccentric research in their software designs for polychronic users. Chronism differences are not trivial
matters of preferences or built up tolerances. The
relationships toward time itself differ greatly. For
example, monochronics: like Scandinavians and
North Americans, prefer to do one thing at a time.
For instance, they do not book several meetings at
the same time. Time is compartmentalised: there’s
time for everything and everything has its time
(Tella, 2000, p. 89).
Hall (1990, p. 6) suggests that monochronic
“Americans think of time as a road [that] has
segments or compartments that are to be kept
discrete.”
In contrast, time in polychronic cultures are
loosely coupled to antecedents and are seen as
occurring mostly in parallel. Polychronic groups
handle multiple tasks and processes concurrently,
like jugglers (Hall, 1966) and show a priority
for personal relationships over timeliness. Tella
(2000) suggests that polychronics: tend to do
several things at the same time or in parallel time
segments. They do not think it odd to answer
their phones while having a videoconference
with foreign partners, or talk to a passing student
by leaving the other videoconference partners to
simply wait (p. 89).
Across sub-Saharan Africa, the cultural tendency to be polychronic is commonly known
as “Africa Time.” Africans find time delineated
more by events than from clocks. Compared to
monochronic Northern Europe or the USA, “in
Africa, time takes on a different meaning: minutes
stretch into hours and timetables are casually discarded” (Katz, 2003). An excerpt of a moderated
discussion by BBC News dealing with African
vs. British senses of time highlights the profound
differences between the cultures in the following
way (Anonymous, 2003):
We Africans exist in time, not for time. Our
life is not defined by seconds, minutes and hours
like machines or robots. Our values are different
from the so-called “developed” world. Time isn’t
money for us. If a friend turns up late for a date,
most likely we won’t be on time either, so we
smile and make the best of the situation. What
we don’t understand is why other people, with
different values and lifestyles, try to impose their
views and call “problems” what we see as natural
and a part of us.
Cultural chronism is a well-understood
characteristic but it has not been definitively
identified to translate into preference and attitude
differences toward SRT. However, sources such
as Hall (Hall, 1989, 1990; van Everdingen &
Waarts, 2003) give no evidence that chronism is
not cultural and therefore we are assuming that
it is cultural and not circumstantial. If there are
indeed cultural effects as they relate to SRT, the
following hypothesis should be true:
H2a: Members of a historically monochromic
culture growing up in a majority monochromic environment will hold more negative
attitudes toward increases in delay than
members of a historically polychromic culture growing up in a majority-polychronic
environment, even if members of both groups
are now living in a majority-polychronic
environment.
Time should be seen as more valuable to
monochronics than polychronics. Per Hall (1989),
monochronics view time as a currency that can
be “saved, spent, wasted, [and] lost” (p. 19). As a
currency, it is valuable and would be spent judiciously on other goods and services. In the case of
software, it could be traded for better functionality. Polychronics are likely to see time as having
less value than monochronics and should be more
willing to trade it for other features. As a result,
if there is variance based on cultural chronism,
a difference in attitudes should be evident when
polychronics and monochronics are offered a
209
Chronism Theory, Culture, and System Delay
tradeoff between delay and content improvements.
The clear, derived hypothesis from this logic is:
H2b: Members of a historically monochronic culture growing up in a majority-monochronic
environment will hold more negative attitudes toward trading increases in delay for
improved functionality than will members of
a historically polychromic culture growing
up in a majority –polychronic environment,
even if members of both groups are now living in a majority-polychronich environment.
Birth with a certain skin pigment or in a certain
household does not negate the impact of the surrounding culture in which a person lives during
her formative years. In fact, if chronism is indeed
a cultural phenomenon, someone who has grown
up in a polychronic environment should exhibit
more tolerance toward increased delay than those
who grow up in a monochronic environment, even
if both are from a historically common subculture.
That is because monochronism must be learned
by experience and it is not a natural state. Per Hall
(1989), growing up within a monochronic culture
is a critical aspect of being monochronic because:
monochronic time is arbitrary and imposed; that
is learned. Because it is so thoroughly integrated
in [a monochronic] culture, it is treated as though
it were the only natural and ‘logical’ way of organizing life. Yet it is not inherent in man’s own
rhythms and creative drives, nor is it existential
in nature (p.20)
As a result, the following hypotheses are
suggested:
H3a: Members of historically monochromic cultures growing up in a majority polychromic
environment will hold more positive attitudes
toward increases in delay than members of
historically monochronic cultures growing
up in a majority monochronic environment,
even if members of both groups are now living in a majority-polychronic environment.
210
H3b: Members of historically monochronic cultures growing up in a majority-polychronic
environment will hold more positive attitudes toward trading increases in delay for
improved functionality than will members of
historically monochromic cultures growing
up in a majority-monochronic environment,
even if members of both groups are now living in a majority-polychronic environment.
Generational differences in attitudes should
not be seen for groups that experienced a common
culture during their formative years. Growing up
in the same cultural chronism conditions, there
should be no significant differences in attitudes
toward increases in delay for members of historically polychronic groups that differ only by age.
Similarly, groups that would have differed historically but instead were integrated during their
formative years should also have no differences
in attitudes toward delay.
H4a: Generational difference in attitudes toward
increases in delay will not be seen within a
historically polychromic culture growing
up in a majority-polychronic environment.
H4b: Generational difference in attitudes toward
trading increases in delat for omproved
functionaliy will not be sen within a historically polychromic culture growing up in a
majority-polychronic environment.
H5a: Members of a historically monochromic
culture growing up in a majority-polychronic
environment will hold similar attitudes
toward increases in delay to members of
a historically polychronic culture growing
up in a majority-polychronic environment.
H5b: Members of a historically monochronic
culture growing up in a majority-polychronic
environment will hold similar attitudes toward trading increases in delay for improved
functionality to members of a historically
polychronic culture growing up in a majoritypolychronic environment.
Chronism Theory, Culture, and System Delay
While we predict generational differences
within groups, there should be no temporal effect
for members of cultural groups in H1-H5. Namely,
once a member of a group takes on cultural characteristics, those characteristics should remain
stable. Culture is a strong influence in people’s
lives, and is particularly resistant to changes over
time (Hofstede, 1980, p. 475). Therefore, we
hypothesize that each of the relationships argued
thus far will hold up over a multi-year study.
Research Model
The research model shown in Figure 1 below
graphically depicts our predictions regarding the
key unresolved issues with respect to chronism
and system response delay.
METHODOLOGY
The hypotheses were tested via a free simulation
experimental method, where the experimental set-
ting duplicates as closely as possible, a real world
situation (Fromkin & Streufert, 1976), (Gefen &
Straub, 2004). The central characteristic of the
free simulation method is to expose participants
to a stimulus of “real-world-like events” (p. 423)
where they are allowed to interact naturally based
on their experience. In a free simulation experiment, variables are not manipulated. They are only
observed and measured (Gefen & Straub, 2004).
Most of the Web system SRT research to date has
been fixed group, controlled experiments using
mock Web sites and browsers with varying degrees
of realism (Ryan & Valverde, 2003). This bias
toward contrived delay environments led (2003)
to conclude that “while it is important to have
absolute control of the variables in experimental
design, a more naturalistic setting might be more
appropriate in order to elicit the true reactions of
subjects” (p. 201).
It is possible that simulated Web browsing may
be the cause of some of the counterintuitive results
in SRT research to date such as non-significance
(Otto, Najdawi & Caron, 2000; Rose, Meuter &
Figure 1. Research Model for Cultural Chronism and System Response Delay
211
Chronism Theory, Culture, and System Delay
Curran 2005; Rose & Straub, 2001), low explained
variation (Galletta et al., 2004; Rose, Meuter &
Curran 2005); and unexpected non-linear relationships (Davis & Hantula, 2001; Galletta et
al., 2004; Otto, Najdawi & Caron, 2000; Rajala
& Hantula, 2000; Rose, Meuter & Curran 2005;
Rose & Straub, 2001). As such, the current design
was selected to stimulate subjects through use of
actual browsers and visits to actual Web sites with
natural delays, similar to Palmer (2002). Because
of the naturalistic methodology chosen, we opted
to measure download delay through estimated and
perceive wait time measures, consistent with other
delay studies (Hui & Tse, 1996; Ranganathan &
Ganapathy, 2002; Rose, Meuter & Curran 2005;
Rushinek & Rushinek, 1986; Straub, Hoffman,
Weber & Steinfield, 2002; Tom et al., 1997;
Weinberg, 2000). As a predictor variable, Davis
et al. (1998) found “actual waits [and] perceived
waits [to be] equally useful” (p. 72). In order to
maximize data sample sizes, a repeated measures
design was adopted, consistent with Rose et al.
(2003). Each subject was exposed to three Web
pages for which a research instrument was completed. These websites were from three different
categories including: a commerce site (Toshiba.
com); a news site (BBC.com); and an entertainment site (Hollywood.com).
For each page, subjects were asked immediately to indicate how many seconds they believed
the page took to load. The purpose of collecting
the perceived wait time (PWT) was to give a
baseline for attitudinal scores and also to look
for outliers that might need to be investigated or
omitted. The plan was to include in our analyses
only those page views with a short response time.
After the PWT was collected, participants were
instructed to answer five attitudinal questions
about delay. For each question, there was a sevenpoint semantic differential scale with anchors of
“not bad” to “very bad.”
The first of the five questions measured the
attitude toward the delay experienced (noted below
as “Att PWT”). This attitude toward the delay was
included to get an anchor point for the subsequent
four questions. The next four questions were designed to be aggregated as a construct measuring
attitude preferences for increases in delay (hereafter referred to as the “Increased Delay Attitude
Construct” or “IDAC”). The items are listed in
Table 1 (in addition to the notation that is used
for each throughout the remainder of the paper).
The IDAC construct includes two items that
measure attitude toward hypothesized increases
in delay of fifteen and sixty seconds. Two additional items measure those increased wait times
in trade for better functionality. The type of functionality was intentionally not defined since that
is beyond the scope of the present project. Subjects
needed to be able to imagine that whatever they
felt could legitimately be categorized as “much
better.” Results would identify different tolerances for tradeoffs in functionality, not prescribe
what the different functionality should be.
Table 1. Items of the Increased Delay Attitude Construct (IDAC)
Notation
Measurement Component
Attitude PWT+15
Consider how long it took for the page to load. If the time it took for the Web page to load were an additional
15 seconds longer, the total wait would then be:
Attitude PWT+60
Consider how long it took for the page to load. If the time it took for the Web page to load were an additional
60 seconds longer, the total wait would then be:
Att trade +15 for better
If the Web page took 15 seconds longer to load, but it offered much better services as a result, the total
wait would then be:
Att trade +60 for better
If the Web page took 60 seconds longer to load, but it offered much better services as a result, the total
wait would then be:
212
Chronism Theory, Culture, and System Delay
Delay times of fifteen and sixty seconds were
chosen for IDAC based on the treatment delay time
of other studies (Dellaert & Kahn, 1999; Ramsay
et al., 1998; Rose, Evaristo & Straub 2003; Rose,
Lees, & Meuter, 2001; Rose, Meuter & Curran
2005; Rose & Straub, 2001; Selvidge, 1999) and
because they captured attitudes and preferences
at both modest and long delay levels. The hope
was to tease out differences between theorized
chronistic groups. Discussions with a group
of older, white South African IT professionals
and academics prior to building the instrument
indicated that these times were long enough to
elicit a change in attitudes from older white South
Africans while still being plausibly long delays.
These IDAC attitudinal questions were then
followed with a scenario asking subjects to find
two pieces of information within the Web site in
a five-minute period. After the tasks were completed or allowed time had elapsed, participants
were asked to evaluate the page using a subset of
eleven items from Palmer’s (2002) Web site assessment instrument. The purpose of the scenario
and evaluation items was twofold. First, it was
included to counteract common methods bias,
and, at the same time provide variety to hopefully
avoid fatigue and hypothesis guessing (Cook &
Campbell, 1979). Second, the items allowed for
a factor analytical test to nomologically validate
the four-item increased delay attitude construct.
Attitudinal data for each of the three pages were
collected and subsequent to this, participants were
asked to share demographic information.
Data were collected in July of 2002 and
February of 2005. Participants were students at
a large, public urban university in South Africa.
Students enrolled in a masters-level informatics
course provided data in 2002 and 2005. First year
students enrolled in an informatics class provided
additional data in 2002. Sample sizes were the
following: Time 1 (2002): older whites (N=24);
older blacks (N=16); younger whites (N=93).
Time 2 (2005): older whites (N=26); older blacks
(N=16); younger whites (N=35). For the testing of
the hypotheses, we set up 5 different subgroups:
younger blacks; younger whites; older blacks;
older whites; and all blacks combined. Younger
groups were 11 years or younger at the Transition. Older groups were 18 years or older at the
Transition. The group of all blacks included any
participant regardless of year of birth (i.e., those
who were 12 through 17 years old at the time of
the Transition were included in addition to the
older and younger participants). There were sufficient data in the 2002 set for analyses for each
of the five subgroups. In 2005, there were not
enough younger black participants for them to be
a separate group in the analyses. The average ages
of members of each group are noted in Table 2.
Machines in the computer lab where the data
were collected were up-to-date, with robust network throughput rates. In fact, machines had been
newly installed when the first data collection in
2002 took place. Since equipment is replaced on
a three-year cycle, the same machines were still
in place for the tests in 2005, which helps to rule
Table 2. Average Age of Participants in 2002 and 2005 Samples
Average Age of Participants
Group
2002
2005
Younger Blacks
18.39
n/a
Younger Whites
18.60
21.35
Older Blacks
33.67
32.80
Older Whites
35.38
34.86
All Blacks
21.19
30.96
213
Chronism Theory, Culture, and System Delay
out an equipment change confound. Configurations in the labs were identical for each station.
ProLine brand computers were used with 1.8 MHz
Intel processors, 40 GB 7200 rpm hard drives,
100 Mbps Intel Ethernet network cards with a
connection to the switch at 100 Mbps, between
switches at 1000 Mbps, between servers at 1000
Mbps, and to the Internet with a 3GB connection.
Each computer had a 17” monitor and was
loaded with Windows XP Professional and Internet Explorer Version 6.
RESULTS
For the analyses below, only South African-born
subject data were included. Of those, only selfreported delays of less than eight seconds were
included in the reported analyses. This cutoff point
was chosen to limit responses to those people
who received a relatively short delay relative to
the IDAC items of fifteen and sixty additional
seconds. Based on this set of conditions, total
sample sizes reported do not equal three page
views per participant. The actual number of valid
data points is proportionately less (roughly 7585% of data points were less than 8 seconds of
perceived wait time).
Reported PWT averages are all less than four
seconds and the differences between mono- and
polychronic groups tested are all less than .67
seconds. These wait times are dwarfed by the
queried IDAC increases of fifteen and sixty
seconds. However, to eliminate the possibility
that differences in PWT were biasing the results,
alternate tests were run to check for any impact
of PWT on the hypothesized outcomes.
In the first alternate tests, PWT thresholds for
groups with significantly higher values were lowered in order to make the PWTs comparable. The
second alternate tests were with all PWTs included.
The third alternate tests flipped which group had
the higher PWT by changing the threshold value
for both groups in different directions.
214
In each of the alternate tests, the directional
hypotheses were confirmed. Likewise, as the
analysis below shows, regardless of which group
has a larger reported PWT in each test, the theorized
monochronic group always has a more favorable
attitude toward the short, initially-reported delay
and the theorized polychronic group always has
a significantly more favorable attitude toward
proposed increases in delay measures. Collectively, differences in reported PWT herein do not
appear to impact the analyses. For parsimony and
clarity, only those with less than eight seconds
are included.
Validation of the Construct
The IDAC was validated with the master’s data set
in 2002. A Principal Component Analysis factor
analysis with a Varimax rotation was performed.
Items included were the four IDAC items and the
eleven items from Palmer (2002). A total of five
components were extracted with eigenvalues > 1.0
(see Table 3). A rotated component matrix found
the IDAC items loaded cleanly in one component
and no other items loaded on that component (see
Table 3). A reliability analysis was run on the four
items to validate their internal consistency. The
Cronbach’s alpha of 0.83 was in excess of Nunnaly’s (1978) recommended threshold of 0.70.
These results indicated the measure was valid and
it was adopted for the entire dataset.
Validation of the Chronism
Manipulation
To validate the groups in our tests each had the
chronism characteristics attributed to them, a
pilot test was run to validate this manipulation.
A ten-item, validated instrument for measuring
cultural chronism was adapted (Bluedorn, Kalliath, Strube, & Martin, 1999) and applied to a
pilot group of South Africans. The instrument
uses a Likert scale from 1 to 7.
Chronism Theory, Culture, and System Delay
Table 3. Factor Analysis Rotated Component Matrix and Eigenvalues
Component
1
Attitude PWT+15
.811
Attitude PWT+60
.829
Att trade +15 for better
.772
Att trade +60 for better
.841
2
3
4
ATTpage A
.775
ATTpage B
.949
ATTpage C
.934
ATTpage D
.682
ATTpage E
.626
ATTpage F
.777
ATTpage G
.625
ATTpage H
.753
ATTpage I
.927
ATTpage J
.909
ATTpage K
Eigenvalues
5
.656
4.017
2.734
1.535
1.481
1.048
Cronbach’s alpha = 0.83
In a confirmatory factor analysis, two items
(items 2 and 10) did not align well with the others and yielded a low Cronbach’s alpha (= .58)
for the reliability of items. Removing these two
items gave clean loadings and an acceptable alpha
(.73). Because of the previously rigorous validation of the instrument (Bluedorn et al., 1999), we
believed elimination of two of ten items was a
modest change and the instrument was still useful
for its current purpose.
The instrument was developed such that higher
scores indicate monochronism and lower scores
indicate relative polychronism. With eight items
and a midpoint of 4 for those items, the mean
chronism score for this instrument was a 32 (=
8 * 4). In our validation test, it was found that
older South African whites (>= 18 years old at
the Transition) were well above this mean with
a chronism score of 37.87 (N = 15). In contrast,
whites who came of age after Apartheid (those
<= 11 years old at the Transition) had a score
below the midpoint with a value of 29.90 (N =
50). An ANOVA comparing these two groups
found significant differences at the .001 level.
These results support our contention that older
whites are more monochronic than their younger
counterparts. Likewise, a test comparing blacks
to older whites found statistical significance at
the .013 level with a relatively polychronic value
of 32.38 for blacks (N = 69) versus the significantly more monochronic 37.87 for older whites.
Collectively, these results validate the chronism
treatment assigning those whites who were 18
years or older at the Transition as monochronic
and blacks and those whites 11 years or younger
at the Transition as polychronic.
215
Chronism Theory, Culture, and System Delay
Table 4. 2002 Attitudes toward Longer SRT by Group
Younger = 19 or less (in 1994 they were 11 or younger)
Older = 26 or more (in 1994 they were 18 or older)
Group
Sample
Size
Mean
Attitude
PWT
Mean
Attitude
PWT+15
Mean
Attitude
PWT+60
ANOVA
p-value
All Subjects
664
2.520
4.400
5.760
0.000***
All Blacks
282
2.730
4.570
5.600
0.000***
All Whites
382
2.370
4.280
5.870
0.000***
Older Blacks
33
2.270
4.000
5.730
0.000***
Older Whites
39
1.900
4.600
6.250
0.000***
Younger Blacks
135
2.770
4.540
5.630
0.000***
Younger Whites
213
2.490
4.170
5.830
0.000***
Hypothesis Tests
Test of Hypothesis H1
Tests of this hypothesis were conducted with
the 2002 data set. ANOVAs were run to test for
significant differences between attitudes toward
PWT and attitudes toward delay if it were increased
by either fifteen or sixty seconds for the page
viewed. All groups held more negative attitudes
toward increases in delay as the proposed delay
increased to an additional fifteen seconds and an
additional sixty seconds. For each of the tested
groups, differences in attitudes were significant
at the .000 level, thus supporting H1. Results of
the analyses are shown in Table 4.
Tests of Hypothesis H2a and H2b
In 2002 older whites had more negative attitudes
toward increases in delay than blacks at the chosen
.05 alpha level. The IDAC measure was 18.27 for
the group of blacks who were 11 or younger at the
Transition (N=135), 17.64 (N=33) for those who
were 18 years old at the Transition, and 18.37 for
blacks of any age combined (N=282). Each was
significantly lower than for those whites who were
18 or older at the Transition (N=39) with a mean
of 21.46, and therefore shows a more negative
216
attitude toward increased delay for older whites
over any category of blacks (at the .001, .003,
and .001 level respectively as shown in Table 5).
In addition, when asked specifically about a
tradeoff for better functionality, black subjects
held more positive attitudes at the alpha level
about trading better functionality for delay than
the older whites. Young blacks, older blacks and
all blacks averaged 3.65, 3.36, and 3.66 respectively for fifteen extra seconds and 4.59, 4.55,
and 4.58 for sixty extra seconds respectively
compared with 4.46 for whites at fifteen seconds
(p-values = .009, .012, and .008) and 6.0 for
sixty seconds (p-values = .000 for each of the
three). It is worth noting that the attitudes of blacks
at a sixty second level is practically the same as
it is for older whites at the fifteen second level.
This level is very close to the middle of the sevenpoint scale and indicates a moderate attitude toward a one-minute delay for improved functionality. Meanwhile, at the sixty-second level, older
whites are near the extreme end of the seven-point
scale and show a strong negative feeling toward
this length of wait.
Differences are more pronounced if the
baseline attitude toward the actual delay is taken
into account. In each case, black groups had less
favorable attitudes (i.e., were more indifferent)
than the older whites for the very short actual
Chronism Theory, Culture, and System Delay
Table 5. 2002 Attitude toward Increased Delay for Older Whites vs. Blacks
Younger = 19 or less (in 1994 they were 11 or younger)
Older = 26 or more (in 1994 they were 18 or older)
Means
Younger
Blacks
Older
Whites
P-value
Older
Blacks
Older
Whites
P-value
All
Blacks
Older
Whites
P-value
Average PWT
3.530
3.950
0.162
3.650
3.950
0.464
3.580
3.950
0.181
Attitude PWT
2.770
1.920
0.001***
2.270
1.920
0.169
2.730
1.920
0.001***
Att trade +15 for better
3.650
4.460
0.009***
3.360
4.460
0.012**
3.660
4.460
0.008***
Att trade +60 for better
4.590
6.000
0.000***
4.550
6.000
0.000***
4.580
6.000
0.000***
IDAC
18.270
21.460
0.001***
17.640
21.460
0.003***
18.370
21.460
0.001***
Net trade +15 for better
0.870
2.540
0.000***
1.090
2.540
0.002***
0.930
2.540
0.000***
Net trade +60 for better
1.810
4.080
0.000***
2.270
4.080
0.000***
1.840
4.080
0.000***
Net IDAC
7.280
13.770
0.000***
8.550
13.770
0.000***
7.480
13.770
0.000***
N=135
N=39
N=33
N=39
N=282
N=39
delay. These differences are significantly so for
young and all-blacks categories at the .001 level
(but not significant for older blacks). Attitudes
toward PWT indicate that short delay was more
favorable to whites and less important to blacks.
Collectively these differences indicate that
black subjects rated various levels of delay in a
narrower band, hovering in the middle of the scales
compared to older whites. This means simply that
black subjects were less concerned with delay at
either end of the attitude scale. Namely, the older
whites viewed the shortest delay with a better attitude than blacks but viewed longer delays more
negatively than blacks. The narrower change in
attitude differences for blacks when compared to
the older whites can also be seen in the net attitude
change measures in Table 5. In each case, the net
change in attitudes between the initial PWT and
proposed increased delays were significantly
smaller for the presumed polychronic subjects than
for the monochronic group. This result is logically
consistent given that polychronics view delay as
a less important phenomenon than monochronics.
Polychronic subjects are apparently less excited
about very short delays and are more tolerant
of very long delay than are their monochronic
counterparts.
The basic patterns of relationships in the 2002
data set were also found in the 2005 data set (see
Table 6). Sample sizes for blacks were too small
for the younger black group (N=11) and as a
result, only older black (N=35) and all black (N
=46) subcategories can be analyzed and reported.
Significant differences were seen in the IDAC
17.85 for older blacks, 18.06 for all blacks, and
21.35 for older whites (p-values = .002 and .001).
Differences were also found for preferences for
trading functionality for extra delay.
Older blacks and all blacks averaged attitudes
of 3.63 and 3.78 respectively for fifteen extra
seconds, and 4.51 and 4.59 for sixty extra seconds
in trade for better functionality respectively.
Older whites averages were 4.50 for an extra
fifteen seconds (p-values = .015 and .027) and
5.94 for an extra sixty seconds (p-values = .000
and .000). Together with the 2002 data results,
these data strongly support H2a and H2b.
The basic patterns of relationships in the 2002
data set were also found in the 2005 data set (see
Table 6). Sample sizes for blacks were too small
for the younger black group (N=11) and as a
result, only older black (N=35) and all black (N
=46) subcategories can be analyzed and reported.
Significant differences were seen in the IDAC
217
Chronism Theory, Culture, and System Delay
Table 6. 2005 Attitude toward Increased Delay for Older Whites vs. Blacks
Younger = 22 or less (in 1994 they were 11 or younger)
Older = 29 or more (in 1994 they were 18 or older)
Means
Older
Blacks
Older
Whites
P-value
All
Blacks
Older
Whites
P-value
Average PWT
2.290
2.700
0.129
2.200
2.700
0.044*
Attitude PWT
2.110
1.720
0.126
2.020
1.720
0.193
Att trade +15 for better
3.630
4.500
0.015**
3.780
4.500
0.027**
Att trade +60 for better
4.510
5.940
0.000***
4.590
5.940
0.000***
IDAC
17.850
21.350
0.002***
18.060
21.350
0.001***
Net trade +15 for better
1.500
2.830
0.002***
1.760
2.830
0.006***
Net trade +60 for better
2.400
4.300
0.000***
2.560
4.300
0.000***
9.400
14.770
0.000***
9.970
14.770
0.000***
N=35
N=69
N=46
N=69
Net IDAC
17.85 for older blacks, 18.06 for all blacks, and
21.35 for older whites (p-values = .002 and .001).
Differences were also found for preferences for
trading functionality for extra delay.
Tests of Hypothesis H3a and H3b
In both 2002 and 2005, our analyses suggest that
cultural chronism has an impact and that the data
support H3a and H3b. Whites who came of age
during Apartheid held significantly more nega-
tive attitudes toward delay than those who were
eleven years old or younger during the Transition.
As indicated in Tables 7 and 8, older whites had
an IDAC score of 21.46 and 21.35 in 2002 and
2005 respectively. Younger whites averaged significantly lower attitudes of 18.31 (p-value = .000)
and 19.29 (p-value = .009) for 2002 and 2005.
Significantly different preferences in tradeoffs
of delay for improved software services were also
seen. Older whites were less favorable to a trade
of an extra fifteen seconds (in 2002 and 2005
Table 7. 2002 Attitude toward Increased Delay for Older Whites vs. Younger Whites
Younger = 19 or less (in 1994 they were 11 or younger)
Older = 26 or more (in 1994 they were 18 or older)
Means
Younger
Whites
Older
Whites
P-value
Average PWT
3.990
3.950
0.884
Attitude PWT
2.490
1.920
0.020***
Att trade +15 for better
3.490
4.460
0.001***
Att trade +60 for better
4.860
6.000
0.000***
IDAC
18.310
21.460
0.000***
Net trade +15 for better
1.000
2.540
0.000***
Net trade +60 for better
2.370
4.080
0.000***
Net IDAC
8.360
13.770
0.000***
N=213
N=39
218
Chronism Theory, Culture, and System Delay
Table 8. 2005 Attitude toward Increased Delay for Older Whites vs. Younger Whites
Younger = 22 or less (in 1994 they were 11 or younger)
Older = 29 or more (in 1994 they were 18 or older)
Means
Younger
Whites
Older
Whites
P-value
Average PWT
3.350
2.700
0.008***
Attitude PWT
1.990
1.720
0.141
Att trade +15 for better
3.500
4.500
0.000***
Att trade +60 for better
5.320
5.940
0.014**
IDAC
19.290
21.350
0.009***
Net trade +15 for better
1.510
2.833
0.000***
Net trade +60 for better
3.333
4.300
0.002***
11.350
14.770
0.001***
N=78
N=69
Net IDAC
respectively: older whites = 4.46 and 4.50;
younger whites = 3.49 and 3.50; p-values = .001
and .000). Likewise, they viewed trading improved
services for an extra sixty-second delay more
negatively (in 2002 and 2005 respectively: older
whites = 6.00 and 5.94; younger whites = 4.86
and 5.32; p-values = .000 and .014). As in comparing blacks to older whites, the differences are
more pronounced when the baseline attitude toward the perceived wait time are considered.
Tables 7 and 8 show a narrower band of attitudes for young whites and a wider difference
for older whites. Once again, the older whites held
more extreme attitudes toward short and proposed
long delays than the hypothesized polychronic
group. Further, it is worth noting that the mean
attitude at a sixty-second level for younger whites
is relatively close to those of older whites at the
fifteen-second level (albeit less so in the 2005
data than in the 2002 data). As was the case with
blacks, attitudes for younger whites for a sixtysecond tradeoff is relatively close to the middle
of the seven-point scale and indicates a moderate
attitude toward a one minute delay for improved
functionality.
Tests of Hypothesis H4a and H4b
Table 9 shows no significant differences between
younger and older blacks and, thus, provides support for both H4a and H4b. In the 2002 dataset,
the IDAC variable averages 18.27 for younger
blacks and 17.64 for older blacks (p-value = .544).
Attitudes toward trading delay for functionality
were also insignificantly different. Tradeoffs at
fifteen seconds evoked an attitude score of 3.65
for younger and 3.36 for older blacks (p-value
.384) and at sixty seconds, scores of 4.59 and
4.55 respectively (p-value.894). Unfortunately,
there were not enough young blacks in the 2005
sample to examine this group.
Tests of Hypothesis H5a and H5b
As shown in Tables 10 and 11, no significant
differences were found between the young white
subjects and any category of black subjects for
the IDAC measure. In 2002, the averages were
18.31 for younger whites, 18.27 for younger
blacks, 17.64 for older blacks, and 18.37 for the
combination of all black subjects (p-values of
.940, .439, and .902 respectively). Similarly, in
2005, the values were 19.29 for younger whites,
219
Chronism Theory, Culture, and System Delay
Table 9. 2002 Attitude toward Increased Delay for Older Blacks vs. Younger Blacks
Younger = 19 or less (in 1994 they were 11 or younger)
Older = 26 or more (in 1994 they were 18 or older)
Means
Younger
Blacks
Older
Blacks
P-value
Average PWT
3.530
3.650
0.695
Attitude PWT
2.770
2.270
0.087
Att trade +15 for better
3.650
3.360
0.384
Att trade +60 for better
4.590
4.550
0.894
IDAC
18.270
17.640
0.544
Net trade +15 for better
0.870
1.090
0.577
Net trade +60 for better
1.810
2.270
0.283
7.280
8.550
0.363
N=135
N=33
Net IDAC
17.85 for older blacks, and 18.06 for all blacks
(p-values of .174 and .207). As stated above,
there were not enough younger black subjects to
conduct tests for that group.
When asked to trade functionality for delay,
no difference is found at the fifteen-second level.
In the 2002 dataset, attitude scores were 3.49 for
younger whites, 3.65 for younger blacks (p-value
= .360), 3.36 for older blacks (p-value = .676),
and 3.66 for all blacks (p-value = .250). In the
2005 data, younger whites averaged an attitude
score of 3.5, while older blacks averaged 3.63
(p-value = .702) and all blacks averaged 3.78
(p-value = .357).
For the sixty-second tradeoff, there are no significant differences in the 2002 dataset, but there
Table 10. 2002 Attitude toward Increased Delay for Younger Whites vs. Blacks
Younger = 19 or less (in 1994 they were 11 or younger)
Older = 26 or more (in 1994 they were 18 or older)
Means
Younger
Blacks
Younger
Whites
P-value
Older
Blacks
Younger
Whites
Pvalue
All
Blacks
Younger
Whites
P-value
Average PWT
3.530
3.990
0.012*
3.650
3.990
0.287
3.580
3.990
0.006***
Attitude PWT
2.770
2.490
0.096
2.270
2.490
0.422
2.730
2.490
0.072
Att trade +15 for
better
3.650
3.490
0.360
3.360
3.490
0.676
3.660
3.490
0.250
Att trade +60 for
better
4.590
4.860
0.114
4.550
4.860
0.261
4.580
4.860
0.063
IDAC
18.270
18.310
0.940
17.640
18.310
0.439
18.370
18.310
0.902
Net trade +15 for
better
0.870
1.000
0.521
1.090
1.000
0.747
0.930
1.000
0.669
Net trade +60 for
better
1.810
2.370
0.016**
2.270
2.370
0.794
1.840
2.370
0.006**
7.280
8.360
0.146
8.550
8.360
0.879
7.480
8.360
0.154
N=135
N=213
N=33
N=213
N=282
N=213
Net IDAC
220
Chronism Theory, Culture, and System Delay
Table 11. 2005 Attitude toward Increased Delay for Younger Whites vs. Blacks
Younger = 22 or less (in 1994 they were 11 or younger)
Older = 29 or more (in 1994 they were 18 or older)
Means
Older
Blacks
Younger
Whites
P-value
All
Blacks
Younger
Whites
P-value
Average PWT
2.290
3.350
0.000***
2.200
3.350
0.000***
Attitude PWT
2.110
1.990
0.654
2.020
1.990
0.892
Att trade +15 for better
3.630
3.500
0.702
3.780
3.500
0.357
Att trade +60 for better
4.510
5.320
0.018*
4.590
5.320
0.019*
IDAC
17.850
19.290
0.174
18.060
19.290
0.207
Net trade +15 for better
1.514
1.513
0.997
1.760
1.510
0.481
Net trade +60 for better
2.400
3.333
0.038*
2.560
3.330
0.065
Net IDAC
9.400
11.350
0.176
9.970
11.350
0.306
N=35
N=78
N=46
N=78
are significant differences in the 2005 data set.
For 2002, attitude scores were 4.86 for younger
whites, 4.59 for younger blacks (p-value = .114),
4.55 for older blacks (p-value = .261), and 4.58
for all blacks (p-value = .063). In the 2005 data,
younger whites averaged an attitude score of
5.32, while older blacks averaged 4.51 (p-value
= .018) and all blacks averaged 4.59 (p-value =
.019). Likewise, the net changes in attitude at the
sixty-second level are significantly different for
two of three groups in 2002 and one group in 2005.
Overall, therefore, it appears that attitudes
between the samples differ when it comes to
trading a very long wait for better functionality.
Analyses indicate that the younger South African
whites may be more polychronic in general than
older whites, but are slightly more monochronic
in their preferences at extreme levels of delay than
South African blacks.
CONTRIBUTIONS AND DISCUSSION
This experiment produces evidence that indeed
chronism impacts attitudes and preferences toward
SRT. Each of the proposed hypotheses in this study
was confirmed in both 2002 and 2005 datasets.
Contributions for Practice
Results from this study can be applied to make
culturally-sensitive application designs in both the
developed and developing economies of the world.
Specifically, it appears there are real cultural differences in preferences for tradeoffs of functionality
for increased delay. The currency of time that is so
valuable in monochronic cultures and often noted
in USA software design guidelines (Galletta et al.,
2004; Nah, 2004; Shneiderman, 1998; Tedeschi,
1999) appears to be less significant in polychronic
cultures. Given that many polychronic cultures are
in lesser-developed economies, there are fewer
resources and poorer access to speedy computers
and networks. When governments or service providers in these countries build applications, they
should be wary of adopting USA design standards.
Instead, the objective appears to be to offer better services even if it means delays of more than
sixty seconds based on the results above. This is
strongly supported by the data, which indicate that
polychronic attitudes toward trading functionality
for an additional sixty second delay is relatively
moderate (a mean of near 4.5). Furthermore, our
results suggest that the barriers to entry to offering
viable service provision are lower in polychronic
221
Chronism Theory, Culture, and System Delay
cultures than in a monochronic cultures. This is
because many of the remedies for slow SRT are
expensive and require speedy servers and broadband connectivity. These expenses are mandatory
based on monochronic specifications. However,
service providers in polychronic cultures apparently could suffice with more affordable narrowband connections and less responsive hardware.
Likewise, it should be noted that there appears to
be fewer benefits for reducing speeds to very low
levels for polychronics relative to monochronics.
Results of the tests herein indicate that attitudes
toward near-zero delay are not as positive for
polychronics as they are for monochronics. As a
result, it seems that not only are there more degrees
of freedom for increasing delay for polychronics,
but also that there is less to be gained by reducing
delays to USA standards of very low levels for
polychronics. Taken collectively, it would seem
that developers should think about content first
and foremost when designing for polychronic
populations. Monochronic-focused applications
in contrast should continue to follow the advice
of the traditional SRT research.
Contributions for Research
Beyond prescriptive benefits to practitioners,
results of this study advance the understanding
of cultural chronism in the academic community.
Research in chronism has been advanced by showing that chronism is not a trivial attribute that is
discarded as circumstances change. Over time,
older whites retained their monochronism even as
the world around them and younger whites turned
polychronic. The fact that younger whites do differ
from older whites shows that a cultural attribute can
change over a single generation within an ethnic
group. It is clear from the data analyses conducted
here that younger South African whites after the
Transition have cultural responses to chronism that
differ from their elders, but are remarkably similar
to those of South African blacks. It is extraordinary that this change would have occurred within
222
a single generation for so many in our samples.
While this possible cultural change reflects only
one cultural value among the many that are fully
experienced by humans (McCoy, 2005; Straub,
Hoffman, Weber & Steinfield, 2002), it does hint
that cultural exposure during formative years can
result in very rapid change from traditional norms
within a subculture.
The final contribution from this research is
specific to South Africans. This research provides
insights into the culture of the new South Africa.
Cultural perspectives from sources such as Hofstede are clearly outdated since they took place
prior to the Transition. This new South Africa is
currently an unknown culture. The current research
project participates in the nascent but important
research into understanding the new “Rainbow
Nation” (Hugo, 2000). In doing so, it contributes to the call to arms by Hugo (2000; 2002)
to understand how to create culturally-sensitive
software applications for the various subcultures
of post-Apartheid South Africa. South Africa can
“no longer afford to exclude any individual or any
sector of society from access to, and benefiting
from information technology” (Hugo, 2000).
While appreciation of the link between culture
and information systems (IS) has been recognized
as important throughout the IS community (see
such papers as Ford et al. (Ford, Connelly, & D.
B. Meister, 2003) or Gallupe and Tan (1999) for
an extensive review of culture and IT literature),
it is all the more critical in South Africa because
of the unique history and scars left by Apartheid.
Specifically, “without proper understanding of the
cultural variables involved, [information systems
and technology] threaten to expand, rather than
resolve, certain cultural conflicts” (Hugo, 2000).
LIMITATIONS AND
FUTURE RESEARCH
The advantage of free simulation experiments
is the realism of the setting for subjects and the
Chronism Theory, Culture, and System Delay
ability of subject responses to range freely over
the independent variables of interest (Fromkin
& Streufert, 1976). We chose to provide a real
Web browsing experience and capture delay
with the PWT construct. Our method did so at
the expense of controlling delay. As indicated
in Rose et al. (Rose, Meuter & Curran 2005;
Rose & Straub, 2001), findings based on single
measures of delay likely need to be repeated with
alternative measures (such as attributable delay or
controlled delay) to validate and understand the
studied phenomena. A lab or field experiment that
controlled the delay time (as in (Davis & Hantula,
2001; Galletta et al., 2004; Nah, 2004; Rajala &
Hantula, 2000; Rose, Meuter & Curran 2005))
or one that capture attributable delay should be
conducted to validate these findings.
In addition for varying the measures of delay,
future research should vary the means of identifying chronism. Researchers interested in chronism
should not only control SRT, but should also consider direct measures of cultural characteristics.
The latter is an area that is still underdeveloped
and cross-cultural IT researchers are now being challenged to go beyond Hofstede and Hall
in their conceptualizations of culture (McCoy,
2005; Straub, Loch, Evaristo, Karahanna, & Srite,
2002). Regardless of how chronism is identified,
research should also be conducted to identify
where pockets of mono and polychronism exist
in target populations. This will allow practitioners
to target software designs appropriately based on
SRT standards.
A question that may arise is whether economic
status confounds chronism. This is possible, but
results herein make them appear unlikely. One
of the key points of the findings of this research
is that young whites, which are still living with a
high economic status in South Africa, have turned
polychronic and stayed that way between 2002
and 2005. This may, however, be a question for
future research to refine.
Future research should also begin to look into
the tradeoffs suggested by this study. If polychron-
ics are indeed more willing to trade delay for some
improvement in services, it is important to find out
what those services are for the targeted groups.
Software developers should offer screens with fast
downloads, for example without the graphics, for
monochronics and another option to download the
full function screen for polychronics (Hoxmeier &
DiCesare, 2000). One needs to look at how much
work needs to be done by software developers to
make noticeable differences in the download time
relative to different software applications. Galetta
et al. (2004) found that attitudinal effects of delay
were alleviated during long delays.
This need for software specification research
outside of the developed economies of the USA
and Northern Europe are made clear by the findings herein. More research needs to be done to
identify design preferences for developing nations.
Future research also needs to look at the fact that
polychronics are naturally good at multitasking and may not mind a longer download time
because they can do other tasks while waiting.
Monochronics are not good at multitasking and
will get frustrated by the longer download times.
CONCLUSION
The current study examines the impact of cultural
chronism on responses to download delay. Knowledge of possible cultural effects can aid designers
in intelligently planning the trade-off between
content size and delay effects. For those cultures
that are more accepting of delay, richer websites
can be created to attract and retain customers.
The vast majority of prior work in this critical area has been in developed countries where
monochronism is the dominant cultural tendency.
Therefore, before the current study, we have had
little scientific evidence for the presence of such
effects in the developing world, which is overwhelmingly polychronic.
Using a robust free simulation experimental design, the present work examined monochronic and
223
Chronism Theory, Culture, and System Delay
polychronic cultural effects in an unconfounded
single network setting. This design allowed us to
rule out the possible confound of infrastructural
differences. Findings were supportive and predicted differences were found between mono- and
polychronics in both the initial data gathering and
in the replicated tests three years later.
Research in the developing world is lagging
in many areas, but this study is an addition to this
small but growing pool. Over time, it is hoped that
an expanding tradition of work will allow us to
learn ways in which systems can be better designed
and networks can be configured to accommodate
cultural features of these regions of the world.
REFERENCES
Anonymous. (2003, October 28). Can Africa keep
time? BBC News. Retrieved from http://news.bbc.
co.uk/1/hi/world/africa/3211923.stm
Barber, R. E., & Lucas, H. C. J. (1983). System
response time, operator productivity and job satisfaction. Communications of the ACM, 26(11),
972–986. doi:10.1145/182.358464
Bayo, M. I., Ekene, N. S., Kenneth, A., & Idowu,
S. F. (2007). Software development: An attainable
goal for sustainable growth in developing nations.
International Journal of Physical Sciences, 2(12),
318–323.
Benson, T. (2005, March 28). Brazil: Free
software’s biggest and best friend. NYTimes
Online. Retrieved from http://www.nytimes.
com/2005/2003/2029/technology/2029computer.
html
Bluedorn, A. C., Kalliath, T. J., Strube, M. J.,
& Martin, G. D. (1999). Polychronicity and the
Inventory of Polychronic Values (IPV): The development of an instrument to measure a fundamental
dimension of organizational culture. Journal
of Managerial Psychology, 14(3/4), 205–230.
doi:10.1108/02683949910263747
224
Cook, T. D., & Campbell, D. T. (1979). Quasi
experimentation: Design and analytical issues
for field settings. Chicago, IL: Rand McNally.
Davis, E. S., & Hantula, D. A. (2001). The effects
of download delay on performance and end user
satisfaction in an Internet tutorial. Computers in
Human Behavior, 17(3), 249–268. doi:10.1016/
S0747-5632(01)00007-3
Davis, M. M., & Heineke, J. (1998). How disconfirmation, perception and actual waiting times
impact customer satisfaction. International Journal of Service Industry Management, 9(1), 64–73.
doi:10.1108/09564239810199950
Dellaert, B. G. C., & Kahn, B. E. (1999). How tolerable is delay? Consumers’ evaluations of Internet
Web sites after waiting. Journal of Interactive Marketing, 13(1), 41–54. doi:10.1002/(SICI)15206653(199924)13:1<41::AID-DIR4>3.0.CO;2-S
Devaraj, S., Fan, M., & Kohli, R. (2006). Examination of online channel preference: Using the
structure-conduct-outcome framework. Decision
Support Systems, 42, 1089–1103. doi:10.1016/j.
dss.2005.09.004
Dowling, M. (2002). The electronic passport to
Apartheid. Retrieved from http://www.mrdowling.com/610-apartheid.html
Ewalt, D. M. (2002, February 19). Count to
10, then read this. Information Week. Retrieved
from http://www.informationweek.com/story/
IWK20020219S20020013
Ford, D. P., Connelly, C. E., & Meister, D. B.
(2003). Information Systems research and Hofstede’s culture’s consequences: An uneasy and
incomplete partnership. IEEE Transactions on Engineering Management, 50(1), 8–25. doi:10.1109/
TEM.2002.808265
Fromkin, H. L., & Streufert, S. (1976). Laboratory
experimentation. In Handbook of industrial and
organizational psychology (pp. 415–465). Chicago, IL: Rand McNally Publishing Company, Inc.
Chronism Theory, Culture, and System Delay
Galletta, D., Henry, R., McCoy, S., & Polak, P.
(2004). Sensitivity to Web delays: A contingency
analysis of users’ tolerance for slow Web sites.
Journal of AIS, 5(1), 1–28.
Hugo, J. (2002). HCI and multiculturalism in
Southern Africa. Retrieved from http://www.
chi-sa.org.za/articles/HCIculture.htm
Gallupe, R. B., & Tan, F. (1999). A research manifesto for global information management. Journal
of Global Information Management, 7(1), 5–18.
Hui, M. K., & Tse, D. K. (1996). What to tell
consumers in waits of different lengths: An integrative model of service evaluation. Journal of
Marketing, 60(2), 81–90. doi:10.2307/1251932
Gefen, D., & Straub, D. W. (2004). Consumer
trust in B2C e-commerce and the importance of
social presence: Experiments in e-products and
e-services. Omega, 32, 407–424. doi:10.1016/j.
omega.2004.01.006
Jacko, J. A., Sears, A., & Borella, M. S. (2000).
The effect of network delay and media on
user perceptions of Web resources. Behaviour
& Information Technology, 19(6), 427–439.
doi:10.1080/014492900750052688
Goble, C. E. (2002). The Internet – An elitist luxury
or the people’s necessity? South African Journal
of Information Management, 4(3). Retrieved
from http://general.rau.ac.za/infosci/raujournal/
student19.14nr13.asp?print=11
Katz, L. (2003, January 16). Off the back of
a truck. Guardian Unlimited. Retrieved from
http://travel.guardian.co.uk/activities/wildlife/
story/0,10234,874764,874700.html
Guynes, J. L. (1988). Impact of system response
time on state anxiety. Communications of the
ACM, 31(3), 342–347. doi:10.1145/42392.42402
Hall, E. T. (1966). The hidden dimension. Garden
City, NY: Doubleday.
Hall, E. T. (1989). Beyond culture (2nd ed.). New
York, NY: Anchor Books.
Hall, E. T. (1990). The silent language (2nd ed.).
Garden City, NY: Anchor Books.
Hofstede, G. (1980). Culture’s consequences:
International differences in work-related values.
Beverly Hills, CA: Sage.
Hoxmeier, J. A., & DiCesare, C. (2000). System
response time and user satisfaction: An experimental study of browser-based applications. Paper presented at the Association of Information
Systems Americas Conference.
Hugo, J. (2000). South Africa’s first HCI Conference: CHI-SA 2000 - A rainbow of opportunity.
Paper presented at the First Conference on Human Computer Interaction in Southern Africa,
South Africa.
Khosrow-Pour, M., & Herman, N. (2000). Webenabled technologies assessment and management: Critical issues, challenges and trends. In
Khosrow-Pour, M. (Ed.), Managing Web-enabled
technologies in organizations: A global perspective (1st ed., pp. 1–22). Hershey, PA: Idea Group
Publishing. doi:10.4018/9781878289728.ch001
Lee, I., Choi, B., Kim, J., & Hong, S. (2007).
Culture-technology fit: Effects of cultural characteristics on the post-adoption beliefs of mobile
Internet users. International Journal of Electronic
Commerce, 11(4), 11–51. doi:10.2753/JEC10864415110401
McCoy, S., Galetta, D. F., & King, W. R. (2005).
Integrating culture into IS research: The need for
current individual-level measures. [CAIS]. Communications of the Association for Information
Systems, 15, 211–224.
Mohammed, T., & Miller, J. (2002). South African
sample country profile: The global competitiveness report 2001-2002.Unpublished manuscript.
Morrison, T., Conway, W. A., & Douress, J. J.
(1999). Dun & Bradstreet’s guide to doing business around the world. Prentice-Hall.
225
Chronism Theory, Culture, and System Delay
Nah, F. F.-H. (2004). A study on tolerable waiting
time: How long are Web users willing to wait?
Behaviour & Information Technology, 23(3),
153–163. doi:10.1080/01449290410001669914
Nunnaly, J. C. (1978). Psychometric theory (2nd
ed.). New York, NY: McGraw Hill.
Otto, J. R., Najdawi, M. K., & Caron, K. M.
(2000). Web-user satisfaction: An exploratory
study. Journal of End User Computing, 12(1),
3–10. doi:10.4018/joeuc.2000100101
Palmer, J. (2002). Web site usability, design,
and performance metrics. Information Systems Research, 13(2), 151–167. doi:10.1287/
isre.13.2.151.88
Prime, N. (1999). Cross-cultural management in
South Africa: Problems, obstacles, and agenda for
companies. Paper presented at the Seventh Cross
Cultural Research Conference, Cancun.
Rajala, A. K., & Hantula, D. A. (2000). Toward
a behavioral ecology of consumption: Delayreduction effects on foraging in a simulated Internet mall. Managerial and Decision Economics,
21(3-4), 145–158. doi:10.1002/mde.979
Ramsay, J., Barbesi, A., & Preece, J. (1998).
A psychological investigation of long retrieval
times on the World Wide Web. Interacting with
Computers, 10(1), 77–86. doi:10.1016/S09535438(97)00019-2
Ranganathan, C., & Ganapathy, S. (2002). Key
dimensions of business-to-consumer Web sites.
Information & Management, 39(6), 457–465.
doi:10.1016/S0378-7206(01)00112-4
Rose, G. M., Evaristo, R., & Straub, D. (2003).
Culture and consumer responses to Web download time: A four-continent study of mono and
polychronism. IEEE Transactions on Engineering Management, 50(1), 31–44. doi:10.1109/
TEM.2002.808262
226
Rose, G. M., Lees, J., & Meuter, M. (2001).
A refined view of download time impacts on
e-consumer attitudes and patronage intentions
toward e-retail. International Journal of Media
Management, 3(2), 105–111.
Rose, G. M., Meuter, M., & Curran, J. (2005).
Online waiting: The role of download time and
attitudinal variables on attitude toward e-retailers.
Psychology and Marketing, 22(2), 127–151.
doi:10.1002/mar.20051
Rose, G. M., & Straub, D. (2001). The effect of
download time on consumer attitude toward the
e-service retailer. e-Service Journal, 1(1), 55-76.
Rushinek, A., & Rushinek, S. F. (1986). What
makes users happy? Communications of the ACM,
29(7), 594–598. doi:10.1145/6138.6140
Ryan, G., & Valverde, M. (2003). Waiting online:
A review and research agenda. Internet Research:
Electronic Networking Applications and Policy,
13(3), 195–205. doi:10.1108/10662240310478213
Ryan, G., & Valverde, M. (2006). Waiting in line
for online services: A qualitative study of user’s
perspective. Information Systems Journal, 16,
181–211. doi:10.1111/j.1365-2575.2006.00214.x
Schleifer, L. M., & Amick, B. C. (1989). System
response time and method of pay: Stress effects
in computer-based tasks articles. III International
Journal of Human-Computer Interaction, 1(1),
23–39. doi:10.1080/10447318909525955
Selvidge, P. (1999). How long is too long for
a Website to load. Usability News, 1(2). Retrieved from http://psychology.wichita.edu/surl/
usabilitynews/1s/time_delay.htm
Shneiderman, B. (1998). Designing the user interface: Strategies for effective human-computer interaction (3 ed.). Reading, MA: Addison-Wesley.
Chronism Theory, Culture, and System Delay
Straub, D. W., Hoffman, D., Weber, B., &
Steinfield, C. (2002). Toward new metrics for
net-enhanced organizations. Information Systems Research, 13(3), 227–238. doi:10.1287/
isre.13.3.227.80
Straub, D. W., Loch, K. D., Evaristo, R., Karahanna, E., & Srite, M. (2002). Toward a theory-based
measurement of culture. Journal of Global Information Management, 10, 13–23. doi:10.4018/
jgim.2002010102
Tedeschi, B. (1999). Good Web site design can
lead to healthy sales. New York Times. Retrieved
from http://www.nytimes.com/library/tech/99/08/
cyber/commerce/30commerce.html
Tella, S. (2000). Achronos: Reflections on timeless time, media and communication. In Tella,
S., & de Vries, J. (Eds.), Media, mediation, time
and communication: Emphases in network-based
media education (Vol. 9, pp. 83–100). Helsinki,
Finland: University of Helsinki.
Tom, G., Burns, M., & Zeng, Y. (1997). Your life
on hold: The effect of telephone waiting time
on customer perception. Journal of Direct Marketing, 11(3), 25–31. doi:10.1002/(SICI)15227138(199722)11:3<25::AID-DIR5>3.0.CO;2-Z
van Everdingen, Y. M., & Waarts, E. (2003).
The effect of national culture on the adoption of
innovations. Marketing Letters, 14(3), 217–232.
doi:10.1023/A:1027452919403
Weinberg, B. D. (2000). Don’t keep your customers waiting at the (virtual) front door. Journal of
Interactive Marketing, 14(1), 30–39. doi:10.1002/
(SICI)1520-6653(200024)14:1<30::AIDDIR3>3.0.CO;2-M
Wines, M. (2003,April 12).Amean scrum on playing
fields of South Africa. New York Times. Retrieved
from http://www.nytimes.com/2003/2009/2012/
international/africa/2012AFRI.html
Wines, M., & LaFraniere, S. (2004, April 26). Decade of democracy fills gaps in South Africa. New
York Times. Retrieved from http://www.nytimes.
com/2004/04/26/international/africa/26AFRI.ht
ml?hp=&pagewanted=print&position=
Wirtz, J. B., & Bateson, J. E. (1995). An experimental investigation of halo effects in satisfaction measures of service attributes. International
Journal of Service Industry Management, 6(3),
84–102. doi:10.1108/09564239510091358
227
228
Chapter 11
A Cross-National Study of
Mobile Internet Services:
A Comparison of U.S. and Korean
Mobile Internet Users
Dong Hee Shin
Sung Kyung Kwan University, South Korea
ABSTRACT
This study surveyed mobile users in the United States and Korea to determine the key differences between
the two countries. Survey questions, developed in two languages, were presented in each country to
explore the influences of informativeness, entertainment, interactivity, and availability on mobile user
dimensions. The study design methods were based on the revision of a uses and gratifications approach,
and a relational model of antecedents and consequences was tested with a structural equation modeling
approach. Mobile Internet service uses and gratifications were analyzed cross-nationally in a comparative fashion focusing on the differences in the composition of motives in the two countries. Based on
the results of this study, practical implications for marketing strategies in mobile service markets and
theoretical implications for cross-country studies are recommended accordingly.
INTRODUCTION
Mobile Internet services have become a major
interest for the information system research community. Current mobile devices provide users
with various advanced services, including mobile
DOI: 10.4018/978-1-60960-605-3.ch011
television service with additional audio and data
services that can be viewed on cell phones, portable
receivers, or vehicle terminals (Shin, 2008). Recent
technological convergence allows cell phone and
personal digital assistant users to watch terrestrial
digital television on their portable communication
devices. Through advanced mobile devices, users
can tune into regular TV programming or watch
Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
A Cross-National Study of Mobile Internet Services
content-on-demand through portable terminals
that include mobile handsets and mini-TVs installed in their cars. Various data services based
on two-way communication capability are also
available. For example, users watching a movie on
their mobile devices can retrieve the background
data of one of the movie’s stars by clicking on
a certain part of the screen. While watching a
cooking program, users can store the recipe in a
computer or get a print-out. When other software
applications are in place, mobile devices will also
enable users to pay for purchases while watching
a TV home shopping program on mobile handsets.
This mobility increases users’ control over mobile
services, allowing them to enjoy interactive services. Industry and academia alike wonder how
users utilize mobile device, what they watch, and
with what gratifications (Shin, 2007, 2008).
Even though the literature on the adoption and
the use of media and the Internet is quite extensive,
few studies have explored the motivations for using the recent convergence mobile services, such
as satellite mobile TV, telematics, and IPTV, and
the associated antecedents and consequences from
a comparative perspective. Traditional models of
media uses and gratifications (U&G) can be applied to convergence technologies, improving our
understanding of U&G of these new services. The
U&G theory claims that users are seen as active and
focuses on the explanations for users’ motivations
and associated behaviors (Eighmey & McCord,
1998; Ruggiero, 2000). The key research question
of the present study is to investigate user attitudes
toward mobile services in a two-country context
in order to investigate differences that may exist
for U&G of mobile services. Although there might
be no universal measurements of user activities,
motivations, and cultural values that is applicable
to different countries, this information is basic
to understanding why certain mobile services
are adopted in some countries and not in others.
Shin (2008) and Kim, Lee, and Lee (2004) urged
researchers to conduct cross-country studies to
determine if U&G were universals or if they serve
different functions in different societies. The innovative and global nature of mobile services has
fostered many visions of comparative understanding among countries, although the research in this
area is still at a very early, exploratory stage. As
mobile services become more and more global,
a cross-national comparison would be useful in
providing practical implications for the industry
for developing global markets. This comparative
study can also be useful in academia by providing theoretical implications for the cross-cultural
studies of technology adoption and usage.
The relational model of the antecedents and
consequences of attitude toward mobile services
is tested by a structural equation modeling (SEM)
analysis. The SEM approach is adopted because
this study vigorously tests the convergent, discriminant, and nomological validity of U&G
constructs or the extent to which predictions
of the model are verified. SEM as compared to
multiple regression has advantages: inclusion of
more flexible assumptions (particularly allowing
interpretation even in the face of multicollinearity),
use of confirmatory factor analysis to reduce measurement error by having multiple indicators per
latent variable, the desirability of testing models
overall rather than coefficients individually, the
ability to test models with multiple dependents,
and the ability to model mediating variables rather
than be restricted to an additive model. Moreover,
where regression is highly susceptible to error
of interpretation by misspecification, the SEM
strategy of comparing alternative models to assess
relative model fit makes it more robust.
The study applies the attitude scales toward
mobile service (Stafford & Gilleson, 2004) to
recently developed mobile services. Consumers’
attitudes toward mobile services have been considered a key indicator of service effectiveness
and consumers’ belief in mobile services (Tsang,
Ho, & Liang, 2004), just as attitude to services has
been supported as a key predictor of subscriptions
in the literature (e.g., Laughlin, 2001). Thus, the
explanatory effects of U&G can be generalized
229
A Cross-National Study of Mobile Internet Services
to a mobile conceptual model. This model guides
the research questions and measurement instruments. When exploring the relationship between
mobile U&G, this study addresses the following
research questions:
RQ1: What are the U&G of mobile services among
users in the two countries?
RQ2: Which variables most significantly affect
satisfaction in mobile services in the two countries?
RQ3: What cross-national differences exist for
the U&G of mobile services? Is there any difference in the structural relations of factors across
the two countries?
With the RQs in place, the goal of this study is
to compare the U&G patterns in the two countries’
users. Shin (2007) argues that greater mobile experience and usage confidence generally broaden
the scope of users’ mobile activities. The present
study examines this premise in a two-country
context, with the underlying assumption that
differing stages of mobile diffusion would be
reflected in the range of activities pursued. The
two countries have in common that both countries
represent one of the highest mobile diffusions in
the world and both countries take the leads in
mobile TV standards, while the two standards are
competing globally to be adapted and evolved for
use in mobile handsets.
This study differs from other descriptions or
comparisons of mobile activities in several ways.
First, it develops a specific attitude and behavior on
mobile service (particularly in convergence technologies) representing the range of user interactions with mobile Internet. Second, to capture both
quantity and quality of online audience activity, it
explores consumer activity participation in terms
of exposure (time and frequency) and gratifications
(cognitive and emotional). Third, its comparison
of the U.S. and Korean mobile users provides a
230
cross-country perspective on user activity at different stages of mobile proliferation. Finally, the
study also addresses the paucity of mobile service
adoption research in the international arena. The
uniqueness of local mobile markets along with the
increasing trend for global mobile-business has
led to a growing need for cross-national studies
on the mobile Internet.
As to contributions, the findings should be
of interest to both academics and practitioners.
From a theoretical perspective, drawing upon
the modified U&G, this study provides a model
to identify antecedents of user intention to use
mobile Internet. The modified model seems well
suited for developing an evaluative framework for
services that are adopted for functional reasons
and services that are directed specifically at innovative user attitudes. Several researchers have
explored U&G applications in the mobile context.
Extensive studies in mobile services have been
conducted by the management of information system fields using a Technology Acceptance Model,
or usability approach. Mobile services from a
media perspective remain an under-researched
area. Although the platform of mobile Internet is
a telecommunication infrastructure, its content
over wireless networks is still broadcasting.
Therefore, it is worthwhile to examine mobile
users’ behavioral aspects from the media usage
perspectives. The present research is intended to
address this literature deficiency by linking the
well-established U&G framework to mobile user
behaviors. Specifically, variables based on U&G
theory are considered the antecedents of attitude
toward mobile Internet services, while the usage
and satisfaction are modeled as the consequences
of attitude toward mobile services.
From a practical standpoint, the findings
of this study provide a good basis for industry
developing a service evaluation framework to
determine the adoption potential of new mobile
services. Mobile service developers are facing the
challenge for how to create an environment with
ease to play and fun to enjoy. Yet, the interface
A Cross-National Study of Mobile Internet Services
designs for related service elements as qualities
are rarely discussed in the context of U&G. The
findings in this study should be useful for mobile
developers to improve their works.
LITERATURE REVIEW AND
HYPOTHESIS DEVELOPMENT
Few studies have taken a U&G approach in studying the mobile services, and even fewer have
narrowed their focus to look at it in a comparative
perspective. This research gap―a lack of studies
applying traditional U&G theory to the motivation
of new mobile services―presents this study with
a research opportunity.
Uses and Gratifications Theory
U&G as a media use theory from mass communications research guides the assessment of user
motivations for media usage and access. Recent
adaptations of U&G research to new convergence
technologies broaden the capability of U&G to
assess new technologies (Shin, 2008). Many studies of technology have extended the U&G theory
to new technologies, including the predicting and
assessing of convergence technologies. Stafford’s
(2004) and Nysveen, Pedersen, and Thorbjørnsen’s (2005) studies of motivations in the use of
the Internet offer examples of the extension of the
U&G approach to information system research.
These pioneering interdisciplinary studies have
been followed by numerous other studies that have
applied a U&G approach to emerging technologies
(Shin, in press). Since the results of Internet U&G
studies based on scales from television research
have subsequently not been very predictive of
Internet use (Lin, 1999), there is a clear need to
extend the process of developing U&G measures
specific not only to the Internet but also to the
unique user interface of mobile devices when
using U&G to understand mobile device users’
motivations. This extension can be a meaningful
task considering the convergence trends.
Hypotheses
Based on the U&G literature, the present study
made a list of possible motives related to the
convergence service via mobile device use such
as entertainment, informativeness, usefulness,
irritation, and interactivity. In addition, possible
gratifications from mobile services, such as
mobility, immediate access, relaxation, escape,
killing time, and expression of identity/fashion/
status, were collected. This study conducted pilot
studies via three focus-group sessions in each
country and in-depth interviews with 10 mobile
Internet users in the two countries respectively.
The results from the two countries were compared and contrasted, and six dimensions emerge:
entertainment, informativeness, interactivity,
usefulness, connectedness, and irritation. These
six dimensions are also recommended by the
experts in mobile communications from the two
countries. Two groups of mobile telecom experts
advised this study on the possible dimensions,
which are also almost identical to the suggestions
by a group of researchers in academia. This study
sought professional advice on the possible dimensions to 23 researchers in the areas of information systems, technology adoption research, and
human-computer interaction. For triangulation
of the dimensions, academia suggestions were
compared with those from industry. Based on a
final list of dimension, exploratory factor analysis
identifies four dimensions (entertainment, informativeness, interactivity, and availability), which
by large confirm previous literature on Internet
and mobile devices.
Entertainment
The entertainment construct has been referred to
as motivation seeking pleasure and enjoyment,
or escapism (Shin, 2007; McQuail, 2000; Eigh-
231
A Cross-National Study of Mobile Internet Services
mey & McCord, 1998; Eighmey, 1997). While
users seem to seek habitual pleasures, such as
environmental reasons of background noise and
regulative uses for the punctuation of time and
activity (Lull, 1990), mobile users appear to
seek occasional entertainment with the ability to
watch short programs whenever the user wants
(Sordegard, 2005). This difference is also found in
viewing pattern and time spent on mobile services.
According to Shin’s (2008) survey, the average
time session using mobile services is less than
five minutes in a commuting time (morning and
evening), and typical usages are short games and
traffic/weather information.
Previous research suggests that providing
higher entertainment value is likely to lead to an
advantage for users and to motivate them to use
technologies more often (Leung & Wei, 2000).
For example, people use media either for the
content (enjoyment) carried by a medium or for
the simple experience and enjoyment of the actual
usage process (Stafford et al., 2004). Coursaris,
Hassanein, and Head (2003) further developed the
relation between entertainment and mobile service
and found that entertainment is positively related
to the mobile media value and attitude to mobile
service. This finding is corroborated by Stafford
and Gilleson’s study (2004) that entertainment
is positively associated with the motivation for
mobile services, although their study has not
vigorously tested (e.g., using the SEM approach)
the scales of attitude toward the mobile Internet
and U&G underlying dimensions. Following this
stream of research, the following hypothesis is
proposed:
H1: Entertainment offered by mobile Internet
service is positively associated with the attitude
toward mobile service.
Informativeness
McQuail (2000) defines information uses as utilizing media to educate users in certain areas, such
232
as learning more about the world, seeking advice
on practical matters, or fulfilling user curiosity.
While information uses in media are generally
broad, information seeking in mobile services
tends to be more specific than those in media.
Leung and Wei (2000) define informativeness in
mobile service as the extent to which the mobile
service provides users with pertinent and useful
information. Leung and Wei (2000) argue that
users consider a mobile phone’s functionality to
provide timely information as the most fundamental reason for adopting mobile services. Similarly,
Stafford et al. (2004) show that the informational
feature in mobile devices is its major function
that draws customers’ attention. Without the time
and space barriers in the mobile era, the mobile
application can provide users with instant and
valuable content. Chen and Wells (1999) find
that informativeness and attitude to the Web site
are positively related, and similarly, Shin (2007)
identifies a substantial and positive correlation
between informativeness and the mobile service
value and attitude to mobile service. Based on this
body of research, it is hypothesized:
H2: Informativeness of mobile Internet is positively associated with attitude toward mobile service.
Interactivity
Media researchers began to pay attention to
interactive media and have studied interactivity.
McQuail (2000) defines interaction as a motivation
to gain insight into the situations of other people
in order to achieve a sense of belonging. Livaditi
and Vassilopoulou (2003) argue that interactivity
is a key factor in adopting interactive multimedia applications. Therefore, it can be reasonably
thought that high elaboration and more information
processing through a high degree of interactivity
on a mobile device might result in better attitudes
toward and evaluations of the mobile service. In
light of this literature, it is hypothesized:
A Cross-National Study of Mobile Internet Services
H3: Interactivity of mobile device is positively
associated with attitude toward mobile service.
Consequences of Attitude
Toward Mobile Services
Availability
Previous research has shown that the attitude
and satisfaction toward the service of technology
are the two most noteworthy indicators of new
technology diffusion and outcomes (e.g., Aaker
& Stayman, 1990; Haley & Baldinger, 1991).
For example, Palmgreen and Rayburn (1985)
discover the significant relation between attitude
and satisfaction of TVs. Kwon and Chidambaram
(2000) also find out that the attitude toward the
technology influences brand attitudes and purchase intentions. Jee and Kim (2005) use a seminal
scale—mobile users’ attitudes toward the mobile
services. Attitude toward a mobile service is considered a key determinant of consumer adoption
and usage of the mobile service and its diffusion,
just as attitude toward technology is a key predictor of emerging mobile service. It is believed that
consumers who have a positive attitude toward
the mobile service would spend more time using
the mobile device for fun or information, and they
feel satisfied with the convenient and resourceful
mobile services. From this line of research, it is
hypothesized (see Figure 1):
Reflecting the feature of advanced mobile services,
this study adds new gratifications related to the
interactivity variable to the analysis. Angleman
(2000) researches the intended uses of the Internet and the expected and latent gratifications.
Latent gratifications are different from manifest
gratifications, which can be obtained from an
actual medium’s content (e.g., watching a specific program); instead, they are gratifications
whereby users feel psychological readiness to
access content or service at any time. Or latent
gratification can be unconscious gratification:
users’ perception of connectedness, that they are
emotionally connected to the world, its resources
and people. In other words, users can be gratified
or comforted with psychological connectedness;
they can enjoy content at their convenience that
used to be available only through traditional media. Shin (2007) investigates users’ perception of
availability in wireless broadband and concludes
that the perception of availability depends not
only on the system’s characteristics, but also on
the user’s behavior. Similarly, Gokhale and Lu
(2006) find the Web-based e-commerce system
and report that users are more influenced by their
perception of the availability of the system than
its actual availability. Recently, this perceived
availability has been applied to mobile devices
(Shin, 2007, 2008). Based on the previous studies, this study proposes perceived synchronicity
(Shin, 2007; Rauh & Renfro, 2004), perceived
connectedness (Shin, in press), and perceived
ubiquity (Shin, 2008; Peters & Allouch, 2005)
as possible variables for latent gratifications or
availability.
H4: Latent gratification of mobile Internet is
positively associated with attitude toward mobile
service.
H5a: Attitude toward mobile Internet is positively associated with users’ mobile usage.
H5b: Attitude toward mobile Internet is positively
associated with users’ mobile satisfaction.
RESEARCH METHODOLOGY
Sample and Data Collection
This study collected data of mobile users (especially for those who use convergence services,
such as mobile TV and wireless broadband) from
Korea and the United States. Respondent samples
were obtained from commercial survey firms
located in the U.S. and Korea. A representative
233
A Cross-National Study of Mobile Internet Services
Figure 1. Result of the research model for Korean and the U.S. users
sample from each country was developed from
these survey firms. The firms developed panels
of representative users who have given permission to be contacted, so called opt-in permission. Trained telephone interviewers carried out
extensive telephone surveys among a randomly
selected 462 people between May 2005 and June
2006. The telephone interviews asked the selected
respondents about their attitudes, satisfaction, and
experience on convergence services via mobile
devices. To refine and increase the validity of
survey data, responses that contained answers
with systematic errors and those with inconsistent
information were excluded.
Measurements
Pilot studies were conducted to ensure consistency
and reliability of measurements of the survey and
remove similar wordings and logically duplicative
items. The initial questions, which consisted of
234
30 questions, were pilot tested in a Web-based
survey with 72 college students who had used
mobile Internet services. Through the pilot tests, 18
items were eliminated because of not meeting the
criteria, such as sufficient correlation coefficients
and reliability. The final 12 measurement scales are
Likert-type with a seven-point format, anchoring
at 1―strongly disagree and 7― strongly agree.
As reported in Table 1, the 12 items measuring the
U&G dimensions in this article are derived from
previous studies (Chen & Wells, 1999; Korgaonkar
& Wolin, 1999). In particular, each of the scales
is measured by five separate items.
To cover the entertainment constructs, a fouritem scale was developed collecting items from
U&G scales. The first of the items covered the
“enjoyment” conception (Leung & Wei, 2000);
the second covered the “relaxation” conception
(Leung & Wei, 2000); the third covered the “excitement” conception, which is also found in the
studies of videogame and TV gratifications
A Cross-National Study of Mobile Internet Services
Table 1. Measurement items
Entertainment
Informativeness
Interactivity
Availability
(latency)
Items
Coefficient
Alpha
Having a mobile Internet is helpful for me to enjoy the pleasure of watching.
0.71
Having a mobile Internet is helpful for me to pass the time.
0.67
Having a mobile Internet is helpful for me because using mobile services relaxes me.
0.59
Mobile Internet gives me quick and easy access to large volume of information.
.88
Information obtained from the mobile is useful.
0.71
I learned a lot from using the mobile service.
0.69
I hope the mobile service to be interactive.
0.59
I like to increase my control over mobile service.
0.68
Using mobile, I like to interactively communicate with others.
0.82
I am gratified with the ubiquity of mobile Internet that I can watch content anytime with
mobile TV.
0.89
Because of mobile Internet, I feel like that I am in connection with my friends and family.
0.71
With mobile Internet, I can enjoy several applications synchronously or I can access realtime news synchronously.
0.69
(Sherry, 2001); and the final item was a general
item covering fun-seeking gratification (Leung
& Wei, 1998).
The interactivity constructs included humandevice (content) interaction (Weilenmann &
Larsson, 2000) and human-human interaction
(Palen, Salzman, & Youngs, 2001). In addition,
the five-item informativeness scale is similar to
Chen and Wells’s study (1999). The informativeness of a Web site is defined by elements such as
informativeness, intelligent contents, knowledge,
resourcefulness, and timeliness. These factors
build upon the informative scales that have
emerged in previous studies of mobile applications. The latency constructs are measured with
perceived synchronicity (Rauh & Renfro, 2004),
perceived connectedness, and perceived ubiquity
(Peters & Allouch, 2005).
Attitude toward the mobile services is measured by the scale modified from a variety of
published sources (Rhee & Kim, 2004; LaRose,
Mastro, & Eastin, 2001; Torkzadeh & Van Dyke,
2001). Examples of this scale include awareness,
usability, and overall impressions. The reliability
and validity of this scale were also confirmed
in their study. Satisfaction is measured by five
Likert-type items used by Lee, Kim, and Chung
(2002), who studied the customer satisfaction
of the mobile Internet services provided by the
mobile telecommunications providers in Korea.
Examples of the items include: “I feel satisfied
with the convenient use of mobile services, such
as mobile TV”; “I am satisfied with information
on the mobile Internet services”; “I am satisfied
with mobile products and services”; “I feel satisfied with the prices on mobile Internet services”;
“Overall, I am satisfied with mobile Internet.”
Finally, mobile usage is measured by one item,
the number of hours spent on mobile services
(Korgaonkar & Wolin, 1999). Instead of asking
respondents to recall the number of hours per
week spent on mobile services, this study used
a scale with 1=rarely to 5=very often. Evidence
of the internal consistency of the constructs was
confirmed in this study. All the items in the scale
were determined to be reasonably reliable, pro-
235
A Cross-National Study of Mobile Internet Services
ducing overall alpha coefficients (Cronbach’s
alpha) of 0.80 for the entertainment factor, 0.64
for the informativeness factor, and 0.69 for the
interactivity factor. Attitude toward the mobile
services and satisfaction of the mobile services
also possessed sufficient reliability, as Cronbach’s
alphas were 0.82 and 0.84 for attitude toward the
mobile services and satisfaction with the mobile
services, respectively. All the measures in the
present study are based upon previously validated measures (Venkatesh & Morris, 2000) and
are considered reliable. The result of correlation
(Pearson’s r) test confirms the validity of each
variable (see Table 2).
RESULTS
Descriptive
On average, the U.S. respondents were younger
(26 years) than Korean respondents (29 years).
The U.S. sample showed a higher representation
of females (64%) as compared to males (36%),
while the Korean sample showed a higher representation of males (62%) as compared to females
(38%). Among the U.S. respondents, a little over
half (50.7%) were college students, 11% had a
bachelor’s degree, 5.7% had a postgraduate degree (master’s degree and in some cases doctoral
degree). Only 2.5% reported having some high
school or completing high school. Among Korean
respondents, education levels were clearly higher,
with a majority indicating that they were graduates
or higher (31% bachelor’s, 20.7% postgraduate);
35.6% were college students, a little over 12% had
some high school or had completed high school.
Mean annual income in U.S. dollars was $29,898
for the U.S. respondents and $28,190 for Korean
respondents. This placed the U.S. respondents in
this study within an average U.S. income group
and Korean respondents in a slightly higher Korean income group. The U.S. respondents reported
about one year of mobile Internet usage and Korean
respondents about two years (see Tables 3 and 4).
Thus, Korean mobile users in this study were older,
more educated, and consisted of more males than
females. The U.S. users were also relatively well
educated but mostly in college, younger, consisted
of more females than males, had average incomes,
and had been online (fixed and mobile Internet)
for about seven years (Koreans had been online
for about five years). Consistently, over 90% of
the sampled users in both countries reported using the mobile devices for communication and
information search activities. While 82% of the
Korean sample had used the mobile device for
information sources (weather information, etc.),
only 62% of the U.S. sample reported mobile use
for this activity. The lowest mobile use was for
entertainment activities, with about 75% of U.S.
Table 2. Summary statistics and correlation among variables
Mean
SD
INFORM
ENTERTA
INTERA
ATTITUD
SATISF
MOBIUSE
INFORM
6.19
0.91
1.00
ENTERTA
5.46
1.20
0.65**
1.00
INTERA
2.60
1.38
0.47*
0.46**
1.00
ATTITUD
5.60
0.99
0.67*
0.78*
0.50**
1.00
SATISF
5.40
1.08
0.58*
0.58**
0.44**
0.63**
1.00
MOBIUSE
8.12
1.07
0.15*
0.26**
0.11
0.23**
0.17*
1.00
LATEN
8.42
1.23
0.32**
0.22*
0.31**
0.11*
0.10*
0.35**
* p<0.05; ** p< 0.01
236
LAT
1.00
A Cross-National Study of Mobile Internet Services
Table 3. The result of hypothesis testing: Parameter estimates for causal paths of Korean users
Hypotheses
Regression Weights
Standardized Parameter
Estimates
t-value
Support
H1
ATTITUDE ← ENTERTA
0.43**
7.50
Yes
H2
ATTITUDE ← INTERACT
0.51**
3.02
Yes
H3
ATTITUDE ← INFORM
0.33*
2.91
Yes
H4
ATTITUDE ← LATENCY
0.61**
3.21
Yes
H5a
SATISF ← ATTITUDE
0.29*
7.58
Yes
H5b
MOBILUSE ← ATTITUDE
0.23
3.39
No
Goodness-of-fix statistics: x (87)=196.97; CFI=0.938; GFI=0.906; AGFI=0.902; RMSEA=0.068; * p<0.05; ** p<0.001.
2
Table 4. The result of hypothesis testing: Parameter estimates for causal paths of U.S. users
Hypotheses
Regression Weights
Standardized Parameter
Estimates
t-value
Support
H1
ATTITUDE ← ENTERTA
0.28*
2.50
Yes
H2
ATTITUDE ← INTERACT
0.24*
2.01
Yes
H3
ATTITUDE ← INFORM
0.51**
7.61
Yes
H4
ATTITUDE ← LATENCY
0.17
1.21
No
H5a
SATISF ← ATTITUDE
0.39**
5.62
Yes
H5b
MOBILUSE ← ATTITUDE
0.11
1.91
No
Goodness-of-fix statistics: x (87)=187.78; CFI=0.826; GFI=0.811; AGFI=0.882; RMSEA=0.052; * p<0.05; ** p<0.001
2
respondents and 68% of Korean respondents
reportedly using mobile for entertainment.
Analysis and Findings
The results indicate that all variables of this
research achieved more than 0.82 reliability
scores. Exploratory factor analysis (EFA) was
conducted to see how U&G variables are grouped.
The coefficients in rotated factor analysis matrix
of all media use variables are more than 0.75;
entertainment (0.82), informativeness (0.81),
interactivity (0.87), and latency (0.71). Next,
EFA examined correlations between the usage
variables and gratification variables. All correlations are significant in p<0.05 and p<0.01. The
correlations with gratification (as measured by
attitude, satisfactions, and mobile use) are 0.85
(entertainment), 0.81 (informativeness), 0.87
(interactivity), and 0.69 (latency). This research
also measured the usage time and frequency to
study users’ behaviors. Using these variables,
the relationship between gratification and mobile
services use is studied (correlation coefficient:
0.72). Finally, the result of SEM indicated that
the overall fit of structural model was acceptable. The results (x2 (87) =196.97; CFI=0.938;
GFI=0.906; AGFI=0.902; RMSEA=0.068)
uncovered significant relationships: between
entertainment and informativeness; and between
interactivity and frequency of accessing mobile
services (r=0.59, p<0.05). These findings suggest
that interactivity may be considered as one of the
important elements for mobile services. Overall,
the findings reveal that national differences are
evident in terms of attitudes to mobile Internet
service and use. And that these differences exist
principally in how individuals perceive and use
mobile Internet services.
237
A Cross-National Study of Mobile Internet Services
Measurements Validation
Since entertainment, informativeness, interactivity, and latency are the dimensions of U&G theory
in this study, the items of these constructs are
included in one measurement model to test the
validity rigorously. A separate measurement model
includes the scales of attitude toward the mobile
services satisfaction. Based on the study of Gerbing and Anderson (1988), it is appropriate to adopt
the two-step procedure of SEM in this study. That
is, one should estimate the measurement model
separately before a simultaneous estimation of
measurement and structural models. All measurement models were evaluated on multiple criteria
such as unidimensionality, reliability, convergent
validity, and discriminate validity.
U&G Theory Measurement Models
Before a confirmatory factor analysis (CFA)
of the U&G measurement model, an EFA was
conducted by a maximum likelihood extraction
method with varimax rotation. In order to decide
the number of factors to be extracted and rotated
in the U&G model, three methods were used: (1)
a cut point of .4 and no significant cross loading
criteria; (2) screen plot tests, and (3) consideration
of eigenvalue magnitude and discontinuity (Hair,
Anderson, Tatham, & Black, 1998). EFA results
of the U&G model suggest a clean four-factor
solution corresponding to informativeness, entertainment, interactivity, and latency (with item
loading > 0.70 and small cross loading). The
total variances explained by the three factors are
69.44% for the Korean sample and 62.31% for
the U.S. sample. Then, a three-factor model with
all indicators of these three constructs of U&G
theory was estimated using CFA, which shows
that both measurement models showed acceptable
fits. All items’ loadings on their corresponding
construct were significant at p<0.05 (i.e., t >
2.0), demonstrating adequate convergent valid-
238
ity. Since the modification indices and estimated
residuals of the U&G model were small, unidimensionality was also achieved (Sujan, Weitz, &
Kumar, 1994). In addition, convergent validity of
the U&G model was supported by checking the
pair-wise correlations in Table 2. As expected,
the factor correlations ranged from -0.46 to 0.65
and significantly different from one, establishing
discriminant validity. For these U&G measurement models, the goodness-of-fit (GFI), adjusted
goodness-of-fit (AGFI), root-mean-square error
of approximation (RMSEA), and comparative fit
index (CFI) were 0.916, .903, 0.069, and 0.949
for the Korean sample, respectively; and 0.902,
0.900, 0.071, and .911 for the U.S. sample, therefore indicating an adequate model fit.
Attitude Toward Mobile Services and
Satisfaction Measurement Model
As in the U&G measurement model, both CFA
and EFA were used with SAS/STAT software
(version 8.1) to test this two-factor model. First,
exploratory factor analysis using maximum likelihood with varimax rotation results clearly show
two factors of attitude toward mobile service and
satisfaction. In addition, CFA was performed on
all variables, and the results support the convergent, discriminant validity of attitude toward the
mobile and satisfaction. Overall, model goodnessof-fit indexes also support this measurement
model (x2 (43)=152.83; CFI=0.919; GFI=0.882;
AGFI=0.863; RMSEA=0.079). It should be noted
that although Stafford and Gillenson (2004) and
Tsang et al. (2004) developed the attitude toward
the mobile service scale, their studies did not
test this scale with confirmatory factor analysis.
Thus, the present study extends their research and
confirms that the attitude toward the mobile scale
achieves convergent and discriminant validity as
well as content validity, which was evidenced in
the previous effort (Stafford & Gillenson, 2004).
A Cross-National Study of Mobile Internet Services
STRUCTURAL PATHS AND
HYPOTHESE TESTS
To test the structural relationships, the hypothesized causal paths were estimated and the two
sets of hypotheses in each case show different
results. While five hypotheses (H1, H2, H3, H4,
and H5a) were supported in the Korean case,
only four hypotheses were supported in the U.S.
case (H1, H2, H3, and H5a). The results showed
noticeable differences in path formation and
composition of items (see Tables 3 and 4). The
overall fit of the each model is acceptable because
the goodness-of-fix statistics (CFI = 0.912, GFI
=0.876, AGFI = 0.847, and RMSEA = 0.074; CFI
= 0.819, GFI =0.795, AGFI = 0.831, and RMSEA
= 0.163; CFI=0.938; GFI=0.906; AGFI=0.902;
and RMSEA=0.068) are satisfactory, with the x2
/df ratio close to 2.0.
Hypothesis 1 states that entertainment is positively related to attitude toward mobile services.
The entertainment−attitude toward Korean mobile
services path loading was 0.68 (p<0.01), supporting H1 for Korea. The entertainment−attitude
toward the U.S. mobile service path loading was
0.42 (p<0.01), supporting H1 in a lesser extent for
the United States. In addition, the results support
H2, as informativeness is positively related to attitude toward mobile services (Korea, loading =
0.29, p<0.05; the U.S. loading = 0.15, p<0.05).
As presumed, interactivity leads to a positive attitude toward mobile services (Korea, loading =
0.26, p<0.05; the U.S. loading = 0.19, p<0.05),
supporting hypothesis H3. A comparison of the
estimated coefficients of all independent variables
shows that entertainment is the most important
determinant for Korean users’ behaviors, whereas
informativeness is the most significant determinant for the U.S. users.
While the Korean users’ latent gratifications are
positively associated with attitude toward mobile
services (H4), which is consistent with a study by
Shin (in press), the U.S. users’ latent gratifications
are not significantly associated with attitude or
satisfaction. Overall, entertainment, informativeness, interactivity, and latency explain 87% of the
variance in attitude toward mobile service. The
final hypothesis states that attitude toward mobile
service is positively related to mobile usage and
user satisfaction. The attitude toward mobile usage path loading was 0.79 (p<0.01) in Korea, 0.62
(p<0.01) in the U.S., while the attitude toward
mobile user satisfaction path estimates is 0.11
for Korea and 0.09 for the U.S., rejecting H5b
for both cases because of the path estimates and
p values of 0.82 and 0.79 respectively.
While users’ attitudes toward mobile services
explain 63% of the variance of mobile satisfaction,
it is interesting to find both countries’ users are
not greatly satisfied with the mobile convergence
services. Overall, the two models of the U.S. and
Korea had good fits with the data, but each model
shows an apparent different pattern.
T-Test and Factor Analysis
As the results of SEM indicated differences, further
analytical tools were used with t-test and factor
analysis. With the t-test, means and standard deviations were used to examine country differences
in each variable (see Table 5). While there was no
significant country differences in entertainment
and informativeness, users did differ on interactivity, with the U.S. users being considerably less
interactive and Korean users indicating fairly high
interactivity levels (interactivity t[235]= -13.04,
p<0.01). In addition, the result of the t-test was
significant for latent gratification, with Korean
users being more gratified than the U.S. users
(t[193]= -9.08, p<0.01). However, the t-test was
not significant for the dependent variables, suggesting that there were no significant differences
between U.S. and Korean users in the extent of
usage time and satisfaction. This is an intriguing
finding that contradicts previous studies: Park
and Jun’s (2003) study found the period/time of
Internet usage was significantly longer in Korea
than the U.S.
239
A Cross-National Study of Mobile Internet Services
Table 5. Difference test results
Variables
t
df
Sig
(2-tailed)
Mean
difference
Std. error
difference
95% C.I.
Lower
95% C.I.
Upper
Entertainment
-13.041
249.44
0.000***
-0.401
0.032
-0.500
-0.349
Informativeness
-10.191
252.92
0.000***
-0.417
0.041
-0.524
-0.310
Interactivity
-13.042
218.23
0.000***
-0.131
0.043
-0.208
-0.042
Latency
-9.083
211.54
0.000***
-0.481
0.039
-0.513
0.293
Attitude
-3.301
201.34
0.000**
0.395
0.037
-0.113
-0.155
Usage
-0.962
231.33
0.338
-0.041
0.041
-0.121
0.039
Satisfaction
1.322
291.41
0.294
0.103
0.041
0.029
0.183
* p< 0.10; ** p< 0.05; ***, p< 0.01
Meanwhile, factor analysis shows further interesting findings. A principal components factor
analysis with varimax rotation of gratifications
showed a different pattern of factor structure for
each country, as summarized in Table 6. The
structure for the combined sample consists of six
out of 12 factors with eigenvalues greater than
one. These factors accounted for 71.6% of the
variance for mobile gratifications. Loadings and
alpha coefficients are reported for each factor.
In previous studies for the U.S. Internet users,
the informational factor was the first and largest
factor (Charney & Greenberg, 2002). In a study
of Korean Web users, interaction was the first
factor loading for gratification obtained (GraceFarfaglia, Dekkers, Sundararajan, Peters, & Park,
2006). In this study, information was the strongest
factor for the U.S. users, with interactivity the
second most important gratification, whereas
entertainment was the primary factor for Korean
users followed by availability and interactivity.
These three factors represent the primary motivations to use mobile service as shown by factor
analysis.
Table 6. Results of principle components factor analysis
Items
Loading
Combined
U.S.
Korea
Variance α/eigenvalues
Entertainment 1
0.886
1
3
1
26.4%/2.89
Entertainment 2
0.859
1
3
1
Entertainment 3
0.832
1
3
1
Information 1
0.617
4
1
4
Information 2
0.612
4
1
4
Information 3
0.603
4
1
4
Interactivity 1
0.721
2
2
3
Interactivity 2
0.732
2
2
3
Interactivity 3
0.747
2
2
3
Availability 1
0.689
3
4
2
Availability 2
0.701
3
4
2
Availability 3
0.676
3
4
2
*Number means the factor number in each rotated component matrix
240
9.3%/0.87
18.4%/1.10
14.5%/0.92
A Cross-National Study of Mobile Internet Services
Usages of Mobile Internet Services
In the survey, respondents were asked to state
which mobile Internet services they currently use.
These additional data were collected to identify
preferred mobile Internet services in each country
and to complement the findings from the SEM
and t-test. As shown in Figure 2, different services
were preferred in the United States and Korea.
This difference also confirms the findings from the
SEM and t-test. The U.S. users clearly express that
their preferences are information-related services
such as e-mail, news, and personal information
management. On the other hand, Korean users in
general prefer hedonic purposes such as downloading, games, and mobile content. This difference
implies that Korean users see mobile Internet as
a hedonic device, whereas the U.S. users tend to
see it as a functional service tool. The different
preferences of the two countries may attribute to
different infrastructure, content, and services―
that Korea has much better infrastructure, quality
of contents, and ranges of services than those of the
U.S. It may be that U.S. users prefer informationoriented services to data-related or entertainmentrelated services because they cannot find satisfying
entertainment services. Thus, the difference may
not be caused by the users’ characteristics but
the mobile Internet infrastructure, content, and
services. However, many comparative studies
on technology adoption support the argument
of different users’ characteristics contributing
Figure 2. Mobile Internet usage (PIM―personal information managing, schedule, calendar, etc.; Information―stock, weather, traffic, etc.)
241
A Cross-National Study of Mobile Internet Services
to different patterns of usage (Shin, 2007, 2008;
Kim et al., 2004; Sun & Su, 2006). It may be
true that different external factors may affect users’ adoption, but users’ characteristics are also
affected by such external factors like pervasive
availability of infrastructure, wide variation of
services, and quality of enjoyable contents. Thus,
such external factors should be considered as a
part of whole contextual factors constituting users’ characteristics and thus contributing to users’
intentions, which are greatly influenced by such
external factors. In other words, external factors
and users’ characteristics are inseparable entities
because they are co-evolving and influencing each
other. For parsimony, this study assumes users’
characteristics embed such external factors. Future
studies may further investigate the relationship
between users’ characteristics and external factors. It may be interesting how user differences
are influenced by the external environment and
to what extent such external environment affects
users’ decisions to adopt.
DISCUSSION
Summary
The continuing rapid convergence of telecom and
Internet presents new opportunities for research
to analyze the ways in which individuals interact
with and act upon new convergence technologies.
This study attempts to investigate the user experience of mobile service in the U.S. and Korea. To
analyze the user dimension from a cross-country
perspective, this study used the U&G framework that was composed of four dimensions of
gratifications. It investigated the influences of
informativeness, entertainment, interactivity, and
latency on various mobile user constructs such as
the attitude toward mobile services and the usage/satisfaction of mobile services. Particularly,
the mobile service usage and satisfaction were
explored as consequences of attitude toward the
242
mobile, while informativeness, entertainment,
interactivity, and latency are the constructs of attitude toward the mobile. This nomological model
was tested with a firm theory testing methodology
of the SEM approach.
The SEM results implied that the U&G theory
effectively explains users’ attitudes toward mobile
services across the samples from both countries. It
is found in both countries that users who perceive
mobile services as entertaining and informative
generally like mobile services and show a positive attitude toward them. In addition, those who
perceive mobile services as interactive are more
likely to report a positive attitude. Attitude then
affected mobile use and satisfaction: users who
have a positive attitude toward the mobile are more
likely to access mobile services for content and
feel more satisfied. This study also shows different
levels of interactivity and latent gratifications in
the two countries.
The results also showed that the users in the
two countries did differ in the way they thought
about and preferred to interact with mobile Internet
service in terms of U&G structures. Results of
the difference tests show that each component of
gratification is different, which implies that the
needs and values of mobile users between the two
countries may be different. The result shows that
Korean users are more influenced by entertainment and interactivity than information, whereas
the U.S. users are more influenced by information
than entertainment and interactivity. The survey
result of mobile service preferences confirms this
difference in that Korean users use entertainment
services more often, such as downloading services
or playing games, while the U.S. users use information services more frequently. It can be inferred
from this difference that Korean users tend to
perceive mobile devices as a functional tool but
also an entertaining gizmo. On the contrary, the
U.S. users tend to see mobile devices as a more
functional than entertaining tool.
In both countries, the effect of attitude on satisfaction is significant, and interestingly, the effect
A Cross-National Study of Mobile Internet Services
of attitude on mobile use is not that significant.
This contradicts previous studies that show the
significant effect of attitude on use. Also, industry
tends to assume the direct link between attitude
and usage. Thus, the finding of the insignificant
effect raises debatable topics as well as important
implications. It may be interpreted in reference to
Shin’s (2007) finding, which shows that customers
become pickier than ever in using and adopting
technologies, and their intention to use/adopt
should be reinforced/mediated or moderated by
special constructs. Given that Shin’s (2007) study
was based on convergence service (wireless broadband), this argument is very much applicable in
mobile Internet because mobile services become
converging integrating data, visual, and voice.
The insignificant effect further implies that
mobile users in these days are becoming more
discriminating in using their mobile services.
Many advanced services over mobile devices
carry a usage-based charge (per access, per usage
time, or per song, etc.), and users may be more
cautious about their usage in mobile Internet services and want to find more distinct usefulness and
enjoyment. As shown, while the users’ attitudes
are very positive, actual use of mobile Internet
services is rather discouraged or moderated by
the usage-based charging system. This explanation is nicely suited to the finding that the effect
of the U.S. users (0.11) is lower than the effect of
Korean users (0.23), because a fixed-based rate
is generally adopted in a Korean mobile market
whereas a usage-based charge is common in the
U.S. market (Shin, in press). Customers become
more discriminating between attitude and actual
use of mobile technologies: to have good attitudes
toward mobile service is one thing and to use the
service is another. To know the usefulness of mobile service is one thing, but to know and intend to
use may be different and may have further steps to
lead customers to use/adopt mobile services. That
is what mobile Internet service truly is, because
almost everybody knows the usefulness of the
mobile Internet, but they may also understand
the actual adoption and use are different dimensions. They may seek specific gratifications and
novel enjoyment. For discretionary use, customers
may want a technology that targets their specific
interests, rather than giving them general options.
Overall, entertainment, informativeness, interactivity, and latency explain 87% of the variance
in attitude toward mobile services for Korean
users and 72% for the U.S. users. While this
finding lends strong support for the nomological
validity of the U&G theory and attitude toward
the mobile scale, the findings also suggest the
different perceptions of mobile services held by
users: Korean users tend to see the mobile phone
as more than just a communications device, and
U.S. users tend to see it as an extension of fixedphones. This implication is conceptually consistent
with previous findings; for example, Leung and
Wei (2000) found that the cell phone is for more
than just talking. A follow-up question is why
U.S. users perceive it differently. This may be
due to the fact that wireless infrastructure is more
widely available in Korea, and wireless spurred
rapid adoption and satisfaction with the mobile
TV (Shin, 2008). It is generally accepted that
the mobile infrastructure in the U.S. is relatively
unstable, and service offerings are limited compared to Asian countries, which would certainly
affect their attitudes and satisfaction (Shin, 2007).
More importantly, the different U&G may be attributed to the cultural and social fabric of media
uses (Grace-Farfaglia et al., 2006). For example,
studies found that U.S. users use mobile service
to enhance individualism like enjoying games
by themselves, whereas Korean users use it as
an interactive tool with friends and peers (Shin,
2008; Leung & Wei, 2000).
Limitations
This study, however, does have its methodological and interpretive limitations, and some of the
findings require further discussion. First, the
responding users might not represent the whole
243
A Cross-National Study of Mobile Internet Services
population because mobile convergence is still
in its early stages. That is, the subjects of the
study were recruited as a representative of early
adopters of advanced mobile services. Thus, this
study does not provide comprehensive comparisons of the two countries, rather just a snapshot
of a subset of subscribers (see Figure 3). It would
have been better to examine the user motivations
from different groups of users with a longitudinal
investigation.
Second, the current study did not include involvement as a study variable, which has been
the common variable in the U&G study. Future
studies may combine the involvement and interactivity variables to examine the role of involvement in the structural relations among motivation,
interactivity, and attitude. At the same time, future
studies may delete some especially skewed items
and report the multicollinearity diagnostics.
Third, this study put satisfaction as a consequence of U&G. This endeavor can be a new
approach, but at the same time it calls for further
research. Most U&G research or research using
Rogers’s (1995) Diffusion of Innovation―decision to adopt, use, and, subsequently, gratifications
obtained from the technology used―is usually the
consequence of attitude toward that technology.
Rogers called this the persuasion stage, which
is an important component in the innovationdecision process―the individual forms a favorable
or unfavorable attitude toward the innovation.
The more positive the technology developed
from these attributes, the more gratifications
users will seek and/or obtain from the use of the
technology. Furthermore, attitudes will also be
influenced by peers, opinion leaders, and trend
setters, and in turn affect their adoption and uses
and gratifications of that technology. Therefore,
putting attitudes as consequences of U&G leaves
room for further discussion. There may be many
other factors influencing the adoption and use of
mobile services, especially in the case of advanced
multimedia convergence services. Future studies
can expand the scope of this study by examining
244
the inhibiting, limiting, and enabling factors influencing mobile service adoption and use at the
individual, social, and system levels.
Finally and most importantly, a key limitation of this study may be that it did not consider
external factors (different infrastructure, service
provisions, and content quality). This limitation
poses a challenge for the analysis, particularly
the different usage and gratification patterns. The
differences can be largely attributed to differences
between the two countries’ infrastructure, industry
structure, and market. Thus, future studies can
investigate the differences further incorporating
such external factors.
CONCLUSION
Practical and Theoretical
Implications
The findings of this study offer several practical
implications to mobile service providers. First,
the most apparent difference between Korean and
U.S. users would be different kinds of gratifications: latent and manifest gratifications. As shown
in the latent gratifications, Korean users tend
to see a mobile device as a tool that is “always
connected” (Peters & Allouch, 2005, p. 239).
Koreans tend to expect the new mobile communication technology to be accessible at all times
and places, leading technology-enabled people to
adapt the new mobile communication technology
almost automatically into their daily lives (Shin,
2007). On the other hand, U.S. users tend to see
manifest gratifications from mobile services such
as entertaining content or informative services.
While Katz and Aakhus (2002) argue that both
manifest and latent functions are required for a
full understanding of the role of personal communication technologies in people’s daily lives,
latent and manifest gratifications influence uses
and adoption differently in this case.
A Cross-National Study of Mobile Internet Services
Figure 3. Profile of respondents by country
This difference of latent and manifest gratification may be associated with different mobile
infrastructure. Mobile users in Korea, for example,
have been able to use their mobile phone for
years, whereas users in the U.S. perhaps have a
rare opportunity to see nice smart phones, and
therefore mainly explore basic functionalities.
In other words, subscribers in less mature mobile
markets do not utilize advanced functions on a
regular basis. Although U.S. mobile operators have
spent a great deal of resources in promoting new
advanced services, the usage levels may have not
caught up with the leading market of Korea yet.
Second, mobile service providers may consider providing infotainment mobile services,
combining information with entertainment. This
recommendation is based on the findings that
both information and entertainment are found to
245
A Cross-National Study of Mobile Internet Services
be reasonably significant antecedents of attitudes.
Possible infotainment services for mobile Internet
may include user polls, quizzes, and contests for
education-related contents. Indeed, demand for
mobile data/visual services grows, and providers are aggressively seeking new infotainment
mobile services that possibly become killer applications. Further, mobile services have become
more versatile and more multi-functional, and this
study confirms that mobile entertainment value
was found to be important determinant of users’
attitudes toward mobile services, which may lead
to more satisfied users and repeated use of mobile
services. The insignificant different satisfactions
between the U.S. and Korean users from the t-test
analysis imply the lack of compelling content. The
insignificant difference of usage time also implies
that U.S. users and Korean users commonly feel
that there is a lack of compelling content and
services. While mobile service providers may be
disappointed by their lack of subscribers, users
may be more dissatisfied with the content, as
much of the video offerings available currently
are limited to clips. This finding implies a topic
for future study of how to remake mobile content
so that it is optimally suited to mobile devices.
Third, given the different formation of path
estimates in two countries, mobile carriers may
focus on localization of mobile services. The
two models show interesting contrasts: (1) the
H1 and H2 of Korean users are supported at the
0.001 significance level, whereas those of U.S.
users are supported at the 0.05 level; (2) while the
H3 and H5a of Korean users are supported at the
0.05 level, those of U.S. users are supported at
the 0.001 level; and (3) the H4 of Korean users is
supported at the .001 level, but that of U.S. users
is rejected. These contrasts imply that the mobile
users in the two countries have different extent
and expectation on mobile Internet services. The
different formations further suggest mobile carriers adopt localization strategies for a particular
country to optimally exploit local markets. As
mobile services become global, mobile carriers
246
may pursue global strategies for the global markets
and they may adopt different localization strategies
for different markets to export mobile service. The
unique characteristics of local mobile markets
along with the increasing trend for global markets
have led to a growing need for better strategies.
In addition to the practical implications, the
present study also contributes to the scholarly
literature in a number of ways. First, it is a new
approach intended to investigate simultaneously
antecedents and consequences of attitude toward
the increase in innovative mobile services. By
using the SEM method, this study supports the
convergent and discriminant validity of various
scales, such as attitude toward convergence services and U&G. These scales and constructs will
be useful for future convergence service usage
motivation and diffusion factor research. SEM
has been regarded as a more promising approach
than multiple regression, which is commonly
used in mass communications research (Cheng,
2001). Issues in convergence media are more
complicated, in that independent variables are
still unknown and many more latent variables
need to be uncovered. A SEM is a good tool for
identifying relationships among manifest and
latent variables. Future studies may use SEM to
advance theory development and testing in the
convergence media by identifying more latent
variables of convergence media.
Second, this study supports an application of
U&G theory to emerging convergence technology. Dickinger, Haghirian, Murphy, and Scharl
(2004) argue that the convergence media uses
and effects process is a complex one that requires
careful attention to antecedent, mediating, and
consequent conditions. In this new, convergence
environment, in which the presence of the mobile is becoming as ubiquitous as television, it is
critical to know why and how people use mobile
phones or wireless Internet to watch broadcasting
content. Furthermore, as users become more and
more proactive, it is important to figure out how
they utilize a variety of mobile services. For this
A Cross-National Study of Mobile Internet Services
purpose, the present study provides a theoretical
model of the relations among motivations for using
mobile service such as interactivity, entertainment,
and informativeness. Subsequently, the results of
this study may help predict and evaluate mobile
service effectiveness for mobile users based on
their motivations and preferred options.
In addition, based on this result, this study
proposes a new gratification specific to mobile Internet, latent gratification or embedded gratification―gratifications that are different from actual
gratifications, which can be obtained from an actual medium’s content accompanied by immediate
pleasure (e.g., watching a specific program), and
are, instead, gratifications that users feel psychological readiness to access Internet resources at
any time. Convergence services via mobile devices
represent a crucial test of the importance of manifest vs. latent gratifications of mobile services.
The results show that even if mobile service has
been designed to provide users with the functional
needs of mobility, latent motivations of “always
connected” are just as important as perceived
usefulness to the individual end user’s adoption/
use of the service. So the latent gratification can
be unconscious gratification―users’ perception
of connectedness, being emotionally connected
with the world, its resources and people. In other
words, users can be gratified or comforted with
psychological connectedness that they can enjoy
at their convenience that used to be available only
through traditional media or technologies.
FUTURE RESEARCH: DIRECTIONS
AND POSSIBILITIES
The strengths of this study lie in the investigation
of new convergence mobile services with U&G
frame and in a cross-country comparison. The
present study sheds light on positive potentials
of the U&G theory in emerging mobile technologies. This underlines that new services should be
rooted in known user interfaces and cultural dif-
ferences. Therefore, any attempt to speculate on
the future direction of social technology theory
should include a U&G approach. Among emerging technologies, convergence services provided
by mobile devices give U&G researchers some
interesting possibilities for verification of their
previous results. This research could be useful to
the U&G field, as it provides a picture of how users attempt to put their media use in the best light.
These findings could then be used in U&G research
where the results are not so readily verifiable; the
findings could provide useful insight into how users responded to the initial survey. This could lead
to a more comprehensive view of how people use
the mobile media, and more importantly, how they
perceive their use of mobile media. In this regard,
future studies should investigate user perceptions
on new concepts such as self-gratification. For
example, convergence media may provide self
gratification to the users who adopt and use the
media. Users may give themselves pleasure or
satisfy other gratifications offered by media. In
addition, the insignificant effect of attitude on
usage is a noteworthy topic for future studies to
delve into possible multi-processes facilitating
customers to adopt technologies.
Future research topics might include: the additional latent and manifest gratifications served
by the mobile Internet; and how individuals make
sense of content from new convergence technologies compared to the cognitions, emotions, and
predispositions people derive from counterpart
legacy technologies. Researchers must also be
willing to explore the direction of causality by
examining changes over time with a longitudinal design (because convergence service is in a
still early stage of diffusion). Future studies may
proceed with a broader inquiry on cultural factors, such as cultural ethnocentrism and cultural
value structure. The emphasis on a cross-country
approach in this study comes essentially from
the intention to avoid cultural ethnocentrism in
building the argument. A longitudinal study of the
dimensions of this study can be undertaken, using
247
A Cross-National Study of Mobile Internet Services
a wider sample of cultures. Finally, future studies may investigate comparative data with other
types of service usage, for example comparisons
between mobile TV and IPTV can provide additional insights on the social and cultural impact
of new technologies.
ACKNOWLEDGMENT
This work was supported by grant No. R31-10062
from the World Class University (WCU) project
of the Korean Ministry of Education, Science &
Technology (MEST) and the Korea National Research Foundation (NRF) through Sungkyunkwan
University (SKK Univ.). The author sincerely
thanks MEST and NRF for proving valuable
resources to this research project. The author is
also very grateful to SKK University for their
administrative support.
REFERENCES
Cheng, E. (2001). SEM being more effective than
multiple regression in parsimonious model testing
for management development research. Journal
of Management Development, 20(7), 650–667.
doi:10.1108/02621710110400564
Coursaris, C., Hassanein, K., & Head, M. (2003).
M-commerce in Canada: An interaction framework for wireless privacy. Canadian Journal of
Administrative Studies, 20(1), 54–73.
Dickinger, A., Haghirian, P., Murphy, J., & Scharl,
A. (2004). An investigation and conceptual model
of interactive marketing. Proceedings of the 37th
Hawaii International Conference on System Sciences.
Eighmey, J. (1997). Profiling user responses to
commercial Web sites. Journal of Advertising
Research, 37(3), 59–67.
Eighmey, J., & McCord, L. (1998). Adding value
in the information age: Uses and gratifications of
sites on the World Wide Web. Journal of Business Research, 41, 187–194. doi:10.1016/S01482963(97)00061-1
Aaker, D. A., & Stayman, D. M. (1990). What
mediates the emotional response to advertising? The case of warmth. In P. Cafferata & A.
Tybout (Eds.), Cognitive and affective reactions
to advertising (pp. 287-304). Lexington, MA:
Lexington Books.
Gerbing, W., & Anderson, J. (1988). An updated
paradigm for scale development incorporating
unidimensionality and its assessment. JMR,
Journal of Marketing Research, 25, 186–192.
doi:10.2307/3172650
Angleman, S. (2000). Uses and gratifications and
Internet profiles: A factor analysis. Is Internet use
and travel to cyberspace reinforced by unrealized
gratifications? Proceedings of the Western Science
Social Association 2001 Conference, Reno, NV.
Gokhale, S., & Lu, J. (2006). Performance and
availability: Analysis of an e-commerce Site.
Proceedings of the 30th Annual International
Computer Software and Applications Conference
(pp. 495-502).
Charney, T., & Greenberg, B. S. (2002). Uses and
gratifications of the Internet. In C.A. Lin & D.J.
Atkin (Eds.), Communication technology and society (pp. 379-407). Cresskill, NJ: Hampton Press.
Grace-Farfaglia, P., Dekkers, A., Sundararajan,
B., Peters, L., & Park, S.-H. (2006). Multinational
Web uses and gratifications: Measuring the social
impact of online participation across national
boundaries. Electronic Commerce Research, 6,
71–96. doi:10.1007/s10660-006-5989-6
Chen, Q., & Wells, W. D. (1999). Attitude toward
the site. Journal of Advertising Research, 39(5),
27–37.
248
A Cross-National Study of Mobile Internet Services
Hair, J. F., Jr., Anderson, R. E., Tatham, R. L., &
Black, W. C. (1998). Multivariate data analysis.
Englewood Cliffs, NJ: Prentice Hall.
Haley, R. I., & Baldinger, A. L. (1991). The ARF
copy research validity project. Journal of Advertising Research, 31(2), 11–32.
Jee, K. Y., & Kim, M. G. (2005). Korean user attitude for adoption of portable Internet services.
Proceeding of the Conference on Communication
Systems and Applications.
Katz, J. E., & Aakhus, M. A. (2002). Perpetual
contact: Mobile communication, private talk,
public performance. Cambridge: Cambridge
University Press.
Kim, J., Lee, I., & Lee, Y. (2004). Exploring
e-business implications of the mobile Internet:
A cross-national survey in Hong Kong, Japan
and Korea. International Journal of Mobile
Communications, 2(1), 1–21. doi:10.1504/
IJMC.2004.004484
Korgaonkar, P. K., & Wolin, L. D. (1999). A
multivariate analysis of Web usage. Journal of
Advertising Research, 39(2), 53–68.
Kwon, H., & Chidambaram, L. (2000, January).
A test of the technology acceptance model the
case of cellular telephone adoption. Proceedings
of the 33rd Hawaii International Conference on
System Sciences.
LaRose, R., Mastro, D., & Eastin, M. (2001).
Understanding Internet usage: A social-cognitive approach to uses and gratifications. Social
Science Computer Review, 19(4), 395–414.
doi:10.1177/089443930101900401
Laughlin, Z. (2001). Mobile phone users and
diffusion rate: A small-scale observational study.
Retrieved from http://www.aber.ac.uk/media/
Students/zgl9901.html
Lee, W. J., Kim, T., & Chung, J. (2002). User acceptance of the mobile Internet. Working paper, IT
Management Research Center, Graduate School
of Business, Sungkyunkwan University, Korea.
Leung, L., & Wei, R. (1998). The gratifications
of pager use: Sociability, information-seeking,
entertainment, utility, and fashion and status.
Telematics and Informatics, 15(4), 253–264.
doi:10.1016/S0736-5853(98)00016-1
Leung, L., & Wei, R. (2000). More than just talk
on the move: Uses and gratifications of the cellular phone. J&MC Quarterly, 77(2), 308–320.
Lin, C. (1999). Online service adoption likelihood.
Journal of Advertising Research, 39, 79–89.
Livaditi, J., & Vassilopoulou, K. (2003). Needs
and gratifications for interactive TV applications.
Proceedings of the 36th Annual Hawaii International Conference on System Sciences.
Lull, J. (1990). Inside family viewing. London:
Comedia.
McQuail, D. (2000). Mass communication theory.
Thousand Oaks, CA: Sage.
Nysveen, H., Pedersen, E., & Thorbjørnsen, H.
(2005). Intentions to use mobile services. Journal of the Academy of Marketing Science, 33(3),
330–346. doi:10.1177/0092070305276149
Palen, L., Salzman, M., & Youngs, E. (2001). Discovery and integration of mobile communications
in everyday life. Personal and Ubiquitous Computing, 5(2), 109–122. doi:10.1007/s007790170014
Palmgreen, P., & Rayburn, J. D. II. (1985). A
comparison of gratification models of media
satisfaction. Communication Monographs, 52,
334–346. doi:10.1080/03637758509376116
Park, C., & Jun, J. (2003). A cross-cultural
comparison of Internet buying behavior. International Marketing Review, 20(5), 534–553.
doi:10.1108/02651330310498771
249
A Cross-National Study of Mobile Internet Services
Peters, O., & Allouch, S. (2005). Always connected: A longitudinal field study of mobile communication. Telematics and Informatics, 22(3),
239–256. doi:10.1016/j.tele.2004.11.002
Rauh, C., & Renfro, S. (2004). Feedback delay
effects in video monitor communication. Proceedings of the Conference of the International
Communications Associations.
Rhee, K., & Kim, W. (2004). The adoption and
use of the Internet in South Korea. Journal of
Computer-Mediated Communication, 9(4). Retrieved from http://jcmc.indiana.edu/vol9/issue4/
rhee.html
Rogers, E. M. (1995). Diffusion of innovations
(4th ed.). New York: The Free Press.
Sherry, J. (2001). The effects of violent video
game on aggression. Human Communication
Research, 27, 409–431.
Shin, D. (2007). User acceptance of mobile Internet: Implication for convergence technologies.
Interacting with Computers, 19(4), 472–483.
doi:10.1016/j.intcom.2007.04.001
Shin, D. (2008). What people do with digital
multimedia broadcasting? Path analysis of
structural equation modeling. Journal of Mobile
Communications, 6(1), 258–275. doi:10.1504/
IJMC.2008.016580
Shin, D. (in press). Virtual gratifications of wireless
Internet: Is wireless portable Internet reinforced
by unrealized gratifications? Telematics and
Informatics.
Sordegard, C. (2005). Mobile television: Technology and user experiences. Retrieved October 2005
from http://www.vtt.fi/inf/pdf
250
Stafford, T. F. (2004). Determining uses and gratifications for the Internet. Decision Sciences, 35(2),
259–288. doi:10.1111/j.00117315.2004.02524.x
Stafford, T. M., & Gillenson, M. L. (2004, December 10-11). Motivations for mobile devices: Uses
and gratifications for m-commerce. Proceedings
of the 3rd Annual Workshop on HCI Research in
MIS, Washington, DC.
Sujan, H., Weitz, B. A., & Kumar, N. (1994).
Learning orientation, working smart, and effective selling. Journal of Marketing, 58, 39–52.
doi:10.2307/1252309
Sun, S., & Su, T. (2006). A comparative study
of value-added mobile services in Finland and
Taiwan. International Journal of Mobile Communications, 4(4), 436–458.
Torkzadeh, G., & Van Dyke, T. P. (2001). Development and validation of an Internet self-efficacy
scale. Behaviour & Information Technology, 20(4),
275–280. doi:10.1080/01449290110050293
Tsang, M., Ho, S., & Liang, T. (2004). Consumer
attitudes toward mobile advertising: An empirical
study. International Journal of Electronic Commerce, 8(3), 65–78.
Venkatesh, V., & Morris, M. G. (2000). Why don’t
men ever stop to ask for directions? Gender, social
influence, and their role in technology acceptance
and usage behavior. MIS Quarterly, 24, 115–139.
doi:10.2307/3250981
Weilenmann, A., & Larsson, C. (2000). On doing
being teenager: Applying ethno-methodology
to the analysis of young people’s use of mobile
phones. In L. Svensson, U. Snis, C. Sørensen,
H. Fägerlind, T. Lindroth, M. Magnusson, & C.
Östlund (Eds.), Proceedings of IRIS 23, Trollhättan, Sweden.
A Cross-National Study of Mobile Internet Services
APPENDIX: A SAMPLE OF SURVEY QUESTIONNAIRE
Part 1. Profiling Questions
Gender: Male or Female
My age:
My highest level of education is:
____Below High School
____High School
____Attended College
____Undergraduate Degree
____Graduate School
My annual household income is:
____Under $20,000
____$20,000-$45,000
____$45,000-$65,000
____$65,000-$90,000
____$90,000-$125,000
____$125,000 plus
____Prefer not to say
My occupation is:
____Professional
____Owner (self-employed)↜____Technician
____Secretarial
____Student
____Housewife
____Retired/unemployed
____Others
Part 2. Mobile Internet Service Questions
Entertainment
1.
2.
3.
Having a mobile Internet is helpful for me to enjoy the pleasure of watching.
Strongly Disagree____ Disagree____ Neutral____ Agree____ Strongly Agree____
Having a mobile Internet is helpful for me to pass the time.
Strongly Disagree____ Disagree____ Neutral____ Agree____ Strongly Agree____
Having a mobile Internet is helpful for me because using mobile services relaxes me.
Strongly Disagree____ Disagree____ Neutral____ Agree____ Strongly Agree____
Informativeness
1.
Mobile Internet gives me quick and easy access to large volume of information.
Strongly Disagree____ Disagree____ Neutral____ Agree____ Strongly Agree____
251
A Cross-National Study of Mobile Internet Services
2.
3.
Information obtained from the mobile is useful.
Strongly Disagree____ Disagree____ Neutral____ Agree____ Strongly Agree____
I learned a lot from using the mobile service.
Strongly Disagree____ Disagree____ Neutral____ Agree____ Strongly Agree____
Interactivity
1.
2.
3.
I like the mobile service to be interactive.
Strongly Disagree____ Disagree____ Neutral____ Agree____ Strongly Agree____
I like to increase my control over mobile service.
Strongly Disagree____ Disagree____ Neutral____ Agree____ Strongly Agree____
Using mobile, I like to interactively communicate with others.
Strongly Disagree____ Disagree____ Neutral____ Agree____ Strongly Agree____
Availability
1. I am gratified with the ubiquity of mobile Internet that I can watch content anytime with
mobile Internet service.
Strongly Disagree____ Disagree____ Neutral____ Agree____ Strongly Agree____
2. Because of mobile Internet, I feel like that I am in connection with my friends and family.
Strongly Disagree____ Disagree____ Neutral____ Agree____ Strongly Agree____
3. With mobile Internet, I can enjoy several applications synchronously or I can access
real-time news synchronously.
Strongly Disagree____ Disagree____ Neutral____ Agree____ Strongly Agree____
Part 3. Actual Use
How many years have your been using mobile Internet services?
How many times do you believe you use mobile Internet service during a week?
____ Not at all
____ Less than about 2 or 3 a week
____ Several times a week
____ About once a day
____ Several times each day
How many hours do you believe you use mobile Internet service every week?
____ Less than 1hr.
____ Between 1-5 hrs.
____ Between 5-10 hrs.
____ Between 10-15 hrs.
____ Between 15-20 hrs.
____ Between 20-25 hrs.
____ More than 25 hrs.n
This work was previously published in International Journal of Global Information Management, Volume 17, Issue 4, edited
by Felix B. Tan, pp. 29-54, copyright 2009 by IGI Publishing (an imprint of IGI Global).
252
Section 3
Trends and Techniques
254
Chapter 12
Flexibility in Offshoring:
Paradoxical Concepts of Control,
Communication and Facilitation
Vanita Yadav
Institute of Rural Management Anand (IRMA), India
Monica Adya
Marquette University, USA
Varadharajan Sridhar
Sasken Communication Technologies, India
Dhruv Nath
Management Development Institute, India
ABSTRACT
Globalization of software development has resulted in a rapid shift away from the traditional collocated,
on-site development model, to the offshoring model. Emerging trends indicate an increasing interest in
offshoring, even in early phases like requirements analysis. Additionally, the flexibility offered by the
agile development approach makes it attractive for adaptation in globally distributed software work. A
question of significance then is what impacts the success of offshoring earlier phases, like requirements
analysis, in a flexible and globally distributed environment? This chapter incorporates the stance of
control theory to posit a research model that examines antecedent factors such as requirements change,
facilitation by vendor and client site-coordinators, control, and computer-mediated communication.
The impact of these factors on success of requirements analysis projects in a “flexible” global setting is
tested using two quasi-experiments involving students from Management Development Institute, India
and Marquette University, U.S.A. Results indicate that formal modes of control significantly influence
project success during requirements analysis. Further, facilitation by both client and vendor site coordinators positively impacts requirements analysis success.
DOI: 10.4018/978-1-60960-605-3.ch012
Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
Flexibility in Offshoring
INTRODUCTION
Globalization has resulted in software development being outsourced to emerging and developing nations (Edwards & Sridhar, 2005). An
increasing range of services and processes are
being delivered by global vendors as per quality,
price, and requirements independent of geography,
suggesting a growing capability and acceptance
of global service delivery (NASSCOM, 2007).
Outsourcing is the largest and fastest growing
category within worldwide IT services spending.
In 2006 the total spending on IT outsourcing was
estimated at over USD 170 billion (more than 36
percent of the total) with an above average growth
at 7.3 percent (NASSCOM, 2007). Accordingly,
software development has moved away from the
traditional co-located model, often called on-site
development, to the offshoring model (Edwards &
Sridhar, 2006) in which global virtual teams collaborate across national borders (Carmel, 1999).
Global software development (GSD) presents
abundant business opportunities as well as challenges in terms of control, coordination, communication, culture, and technology. To address
these challenges many researchers propose that
firms must have ambidextrous capabilities (Lee,
Delone & Espinosa, 2006) and combine the
flexibility offered by the growing agile development approaches with the traditional plan-based
approaches (Lee, Banerjee, Lim, Hillegersberg
& Wei, 2006; Agerfalk & Fitzgerald, 2006; Ramesh, Cao, Mohan & Xu, 2006). As organizations
become more virtual, distributed development
will become increasingly apparent throughout
the entire software development life-cycle, particularly so in early stages such as requirements
analysis (Evaristo, Watson-Manheim & Audy,
2005). Despite the abundance of literature on
globally distributed virtual teams (see Powell,
Piccoli & Ives, 2004) and IT outsourcing (see
Yadav & Gupta, 2008; Dibbern, Goles, Hirschheim
& Jayatilaka, 2004), very few studies address
the critical requirements analysis phase of GSD
(Yadav, Nath, Adya & Sridhar, 2007).
Requirements analysis refers to that stage of
the system development life cycle wherein the
information and information processing services
needed to support select objectives and functions
of the organization are (i) determined and (ii) coherently represented using well defined artifacts
such as entity-relationship diagrams, data flow
diagrams, use cases, and screen prototypes (Hoffer, George & Valacich, 1999). Typically in GSD
this phase is conducted at the client location since
this phase requires significant interaction between
users and developers. Business and systems analysts are physically located at the client site to
perform this activity. Depending on the nature of
the project, high-level design is conducted in both
on-site and off-shore mode due to comparatively
lower interaction needs with the client. Detailed
design, coding, and testing are executed at the
off-shore site (Carmel & Tijia, 2005).
Damian and Zowghi (2002) report that in
global projects consultant teams from the offshore location travel to the user site to gather and
analyze requirements in face-to-face meetings.
The consultants then communicate the requirements to the development staff at the offshore
locations. An emerging stream of research, on
the other hand, puts forward the phenomenon
of distributed requirements engineering (Bhat,
Jyoti, Gupta & Murthy, 2006; Nath, Sridhar,
Adya & Malik, 2006; Edwards & Sridhar, 2005;
Evaristo, Watson-Manheim & Audy, 2005). An
interesting alternative being considered by software companies is the possibility of off-shoring a
larger part of the software development process.
Specifically, is it possible to effectively conduct
the requirements analysis phase from offshore
location which is traditionally done on-site? In
such a scenario, analysts and developers located
at the off-shore location would interact in a virtual
mode with the clients located at their premises to
determine and structure the requirements. Such a
shift could potentially improve the cost arbitrage
255
Flexibility in Offshoring
of the projects for instance, by cutting down the
travel costs incurred for sending analysts to the
client site for face-to-face meetings and reduced
staffing needs at the client end. It would also
provide an opportunity to build client and developer relationships using computer-mediated
communication.
Recent years have also witnessed growth of
the agile movement that focuses on producing a
working solution in conjunction with changing
user requirements. The flexibility and responsiveness of the agile approach makes it attractive for
adaptation in globally distributed software work.
Flexibility in GSD can be incorporated by adapting some of the principles of agile development
like simplified project planning, acknowledging
requirements change, lesser emphasis on processes
and documentation, and supporting informal as
well as formal communication (Yadav et. al.,
2007).
In the requirements analysis phase of offshore
GSD projects, the absence of analysts and developers at customer premises is likely to create a
need to exercise control even in a flexible environment to ensure that the project meets defined
goals. Offshore vendors often deploy liaisons
who coordinate activities between on-site users
and offshore development team. These liaisons
are critical for effective communication and coordination between users and developers (Battin,
Crocker, Kreidler & Subramanian, 2003).
This study proposes to examine two research
questions related to flexible GSD: what are the
antecedents of requirements analysis success in a
flexible GSD environment? How do these antecedents impact success? Founded on control theory,
this paper posits a research model examining
relationships between process facilitation (by client/vendor site-coordinators), computer-mediated
communication, control, change in requirements
and requirements analysis success.
Our interactions with managers in client and
vendor firms engaged in GSD indicate that ‘total’
offshoring of requirements engineering is still
256
uncommon. However, they assert that in some
smaller projects, up to 75% of the requirements
analysis is carried out in offshore mode. For this
reason, it might not be feasible to analyze ‘total’
offshoring of requirements analysis phase in a
real-life setting. Furthermore, as client nations
face a growing shortage of business and systems
analysts, organizations may be compelled to
consider offshoring of early GSD phases. Given
these arguments, we designed exploratory quasiexperimental research studies in an academic
setting involving management students enrolled
in a graduate level information systems course
at Management Development Institute (MDI),
India, and management students enrolled in a
graduate level IT Project Management course at
Marquette University (MU), U.S.A. MU students
role-played clients while MDI students role-played
systems analysts.
The paper is organized as follows. In the next
section we review the background literature.
Subsequently, the theoretical foundation, research
hypotheses and conceptual model are presented,
followed by research methodology, and an overview of the findings. The paper concludes with
implications and directions for future research.
REQUIREMENTS ANALYSIS IN
FLEXIBLE GSD PROJECTS
One of the most challenging aspects of system
development is ascertaining ‘what the system
should do’, that is in determining the system
requirements (Crowston & Kammerer 1998, p.
227). Globally distributed requirements analysis generally includes a team of analysts and
users working together using technologies like
computer-mediated conferencing, instant messaging, emails, teleconferencing and web based
group support systems.
There is a growing debate on what can be and
what cannot be offshored? One school of thought
suggests that certain activities, like coding, are a
Flexibility in Offshoring
better fit for offshore locations while other activities, like requirements gathering, are better to be
carried out onshore within the client’s country
(Carmel & Tijia, 2005). On the other hand, there
is a growing stream of researchers who reason
in favor of distributed requirements engineering.
Few studies have examined the use of virtual
and globally distributed teams for requirements
analysis. Edwards and Sridhar (2005) studied the
effectiveness of virtual teams in a collaborative
requirements analysis practice. In this study virtual
GSD teams at near and far locations participated
in requirements analysis phase of the project.
This typically is applicable in collaborative global
product development exercises as described in
Battin et al. (2003). However, in our research
we study the requirements analysis phase of offshored software projects consisting of (i) users of
the client who specify the requirements, and (ii)
developers/analysts of the vendor located at an
offshore development center who determine and
document these requirements. We define a virtual
GSD team to comprise of these two protagonists
who rarely meet face-to-face and who work together using computer mediate communications.
Damian and Zowghi (2002) studied the interplay between culture and conflict and the impact
of distance on the ability to reconcile different
viewpoints with respect to “requirements negotiation” processes. They find that lack of a common
understanding of requirements, together with
reduced awareness of local context, trust level,
and ability to share work artifacts significantly
challenge effective collaboration among remote
stakeholders in negotiating a set of requirements
that satisfies geographically dispersed customers.
Damian et al. (2000) examined the effect of the
distribution of various stakeholders in requirements engineering process. They found that highest group performance occurred when customers
were separated from each other and collocated
with the facilitator or system analyst. Our study
aims to contribute to the literature on globally
distributed virtual teams engaged in off-shored
software requirements analysis.
FLEXIBILITY IN GSD
Continued dissatisfaction with the traditional
plan-driven (heavyweight) software development methodologies have led to the introduction
of various agile (lightweight) methodologies,
like eXtreme programming, Scrum, Crystal etc.
(Fruhling & Vreede, 2006; Lindstrom & Jeffries,
2004). Practice-led agile methods have been
proposed as a solution that addresses problems,
such as budget/schedule overruns and poor quality levels, by promoting communication, flexibility, innovation and teamwork (Agerfaulk &
Fitzgerald, 2006; Augustine, Payne, Sencindiver
& Woodcock, 2005). The agile alliance movement
was motivated by the observation that software
teams in many corporations are entrapped in an
ever-increasing amount of process and documentation (Fruhling & Vreede, 2006). Agile approaches
focus on fast deliverables, dynamic management
of requirements, and fast iterations and increments
(Fruhling & Vreede, 2006). While many benefits
of these newer approaches have been proposed,
very few field studies have empirically operationalized agile GSD (Fruhling & Vreede, 2006)
and developed theories in this area (Agerfaulk &
Fitzgerald, 2006).
Our informal interactions with client and vendor organizations in US and India also indicate that
in reality what exits is something in between the
spectrum of the traditional and agile approaches.
Though industry continues to debate whether or
not agile processes are appropriate for distributed
teams, let alone large and offshore distributed
teams, organizations like Sapient have been successful in tailoring the agile approach to incorporate flexibility using a mixed focus on people,
process, and tools (Barnett, 2006). Therefore on
a spectrum, flexible GSD can be visualized as
257
Flexibility in Offshoring
Figure 1. Flexibility spectrum
an ‘agile-rigid’ approach that lies in between the
traditional and agile approaches (see Figure 1).
To highlight the differences, the key characteristics of the traditional, agile and flexible GSD
environments adapted from a current stream of
research (Holmstrom, Fitzgerald, Agerfalk &
Conchuir, Eoin, 2006; Lee, Delone & Espinosa,
2006; Ramesh et al., 2006; Lee et al., 2006; Furhling et al., 2006) are presented in Table 1. Flexible GSD offers ‘rigor’ of the traditional approaches and ‘flexibility’ of the agile approaches
within the teams (Yadav et al., 2007). Rigor is
provided by incorporating formal structures of
the traditional approach, such as, development of
a project plan, communication plan and project
status tracking. Agility is allowed through simplified project planning. Elaborate project management techniques are tailored to make them ‘lightweight’. ‘Simple rules’ (Augustine et al., 2005)
are adopted. Formal as well as informal channels
of communication are encouraged. Additionally,
it adopts the agile philosophy of embracing requirements change. The agile principle states that
we should welcome changing requirements in
software projects as they harness change for the
customer’s competitive advantage (Lindstrom &
Jeffries, 2004). However, Fruhling and Vreede
(2006) indicate that the impact of requirements
change has not been studied empirically.
Table 1. Characteristics of the traditional, agile and flexible approaches
Traditional Development
Approach
Agile Development
Approach
Flexible GSD
Approach
Fundamental Assumption
Systems are fully specifiable,
predictable and can be built
through meticulous and extensive
planning.
High-quality, adaptive software can
be developed by small teams using
principles of continuous design improvement and testing based upon
rapid feedback and change.
High-quality, adaptive software can
be developed by globally distributed
teams using principles of continuous
improvement based upon feedback
and change having some amount of
planning and control.
Communication
Formal
Informal
Formal and informal
Project Management
-Process-centric
-Processes over people
-Extensive milestone planning
- Extensive documentation
-People-centric
-People over processes
-Respond to change over following
a plan
- Lack of documentation
-Equal importance to people and
processes
- Medium project planning
-Medium documentation
Requirements
-Knowable early; largely stable
-Detailed specification of requirements
-Largely emergent; rapid change
- Iterative development that produce
working solutions at the end of
each iteration to capture emerging
requirements
- Emergent
- Use prototypes or incremental
working solutions to capture emerging requirements
Size
Larger teams (>10 team members)
Smaller teams (2-8 team members).
Smaller teams or large teams broken
down to small sub-teams (<10 team
members).
258
Flexibility in Offshoring
UNDERLYING THEORETICAL
FOUNDATION
Global software development is not just a technical process of building software or information systems but also a social process involving
stakeholders from multiple organizational units.
Our paper draws upon control theory to study the
impact of antecedents on success during requirements analysis in a flexible GSD setting. Academic
(Kirsch & Cummings, 1996) and practitioner
(PMBOK, 2000) literature suggests that control
plays an effective role in managing projects. Control theory attempts to explain how one person or
group ensures that another person or group works
toward and attains a set of organizational goals.
Aligned with prior research on control (Kirsch,
1997; Choudhury & Sabherwal, 2003), our study
views control broadly as attempts to ensure that
individuals act in a manner consistent with achieving desired project goals. Control modes can be
“formal or “informal” (Crisp, 2003) in nature,
where former are documented and initiated by
management, and the latter are often initiated by
employees themselves.
Control theories suggest that controllers utilize
two modes of formal control: behavior and outcome (Eisenhardt, 1985; Kirsch et al., 2002). In
behavior control, appropriate steps and procedures
for task performance are defined by controllers
and then controllees’ performance is evaluated
according to their adherence to the prescribed
procedures. In outcome control, controllers define
appropriate targets and allow controllees to decide
how to meet those output targets. Controllees’
performance is evaluated on the degree to which
targets were met, and not the processes used to
achieve the targets (Kirsch et al., 2002). In this
study, we focus on the impact of formal modes
of control (outcome and behavior) to on success
of GSD projects during the requirements analysis
phase.
Kirsch et al. (2002) extended the control theory
to the role of client liaisons/coordinators exercis-
ing control of IS project leaders to ensure that IS
projects meet their goals. Lee et al. (2006) also
proposed assigning ‘point persons’/ coordinators
to offshore sites for effective management of GSD
projects. We draw upon this literature to study the
effect of process facilitation by both the client and
vendor site coordinators.
RESEARCH MODEL
To reiterate, this paper proposes to examine
the following research questions - what are the
antecedents of requirements analysis success
in a flexible GSD environment? How do these
antecedents impact requirements analysis success? Specifically, we explore how incorporating discipline through formal modes of control,
process facilitation by site-coordinators (at both
the client and vendor sites), and task related
computer-mediated communication in a flexible
GSD environment impacts requirements analysis
success. We also explore how change in requirements, which is an inherent assumption behind the
agile philosophy, impacts requirements analysis
success. Control theory guides our hypotheses
and model development process (see Figure 2).
The research model variables and hypotheses
specifying relationships between these variables
are presented in the subsequent sections.
Control
Outsourced projects in GSD pose unique problems
that make the task of controlling them particularly
challenging (Choudhury & Sabherwal, 2003). In
the case of outsourced GSD projects, the controller
and controllee may not be single individuals but
teams of individuals representing the client and
vendor organizations at globally distributed locations. Lee, Delone and Espinosa (2006) suggest
that agile methods should be tailored to embrace
more discipline in GSD. Derived from control
theory, we hypothesize that incorporating disci-
259
Flexibility in Offshoring
Figure 2. Research model
pline by using formal modes of control (outcome
and behavior) during requirements analysis in
flexible GSD projects is likely to help in achieving success (Yadav, Nath, Adya & Sridhar, 2007).
H1:Formal modes of control are positively related
to requirements analysis success in a flexible
GSD environment.
Computer-Mediated Communication
Computer-mediated communication uses computers to structure and process information, and
uses telecommunication networks to facilitate its
exchange (e.g. e-mails, computer conferencing,
e-groups). Rice (1987) indicates that computermediated communication is preferred in crosslocation interdependency tasks because of greater
freedom from temporal and geographic constraints
(e.g. offshoring). Computer-mediated communication allows group members to collaboratively
create meaning out of diverse sources of information leading to better outcomes (DeSanctis &
Gallupe, 1987).
260
The virtual GSD environment presents considerable challenges to effective communication
including time delays in sending feedback, lack
of a common frame of reference for all members,
differences in salience and interpretation of written
text, and assurance of participation from remote
team members (Crampton, 2001). Hulnick (2000,
pp. 33) noted that “if technology is the foundation
of the virtual business relationship, communication is the cement”. Information processing
perspective of communication indicates that task
uncertainty and work flow interdependence lead
to higher frequency of interaction between work
units (Putnam & Cheney, 1985). Extending this
to a GSD scenario during requirements analysis
there exits work flow interdependence between
the remote client and the vendor teams for accurately capturing and documenting requirements.
This is likely to lead to frequent task related
computer-mediated communication between
globally distributed team members for successful
completion of requirements analysis. Thus we
hypothesize that:
Flexibility in Offshoring
H2: Task-related computer-mediated communication is positively related to requirements
analysis success in a flexible GSD environment.
Process Facilitation
Lee, Delone and Espinosa (2006) indicate that
assigning ‘point persons’ in offshore software
development plays pivotal role in sensing and
responding to emergent problems on a real time
basis. Borrowing from IS Literature on Group
Support Systems (GSS) process facilitation is
defined as the provision of procedural structure and
general support to groups (Miranda & Bostrom,
1999). Our research acknowledges Crisp’s (2003)
view on structuring borrowed from control theory
and applies it in the context of GSD process facilitation where structuring refers to “any explicit
or implicit means of developing structures for
control”. In the case of outsourced GSD projects,
process facilitation can be provided by assigning
point persons/liaisons/site-coordinators at both
the client and vendor sites. We hypothesize that
process facilitation provided by site-coordinators
at offshore sites (client/vendor) can be a considered
as a structure that is more likely to assist productive outcomes (Yadav et al., 2007).
H3a: Process facilitation by vendor site-coordinator is positively related to requirements
analysis success in a flexible GSD environment.
H3b: Process facilitation by client site-coordinator is positively related to requirements
analysis success in a flexible GSD environment.
Structures are formal and informal procedures,
techniques, skills, rules, and technologies that
organize and direct group behavior and processes
(Anson, Bostrom & Wynne, 1995). We therefore
hypothesize that process facilitation provided by
site-coordinators (client/vendor) at offshore sites
can be a considered as a structure that is more
likely to assist in formal modes of control:
H4a: Process facilitation by vendor site-coordinator is positively related to formal modes
of control during requirements analysis in
a flexible GSD environment.
H4b: Process facilitation by client site-coordinator is positively related to formal modes
of control during requirements analysis in
a flexible GSD environment.
Ramesh et al. (2006) report that a primary
‘point of contact’/ a project lead, for each location in GSD helped in facilitating communication
across the teams. Griffith, Fuller and Northcraft
(1998) assert the fundamental role of a process
facilitator is to improve the effectiveness and
efficiency of communication and interaction of
group members in order to help the group achieve
outcomes. Therefore, we hypothesize that process
facilitation by vendor site-coordinator and also by
client site-coordinator is likely to lead to increased
computer-mediated communication:
H5a: Process facilitation by vendor sitecoordinator is positively related to taskrelatedcomputer-mediated communication
during requirements analysis in a flexible
GSD environment.
H5b: Process facilitation by client site-coordinator is positively related to task-related
computer-mediated communication during
requirements analysis in a flexible GSD
environment.
Requirements Change
The task of specifying requirements has ‘high
dynamic complexity’ (Briggs & Gruenbacher,
2002). This complexity stems from the evolutionary nature of requirements, which are clarified
only through multiple iterations of information
gathering (Evaristo et al., 2005). As organizations
261
Flexibility in Offshoring
become more global and stakeholders more distributed, getting the requirements right will pose
a greater challenge (Damian et al. 2000).
Flexible perspective borrowing from Agile
philosophy states that “changing requirements
are not necessarily bad but are welcomed as an
opportunity to satisfy customer needs even better
than inflexibly sticking to old requirements” (Turk
et al., 2005). Although many positive benefits
have been speculated for dynamic management
of requirements and fast iterations few have been
empirically examined (Fruhling & Vreede, 2006).
Founded on the agile philosophy of considering
requirements change as good opportunity to satisfy
customer needs, we hypothesizeH6a: Members of teams that have changes in
requirements will perceive higher level of
requirements analysis success than those
who have no changes.
When requirements change, it is assumed that
client team members are likely to experience a
greater need to communicate and monitor vendor
team members’ behavior to ensure changes are
incorporated. Therefore, it is likely that changes
in requirements would probably lead to increase
in control and task-related communication.
H6b: Members of teams that have changes in
requirements will perceive higher level of
control than those who have no changes.
H6c: Members of teams that have changes in
requirements will perceive higher level of
task related communication than those who
have no changes.
Tan & Teo (2007) recently pointed out that
project team using agile methodology could be
effective in responding to change in user requirements if it stressed collaboration among members
and users. Thus it is likely that changes in requirements leads to increased process facilitation by
site coordinators (client/vendor).
262
H6d: Members of teams that have changes in requirements will perceive higher level of process facilitation by vendor site-coordinator
than those who have no changes.
H6e: Members of teams that have changes in
requirements will perceive higher level of
process facilitation by client site-coordinator
than those who have no changes.
RESEARCH METHODOLOGY
Faculties in many universities and business schools
have set up distributed software engineering laboratories for conducting virtual team exercises in
their courses (Edwards & Sridhar, 2006; Nath et al.,
2007). Powell et al. (2004) have listed a number of
studies involving students in global virtual teams.
A controlled experimental approach provides three
benefits. Firstly, it makes available several teams
that work in parallel thereby generating rich data
for drawing conclusions. Secondly, it permits
researchers to experiment with newer approaches
which may not yet have been explored by the
industry. Finally, it equips and trains software
engineering students to understand and to handle
the challenges of working in global software teams
(Favela & Pena-Mora, 2001).
We conducted a quasi-experiment in a globally distributed academic setting involving MBA
students from two countries- India and USA.
Students from MDI, India posed as analysts from
vendor side while those from MU, USA posed as
clients. Vendor role was assigned based upon the
rationale that India is still the preferred sourcing
destination with a 58 percent share in worldwide
offshore IT-BPO market in spite of the expansion
of global sourcing arena (NASSCOM, 2007).
Similarly, client role was assigned based upon
the rationale that US alone accounts for about
two-thirds of the software and services exports
from India (NASSCOM, 2007).
Defining quasi-experimental designs, Campbell and Stanley (1966, pp. 34) state that “there are
many natural settings in which a research person
Flexibility in Offshoring
Table 2. ANOVA results
Variable
Levene Test (p>.05)
F-value
Significance*
€€€€€Age (years)
€€€€€€.908
€€€€€0.782
€€€€€€0.726
€€€€€Total Work Experience
€€€€€€.286
€€€€€1.592
€€€€€€0.908
€€€€€IT Work Experience
€€€€€€.083
€€€€€0.490
€€€€€€0.964
* Significance >.05 indicates no difference between the GSD teams for the variables that were controlled
can introduce something like experimental design
into his scheduling of data collection procedures
(e.g., the when and to whom of measurement), even
though he lacks full control over the scheduling
of experimental stimuli (the when and to whom
of exposures) which makes a true experiment
possible.” We designed a post-test control group
quasi-experiment design of the following form:
Mâ•…â•… Xâ•…â•… O1
Mâ•…â•… â•…â•… O2
‘M’ stands for matching (a priori equalization
of the two groups for the factors that have to be
controlled), ‘X’ stands for treatment and ‘O’ stands
for observation or measurement.
Marquette University has a service learning
office which obtains IS development projects from
non-profit organizations and small businesses in
and around Milwaukee. These real life projects
at Marquette University were used to create the
simulated flexible GSD requirements analysis
projects. This enabled the projects to closely
mirror real business environments. Examples of
these projects include a donation management
system for a non-profit organization, a volunteer management system, an alumni website, a
tracking system for battered and abused women,
and a book inventory management system. Only
high level requirements were provided, such as,
the system must be secure and accurate; it must
track certain information etc. Many of the detailed
level requirements were expected to emerge only
through remote team member communications for
requirements gathering and analysis.
The Experiment
The clients in US enrolled for a course in IT
Project Management and the analysts in India
enrolled for Management Information Systems
(MIS) course, having comprehensive coverage
of systems analysis and design, in Jan 2007. This
quasi-experiment was designed as a part of the
course project and its duration was 8 weeks. Students developed project artifacts iteratively using
structured software development methodology. A
flexible offshore GSD project environment was
simulated based upon characteristics highlighted
in Table 1. Each client team was paired with an
analyst team. The teams were controlled in terms
of age and work experience of team members and
team size (see Table 2).
MDI teams elicited project requirements from
MU teams using web-based communication
technology like e-groups (Google/yahoo groups),
text/voice/video chat (Skype/msn/yahoo) and
email (see Table 3 and Figure 3). The gathered
requirements were then structured using the process modeling tools such as Context Analysis
Diagram (CAD), Data Flow Diagrams (DFDs)
and Process Specifications. MDI teams also modeled the data and associated relationships using
Entity Relationship Diagrams (ERDs). Further,
to give the users a feel of what the final system
would look like, MDI teams also created screen
based prototypes as part of the requirements
analysis exercise. All these artifacts were submitted by the MDI teams to MU user teams in two
iterations (see Table 4).
263
Flexibility in Offshoring
Table 3. Experiment setting and treatment
Setting
Non-control Group
(Treatment: Changes in Requirements)
Control Group
(No Treatment)
Flexible GSD
All teams matched in
terms of age, size, work
experience, technology,
standard development
methodology
Sample Size= 90
Analysts=59, Clients=33
10 client-analyst GSD teams, each team having 5-6
analysts (1 analyst appointed, by consensus, as MDI
site-coordinator) and 3-4 clients (1 client appointed,
by consensus, as MU site-coordinator)
Sample Size= 91
Analysts=56, Clients=33
10 client-analyst GSD teams, each team having 5-6
analysts (1 analyst appointed, by consensus, as MDI
site-coordinator) and 3-4 clients (1 client appointed,
by consensus, as MU site-coordinator)
Figure 3. Experiment design
Table 4. Project artifacts
MDI Analyst Project Deliverables
MU Client Project Deliverables
Week 1: Simplified Project Management Plan (to MU clients with
a copy to MDI faculty and researcher). No formal resource allocation at the analyst end.
Week 1: Simplified Project Charter, Project Schedules, Resource
Allocation and Communication Plans (to the MU faculty).
Week 3: Simplified Project Status Reports to MU Clients.
Week 3: Simplified Risk Assessment and Contingency Plans (to
the MU faculty).
Week 4: First Iteration (Context Analysis Diagrams, Data Flow
Diagrams, Entity Relationship Diagrams, Process Specifications
and Screen-based Prototypes of the business IS) to MU Clients.
Week 4: Feedback to MDI analysts.
End of Week 4
Main Experiment: TREATMENT- Changes in Requirements for the Non-control Group
Week 6: Simplified Project Status Reports to MU Clients.
Week 6: Simplified Project Status Reports to MU faculty.
Week 8: Second Iteration (Context Analysis Diagrams, Data Flow
Diagrams, Entity Relationship Diagrams, Process Specifications
and Screen-based Prototypes of the business IS) to MU Clients.
End of Week 8: Simplified Project Closure Report to the MU
faculty.
264
Flexibility in Offshoring
All the GSD teams had appointed site-coordinators at both the client (USA) and vendor (India)
sites. The quasi-experimental design was a posttest control group design. The treatment for this
experiment was ‘requirements change’ where
non-control groups had major changes in requirements and control groups did not have changes
in requirements. The analysts in the non-control
group developed the first iteration artifacts based
on a set of high level and ambiguous requirements.
A second set of more detailed requirements were
then given to the MU client teams in the noncontrol group by the course faculty and the analysts
then created the second iteration incorporating
the changed set of requirements. The control group
teams also developed the projects in two iterations
but they were given detailed and clearly specified
requirements at the start of the project.
Measures
Requirements Analysis Success:Mahaney and
Lederer (2006) developed a comprehensive instrument for measuring IS project success based on
three dimensions- client satisfaction, perceived
quality of the project and success with the implementation process. Baroudi and Orlikowski (1988)
presented a short-form measure of User Information Satisfaction (UIS). Extending the dimensions
of project success to the success of a particular
phase of a project (requirements analysis), we
define requirements analysis success in terms of
client satisfaction with the requirements analysis
phase, perceived quality of the requirements
analysis deliverables and perceived success of
the requirements analysis process. This research
refers to the items developed by Mahaney and
Lederer (2006), and Baroudi and Orlikowski
(1988) to measure requirements analysis success.
Specifically, this study proposes to measure team
members’ perception of success in flexible GSD
requirements analysis projects.
Control: Items to measure control for this study
have been adapted from literature on formal modes
of control (behavior and outcome) in information
systems development projects (Kirsch, 1996;
Kirsch et al., 2002; Piccoli et al., 2004).
Computer-mediated communication: We
extend the task related communication items
developed by Espinosa (2002) to measure task
related computer-mediated communication for
this study.
Process Facilitation: Items to measure process
facilitation for this study have been adapted from
GSS literature on process facilitation (Miranda
et al., 1999; Anson et al., 2004). The items on
process facilitation were used twice. First, the
items were used to measure process facilitation
by the client site-coordinator and then they were
used to measure process facilitation by the vendor
site-coordinator.
Requirements Change: This variable was
introduced as a treatment in the experiment to
study whether there is an impact of changing
requirements on the proposed model variables.
Requirements change was thus treated as a dichotomous variable (1=changes in requirements
and 0=no changes).
ANALYSIS
A survey instrument was used to collect data
pertaining to the above measures at the end of the
quasi-experiments. Two separate versions were
developed for client and vendor team members.
Both versions included the same items for each
construct but were worded differently to conform
to the participant roles. The items were derived
from available literature as described earlier and
adapted for this study. Questions were randomized prior to instrument administration to counter
possible order effects in the responses. All items
were measured on a seven point Likert-type scale,
where one indicated strong disagreement and
seven indicated strong agreement with the item.
Items of the vendor version are given in Appen-
265
Flexibility in Offshoring
Table 5. Pilot study setting
Setting
Sample
Flexible GSD Requirements Analysis
All teams matched in terms of age, size, work experience, technology,
standard development methodology
Total Sample Size= 102
Analysts=52, Clients=50
(16 client-analyst GSD teams. Each having 3-4 analysts and
2-3 clients)
dix 1 for reference. Demographic data was also
collected through direct questions and analyzed.
Our unit of analysis was at the individual level.
We measured the effect of predictor variables on
the outcome variable as perceived by the individual
team members. A majority of research in organizational behavior has been conducted at individual
level of analysis (Schnake & Dumler, 2003). For
example, leadership style has been examined at
both individual and group levels in leadership
research, but primarily at the individual level
in organizational citizenship behavior research
(Schnake & Dumler, 2003). Similarly, task scope
(Griffin, 1982) and work context (Comstock &
Scott, 1977) have individual and group level effects. Our view is consistent with the assertion
by Choudhury and Sabherwal (2003) that in an
outsourced software project although the overall
context is the contract between the two firms (client and vendor), the focus is still on individual(s)
evaluating and influencing the actions of other
individual(s).
Standard psychometric techniques were employed to validate measures. For construct validity
Cronbach alpha was used and all constructs that
had Cronbach alpha closer to or greater than 0.7
confirmed construct validity (Cheung & Lee,
2001). Factor analyses was conducted to examine
convergent and discriminant validity. Principle
component method of extraction with varimax
rotation was used. The items that did not load well
were removed (see Appendix 1). For ensuring
convergent and discriminant validity we retained
the indicators that loaded onto their proposed
factors for the study.
266
Pilot Testing
As a first step, we carried out a pilot study to gain
an initial understanding and to pilot test the questionnaire items. Subsequently, we conducted the
main experiment to test our hypothesized model.
The pilot study was designed as a part of MU and
MDI course projects and its duration was 8 weeks
(see Table 5). The client teams in US enrolled for
an undergraduate course in IT Project Management and the analyst teams in India enrolled for a
graduate course in Systems Analysis and Design
in Oct 2006. Each client team was given a set of
requirements and the analyst team was required to
develop project artifacts and screen-based prototypes of a business information system requested
by the client team. This study was a simple posttest quasi-experiment design intended to pilot test
the data collection instrument and to comprehend
the flexible GSD environment.
The Experiment
Subsequently in our main experiment, we estimated the complete hypothesized structural model
using Arbuckle’s (2006) AMOS 7.0 program. A
common practice used in conducting SEM analyses with latent variables involves creating “item
parcels” based on sums or means of responses to
individual items and then using scores on these
parcels in the latent variable analysis (Russell,
1998). We created item parcels for success and
control based upon rationale grounds (Kline,
2005) for our structural model analysis. For
success (total 7 questionnaire items) we created
three item parcels- client satisfaction (2 items),
Flexibility in Offshoring
had Cronbach Alpha values above 0.70 indicating
construct reliability. The dependent variable perceived success showed significant correlations
with all the antecedent variables proposed in the
model, i.e. control, communication, process facilitation vendor, and process facilitation client.
There are significant correlations between the
antecedent variables themselves which needs to
be explored further.
perceived quality (three items) and success of
implementation process (2 items). For control,
measured by total 6 questionnaire items, we created two parcels- behavior control (3 items) and
outcome control (3 items).
RESULTS
Survey responses for 20 items were generated
from 181 respondents (client team members and
vendor analysts) for the experiment. Demographic
data was also collected through direct questions
and examined. Four items to measure process
facilitation and two additional items to measure
control were added to the same instrument used in
pilot study. The descriptive statistics is outlined in
Table 6 and Pearson’s product-moment correlation
matrix and reliability analysis is shown in Table 7.
Reliability of the measures of (i) success (ii)
control (iii) communication (iv) process facilitation (vendor) and (iv) process facilitation (client)
Multigroup Analysis
As stated earlier, our experiment was a control
group quasi-experiment design. One of the methods to incorporate group membership (treatment
vs. control) into the model involves separating the
data from members of the treatment and the control
groups and conducting a multiple-group analysis
(Russell, 1998; Sorbom, 1981). We referred to the
multigroup analysis specified by Byrne (2001, pp.
226-43) to test for invariant latent mean structures and Ho (2006, pp. 321-56) to estimate our
Table 6. Descriptive statistics
Total
Std. Dev.
Analyst
Client
Sample Size (Control + Non-control)
181
0.48
115
66
Age (mean value in years)
22.8
1.9
23.5
21.6
Total Work Experience (mean value in years)
2
2
1.2
3.3
IT Work Experience (mean value in years)
0.7
1.3
0.72
0.76
Table 7. Correlation matrix and reliability
Items
1
2
3
4
1. CONTROL
6
(0.837)***
2. COMMUNICATION
3
.178*
(0.828)***
3. PROCESS FACILITATION
€€€€€(CLIENT)
2
.168*
.464**
(0.755)***
4. PROCESS FACILITATION
(VENDOR)
2
.584**
.207**
.178*
(0.789)***
5. SUCCESS
7
.612**
.337**
.584**
.504**
5
(0.790)***
*** Reliability Analysis (Cronbach alpha) on the diagonal
** Correlation is significant at the 0.01 level (2-tailed)
* Correlation is significant at the 0.05 level (2-tailed)
267
Flexibility in Offshoring
hypothesized structural equation model for the
control group and non-control groups.
Byrne (2001) suggests that the usual test for
multigroup comparisons typically focuses on
the extent to which differences are statistically
significant among observed means (calculable
from raw data) representing the various groups.
In contrast, means of latent variables, as in our
case, are unobservable. Thus our focus is on testing
for differences in the latent means for hypotheses
6a-e, i.e. team members having changes in requirements will have higher perceptions of the latent
variables than team members having no changes
in requirements. Specifically, we test for differences in the latent means of control, computermediated communication, process facilitation by
vendor site-coordinator, process facilitation by
client site-coordinator and requirements analysis
success across control and non-control groups.
In testing for differences in latent means our
baseline measurement model for each group is
shown in Figure 4. The variances of the five latent
factors are freely estimated in each group. Means
of the error terms are not estimated and remain
constrained to zero. Except for those fixed to 1.00,
all factor loadings are constrained equal across
groups. All intercepts for the observed measures
are constrained equal across groups. The five
factor means are freely estimated for non-control
group and constrained equal to zero for the control
group. The latent mean estimates indicate that
team members having changes in requirements
have significantly higher perceptions of control
(H6b), computer-mediated communication (H6c),
process facilitation by vendor site-coordinator
(H6e). But there appears to be little difference in
perception for requirements analysis success (H6a)
and process facilitation by client site-coordinator
(H6d). These results are summarized in Table 8.
For our multigroup analysis there were two
data sets (control and non-control groups), each
having 11 measurement variables. The two covariance matrices generated from the two data sets
contained 132 sample moments. In multigroup
Figure 4. Structured means model for control and non-control groups
268
Flexibility in Offshoring
Table 8. Results summary (latent mean estimates)
Hypothesis
Latent Variable
Estimate
Critical Ratio
(CR>+1.96)
Sig.
(p<.05)
Hypothesis
Supported?
H6a
Requirements analysis success
.232
1.760
.078
Not Supported
H6b
Control
.364
2.337*
.019*
Supported
H6c
Computer-mediated communication
.510
2.227*
.026*
Supported
H6d
Process facilitation (vendor site-coordinator)
.334
2.186*
.029*
Supported
H6e
Process facilitation (client site-coordinator)
-.036
-.217
.828
Not Supported
confirmatory factor analysis (measurement)
model, the critical ratio test for control and noncontrol group differences among regression
weights yielded no significant difference for the
16 regression weights. Therefore, we constrained
16 regression weights to equality in the multigroup
path analysis model. The computation of degrees
of freedom and chi-square goodness-of-fit statistics for the group-invariant and group-variant
model are outlined in Appendix 2. The baseline
comparison fit indices of NFI, RFI, IFI, TLI and
CFI for both the models are close to or are above
0.90 (see Table 9). These values show the improvement in fit of both models relative to the null
model. Root mean square error of approximation
(RMSEA) fit index values for group-invariant
(0.074) and group-variant path models (0.077)
indicated adequate fits for both the models. RMSEA values ranging from 0.05 to 0.08 are deemed
acceptable (Ho, 2006, pp. 349).
The nested model comparisons statistics indicate that chi-square difference value for the two
models is 8.847. This value is not significant (p<
.05) with 8 degrees of freedom (see Table 10).
Thus the two models do not differ significantly
in their goodness-of-fit. The AIC measure for
group-invariant model (261.380) is lower than
that for the group-variant model (268.532), indicating that group-invariant model is more parsimonious and better fitting than the group-variant
model. Hence we refer to the estimates of the
group-invariant model for our analysis.
Table 9. Baseline comparisons
Model
NFI
Delta1
RFI
rho1
IFI
Delta2
TLI
rho2
CFI
Group Invariant
.859
.811
.925
.897
.923
Group Variant
.867
.802
.927
.887
.924
Saturated model
1.000
Independence model
.000
1.000
.000
1.000
.000
.000
.000
Table 10. Nested model comparisons (assuming model Group Variant to be correct)
Model
DF
CMIN
P
NFI
Delta-1
IFI
Delta-2
RFI
rho-1
TLI
rho2
Group Invariant
8
8.847
.355
.008
.008
-.009
-.010
269
Flexibility in Offshoring
Figure 5. Non-control group (req. change) structural path model *Standardized estimates and significant
relationships (p<.05, CR>+.96)
Figure 5 and 6 present the path models for the
control and non-control groups together with
standardized regression weights (beta coefficients)
associated with the hypothesized paths. First, we
analyzed critical ratios for differences test for
experimental treatment impact on the path coefficients. Table 11 shows results of the individual
hypotheses for both control and non-control
Figure 6. Control group (no req. change) structural path model *Standardized estimates and significant
relationships (p<.05, CR>+.96)
270
Flexibility in Offshoring
groups. Appendix 2 presents the control and noncontrol group estimates in detail. H1 was supported which posits that formal modes of control
are positively related to perceived success during
requirements analysis in a flexible GSD environment (control group β= .55 and non-control group
β= .43). Surprisingly, as proposed in hypothesis
2, frequent task-related communication was not
found to be significantly related to perceived
success (control group β= -.13, non-control group
β= -.13). Though process facilitation by vendor
site-coordinator did not have a positive impact
on success (H3a: control group β= .08, noncontrol group β= .05), the positive impact of
process facilitation by client site-coordinator was
found to be significant (H3b: control group β=
.60, non-control group β= .71).
The proposed positive relationship between
process facilitation by vendor site-coordinator
and control was found to be significant (H4a:
control group β= .70, non-control group β= .60)
but was insignificant for process facilitation by
client site-coordinator and control (H4b: control
group β= .09, non-control group β= .13). Again,
Table 11. Experiment 2 Results Summary
Hypothesis
Path
Hypothesized
Relationship
H1
control → req analysis success
Control Group:
Non-control Group:
+
H2
computer-mediated communication → req analysis success
Control Group:
Non-control Group:
+
process facilitation(vendor) →
req analysis success
Control Group:
Non-control Group:
+
process facilitation (client) →
req analysis success
Control Group:
Non-control Group:
+
process facilitation (vendor) →
control
Control Group:
Non-control Group:
+
process facilitation (client) →
control
Control Group:
Non-control Group:
+
process facilitation (vendor) →
computer-mediated communication
Control Group:
Non-control Group:
+
process facilitation (client) →
computer-mediated communication
Control Group:
Non-control Group:
+
H3a
H3b
H4a
H4b
H5a
H5b
Path Coefficient
(Standardized
beta)
Critical Ratio
(CR>+1.96)
Sig.
(p<.05)
.547
.434
4.079*
4.079*
.000*
.000*
Hypothesis
Supported?
Supported
Not Supported
-.131
-.131
-1.746
-1.746
.081
.081
Not Supported
-.079
-.054
-.607
-.607
.544
.544
Supported
.599
.710
6.754*
6.754*
.000*
.000*
Supported
.702
.603
6.509*
6.509*
.000*
.000*
Not Supported
.088
.132
1.274
1.274
.203
.203
Not Supported
.177
.120
1.768
1.768
.077
.077
Supported
.399
.471
4.631*
4.631*
.000*
.000*
271
Flexibility in Offshoring
process facilitation by vendor site-coordinator did
not have a positive impact on frequent task-related communication (H5a: control group β= .18,
non-control group β= .12) but the positive impact
of process facilitation by client site-coordinator
was found to be significant (H5b: control group
β= .40, non-control group β= .47). Based on the
results displayed in Table 11, it can be concluded
that for both the control and non-control groups,
process facilitation by vendor site-coordinator is
related indirectly to perception of success, being
mediated by formal modes of control. Furthermore, the greater the process facilitation by client
site-coordinator, the greater is task-related communication, and perception of success.
DISCUSSION
This study was driven by the need to better
understand which antecedent factors impact
requirements analysis success in flexible GSD
projects and how. We proposed a conceptual model
grounded in control theory to find preliminary
answers to our research questions- what are the
factors that are of importance in the requirements
analysis phase of flexible GSD projects? How
do they relate to requirements analysis success?
We find that control (both behaviour and
outcome) positively impacts success during
requirements analysis. Prior studies have primarily focused on behavior and outcome controls
in internal software projects. Few studies, such
as Choudhury and Sabherwal (2003), reported
that outcome control in outsourced IS projects
resembled that of internal IS development projects. They also state that behavior control entailed
monitoring behavior that was ‘explicitly’ as well
as ‘not explicitly’ prescribed and requires further
investigation in other contexts. Consistent with
their view, in a flexible GSD context our study
reports that having some amount of structure in
the form of formal modes of controls positively
impacts success.
272
Applying the information processing perspective of communication (Putnam & Cheney, 1985)
to a GSD scenario, we suggest that there exits work
flow interdependence between the remote client
and the vendor teams for accurately capturing
and documenting requirements. This is likely to
lead to frequent task related computer-mediated
communication for successful completion of requirements analysis. Surprisingly, the results show
that increased task-related communication did not
positively impact success. A likely explanation
for such a finding can be that informal means
of communications across the teams could have
probably sufficed the need of task-related online
information exchange.
Consistent with prior literature and practice,
we find that increased process facilitation by client
site-coordinators in a flexible GSD environment
increases task-related communication and also
leads to greater success of requirements analysis
projects. However, increased process facilitation
by vendor site-coordinators did not lead to greater
task-related communication. Rather, increased
process facilitation by vendor site-coordinators
lead to increased control which further impacted
requirements analysis success. Consequently,
the effect of process facilitation by vendor sitecoordinators on requirements analysis success
is found to be mediated by control whereas process facilitation by client site-coordinators has a
direct impact on requirements analysis success.
Further, increased process facilitation by client
site-coordinators did not lead to increased control.
An inherent assumption behind flexible (agile) philosophy is changes in requirements are
good since requirements evolve over time and
are best clarified through multiple information
gathering iterations. In our experiment we introduced ‘changes in requirements’ as a treatment
for the non-control group. Our findings suggest
that team members who experienced changing
requirements also experienced greater control,
greater computer-mediated communication, and
greater process facilitation by vendor site-coor-
Flexibility in Offshoring
dinator. However, these team members did not
experience greater process facilitation by client
site-coordinator. Since changes in requirements
are client initiated, it is likely that the increased
role of the client site-coordinator was perceived
to be less significant. The latent mean scores also
indicated lower difference in perception of requirements analysis success. A probable explanation in
this regard could be that changes in requirements
might not have a direct impact on the success
of requirements analysis phase. However, it is
likely that the changes have greater impact on
subsequent phases of software development and
hence on the complete project. Also, there is likely
to be an indirect impact via mediating factors like
control, communication and process facilitation.
This needs to be further explored by including
requirements change as a model variable.
LIMITATIONS AND FUTURE
PROPOSITIONS
Use of Experiments
Literature in the area of virtual teams has mainly
followed three research methodologies– case
studies, industry survey, and experiments. Experimental methods make possible the careful
observation and precise manipulation of independent variables, allowing for greater certainty with
respect to cause and effect, while holding constant
other variables that would normally be associated
with it in field settings (Damian, Eberlein, Shaw
& Gaines, 2000). They also encourage the investigator to try out novel conditions and strategies
in a safe and exploratory environment before
implementing them in the real world (McGrath,
1984). The industry is yet to adopt off-shoring of
the requirements analysis phase. This precludes
the use of case study or industry survey for this
research. Hence, we used experiments where we
can try out this new method. While this limits
the generalizablity of our findings, our approach
provides a foundation from which to build a future
empirical industry assessment as the time is ripe.
Use of Students as Surrogates
There are criticisms for the use of students in
academic experiments as surrogates. However,
MBA students have been used as surrogate users
in experiments conducted by Hazari (2005), and
Briggs, Balthazard and Dennis (1996). Even in
requirement negotiation phase, students with work
experience were taken as users for developing a
small system (Damian et al., 2000). Remus (1986)
argued that graduate students could be used as
surrogates for managers in experiments on business decision making. Students often represent a
typical working professional and organizational
member due to the variety of backgrounds and
goals (Dipboye & Flanagan, 1979). Studies in
industrial organization psychology and organization behavior suggest that results obtained from
students are similar to those from managers (see
for example, Locke, 1986). Despite the fact that
clients and analysts in our experiments had 2-4
years of work experience, limitations of using
students as surrogates are still applicable in our
study.
The potential lack of realism in laboratory
experiments using student surrogates can be addressed through multiple methods (McGrath,
1984) so that strengths of some compensate weaknesses of others (Jarvenpaa, 1988). Simulated
laboratory negotiations could be complemented
by field studies or validations, if the lack of realism is an issue. As a next step in our research, we
plan to complement the findings of our laboratory
experiments with field validations.
Project Complexity
We acknowledge that requirements analysis is
an intensive phase and hence is not possible to
completely replicate it in student experiments.
However, our objective was to study the research
273
Flexibility in Offshoring
questions on comparable relatively well-defined
small projects in which complexity of requirements analysis is not high. Though the experiments
were carefully designed, the projects done were
limited in scope and size compared to large scale
industrial projects. However, no formal measures
of complexity were used in the study so that we
could compare the projects used in the experiments with real world industrial projects. Further
research is needed to assess the impact of these
findings on large scale industrial projects with
complex requirements.
Communication Technology
The communication technology used in our
experiment was mainly freely available internet
based information and communication technology
like chat, email, egroups and web conferencing.
Moreover, in industry other technologies like
teleconferencing and video conferencing are also
used. Further research is required to study the
impact of different communication technologies
during requirements analysis.
IMPLICATIONS FOR PRACTICE
The use of GSD in organizations is becoming
increasingly commonplace as corporations seek to
take advantage of geographically dispersed talent
for multi-location operations. This study yields
several interesting implications for practice that
can assist organizations in managing their offshore
GSD projects more effectively.
Despite evidence of successful agile software
development, its application in GSD warrants
some planning. Although there exists some
preliminary research on applications of Extreme
Programming (XP) in GSD (Nago-The, Hoang,
Nguyen & Mai, 2005), it is a common perception
that agile methods are not applicable in GSD. There
is a need for increased understanding of the characteristics of agile methods and how they can be
274
applied to mitigate negative influence of distance
(Holmstrom et al., 2006). There are emerging
examples of organizations like Sapient that have
demonstrated the ability to run complex, largescale distributed projects and have leveraged the
benefits that agile processes offer (Barnett, 2006).
Prior studies indicate that creation of flexible
environments by incorporating the principles of
agile development can help organizations mitigate
communication and control related risks inherent
in GSD. Our study also finds that client as well as
vendor organizations need to pay adequate attention to incorporating discipline even in flexible
GSD projects, specifically during requirements
analysis. Our paper makes an important contribution of delineating the differences of the traditional,
agile and flexible approaches. In an attempt to
portray reality, we explain the hybrid ‘flexible
GSD’ approach which most organizations are
adopting. This hybrid approach incorporates the
discipline offered by the traditional approaches
and the flexibility offered by the agile approaches.
Companies that are engaged in offshore GSD
have developed strong processes around their
global delivery model. However, whether the
same processes and project monitoring discipline (control) will lead to success of projects
conducted in ‘flexible and pure offshore mode’
in virtual team setting during the early stages of
system development work has not been explored.
Hence, we examined the antecedents of success
during requirements analysis in such an environment. Our research indicates that client sitecoordinators play an important role in increasing
task related computer-mediated communication
and requirements analysis success. Further vendor
site-coordinators play a critical role in increasing
control which indirectly impacts requirements
analysis success.
Agile philosophy advocates that requirements
change is a business world reality. Our study
provides empirical support for the impact of
requirements change even in the early phases of
software development like requirements analysis.
Flexibility in Offshoring
The results indicate that changes in requirements
resulted in greater control, greater computermediated communication, and greater process
facilitation by vendor site-coordinator. These
factors are likely to further impact requirements
analysis success in a GSD setting.
Another important message that this study conveys is that site-coordinators need not be project
managers or team leads. Any experienced team
member who gets nominated by consensus can
facilitate project processes and impact success of
distributed projects. Further research is needed to
confirm our exploratory findings.
IMPLICATIONS FOR RESEARCH
Contributing to existing research, this study
empirically demonstrates the direct and indirect
relationships between antecedent factors (control,
computer-mediated communication, process facilitation by vendor site-coordinator and process
facilitation by client site-coordinator) and success
in flexible offshore GSD projects. It may not always be feasible to make experimental and control
groups adhere to experimental requirements in a
classroom setting, hence a flexible approach is
needed in experimental design.
The findings highlight the key role that formal modes of control play even in a flexible
GSD environment. The scope of our research
was limited to examining only formal modes of
control whereas agile philosophy can probably
also support informal modes of control, such as,
self and clan control. Future research is needed
to understand the conditions under which self and
clan control is used in flexible GSD projects, and
hence their impact on success.
By implementing dedicated client as well as
vendor site-coordinators at each distributed site
we also empirically determined the direct and
indirect (via control) effects of process facilitation
on success. Contrary to our belief we found that
there was no significant impact of task-related
computer-mediated communication on success.
Informal communication and socialization is one
probable explanation that could have met the need
for such information exchange. Also, given the
fact that students were used as surrogates, informal
online interaction is more likely to occur rather
than planned formal interaction. This needs further investigation by researchers in field settings.
Although several benefits have been speculated, our research is one of the first studies to
empirically validate the impact of changes in
requirements, which resulted in greater perception
of control, computer-mediated communication,
and process facilitation by vendor site-coordinator.
However, requirements change calls for more
detailed investigation as we studied this using a
dichotomous variable in our multigroup analysis.
Future research can involve measuring the level
of requirements change and further adding it as
a variable in our research model. Moreover, to
further our knowledge about success of flexible
offshore GSD projects and their sub-phases, additional studies are needed that move beyond presently conceptualized variables, such as, impact of
motivation, cohesion and trust between offshore
GSD team members, and emotional intelligence
of individual team members.
REFERENCES
Agerfalk, P. J., & Fitzgerald, B. (2006). Flexible
and distributed software processes: Old petunias
in new bowls? Communications of the ACM,
49(10), 27–35.
Anson, R., Bostrom, R., & Wynne, B. (1995). An
experiment assessing group support system and
facilitator effects on meeting outcomes. Management Science, 41(2), 189–208. doi:10.1287/
mnsc.41.2.189
Arbuckle, J. L. (2006). AMOS 7.0 computer software. Chicago, IL: SmallWaters.
275
Flexibility in Offshoring
Augustine, S., Payne, B., Sencindiver, F., & Woodcock, S. (2005). Agile project management: Steering from the edges. Communications of the ACM,
48(12), 85–89. doi:10.1145/1101779.1101781
Barnett, L. (2006). Scaling agile worldwide: Sapient’s distributed agile methodology. Agile Journal.
Retrieved October 23, 2007, from: http://www.
agilejournal.com
Baroudi, J. J., & Orlikowski, W. J. (1988). A shortform measure of user information satisfaction: A
psychometric evaluation and notes on use. Journal
of Management Information Systems, 4(4), 44–59.
Battin, R., Crocker, R., Kreidler, J., & Subramanian, K. (2003). Leveraging resources in global
software development. IEEE Software, (March/
April): 70–77.
Beath, C. M., & Orlikowski, W. J. (1994). The
contradictory structure of systems development
methodologies: Destructuring the IS user relationship in information engineering. Information
Systems Research, 5(4), 350–377. doi:10.1287/
isre.5.4.350
Bhat, J., Gupta, M., & Santhosh, N. (2006). Overcoming requirements engineering challenges: Lessons from offshore outsourcing. IEEE Software,
(Sept): 38–43. doi:10.1109/MS.2006.137
Briggs, R. O., Balthazard, P. A., & Dennis, A. R.
(1996). Graduate business students as surrogates
in the evaluation of technology. Journal of End
User Computing, 8(4), 11–17.
Briggs, R. O., & Gruenbacher, P. (2002). Easy
win-win: Managing complexity in requirements
negotiation with GSS. Proceedings: 35th Annual
Hawaii International Conference on Systems Science. Hawaii.
Byrne, B. M. (2001). Structural equation modeling with AMOS. London, UK: Lawrence Erlbaum
Associates.
276
Campbell, D. T., & Stanley, J. C. (1966). Experimental and quasi-experimental designs for
research. Chicago, IL: Rand McNally.
Carmel, E. (1999). Global software teams: Collaborating across borders and time zones. Upper
Saddle River, NJ: Prentice Hall.
Carmel, E., & Tjia, P. (2005). Offshoring Information Technology: Sourcing and outsourcing to a global workforce. Cambridge, UK:
Cambridge University Press. doi:10.1017/
CBO9780511541193
Cheung, C., & Lee, M. (2001). Trust in Internet
shopping: Instrument development and validation
through classical and modern approaches. Journal
of Global Information Management, 9(3), 23–35.
doi:10.4018/jgim.2001070103
Choudhury, V., & Sabherwal, R. (2003). Portfolios
of control in outsourced software development
projects. Information Systems Research, 14(3),
291–314. doi:10.1287/isre.14.3.291.16563
Cockburn, A., & Highsmith, J. (2001). Agile software development: The business of innovation.
IEEE Computer, 120-122.
Comstock, D. E., & Scott, W. R. (1977). Technology and the structure of subunits: Distinguishing individual and workgroup effects.
Administrative Science Quarterly, 22, 177–202.
doi:10.2307/2391955
Crampton, C. (2001). The mutual knowledge
problem and its consequences for dispersed collaboration. Organization Science, 12(3), 346–371.
doi:10.1287/orsc.12.3.346.10098
Crisp, C. B. (2003). Control enactment in global
virtual teams. UMI published doctoral dissertation. University of Texas, Austin.
Crowston, K., & Kammerer, E. E. (1998). Coordination and collective mind in software requirements development. IBM Systems Journal, 37(2),
227–241. doi:10.1147/sj.372.0227
Flexibility in Offshoring
Damian, D. E., Eberlein, A., Shaw, M. L. G., &
Gaines, B. R. (2000). Using different communication media in requirements negotiation. IEEE
Software, (May/June): 28–36.
Espinosa, A. J. (2002). Shared mental models and
coordination in large-scale, distributed software
development. UMI published doctoral dissertation.
Carnegie Mellon University.
Damian, D. E., & Zowghi, D. (2002). An insight
into interplay between culture, conflict and distance in globally distributed requirement negotiations. Proceedings: 36th Hawaii International
Conference on System Sciences. Hawaii.
Evaristo, R., Watson-Manheim, M. B., & Audy,
J. (2005). E-collaboration in distributed requirements determination. International Journal
of e-Collaboration, 1(2), 40–55. doi:10.4018/
jec.2005040103
DeSanctis, G., & Gallupe, R. B. (1987). A foundation for the study of group decision support
systems. Management Science, 35(5), 589–609.
doi:10.1287/mnsc.33.5.589
Favela, J., & Pena-Mora, F. (2001). An experience in collaborative software engineering
education. IEEE Software, (March/April): 47–53.
doi:10.1109/52.914742
Dibbern, J., Goles, T., Hirschheim, R., & Jayatilaka, B. (2004). Information Systems outsourcing: A survey and analysis of the literature. The
Data Base for Advances in Information Systems,
35(4), 6–102.
Fruhling, A., & de Vreede, G.-J. (2006). Field
experiences with eXtreme programming: Developing an emergency response system. Journal of
Management Information Systems, 22(4), 39–68.
doi:10.2753/MIS0742-1222220403
Dipboye, R. L., & Flanagan, M. F. (1979). Research
setting in industrial and organization psychology:
Are findings in the field more generalizable than
in laboratory. The American Psychologist, 32,
141–150. doi:10.1037/0003-066X.34.2.141
Griffin, R. W. (1982). Task design: An integrated
approach. Glenview, IL: Scott Foresman.
Edwards, K., & Sridhar, V. (2005). Analysis of
software requirements engineering exercises in a
global virtual team setup. Journal of Global Information Management, 13(2), 21–41. doi:10.4018/
jgim.2005040102
Edwards, K., & Sridhar, V. (2006). Collaborative
software requirements engineering exercises in a
distributed virtual team environment. In Hunter,
G., & Tan, F. (Eds.), Advanced topics of global
information management (Vol. 5, pp. 178–197).
Hershey, PA: Idea Publishing.
Eisenhardt, K. M. (1985). Control: Organizational
and economic approaches. Management Science,
31(2), 134–149. doi:10.1287/mnsc.31.2.134
Griffith, T. L., Fuller, M. A., & Northcraft, G.
B. (1998). Facilitator influence in group support systems: Intended and unintended effects.
Information Systems Research, 9(1), 21–35.
doi:10.1287/isre.9.1.20
Hazari, S. I. (2005). Perceptions of end-users on
the requirements in personal firewall software: An
exploratory study. Journal of Organizational and
End User Computing, 17(3), 47–65. doi:10.4018/
joeuc.2005070103
Ho, R. (2006). Handbook of univariate and
multivariate data analysis and interpretation
with SPSS. Boca Raton, FL: Chapman & Hall.
doi:10.1201/9781420011111
Hoffer, J., George, J., & Valacich, J. (1999). Modern systems analysis and design. Addison Wesley.
277
Flexibility in Offshoring
Holmstrom, H., & Fitzgerald, B., Agerfalk, Par,
J., & Conchuir, E. O. (2006). Agile practices
reduce distance in global software development.
Information Systems Management, (Summer):
7–18. doi:10.1201/1078.10580530/46108.23.3.
20060601/93703.2
Hulnick, G. (2000). Doing business virtually.
Communication World, 17(3), 33–36.
Jarvenpaa, S. (1988). The importance of laboratory experimentation in IS research. Communications of the ACM, 31(12), 1502–1505.
doi:10.1145/53580.374700
Kirsch, L. J. (1997). Portfolios of control modes
and IS project management. Information Systems
Research, 8(3), 215–239. doi:10.1287/isre.8.3.215
Kirsch, L. J., & Cummings, L. L. (1996). Contextual influences on self-control of IS professionals
engaged in systems development. Accounting.
Management Information Technology, 6(3),
191–219.
Kirsch, L. J., Sambamurthy, V., Dong-Gil, K.,
& Purvis, R. L. (2002). Controlling Information
Systems development projects: The view from
the client. Management Science, 48(4), 484–498.
doi:10.1287/mnsc.48.4.484.204
Kline, R. B. (2005). Principles and practice of
structural equation modeling (2nd ed.). New
York, NY: Guilford.
Lee, G., Delone, W., & Espinosa, J. A. (2006).
Ambidextrous coping strategies in globally distributed software development projects.
Communications of the ACM, 49(10), 35–40.
doi:10.1145/1164394.1164417
Lee, O.-K., Banerjee, P., Lim, K. H., Hillegersberg,
J. V., & Wei, K. K. (2006). Aligning IT components
to achieve agility in globally distributed system
development. Communications of the ACM,
49(10), 49–54. doi:10.1145/1164394.1164419
278
Lindstrom, L., & Jeffries, R. (2004). Extreme
programming and agile software development
methodologies. Information Systems Management, (Summer): 41–52. doi:10.1201/1078/444
32.21.3.20040601/82476.7
Locke, E. A. (1986). Generalizing from laboratory
to field setting: Research finding from industrial
organization, organization behavior, and human
resource management. Lexington, MA: Lexington
Books.
Mahaney, R. C., & Lederer, A. L. (2006). The effect
of intrinsic and extrinsic rewards for developers
on Information Systems project success. Project
Management Journal, 37(4), 42–54.
McGrath, J. (1984). Groups: Interaction and
performance. New Jersey: Prentice-Hall.
Meso, P., & Jain, R. (2006). Agile software development: Adaptive systems principles and best
practices. Information Systems Management,
(Summer): 19–30. doi:10.1201/1078.10580530
/46108.23.3.20060601/93704.3
Nago-T., A., Hoang, K., Nguyen, T., & Mai, N.
(2005). Extreme programming in distributed
software development: A case study. Proceedings
of the International Workshop on Distributed
Software Development. Paris.
Nath, D., Sridhar, V., Adya, M., & Malik, A.
(2006). The effect of user project monitoring on
the performance of virtual teams in the requirements analysis phase of off-shored software projects. Proceedings of the INFORMS Conference
on Information Systems and Technology (CIST).
Pittsburgh, PA, USA.
National Association of Software and Service
Companies (NASSCOM). (2007). Strategic review: The IT industry in India. Retrieved March
19, 2007, from http://www.nasscom.org/
Flexibility in Offshoring
PMBOK®. (2000). A guide to project management
body of knowledge (PMBOK). Newtown Square,
PA: Project Management Institute.
Powell, A., Piccoli, G., & Ives, B. (2004). Virtual
teams: A review of current literature and directions
for future research. The Data Base for Advances
in Information Systems, 35(1), 6–36.
Putnam, L. L., & Cheney, G. (1985). Organizational communication: Historical development
and future directions. In Benson, T. W. (Ed.),
Speech communication in the 20th century (pp.
130–156). Carbondale, IL: Southern Illinois
University Press.
Ramesh, B., Cao, L., Mohan, K., & Xu, P. (2006).
Can distributed software development be agile?
Communications of the ACM, 49(10), 41–46.
doi:10.1145/1164394.1164418
Remus, W. E. (1986). An empirical test of the use
of graduate students as surrogates for managers in
experiments on business decision making. Journal
of Business Research, 14(1), 1986, 19–25.
Rice, R. E. (1987). Computer-mediated communication and organizational innovation.
The Journal of Communication, 37(4), 65–94.
doi:10.1111/j.1460-2466.1987.tb01009.x
Russell, D. W., Kahn, J. H., & Altmaier, E. M.
(1998). Analyzing data from experimental studies:
A latent variable structural equation modeling approach. Journal of Counseling Psychology, 45(1),
18–29. doi:10.1037/0022-0167.45.1.18
Schnake, M. E., & Dumler, M. P. (2003). Levels
of measurement and analysis issues in organizational citizenship behavior research. Journal of
Occupational and Organizational Psychology,
76, 283–301. doi:10.1348/096317903769647184
Shaila, M., & Bostrom, R. P. (1999). Meeting
facilitation: Process versus content interventions.
Journal of Management Information Systems,
15(4), 89–114.
Sorbom, D. (1981). Systems under indirect observation: Causality, structure and prediction. In
Joreskog, K. G., & Wald, H. (Eds.), Structural
equation models with structured means. Amsterdam, The Netherlands: North-Holland.
Tan, C., & Teo, H. (2007). Training future software
developers to acquire agile development skills.
Communications of the ACM, 50(12), 97–98.
doi:10.1145/1323688.1323715
Yadav, V., & Gupta, R. (2008). A paradigmatic and
methodological review of research in outsourcing. Information Resources Management Journal,
21(1), 27–43. doi:10.4018/irmj.2008010102
Yadav, V., Nath, D., Adya, M., & Sridhar, V. (2007).
Investigating an agile-rigid approach in globally
distributed requirements analysis. Proceedings:
11th Pacific-Asia Conference on Information Systems (PACIS). Auckland, New Zealand.
279
Flexibility in Offshoring
APPENDIX A. QUESTIONNAIRE ITEMS
Two separate versions of the same questionnaire were created having identifiers, such as ‘MU’ for the
client questionnaire and ‘MDI’ for the vendor questionnaire. The items were the same only the identifiers
(MU/MDI) were different for the client and vendor team members. The items presented in the appendix
are from the MDI vendor questionnaire.
Response Scale: “Please answer each of the following questions related to the globally distributed development project by encircling the appropriate response”
Seven point scale, with 1= “Strongly Disagree”, 4= “Neutral” and 7= “Strongly Agree”
Process Facilitation
Process Facilitation- Vendor
1.
2.
3.
The MDI site-coordinator helped coordinate the workflow between MDI and MU team members.
My team members could have prepared as good project deliverables even if the MDI site-coordinator
had not been present.*
The MDI site-coordinator constructively responded to our team’s needs for assistance.
Process Facilitation- Client
1.
2.
3.
The MU site-coordinator helped coordinate the workflow between MDI and MU team members.
My team members could have prepared as good project deliverables even if the MU site-coordinator
had not been present.*
The MU site-coordinator constructively responded to our team’s needs for assistance.
Computer-Mediated Communication
1.
2.
3.
We had frequent online-meetings with MU team members for coordination and planning.
We had frequent online-meetings with MU team members for requirements analysis.
We had frequent online formal review meetings with MU team members. 4. I participated actively
in the on-line discussions with MU team members.*
Control
Behavior Control**
1.
2.
280
The MU team members insisted on complete and on-time submission of project status reports.
The MU team members insisted on complete and on-time submission of intermediate project
deliverables.
Flexibility in Offshoring
3.
The MU team members insisted on complete and on-time submission of final project deliverables.
Outcome Control**
4.
5.
6.
The MU team members regularly monitored the project progress.
A detailed project management plan was developed between our MDI team members and MU team
members.
The communication process between our MDI team members and MU team members was well
defined.
Success
Perceived Quality**
1.
2.
3.
4.
5.
MDI team members have been able to come out with the best possible deliverables for capturing
MU client requirements.
The final project deliverables were readily accepted by the MU team members.
The project deliverables clearly specified MU client requirements. Client Satisfaction**
The MU team members clearly understood the project deliverables submitted by MDI team members.
The commitment of the MU team members in favor of the project directed goals and tasks were
positive.
Success of Implementation Process**
6.
7.
The virtual team project was completed within its original schedule.
MU client was satisfied with the process by which this project was completed.
* Item dropped from analysis
** Entered as item parcels in the model for analysis
APPENDIX B. RESULTS OF MULTIGROUP ANALYSIS
Computation of degrees of freedom and chi-square goodness-of-fit statistics for the models
Model (Group-Invariant)
Model (Group-Variant)
Computation of degrees of freedom (Group-Invariant)
€€Number of distinct sample moments :132
€€Number of distinct parameters to be estimated: 50
€€Degrees of freedom (132 - 50) : 82
Computation of degrees of freedom (Group-Variant)
€€Number of distinct sample moments :132
€€Number of distinct parameters to be estimated: 58
€€Degrees of freedom (132 - 58) : 74
Result (Group-Invariant)
Minimum was achieved
Chi-square = 161.380
Degrees of freedom = 82
Probability level = .000
Result (Group-Variant)
Minimum was achieved
Chi-square = 152.532
Degrees of freedom = 74
Probability level = .000
281
Flexibility in Offshoring
Estimates (group invariant)
CONTROL GROUP (No Req. Change)
NON-CONTROL GROUP (Req. Change)
Scalar Estimates (Maximum Likelihood Estimates)
Regression Weights
Scalar Estimates (Maximum Likelihood Estimates)
Regression Weights
Estimate
S.E.
C.R.
P
Label
Estimate
S.E.
C.R.
P
Label
Com
<---
Fac
Vendor
.238
.135
1.768
.077
nc1
Com
<---
Fac
Vendor
.238
.135
1.768
.077
nc1
Contrl
<---
Fac
Client
.102
.080
1.274
.203
nc6
Contrl
<---
Fac
Client
.102
.080
1.274
.203
nc6
Contrl
<---
Fac
Vendor
.653
.100
6.509
***
nc3
Contrl
<---
Fac
Vendor
.653
.100
6.509
***
nc3
Com
<---
Fac
Client
.672
.145
4.631
***
nc4
Com
<---
Fac
Client
.672
.145
4.631
***
nc4
Success
<---
Contrl
.476
.117
4.079
***
nc7
Success
<---
Contrl
.476
.117
4.079
***
nc7
Success
<---
Com
-.078
.045
-1.746
.081
nc8
Success
<---
Com
-.078
.045
-1.746
.081
nc8
Success
<---
Fac
Client
.605
.090
6.754
***
nc5
Success
<---
Fac
Client
.605
.090
6.754
***
nc5
Success
<---
Fac
Vendor
-.064
.105
-.607
.544
nc2
Success
<---
Fac
Vendor
-.064
.105
-.607
.544
nc2
fac2
<---
Fac
Vendor
1.002
.124
8.075
***
ncv2
fac2
<---
Fac
Vendor
1.002
.124
8.075
***
ncv2
outcnt
<---
Contrl
.717
.085
8.426
***
ncv3
outcnt
<---
Contrl
.717
.085
8.426
***
ncv3
fac1
<---
Fac
Client
.984
.110
8.924
***
ncv1
fac1
<---
Fac
Client
.984
.110
8.924
***
ncv1
fac5
<---
Fac
Client
1.000
fac5
<---
Fac
Client
1.000
fac6
<---
Fac
Vendor
1.000
fac6
<---
Fac
Vendor
1.000
behcnt
<---
Contrl
1.000
behcnt
<---
Contrl
1.000
com4
<---
Com
.992
com4
<---
Com
.992
.115
8.652
***
ncv6
com3
<---
Com
1.000
com3
<---
Com
1.000
uis
<---
Success
.684
.067
10.225
***
ncv4
uis
<---
Success
.684
.067
10.225
***
ncv4
proc
<---
Success
.967
.058
16.565
***
ncv5
proc
<---
Success
.967
.058
16.565
***
ncv5
qlty
<--
Success
1.000
qlty
<---
Success
1.000
.115
8.652
***
Standardized Regression Weights
ncv6
Standardized Regression Weights
Estimate
Estimate
Com
<---
Fac Vendor
.177
Com
<---
Fac Vendor
.120
Contrl
<---
Fac Client
.088
Contrl
<---
Fac Client
.132
Contrl
<---
Fac Vendor
.702
Contrl
<---
Fac Vendor
.603
Com
<---
Fac Client
.399
Com
<---
Fac Client
.471
Success
<---
Contrl
.547
Success
<---
Contrl
.434
Success
<---
Com
-.131
Success
<---
Com
-.131
continued on following page
282
Flexibility in Offshoring
Estimates (group invariant), continued
CONTROL GROUP (No Req. Change)
NON-CONTROL GROUP (Req. Change)
Success
<---
Fac Client
.599
Success
<---
Fac Client
.710
Success
<---
Fac Vendor
-.079
Success
<---
Fac Vendor
-.054
fac2
<---
Fac Vendor
.819
fac2
<---
Fac Vendor
.717
Out cnt
<---
Contrl
.717
Out cnt
<---
Control
.742
fac1
<---
Fac Client
.761
fac1
<---
Fac Client
.824
fac2
<---
Fac Client
.726
fac5
<---
Fac Client
.788
fac6
<---
Fac Vendor
.877
fac6
<---
Fac Vendor
.742
Beh cnt
<---
Control
.797
Beh cnt
<---
Control
.831
com4
<---
Com
.864
com4
<---
Com
.904
com3
<---
Com
.892
com3
<---
Com
.925
sat
<---
Success
.611
sat
<---
Success
.691
proc
<---
Success
.962
proc
<---
Success
.923
qlty
<---
Success
.808
qlty
<---
Success
.926
Squared Multiple Correlations
Squared Multiple Correlations
Estimate
Estimate
Com
.216
Com
.270
Control
.522
Control
.427
Success
.649
Success
.735
proc
.925
proc
.852
com3
.795
com3
.856
Out cnt
.514
Out cnt
.551
Beh cnt
.635
Beh cnt
.690
sat
.373
sat
.478
qlty
.652
qlty
.858
com4
.747
com4
.817
fac2
.672
fac2
.514
fac6
.769
fac6
.551
fac1
.580
fac1
.680
fac5
.527
fac5
.621
283
284
Chapter 13
Taking Information
Systems Business Process
Outsourcing Offshore:
The Conflict of Competition and Risk
Georgia Beverakis
University of New South Wales, Australia
Geoffrey N. Dick
University of New South Wales, Australia
Dubravka Cecez-Kecmanovi
University of New South Wales, Australia
ABSTRACT
As Business Process Outsourcing (BPO) moves offshore and now includes Information Systems (IS)
Processes, there is a need to consider a combination of the two. This article explores the factors that
a multinational organisation considered when it “offshored” its IS business processes to lower-cost
destinations. It focuses on determining the driving factors and challenges faced during the offshore
sourcing project. A single, in-depth interpretive case study approach was used to explore this research
topic. The results of this study show that the organisation under investigation was primarily driven to
offshore its IS business processes in order to become more competitive in the marketplace. This was
assisted by the organisation reducing its operational costs, and establishing a global presence in many
lower-cost locations offshore. A model was developed, which illustrates the interrelationships that exist
between these concepts.
INTRODUCTION
Offshore sourcing (the purchase of services
from overseas) is not a new phenomenon; it has
DOI: 10.4018/978-1-60960-605-3.ch013
existed for the last 2 decades in many industries,
particularly in the manufacturing industry, and
more recently in the financial services industry.
Despite being active for a number of years, offshore sourcing only gained momentum during
Y2K when there was an abundance of capable
Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
Taking Information Systems Business Process Outsourcing Offshore
professionals offshore, more “than those available
in the first world” (Dibbern, Goles, Hirschheim,
& Jayatilaka, 2004, p. 13; Hirschheim, George,
& Wong, 2004, pp.12-13). Since then, the phenomenon has gained increasing popularity in
many varied industries. Offshore sourcing in
Information Systems is defined as the handing
over of “responsibility for [the] management and
delivery of information technology services” to a
vendor located in another country (Pfannenstein
& Tsai, 2004, p. 72; Sabherwal, 1999).
Business Process Outsourcing (BPO) is an additional outsourcing concept, which has developed
in recent years. It is an outsourcing agreement
“where a third party provider is responsible for
performing an entire business function for the
client organization” (Dibbern et al., 2004, p.
11). BPO, like offshore sourcing, is a concept
familiar to both the manufacturing and financial
services industries. Business processes that have
typically been outsourced are IS/IT enabled business processes and not the actual IS/IT business
processes themselves, making it apparent that the
commonality of offshore sourcing and BPO is that
the business processes and services targeted for
outsourcing are those that are supported by IS/IT
(Willcocks, Hindle, Feeny, & Lacity, 2004). As
an example, loan approvals, debt collection, and
data input are often sent offshore, but related IS/
IT processes of reporting, modification of output,
and database administration generally have not
been. A diagrammatic representation of the evolution of these practices is included a little later
in the article after some further explanation, as
in Figure 2.
In recent years, there has been a significant
increase in the attention paid to BPO and offshore
sourcing. While the concept of offshore BPO has
been given some consideration, there is little
literature on the newer offshoring of IS business
processes. There is, however, some research that
has been conducted into the offshoring of other
business processes, for example, human resources services, finance and accounting services, and
customer support functions, to name a few (Whinston, 2004). The offshoring of these business
processes has typically been enabled by IS/IT
(Hirschheim et al., 2004, p. 23). But, there is a
further development—the offshoring of IS business processes themselves (such as customising
and modelling SAP reports, changing management, and modifying work roles and tasks by
sending certain components offshore). The research reported here has the aim of providing an
in-depth understanding of its drivers and motivations and is based on an intensive case study of
IS business processes offshoring by an Australian
arm of a multinational organisation—ComputerInc (a pseudonym, as are the names of staff), a
leading supplier of IT hardware, software, and
services.
This study identifies both the drivers (expected
benefits) and challenges of offshoring IS BPO.
It is perhaps relevant that one of the sites being
considered for taking on some of the work was
in fact on-shore—in Australia, but removed from
the main office. This was seen as significant at
the time as ComputerInc had several clients who
were expected to have concerns about data and
data interpretation going offshore. The resulting
framework grounded in empirical data contributes
to better understanding of the offshoring IS business processes which may aid organisations in their
decision-making surrounding offshore sourcing. It
is hoped that this framework will be employed in
future research to develop a more general theory
into offshore BPO. The authors firmly believe
that the interplay between the benefits and challenges explored here will be useful for managers
considering similar decisions.
BACKGROUND
Outsourcing
The outsourcing phenomenon has today evolved
from arrangements, which involve one client
285
Taking Information Systems Business Process Outsourcing Offshore
and one vendor where the vendor is responsible
for providing all IS-related services to the client;
to arrangements that involve multiple vendors
and clients and encompass more complex tasks.
Moreover, we now see “significant partnerships
and alliances … where client and vendor share risk
and reward” (Dibbern et al., 2004, p. 8). Furthermore, outsourcing deals now entail “value-based
outsourcing, equity based outsourcing, eBusiness
outsourcing, and business process outsourcing”
emphasising the evolution that has taken place in
the outsourcing arena (Dibbern et al., 2004, p. 8).
Further highlighting this evolution, IS outsourcing
used to typically focus on “single-system contracts
comprising a small portion of the IS budget”
and now IS outsourcing spans multiple systems
and embodies the “transfer of assets, leases, and
staff to a vendor that assumes a profit and loss
responsibility” (Lacity & Hirschheim, 1993a, p.
74). Additionally, there are no signs that indicate
that this growth in IS outsourcing will cease, in
fact it is likely to continue well into the future
(Willcocks et al., 2004, p. 7).
Motivations for outsourcing include cost
savings and strategic benefits (Costa & Beaumont, 2001; Dibbern et al., 2004; Hirschheim
et al., 2004; Jurison, 1995; King & Malhotra,
2004; Martinsons, 1993; McLellan, Marcolin, &
Beamish, 1995; Quelin & Duhamel, 2003; Sobol &
Apte, 1995, p. 270). Other commonly recognised
motivations include the ability to leverage the IS
expertise of its vendors (Lacity & Hirschheim,
1993b, p. 13; Martinsons, 1993, pp. 19-20) and
the ability to focus more attention on its core
competencies, making the organisation more
efficient and effective in its operations (Grover,
Cheon, & Teng, 1996; Harris, Giunipero, & Holt,
1998). Challenges include the irreversibility of
the outsourcing decision, loss of control, hidden
costs, challenges associated with the vendor, and
a loss of skills/knowledge/expertise (Apte et al.,
1997; Aubert, Rivard, & Patry, 2001; Earl, 1996;
Jurison, 1995; King & Malhotra, 2000; Lacity &
Hirschheim, 1993b; Lacity, Willcocks, Feeny,
286
1996; Martinsons, 1993; Quelin & Duhamel,
2003).
Offshore Sourcing
Offshore sourcing, commonly known as offshoring, refers to the outsourcing of services to a
third-party vendor in another country “as well as
offshore insourcing to an internal group within a
global corporation” (Carmel & Agarwal, 2002, p.
65). It thus does not necessarily mean “outsourcing” in the typical sense of the word, because it
does encompass a wide range of options, including
offshore insourcing, joint venture/cosourcing, and
outsourcing (Bhide, 2005, p. 39).
Offshoring grew in significance during the
1990s when organisations outsourced Y2K compliance work to offshore vendors (Pfannenstein
& Tsai, 2004, p. 73). Since then, the offshoring
phenomenon has continued to grow as it offers
companies several benefits, including significant
cost savings and access to a larger supply of more
sophisticated skills than those available onshore
(Dibbern et al., 2004, p.90; Hirschheim et al., 2004;
Pfannenstein & Tsai, 2004; Tas & Sunder, 2004,
p. 51). The term onshore refers to the location of
the organisation from which the services are being
sent. Typically, developed countries such as the
United States of America, the United Kingdom,
and even Australia offshore their services to
lower-cost destinations in more developing nations, such as India and China (Devata, Kumar,
& Stratopoulos, 2005, p. 102).
The growth of offshore sourcing has not
diminished and does not look likely to. In 2003,
offshore sourcing contracts made up a mere 1.4%
of all total outsourcing contracts, however, in 2004
“the value of offshore outsourcing contracts rose
890% from the previous year to $1.66 billion”
(Hirschheim et al., 2004, p. 8). Furthermore, the
literature identifies that this growth is set to continue, with offshore sourcing growing in both use
and popularity (Pfannenstein & Tsai, 2004, p. 75).
Taking Information Systems Business Process Outsourcing Offshore
There have been several factors that have
enabled the move offshore to take place. The
communications infrastructure available offshore
has significantly improved and is being provided
to organisations at a lower cost than is available
onshore. Additionally, the availability of welleducated and talented staff in countries such as
India and China provide companies with a large
supply of vendors and significant costs savings of
30% - 50% (Tas & Sunder, 2004, p. 51). Moreover,
the costs associated with coordinating the onshore
client and the offshore vendor have significantly
decreased, and there are technologies available to
assist organisations in managing this relationship;
which have also matured (Carmel & Agarwal,
2002, p. 66).
In summary, an organisation offshores its
services to achieve cost reductions, have access
to more sophisticated skills and labour markets,
improve its operational performance, achieve
flexibility and operating efficiency, and to gain a
strategic advantage in the marketplace (Aron &
Singh, 2005, pp.135-136; Dibbern et al., 2004,
p. 90; Gopal, Mukhopadhyay, & Krishnan, 2002,
p. 193; Krishna, Sahay, & Walsham, 2004, p. 64;
Pfannenstein & Tsai, 2004, p. 73; Tas & Sunder,
2004, p. 51).
There are significant challenges too, which
affect the risk associated with the offshoring project. Privacy and security can become problematic
adding an extra dimension to the management
process, as it is difficult to control and monitor the
relevant activities of offshore vendors (Davison,
2004; Kleim, 2004; Tafti, 2005). There are a range
of hidden costs associated with such arrangements,
including, vendor selection, establishment of contracts, transitioning the work offshore as well as
bringing it back onshore at the conclusion of the
contract (Pfannenstein & Tsai, 2004; Tafti, 2005).
A loss of control over factors associated with the
project can affect quality of work, loss of business knowledge as experienced employees leave
and take that knowledge with them, and resulting
inability to re-build should the work come back
onshore, all add to the risk (Herbsleb & Moitra,
2001; Pfannenstein & Tsai, 2004; Sobol & Apte,
1995; Tafti, 2005;). Furthermore, cultural differences (Davison, 2004; Hirschheim et al., 2004),
geographical distance, employee morale, and the
political environment of the offshore location
(Gupta & Raval, 2000; Krishna et al., 2004) are
also of concern.
Offshore Business Process
Outsourcing (BPO)
BPO is an outsourcing agreement which sees a
vendor perform an entire business function for
a client organisation (Dibbern et al., 2004, p.
11). The premise behind BPO is to outsource IT
enabled, noncore, largely back-office processes
to a vendor who has “superior structural and human capital in the areas of business process and
specific expertise” (Willcocks et al., 2004, p. 11).
Business processes appropriate for offshoring are
“well-defined, self-contained, and measurable,” as
well as being stable processes with little variance
(Tas & Sunder, 2004, p. 51). The BPO market has
experienced significant growth in recent years,
as an example in the UK and Europe and it was
estimated that revenues from BPO will rise to
72 billion Euros (from 43 billion Euros) during
the period from 2002 to 2005 (Willcocks et al.,
2004, p. 7).
Offshore BPO is simply a BPO agreement
with a vendor located in different country to that
of the client (Dibbern et al., 2004, p. 11). Offshore
BPO has been “made possible by the dramatic
fall in telecommunications costs and the ability to
transform paper-based activities into digital ones
requiring only a telephone and a computer” (Agarwal, Farrell, & Remes, 2003). Although it is now
common for business processes to be offshored
to lower-cost destinations, the literature identifies
that there has been little focus on the offshoring
of IS business processes, such as SAP functions,
database administration, as well as problem and
287
Taking Information Systems Business Process Outsourcing Offshore
Figure 1. Comparison of US and Indian wges (adapted from Devata et al., 2005, pg. 103)
change management processes, for instance (Tas
& Sunder, 2004, p. 51).
Motivations for such offshoring are of an
economic nature. First, a drastic difference in salaries—for instance, U.S. vs. Indian annual salaries
are graphically illustrated by Figure 1—motivates
organisations to undertake BPO offshoring.
Second, many organisations find offshore BPO
a desirable solution to become more efficient
because it enables the offshoring of noncore
business processes and thus lets the organisation
focus on its core-competencies, thus facilitating
the delivery of better products and services for its
customers (Devata et al., 2005, p. 103). It should
also be noted that the offshoring of BPO assists
organisations in making fixed costs variable,
which gives the organisation greater flexibility
in regards to the changing economy (Devata et
al., 2005, p. 102).
In addition to the challenges mentioned, an
organisation looking to offshore its business
processes is faced with extra problems, which are
primarily focused on costs, as well as the loss of
knowledge and expertise. Many organisations are
in such a rush to partake in business process
offshoring to remain on par with the competition,
288
that these organisations do not realise there are
greater benefits than simply reducing costs, such
as opportunities to improve the quality of the
organisation’s services. For the reason that reducing costs is often cited as the primary driver of
offshore BPO, vendor organisations often focus
on delivering a service for the clients at the
lowest-possible rate, as opposed to providing the
client with the most innovative and up-to-date
solution, because the clients provide the vendor
organisations with no incentive to do so. This can
lead to challenges regarding a reduction in the
quality of service, as well as additional costs for
any amendments that need to be made (Devata et
al., 2005, p. 117).
Additionally, an organisation also risks losing
its knowledge and expertise in the areas for which
it has offshored business processes, because the
vendor has assumed responsibility for providing
those services (Willcocks et al., 2004, p. 9). As
with offshore sourcing and IS outsourcing, this
can prove troublesome when the organisation
brings the business processes back onshore, and
can prove detrimental to the future health of the
organisation (Tafti, 2005, p. 556).
Taking Information Systems Business Process Outsourcing Offshore
Figure 2. The evolution of outsourcing concepts
The evolution outlined above, in particular, as
suggested by Hirschheim et al. (2004), Dibbern et
al. (2004), Carmel and Agarwal (2002), and Tas
and Sunder (2004), of the outsourcing concepts as
well as the motivations behind these concepts can
be illustrated as in Figure 2, showing the movement from outsourcing to parallel development
of offshore sourcing and BPO to the eventual
offshoring of the IS business processes themselves.
At this point, it maybe useful to further distinguish between traditional business processes
and IS business processes. Traditional business
processes are those that are typically enabled by
IS/IT (Hirschheim et al., 2004, p. 23), such as human resources services, finance and accounting
services, and customer support functions (Whinston, 2004), whereas IS business processes are
the actual IS/IT business processes themselves,
such as customising and modelling ERP reports,
database administration, and modifying roles
and tasks. While offshoring traditional business
processes has been well researched, offshoring
IS BPO as the most recent outsourcing concept
is still quite unknown.
As the processes of IS BPO offshoring are not
well understood, the specific challenges and risks
that an organisation is likely to face are largely
unknown. Both the paucity of research in this
domain and the relevance of these issues for IS
practice motivated us to examine:
1.
2.
What are the motivations and driving forces
involved in an organisation’s decision making to offshore its Information Systems
business processes?
What are challenges and risks encountered
during such decision making?
To answer these research questions (notwithstanding that the first may be driven by a management decree to do so) required an in-depth study
of IS BPO offshoring decision making. To this
end, we selected a case organisation that has just
began its IS BPO offshoring decision making and
where such offshoring phenomenon was present
in its exemplary form (complex IS/IT operations
managed globally). Such an in-depth case study,
the first to authors’ knowledge in relation to IS
BPO offshoring, enabled us to gain insights into
managerial thinking and decision making and to
achieve a deeper understanding of the interplay
289
Taking Information Systems Business Process Outsourcing Offshore
between different motivations and forces as well
as challenges and risks..
Research Methodology
Our choice of the case organization ComputerInc—a multinational company with over 300,000
people, operating in over 70 countries worldwide—was motivated by several key reasons.
First, as ComputerInc competes and manages its
IS/IT operations globally with clients ranging in
size and vary across a number of industries, it
experiences a full measure of global competition
and pressures to lower costs. Second, at the end of
2005, ComputerInc made a decision to offshore its
IS business processes to lower-cost destinations.
They were aware of the long term importance of
this decision and devoted the following 8 months
of 2006 to determine how to do it and which
processes to offshore. As such, ComputerInc’s
decision making regarding offshoring IS business
processes was an exemplar of the phenomenon we
aimed to study. We were also fortunate that one
of the authors gained access to ComputerInc and
was granted permission to conduct an in-depth,
intensive field study of its decision making processes that led to determining what IS business
processes were suitable to be offshored (during
6 months Feb-Aug 2006).
The methodology adopted for this study was
an interpretivist case study (Klein & Myers, 1999;
Walsham, 1995). Data were collected from multiple sources, including semistructured interviews,
documentary materials, and notes from field observation. The interview protocol followed was
the preparation of a range of discussion questions
(piloted and tested) which provided the interviewer
with the opportunity of addressing certain topics
in more depth. The transcriptions of the interviews
were provided to the interviewees to ensure authenticity. Using multiple sources of information
provided the authors with different interpretations
of the same phenomenon, addressing the problem
of construct validity, which is inherent in the use of
290
case studies (Benbasat, Goldstein, & Mead, 1987,
p. 374; Yin, 2003). The use of multiple methods
of data collection has ensured the production of a
high quality case study (Yin, 2003, p. 99).
Semistructured interviews were conducted
with a range of managerial staff covering the
areas of IS management, Projects Operations,
Service, International Programs, and Database
Management in ComputerInc (10 in total) lasted
between half an hour and 2 hours. Collected documents regarding the IS business process offshoring
project were also examined as part of the research.
These included spreadsheets, presentations, and
meeting minutes. Most of these documents were
confidential and available only on the premises
of the company. The researcher in the field also
made notes on events and impressions about the
meetings and the people involved. The notes were
useful in interpreting data and reconstructing the
story afterwards.
The collection and analysis of empirical data
were inspired and governed by grounded theory.
Although there are many relevant concepts and
frameworks to analyse and explain processes of
IS-enabled BPO outsourcing, they do not necessarily apply to IS BPO offshoring. Without an
obvious conceptual framework available from the
literature our in-depth study led of necessity to
grounded theory (Strauss & Corbin, 1990, 1998).
While grounded theory is broadly understood as
an approach to developing a theory from the data
and be “faithful to the evidence”, where “the researcher compares unlike phenomena with a view
toward learning similarities” (Neuman, 2006, p.
60), the adoption of grounded theory as a method
differs across studies. In this study, grounded
theory was adopted in line with the principles
and analytic guidelines proposed by Strauss and
Corbin (1990) and reinterpreted by Charmaz
(2000). Theory development is grounded in data
in a particular way: through the observation and
interpretation of researchers engaged in lives of
subjects studied. Namely, by interacting with
the subjects and by analysing and comparing the
Taking Information Systems Business Process Outsourcing Offshore
collected data, researchers develop understanding
of subjects’ actions and meanings, which evolve
into further, more abstract conceptual categories
and interpretations. By defining (and refining)
categories and deriving relationships between
them, researchers develop a substantive theory of
a particular phenomenon grounded in the data. As
the research progresses, the researcher continuously interprets the collected data, which is then
used to refine the proposed theory being developed
(Charmaz, 2000, p. 509).
In this study, the empirical data were gathered
and analysed to identify themes, patterns, and common categories. The coding process involved three
stages, open, axial and selective coding (Strauss &
Corbin, 1998). Open coding included line-by-line
examination and interpretation of texts (interviews
and documents) during which interesting themes
and ideas were identified and named (i.e., coded).
Open coding enabled the authors to become familiar with the views of the research participants
“rather than assume that we share the same views
and worlds” (Charmaz, 2000, p. 515). Open codes
reflected both the terms people in the field used
(these are called “in vivo” codes) and the terms
found in the literature. This approach to coding
in grounded theory recognizes that researchers do
not enter the field with a “blank slate” and that
they are necessarily sensitised by the concepts
and findings from the literature. While we aim to
ground codes and categories in the empirical texts
the coding process does not and cannot happen in
a conceptual vacuum and is therefore inevitably
linked to knowledge in the field of study. This is
in fact beneficial when the codes are grounded in
the empirical data as well as are related to concepts
from the literature. This allows comparisons of
a derived theory with literature or in some cases
testing theories from the literature.
It is important to emphasize that open codes
were identified in close relation to the text and
only later they had been compared, redefined, and
restructured. Furthermore, as the open codes began
to accumulate, they were grouped and regrouped
under higher order categories (Strauss & Corbin,
1998, p. 113). To have a better overview of the
open codes and how they can be interrelated a
particular technique of graphical representation
was used in this study. Each code and associated
text was printed on colored paper (colors corresponding to different sources of texts) and arranged
on a wall. Grouping and regrouping of the codes
(and associated texts) under different categories
on the wall provided a visual presentation and
assisted in the sharing of ideas and discussion
among researchers.
During the next stage, axial coding, categories
were analysed and constantly compared in terms
of properties and dimensions, thus enabling the
researchers in exploring the relationships between
categories and identifying their similarities and
differences. Having the codes and categories arranged on the wall was even more useful in this
stage as it allowed hypothesizing and exploration
of different relationships among the categories. By
moving and re-organising the codes and categories
around the wall, the discussion easily focused on
their properties and dimensions thus enabling a
grounded testing of axial codes, categories, and
their relationships. As categories and their relationships emerged, the empirical data were seen
in a new way, with each step going further from
the raw empirical data and towards a substantive
theory (Blaikie, 2000; Strauss & Corbin, 1998).
This was followed by selective coding where
the aim was to identify a central category or a
main theme of the research. Strauss and Corbin
(1998) define selective coding as the “process of
integrating and refining theory” (p. 143). During
the process of integration, a central category is
decided upon, which represents the main theme
of the research. This category must appear regularly in the data and as this category is related to
other categories, the evolving explanation must
be logical and consistent (Strauss & Corbin,
1998, pp.146-147). To assist the researchers in
identifying a central category, a few techniques
may be employed. One such technique, which
291
Taking Information Systems Business Process Outsourcing Offshore
was adopted for this article is that of writing a
storyline, which involves writing a few sentences
about “what seems to be going on here” (Strauss
& Corbin, 1998, p. 148). The result of relating
categories to the central category is illustrated in
Figure 3. This process was employed for each of
the higher-order categories that surfaced during
open and axial coding. From this, two central
categories emerged. As a result, the authors were
able to develop an understanding of how each
of the higher-order categories were interrelated
and how they related to the central categories.
By identifying the central categories as well as
the interrelationships between them and other
categories, we developed the substantive theory of
IS business processes offshoring in ComputerInc.
While the proposed substantive theory is context
bound and applicable in a single setting, it may be
a starting point for the subsequent development
of a more general, formal theory applicable in a
variety of settings.
Findings
Table 1 gives the axial codes and categories that
emerged from the open coding.
An example of how a category is developed,
starting from the comments and open coding,
through the axial codes is given in Appendix A.
Table 1. Emerging categories
Axial Codes
Category
24 Hour organisation
Future offshoring
Expand globally
Global capability
Virtual teams
Establish a Global Capability
Region driven
Contractual obligations
Existing infrastructure
Low cost
Location Choice
Research driven
Market research
Past experience
Research
Management driven
Management’s Decision
Behind competition
Create competition
Efficiency/effectiveness
Remain competitive
Become More Competitive
Cost driven
Increase profit
Reduce Costs
Risk mitigation
Employee support and changes
Public opinion issues
Challenges
Customer impact
Team impact/morale
Challenges / Risks
Type of work
Skills availability
Improve onshore skills
Offshore skills
Access to more sophisticated skills
Inferior skills
Skills & Type of Work Available & Targeted
292
Taking Information Systems Business Process Outsourcing Offshore
Through the analysis of these categories and
their interrelationships, the authors were able
to understand why ComputerInc was driven to
offshore its IS business processes to lower-cost
destinations; and also understand the challenges
that the organisation faced during the offshoring
process.
Becoming more competitive in the marketplace
was the key finding relating to Research Question
1. Other key findings were that ComputerInc was
also driven by reducing its operational costs and
establishing a global capability. Although drivers
themselves, these concepts enable ComputerInc
to achieve its primary aim of becoming more
competitive in the marketplace. Furthermore,
there were a number of factors identified that had a
significant influence on these drivers. These were
management’s overriding decision to offshore the
IS business processes, the skills and the type of
work available in the offshore locations and the
decisions surrounding the offshore location choice.
It is evident that central to the offshore sourcing
project is ComputerInc’s aim of becoming more
competitive in the marketplace, that is, to hold
the majority of the market share. our competitors
are still holding more of the market share than
what we are… so in order to remain competitive
and be a leading force, then we need to go down
that path and make it happen. (Janet, personal
communication)
The company saw that in order to become
more competitive, it must remain competitive
and, further, that by reducing its operational costs,
expanding globally and improving the efficiency
and effectiveness of its operations. The organisation considered that offshoring its IS business
processes would enable the organisation to achieve
this goal as is highlighted through the statements
of the respondents.
ComputerInc has historically been the leading and dominant force within its industry. As
such, it is imperative that the organisation at
least maintains this status to ensure its success;
this has been recognised by many of the service
managers, who identify that “we need to be, as
an organisation, competitive” (Andrew, Service
Manager, personal communication) and “we need
to remain competitive in a very cost competitive
business” (David, Service Manager). Janet, Service Manager (personal communication), further
highlights this need as well as describing the
reasons underlying it:
the rest of the world is doing it [offshoring] so
in order to remain competitive and be a leading
force then you need to go down that path and I
guess make it happen.
Improving the efficiency and effectiveness
of the organisation’s operations, which entails
establishing a global set of processes, tools, and
standards for the global organisation has also
made it possible for ComputerInc to become more
competitive in the marketplace for the reason
that now there is only one way to operate and
complete functions, the organisation has become
more consistent, enabling the improvement and
refinement its processes:
If we can get it right and we get a global set of
processes, then obviously we are going to be much
more efficient and effective at what we do… and
if we are more effective and we are more efficient,
then that would mean we are more competitive
in the marketplace. (Janet, Service Manager,
personal communication)
Reducing the operational costs of the organisation is not only a means to becoming more
competitive, but it is also a key driver in itself,
with all of the service managers identifying it as
a core motivation for the organisation offshoring
its IS business processes. For instance, Service
Manager Peter went so far as to suggest that “the
bottom line… is to reduce costs” (personal communication). Furthermore, Craig, Project Owner,
293
Taking Information Systems Business Process Outsourcing Offshore
identified that this offshore sourcing project
enables the organisation to be successful in that
respect, that it “allows us [ComputerInc] to reduce
cost” (personal communication). These efforts
are predominately focused on reducing operating
costs and in particular, salaries and labour costs.
[x] is a location where we have a series of centralised functions coming out of there and so if
those functions that we are looking at up here
could be centralised …. and the cost of living is
cheaper down there and free up resources up here.
(Janet, personal communication)
by spreading our the processes and the knowledge
base across various countries, obviously… this is
meant to drive the costs down because the work
can be done at a lesser cost. (Vijay, personal
communication)
These countries, the salaries and cost to live is
much lower than in other European countries as
well as America and Australia. (Vijay, personal
communication).
For this particular project, the offshore locations that were chosen are India, Chin, and a
lower-cost destination in Australia. Craig (personal communication) highlights the importance
of establishing this global capability and electing
to offshore to growing markets:
where we see our strengths in the future, is supporting Shenzhen… China is very much a growing
market, and the organisation needs to be able to
demonstrate to the big businesses in China that
from a service perspective that we have a capability in China.
As observed by one of the authors, ComputerInc’s executive management team was directly
involved in making the highly essential decisions
294
for this project. The reduction of the organisation’s
operational costs and becoming more competitive
in the marketplace came as a decree from the
executive management team. Furthermore, Craig
(personal communication) highlights that “it’s not
just an Australian decision… competitors are doing it; it’s a management decision, worldwide.”
The skills that were being sourced and the
skills that were available in the offshore locations
worked in both ways to influence the key drivers
of the offshore sourcing project, directly influencing the organisation becoming more competitive
and establishing a global capability. ComputerInc
has moved many processes offshore that were
identified by a number of characteristics; some of
these characteristics include “highly repetitive and
standardised” processes (Janet, Service Manager,
personal communication). However, the skills and
the type of work targeted for offshoring directly
influenced where the organisation was able to
establish its global capabilities. These locations
were dependent on many factors, including where
the growing markets were and what locations had
the correct skill sets available.
ComputerInc decided on appropriate offshore
locations based on a number of factors, one being the type of work these offshore locations
specialised in and the skills they had available.
ComputerInc’s research identified that the chosen
locations, India, China, and the lower-cost onshore
location in Australia are endowed with talented
and educated resources, and these locations were
advertising their skill sets to the world.
The key findings relating to Research Question 2 were that during the offshoring of its IS
business processes, ComputerInc were faced
with a number of challenges. Employee changes,
impact and support onshore affected staff—it not
only impacted the employees who lost their roles,
but the team also felt the impact of the changes
because “a lot of the team members are nervous
about what it would mean for them” (Andrew,
Service Manager, personal communication).
Many of the remaining team members struggled
Taking Information Systems Business Process Outsourcing Offshore
with “the survivor syndrome, when they say, why
wasn’t it me? And they’re constantly looking over
their shoulders” (Janet, Service Manager, personal
communication). Furthermore, the overall team
morale was impacted, which has been very difficult for the service managers to control (Janet,
Service Manager).
The geographic distance between the onshore
and offshore locations, cultural barriers, and
connectivity and infrastructure were challenges,
too, in that relationship management became difficult, the language barrier was problematic, and
if the network connectivity in the offshore location failed, there was no mitigation technique or
backup plan in place. As David, Service Manager
(personal communication), explains:
no I don’t know of any other backup if we lose
connectivity, the only thing is probably still having
a number of people back in country to avert that.
ComputerInc also faced difficulties in terms
of security and privacy (some customers were
reluctant to, or prohibited from, allowing data
offshore), service delivery, and reduced quality
of service (there was the possibility that during
the early stages of the offshoring project, there
would be dissatisfied customers and that the
quality of service would diminish), the social
and political environments of the offshore locations (potentially less stable than might be the
case in Australia) and the impact of offshoring
on the customer (some degree of dissatisfaction
was to be expected). The following factors were
identified as having a significant influence on the
above challenges: the unavailability of skills or
the substandard skills available in the offshore
locations (there was not a complete match with
what was desired), the research conducted during the pre-implementation phases of the project,
and the decisions made surrounding the offshore
location choices, in that tasks were chosen to fit
predetermined locations.
Discussion
The selective coding and the in-depth analysis
described above led to a substantive theory of IS
BPO offshoring in ComputerInc, which we present
in a form of a substantive framework in Figure
3. The substantive framework emerged through
coding, analysis, interpretation, and constant
Figure 3. A substantive framework of ComputerInc IS BPO offshoring
295
Taking Information Systems Business Process Outsourcing Offshore
comparisons of the data collected (interviews,
observations and documents). The framework in
Figure 3 below describes specific relationships
between the key eight categories that emerged
from the data: Become More Competitive, Challenges/Risks, Reduce Costs, Establish a Global
Capability, Skills & Work Targeted & Available,
Location Choice, Management Decision, and
Research. Among them Become More Competitive and Challenges/Risks emerged as central categories. Central to the framework is the inferred
relationship between the goal of becoming more
competitive and the challenges and risks involved
in doing so. Becoming more competitive is driven
by managerial decisions taking into account a
desire to reduce costs establish a wider global
presence and the skills and work available. The
challenges and risks are dependent on where the
offshoring takes place, what needs to be done,
and the extent of research governing the decision.
The framework demonstrates the relationships
that exist between the categories. There are four
types of relationships present in the framework—
significant (heavy black arrow) illustrating the
concepts that have a direct influence on either of
the central categories Become More Competitive
or Challenges/Risks. The relationship between
the central categories themselves—the impact
of Challenges/Risks on Become More Competitive—is not directly observable from the data but
is nevertheless implied, which we inferred from
the broader analysis and insight (indicated by
the heavy dashed black arrow). The red arrows
represent relationships between concepts where
concept A has a direct relationship with concept
B and concept B is related to either Become
More Competitive or Challenges/Risks. Finally,
the blue arrows illustrate the remaining relationships presented by the framework, which do not
directly impact the central categories. As a result
of the data analysis and the model, two primary
propositions are put forward.
Proposition 1: The central driver of the
offshore sourcing project is for ComputerInc to
296
Become more Competitive. This concept came
as a directive from ComputerInc’s executive
management team, which additionally identified
the organisation’s need to reduce its operational
costs. Becoming more competitive not only focuses on improving the competitive standing of
ComputerInc but also looks at improving how
efficient and effective it is with its operations.
ComputerInc was able to achieve its goal of becoming more competitive, with the assistance of
two factors, namely reducing its operational costs,
and establishing a global capability, which are
acknowledged as two of the main advantages of
offshore sourcing (Sobol & Apte, 1995, p. 274).
Reducing operational costs is one key advantage
of offshoring, but another is the ability for organisations to develop and operate global information
systems (Sobol & Apte, 1995, p. 271). Establishing
capabilities in locations around the world provided
ComputerInc with access to a highly skilled and
readily available work force allowing the organisation to “improve their ability to compete in their
fast-paced domestic and international markets”
(Wilson, 2003, p. 102). Furthermore, the skills
and the type of work that are available in the offshore locations further enabled ComputerInc to
establish its global capability as well as become
more competitive in the marketplace because
“having workers overseas in close proximity to
a new pool of consumers can assist a company
in expanding its global presence” (Pfannenstein
& Tsai, 2004, p. 74).
Proposition 2: The Challenges/Risks faced and
anticipated by ComputerInc may prove to impede
the organisation from Becoming more Competitive. This relationship is a research inference made
based on the empirical data that emerged during
analysis. Although there is no direct evidence in
the data linking these concepts, the analysis of
all the data and the observations indicated the
existence of this relationship. Furthermore, the
authors have identified this as one of the core
note-worthy relationships present in the model
presented. ComputerInc conducted a thorough
Taking Information Systems Business Process Outsourcing Offshore
analysis into the potential challenges the organisation might have been faced with, prior to partaking
in this offshoring project, through in-depth market
research and assessments of past experiences both
within the organisation and externally. However,
the organisation developed very few risk mitigation techniques and as a result, these challenges
impeded its ability to become more competitive
in the marketplace, that is, until these challenges
were adequately managed. It is absolutely necessary for ComputerInc to act on the findings of the
research it conducts, and to develop a series of
techniques to assist the organisation in preventing,
mitigating, and/or managing these challenges and
risks, as is addressed by Hirschheim et al. (2004)
who stipulate, “It is critical that an organization
considering offshore outsourcing take the time to
evaluate the challenges and study how other organizations have handled these challenges” (p. 22).
This article has identified a number of challenges and risks that projects of this nature face.
The authors believe that the list developed provides a useful starting point for other organisations considering a similar decision. The way an
organisation is able to mitigate and/or manage
these challenges and risks would to a large extend
impact the goal of becoming more competitive in
the marketplace.
CONCLUSION
Based on a grounded theory study of the IS BPO
offshoring in ComputerInc, the article proposes
a substantive framework that identifies key categories and their relationships (as described in
Figure 3). A key contribution of this research is
that it highlights the complex relationship between
the offshoring objectives and risk management.
In this way, and with the development of the
framework depicting the interrelationships, the
study takes the prior research further. Offshore
BPO is not a panacea. There are many variables
involved and these will differ significantly from
project to project. Organisations considering such
action need to carefully evaluate the risks and
their potential effects.
A further contribution of this study is the identification of areas for further study. The impact of
personal opinions and bias could prove to be an
impediment to the success of such of an offshore
sourcing project. A study of this impact could
prove extremely valuable and make a significant
contribution to the offshore sourcing field, in both
theory and practice. Additionally, the principle
researcher saw very little questioning of the decisions made by the executive management team.
The decisions were simply accepted as fact even
though many of the people involved had first-hand,
on-the-ground experience with the roles being sent
offshore. It may be that this could be the reason
behind why a number of the employees did not
fully comprehend the organisation’s reasons for
going offshore. The authors believe management
gathering ideas from those directly involved would
be an extremely interesting area for research. Other
limitations include the potential opportunity for
the interviewees to mislead the researcher or push
a point of view, the relatively brief time-frame in
which the study was conducted and the organisation to which the tasks were outsourced being
outside the scope of the study.
The authors believe that the proposed substantive framework of IS BPO offshoring in ComputerInc provides a sound basis for the examination of
such projects in other companies as it conceptualizes the key drivers associated with the offshore
sourcing of IS business processes, as well as the
challenges that are faced in such a situation. The
framework suggests that becoming more competitive in the marketplace primarily drove ComputerInc to offshore its IS business processes, and in
order to achieve this, the organisation reduced its
operating costs and established global capabilities. While this framework is grounded in the data
and therefore context bound it is also found to be
supported by the outsourcing and offshore BPO
297
Taking Information Systems Business Process Outsourcing Offshore
literature, which highlights its potential use in the
development of a more general theory.
For practitioners, the authors believe that the
study has presented a series of potential challenges
that may arise during an offshore sourcing project,
has highlighted the importance of understanding
the goals of the project and conducting thorough
and appropriate research, and demonstrates the
importance of communicating significant decisions and in considering the knowledge of all
parties involved. As such, it is hoped that this
study will go some way towards helping managers note and understand the pitfalls and provide a
greater understanding of how the conflict between
benefits and challenges might be resolved.
Bhide, D. (2005). Strategic offshoring: Decision
analysis, best practices, and emerging trends.
In Brudenall, P. (Ed.), Technology and offshore
outsourcing strategies (pp. 37–64). Great Britain:
Palgrave Macmillan.
REFERENCES
Costa, C., & Beaumont, N. (2001). Information
technology in Australia. Working Paper.
Agarwal, V., Farrell, D., & Remes, J. (2003).
Offshoring and beyond [Special edition]. The
McKinsey Quarterly, 4.
Apte, U. M., Sobol, M. G., Hanaoka, S., Shimada,
T., Saarinen, T., Salmela, T., et al. (1997). IS
outsourcing in the USA, Japan and Finland: A
comparative study. 12, 289-304.
Aron, R., & Singh, J. V. (2005). Getting offshoring right. Harvard Business Review, 135–143.
Aubert, B. A., Rivard, S., & Patry, M. (2001). A
transaction cost model of IT outsourcing. Information & Management, 41, 921–932. doi:10.1016/j.
im.2003.09.001
Babbie, E. (2004). The practice of social research
(10th ed.). CA: Wadsworth/Thomson Learning.
Benbasat, I., Goldstein, D. K., & Mead, M. (1987,
September). The case research strategy in studies
of information systems. Management Information
Systems Quarterly, 369–386. doi:10.2307/248684
298
Blaikie, N. (2005). Designing social research.
Cambridge, UK: Polity Press.
Carmel, E., & Agarwal, R. (2002). The maturation
of offshore sourcing of information technology.
MIS Quarterly Executive, 1(2), 65–77.
Charmaz, K. (2000). Grounded theory: Objectivist and constructivist methods. In Denzin, N., &
Lincoln, Y. S. (Eds.), Handbook of qualitative
research (2nd ed., pp. 509–535). Thousand Oaks,
CA: Sage.
Davison, D. (2005). Top 10 risks of offshore outsourcing. CIO – Analyst Corner. Retrieved July
28, 2008, from http://www2.cio.com/analyst/
report2224.html
Devata, A. C., Kumar, R., & Stratopoulos, T.
(2005). Business process outsourcing: A manager’s guide to understanding the market phenomenon. In Brudenall, P. (Ed.), Technology and
offshore outsourcing strategies (pp. 97–115).
Great Britain: Palgrave Macmillan.
Dibbern, J., Goles, T., Hirschheim, R., & Jayatilaka, B. (2004). Information systems outsourcing: A survey and analysis of the literature. The
Data Base for Advances in Information Systems,
35(4), 6–102.
Earl, M. J. (1996). The risks of outsourcing IT.
Sloan Management Review, 37(3), 26–32.
Glaser, B. G., & Strauss, A. L. (1968). The discovery of grounded theory strategies for qualitative
research. London: Weidenfeld and Nicolson.
Taking Information Systems Business Process Outsourcing Offshore
Gopal, A., Mukhopadhyay, T., & Krishnan, M. S.
(2002). The role of software processes and communication in offshore software development.
Communications of the ACM, 45(4), 193–200.
doi:10.1145/505248.506008
Grover, V., Cheon, M. J., & Teng, J. T. C. (1996).
The effect of service quality and partnership on
the outsourcing of information systems functions.
Journal of Management Information Systems,
14(4), 89–116.
Gupta, U. G., & Raval, V. (1999). Critical success factors for anchoring offshore projects.
Information Strategy: The Executive’s Journal,
15(2), 21–27.
Harris, A., Giunipero, L. C., & Hult, G. T. M.
(1998). Impact of organisational and contract flexibility on outsourcing contracts. Industrial Marketing Management, 27, 373–384. doi:10.1016/
S0019-8501(97)00085-0
Herbsleb, J. D., & Moitra, D. (2001). Global
software development. IEEE Software, 16–20.
doi:10.1109/52.914732
Hirschheim, R., George, B., & Wong, S.F. (2004,
December). Information technology outsourcing:
The move towards offshoring [Special issue].
Indian Journal of Economics and Business, 1-30.
Jurison, J. (1995). The role of risk and return in
information technology outsourcing decisions.
Journal of Information Technology, 10, 239–247.
doi:10.1057/jit.1995.27
Kern, T., Willcocks, L. P., & van Heck, E. (2002).
The winner’s curse in IT outsourcing: Strategies
for avoiding relational trauma. California Management Review, 44(2), 47–69.
King, W., & Malhotra, Y. (2000). Developing a
framework for analyzing IS sourcing. Information & Management, 37, 323–334. doi:10.1016/
S0378-7206(00)00046-X
Kleim, R. (2004, Summer). Managing the risks
of offshore IT development projects. Information
Systems Management, 22–27. doi:10.1201/1078/
44432.21.3.20040601/82473.4
Klein, H., & Myers, M. (1999). A set of principles
for conducting and evaluating interpretive field
studies in information systems. Management
Information Systems Quarterly, 23(1), 67–88.
doi:10.2307/249410
Krishna, S., Sahay, S., & Walsham, G. (2004).
Managing cross-cultural issues in global software
outsourcing. Communications of the ACM, 47(4),
62–66. doi:10.1145/975817.975818
Lacity, M. C., & Hirschheim, R. (1993a). The
information systems outsourcing bandwagon.
Sloan Management Review, 35(1), 73–86.
Lacity, M. C., & Hirschheim, R. (1993b). Information systems outsourcing, myths, metaphors
and realities. West Sussex, England: John Wiley
& Sons.
Lacity, M. C., Willcocks, L. P., & Feeny, D. F.
(1996). The value of selective IT sourcing. Sloan
Management Review, 37(3), 13–25.
Martinsons, M. G. (1993). Outsourcing information systems: A strategic partnership with
risks. Long Range Planning, 26(3), 18–25.
doi:10.1016/0024-6301(93)90003-X
McLellan, K., Marcolin, B. L., & Beamish, P. W.
(1995). Financial and strategic motivations behind
outsourcing. Journal of Information Technology,
10, 299–321. doi:10.1057/jit.1995.32
Neuman, W. L. (2006). Social research methods
qualitative and quantitative approaches (6th ed.).
USA: Pearson Education.
Pfannenstein, L. L., & Tsai, R. J. (2004, Fall).
Offshore outsourcing: Current and future effects
on American IT industry. Information Systems
Management, 72–80. doi:10.1201/1078/44705.
21.4.20040901/84190.9
299
Taking Information Systems Business Process Outsourcing Offshore
Quelin, B., & Duhamel, F. (2003). Bringing
together strategic outsourcing and corporate
strategy: Outsourcing motives and risks. European Management Journal, 21(5), 647–661.
doi:10.1016/S0263-2373(03)00113-0
Roy, V., & Aubert, B. A. (2002). A resourcebased analysis of IT sourcing. The Data Base for
Advances in Information Systems, 33(2), 29–40.
Sabherwal, R. (1999). The role of trust in outsourced
IS development projects. Communications of the
ACM, 42(2), 80–87. doi:10.1145/293411.293485
Tas, J., & Sunder, S. (2004). Financial services business process outsourcing. Communications of the
ACM, 47(5), 50–52. doi:10.1145/986213.986238
Walsham, G. (1995). The emergence of interpretivism in IS research. Information Systems Research,
6(4), 376–394. doi:10.1287/isre.6.4.376
Whinston, A. (2004). Offshoring statistics – dollar
size, job loss, and market potential. EBStrategy.
Retrieved July 28, 2008, from http://ebstrategy.
com/Outsourcing/trends/statistics.htm
Sobol, M. G., & Apte, U. (1990/1995). Domestic
and global outsourcing practices of America’s
most effective IS users. Journal of Information
Technology, 10, 269–280. doi:10.1057/jit.1995.30
Willcocks, L., Hindle, J., Feeny, D., & Lacity,
M. (2004, Summer). IT and business process
outsourcing: The knowledge potential. Information Systems Management, 7–15. doi:10.1201/1
078/44432.21.3.20040601/82471.2
Strauss, A., & Corbin, J. (1998). Basics of qualitative research: Grounded theory procedures and
techniques. Newbury Park: Sage Publications.
Wilson, R. (2003). Outsource tide won’t ebb –
adapt to new order, CICC panel tells designers.
Electronic Engineering Times, 1289, 1/102.
Tafti, M.H.A. (2005). Risks factors associated with
offshore IT outsourcing. Industrial Management
+ Data Systems, 105(5), 549-560.
Yin, R. K. (2003). Case study research: Design
and methods (3rd ed.). Thousand Oaks, CA: SAGE
Publications.
300
Taking Information Systems Business Process Outsourcing Offshore
APPENDIX
Figure 4. Example of support for a “Category”
This work was previously published in International Journal of Global Information Management, Volume 17, Issue 1, edited
by Felix B. Tan, pp. 32-48, copyright 2009 by IGI Publishing (an imprint of IGI Global).
301
302
Chapter 14
Managing Knowledge
Transfer in Offshore
Software Development:
The Role of Copresent and
ICT-Based Interaction
Aini Aman
University Kebangsaan Malaysia, Malaysia
Brian Nicholson
Manchester Business School, UK
ABSTRACT
The aim of this article is to examine the role of copresent interaction and the extent to which this can be
supplanted by information and communication technology-based interaction for managing knowledge
transfer in distributed settings. This study draws on two case studies of small UK firms sourcing software
development from India and Bangladesh. Using Nonaka and Konno’s knowledge creation theory, the
role of copresent and ICT-based interactions in managing knowledge transfer is explained. The article
contributes an extension of the concepts of knowledge creation theory by providing evidence of the
role of copresent and ICT-based interaction for knowledge transfer in the context of offshore software
development.
INTRODUCTION
Offshore software development refers to the
process of sending tasks at various stages of the
software development lifecycle to suppliers in
DOI: 10.4018/978-1-60960-605-3.ch014
low-wage countries. This may involve direct
ownership of foreign facilities, partnering, or outsourcing (Shao & Smith David, 2007; Carmel &
Tija, 2005). There is a large, expanding literature
on knowledge transfer (see Schultze & Leidner
[2002] for a review) and an increasing focus on
Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
Managing Knowledge Transfer in Offshore Software Development
the complexities of knowledge transfer between
software team members that are separated across
time and space (e.g., Nicholson & Sahay, 2004;
Kotlarsky, Oshri, Hillegersberg, & Kumar, 2007;
Cha, Pingry, & Tatcher, 2008; Vlaar, Fenema, &
Tiwari, 2008). Other authors have questioned
the need for co-present interaction for effective
knowledge transfer in such scenarios (Boden &
Molotch, 1994; Maznevski & Chudoba, 2000;
Nardi & Whittaker, 2002; Oshri, Kotlarsky, &
Willcocks 2007). Boden and Molotch (1994)
emphasize how co-present interaction contributes
to the development of trust. This concurs with
the findings of Maznevski and Chudoba (2000),
who describe the importance of initial co-present
meetings to establish effective temporal rhythms
like ‘heartbeats’ which support subsequent stages
of global virtual teamwork. Nardi and Whittaker
(2002) point out how copresent interactions allow
bodily shared activities such as touching, eating
and drinking together, or engaging in mutually
meaningful experiences through socialization in a
common physical place. They argue that copresent
interaction is particularly useful as it is ‘thick’ with
information and can enable a transfer of greater
degree of contextual knowledge than ICT-based
interaction. Oshri et al. (2007) also emphasize
the importance of copresent meetings for creating
interpersonal ties and facilitating socialization.
However, in their study, socialization in globally
distributed teams is also supported by the use
of various collaborative technologies employed
before and after copresent meetings.
A gap in the existing academic literature and
a problem facing practitioners is that although
copresence is desirable for knowledge transfer,
arrangements may be logistically impossible or
cost prohibitive. This article aims to improve our
understanding of the practices firms are adopting
to overcome the dilemma between the need for
co-presence on the one hand and the drive for
lowest cost ICT-mediated offshore outsourcing
on the other. The study is guided by the following
research questions:
•
•
What is the role of co-present interaction
for knowledge transfer?
How can ICT-based interaction enable
knowledge transfer?
The empirical study draws on two cases of
UK-based software companies sourcing software
development from India and Bangladesh. The contribution of the article lies in the application and
extension of the concepts of knowledge creation
theory or “SECI-Ba” (Nonaka & Konno, 1998),
enabling an analysis of the knowledge transfer
practices and difficulties experienced. The article
is organized as follows. The next section presents
the literature review and theoretical framework
drawing on concepts derived from SECI-Ba. Subsequently in the research methodology section, we
explain the data collection and analysis approach.
Following this is the discussion and analysis of the
findings, and finally we present the conclusions
and limitations of the study.
THEORETICAL FRAMEWORK
The theoretical basis draws on knowledge creation
theory (Nonaka, 1994; Nonaka & Konno, 1998).
This theory presents knowledge as created and
transferred through a continuous dialogue between
tacit and explicit knowledge along four patterns of
interaction―socialization, externalization, combination, and internalization (SECI)―that create
a ‘spiral’ model of knowledge. Tacit knowledge is
broadly understood as intuitive and unarticulated,
while explicit knowledge is knowledge that can be
codified or documented (Nonaka & Konno, 1998).
The concept of ‘Ba’ refers to a “shared space for
emerging relationships,” which can be physical
(e.g., office, dispersed business space), virtual
(e.g., e-mail, teleconference), mental (e.g., shared
experiences, ideas, ideals), or a combination of all
three. For Nonaka and Konno (1998), knowledge
is embedded in Ba and acquired through one’s
own experience or reflections on the experiences
303
Managing Knowledge Transfer in Offshore Software Development
Table 1. SECI-Ba concepts
Socialization―Originating Ba
Exchange tacit knowledge through joint activities. Share feelings and emotions in
copresent meetings.
Externalization―Interacting Ba
Articulate hidden tacit knowledge through interaction. Express ideas in understandable forms.
Combination―Cyber Ba
Combine explicit knowledge through sorting, adding, combining, and categorizing.
Internalization―Exercising Ba
Convert explicit into tacit knowledge through learning by doing or training.
of others within Ba. The authors identify four Ba
types that correspond to the stages of the SECI
model: originating Ba (copresence), interacting
Ba (peer-to-peer), cyber Ba (group-to-group), and
exercising Ba (onsite). Thus, each Ba supports a
particular knowledge conversion process in the
SECI model, as summarized in Table 1.
Nonaka (1994) argues that tacit knowledge
cannot be formally articulated between individuals and can only be exchanged through joint activities. Originating Ba provides a physical context
for ‘socialization’, an existential space where
individuals can share feelings, emotions, and
experiences, hence removing barriers, as well as
increasing trust. Examples are group physical
activities or educational visits (Kostiainen, 2002).
Secondly, ‘externalization’ is the process of converting tacit knowledge into explicit knowledge
through reflection and dialogue where members
articulate hidden tacit knowledge (Nonaka, 1994).
Interacting Ba is the space that supports the process, typical examples being joint meetings, forums, discussions, memos, or other forms of dialogue (Kostiainen, 2002). The formation of Ba
can be encouraged by identifying an intermediary
or facilitator (Reich, 1995) to steer dialogue and
refine information. Thirdly, ‘combination’ is the
process of combining explicit knowledge through
sorting, adding, combining, and categorizing
(Nonaka, 1994). Cyber Ba provides a place of
storage, updating, querying, and interaction using
ICTs as encapsulated by integrated groupware
such as Lotus Notes. Finally, ‘internalization’ is
the process of converting explicit knowledge into
tacit knowledge through ‘learning by doing’ or
304
‘action’ (Nonaka, 1994). Exercising Ba supports
the process through focused training or on-the-job
learning by doing (Nonaka & Konno, 1998; Nonaka, Toyama, & Nagata, 2000). Shared bodies of
knowledge are internalized as “technical knowhow” (Nonaka & Konno, 1998). A limitation of
the theory as an appropriate conceptual lens for
our data is the lack of attention to knowledge
transfer between individuals or groups in distributed settings. Knowledge transfer between globally distributed software team members takes
place in shared electronic spaces (Carmel, 1999;
Carmel & Tija, 2005; Leonardi & Bailey, 2008),
which are supported by various types of ICTs such
as videoconferencing, telephone, text conferencing (e-chat), e-mail, databases, or via an integrated groupware application such as Lotus Notes.
However, the capacity of ICTs to transfer knowledge unproblematically is highly controversial
and has been subject to critique (Walsham, 2002;
Oshri et al., 2007; Leonardi & Bailey, 2008).
We extend the SECI-Ba model by drawing
on previous studies related to copresent and ICTbased interaction in software development. Table
2 shows the SECI-Ba concepts alongside findings from related studies showing strategies for
knowledge transfer in copresent and ICT-enabled
offshore software development settings.
In socialization-originating Ba, copresent
meetings at the start or during the project are
recommended by Carmel, (1999). Periodic copresent interaction provides a ‘rhythm’ that permits
intense focus during short intervals of remote
communication using ICTs (Maznevski & Chudoba, 2000). ICT-based interaction could provide
Managing Knowledge Transfer in Offshore Software Development
Table 2. Theoretical framework
SECI-Ba
Copresent Interaction
ICT-Based Interaction
Socialization―Originating Ba
Exchange tacit knowledge through joint activities.
Share feelings and emotions in copresent meetings.
Copresent meeting at the start or during
the project establishes understanding
between team members (Carmel, 1999;
Oshri et al., 2007).
Using text-based virtual realities
or posting static pictures during
e-chat establishes human connection (Robey et al., 2000; Towell &
Towell, 1997, 2001; Sarker & Sahay,
2004).
Externalization―Interacting Ba
Articulate hidden tacit knowledge through interaction.
Express ideas in understandable forms.
Using an intermediary to steer dialogue
and refine information (Reich, 1995).
Using team leader to refine information during online communication
(Sarker & Sahay, 2004).
Combination―Cyber Ba
Combine explicit knowledge through sorting, adding,
combining, and categorizing.
Gathering team members in a physical
place for preparation of documents
(Chua & Pan, 2008).
Posting documents and updating the
current set of documentation in the
groupware (Sarker & Sahay, 2004).
Internalization―Exercising Ba
Convert explicit into tacit knowledge through learning
by doing or training.
Sending experts to offshore location for
a period of 1-3 weeks to train the teams
(Chua & Pan, 2008).
Online training on new standards or
methods using groupware application (Deek & Espinosa, 2005).
‘virtual presence’ by members posting static
pictures and displaying names during synchronous
e-chat sessions in order to enhance social presence
and establish human connection (Robey, Khoo,
& Powers, 2000; Sarker & Sahay, 2004). The use
of text-based virtual realities such as description
of themselves, rooms, or objects (Towell & Towell, 1997, 2001) could enhance the feelings of
‘being there’ or presence. Oshri et al. (2007) stress
the importance of copresent meetings in socializing remote team members as well as the resocializing them throughout project lifecycle. Oshri
et al. (2008) further investigate how these globally distributed teams support the re-acquisition
of norms and attitudes over time. The main challenge they identify is that copresent meetings are
usually short and tend to offer only limited social
space that accommodates cultural differences.
However socialization can be developed and
sustained through various activities that take place
before, during, and after copresent meetings which
ensure the renewal of socialization using various
collaborative technologies.
In externalization-interacting Ba, there is
considerable literature on the positive effects of
copresent interaction by an intermediary: making
regular visits to offshore centers may facilitate
communication and refine information (Reich,
1995). Intermediaries (sometimes referred to as
“boundary spanners” or “straddlers”) can assist
in bridging distributed groups, coordinating tasks
(Carmel & Tija, 2005; Sarker & Sahay, 2004;
Waterson, Clegg, & Axtell, 1997), establishing
interpersonal relationships (Crowston, Howison,
Masango, & Eseryel, 2005), and developing social
ties between team members (Kotlarsky, Fenema
& Willcocks, 2008).
In combination-cyber Ba, copresent interaction
may help the process of preparing various documentation and forms of codified knowledge by
gathering resource persons who have the knowledge in one place (Chua & Pan, 2008). Sarker
and Sahay (2004) demonstrate how ICT-based
interaction may facilitate this process by making
the ongoing tasks visible to remote members by
posting detailed minutes of meetings or documents
held by team members of any location to online
databases: updating the current set of documentation or downloading documents that are required.
In internalization-exercising Ba, copresent interaction during the training by experts travelling
to the offshore location enables a large group of
offshore staff to be copresent and observe how the
software development work is carried out (Deek
305
Managing Knowledge Transfer in Offshore Software Development
Table 3. Summary of cases
Alp
Das
Founded
1997
1994
Employees
25-35
40-50
Headquarters
Manchester, UK
London, UK
Offshore Center
India
- Started a partnership with Indian companies in
1999
- Established a wholly owned software development center in 2002
Bangladesh and Malaysia
- Opened a wholly owned software development
center in Bangladesh in 1997
- Opened a wholly owned subsidiary in Malaysia in
2000
Major Product
Custom software development
Custom software development
Onshore/Offshore Work
Split
Sent different types of work including design to
offshore
Only sent coding to offshore
Communication Medium
Copresent meetings, telephone, e-mail, e-chat,
documents, and reports
Groupware application (Lotus Notes), telephone,
e-mail, e-chat
& Espinosa, 2005). Chua and Pan (2008) describe
how knowledge experts take turns travelling to
the offshore location to train the offshore teams.
The offshore team members are learning-by-doing
when they perform activities together, engaging in
active participation such as explaining, discussing, arguing opinions, and answering each other’s
questions. Deek and Espinosa (2005) posit that
internalization-exercising Ba may be achieved
during online training.
METHODOLOGY AND
CASE DESCRIPTION
We adopt an interpretive case study approach
(Walsham, 1995; Yin, 1994) carried out in two
UK-based companies of similar size engaged in
offshore software development. Both companies
were selected, as they offer some similarities and
differences that enable cross-case analysis. The
foremost similarity is that both are UK-based
software companies with offshore software development centers in Asia (Alp in India, Das in
Bangladesh and Malaysia) and their head offices
are based in the UK. Over the past decade, India
and Malaysia have emerged as a preferred location
for offshore outsourcing. According to a study on
306
the performance of the top four Asian offshore
countries, India will continue to lead the offshore
segment through 2010 with at least a 15% share,
while Malaysia is estimated to achieve 1.2% of the
global market share (XMG, 2007). Bangladesh is
still an untapped market for offshore outsourcing.
Table 3 displays a summary of both case study
companies.
Alp (a pseudonym) was established in 1997
with a total of 40 staff in the UK and India. During 1999, Alp entered into a partnership with a
local company in Mumbai for offshore software
development in order to take advantage of low
labor costs. The partnership with the small local
company called ABS in India lasted for almost
two years. Alp (UK) set up a wholly owned offshore software development center in India in
July 2002. In Alp, the type of software development work sent to India varies dependent on the
type of project. Typically, the onshore team in the
UK will perform the requirements elicitation and
some high-level technical architecture. The offshore team in India will undertake the design,
coding, and testing. Communication between
onshore and offshore involves copresent meetings,
telephone, e-chat, and e-mail.
The second case, Das (a pseudonym), was
established in 1994 with a total of 50 staff in the
Managing Knowledge Transfer in Offshore Software Development
UK, Malaysia, and Bangladesh. In 1997, Das
commenced operations in Bangladesh.. The offshore center in Bangladesh serves both the UK
and Malaysia, and the subsidiary in Malaysia
started in 2000 for local project delivery. Das
UK only sent coding work to Das offshore and
onshore teams to prepare requirements and design,
which are subsequently completed by the offshore
teams with time estimation and technical design.
The onshore and offshore teams use Lotus Notes
Groupware, which contains features such as messaging, calendaring and scheduling, e-meetings,
team room, and e-mail.
Fieldwork and Data Collection
Field research for both companies was carried
out between 2002 and 2005. The research began
with historical reconstruction from inception of
operations offshore and following the subsequent
operations during the period of the study. Data
collection used a variety of methods including
unstructured and semi-structured interviews,
documentation reviews, and observations taken in
the UK and during field trips to India, Bangladesh,
and Malaysia. Interviewing, repeat interviewing
and triangulation of sources over time using various techniques of data collection provided multiple
perspectives (Eisenhardt, 1989; Strauss & Corbin,
1990). There was no specific interview protocol,
but the first stage of interviews covered fact finding on company background, client background,
as well as their onsite and offsite interactions.
The second stage of interviews covered detailed
issues on projects and management in terms of
the processes, practices, and discussion of the
challenges encountered.
Documents such as functional specifications,
project plans, project reports, test cases, e-chat
between onshore and offshore teams, and other
documents related to software projects and work
procedures were reviewed. In addition, observations of team interactions and meetings during
projects were undertaken during the fieldwork in
the UK, Malaysia, Bangladesh, and India. Interview information collected at periodic intervals
was cross-checked where possible against other
sources of evidence such as company reports.
Data collection consisted of two stages, the first
stage being largely based on the interviews and
an external documents review, while the second
stage was based on the fieldwork in Bangladesh,
India, and Malaysia. The first stage of data collection was accomplished in the first quarter of
2002, and the second stage started at the end of
2003 and was completed in early 2004. The data
collection process included repeated interviews,
which are important in a longitudinal research
design to identify changes over time. Overall, there
were 111 total hours of interviews. Interviews in
Das involved 34 hours of interviews from January
to March 2002 and from December to February
2004 (summarized in Table 4). Interviews in Alp
involved 77 hours of interviews in June 2002 and
from August to December 2003 (summarized in
Table 5). Interviews lasted for at least one and a
maximum of two-and-a-half hours.
The process of data analysis involves coding,
theorizing, interpreting, and reporting. Data from
interviews, observations, and documents were
coded using the open coding technique, which
adopts a form of content analysis where the data
are read and categorized into concepts suggested
by the data (Strauss & Corbin, 1990). The data
are then re-examined and re-coded using the
proposed scheme where the goal is to determine
the set of categories and concepts that cover as
much of the data as possible. This iterative examination yields a set of broad categories and
associated concepts that describe the events,
experiences, and consequences. Once all the data
were examined, the concepts were organized by
recurring themes. The themes were categorized
and sorted according to their relevance to the
process of managing knowledge in offshore software development. This was followed with axial
coding to relate the main themes to the sub-themes
using stable categories to construct a more com-
307
Managing Knowledge Transfer in Offshore Software Development
Table 4. Details of interviews in Das
Number of Interviews
Number of Hours
2
6
Director, Das Asia
2
3
Offshore Project Manager
1
2
1
1
Director, Das Asia
1
2
Project Manager
1
2
Deputy Director
1
2
Country Manager
1
2
Project Manager
1
2
Project Leader
1
2
Senior Developers
4
8
System Administration
1
2
Total
17
34
January-March 2002
UK, Copresent interview
Director, Das UK
Malaysia, Copresent interviews
December-February 2004
UK, Telephone interview
Director, Das UK
Malaysia, Copresent interviews
Bangladesh, Copresent interviews
prehensive scheme. The emerging themes were
further analyzed in relation to related theory (Klein
& Myers, 1999). In the process of theorizing, the
goal was to seek a theory that was compatible
with the evidence and enabled explanation (Neuman, 1994). The process of relating the data to
theory involved a degree of ‘trial and error’ as the
researchers often had to refer to the related literature reviews during the process of data collection and analysis. Theoretical analysis involved
continuous reflecting on the data collected, interpretations, feedback from the case participants,
discussions with colleagues, evolving literature
reviews, and relating the data to relevant theoretical concepts (Nicholson & Sahay, 2001;
Walsham & Sahay, 1999). Finally, although not
part of an explicit action research approach, an
intermediate report was presented to the operations
308
manager in one of the case companies for further
explanation and verification.
FINDINGS AND ANALYSIS
This section presents specific episodes in Alp and
Das to illustrate the role of copresent interaction
and ICT-based interaction in managing knowledge transfer between onshore and offshore team
members. The analysis draws firstly from Alp
and subsequently the Das case. Episodes from
both cases are framed and explained using the
conceptual lens of SECI-Ba.
Managing Knowledge Transfer in Offshore Software Development
Table 5. Details of interviews in Alp
Number of Interviews
Number of Hours
Director, Alp UK
3
6
Operations Director
2
4
June 2002
UK, Copresent interviews
Principal Consultant
1
2
Business Consultants
2
4
Systems Administration
1
2
1
1.5
India, Conference call & e-chat interviews
Director, Alp India
Project Leader
1
1.5
Quality and Testing
2
2
Client
1
2
Director, Alp UK
3
6
December-February 2004
UK, Copresent interviews
Operations Director
3
6
Business Consultants
2
4
Chinese Translator
1
2
1
2
India, Copresent interviews
Director, Alp (India)
Project Manager
1
2.5
Project Leader
1
2.5
Senior Developers
5
10.5
Developers
3
6
Junior Developer
2
3
Finance
1
1.5
Administration
2
4
Systems Administration
1
2
Total
40
77
Case 1: Alp
Socialization-Originating Ba
Socialization took place in copresence between
onshore and offshore team members. When
commencing operations in India, Alp management brought Indian developers to the UK for
six months. During that period, the onshore and
offshore team members shared office and living
accommodations with UK staff. During their time
together living in the same environment, the teams
developed mutual understanding. For instance, one
of the offshore developers in Alp (India) tells the
story of how he had gained a greater understanding of UK norms:
309
Managing Knowledge Transfer in Offshore Software Development
I have learned a lot from the onshore team in the
UK, when we were there. Now, I can talk to the
team in the UK about so many things. We played
table tennis, squash and did Yoga with them. In
India, even if you know the person very well you
might not say ‘thank you’ or ‘please’ all the time.
Now, that has become common language for me
and the cultural difference is reducing.
Informal socialization enabled originating Ba
in the form of mutual understanding of mental
frameworks of local norms and for personal ties
to be fostered in community activity. However,
when the Indian team returned home, some difficulties endured in sustaining originating Ba
even after the period of copresent socialization.
In one episode after an early project failure, the
onshore team blamed the offshore teams’ ‘lack of
questioning behavior’ and their ‘low self-esteem’
when dealing with their English colleagues. One
of the team members in UK explains:
When you suggest something, they (Indian developers) often don’t question you. They just accept it
as correct. I also feel that there is some inferiority
complex that they have. They think that you are
an English person and you should know better
than they.
The offshore director in India explained that
this situation arose because Indians tended to be
relatively “more reserved and reluctant to pose
questions.” She explains further:
It takes some time for you to get into the relationship so that you can ask. Otherwise, people tend
to assume, because you think ‘well, what will the
people think if I go back and ask again?’
An explanation of this can be related to the
enduring effect of norms of status consciousness
in India. Once back in India, team members did
not want to give a ‘negative’ impression. Their
reluctance to interrogate instructions from UK
310
staff meant that incorrect assumptions were often
made which negatively affected outcomes. Chung
(2001) points out that such a cultural issue is
common between Western and Asian managers.
Western managers often assume that “direct talk”
and ‘getting to the point’ are the best way to negotiate, while the Asian manager prefers indirect
communication, absolute politeness, and reserve
with self-control. In India, social relations tend
to be hierarchical, and relative status is related
to issues of caste, which, although diminishing
in importance, still have effect on perceived
perceptions in social pecking order (Nicholson
& Sahay, 2001). This finding points to the limits
of socialization in sustaining Ba, the need for and
enduring impact of social structures affecting
knowledge transfer.
Externalization-Interacting Ba
Since the cost of moving large numbers of team
members around the globe is considered cost
prohibitive, Alp assigns a project manager to act
as an intermediary. The travel of this person is
bi-directional, for example at the early stage of
the project the UK-based onshore project manager
will visit India for a period (usually one or two
weeks) to interpret the requirements of the project
to the offshore teams. On occasion, an offshore
project leader will travel to the UK to facilitate the
India team’s understanding of client requirements.
An onshore project manager in the UK explains:
We bring a systems analyst from India to the UK.
We go together to the customers and try to understand the requirements. Then he takes all the
requirements back to India. Since he is Indian, he
can explain better to the offshore teams. Sometimes
I go to India to explain the requirements.
This finding concurs with previous research in
that the intermediary was responsible for answering queries from clients, onshore and offshore
teams, disseminating information and checking
Managing Knowledge Transfer in Offshore Software Development
progress (Reich, 1995; Sarker & Sahay, 2004).
However, intermediaries were not always able to
create the conditions of interacting Ba and externalize tacit knowledge. This is demonstrated in an
episode involving Jeevan, a UK-based intermediary between client and offshore teams. Jeevan had
difficulty externalizing knowledge to the offshore
team as he did not hold the technical knowledge
to communicate the client requirements to the offshore teams using technical terms that the Indian
team would understand. The consequences of this
are described by one offshore developer in India:
Most of the time, we have to assume. Some developers felt that they didn’t have enough information about the system because questions were not
adequately answered.
Jeevan, as intermediary between client and
offshore team, ought to have been able to make his
tacit knowledge of the client requirement explicit
in a meaningful technical specification that could
be understood by the offshore team. However, the
offshore team felt he was insufficiently technically trained and inexperienced to perform the
externalization. Pratap, an India-based project
leader, explains:
The person in the UK (onshore team) should understand what the client wants and communicate
the same line of information down to the Indian
team. The very essence of understanding is the
ability to explain with data, to foresee the possible design or coding and to communicate it to
the offshore teams.
Nicholson (1999) describes how an intermediary in similar circumstances had difficulty externalizing knowledge due to a lack of contextual
knowledge of UK social security benefits, an
application context unknown to many Indians.
India has no equivalent concept of social security
unemployment or housing benefits, and thus the
Indians had no common understanding with their
UK counterparts of the functions and terminology
of such as system. The UK intermediary was not
aware of the gaps in the Indians’ knowledge when
documenting client requirements into specifications for the Indians, and thus misunderstandings
occurred.
Combination-Cyber Ba
A central practice for knowledge combination
at Alp involves preparation of detailed progress
reports at stages in the project. The report is
signed by the client in a copresent meeting and
subsequently posted in an online database. It is
intended to provide “synchronization” between
offshore, onshore, and the client. A project leader
in India explains:
We have to prepare a project plan, at the end of
every important deliverable and milestone there
is a review and approval process. If the deliverables have been completed, it will be reviewed
and certified by the project leader, the project
manager in India and the project manager in
UK. They will meet a customer face-to-face to
sign off the deliverables. If the customer has any
issues he should raise it before we can continue
to the next stage.
The client felt uncomfortable with the formal
process of reporting:
I felt that it was just not necessary to do it because
what I need is just some informal reporting and
for them to treat me as their team leader and not
as a customer so that there is not so much time
wasted on reporting. I don’t have too much time
to read their reports.
The quote indicates that there was a tension
between the formal knowledge combination
mechanism and the client’s wishes. Yates and
Orlikowski (1992, 1999) posit that the use of
document as genre should be created based on
311
Managing Knowledge Transfer in Offshore Software Development
substance or topics and should focus on the physical and linguistic feature of the communication.
The document was considered inappropriate by
the client, who did not read it as he wanted to be
treated in the same manner as an internal team
leader and updated informally rather than as a
distanced, formal client-vendor relationship.
The episode underlines the importance of shared
mental frameworks in creation of Ba. The practice
of project reporting physical and virtual arrangements alone did not create Ba. Mental frameworks
around the client’s role and position in the team
were not taken into account, meaning that Ba was
not created and combination not achieved.
Internalization-Exercising Ba
Internalization involves mentoring of Indian staff
and Ba created through bi-directional travel, regular telephone, calls and e-chat between onshore
and offshore project managers. One of the onshore
project managers in the UK explains:
Project management is not something that you
just follow a guide to. It is extremely difficult and
needs experience. So giving the Indian teams a
process is not enough. We have to give them a
mentor; we need to give them someone who is
really experienced who actually can be a mentor
to them.
The process of reporting, review, and approval
is designed in such a way to enable ‘learning by
doing’ in virtual Ba. An example is provided by
one of the mentors:
When a document is produced in India, it is produced in Indian English and the grammar is not
always good. When I received a document from
India, I will meticulously go through it, change
the grammar, and send it back to the person with
a copy to the team leader. The purpose is to make
sure they learn their mistakes. In practice, the next
version has the same English errors. We don’t have
312
any sort of formal way of assessing individuals’
English and I think we really need to do that.
Other approaches use copresence; for example
an episode during the early stages of software
development involved an offshore team leader
travelling to the onshore location to be trained
by a mentor. The offshore team leader had the
opportunity to sit next to the expert and engage
in and observe how work is performed. The team
leaders later went back to the offshore location
and trained other team members. Any questions
raised after return were handled through regular
telephone calls and e-chat. These findings concur with previous research on the importance of
one-to-one on-the-job training for internalization
(Deek & Espinosa, 2005) and active participation
(Chua & Pan, 2008) for knowledge transfer.
Case 2: Das
Socialization-Originating Ba
Das does not engage in copresent interaction
and instead relies wholly on ICT-based interaction. Originating Ba is created in the team room,
which is one of the features built into Lotus
Notes Groupware. It provides a shared space for
team members to socialize and interact using the
discussion forums and e-chat, and by placing
personal information online (e.g., hobbies, etc.)
to establish social presence (Robey et al., 2000;
Sarker & Sahay, 2004). Text-based interaction
overcomes disadvantages of verbal communication being immediately lost. An offshore team
member explains:
We feel more comfortable using e-chat because
we will have more time to think and we can document it. We will put the chat conversation in the
team room so that other team members can refer
to it later.
Managing Knowledge Transfer in Offshore Software Development
Das requires team members to follow standard
communication procedures to overcome the lack
of non-verbal confirmations which could be encountered in copresent interaction. The director
of Das (UK) told us:
We develop methods and standards…Whenever we
discussed something, we (onshore team members)
would get them (offshore team members) to confirm
that they understood or agreed. When they didn’t
say anything, we would ask. ‘Did you understand
it, did I explain it right and are you happy with
it?’ Whatever it is, we always get them to agree.
If we have a very long discussion, we always
summarize what we agreed in the course of that.
An example of problems in maintaining Ba is
shown in the functional requirement specification
(FRS). This is a key document that must be clearly
understood by the offshore teams for accurate
coding. The offshore teams are not regarded as
‘good’ in English, and there were occasions when
they experienced difficulty in interpreting the
FRS and thus the onshore teams must use “simple
language.” One of the onshore project managers
told us that she had learned to communicate in
e-chat for clarifications as if she was “talking” in
a copresent setting:
When I type, I don’t use text abbreviation so
much. Literally, the words I use through e-chat
will be the same words I use when I talk to the
offshore teams face-to-face. It doesn’t inhibit me
and I am quite happy. I want to have a personal
connection, so I try to foster the idea that when
we are talking online we are doing it as though
we are face-to-face.
The techniques and rules around standardization, use of clear language, databases in the team
room, and “talking” in e-chat facilitate socialization in originating Ba.
Externalization-Interacting Ba
The venue for externalized knowledge is the Lotus
Notes team room, which forms the interacting Ba.
It is here where stores of discussions related to
the project are held. For example, onshore teams
will prepare the functional requirement specification and later send this to the offshore teams for
technical design and coding. The externalization
process involves intensive discussions over the
FRS in the team room. The same procedures
apply for project delivery from the offshore to
onshore teams. One of the offshore developers
in Bangladesh explains:
In the team room we can check the documents and
discussions related to the project. We can track who
is doing what, what has been done on which day,
what are the errors and whether they have been
repaired. When we deliver the project, we zip the
document in the database and the UK team will
unzip and open the document from their database.
If there is a bug, the onshore teams will create a
document in the team room. We will respond to
them and if it is really a bug we give reasons and
repairs. If it is not a bug, we will explain and give
structure on how it works. At this stage, we are
guiding them on how to repair the bugs properly.
Records saved in the team room allow all
authorized employees to have access to the latest
versions of documents. The onshore and offshore
team members are linked together and connected to
the team room as a shared knowledge base. One of
the senior developers in Bangladesh describes this:
In the team room we can check the document and
the discussion related to the project. If we chat on
e-messenger, even a single line, we will put it in
the team room. We can track who is doing what,
on which day, what has been done, what the errors are and whether it has been improved or not.
The team room shows project steps, discussions,
and deliveries.
313
Managing Knowledge Transfer in Offshore Software Development
Over time, the offshore teams became comfortable with the use of the groupware application for
communicating technical knowledge. In addition,
the use of the medium supported and integrated
knowledge as well as encouraged learning between
users (Ciborra & Patriota, 1998). For example,
when someone in an onshore team wants to
communicate about ‘bugs’, he or she must only
create a document in the team room and commence discussions. However, a team leader was
needed to act as facilitator between onshore and
offshore team members to facilitate the process
of interacting Ba.
Similar evidence was found in the Alp case in
relation to the questioning ability of the offshore
team. Clara, a UK director, explains:
It will be a painful process to make them say
something.
Bangladeshi developers revealed that they felt
unable to question the onshore teams because they
did not want to feel ‘discredited’ for asking questions. Consequently, they tried to avoid negative
perceptions from the onshore teams that questioning meant they had no cognizance. According to
one offshore team leader in Bangladesh:
‘sometimes it is this’, instead saying ‘sometimes
it is this because of that’, which we don’t necessarily have to explain to other people.
In reaction to the perceptions of relative status,
team leaders were encouraged to apply a “parentstick game.” Clara, the UK director, explains:
There is high value placed on success especially
in Bangladesh. So, you can play this ‘parent-stick’
game quite a lot. When you praise them, [for example] ‘that is really fantastic’, they really take
it up, and equally if you tell them they are bad,
they really take it quite seriously. I pretended to
be disgusted, disappointed and I really played
on their emotional response to feel as though it
is very difficult to manage when they did something wrong.
Combination-Cyber Ba
In my early years with Das, as an inexperienced
developer, I took more time to understand the FRS.
We did not ask many questions, even if sometimes
we did not understand the FRS. We used to think
that if we asked, it might be a ‘discredit’ to us.
At Das, combination takes place when knowledge
of a specific project is documented and stored
in the ‘team room’ along with e-chat and e-mail
records. The use of templates and standard forms
is considered important by both on- and offshore
groups. However, there is the problem of ‘how
much and what’ is to be documented to avoid
information overload. The problem has been
resolved over time and Ba in the evolution of
key documents. According to the Das offshore
project manager:
Thus, in reaction to this the director in the UK
told us that onshore team leaders are required to
make their “assumptions clear and to express
knowledge that might otherwise be assumed
to exist between co-present individuals.” One
onshore team leader in the UK explained further:
In the beginning, UK people tested our work and
then sent us a list of errors including silly mistakes.
Now, we have to prepare test scripts or procedures
and have a process to follow. As a result, UK
people don’t have to test so much. Every delivery
contains documents and installation procedures.
…we (onshore teams) learn how to talk with the
Bangladesh team…we know that we have to explain things in a particular way: rather than saying
314
Managing Knowledge Transfer in Offshore Software Development
Internalization-Exercising Ba
To enable internalization, Das performs online
mentoring and training. In the early years since Das
was formed, onshore teams personally coached
the offshore teams and ensured they learned from
their mistakes. One of the offshore team leaders
explains:
Previously, when we delivered it, they used to
re-check. If there were bugs, they used to re-do it
themselves. After some time, they sent the errors
back to us and asked us to do the correction. We
did the correction and we sent it back to them.
The process went on until it was perfect before
delivery.
This ‘one-on-one’ online training was regarded
as much slower than that would be encountered
in copresent settings. This was due to the time
taken to explain, send, re-explain, and re-send
until the explanation became clear. This results
in delay compared to the instantaneous nature of
co-present feedback. The process is also more
difficult according to a UK-based manager:
When you try to do things online, you can make
progress but ultimately teaching them down to
absolute detail is a lot harder compared to doing
it face-to-face.
However, over time, the offshore teams’ ability
to internalize their understanding of client requirements and processes improved as the communication with the onshore teams developed. Thus, the
exercising Ba was not created instantaneously but
gradually through experience. One of the longer
experienced offshore team leaders explains:
In the early years we used to make mistakes quite
often. We often misunderstood the requirements
because of: (1) we could not understand the client,
(2) we could not understand the process, and (3)
we lacked experience. As time passes, we are get-
ting better and now we can visualize the scenario
because now we have more experience and can
communicate better with the onshore teams to
visualize the whole system.
To train the offshore team, the onshore project
manager asks the offshore teams to try several
solutions even though she knew from experience
the best solution. As explained by the offshore
senior developer:
She (UK manager) has a lot of experience in
programming and she will suggest some other
solution, saying ‘you may try this’, but I always
find that her suggestion is actually the best solution. She never says that it is the best solution
and asks me to try other solutions too. I will try
other ways but after comparing with her solution,
I usually agree with her solution.
Although the project manager who acts as a
mentor had direct answers to the questions, she
would encourage the offshore teams to think about
and try several solutions before reaching a decision as to which solution was more suitable for
coding. Taylor (1990) has stated the importance
of considering variation between cultures to avoid
misinterpretation of knowledge being communicated. The techniques of asking the offshore teams
to think and try several solutions was deemed to
encourage creativity of the offshore team.
Table 6 summarizes the findings showing the
role of copresent interaction and ICT-based interaction for managing knowledge transfer between
globally distributed team members.
DISCUSSION AND CONCLUSION
The Role of Copresent Interaction
in Knowledge Transfer
Our first research question was to understand the
role of copresent interaction in the knowledge
315
Managing Knowledge Transfer in Offshore Software Development
Table 6. Summary of findings
Alp
Das
Socialization–Originating Ba
Develop understanding during copresent interaction
at the early stage of relationship.
Set guidance for ICT-based interaction especially
at the early stage of relationship.
Externalization–Interacting Ba
Use intermediary to allow copresent interaction.
Use team room features for ICT-based interaction.
Combination–Cyber Ba
Gather team members with knowledge for document/
report preparation.
Online documentation using team room features
and standard templates.
Internalization–Exercising Ba
On the job training―Bring offshore teams to the UK
or send a project manager to India to train offshore
teams. Learning by doing.
Online training, mentoring, and coaching.
transfer process. We found that the copresent
interaction plays important roles in managing
knowledge transfer. Firstly, copresent interaction
allowed the development of mutual understanding between team members. The onshore and
offshore teams in Alp were colocated in the UK,
allowing them to socialize through formal and
informal activities which developed understanding
of each other’s mental framework. However, the
creation