THE IMPLEMENTATION OF STRUCTURED METHODOLOGIES
AS A PRODUCTIVITY TECHNIQUE IN SYSTEMS DEVELOPMENT
by
Walter J. Popper
B.A.
Swarthmore College
(1966)
Ed.D.
University of Massachusetts, Amherst
(1982)
Submitted to the
Sloan School of Management
in Partial Fulfillment of the
Requirements of the
Degree of
MASTER OF SCIENCE
IN MANAGEMENT
at the
MASSACHUSETTS INSTITUTE OF TECHNOLOGY
May 1983
ý(
Massachusetts Institute of Technology 1983
Signature of the Author
Slc(rVSchool of Management
May 19, 1983
/
Certified by
,/Dr. Dorothy Leonard-Barton,
Thesis Supervisor
Accepted by
Jeffrey A. Barks
Director of Master's Programs
Archivo'
MASSACHUSETTS INSTITUTE
OF TECHNOLOGY
JUN 27 1983
LIBRARIES
THE IMPLEMENTATION OF STRUCTURED METHODOLOGIES
AS A PRODUCTIVITY TECHNIQUE IN SYSTEMS DEVELOPMENT
by
WALTER J. POPPER
Submitted to the Sloan School of Management
on May 19, 1983 in partial fulfillment of the
requirements for the Degree of Master of Science in
Management
ABSTRACT
study of the implementation of structured systems development
A
methodologies by analysts and programmer-analysts was conducted at three
sites in a large industrial firm. A sample of 145 was selected and
A questionnaire was used to obtain information on the
interviewed.
extent of use of the methodologies, and perceived advantages and
disadvantages and the reasons for using alternative approaches. Data on
the background, attitudes and perceptions of the analysts was gathered.
implementation of these methodologies is
The results indicate that
related to specific organizational, individual and opportunity factors.
The degree to which supervisors and clients support use of the technique
is related to the analyst's decision to implement it, for the structured
analysis methodology but not for the structured programming methodology.
Similar relationships were observed between analysts' attitudes and
technical orientation and extent of use. Finally access to training, as
an opportunity variable, was related to use of structured systems
analysis.
Recommendations were made for the more effective implementation of the
They included
methodologies in the systems development organization.
improving the measurement and monitoring of use and matching of access
to training with expected extent of use.
Thesis Supervisor:
Dr. Dorothy Leonard-Barton
Title:
Professor of Management
TABLE OF CONTENTS
Chapter 1: Introduction
.....................
The Need for Productivity Improvement .......
Purpose of the Study ........................
Design and Hypotheses .......................
Definitions .................................
Chapter 2: Review of the Literature
.........
Individual Implementation ...................
Information Systems Productivity ............
Chapter 3: Methods
..........................
Population ..................................
Questionnaire ...............................
Data Analysis ...............................
Chapter 4: Implementation
...................
Extent of Use ...............................
Access to Training ..........................
Benefits of the Methodologies ...............
Reasons for Using Alternatives ..............
Chapter 5: Factors Related to Extent of Use
.
90
Organizational Support .......................
Analysts' Atti t udes ..........................
Analysts' Ability, Experience and Orientation ..............
Opportunity Factors ........................................
101
109
.....................
113
Findings ...................................................
Recommendations for More Effective Implementation ..........
Recommendations for Future Research ........................
113
117
120
..................................
128
Chapter 6: Summary and Recommendations
Bibliography and Appendix
98
INTRODUCTION
years
the
with
in recent
concerned
Corporate management has become increasingly
of productivity improvement techniques
implementation
the
of
Because
among professional and white collar workers.
growing
importance of information systems organizations in most large firms, and
because
increasing
of
costs
of
improvement among systems analysts
interest.
in this
labor
and
productivity
area,
is of
programmers
particular
This study is an examination of the use of structured systems
development
The
methodologies as a productivity improvement technique.
introductory chapter provides an overview of the study in four sections.
The
describes
section
first
information
systems
problem
the
organization
in general
section, the purpose of the study is stated.
in
productivity
of
terms.
an
In the second
Section three outlines the
design of the study and sates the hypotheses to be
tested.
The
final
section defines the terminology used.
The Need for Productivity Improvement in Systems Development. Large
corporations have been increasing their expenditures for electronic data
processing steadily ever since the first
were
installed.
As
expenditures
have
dollars and in proportion to other costs of
managers
computing
electronic
increased,
doing
both
business,
systems
in absolute
corporate
have become increasingly concerned about the efficiency of the
5
been
an
result
One
information systems organizations within their firms.
has
increased interest on the part of information systems managers
in establishing
modern
management
to
practices
the
that
insure
system organizations for which they are responsible provide
information
the most effective services for the least cost.
Several developments over the last fifteen years have resulted in a
focus of these concerns in the area of labor productivity in information
development
systems, a and in particular in the productivity of systems
professionals
responsible for analyzing business problems and designing
These
computer-based information systems to help solve these problems.
professionals
are
the
systems
analysts
and programmer-analysts, the
technical people who develop new software for a firm's computers and who
design and implement improvements in existing software.
In large corporations,
systems
information
group,
departments in the business.
and
systems
analysts
The
latter
applications
are
packages
systems,
programs
analysts.
These end users or clients initiate
contract
for
on
services
providing
the
in the
work
generally
to other
request
end
users
developed
requests
for
of
by
systems
services,
these services and pay for the services provided.
products and services of the information systems organization
the
are
As the
ever
more widely used within the firm, costs of information processing become
more visible to management and concern with productivity in this area is
more widely distributed.
distribution
Meanwhile technical developments have changed the
costs
As computer hardware has become more
within information systems.
inexpensive to manufacture the costs of processing time and memory
to
relative
decreased
the
and
workers
complicated
increasingly
with the increase of
to
productivity
to
the
labor.
labor
costs have
of
trained
technical
to
perform
to
required
training
necessary
tasks
Unit
shortage
of
have
sophisticated
develop
result of these developments is that a manager concerned
The
systems.
of
level
higher
the
cost
due
simultaneously increased
of
productivity
must
more
look
labor
at
closely
whether the output per unit of labor input can be
see
increased.
the
underlying
change
Another
is the gradual change over time in the content of systems
productivity
development work from the development of entirely
and
enhancement
As
development.
programs
was
systems.
proportion
1981;
some
improvements
the
supplements
have
been
In the process, managers
maintenance
oversights on the part
original
in new
to
previously
spending
an
requests
developed
increasing
their time on this type of activity (Synnott and Gruber,
Zmud, 1980).
of
and
Analysts and programmers
of
the
implemented and software applications
were
packages adopted end users returned to systems developers with
changes,
to
When computers were
of existing systems.
maintenance
systems
new
first introduced in business, all systems development work
for
labor
with
concern
increased
product.
of
They
have
become
aware
that
and
enhancement work results from errors or
the
systems
developers
who
designed
the
attempt to improve the quality of work in the
initial stages of development in hopes that the need for subsequent work
Errors caught in initial development can avoid
will be reduced.
costly
maintenance later in the life cycle of a system and thus improve overall
productivity of the information systems operation.
One aspect of the problem which is particularly troublesome is the
of measuring productivity in systems development.
difficulty
common problem in managing
white
and
technical
collar,
professional
The industrial
in general and knowledge workers in particular.
workers
approach to productivity measurement, to calculate a ratio of
in which the inputs and outputs are often
of
intangible,
unobservable
or
The debate in systems development for years has been
to quantify.
hard
units
to units of factor inputs is difficult to apply to work
output
product
This is a
whether the one easily quantifiable output, lines of code in a program,
measure
legitimate
is a
worker.
of
the
of the information systems
product
Without a meaningful, measurable output
a
calculate
productivity
ratio
it is impossible
to
no matter how precisely the inputs of
labor, machine time or other factors can be determined.
An alternative in this situation is to address the problem
of
improving
to
the
firm
The
(Crawford,
of
output
is that
what
is the quality of services provided to clients,
1982).
Effectiveness
meet
in the
the
needs
of
the
can be a particularly useful
concept for long-lived, complex products in that
measures
here
argument
their timeliness and the degree to which they
client
one
effectiveness of the information systems operation rather
than improving just its efficiency.
matters
as
it suggests
multiple
environment in which the product is used,
potentially including measures across a range of
the
across
and
uses
life of the product.
As information systems managers have addressed the
of
efficiency
productivity,
individual
process
the
on
focussed
of
systems
development
of
of
role
the
and
Since the
analysts and other professionals in this process.
late 1960's the literature has provided evidence
most
they have necessarily
effectiveness
and
issues
related
the
with
concern
organization of the work process, project management,
appropriate
process
Systems
innovations.
time
and
cost
estimates
and
development
has
been
described
more frequently as a process in which
labor
work could be divided into well-defined stages, and
have
methods
been
tried for organizing and managing the work at each stage.
One set of innovative
systems
which
methods
has
emerged
This methodology is based on the assumption that
development.
increased
the effectiveness of analysts and programmers can be
development
degree,
process
be
can
a
providing
standardized
approach,
common
Standardization
representation.
is structured
and
if the
and engineered to a greater
language
graphical
and
engineering, as well as language
and graphics, are seen as helpful in improving communications both among
analysts and between analysts and their
also
presented
as
improving
clients.
The
methodology
the effectiveness of systems development
management by providing managers with a framework within which
estimates,
assess
inputs and results.
address
the
is
to
make
progress, evaluate teams and individuals and compare
In short,
productivity
structured
problem
by
methodologies
improving
promise
information
to
systems
effectiveness while at the same time establishing a framework
in which
effectiveness can be analyzed and perhaps measured.
Purpose of the Study. Structured systems development
have
adopted
been
information
by
systems literature, along with
These reports are
improvement.
studies
productivity
of
adoption,
The reports analyze adoption as an event at the level of
Mendes, 1980).
the organization, and discuss
They
organization.
of
implications
the
the
for
ad option
do not specifically examine the role of individual
adoption
analysts in the
or
decision
in the
subsetquent
decisions
the use of newly adopted methodologies in the work of systems
development (Goldstein, 1982).
previous
complement
decision
t:o
approaches
case
these
advantages
methodology and its benefits for the ore;anization (e.g.
the
describing
primarily
the
of
descriptions
alternative
over
off er
methodologies
to
the
of
Reports
and adoption of these innovations appear in the information
development
regarding
systems departments in a number of
Goldstein, 1982).
large organizations (Johnson, 1977;
methodologies
to
implement
It is the purpose of the
reports
structured
current
study
by examining the indi,vidual analyst's
once
methodologies
the
official
adoption decision has been made at the policy level.
Information systems organizations
educated
staff
decision making.
and
are
characterized
employ
by
a
professional,
substantial
highly
dispersion of
Analysts make many of the decisions about the
way
in
which they work, the tools and techniques they use, and the way in which
their
In particular, in applying structured systems
work is organized.
have
analysts
methodologi es to the work of systems development, the
a
great deal of discretion in the extent of use of the methodology and the
determinati on
instead.
official
organizational
approach,
analysts
have a certain 1atitude in interpreting
in
development as in other
innovation
by
around
getting
ways
other
areas
professional
of
is only
organization
the
work
an
of
adoption
the first step in effective
use.
guarantee
not
It does
innovation.
that
implementat ion of
In systems
policy.
official
tasks,
the
the guideli nes, altering the situation or the def inition of
and
a
specify
guidelines
when
Even
particular
situations in which they will use an alternative
the
of
Effective implementati on is of particular interest to management because
it
crucial
is
to the
methodologies are form ally adopted
adopted
generate
methodologies
structured
of
success
without
being
If
methodologies.
if
or
implemented,
effective resistance among analysts in
An
their daily work, adop tion will not result in the desired outcomes.
of
understanding
process
implementation
the
at
the
individual analyst is a necessary pre-requisite to effective
the
of
level
management
of this change in the information systems organization.
There is the poss ibility, too, that individual analysts
some
insight
structured
into
the
methodology,
strengths
arising
and
from
weaknesses
their
of
will
have
a particular
experience
with
that
methodology in their work, and this information might be useful to those
responsible
for the further refinement of the innovation and subsequent
changes in the ways in which it is used by the organization.
There
is
(Mendes,
evidence
that
1980)
experiences
analysts
been
have
used
intentionally by methodology development groups in some organizations in
just this way.
study
This
implementation
is designed
of
understanding
provide
to
the
of structured methodologies by analysts in order to heJp
The study is
managers improve the effectiveness of systems development.
1) to
objectives:
following
organized to achieve the
describe
the
implementation of a set of structured systems methodologies
micro-level
by systems analysts across a large information systems organization in a
multi-national environment;
selected
2) to
personal
organizational,
the
analyze
opportunity
and
extent of use of these methodologies;
relationships
between
variables and the
3) to summarize and
analyze
the
reasons given by analysts for use of alternatives to these methodologies
at
stages
various
of
the systems development life cycle;
4) to make
recommendations for more effective implementation of structured
methodologies
systems
in this and similar settings, with the goal of improving
the overall effectiveness of information systems organizations in large
firms.
The study is based on
methodologies
themselves,
environments in which the
assumes
that
them by
their
assessment
of
the
a
the
number
of
users
methodologies
of
assumptions
these
are
related
to
the
methodologies and the
implemented.
First,
it
methodologies have the characteristics attributed to
developers
and
their
users.
The
study
is not
an
these techniques at a technical level but an examination
of the decisions leading to their use and the factors in the environment
It is taken for granted
related to their use.
are
they
function
as
methodologies
in some situations and for some tasks, and that
least
at
useful,
the
that
described
in the
as
and
literature
by
stated
participants in the study.
Secondly, it is assumed that the existence of
the use of these methodologies in the organization was not in
regarding
itself sufficient to guarantee full and effective use.
who
constitute
to
opportunity
methodologies
make
study
in the
subjects
the
assumed to have the
they
their own work
training.
The
the
of
application
the
to
related
decisions
are
individuals
The
It is further assumed that
in specific work situations.
the subjects accurately responded to questions and that
analysts
guidelines
official
as
experienced
were in a position to comment on their own decisions and
as
well
their
on
as
skill
and
technical
previous
responses of analysts, while to some extent subjective,
are a unique source of data related to the
and
process
implementation
assumed valuable in this regard.
Finally,
methodologies
it is assumed
and
that
There is ample evidence of
resources
devoted
to
that
management
the
intends
organizational
the
use them effectively.
to
support
amount
in the
testing,
development,
these
values
organization
revision
implementation of the methodologies over the past eight years.
of
and
There is
additional evidence in the resources devoted to the study in the form of
the time of a large number of
individual
analysts
training, upon
the
employees.
is assumed
approval
of
to
the
take
Organizational
the
analysts
form
support
for
of assignment to
immediate
supervisor.
official support is evident in the mention of the methodologies
Further
in guidelines, memos and other written material.
large organization and
of
between
relationships
gathered from analysts.
are
use
to
and
variables
opportunity
and
believed
Data are analyzed in the
among
framework of hypotheses suggesting relationships
individual
selected
of these methodologies by individual
use
Data on the use of the methodologies and on factors
analysts.
related
the
asses
to
extent
the
and
the
of two structured systems development methodologies in a
implementation
factors
describe
to
is designed
study
Design and Hypotheses. This
extent
organizational,
of
use
of
the
methodologies.
of
The dependent variable in this study is use
for use by the applications development
in
analysts
that
firm.
The
similar in that they are both process innovations in
are
methodologies
methodologies
computer technology group in a Iarge industrial firm
the
by
developed
two
the organization and in that they are based
need
the
for
on
a
structured,
a more
common
philosphical
more engineered work
approach
to
process.
They differ, however, in that each is appr opriate for use at a
stage
different
Methodology
in the
sequence
of
systems
development
1 is a technique for improving the prod uctivity of analysts
in the early stage of development, when the initial exploration
business
operation
leads
and
of
the
to an assessment of alternative solutions to
the problem posed by the client.
client
tasks.
At this stage communication
with
the
accurate representation of the problem are central concerns
of the analyst.
Methodology 2 is a technique for improving productivity
toward the end of the development sequence, at a stage at which analysts
At this
and programmers design, code and document the computer program.
stage
the
use
of
common
and
logic
and
notation
the
clarity
of
relationships among the many components of a complex program are central
concerns.
as
The methodologies, while used
self-contained
techniques,
are
conceived of as two parts of a system of techniques intended by the firm
to
structure
the
system
(for
example,
an
and
methodology
entire development process.
the
a
logical
structured
computerized
on-line
The other components of
design
data
design
base
not be
will
tool)
addressed in the present study.
These
innovation
two
methodologies
officially
adopted
discretion of individual analysts
by
firm
the
and
a
comprise
together
taken
but
programmer
process
at the
implemented
analysts
and
their
These analysts are the universe of subjects to be sampled
supervisors.
in the study.
Implementation is assumed in this model to be influenced
major clusters of variables:
variables.
by
three
organizational, individual and opportunity
Within each cluster a few representative measures have been
selected for this study to represent the whole
cluster.
The
measures
were selcted on the basis of extensive open-ended conversations with the
developers
of
methodologies.
degree
to
the
methodologies
and
They were chosen according
which
each
represented
related to extent of use and
2)
the
analysts
with
to
two
who
criteria:
use
the
1) the
a variable thought to be strongly
degree
to
which
analysts
were
thought
to
have
the
information.
necessary
organizational,
The
individual and opportunity clusters are discussed
briefly
below.
The
selcted measures are explained in more detail in Chapter 3.
Organizational factors represent the influence of
on
the
implementation
of
methodologies.
the
the
environment
In particular,
the
as
the
attitudes and values of key people in the organization as
well
specific resources provided by the organization are included here.
respect
to
systems
analysts,
key people in the organization are
the
those for whom these analysts regularly perform tasks and those who
the
users of the applications products the analysts produce.
supervisors and clients.
With
are
These are
The resources provided by the organization are
those training, consulting and support activites which together comprise
the
efforts
organization's
their
encourage
use
by
to
disseminate
methodologies
the
and
One tangible resource, evidence of
analysts.
organizational support, is the written guidelines for the use of systems
development
methodologies
and
techniques,
in which
both
of
the
structured methodologies are included.
Individual factors which may be related to use of the methodologies
are grouped in the model as attitude and
ability
variables.
Attitude
variables include the attitude of individual analysts toward their jobs,
toward
the idea of structured analysis and programming methodologies in
general, and toward Methodology
Ability
variables
include
I
and
Methodology
2
in particular.
measures of past training and experience in
systems development, skills in various
systems
development
and orientation to technical or non-technical careers.
techniques
Opportunity factors, in this model, are
the
type
characterize
which
those
tasks analysts are engaged in as well as the implied need
of
for productivity improving methodologies inherent in the organization of
These factors are an attempt to explore
these tasks.
the
between
fit
task-related needs and the capabilities of the two methodologies, and to
test
the
assumption that decisions to use a methodology are related to
the strength of the need for that methodology.
The hypotheses are stated and explained below.
1. Extent of use is related to organizational support.
will
1.1 Client attitudes favorable to the use of the methodologies
associated
greater
with
of use.
extent
individually or as members of work
providing
a
service
to a client.
teams,
product
receives
and
professionals,
and
will
uses
are
modify
or
service,
the
end
Analysts perform their work,
for
the
goal
ultimate
of
In the context of the organization,
the client contracts with the information
specific
be
generally
product.
systems
organization
for
a
an applications package, and
Analysts,
just
like
any
sensitive to the needs and desires of their clients
their
attitudes of the clients.
behavior
in accordance
with
the
perceived
Supervisor attitudes favorable to the use of the methodologies will
1.2
be
with
associated
to
supervisor
analyst
of use.
extent
greater
is direct
and
of
relationship
The
the consequences of supervisor
preferences are likely to include changes in the
expressed
preferences
and in the behavior of analysts.
1.3
General
organizational
attitudes
consistent
with the use of the
The
methodologies are likely to be associated with more extensive use.
perceived
upper
favorable
level
managers
attitudes
of
the
of
work
systems
group peers as well as of the
development
are
organization
predicted to be positively related to extent of use of methodologies.
2.
Extent of use is related to individual attitudes about
systems
development work and structured systems methodologies.
2.1
Analysts
methodologies
methodologies.
whose attitudes are consistent with the use of structured
in general
will
be
more
extensive
users
of
these
2.2
whose attitudes toward these particular methodologies are
Analysts
favorable will be more extensive users of these methodologies.
ability,
3. Extent of use is related to personal
and
experience
career orientation in systems development work.
3.1 There is a relationship between skill in each methodology and extent
of
of
use
that methodology.
analyst's skill, the less
therefor
the
It is possible here that the greater the
using
of
cost
the
the more likely that use will increase productivity.
who want to be more effective will use their skills.
the
relationship may work in reverse.
and
methodologies
Analysts
On the other
hand
Supervisors may insist on use of
the methodology, and individuals may subsequently develop
skills
as
a
result of extensive use.
3.2
Analysts
who
are
skilled
in other
development techniques will be less likely
These
analysts
are
those
a new method which promises to
too
that
use
the
and
systems
methodologies.
who are relatively effective in using other
programming and analysis techniques .
likely
to
programming
They will not perceive a need for
their
increase
effectiveness.
It is
supervisors of more technically skilled analysts will
not push them to use the methodologies
to
improve
their
performance.
The
more
skilled
may
analysts
methodologies which threaten to
be
resistant
their
present
also
make
to
learning new
obsolete.
skills
They may, as a result, use the innovation less extensively.
3.3
Technical work experience and technical orientation of the analysts
will be related to extent of use.
field
The more experience in the
of
systems development and the greater the degree of commitment to pursuing
a
technical
career,
methodologies.
the
more
likely
that
an
analyst will use the
In addition, analysts with a technical, computer science
degree will be more likely to use the methodologies.Among people with
strong
technical
education
it
a
is likely that the appeal of using the
methodologies will arise from an appreciation of the structured approach
to systems development work as a "state of the art" technique.
4.
Extent of
use
is related
to
the
implied
opportunity
for
increased productivity in the analyst's work.
4.1
The
type
of task to which analysts are assigned and in which they
spend major portions of their time will be related to
use
of these methodologies.
extent
of
The methodologies will more likely be used
on new applications development tasks than on
tasks.
their
applications
maintenance
In the latter tasks, the applications will have been developed
for
in many cases at a time before the methodologies were available and
The perceived
this reasons an alternative approach will have been used.
costs
of
employing the methodologies in this situation due to the need
to modify or
previous
replicate
work
may
be
high
sufficiently
to
discourage extensive use.
The
4.2
size of the work group to which an analyst is assigned as well
on
as the length of the project the analyst is working
to extent of use.
related
positively
communications.
that
an
analysts'
improve
to
designed
methodologies
The
improving
by
efficiency
The more complex the communications,
opportunity
be
Large jobs and long projects are
characterized by relatively complex communications.
are
both
will
the
more
likely
exists for productivity improvement due to use of
the methodologies.
Definitions. A number of terms will be used in the context of
study with specific meanings.
These terms are defined below.
Implementation: Implementation is the stage of innovation at
a
this
which
new product of process, having been developed and formally adopted by
the organization, is incorporated into the
Students
of
innovation
innovation is a
developed
into
of
the
organization.
and organizational change generally agree that
sequential
a
work
process
recognizable
in which
product
a
new
idea
is first
or process, then more or less
formally adopted through some decision-making process
into
practice
by
The number of stages varies,
one or more end users.
according to the situation and the researcher's
distinctions.
to
ability
fine
make
For the purposes of the present study the number matters
We are examining here a process
less than the sequence off the stages.
innovation
put
finally
and
which has clearly gone through numerous development, testing
and revision cycles so that it can safely be stated that the development
stage is complete.
The fact that
methodologies
the
under
are
study
in guidelines published by the organization for its analysts
described
indicates that the innovation has also gone through the adoption
stage.
It is implementation of the methodologies by individual analysts, in the
context
of their every day work assignments, which will be the focus of
this study.
Structured methodologies: procedures used
work
development
in systems
to provide a common approach, terminology, sequence of activities,
set of graphics and logical structure for undertaking the full range
applications
whose
from scoping of new projects through design,
development,
coding and documentation
of
programs.
is examined
implementation
of
structured
The
methodologies
study were developed by a
in this
computer technology development group in a large multi-national firm for
use by analysts employed by the firm.
step
steps
toward
include
methodologies,
design tools.
improved
the
analyst and programmer productivity.
development
essentially
of
as
The methodologies are seen
automated
computer-aided
tools
systems
to
and
Two specific methodologies are the focus of
a
Subsequent
support
the
analysis
and
the
current
study;
they are described briefly below, under slightly altered names.
The first methodology is used early in the process of
Methl:
systems development, when analysts are involved in scoping and
exploration, understanding the business operations and selecting
an appropriate applications development approach.
The second methodology is used toward the end of the
Meth2:
systems development process, when programmer-analysts are
conducting program design, coding and documentation tasks.
Analyst: the professional employees of the
the
responsibility
the
for
purposes
supervisory positions in the
organization
working,
maintenance
including
and
They
have
and
support.
study, include some employees in
the
of
information
systems
departments
of
the
for example, as project leaders or group leaders
which
with responsibility for several projects on
working.
who
the entire range of systems development tasks,
for
from analysis through coding
Analysts,
organization
include,
analysts
are
who in o other organizations
employees
too,
other
might have the title of programmer.
Client: end user of the applications packages or soft ware developed
in the
information
organization,
all
request of a client
systems
systems
in one
organization.
development
of
corporate
the
work
other
In
genera 1,
is undert aken
functional
headquarters,
this
upon the
in the
financ e, marketing,
organization
(i.e.
engineering).
The two parties agree on a time frame and
the work, and the client is billed for the service.
areas
in
a budget
for
REVIEW OF THE LITERATURE
of
process
The
implementing
tecnological
organizational, individual and
a
for
literature
this
of
study
behavior,
there
are
address the issues specifically related to the behavior of
which
the
First, among
type is in two areas.
individuals in organizational settings.
of
an
relevant
The
variables.
numerous studies of innovation and organizational
those
in
innovation
technical
complex one involving interactions among
is a
setting
organizational
a
on
research
This work is only a small
part
innovation, most of the rest being devoted to the
decisions and actions at the organizational management and policy level.
The topic of individual decision-making in relation to innovation in an
phenomenon D. Anderson (1981) refers to as
a
context,
organizational
because
the
policy
approaches
and
and
allowing
decision-making,
organization
in the
encouraging
even
of
is one
decentralized
individuals
also
to
use
methodologies of their own choosing from among a set of
such approaches identified as acceptable by the organization.
interest
study
current
micro-level innovation, is of particular interest in the
because
analyzed here are college
the
analysts
educated
It is of
whose decisions and behavior are
professionals,
a
group
generally
given a relatively high degree of responsibility for decision-making and
relatively more autonomy on the job than the employee who is not as well
educated.
These
are people who have been trained to think and act for
themselves and who tend to be expected by the organization to do so.Thus
the focus of the review, in this area, will be on implementation by
the
individual, in the environment of the organization.
helpful
The second area in which a review of previous work will be
is the
collar
use
issues.
in general,
workers
particular.
technology
new
of
methodologies
Information systems and structured
as
methodologies
recent arrivals in the workplace and are limited to the
largest work organizations.
studies
area,
They have been little studied to date.
are
work
being
described
For this reason, it is advisable to include in the
1982).
review some studies of more general interest though they may be
relevance
in specific
In
limited primarily to case descriptions
written by individuals directly involved in the
(Goldstein,
in
tools will help frame both the hypotheses and the analyses
relatively
latter
employees
development
In particular, studies of structured systems
of the current study.
the
systems
These studies may help in identifying micro-implementation
productivity
are
and
in improving productivity among white
lacking
Studies in
to the systems development process.
the area of white collar productivity will be reviewed to understand the
to
need for productivity improvement and the issues related
and
measurement,
with
specific
related to micro-implementation.
1982)
attention
once
These are the
again to the factors
human
(Keene,
factors
about which there has been increasing concern among management as
information technology changes the ways work is conducted.
the
definition
study
current
complements
productivity implementation.
attributes
of
new
the
major
Previous studies
technologies
In a
sense,
thrust of previous work in
emphasize
the
specific
and explore the relationships between
use
these attributes and the outcomes that result from the adoption and
of
The current study is concerned with the conditions
innovation.
the
under which adoption can be effective, in particular the factors related
of
is necessary
areas
both
to
an
Of course,
to an individual's decision to implement.
a
develop
understanding
complete analysis of the
process.
In the framework of innovation research, the
stages
of
sequence
a
postulate
of
in the
field
in the process of innovation.
Early
studies
Most theoretical and empirical
implementation.
is one
and different levels of
innovation
of
stages
different
on
emphasis
question
dimension
in
Reviews
of
subsequent work (Zaltman, et al., 1973) make it clear that some sort
of
the
studies (Rogers, 1962) noted the importance of
developmental
field.
of
process
the
describing
or
sequential
stage
diffusion.
and
innovation
time
theory is explicit in most work in
The number of stages varies according to the the
personal
situation
and
preference of the author, but even in a simple two stage model
common
there is a
sequence,
or
initiation
implementation or adoption later.
invention
coming
first,
The differences are important in that
different organizational, individual and technological factors appear to
at
effectiveness
facilitate
stages
different
(Rowe and Boise, 1974;
Downs and Mohr, 1976).
Most studies to date of productivity improvement
in information systems
development
implementation
of
a
new
technology.
explores the implementation process
decision
to
implement.
focus
and
on
outcomes
subsequent
to
The present study, in contrast,
the
factors
related
to
the
This focus on a particular sequential stage in
the Ife
cycle
of
the
organizational
innovation,
however,
is not
intended to limit the study to a narrow range of individual behaviors in
relation
to
use
of
the
methodologies.
At the individual level, the
entire process repeats itself as the individual conducts his or her
intitiation
or
search
own
activities prior to the decision to implement a
new technology (Anderson, 1981).
The review which follows is divided into three sections.
First the
literature on individual implementation in and organizational setting is
revie wed for suggestion
facto rs
which
may
of
applicable
and
organizational
individual
be found to relate to the use of the methodologies.
SeconId recent studies on
the
need
for
and
results
of
productivity
impro,vement in knowledge work in general and in informa tion systems work
in
are
particular
summarized
and
for
implications
micro-level
Third, work on the characte ristics
imple mentation identified.
and
use
studies
of
of structured systems methodology is reviewed.
Individual
organ izational
Implementation.
behavior,
innovration
lit erature,
appli cable
in a
while
establish
situation
A
number
not
a
of
directly in the mainstream of the
theoretical
in which
the
their
theory
on
framework
decisions
individuals are the outcomes under exami nation.
base
ear ly
and
particular
environments.
This
(1958)
the assumption than individuals act rationally,
they perceive in their work.
in
is
actions of
March and Simon
given their own understanding of their s ituation, to solve the
and
which
the
problems
The theory emphasizes individual cognition
individual's
suggests
that
perception
in
of
explaining
organizational
individual
is those aspects of the organization experienced by
it
decision-making
and perceived by the individual which are important, and that using
individual as the source of data about an organization is justified.
the
It
is after all the individual's perception of the organization, as much as
the
ultimately
behavior.
cognition,
including
which will influence decisions and
organization,
the
of
reality
individual
that
further
suggests
work
This
skills and frame of reference, will interact with
perceptions to influence behavioral outcomes.
Organizational issues
situations
rational decision process.
or
the
decisions
the
clients
confusing.
may
be
element
or
1977)
Delbecq,
is
if political
in decision-making
(Markus,
to
uncertainty
If, for
And
individual.
example,
otherwise
or avoided due to
delayed
complicated,
the
of
respond
will
individual
and
In each case one would expect that
might
organizational issues.
managers,
for
intended implementors do not have the clear authority to
uncertainty on the part
individual
This may be the case if the organization
important
implement (Knight, 1967).
rational
in
goals are not entirely clear
organizational
(Pierce
decentralized
considerations are an
1981)
important
environmental factors necessary for the completion of the
relevant
highly
more
in that the individual is forced search more actively
(March, 1981)
the
the
in which
become
to
likely
are
in each
case
the
by attending more closely to
the
goals
of
supervisors,
and work group members are not internally consistent
experience
the
goals
of
the
organization
as
In this situation one might expect both increased attention
by the individual to organizational cues and a reluctance
to
implement
an
is perceived to have mixed backing on the part of
which
innovation
significant organization members.
the
In addition to
perceives
of
goals
the
organization,
the
individual
the extent to which organizational resources are allocated to
a particular innovation and may in fact be the direct recipient or
In the
resources.
case of a process innovation, the resources may be
training time, or access to consultants who are experts in the
or
196 3;
front
the
of
end
it
However
experimentation.
in
process,
which
in the
resources
captures
innovation
that slack is most important at the
found
is generally
It
is one
1976)
Mohr,
and
Downs
metaphorically the significance of spare
process.
and
The concept of organizati onal slack (Cyert
he process.
modification of ti
March,
development
in the
process implement;ation of even to be involved
and
process,
to be involved in the development of guidelines for
opportuni ty
an
such
initiation
with
connection
and
that slack is necessary at
is possible
As an individual is deciding whether to implement
later stages as well.
decision
positive
a new technology, the likelihood of a
is increased
when the organizational resources are sufficient to support initial use.
The
available
resources
for
implementation
be
will
a factor worth
considering (Mohr, 1969) and the incentives to implement are
make a difference as well
Several
to
(Kerr, 1975).
suggest
studies
likely
and
values
the
that
orientation
of
organization members can affect the implementation process, particularly
when
they
outcome.
are
directly
Research
on
related
the
size
to
and
the
innovation
degree
of
or its intended
specialization
of
organizations suggest that with specialization and expert knowledge in a
unit of an organization, the attitude of professionals is more likely to
be
an
1977).
important
factor
the
In general,
important
in
a
For
example,
more
would
one
systems analyst to care more than a programmer about what the
client says, since the analyst's job puts him in closer,
and
be
(Rogers, 1962) as well as the type and degree of access of an
individual to the members of that community.
expect
will
community
relevant
the
of
norms
Moch and Morse,
(Moch, 1976;
innovation
contact
intensive
On the
programmer.
other
with
frequent
the client than does the job of the
both
hand,
more
will
be
influenced
by
their
perceptions of the attitudes, opinions and desires of their supervisor.
The organizational factors which emerge from the research described
above are contextual variables in the
Equally
micro-level.
implementation
process
for
Individual factors have been categorized in numerous ways.
simplicity
a
two
category
scheme
adopted by D. Anderson (1981) is used
perceptions
the
are the individual factors, for it is
important
the individual whose perceptions of reality we are taking
of
at
reality.
For the sake
suggested by Knight (1967) and
in this
study.
Individuals'
are assumed to be influenced by their motivations and their
cognition.
Motivation to implement an innovation may be assumed in situations
in which
the
organizational
incentives
directly related to
the
relevant
circumstances
the
point
from
of
are
behaviors.
clear,
Even
consistent
in the
best
and
of
view of the organization wishing to
encourage innovation, however, the effect of incentives is mediated
by
individuals
of
motivation
personal
the
Locke et al.,
(Locke, 1977;
those
Unless an individual's goals are consistent with
1978).
the
of
organization, or unless that individual desires to obtain the incentives
may
outcome
the
offered,
not be exactly as desired.
Early theorists
of
call attention to the significance of the general frame of reference
may influence motivation.
which
individual
the
status and Rogers (1962) explores the importance of
conceive
work, one can
experience
previous
as
of
an
individual
analyst's
a better
increase.
example
and
orientation
professional by using the most modern methods, and if
On
the
other
hand
limiting the analyst's opportunities
for
In the current
If the motivation is to
the methodologies are perceived as modern, then the
will
ideology
a systems development professional influencing
his or her motivation to use new technologies.
become
attitude,
of cosmopolitan orientation of individuals.
extent
and
Blau (1963) discusses
likelihood
use
of
if the methodologies are seen as
for
professional
accomplishment,
by reducing the level of professional skill and creativity
required in the job (Kraft, 1977) then a lower level of use is the
more
likely outcome.
Another way in which motivation may influence implementation is as
a
result
of contact with or communications from others.
In a study of
homeowners' decisions to use solar energy devices, Leonard-Barton (1979)
found that knowing another solar owner was the
use.
Mohr
more
Similarly,
strongest
predictor
of
in a study of professionals in public health agencies
(1969) found that his subjects' attitudes toward change,
and
even
strongly their motivation to change, were related to the extent of
was
Others have found
an important contributing factor.
be
to
found
orientation
professional
In this case a cosmopolitan,
implementation.
that communications from respected professionals in the field influenced
Dewar,
in the
And
1973).
study,
present
extent
and
and
(1983) found a
McErlean
advocate
significant relationship between contact with a local
methodologies
Hage
1970;
Aiken,
the degree of use of new technology (Hage and
of
the
of use, and Leonard-Barton (1983) found that
informal organizational influences were a significant factor in relation
to
use.
of
extent
In these
communications
resulted
skill or both.
This
it is not
examples
clear
whether
increase in motivation, an increase in
in an
be
may
distinction
worth
in future
exploring
research.
Individual cognitive factors related to implementation include, but
are not limited to, awareness of and
process.
a
innovation as
use
of
problem
solutions
new
implement
the
(1967)
Knight
those
that
argues
solving
pro cess
they
but
a lso
to
the
solved.
postulates the effects of expertise at two
levels.
specific
solutions
is seen
as
process.
can
be
expected
to
problem
extent
they
are
D. Anderson (1981)
First,
in search
as well as general awareness of relevant principles
important.
or
new
at the level of initi ation of innovation) expertise in
activities (i.e.
detachment,
the
to not only to the extent they are skilled in
solutions,
the
of
when individuals are engaged in
to be
about
knowledgeable
use
in the
skills
far-out
Second,
Involvement,
expertise,
expertise
will
or
expertise,
and
each contribute to the search
can
be
close-in
a
factor
at
the
level
of
in that individuals who understand a new technology and
implementation
that
use
to
likely
more
are
appreciate the need and potential uses
Similarly Pelz and Andrews (1966) in a study of innovation
technology.
so
become
skilled
they are not open to new ways of
that
study
in a
obtained
A similar result was
approaching a problem.
be
when
limited
be
can
skills
that the positive effects of specialized
individuals
can
the extent of use of new technology, suggesting
to
related
positively
periods
short
for
specialization
among scientists show that
of
manufacturing engineers (Bigoness and Perreault, 1981).
The interaction of motivational
suggested
boundaries
Specifically,
for
perspective,
those
outcomes
(Porter
succeed
the
definition
to
technology in relation
(Campbell,
of
that
Lawler,
and
in light of past experience.
Information Systems
Productivity
can
are
1970).
influence
another
From
assessment
constructs
of
a new
individual
the
An implementation, just like any
change in behavior, will be evaluated in relation to
of known alternatives (Knight, 1967;
al.,
1968).
the
and
need
need
et
in innovation
success
of
expectations
to
motivation
develops
past
and
skills
individuals
an
shape the expectation of future behavioral outcomes and thus
experience
set
in which
ways
the
explored
have
Some motivation researchers
expectancy theory.
on
in work
been
has
factors
cognitive
and
the
effectiveness
Cyert and March, 1963).
Improvement
Methodologies.
The
type of innovation studied and the context in which implementation takes
place,
as
degree to
well
which
as
the stage in the innovation cycle, will affect the
various
organizational
and
individual
factors
are
An understanding of
the extent of use of a new technology.
to
related
general issues related to productivity improvement among white collar or
section, the
this
for
need
and
issues
some
and
to
related
will
resistance,
overcoming
In
systems
in
the
factors,
The importance of human
measurement of productivity reviewed.
motivation
improvement
productivity
summarized
be
will
development
including
specifically
the implementation and use of structured methodologies.
to
related
issues
in identifying
knowledge workers will be helpful
addressed
be
briefly.
There is general
over
that
makers
agreement
the
past
social
among
years
thirty
and
scientists
productivity
policy
increases in
professional, technical and office work have lagged those in other areas
(Strassman, 1982;
category
has
Thurow, 1982).
increased,
employees
As the number of
managers
have
become increasingly concerned
about controlling costs, monitoring and coordinating work and
output (Gremillion and Pyburn, 1983;
in this
measuring
Martin, 1982).
The interest in productivity improvement in systems development
a
special
rapid
case.
growth
technology
of
which
is
The need in this area is highlighted by the extremely
employment
coupled
with
the
drastic
changes
in
have taken place over a relatively short period, with
the result than in many large systems development groups there are
elementary management systems in place (Zmud, 1980).
only
and
The noticeable increase in the cost of developing new software
maintaining and enhancing applications which are now in operation is one
Costs of analysis and programming, relative to
the problem.
of
aspect
the cost of hardware for computing
are
substantially
and
Mendes, 1980;
Mcgowan,
expected
1975;
power
to
increased
have
memory,
and
continue to do so (Yourdan, 1976;
1982;
Benjamin,
In
1975).
Brooks,
addition the demand for software has increased, as evidenced by the ever
growing
work
of
backlog
In part the
in most systems organizations.
applications
demand is a result of previous successes, in that users of
developed begin to find new uses for information and request
previously
minor or major changes
Johnson,
the
At
1977).
in the
same
programs
time
(Synnott and
that
both
is problematic
work
(Synnott
1981;
and
total
relative
labor
development costs are increasing, the availabilty of
this
Gruber,
to
perform
The field has
and Gruber, 1981).
expanded more rapidly than the capacity of training institutions so that
it is at times difficult for a firm to find enough skilled employees
to hold current employees on the job.
or
Turnover is high, frequently more
than 25 per cent annually, and the costs of orientation and training for
new
personnel
be
can
significant (McLaughlin, 1977).
These problems
apply less to the programmers, who perform relatively less skilled jobs,
than to the analysts.
programming,
are
the
The latter, while not always
skilled
in
professionals in systems development, the people
who must accurately model the information flow
operation
highly
of
an
entire
business
and represent it in an understandable and workable plan which
will become the framework for the design of the program.
the
What magnifies the difficulty is that
related
and
analysis
occuring at the front end of the development process are the
activities
to
most important with respect
of
effectiveness
the product.
qual ity,
dependability
long
and
run
Anal ysts, even the most skilled and the
most tenured in a job, face a great
deal
of
to
due
uncertainty
the
amount of information, the complexity of information flows, the changing
nature of the business they are model ling, the changes in the technology
on which the applications will be run 1. To the extent that they postpone
decisions
or make incorrect decisionIs based on insufficient information
recommendations
their
Coming
application.
decreaýse
will
the
effectiveness
project
is a
and most likely in ways which are difficult to detect until
one,
the program is installed.
At that point what errors
or
inefficiencies
there are wil 1 lead to requests for more work (maintenance).
is a
the
in a complex process, their decisions will
early
affect the wc,rk of many others, over a long time if the
large
of
lower
The result
quality product at a higher cost than might otherwise have
been the case (Zmud, 1980).
Given thie cost and quality problems,
concerned
managers
are
understandably
with finding ways to establish more effective controls.
They
would like to have an analysis and design sequence which is predictable
in cost and time, observable by supervisory personnel at each level, and
standardized
to
the degree necessary to allow new personnel to pick up
where previous employees left.
They would like a sequence which results
in a higher quality, more maintainable product.
of
reliability
and
cost
reduction,
along
It is the
with
the
combination
possibility of
development
systems
structured
resulted, eventually, in the use
Kelly, 1975;
Mendes, 1980;
interest
generated
which
effectiveness,
management
improved
in
methodologies in the middle 1970's and
of
such
(McGowan and
methodologies
Nolan, 1977 and 1979).
To comment on the effectiveness of any new technology in increasing
unit,
measuring
desired
a
and
outcome
or
of
productivity
number
concept
efforts
present
of
to
programmer productivity such measures are used (Johnson, 1977).
The input is generally taken as units of
some
The economic
measurement along two dimensions, inputs and
And in some early and even a
outputs.
measure
suggests
least a well defined
at
dimension on which productivity will be measured.
standard
some
point,
starting
productivity it is necessary to have a
variation.
The
time,
working
programmer
or
output is lines of code, or source code, or some
other unit of this sort.
stated
The need for a system of measurement is frequently
journals
(Keene,
1982;
Hammer,
1981).
in the
But convincing measures are
missing from most studies and appear to be unavailable to management
most
organizations
(Khalil,
1980;
Strassman,
The
1982).
and
appears directly related to the difficulties in defining
the
work
of
systems
development,
particularly
structured stages of analysis (Parikh, 1981).
at
the
in
problem
managing
early, less
One approach which may be
applicable has been used in assessing the impact of technology on office
work (Crawford, 1982).
electronic
Crawford (1982) describes
the
introduction
of
mail in a large industrial firm, addressing the productivity
issue in two parts.
First he uses quantifiable measures
of
input
and
output
as
extremely
an
rough
for
basis
efficiency
calculating
using
Then he addresses the improvement of effectiveness
improvements.
qualitative data.
Human factors in implementation of new technology are recognized as
important in many of the ways described above.
here are those suggesting
increase
extent
the
of
changes
management
The findings of interest
which
can
be
made
to
use of the technology, as well as those about
which additional data can be gathered in the current study.
This
data
may
have implications for selection and job assignment (McGowan, 1976),
for
reducing
(Strassman,
job
turnover
1982;
Campbell,
and rewards (Campbell, et al.
Structured systems
improve
the
personnel,
They
were
(Goldstein,
1971;
for
job
Yourdan, 1976) and for incentives
and
methodologies
efficiency
of
were
developed
systems
in large part in response to the problems already
intended
to
improve the ability of
process (McGowan, 1976:
training
1970).
development
effectiveness
1982),
improve
managers
development
described.
productivity, and at the same time to
to
control
Yourdan, 1976:
the
systems
development
Mendes, 1980).
Studies of these methodologies ascribe numerous benefits
use.
to
to
their
Some studies describe methodologies used early in the development
process.
Mendes (1980) discusses how these techniques
improve
product
quality, increase the effectiveness of programmers and analysts, improve
the
accuracy
projects.
the
point
of
time
estimates
and
the
She describes a methodology which
of
amount
of time needed for
approaches
analysis
from
view of the business operation rather than from the data
component of project management.
not
that
argues
(1981)
Anderson
C. M.
potential
has
methodology
same
this
is presented, in addition, as a
methodology
The
client.
the
with
communication
analyst
improves
processing perspective and as a result
as a communications,
just
productivity or management technique but also as a way of involving
He argues that both
more integrally in the development process.
client
the
because of increased participation and due to the increased accuracy
of
the outcome, client acceptance and use of the end product will increase.
Other studies focus on
process.
In a paper on one
programming
for
methodologies
toward the end of the development
place
take
which
those
activites,
structured
structured
programming
technique,
Menard
increase
of
productivity, the reduction of maintenance costs and the improvement
of
benefits
as
includes
(1980)
the
use
from
resulting
in development and in maintenance due to improved
communications
both
documentation.
She notes that
methodology
this
time
more
takes
in
design work than would an alternative approach, but argues that the time
is made
up
time is reduced due
automated
In addition, maintenance
in subsequent development steps.
version
to
of
the
the
presence
fewer
of
time
methodology,
errors,
more
In an
required for its use is
expected to be reduced still further.
Other studies reinforce these findings.
see
advantages
in the
development, improving
seriousness
of
errors,
for
potential
overall
design
improving
and
Synnott and Gruber
integrating
reducing
maintainability
the
the
(1981)
stages
number
of
and
and readability of
programs, and improving the effectiveness of project management.
Holton
(1977) argues more generally for improved quality,
from
resulting
methodologies.
He warns, as does Yourdan (1976),
programmers
structured
the
of
introduction
productivity
and
maintainability
all
against expecting
to benefit since significant skill is required to learn and
use these techniques correctly, and not all programmers and analysts are
selection,
These studies suggest that
sufficiently talented.
training
and assignment to tasks on which the methodologies are to be used may be
more important than is generally admitted.
Yourdan goes further in discussing
implementation.
First,
the
the
methodologies
potential
may
conflict with classic
design and programming methods so that personnel who
forced
to
unlearn
old
ways
of working.
with
problems
use
them
may
For analysts with extensive
technical training or many years of work experience, this situation
result
be
may
in resistance to the use of the newer techniques (Holton, 1977).
Secondly according to Yourdan the methodologies, being time consuming to
set up, may not be appropriate for use in small projects and may
be
of
only limited use in projects of medium size, those requiring the time of
a
few
Finally Yourdan suggests that use
programmers for a few months.
of these techniques may expose some organizational issues related to the
lack of
focus
designers.
and
Yourdan
specific
expects
expectations
that
for
management
analysts
will
and
program
discover that the
responsibility for design is not clearly delegated, and he even suggests
that use of the methodologies may
categories to correct the problem.
lead
to
the
creation
of
new
job
as
Structured methodologies have been described
will
improve
the
which
As such they would be
analysts.
of
effectiveness
techniques
appealing to those analysts who see themselves as professionals having a
the
responsibility to use
methodologies
have
most
techniques
modern
available.
These
alternately been presented as techniques which will
increase management control over analysts.
As such they may be resisted
by analysts, especially those whose current skill level is adequate
the
job.
Given
the
possibility
methodologies may be perceived,
implement
of
these
it is likely
two
that
ways
the
for
in which the
decision
will be made on more than just technical criteria.
to
Analysts'
attitudes as well as analysts' perceptions of organizational support may
both influence the decision and may be related to extent of use.
METHODS
in computers and computer programming create models of
expertise
with
professionals
in which
process
Systems development is a complex
These
information use and information flow in a business.
are
models
the framework for the design of applications programs which will support
the
A study of the systems development process is
operation.
business
second,
and,
itself
technology
the
understand
to
organizational
users
relationships among the developers and between developers and end
of the systems they develop.
the
of
necessarily complicated by the need, first, to grasp the basics
A study of innovation, in this setting, is
similarly complicated.
The
study
current
under
researchers,
the
was
conducted
and
designed
in which
a
team
science
the
research
was undertaken.
staff
in the
It was designed
under the assumption that the people in the best position to comment
the
extent
of
use
of
of
supervision of a university faculty member and
with the cooperation of members of the computer
organization
by
the
innovation
responsible
those
were
on
for
implementing that innovation, the analysts.
will
In this chapter the methods used in the study
In the
given,
first
explanation
of
presented.
section, a description of the subject population will be
with
along
be
how
summary
the
characteristics
sample
was
describe the questionnaire, highlight
chosen.
the
of
the
sample
and
an
The second section will
sections
relevant
for
the
the
describe
and
study
current
pilot.
The third section provides a
used
methods
the
section
fourth
In the
data.
description of the interview used to gather the
for analyzing the data are presented.
section includes a description of the dependent variables and a
This
summary
of the independent variables, as well as a definition of the opportunity
to specify the subset of the sample considered eligible
used
variables
users.
universe
Population. The
of
population
systems
of
a
long
All the subjects of the study will be referred
to hereafter as analysts although they held more than a
titles
in the
in
These analysts were responsible for
organization.
organization.
supervisory
of
support
The
clients
population
positions,
some
included
several
include
any
senior
level
who
analysts
of
worked
the
in
levels up from the personnel who
The population
managers
or
and revising the methodologies under study.
did
not,
any of the computer
technology development team which had been responsible
testing
consulting
areas
functional
in other
wrote computer program code or documentation.
however,
official
dozen
applications development and related systems development and
work,
corporation
standing and large scale computer science and information
processing organization.
job
the
analysts, programmer-analysts and other systems
development professionals employed by a large multinational
with
was
study
this
for
subjects
for
developing,
Nor did the sample
include any personnel outside the information systems organization.
in
The sample was chosen from a total population of more than 1000
two
First
steps.
three locations were selected to provide a range of
Each location had a minimum
organizational environments for the study.
of
analysts
thirty
in the
organization and a minimum of three years
In addition,
experience with the structured programming methodologies.
each
site
of
process
in the
was
of
none
at
although
computerized tools for these methodologies,
installation of
the
considering
the
sites was that version widely available.
Location
1 was
the
corporate
applications
development
group,
to,
but
organizationally distinct from, the computer technology group which
had
located
designed
a
in
corporate
large
the
developed
and
close
physically
complex
methodologies.
There
were some working
relationships between the two groups and there had been some exchange of
personnel, most notably at the management
Location
1
proximity
to
information organization.
the
There
level.
was
at
also
upper level corporate management of the
It should be noted, too, that at
Location
1
the applications group had only limited internal consultation and had to
In the other
rely on the technology development group for this support.
locations,
in contrast, the applications groups had their own internal
technical consultation staff members.
Location
relative
2 was
independence
an
international
since
selected
for
its
from the American corporate headquarters and the
difference in the background of
extreme,
affiliate,
the
analysts.
Differences
were
not
the site was in Canada, but it was clearly important in
the corporate culture that this site maintain
itself
as
a
firm
with
management
distinctly separated from that of the corporate parent.
For
of
the
rather
name
own
example, this affiliate used its
that
than
parent.
Location 3 was an American
two
other
the
from
affiliate
geographically
removed
far
There had been some transfers in past
locations.
years from the Location 1 applications group, but these were limited
3, like the other two sites, had officially approved
Location
number.
in
the methodologies as part of a set
techniques
development
systems
of
recommended for use.
The sample size was targeted between 120 and 150.
corporate
the
management
at
study explained.
technology
development
group,
With the help of
systems
information
each site was contacted and the purpose and scope of the
Each manager was asked to participate in the study
by
selecting a random sample of employees below the senior management level
as potential subjects.
I
and
60
each
at
The target size, by location, was 30 at Location
Locations
2 and 3. The sample was to be drawn by
selecting every Nth name from a list
determined
by
of
all
if they
N was
These employees were
then
would be willing to participate in a one hour interview
and written questionnaire session as part
agreed
where
dividing the number of analysts employed at the location
by the target sample size for that location.
asked
personnel,
of
the
study.
Almost
all
to participate, and of these almost all completed the interview.
Schedules were arranged by administrative support staff in the firm, and
gaps were filled with back-up respondents drawn from the same
lists.
personnel
The sample was relatively balanced by sex and relatively young
well educated.
per
cent
were
and
Of a total of 145 subjects, 60 per cent were male and 40
Slightly more than half were in their twenties
female.
and 87 per cent were less than forty years old.
All but 5 of the sample
had attained a level of education of college or beyond (Table 1).
The sample can be characterized by type of education, previous work
Table
experience and expressed interest in a technical career path.
shows
that
science.
more
than
however,
permanent
42 per cent had a bachelor's or master's degree in computer
Seventy eight per cent had worked in systems
two
years
subjects
jobs
indicated
that
they
in systems development.
(Table 3).
64
per
for
development
In describing their career plans,
(Table 2).
pursue a technical career,
management
1
cent
were
not
all
to
committed
While 36 per cent intended to
were
planning
a
in
career
Asked to project their plans over the next five
years, only 15 per cent expected to be in non-managerial jobs in systems
development at the end of that time.
to be in general management positions.
analysts
expected
In contrast, 49 per cent
The background data suggest that
in this sample are young, well educated, technically oriented
in the field of systems development, and experienced in the field.
suggest further that these subjects are looking forward to a
too
far
in the
They
time,
future, when they will be in management positions.
majority expect that they will be general managers rather
development managers.
than
not
A
systems
Table 1
Demographics
Male
Sex
20-29
Female
n
%
n
%
87
60
57
40
30-39
40-49
50-59
Total
Missing
U
60 or over
n
%a
145
100
Total
Missing
n
%
n
%
n
%
n
%
n
%
n
n
77
53
49
34
13
9
4
3
1
1
1
145 100
Education
Nontechnical
Technical
Bachelor's
Master's
Bachelor's
Master's
Other
N
%
N
%
N
%
N
%
N
%
46
32
14
10
43
30
29
20
12
8
aMissing is not included in percentage data.
%
Table 2
Work Experience
Years
Years in Systems Development
Number
Under 1
Percent
Years at the Firm
Number
Percent
4
3
10
7
1,0 to 1.5
17
12
20
14
1.6 to 2.0
11
8
19
13
2.1 to 3.0
20
14
26
18
3.1 to 4.0
13
9
9
6
4.1 to 5.0
8
6
8
5
5.1 to 7.0
20
14
13
9
7.1 to 10.0
15
11
10
7
10.1 to 13.0
15
11
11
7
13,1 to 16.0
11
8
7
5
7
12
8
More Than 16.0
9.
Missinga
6
Total
145
aMissing is not included in percentage data.
0
100
145
100
48
Table 3
Career Orientation
Percent
Number
Managerial
64
Technical
36
5
Missinga
145
Total
100
Career Path
What type of work do you see youself doing in
In 1 Year
N
Systems Development
Technical Management
General Management
Other
Missing
Totala
103
23
3
8
8
145
years?
In 3 Years
In 5 Years
%
N
%
N
%
35
37
19
9
20
36
64
10
15
145
15
28
49
8
100
47
49
26
12
11
145
aMissing is not included in percentage data.
100
100
and
The sample can be further characterized according to the level
type
of
study
under
methodologies
the
because
interest
is of
information
context
in the
performed
work
of
the organization.
This
of
the
here is thought to relate to the individual
or
analyst's need for a particular methodology in his
work
of
use
of
extent
her
The
work.
an analyst is not easy to observe, let alone categorize, so an
categories
to
study
attempt was made in this
identify
key
several
descriptive
within each of three dimensions of systems development work:
the function which the application software will perform for the client,
and
the
cycle at which the work was performed.
The
the type of application or system on which the analyst worked,
level
development
in the
distributio n
of
working
within
time
these
of
each
dimensions
summarized in Table 4 and described briefly here.
In the functional dimension, the mean time
application s
will
which
used
be
worked
in financial
or
is largest
other
on
business
application s. The mean is only 14 per cent for those products whose end
use will be in engineering, and 15 per cent for products to be
such
areas
other
as
personnel,
among
time
different
In reporting the split
a mean
of
57
per
of
types of systems, analysts indicated that
operational or transanctions systems comprise the bulk
with
in
text processing or technical support
within the systems development organization.
their
used
cent
of
work,
their
of time in this systems type.
Finally
analysts report that on average they spend substantial time on work that
is related to existing systems, relative to work on
mean
per
cent
new
systems.
The
of time on maintenance and enhancement, taken together,
Table 4
Distribution of Working Time By Functional Area,
Systems Type and Project Level: Summary Statistics
Mean Percent of Time
Worked in this Category
Functional Area
Engineering
Finance
Business Logistics
Other
14
34
34
15
Systems Type
Operational, Transactions
Stewardship, Reporting
Planning, Analysis, Decision Support
Other
57
20
17
4
Project Level
Scoping, Exploration
Systems Development, Acquisition
Major Enhancement
Application or User Support,
Maintenance
t-harL
vc~rr
17
35
15
27
account for more than 40 per cent of total time.
working
was
time
according
breakdown
This
the
to
defined for the
tasks
of
their
programmer-analysts
is of interest in defining an opportunity
to
Prior
variable for the study.
work
down
break
to
asked
specific
the
comprising
purposes of the study as
(Table 5).
were
subjects
Another way in which
of
use
of
extent
examining
the
methodologies, it was necessary to define the subset of the sample which
had
the
Since each
to use each of the two methodologies.
opportunity
methodology is considered appropriate for use in performing only certain
tasks, only those analysts who spent time on those tasks were considered
subset
Specifically, analysts were included in the
eligible users.
of
eligible users for each methodology if they reported spending 5 per cent
or
more
time
their
of
on
at
least one of the tasks for which that
methodology is considered appropriate.
tasks
For the purpose of this study, ten
programmers
assigned
are
these tasks, the first
client
needs,
three
(understand business
Among
define
operation,
recommend solution) are those for which Methodology 1 is
analysis
in the
tasks
These
organization.
the
comprise
work
of
A large majority of analysts
reported that they spent more than 5 per cent of
tasks.
and
analysts
which
in this organization were identified.
considered an appropriate method.
systems
to
their
time
on
these
It is analysts in this group who were considered to have had the
opportunity
program,
programmer
to
code
in
use Methodology 1. Similarly three other tasks (design
program)
program,
document
the
organization.
comprise
It is these
the
tasks
work
for
of
a
which
Table 5
Proportion of Time Spent on System Development Task
Proportion of Time
Less than 5% 5% or More
Frequency
1. Understand Business
Operationa
2.
Define Client Needsa
Percent
21
115
79
30
21
115
79
107
93
Document Solution
Produce Data Base Design
6. Analyze role of
Information
9.
106
73.
39
27
76
52
69
48
Design Programb
105
Code Programb
105
Document Programb
10. Provide User Support
Percent
30
Recommend Solutiona
5.
Frequency
50
35
aTasks for which Ml is an appropriate methodology.
bTasks for which 112 is an appropriate methodology.
74
51
95
65
is considered
2
Mehthodology
A
or more of t;heir time on each of these tasks.
were
whc
sample
considered
to
have
had
It is this subset of
the
opportunity
to
use
organization
and
the
Methodology 2. A third methodology exits in the
considered
still
but
smaller,
of respondents reported having spent 5 per cent
proportion
substantial
appropriate.
Use of that
for use in tasks four through six.
appropriate
methodology will not be considered in this study.
is
Finally, no
specific
methodology is considered applicable to task ten.
A final set of descriptive
an
gives
iindication
organization
of
the
in which
they
time
in the
responsibility,
for
variables
relationship
work
of
the
sample
the respondents to the
of
level
of
dimensions
on
population
organization, type of client and size of
project.
Most respondents in the sample worked at the level of programmer or
Seventy
analyst in the organization, rather than at a management level.
two per cent of respondents were classified
with
variations
of
these
titles and only 27 per cent were in supervisory or specialist jobs.
Respondents typically had
organization
and
spent
relatively
Only
one
time
in the
in their current jobs at the time of the study.
than half of the sample had been in the
fewer.
little
fifth
had
been
organization
there
more
three
years
than ten years.
More
or
This
distribution is not surprising, given the relatively recent emergence of
systems development as an area of employment in most organizations.
is related,
too,
with
It
relatively low age distribution in the sample.
Tenure in the current work group is even shorter, with 61
per
cent
of
having
respondents
fewer.
worked
their
with
Tenure in the current job is similarly relatively short.
Questionnaire. During the two
study,
current group for two years or
of
members
the
research
team
issues
the
pilot
members of the technology
and
twice
met
methodologies
development group which had developed the
identify
to
prior
period
month
which were to be addressed in the study.
to.
Based on the
results of these sessions and on conversations and open-ended interviews
with individual applications development group members representative of
the population to be sampled, the team wrote an initial
interview
protocol
an
of
version
This protocol was revised
to be used in the pilot.
page
after the pilot and finalized as a twenty-two
questionnaire
(see
Appendix A).
The questionnaire was designed for two
intended
structure
to
the
gathering
First,
purposes.
it was
relatively large amounts of
of
specific data on independent and dependent variables under investigation
Since four team members were each planning to
by the team.
different
aspect
in the
form
of
To
the
short
the
that
extent
written
answer,
gathering process would be less time consuming.
could
data
lack
of
For some aspects of the
questions.
and
well-defined measures suggested that open ended questions
would provide more valuable
questionnaire
be
responses the data
problem, however, both the complexity of the implementation process
the
a
of the problem, the topics covered were broad and the
number of questions large.
obtained
research
was
data.
Thus
the
second
purpose
of
the
to provide a format for recording responses to these
The format was
self-explanatory
so
that
each
respondent
complete
could
found
generally
members
the
with
interview,
interactive
an
that
notes in the appropriate spaces, was most effective
writing
researcher
However in practice, team
in writing.
questions
all
on these open-ended questions.
part
the
on
resistance
of
to a long and demanding data
respondents
of
difficulty
the
minimize
to
organized
The questionnaire was
gathering process by ordering questions so that easy-to-answer questions
middle
sections
and necessary but perhaps less interesting demographic
questions at the end.
for
asking
at work,
in the
questions
threatening
came first, challenging and potentially
The early questions dealt with recent experiences
and
descriptions
of
categorizations
responsibilities over the preceding twelve months.
analysts'
These questions were
factual and could be answered without analysis or evaluation on the part
of
the
respondents.
The
middle sections contained the more complex,
more analytic or judgemental questions which might require some time
mental
effort
which
and
might
be
difficult
accustomed to reflecting on their experiences.
that most of
researchers
the
comments
found
and
questions,
supportive
necessary.
The questionnaire continued
that
with
not
respondents
It was in these sections
some
in some
for
or
probing
interaction,
cases interviewing were
several
pages
of
short
answer attitudinal questions and ended with a page of demographic data.
The questionnaire included eight sections.
job
title
and
The first
on
responsibilities as well as project characteristics for
work to which the respondent had been assigned over the
Included
focussed
in this
section
previous
year.
were questions about the way each analyst's
In the second section
system developed.
extent
use
end
of
type
which
for
been developed, level of development task and type of
had
applications
by
categorized
spent,
been
had
time
reported
respondents
on
to which they used the methodologies, the alternative approaches
they used when they did not use the methodologies and their reasons
selecting
organization's assessment of their work.
advantages
own
in
and
with
satisfaction
Fifth they assessed their own skills
of
sources
organization,
analysts'
scaled
of
attitude:s
response
perceptions
and
self-perceptions
analys ts'
final
The
satisfa Lction.
job
level
and
in
The
of
tra ining, and sources and extent of coimmunication in
sets
methodologies,
extent
instruction,
the organization relati ve to the methodologies.
several
the
In the fourth section analysts
various development techniques, including the methodologies.
sixth section covered
were
work
and disadvantages of the methodologies and gave each
methodology an overall rating.
using
for
third section requested analysts'
in their
important
factors
of
The
alternatives.
these
perceptions
listed
the
In the seventh
section
values
in the
on
items
related
of
learnin g
section
structured
to
styles
requested
and
such
demographic data as age , education level and work experien ce.
The questionnaire was pilot tested at Location I
team,
in interview
analysts.
with
respcindent
Each
questionnaire.
respect
sessions
was
with
twenty
asked
to
by the
research
application s development
read
and
complete
the
The teaim member then reviewed questions with respondents
to
clarity
of the question, agreement on the meaning of
certain terminology, sequence and length of the protocol, unintended
or
emphasis
misplaced
and
omissions.
The team members then individually
and as a group reviewed the results of these sessions and
In particular, several questions which had until this time
instrument.
structured
remained open-ended were rewritten in a
other
structured
For these
was altered to
format
the
questions
response
pilot
using
form,
session data for generation of multiple responses categories.
and
the
modified
include a category for "other" responses.
insure
To
locations,
the
that
the
used
terminology
questionnaire
was
further
gathering visits to Locations 2 and 3 by
conducting
a
telephone
conference
with
of
across
applicable
prior
tested
to the data
review
and
of
the
For both locations, on
the
sending
applications development group at the site.
suggestion
was
a
it for
representative
local personnel, the questionnaire was slightly modified
to reflect local terms and titles.
Interviews. The questionnaire was administered to
face
in a
to face interview scheduled to last one hour.
participation was voluntary.
regular
each
Respondents'
Their time at the session was part of
the
working their involvement had the prior approval of management.
Upon arriving at the interview, each respondent was given a
summary
from
respondent
of
the
voluntary.
the
team
brief
oral
project, its purpose and its scope as well as a letter
leader
once
again
stating
that
participation
was
file
data
for
within
processing
two
weeks
completion
of
a
to
All coded responses were transferred from the questionnaires
the
of
interviews.
P lan for Data Analysis. Data gathered in the course
was
a nalyzed
prepared
in papers
of
Use.
study
separately by each of the four team
member s. Each study addressed a different aspect of the
studie s
the
The
problem.
shared a common definition of the dependent variable, Extent of
to
Independent variables were selected and defined according
the
design of each author.
variable
n this section the definition of the dependent
given
and
the
method
of
indepe ndent variables will be defined.
select ing
from
the
it will
calculating
be
will
explained.
Finally, the reasoning
be
The
used
in
sample those analysts considered to be eligible to
use the methodologies will be explained.
There were two dependent variables in this study, each representing
Methodology 1,
the extent of use of one structured systems methodology.
the
technique
develo pment
for
increasing
sequence,
was
analysis
systems
measured
by
This
MI.
early
work
in the
was
methodology
consid ered by the organization to be appropriate for any one or more
three
the
tasks which comprise the work of systems analysis:
b usiness
soluti on.
An
operation,
analyst
defining
was
client
needs
and
of
understanding
recommending
a
given a score on Ml based on the amount of
time he or she had spent on the tasks for which
the
methodology
could
time
working
or
If a respondent reported having spent 5 per cent
methodology.
of
used and on the self-report of the extent of use of the
be
potentially
a
be
considered eligible to
the
of
one
in any
user.
The
more
tasks, that analyst was
three
score
for
calculated
that
respondent was an average of the reported extent of use across the three
where
tasks,
of
extent
for
was
"Always".
representing "Never" to
grouped,
use
into
analysis,
two
measured
The
on
a
calculated
categories
scores
roughly
scale
point
five
were
then
equal in size,
representing low and high use.
Similarly, a dependent variable for extent of use of Methodology 2,
the
the structured programming technique, was calculated based on
spent
the
on
appropriate:
tasks
three
for
which
that
method
was
question.
related
to be
said
designing, coding and documenting the program.
Independent variables were developed for each part of each
hypotheses.
time
cases
In some
the
The
responses
to
several
independent variables are given below, in the
order of the hypotheses which they which they are used to test.
Hypothesis 1:
the
variable is the response to a single
In others it is a scale combining the
questions.
of
Organizational Factors
two
with
measured
Attitude,
Client
1.1
a)"My client users want
agree-disagree questions:
maintainability more than efficiency in code."
b)"If and when I use the new system development
technologies...my client users appreciate my work
more."
two
with
measured
Attitude,
Supervisor
1.2
agree-disagree questions, one for each methodology:
(over the past year) would have
a) "My supervisor
liked me to use Ml." b) "My supervisor (over the
past year) would have liked me to use M2."
1.3 Organization Attitude, measured with a scale
comprising five agree-disagree items, reliability
tested at the alpha = .58 level (Questionnaire items
38.1,13,15,20,28; see Appendix A).
Hypothesis 2:
Individual Attitudes
Structured
toward
Attitude
General
2.1
Methodologies, measured with a scale comprising five
agree-disagree items, reliability tested at the
items
(Questionnaire
level
.56
=
alpha
38.2,11,19,21,24).
2.2 Attitude toward Local Methodologies, measured
with a scale comprising six agree-disagree items,
reliability tested at the alpha =
Individual
3:
Hypothesis
Experience and Orientation
Technical
Ability,
3.1 Skill in Systems Development, measured with a
five point self-rating on skill in programming in
Fortran and PL1.
in Methodologies, measured with a five
3.2 Skill
point self-rating on skill in Ml and M2.
3.3 Technical Experience and Orientation, measured
a) education as a
for each factor as follows:
or
technical
dichotomous variable representing
nontechnical college degree, b) experience as years
of experience in systems development, c) orientation
as dichotomous variable representing responses to
question "In general do you see yourself
the
pursuing a managerial or technical career path?".
Hypothesis
Productivity
4:
Opportunity
for
Increased
4.1 Type of Task, measured by per cent of time spent
on
tasks
at each of four levels of systems
development.
4.2 Complexity of Communications, measured by
of work group and length of current project.
The independent
related
to
variables
above
are
to
thought
However, before
any
of
tests
statistical
are used it is necessary to examine more c losely the logic
significance
of the situation in which the
the
entire
sample
respondents
methodologies.
were
workin Ig to
determine
a subset is the relevaInt group.
or
possible that not all the analysts had
the
opportunity
to
It is
use
these
In this event, it would be of interest to find out what
distinguished those who had the opportunity from those w,ho did not.
analyzing
be
It is the purpose of this study to
the dependent variable.
explore that relationship.
whether
summarized
size
relationships
of
other
fact ors
to
extlent
of
methodologies, it would be advisable to limi t the analys is to
In
use
of
analysts
who had a meaningful opportunity.
There are two variables which determine the opportunity of analysts
to use the methodologies.
source
of
skill
development
either trained or skilled
although
One is access to training, or to
in the technique.
in the
methodology
some
other
An analyst who is not
will
not
be
a user,
it is likely that skill will be further developed through use
The other opportunity variable is time spent on a task
on the job.
the
which
are
methodologies
of
use
are
they
however,
organization,
the
for
conditions
necessary
training program conducted by the firm.
Since
by
approval
to
rather than a
control
the
job
the
are
methodologies
solely,
dependent
on
the
systems
is responsible
are
closely
and this is largely an organizational
assignment
personal
in the
the individual
Similarly,
supervisor.
a
development tasks for which an individual
related
which
if not
is primarily,
training
or
assignment
over
In this
is no opportunity to get training outside the firm.
there
And access to
variables are
methodologies.
the
factors
two
Skill is most commonly acquired in a formal
analyst has little control.
proprietary,
These
appropriate.
for
(The analyst
decision.
does
some
exercise
process, for example by turning down an opportunity to
Several respondents reported that when
attend training sessions.
their
turn to attend training came they were "too busy" to go.)
There is evidence that a substantial proportion of the sample
not had training in the methodologies (Table 7).
respondents
report
For Ml, 38 per cent of
having had a day or less of training.
which has been in use for four years or more, 19 per cent
training
or
only
one
day
of
training.
have
Even for M2,
have
had
no
There are, of course, some
respondents who have not had training but have developed skills on their
own or learned informally from others.
group
having
criterion.
opportunity
to
use
the
These have been included in the
methodologies, on the training
6,3
Table 7
Training in Methodologies
Methodology
M1
Mpthodol
Days
Training
None
1
3
4
5
Missing
Total
Number
of Cases
51
3
10
42
35
145
apercent of Non-Missing Data Only
100
17
2
8
46
14
12
None
1
2
3
4
5
10
Missing
Total
Percenta
1
145
100
with
To distinguish between opportunity and no-opportunity groups,
respect
training,
to
a
created which was based on the
was
variable
amount of training as well as the self-reported skill in the use of each
a) the analyst
individual met at least one of the following conditions:
having
reported
had
at
least
if that
opportunity
A respondent was considered to have
methodology.
one day of training, or b) the analyst
least
reported being skilled at a level of at
a five
on
four
point
scale.
An analyst who met this test in regard to MI was included in the
subset
of
used for M2.
offered
Wh ile this aplproach may exclude
trainin
and
did
in d efining
conservative
b etw een
relationships
The same approach was
sample used in exploring M1 use.
the
not
who
analysts
were
it has the advantage of being
attend,
opportunity.
By
if
that
It assures
opportunity.
in defining
some
any
controlling for
training/skill, it assures that those analysts for whom extent of use is
a
as
measured had at lea st a familiarity with the methodologies
basis
for the decision to use or not to use them.
The other
methodologies
opp ortunity
on
time
on
for
tasks
which
the
a re appropriate, is taken into account in the definition
of the dependent variable.
working
variable,
It is assumed
that
anyone
who
reported
one or more of the tasks for 5 per cent or more of the time
had the opportunity to use the methodology in question.
The following chapters make use of the dependent,
opportunity
findings.
variables
discussed
here
to
independent
and
analyze the data and report
IMPLEMENTATION
The use of structured systems development methodologies by analysts
at various stages in the systems development
process
is described
first
section
presents
The
parts.
in four
chapter
this
describing the extent of use of Ml and M2 in the tasks
are
for
in
results
which
they
appropriate and discusses the logical grounds for the creation of a
single variable to measure the extent of use of each
second
section
describes
In third
The
analysts'
access
to
training and explores
the
amount
of
training
factors which may be related
receives.
methodology.
to
fourth
and
sections
the
regarding the benefits and disadvantages of each
an
analyst
analysts' perceptions
of
the
methodologies
are presented and the stated reasons for using alternative methods, when
M1 and M2 are not used, are summarized and discussed.
The results presented in this chapter are primarily descriptive
the
aspects
study.
and
of
implementation
which data were gathered in the
The descriptions of extent of use, access to training,
disadvantages
and
background necessary for
factors
about
to
reasons
for
understanding
implementation.
using
the
benefits
an alternative provide the
relationships
of
various
In Chapter 5 a more analytic approach is
taken in identifying specific factors thought to be associated with
extent of use.
of
the
in the
Extent of Use. Each of the two methodologies was described
guidelines for systems development in the organization as applicable for
in three specific tasks.
use
In describing the extent of use of MI and
M2 we would like to know, first, whether analysts worked on these tasks.
If they worked on one or more task and therefor were eligible to use the
methodology, we are interested in whether they did in fact use
to
what
extent.
it, and
Further, it will be important to find out whether the
extent of use differed across tasks or whether analysts who
were
users
on one task in a set tended to be users on all three tasks in a set.
In
the latter case, it will be possible and logically justifiable to create
a
single extent of use variable for Ml and another for M2 averaging the
extent of use measures across the three tasks for which each methodology
is appropriate.
Extent of use is given separately for each methodology on
These
data
sets
8
show responses to a question on extent of use, condensing a
five point scale into three groups.
both
Table
of
tasks,
the
Several patterns
stand
out.
For
large majority of respondents were eligible
users as judged by the amount of time they had spent on the task (5 per
cent
or
more).
For each of the Ml tasks, more than 85 per cent of the
sample were eligible users.
For each of the M2 tasks, more than 70
cent of the sample were eligible, by this criterion.
per
68
Table 8
Extent of Use, by Task
Ml Task a
Define
Client Needs
Understand
The Business
Number
Evaluate
Alternative
Percentb Number
Number
Percentb
Always or
frequently
Sometimes or
infrequently
Never
Not Applicable c
or Missing Data
44
34
32
25
22
17
28
22
29
23
35
27
56
44
66
52
67
52
17
-
18
-
21
TOTAL
145
100
145
100
145
Extent of Use
Percent b
100
M2 Taskd
Extent of Use
Always or
Frequently
Sometimes or
Design Program
Code Program
Number
Number
Percentb
Number
Percentb
47
41
33
30
42
40
26
23
26
24
21
20
46
42
Infrequently
Never
Not Applicablec
or Missing Data
Total
Percentb
Document Program
-
3
10o
14
100
40
145
100
aTasks listed are those for which Il may be appropriate.
bPercentages are based on adjusted totals, not including "Not
Applicable" or Missing Data.
CNot Applicable: respondent spent less than five percent of working
time on this task.
dTasks listed are those for which 142 may be appropriate.
However, given membership in the population of
eligible
users,
a
relatively large proportion of analysts reported that they never use the
per
50
almost
methodologies:
of
cent
eligible
in this
approximately 40 per cent of eligible M2 users were
It is possible
by
the
category.
that MI is little used by some analysts because it has
has
M2
been approved for use relativly recently.
approved
and
users
Ml
been
and
available
organization for a number of years, however, so other
factors are likely to be important.
be
Extent of use appears to
within
each
grouping.
consistent
relatively
tasks
across
For example, the proportion of respondents who
the
never use Ml on the task "Understanding the Business" is similar to
who
proportion
never
use
MI
on "Define Client Needs".
To test this
relationship, Pearson Correlations were computed within each cluster
The
tasks.
result
indicates
a
of
relationship among responses
strong
within each cluster, with a correlation of .7 or higher at the .01 level
of significance.
generally
heavy
Analysts who are heavy of
users on the other two tasks;
on one task are likely not to be users on
consistent
Ml
the
on
any
one
those who are not users
other
evaluating
of alternatives.
not spent in discrete blocks.
two,
and
other two.
is
Analysts in their
work integrate their understanding of the business, defining
and
This
tasks.
with the impression given by analysts in the interviews that
the tasks within each cluster are closely associated.
needs
are
task
of
client
The time spent on the tasks is
Work in one area weaves among
the
other
techniques used in one task tend also to be those used in the
Based both on the high Pearson Correlations and on the logic
of
use
of
of the situation, it makes sense to average the reported extent
MI across the MI tasks, and likewise to average the extent of use of
M2.
to
criterion,
time-on-task
Since many analysts were eligible, on the
both MI and M2 there was a possibility that analysts had common
use
clusters
patterns of use across the two
tasks.
of
words,
In other
analysts might be either high or low users, regardless of the particular
the
again
statistic was used to test the
Pearson
methodology.
Once
relationship.
Using the averaged measure of
methodology,
a
of
Correlation
Pearson
extent
there
of
for
use
is a
each
significant
relationship between use of Ml and use of M2, given eligibility
to
use
Some analysts are high users of structured systems methodologies
both.
in general, while others tend to be lower level users or non-users.
The
relationship here appears to be not as strong as the relationship within
a task cluster, however.
Access to Training.
organization
Formal
training
sessions
conducted
by
the
are the primary source of skill acquisition for Ml and M2.
In Chapter 3, Table 7, the frequency distribution of training in each of
the two methodologies was given.
analysts
had
two
similarly in M2 (80
minortiy
to
per
five
days
cent).
It indicated
of
There
that
training in MI
was,
majority
the
of
(62 per cent) and
however,
a
significant
who had not received training, particularly in Ml (38 per cent
with one day or less).
In another study conducted as part of this project, McErlean (1983)
training
formal
both
of
has explored in detail the relationship
and
access to consulting and advocate support to the extent of use
informal
of the methodologies, finding significant relationships between training
in the previous chapter of this study, that access to
argued,
further
It was
other.
the
on
and support on the one hand and extent of use
training should be considered an opportunity variable in that an analyst
without skill in M1 could not be expected to use Ml,
M2.
From
for
points of view, training is an important aspect of the
both
implementation process.
the
likewise
and
effectiveness
of
related
Access to training is likely to be
and
innovation
the
to
understanding of which
an
analysts get training may be useful in interpreting analysts' assessment
of the benefits of the methodologies.
analysts
which
To provide some understanding
received training, and of how training and skill in the
methodologies are related, this section will explore
between
level
of
of
relationships
the
and a number of factors characterizing the
training
analysts.
It is conceivable
organizational
factors
regarding
the
attitude
methodologies.
of
the
background
be
related
to
It is possible,
(technical
orientation, or the analyst's age.
and the results are reported below.
or
These
supervisor
respondent's
training may be related to individual variables such
educational
may
as location, respondent's position in the
such
organizational hierarchy or
training
to
access
that
too,
as
nontechnical)
possibilities
that access to
analyst's
the
or
were
career
explored
in days
The relationship between access to training, measured
several
and
training,
organizational
Pearson Correlation Coefficient.
factors
For MI training, no
found between level of training and location.
with
Title
Job
was
using the
tested
relationship
(a seven point scale representing the hierarchical job
level.
The
level analysts were more likely to have received training.
senior
This result may be due, in part, to the fact
in senior
analysts
that
tend to be older (Pearson r of .53, at the .00 level, for the
positions
relationship between age and Title) and therefor to
have
in the
been
organization longer and more exposed to the opportunity of training.
the
was
However, the relationship
level within the organization) gave an r of .32 at the .00
more
of
On
other hand, it may be that the organization trains its higher level
analysts to a greater extent than it trains those at lower levels.
Another organizational factor related to training was the degree to
which the supervisor wanted the respondent to use MI.
consistent
with
the
reasoning
This
finding
is
that it is the supervisor who is a key
decision maker in the process of allocating training resources among the
analysts.
An
exploration
organizational
factors
relationships
the
of
and
training was found to vary to
between
these
M2 did not result in similar findings.
a
small
degree
according
to
same
M2
location
(McErlean, 1983) but to have no significant relationship to Job Title or
to the attitude of the supervisor to the use of M2.
be
These findings must
interpreted in light of the fact that the great majority of analysts
had training in M2 at more than the minimum
level.
The
organization,
policy or at the level of an individual supervisor's decisions,
through
is likely to influence the initial access to training,
have
analysts
that
access
M2
to
but
all
almost
Beyond initial access,
training.
variation in the number of days of training may be
influenced
by
more
other factors than by the organizational factors measured here.
Two other organizational factors were examined and found to have no
relationship to either Ml or M2 training.
Analysts in large groups,
group and the length of the project.
on
long
projects,
thought
to
to
improve
communications
coordination
effective
across
working
potentially more use for a
were
designed
methodology
They are the size of the work
have
and
allow
to
and through time.
individuals
were true, one would expect the organization to
methodology in order to increase productivity.
encourage
use
more
If this
of
the
One way to encourage use
in these situations would be to attempt to provide training to analysts
in methodologies appropriate
here.
The
that
fact
training
to these factors suggests that either the organization does not
related
currently act in this way or that other variables mask
a
long
project.
the
use
on
which
MI
and
M2
were
not
of Ml or M2 now might not be appropriate.
case, a rational supervisor might decide
training.
results
of
If the thrust of that project were to maintain or
enhance an existing application
initially,
the
For example, suppose an analyst had an assignment to work
such actions.
on
is not
not
to
send
an
used
In that
analyst
to
or
Access to training may be related to an individual's background
Ml
both
For
orientation.
technical
and
possible
training,
M2
were
relationships with age, technical education and career orientation
was found to be significantly related to Ml training (r =
Age
tested.
.21 at the .01 level) but not to M2 training.
to
found
training.
negatively
be
related
Technical
education
was
to Ml training but not related to M2
Those analysts who have a bachelor's or a master's degree
in
computer science are less likely to have had Ml training than those with
It is possible that these analysts do
degrees (Table 9).
non-technical
not feel the need to acquire skills in Ml, or that
are
concerned
their previously learned skills will become obsolete if they learn
that
Ml and are therefor resistant to change.
the
they
organization
On the other hand,
if it is
which is making decisions about access to training it
is likely that supervisors are assessing the need of technically skilled
analysts for M1 as being relatively low and are therefore not
assigning
these analysts to training.
The final individual factor explored in relation
with either Ml
or
training
was
In this case no significant relationship was found
orientation.
career
to
M2
training;
i.e.
analysts
oriented
toward
a
technical career are no more likely to have received training than those
oriented toward a career in management.
One
aspect
of
training
worth
investigating
further
relationship of training level with skill level for M1 and M2.
shows
that
is
the
Table 10
as expected the relationship is strong and positive for MI,
but contrary to expectations it is not nearly as strong for M2.
It is
75
Table 9
Access to Training, by Educatioa
M2 Tratnfn!
Hl Training
1+ Days
0 Dhys
0 Days
1+ Days
N
I
N
I
Nontechnical'
17
35
53
65
10
48
59
55
Technicalb
31
65
28
35
11
52
48
45
Total
48
100 81
100
21
100 107
ducation
100
M2Statistics
M1 Statistics
K "128
" 129
Hissing Cases * 16
Missing Cases - 17
Chi Square
- 9.8
Chi Square
Significance
- 0.00 Significance
- 0.2
- 0.7
"aBachelor's or Easter's, other than Systems Developýent.
or
o.S.
H.S., Systems Development.
Table 10
Skill Level by Training
Days of Trairing
12
O0
1+
0
Low
96
65
72
Levele
High
4
25
28
100
90
100
Total
Hissing, NIA
Chi Square
Significance
n -I
18
75
69
60
6
25
46
40
24
100
115 100
112
Statistics
N
%
n-X
Skill
1+
Z
141
139
4
6
12.6
0.00
2.1
0.35
aSkill Level based on self-rating on a 5-polnL scale, 1 to 3 low, 4
to 5 high.
)licsing, N/A includes respondents not ansvering this Itec.
striking that a large number of respondents (65 for MI and
particularly
31 for M2)
report
that
are
they
of training used here was relatively liberal
standard
company
the
either
whom
a
training
was
not
This latter phenomenon was reported in a
that
number of interviews by analysts who explained
at
Among these
training took place so long ago that the skills developed
or
in training have deteriorated.
given
in the
(one day or more), compared
the M1 training of 4 days.
for
analysts, however, are likely some for
effective
skilled
not
In part this finding is due to the fact that the measure
methodologies.
to
but
trained
time
the
training
was
when there was no immediate opportunity to apply the
skills learned, so
that
the
skills
were
eventually
forgotten.
In
addition, several respondents commented that the training program itself
would
benefit
the introduction of more practical examples in the
from
curriculum and from a closer integration of MI or M2 techniques with the
work in which trainees were currently engaged.
Also of interest in these data are the respondents who are
but
not
trained.
These
analysts (2 for MI and 19 for M2) are living
proof that the methodologies can be learned outside the formal
program,
suggests
that
those
The
existence
effective
increase
their
ways
to
use
of
increase
the
of
analysts
in this
in the organization concerned with more
effective implementation of the methodologies, and
more
training
when the methodology has been in use over a number
especially
of years as is the case with M2.
category
skilled
the
in particular
with
level of skills, might plan to
non-formal
training,
support
and
communications systems through which some analysts are currently getting
use
The
development.
skill
skill-refresher
may
techniques
tutorials and other support
providing
of consultants, local advocates, on-line
activities
in particular,
help,
in
which may address the problem of
the deterioration over time of the effects of formal training.
the
Benefits of the Methodologies. One of
underlying
individual analysts, acting rationally within the
is that
study
this
assumptions
own
their
and
abilities
opportunities, will assess the
work-related
improve
value of each of the methodologies and the potential of each to
analyst
the
reports
section
This
effectiveness.
their
by
boundaries set by the organzation and the constraints imposed
These
disadvantages of the methodologies, as perceived by the analysts.
technology
computer
dissemination.
creation and
to
methodologies
group
development
find
common
which is responsible for their
compared
results
are
also
advantages
and
disadvantages
The
the
by
methodology
results are compared with the claims made for each
and
benefits
and
Finally
identify those benefits which are specific to M1 and M2.
across
to
those
benefits not frequently chosen will be reviewed and the implications for
the implementation process discussed.
On Table 11 the four most heavily weighted advantages of M1 and M2,
as perceived by respondents, are given.
listed
for
MI
are
defining
the
four
advantages
benefits related to the central tasks of the early
stages of systems development.
accurately
Three of
the
Understanding
requirements
the
client's
business,
of the system and communicating
effectively are all necessary parts of the analysts work.
As
indicated
Table 11
Principal Advantages of the Methodologies According
to the Survey Respondents (N = 145)
Weighted
Weighted
MI Advantages
Rankinga
1. Helps me understand
clients' business
3.7
i. Provides
structured
design
3.8
2. Structured approach
4.7
2. Structured
approach
4.9
3.
4.8
3. Application
more
maintainable
4.9
5.6
4. Provides
better
documentation
5.1
Improves requirement
definition
4. Provides improved
communication
M2 Advantages
Rankinga
aRespondents were asked to select and record in ranked order, the
three most important advantages of each of these methodologies.
Responses were weighted by their rank order (e.g., third ranked = 3);
responses not selected were arbitrarily assigned the average ranking
of 8 on the assumption that responses not selected, if ranked, would
be randomly distributed among all possible remaining rankings (i.e., 4
to 13); the rankings were then aggregated and averaged. The smaller
the number recorded here, the higher the rank, on average, given to
this response.
Range 1 - 8.
on
The fourth advantage given by
advantages of the methodology.
approach",
"structured
of MI agree with this assessment of the
developers
the
12,
Table
descriptive of the methodology itself
is more
than suggestive of the benefits it brings
noting,
to
that structure is seen as an advantage by respondents
however,
orientation
general
the
It is worth
analysis.
rather than an irrelevant factor or a disadvantage.
that
analysts,
analysts
of
The implication
to
is favorable
structured approach to the early stages of systems development:
clearly
a
Ml,
a more
MI
is
structured approach, and respondents see that characteristic
as an advantage.
of
is
The developers do not list structure as
advantage
an
likely in that they take for granted that structure is an
most
advantage since
it is the
central
characteristic
of
both
of
the
methodologies.
Advantages given for M2 are similar to those given for MI
both
structure
and
two more specific advantages are heavily weighted.
The implications of the listing of "structure" as an advantage
same as above.
design,
coding
are
the
The other items address specific desired outcomes of the
and documentation sequence of work which takes place at
the later stages of systems development.
The mention of maintainability
is interesting in that we know that this feature is a major
managers
in that
concern
of
attempting to improve systems development productivity, yet it
is not a characteristic which is easily observed, described or measured.
An application or a system can be demonstrably maintainable only in the
long
run.
While
it would
be
relatively
documentation", and thus to know whether
M2
easy
to, observe "better
contributes
to
a better
80
Table 12
Principle Advantages and Disadvantages of the
Methodologies According to Their Developers
MI Advantages
M2 Advantages
1. Provides improved communications
1. Application more
maintainable
2. Helps me understand client's
business
2.
Provides better
documentation
3. Improves requirement definition
3.
Provides structured
design
M2 Disadvantages
M1 Disadvantages
Time Consuming to use
1.
Time cosuming to use
2. Too expensive for client
(time or budget)
2.
Incompatible with
programming language I
1.
usea
3.
Unfamiliar
3. Not oriented to my
applicationa
___
aln these cases, the developers chose as disadvantages, ones that
actually ranked quite low on the respondents' lists, and left out ones
the respondents felt were important.
product , it may
documented
demonstrate
to
difficult
be
a similar
relationship betwee n the use of M2 and the maintainability of a product.
It is a
fact
is intended
M2
that
to
provide
years duri ng which M2 has been in use increased maintainability
However, of
has been observed.
appears
The
It may even be that over the
developers make the claim here (Table 12).
several
benefit.
this
less
like ly
the
all
given,
advantages
this
one
to be grounded in the immediate experience of the
respondents and mor e likely a restatement of generally accepted opinion.
by
The two most important disadvantages perceived
respect
to both
M1
and
M2,
analysts,
their lack of familiarity with the
are
methodologie s and the time it takes to use these techniques (Table
Time
is a
central
familiar wit h a new approach and to develop skill in applying
analyst
is aware
it.
in this
field
since
work
(Budget, itself, transl ates
is completed on time and within the budget.
time
The
of the importance of time in that one of the pri mary
criteria use d in performance evaluation is the extent to which the
into
13) .
in that it requires time to be,come
here,
concern
with
it is the labor cost of analysts and
programmers which constitute the largest proportion of variable costs of
any project.)
section
The issue of familiarity was addressed
in regard
to
access
and
to. training
in the
previous
the need to schedule
training in a way more closely coordinated with use of the methodologies
on a project.
Here analysts are reporting directly that they
recognize
that their own lack of familiarity with the methodologies as well as the
time
demands of use both cause problems.
concern for the fact that the
client
will
For Ml, in addition, there is
be
required
to
bear
the
Table 13
Principal Disadvantages of the Methodologies According
to the Survey Respondents (N
=
145)
Weighted
Rankinga M2 Disadvantages
M1 Disadvantages
Weighted
Rankinga
1. Unfamiliar
2.2
1. Time consuming to
use
3.8
2. Time consuming to
use
3.0
2. Unfamiliar
4.0
3. Too expensive for
client (time or
budget)
5.6
3. Not useful for
maintenance
5.3
4. Not oriented to
my application
5.8
4. Restrictive
(inhibits
creativity)
5.7
5. Not useful for
maintenance
5.9
aRespondents were asked to select and record in ranked order, the
three most important disadvantages of each of these methodologies.
Responses were weighted by their rank order (e.g., third ranked = 3);
responses not selected were arbitrarily assigned the average ranking
of 7 on the assumption that responses not selected, if ranked, would
be randomly distributed among all possible remaining rankings (i.e.,
4 to 11); the rankings were then aggregated and averaged. The smaller
the number recorded here, the higher the rank, on average, given to
this response.
Range 1 - 7.
increased expense due to use of the methodology.
Another disadvantage which analysts mention
methodologies
is the
maintenance.
or
to
previously
developed
systems,
whether
to
correct
enhance the systems, M1 and M2 cannot in general be used
an
excessive
original
amount
design
even
for
of
of extra work.
necessary to retrofit an existing system
prohibitive,
both
perception that neither M1 nor M2 are useful for
unless they were use in the
undertaking
to
Analysts reported in the interviews that on projects which
are building onto
errors
in relation
relatively
with
large
and
M1
the
system
without
The cost of the work
or
M2
is seen
as
somewhat self-contained
components newly designed to augment systems currently in operation.
In
this area there appears to be signif icant latitude
for
or
project
or
leaders
to
make
the
dec ision
to
use
individuals
not
to
methodologies, and most individuals, at least, are reluctant
use the
to
commit
themselves to the added costs of use
It is clear that the developers disagree with the analysts on
point.
The item "not useful for mai ntenance" is noticeably missing from
their list of disadvantages.
M1
and
this
one
While their other listed disadvantages for
of their disadvantages for M2 are the same as those of the
analysts, they do not share the pers pective of the analysts on the issue
of
maintenance.
reasonable
There
compromise
may
between
beyond the scope of this study.
be
a
the
correct
two
perspective
positions,
or
a
but to find it is
In practice, we know that use of MI and
M2 in the context of a maintenance project is a judgement
making
here,
the judgement are the analysts.
call.
Those
Their assessment, in this case,
is the one which is the most directly related to behavior (extent of use
of the methodologies) and is for that reason worth noting.
disadvantages
Underlying the assessment of advantages and
methodologies
One of
set of questions about costs and benefits.
is a
the
of
the missing pieces in this puzzle is the actual cost of using MI and M2,
and the actual advantages of
improvement measurement issue.
using
each.
This
is the
productivity
While no specific measures are suggested
here, there is in these findings a suggestion of a possible next step in
developing
a
such
measure.
The step is to make a distinction between
long and short term costs, and similarly long and short term benefits.
these
MI
using
Analysts are concerned about the costs of
concerns may well reduce their extent of use.
The costs they are
client
considering are short term costs, essentially the charges to the
on
This is
an applications development or systems development project.
incentives
not surprising, since analysts are on the whole given
on
term
short
their
and
M2,
and
based
They recognize the benefits of the
production.
structured methodologies and will most likely use the methodologies when
However
the perceived benefits are greater than the perceived costs.
number
of
benefits
the
identified,
as
well
as
some
a
noted by the
developers but not weighted heavily by analysts, are long term benefits.
extent
that
operationalized
and
To
the
analysts
can
disadvantages,
be
these
long
measured,
developed.
a
The
term
modified
balance
advantages
incentive
between
are
defined,
structure
advantages
for
and
perceived by the analysts, can be altered and the extent
of use of the methodologies can be increased.
for
alternative approaches, in those cases in which they did not
using
Their
use the structured methodologies.
that they did not have
had
been
ss
roe 10
implementat ;^n
to
learn
willing to pay
i
Analysts
projects.
for
for
increaisinal
decreasing
the
thei
time
the
high
The
underscores
This
the
to
use
than
the
form
of
lieve
of
this
difficulty
M1
in
more
is
the organization is aware or is
additional
"overhead"
time
on
there is an experience curve, a potential
skill
level
required
for
with
additional
effective
use
dependent on the organization, however, to cover the
thought
of
use
for
perhaps
that
ests
it
practice
reason
this
of
seriousness
su
findin
rank
to
or
properly
M1
learn
to
time
trained.
alternative approaches
difficult
the
of
use
Respondents often expressed in the interviews the feeling
methodology.
they
in the
experience
and
skill
was the lack of training,
when
on
summarized
are
responses
For M1 the most frequently mentioned reason, by far,
14 and 15.
Tables
reasons
their
Reasons for Using Alternatives. Analysts were asked
experience
of
Ml.
cost
of
and
They are
what
is
of, in effect, as on the job training necessary for them to get
that experience.
The next two reasons, in order of relative frequency,
to
the fit between MI and the analysts' work.
are
related
Ml is seen as too formal
or too detailed for use on the tasks for which they are responsible.
It
is not perceived to be justifiable for use on relatively small projects.
In both cases the underlying concern seems once again to be one of cost,
in that the analyst must spend extra time to use MI in relation
to
the
6
Table 14
Reasons for Using Alternative to M1
Task
Understand Business
Reasons
a
Define Client Needs
Number
Percentb
Evaluate Alternatives
Number
Percentb
Number
Percentb
Respondent not sufficiently trained/
skilled/experienced in M1
39
38
28
26
38
M1 too formal/detailed
19
19
29
28
15
Project too small to justify M1
12
12
13
13
14
Precedent set earlier in project
15
15
8
8
7
M1 not suitable/compatiablewith hardware
or software
4
4
9
9
16
Alternative saves time
6
6
7
7
7
7
Other
7
7
12
12
7
7
Not Applicablec or Missing Data
43
37
41
39
Total
145
100
145
100
aTasks listed are those for which Ml may be appropriate.
bpercentages calculated on the basis of adjusted totals, not including "Not Applicable" or
Missing Data.
CNot Applicable:
Respondent spent less than 5% of working tiome on this task.
145
100
-
100
Table 15
Reasons for Using an Alternative to M2
Taska
Design Program
Reasons
Code Program
Percentb
Document Program
Number
Percentb
13
19
15
9
13
10
9
13
M2 too formal/detailed
9
11
16
Precedent set earlier in project
9
4
6
14
12
18
Percentb
Number
M2 not suitable/compatible with
hardware or software
Respondent not sufficiently trained/
skilled/experienced in M2
Alternative saves time
23
17
Number
17
-17
10
1
13
Project too small to justify M2
4
Other
22
Not Applicablec or Missing Data
70
Total
145
5
11
76
66
100
145
100
aTasks listed are those for which M2 may be appropriate.
bpercentages based on adjusted totals not including "Not Applicable" or Missing Data.
CNot Applicable:
respondent spent less than five percent of working time on this task.
145
100
approach,
alternative
time which is apparently not seen as justifiable
given the requirements of the project ("too formal") or the size of. the
too
("project
project
The
small").
analyst is held responsible for
cost, and this responsibility provides the
offer
approaches
better
definition
communications,
remains a favored approach.
areas,
these
of
business, and so forth.
client's
however,
between
solutions
within
reference
use
not
of
MI
or an
that
indicate
to the underlying problems:
understanding
requirements,
the
of
In these areas we must assume that Ml
in
The potential benefits attributed to Ml
are not observable in the short run and are not
outcomes for which analysts
choosing
the
regarding
The reasons for using alternatives do
alternative.
other
decisions
makes
she
or
he
which
of
frame
will
be
in
words,
In other
rewarded.
Ml and the available alternatives analysts are acting
according to short term incentives, although they
were
that
recognize
to use Ml more extensively they might contribute to an increase in
they
the effectiveness of systems development in the long run.
applications
In the case of M2, incompatibility with
the
with
hardware
environment
in which
the
software
application
is being
developed replaces lack of training as the most commonly sighted
for
use of an alternative approach.
or
reason
However training is a close second
in rank, a surprising result in that such a large proportion of analysts
is trained in M2.
Once again it appears that either the training is too
distant in time from the implementation or the job does not provide
chance
to
the
practice and gain experience in using the M2 skills learned.
The other reasons are similar to those for Ml,
and
the
interpretation
above appears to apply here too.
In gathering
the
data
section,
this
for
left
were
questions
relatively unstructured in hopes of allowing for more open expression of
on
concerns
the
part of analysts and in an attempt to shed light on a
complicated decision making process.
of
value
The
this
data,
more
qualitative and less suitable to statistical analysis, is in the insight
it may provide into the implementation process.
The data indicate that
analysts are making decisions about both MI and M2 on
number
of
criteria
related
dollars, and compatibility.
reasons
primarily
To the
to
extent
cost,
that
the
basis
including
these
are
of
time and
the
real
for choices of alternatives, and we have little if any evidence
to indicate that they are not, it is likely that appropriate changes
the
a
in
management of this implementation will result in an increase in the
extent of use of the methodologies.
FACTORS RELATED TO EXTENT OF USE
The implementation of change at the level of an individual
is
factors.
varic )us organizational ,
to
related
In this study
',
represerIt each
to
variables
four
hypotheses
these
of
and
results of the tests ar"e reported
reports
relationships
supervisors
the
and
methodologies.
found
individual
organization
and
extent
the
orientatic)n are
of
clients,
of
of
use
associated
with
education,
use.
the
experience
The final section
presents data on two selected opportunity variables, type
complexity
section
The third section reports and
discusses the ways in vihich analysts' abi lities,
career
of
first
In thie second section, relationships between analysts'
attitudes and extent of: use are reported.
and
The
attitudes
the
selected
In this chapter the
factors.
d iscussed.
opportunity
using
tested
were
between
and
analyst
of
task
and
project, in relation to use of structured methodologies.
In each case, discussion of the findings is included with the report
of
results.
Hypothesis 1:
Organizational Support. There is a strong indication
that the level of organizational support perceived by an analyst in the
form of favorable attitudes of clients, supervisors and the organization
in general
toward MI is associated with more extensive use of Ml.
same relationship appears to be present for
perceived supervisor support.
M2
only
in the
case
The
of
Client and general organizational support
are
(See Chapter
not significantly associated with more extensive use.
2 for explanation of attitude variables.)
1.1 Client attitudes favorable to the use of the methodologies will
16
Tables
be associated with greater extent of use.
indicate
17
and
that the extent of use of M1 is significantly greater when the client is
appreciating
as
seen
preferring
as
perceived
the
of
efficiency
structured
methodologies and when the client is
maintainability
No
code.
over
application
an
of
similar relationships are found between
these client attitude measures and the extent of use of M2.
Discussion:
work
development
with clients.
systems
It is analysts engaged in the early stages of
who have the most frequent and most intensive contact
The analyst and the client are dependent on each other in
the process of representing the business accurately, defining the
of
business
the
and
recommending
applications software package,
product.
In this
develop respect for
perhaps,
an
a
solution
information
system
the
the
attitudes
of
the
client.
M1
some
other
More
important,
the analyst is dependent on the client's satisfaction with the
analyst
is willing
to
It is likely
act on the client's preferences for
certain product characteristics, if there is a way to do
case,
or
an
close working relationship the analyst is likely to
relationship in keeping his or her own supervisor happy.
that
of
form
in the
needs
so.
In this
is a technique which promises to deliver the product with the
attributes the client wants.
92
Table 16
Extent of Use by Client's Appreciation
Extent of Usea
M2
Ml
~~~
Moderate
n
%
Client's
Appreciationb
Low
28
66
High
n
%
23
54
8
26
High 14
33
24
77
20
46
23
74
Total 42
100
31
100
43
100
31
100
NC
74
74
Missing, N/A
71
71
14.7
4.6
Significance
%
23
M2
Chi Square
High
n
7
M1
Statistics
Moderate
n
%
0.00
0.03
aExtent of Use based on 5-point scale, 1 to 3 moderate, 4 to 5 high.
bClient's Appreciation based on responses to an agree-disagree
question with a 5-point response scale, grouped 1 to 3 low, 4 to 5
high.
CN includes only those respondents with 1 or more days training or
methodology skill level of 4 or 5, on a 5-point scale.
93
Table 17
Extent of Use by Client's Preference
for Maintainability Over Efficiency
Extent of Usea
M2
M1
Client's
Preferenceb
Statistics
n
%
n
%
8
26
24
44
11
46
46
23
74
30
56
13
54
100
31
100
54
100
24
100
%
23
54
20
Total 43
High
MI
High
%
n
Low
Moderate
High
Moderate
n
M2
NC
74
78
Missing, N/A
71
67
Chi Square
4.6
0.0
Significance
0.03
1.00
aExtent of Use based on 5-point scale, 1 to 3 moderate, 4 to 5 high.
bBased on responses to an agree-disagree question with a 5-point
response scale, grouped 1 to 3 low, 4 to 5 high.
CN includes only those respondents with 1 or more days training or
methodology skill level of 4 or 5, on a 5-point scale.
who
analysts
For
programming
later
at
work
of
the
development,
stages, contact with the client and interdependence between
analyst and client are less intense.
than
important
stages
well
how
the
What the client wants may be
less
and
cost
because
they
program
Technical
works.
considerations may be relatively more important, in part
For these reasons, the clients'
are more easily observed at this stage.
attitudes may matter less.
this
Perhaps it is
hypothesis and test there is no proof of causation.
attitudes,
in the
resulting
clients'
who
influence
their
association
discussed
above.
analysts who are extensive users of
MI
understood that this possibility is present
in all
In
way.
other
(It is possible that the association may work the
It is
relationships
the
explored here, although it will not be explicitly mentioned in each.)
methodologies
1.2 Supervisor attitudes favorable to the use of the
will
be
associated
with
more
extensive use.
The relationship found
between supervisor's attitude and use of both Ml and M2 was positive and
highly significant.
methodologies,
When the supervisor did
hardly
any
analysts
not
support
used the methodologies.
supervisor was seen as wanting analysts to use MI or M2
use
of
each, respectively, was higher.
number of analysts who were not heavy
supervisors wanted (Table 18).
the
the
of
use
When the
extent
of
There were still a substantial
users
in spite
of
what
their
95
Table 18
Extent of Use by Supervisor's Orientation
Extent of Usea
M2
Ml
Moderate
Supervisor's
Orientationb
Low
High
n
%
4
13
28
52
42
27
87
26
100
31
100
54
%
25
58
18
MI
Statistics
%
n
Total 43
Moderate
High
n
74
78
Missing, N/A
71
67
13.6
0.00
Significance
n
%
4
17
48
20
83
100
24
100
M2
NC
Chi Square
High
7.1
0.01
__
aExtent of Use based on 5-point scale, 1 to 3 moderate, 4 to 5 high.
bBased on responses to an agree-disagree question with a 5-point
response scale, grouped 1 to 3 low, 4 to 5 high.
CN includes only those respondents with 1 or more days training or
methodology skill level of 4 or 5, on a 5-point scale.
Discussion:
The supervisor is the single most influenital point of
contact most analysts have with the organization.
that
supervisor
individual receives incentives and communications
the
about what is expected.
attitudes
supervisors'
the
It is through
are
associated
that
finding
There is little surprise in the
with extent of implementation.
The fact that both M1 and M2 are affected is evidence of the strength of
supervisor influence relative to the influence of other variables.
few of the
tested
factors
here
statistically
a
showed
Very
significant
relationship with M2.
1.3 General organizational attitudes consistent with the use of the
methodologies are likely to be associated with more extensive use.
in Table 19 is that organizational values consistent
reported
finding
The
with the methodologies are associated with extensive use of M1
but
are
In the case of Ml, even if
not significantly associated with use of M2.
the organization is perceived as supportive of the methodologies, 31 per
cent of respondents make little or no use of the methodology.
Discussion:
organizational
The influence of perceived
values
on
the decision of an analyst to use a particular technique is difficult to
event.
can
be
particular
any
to
attribute
individual
or
to isolate in a specific
The perception of dominant values or beliefs in the
associated
with
behavior, however, and the association may be
strengthened if the clarity and internal consistency of
increased.
If management
commitment
to
improving
environment
were
long
able
run
the
values
is
to establish, in this case, the
productivity
by
improving
maintainability, or client-analyst communications, and if the structured
97
Table 19
Extent of Use by Organization Values
Extent of Usea
M2
M1
Moderate
n
Organization
Values
Consistency
With
Methodologyb
Statistics
%
26
High 17
Total 43
Low
M1
61
39
100
Moderate
High
n
%
29
71
100
n
%
High
n
54
46
100
%
42
58
100
M2
NC
74
78
Missing, N/A
71
67
Chi Square
5.9
0.5
Significance
0.01
0.46
aExtent of Use based on 5-point scale, 1 to 3 moderate, 4 to 5 high.
bBased on responses to a scaled set of agree-disagree questions,
.55)
reliability tested at alpha
CN includes only those respondents with 1 or more days training or
methodology skill level of 4 or 5, on a 5-point scale.
methodologygies
perceived as useful in achieving these ends, then
were
increase.
use of the methodologies might well
Individual
2:
Hypothesis
·
Attitudes.
The
analyst, both toward structured methodologies in general and
individual
M2.
positively
were
toward the local versions of structured methodologies,
related
to extent of use of MI but were not related to extent of use of
(For explanation of attitude variables, see Chapter 2, above.)
2.1 Analysts
and M2.
are
attitudes
whose
methodologies
structured
consistent
for
MI
use
the
with
level
using
the
Square
Chi
MI more extensively (Table 20).
of
Analysts who are
test.
favorably inclined toward structured methodologies in general will
use
of
in general will be more extensive users of MI
It was found that this statement was true at the .01
significance
to
the
of
attitudes
tend
The relationship does not hold
for M2.
2.2 Analysts whose attitude toward the local versions of structured
It was
methodologies are favorable will be more extnesive users.
that
favorable
found
attitudes were once again associated with significantly
greater extent of use.
In this case, the direction of
the
association
was present in M2 as well as in Ml, but the relationship was significant
only in the latter (Table 21).
Discussion:
influences,
and
The
respondent's
attitude
to
the
is influenced by, use of the methodologies.
that the relationship is not stronger in the case of M2
although
this
methodologies
result
fits
the
The fact
is surprising,
pattern noted above that many factors
99
Table 20
Extent of Use by Respondent's Attitude
Toward Structured Methodologies
Extent of Usea
Ml
High
Moderate
n
%
31
72
28
100
Respondent's
Low
Favorable
High 12
Total 43
Attitudeb
Statistics
M1
n
Moderate
%
32
68
100
74
78
Missing, N/A
71
67
Significance
10.0
0.00
%
n
69
31
100
%
54
46
100
M2
Nc
Chi Square
n
High
0.9
0.34
aExtent of Use based on 5-point scale, 1 to 3 moderate, 4 to 5 high.
bBased on responses to a scaled set of agree-disagree questions,
.55)
reliability tested at alpha
cN includes only those respondents with 1 or more days training or
methodology skill level of 4 or 5, on a 5-point scale.
100
Table 21
Extent of Use by Respondent's Attitude
Toward M1 and M2
Extent of Usea
Mi
Moderate
n
Respondent's
Favorable
Attitudeb
Statistics
23
Low
High 20
Total 43
M1
M2
--
Moderate
High
High
%
U
%
n
%
n
%
54
46
100
31
100
54
100
24
100
M2
NC
74
78
Missing, N/A
71
67
Chi Square
9.2
1.6
Significance
0.00
0.20
aExtent of Use based on 5-point scale, 1 to 3 moderate, 4 to 5 high.
bBased on responses to a scaled set of agree-disagree questions,
.55)
reliability tested at alpha
CN includes only those respondents with 1 or more days training or
methodology skill level of 4 or 5, on a 5-point scale.
101
interesting
that
for
neither
methodology
attitude
favorable
there
of
presence
the
a
with
In both cases among analysts
favorable attitude assure heavy use.
a
does
It is
M2.
with
associated are associated more strongly with MI than
more users in the "none to moderate"
are
category than there are heavy users.
3. Personal Ability, Experience and Career Orientation. There were
statistically significant but complex relationships
result
Years
methodologies.
times
a
Skill in programming appeared to be
of testing this hypothesis.
at times positively and at
as
demonstrated
use
to
related
negatively
of
the
of experience was found to be related positively
education
was
methodology
and
of
And orientation
to use, but only in the case of MI.
not found to be related to use of either methodology.
3.1 There is a relationship between skill in each
extent
of that methodology.
use
of
The results given in Table 22 are
consistent with this hypothesis for both MI and
analysts
low
with
to
moderate
skills
are
For
M2.
low
level
the
former,
users of the
methodology by a ratio of more that four to one, while analysts who
highly
skilled are found mostly among the high users.
is found for M2.
The results are tested with the Chi
are
The same pattern
statistic
Square
and found highly significant.
3.2 Analysts who are
development
techniques
skilled
will
be
The hypothesis was tested for four
Nomad/Ramis.
in other
program ming
and
systems
less likely to use the methodologies.
skills:
Fortran,
PLI,
Cobol
and
No statistically significant relationsh ips were found for
102
Table 22
-
Extent of Use by Skill in M1 and M2 -
Extent of Usea
M1
Moderate
Level
of
Skill a
24
100
Low
High
Statistics
Nc
High
Moderate
High
n
%
n
%
n
%
n
%
34
81
14
45
43
80
8 19
Total
17
42
55
100
11
31
20
100
4
20
54
17
83
100
M1
73
M2
78
Missing, N/A
Chi Square
Significance
10.9
0.00
24.9
0.00
aExtent of Use based on 5-point scale, 1 to 3 moderate, 4 to 5 high.
CN includes only those respondents with 1 or more days training or
methodology skill level of 4 or 5, on a 5-point scale.
103
However
positive
significant
a
finally,
And
(Table 23).
for
found
was
relationship
MI
and
significant
a negative
and
significant
Given these contradictory results,
relationship for M2 (Table 24).
Ml.
positive
a
and
of
was found with use of M2
association
Fortran,
for
use
with
found
was
For PL1 no relationship
the latter two.
the
hypothesis can neither be confirmed nor denied.
The relationship between technical skill level and use
Discussion:
that
is
skills have specific effects on use of each methodology
technical
due to commonalities or differences.
relatively
For example, PL1 is a
p rogramming language, possibly similar in orientation to M2.
structured
with
familiarity
here
investigated
the approach.
to
due
M2
use
to
Another possibil ity is that the skills
themselves
may
willing
more
An analyst skilled in PLI might
underlying
possibility
One
simply.
of the methodologies cannot be explained
be
associated
strongly
with
an
This idea was tested with both age and technical
variable.
found,
educational background, and significant associatio ins were
Both
Fortran and PLI skills are negatively related to the age of the analyst;
i.e.
analysts
younger
than
languages
programming
are
more
older
b)e skilled in these two
to
likely
for
S imilarly,
analysts.
both
languages analysts with a technical degree are mor e likely to be skilled
than
with
those
a
No sig Inificant relationships
nontechnical degree.
were found between age and education, on the
Cobol
or
Focus/Ramis.
So
to
the
hand,
and
skill
in
certain patterns of skill distribution may
represent patterns of distribution of age
relationship
one
contradictory
or
of
associations
education.
found
But
the
in the study
104
Table 23
Extent of Use by Skill in P1
Extent of Usea
Ml
Moderate
n
Level
Low
of
High
Skill a
24
%
23
55
19
45
Total
100
Statistics
NC
Ml
73
Moderate
High
n
%
n
%
n
%
18
13
58
42
29
71
100
61
39
100
7
17
42
33
21
31
54
100
M2
78
Missing, N/A
Chi Square
0.8
5.6
Significance
0.67
0.02
aExtent of Use based on 5-point scale, 1 to 3 moderate, 4 to 5 high.
CN includes only those respondents with 1 or more days training or
methodology skill level of 4 or 5, on a 5-point scale.
105
Table 24
Extent of Use by Skill in Fortran
Extent of Usea
M1
Moderate
Low
High
Statistics
High
%
n
%
n
%
27
66
34
14
Total
74
26
100
28
25
31
53
47
100
19
5
53
79
21
100
M1
M2
n
Level
of
Skill a
100
24
Moderate
High
%
n
NC
72
78
Missing, N/A
73
67
Chi Square
2.1
5.3
Significance
0.36
0.07
aExtent of Use based on 5-point scale, 1 to 3 moderate, 4 to 5 high.
CN includes only those respondents with 1 or more days training or
methodology skill level of 4 or 5, on a 5-point scale.
106
remains unclear.
be
3.3 Technical work experience of the analysts will
The relationship was tested using years of experience in
extent of use.
the
of
field
be
to
26) .
as
technical
Years
of
to
compared
exper ience
was
related to use of Ml, w ith Chi Square equal to 5.7 at the
t:end
to
be
in systems
Analysts with longer experience
.02 level of s;ign ificance.
development
and
development
syste ms
nontechnical e!ducation (Tables 25 and
found
to
related
extensive
more
users.
There were no other
statistically sig nificant relationships under this hypothesis.
Discussicon:
from
degree
While technical orientation, as measured by
is apparently related to access to training in Ml
col lege
skil led
no more or lerss
nontechnical
bac kground.
to
methodologies
the
use
methodologies
may
analysts
than
one
with
a
This suggests that the technical orientation
of an analyst is not in itself a factor in determining
but
among
An analyst with a technical background is
or trained.
likely
the
of
and M2 it is not related to use
already
technical
extent
of use,
be indirectly involved in the sequence of events through which
the patterns of use
are
established
in the
organization.
If, for
example, supervisors are instrumental in assigning analysts to-training,
and
if the
supervisors
act
on
the
assumption that analysts with a
technical background do not need training in MI and M2, the analysts
this
in
group will receive less training in the methodologies and will for
that reason appear to be less extensive users.
107
Table 25
Extent of Use by Experience in Systems Development
Extent of Usea
M2
Ml
Moderate
Years
0-3
3+
of
Experience
Statistics
Low
High
Total
High
I
n
%
n
8
26
35
66
14
58
23
74
18
34
10
42
100
53
100
24
100
n
%
13
31
29
69
42
100
31
M1
Moderate
High
n
%
M2
NC
Missing, N/A
Chi Square
0.0
0.2
Significance
0.83
0.69
aExtent of Use based on 5-point scale, 1 to 3 moderate, 4 to 5 high.
CN includes only those respondents with 1 or more days training or
methodology skill level of 4 or 5, on a 5-point scale.
108
Table 26
Extent of Use by Education
Extent of Usea
M2
Ml
Moderate
n
Nontechnical
Technicalb
Low
28
High 12
Total 40
Statistics
M1
MI
Moderate
High
High
%
n
%
n
%
n
70
19
70
24
47
53
100
12
30
100
30
100
%
55
45
100
M2
NC
67
73
Missing, N/A
78
72
Chi Square
0.0
0.1
Significance
1.0
0.7
aExtent of Use based on 5-point scale, 1 to 3 moderate, 4 to 5 high.
bNontechnical includes bachelor's or master's in fields other than
computer science. Technical includes bachelor's or master's in
computer science.
CN includes only those respondents with 1 or more days training or
methodology skill level of 4 or 5, on a 5-point scale.
109
Factors.
Opportunity
4.
in
obtained
Results
this
testing
hypothesis indicate that opportunity fac tors of two types may be related
to
use
structured
of
assigned as well as the amou nt
associated
with
use.
The task to which an analyst is
methodologies.
of
time
spent
on
can
task
that
be
And the size of the project on which the analyst
is working may be similarly related to use.
4.1 The type of task to which analysts are assigned
they
in which
spend major portions o f their time will be related to their extent
of use of the methodologies.
proportion
the
the
of
This relationship was tested by
comparing
on
Scoping,
analyst's
time
spent
Development/Acquisition, Enh ancement and Maintenance to
use
and
of
both
MI
and
M2.
given in Tables 27 and 28.
related
to
more
the
extent
The results for Scoping and Maintenance are
They show more
time
use of MI and less use of M2.
spent
in Scoping
related
is
Scoping is an acti vity
which takes place early in the development cycle and is appropriate
the use of Ml.
of
for
They further show that more time spent in Maintenance is
to less use of both methodologies, although the relationship is
not statistically significant for M2.
In maintnenance work
the
system
being modified has frequently been in use for many years and is unlikely
to
have been developed with either methodology originally.
use the methodologies now may not
be
as
easy
as
using
To begin to
them
on
an
entirely new system and this may be a factor in the relatively low level
of use.
110
Table 27
Extent of Use by Time on Scoping Tasks
Extent of Usea
M1
M2
Moderate
Percent Time
on Scaping
Statistics
n
%
n
Z
42
23
58
100
77
100
30
24
54
56
44
100
8
16
24
33
67
100
%
18
100
25
Total 43
M1
NC
74
Missing, N/A
71
High
%
n
0
Moderate
High
n
M2
78
Chi Square
0.9
0.7
Significance
1.63
0.8
aExtent of Use based on 5-point scale, 1 to 3 moderate, 4 to 5 high.
CN includes only those respondents with 1 or more days training or
methodology skill level of 4 or 5, on a 5-point scale.
111
Table 28
Extent of Use by Time on Maintenance
Extent of Usea
M2
M1
Moderate
Percent Time
on Maintenance
Statistics
Moderate
High
n
%
n
%
n
%
n
%
18
10
42
23
14
45
13
24
10
42
50+
15
Total 43
35
100
31
100
54
100
24
100
0
1-49
M1
M2
NC
74
78
Missing, N/A
71
67
Chi Square
7.4
3.4
Significance
0.02
0.18
aExtent of Use based on 5-point scale, 1 to 3 moderate, 4 to 5 high.
CN includes only those respondents with 1 or more days training or
methodology skill level of 4 or 5, on a 5-point scale.
112
4.2 The size of the work group to which the analyst is assigned
well
as
the length of the project on which the analyst is working will
both be related to extent of use.
were
tested
and
the
results
While there appeared to be come
comments
made
Both
relationships
predicted
here
found not be statistically significant.
support
for
this
hypothesis
in the
by several respondents during the interviews, the extent
of use measure used here was not specific to a
rather
as
referred to use during the entire year.
particular
but
For analysts who worked
on more than one project, the relationships tested
been represented in the data.
project
here
may
not
have
113
SUMMARY AND RECOMMENDATIONS
This study was undertaken for the purpose of
understanding
of
implementation of structured systems development
the
process
individual analysts.
selected
the
from
described
was
point
The
firm.
large
a
of
environment
methodologies by analysts in the
implementation
better
a
developing
view
of
Fou r hypotheses regarding the association
of
between
organizational, attitudinal, ability and opportunity variables
and the extent of use of two structured methodologies were tested.
In this chapter
the
results
of
the
are
study
summarized
and
implications for the orga nization and for future research are presented.
In the
provides
section
first
the findings are presented.
action
for
recommendations
The second section
management
by
of
systems
the
development organization to increase the effectiveness of implementation
of
the
two
methodologies.
In the third section directions for future
research are suggested.
Findings of the Study. The findings of this
here
in three
parts:
extent
benefits of the methodologies and
and
opportunity
variables
study
are
summarized
of use of the methodologies, perceived
selected
related
methodologies by individual analysts.
to
the
organizational,
implementation
individual
of
these
114
Two structured systems
in a
group
technology
methodologies
large
information systems organization for
methodologies
These
techniques.
firm
industrial
use
as
were
adopted
to
effectiveness of analysts at two stages in the applicat ions
sequence.
first
The
methodology,
MI,
the
analyst
is designing,
coding
program which will become the application
delivered
to
the
development
the
client.
stage
in the
or
software
proc uct
to
be
M2 was developed in the mid 1970's and has
Ml was developed
past
few
The organization provides formal training of three to five days
duration for its analysts in each of the methodologies
does
sequence
and documen ting the computer
in the late 1970's and has been adopted for use only over the
but
business,
client's
been in use in the organization for a number of years.
years.
the
needs and recommending a solution to meet those needs.
client
The second methodology, M2, is used at a later
when
mprove
he ear ly stage of
is used at tl
development when the analyst is learning about
defining
by the
improvement
productivity
designed
were
computer
the
by
developed
and
encourages,
not require, use of the methodologies in tasks for which they
are appropriate.
For the purpose of this study it was assumed that
the
decision to use a methodology or an alternative approach was made by the
analyst.
The extent of use of both MI and M2 by analysts in the organization
varied.
Overall approximately half of the respondents reported using Ml
at least some of the time when they were working on tasks for
is applicable.
which
MI
Approximately 60 per cent of respondents reported using
M2 at least some of the time when they were working on those
tasks
for
115
which
M2
This means that about 50 per cent of analysts
is applicable.
who work on MI-appropriate tasks never use Ml, and about 40 per cent
of
analysts who work on M2 tasks never use M2.
Analysts and members of the computer
asked
what
they
perceived
both
were
as the benefits and disadvantages of these
that
agreement
was
group
technology
improved
provides
Ml
methodologies.
There
communications,
helps understand the client's business and improves the
Analysts also saw the fact
definition of requirements.
structured
approach
advantages of M2.
as an advantage.
by
is a
MI
There was agreement, too, on the
both
application
the
making
documentation,
seen
It was
that
groups
for
providing
as
which
better
used more
it was
maintainable and providing a structured design.
a
Both analysts and developers of the methodologies saw as
disadvantage
of
MI
and
M2
common
that these techniques are time consuming.
Analysts reported further that the methodologies are unfamiliar and that
they are not useful for maintenance (i.e.
of
previously
modification
or
enhancement
existing applications or systems, as opposed to creation
of entirely new systems).
Additional
insight into the decision
to
use
alternative approaches was provided when analysts gave their reasons for
selecting alternatives.
Considerations of short run costs, particularly
in terms of additional time needed by analysts relatively inexperienced
in using the methodologies and in relation to the size and complexity of
the project, appear to dominate the decision.
116
this study was considered an opportunity variable, was the
of
purposes
the
for
which
and
use,
One fac tor which was found to relate to
the
of
portion
substantial
A
received.
extent of training
sample
having had no access to training in these methodologies (about
reported
20 per cent in the case of MI and 40 per cent in the case of M2).
A
found
were
variables
use
client's
values.
Individual variables which proved to
attitude
were
toward
toward Ml and M2 specifically, and certain skill
was
methodology
each
with
associations
language,.
years
of
use,
but
in opposite
systems
nontechnically
experience
development
was
nontechnical,
directions depending on the
by
educated
methodologies,
there
associated
analysts
were
and
associated with use.
complexity
found to be positively
access
with
further
no
of
training,
to
with
more likely to have had high access.
technical education and extent of use.
tasks
was
Type of education, technical as compared
However, given at least a minimum level of
type of
significant
Finally the technical background of analysts as measured
associated with extent of use.
to
showed
Fortran)
and
(PLI
in
Skill
variables.
to use of that methodology.
related
positively
Skill in programming languages
in general, attitude
methodologies
structured
use
with
associated
be
found
was
organizational
and
desires
supervisor's
include
the methodologies.
of
association
significant
Organization variables for which a
attitude,
to
related
be
to
opportunity
and
individual
organizational,
other
of
number
or
skill
in the
associations
found
between
training
Opportunity variables, including
project,
were
both
found
to
be
117
extent
was
use
of
with
MI
Desires of the
rather than M2 use.
use
supervisor and time spent on maintenance, as well
with
association
In most of the relationships reported above, the
own
respondent's
as
skill level, were the exceptions to this pattern.
Recommendations for More Effective Implementation. Data analyzed in
this study, along with the comments of respondents on the implementation
process, suggest a number of actions on the part of systems
the
increase
might
which
managers
of
extent
development
use of the structured
These
methodologies and improve the effectiveness of analysts.
actions
address the disadvantages currently perceived by analysts as well as the
more
potential
for
potential
action
extensive
use
Five areas for
section:
measurement,
in this.
identified
are
of the techniques.
training, motivation, monitoring and user involvement.
A
large
disadvantages
number
Each analyst
and
each
of
the
computer
there
is
Many were interested in discussing the
general
agreement
measurement is not possible, it appears that
constructing
at
that
precise
there
are
issue.
productivity
advantages
to
least approximate measures of inputs and outputs using
the methodologies compared with equivalent
performed
member
group, speaking from his or her own experience, had opinions
about cost and benefits.
Although
and
advantages
both
regarding
comments
of the methodologies touch on the related issues of costs
of use and benefits.
technology
of
without the methodologies.
measures
of
similar
tasks
How much "overhead" labor cost is
118
required?
of
size
project
long
set
to
for
minimum
a
as
appropriate
is an
the
of
use
requiring
with
What are the short run costs, how do these compare
methodologies?
the
What
How steep is theexperience curve?
An attempt to measure some of these effects, over
run gains?
focusing
time, would at least have the advantage of
the
attention
of
action
is
analysts on the relevant dimensions.
Skill acquisition is a second area in which
If the
needed.
management
organization favors the use of M1 and M2, it is up to
management to train all analysts and to be sure that the skills acquired
.are kept current.
courses,
This
can
and
consultants
local
done
be
with
networks
a
of
experienced
advocates, and printed or on-line documentation, tutorials
formal
of
combination
or
users
or
updates.
Project leaders and others in supervisory positions are most likely in a
good
position
to
keep track of the skill level of their team members.
They should be encouraged to do so, perhaps by being
form,
content
and
timing
of
training
activities
on
consulted
the
or by being given
responsibility for conducting the training themselves,as was done at one
of the locations.
It is possible that simply reorienting
the
training
to include more practical applications, and scheduling training so as to
be closer in time to each analyst's opportunity for use of the skills on
the
job will be enough of a change to produce an increase in use of the
techniques.
the
required
Clearly training is not an isolated activity through
which
skills are transmitted on a one-shot basis, but rather an
integral part of the productivity improvement process.
119
area
another
Motivation and performance incen tives constitute
in
which management actions are possible . Currently analysts perceive that
are
they
primarily
judged
basis
the
on
of
client needs,
meeting
While many
performing according to time plan and working within budget.
expressed a
to
commitment
professi onal
structured
maintainability,
quality,
as
such
standards
and similar features which tend to
design
make an applicat ion more effective in the long run, few stated that they
were
features,
as
success
their
improve
the
cost
in the
short
Until
run.
the
finds a way to cover the cost, or to provide motivation or
mak:ing
decisions
admitting
that
the
analysts,
of
the
on use of the methodologies will tend to
select the lower cost alternatives.
while
other
and
quality
incentives which will outweigh the costs in the mind
indi.viduals
The
areas.
in these
helping
definite
a
at
but
organization
seen
akre
methodologies
for
primarily
evaluated
result
They
may
may
be,
make
in the
that
choice
even
long run, a less
effective product.
Management must also find ways to monitor more closely the
M1
and
M2,
phase
A simple checkoff as part of completion
entail
of
of a project is used at one of the locations and it appears
to have this effect, according to several respondents.
not
of
if only to insure that the decision not to use one of the
methodologies is justified.
each
use
requiring
use
ofany
technique.
allowing analysts to make choices according to
Monitoring
need
There are advantages of
their
needs.
But
the
visibility of the technique, and the message that it is approved for use
by
the
organization,
are
both
increased through use of a monitoring
120
system.
with increasing systems
implemented in an organization.
resources
conserve
productivity
development
researchers
to
and
the process through which change is
understanding
the
with
concerned
has
study
current
The
issues of interest both to managers concerned
of
number
a
identified
Research.
Future
for
Recommendations
the
The study was limited by
need
to
and for this reason focused on only a small part of
the implementa tion problem.
The current study leaves for later research
efforts a numb er of promising leads.
First, the
literature
clearly
implementation
on
suggests
that
factors relate d to organizational environment are important determinants
of
In regard to differences in organizational factors across
change.
the three loca tions sampled here, this study has only begun
the
for
implicati ons
effective
explore such issues as the following:
policy
in regard
differ
assignment
to
training,
is management in the
vs.
methodologies) ?
between
the
In what ways does
use of the methodologies, for example in
access
various
to
help,
consulting
locations
distributed
To
what
applications
up work can
organizational
in written systems development guidelines?
methodologies
centralized
to
Follow
implementation.
explore
to
extent
organized
authority
are
there
for
status
In what ways
differently
selecting
differences
of
these
teams?
(e.g.
appropriate
in mobility
development teams and the various.
organizations who are the clients
of
Previous
end-user
studies
121
suggest
that
organizations
are characterized by top management
which
support of an innovation, which are centralized in their decision-making
into
related to an innovation, and which evidence movement by employees
and
out
of the group adopting an innovation will tend to implement the
innovation more readily than is the case where
To
present.
want
to
not
are
implementation more effective, the corporation will
make
identify
factors
these
specific
local
level
changes
organizational
in
structure or policy which can facilitate change.
Two
recently
importance
of
published
linking
the
provide
papers
of
study
assessment of the desired and expected
in which
organization
some
insight
change
technological
impact
it is implemented.
of
the
into
the
with
change
an
the
on
Both discuss organizational
impact of technological change in terms of risk or uncertainty, and both
suggest that management of that change
problem
of
organization
best
be
the
of installing a new technology (Gibson et al., 1983).
The
implementation,
that
strategy
twenty
in terms
cases
of
of
new
the
risk
information
impact
system
traditional implementation and project management
techniques work best in situations in which the risk of failure
are
low.
In other
situations,
due
the
more
They suggest that the appropriate strategy for implementing
a technological change should be selected only after
impact
to
when
organizational risk is higher, different management strategies are
effective.
a
to
authors find, in a study of
organizational
as
understood
The first is a model which describes
managing uncertainty.
choice of an implementation
can
assessment
is conducted
and
an
organizational
the level of risk understood.
A
122
of
similar conclusion emerges from a conceptual study on the management
software development efforts (Zmud, 1980) which places structured
large
by
address productivity issues.
to
management
to
the productivity problem is largely due
of
stages
systems
development,
used
methodologies
systems development methodologies among a number of
The author argues that
uncertainty
at
early
the
when systems analysts are expected to
make decisions which will determine the way the remainder of the work is
conducted.
complexity
These uncertainties are due in large part to the
They
of communications among analysts and between analysts and clients.
are
exacerbated, over time, by personnel turnover and changes in system
but
methodologies,
implemented
of
The resolution
requirements.
on
depends
difficulties
new
these methodologies are seen as effective only when
in the
of
context
In short,
changes.
organizational
are part of the problem and necessarily part of
factors
organizational
these
the solution.
and
Another way to build on the current study is to verify
on
the findings on extent of use, attitudes and perceptions of analysts
and perceived values in the
include
clients
the
on
clearly
influenced
organization
one
population to be sampled.
extending
understanding
developed
the
by
both
to
use
client
the
and
methodologies
responses
to
is
supervisor, and an
complement
the
in this paper regarding the views of analysts.
There is further value in including these groups in future
their
study
hand and supervisors on the other in the
The decision
strongly
by
understanding of the views of both of these groups will
that
expand
can
serve
to
check
and
perhaps
studies,
modify
in
the
123
conclusions drawn from the data gathered here, limited to the admittedly
implementation
in the
subjective responses of just one set of players
game.
at
the
and
population
A particularly promising way to expand the subject
same time improve the design of the study is to change the unit
appears
in most cases where a decision is made to use one of the
that
in the
methodologies or an alternative approach, that decision is made
context
tasks comprising a project and conducted over a
of
set
a
of
of
period of months for a client, most often by a team
developed
variables
The
analysts.
relationships tested in the current study can
and
easily be apapted to a study of implementation by
addition
It
project.
or
job
the
to
of analysis from the individual analyst
The
teams.
project
parallel sets of independent variables to analyze the role
of
of the project leader, group
leader
and
client
decision
in the
to
implement the methodologies would greatly enrich the findings.
It would be of interest, too, to explore in more depth the unstated
the
new
rationale
analysts'
for
statements
respondents'
current
The
methodologies.
choices
questions.
that
they
or
did
while
focussed
on
accepting
the stated
unquestioned
that the methodologies were "not appropriate",
or their assertion in some cases
methodologies
study
using
or
and perhaps unconscious resistance of some analysts to trying
that
they
had
never
heard
of
not know enough about them to respond to certain
The study did not probe, in many cases, respondents'
used
the
claims
their own methods to determine just what those methods
were, and to what extent an explicit choice had been
made
rather
than
124
just
decision
a
to avoid the new, unfamiliar and perhaps difficult or
threatening alternative.
that
respondents
expressed reluctance to use the methodologies because
sufficient
elsewhere
developed
the methodologies had been
without
found
team
study
the
In several interviews
consultation
with
in the
organization,
analysts (the Not Invented Here
In other cases, respondents claimed they had no skill in the
syndrome).
methodologies because they had been prevented by their
Data gathered in the current study is not
training sessions.
attending
from
supervisor
nor
sufficient to allow any sophisticated analysis of these statements,
would
it justify
these factors.
some
analysts
drawing
any conclusions regarding resistance due to
It is likely, however, that resistance
and
is present,
in
in some parts of the organization more than others,
and it would be helpful 'in improving analyst productivity to isolate and
counteract this resistance.
in which
sessions
analysts
of
observation
Structured
post-training
with
trainers could
lead
the
an
interviews
training,
and
with
interviews
understanding
training
the
metholdologies initially.
the
encounter
participants,
to
be
One source of data might
of
types
and
and
pre-
experienced
sources
of
resistance, impact of resistance on learning and strategies effective in
overcoming resistance prior to and during implementation.
It is quite possible that some of
based
in
quite
reasonable
objections
study
methodologies;
the
current
felt
the
methodologies
analysts
the
resistance
to
failed
characteristics
short
stopped
to
encountered
of
examining
meet their needs.
particularly important for the technology development
of
group
to
is
the
ways
It is
assess
125
the
of any such objections, since that group is continuing to
validity
widespread
and
effective
their
by
made
or
his
of
users
the
Nevertheless, end-user
product.
her
comments
critical
developer actively to listen to, hear and respond to
in
difficult for a
It is often
use.
interest
an
have
to
modify these methodologies and will continue
acceptance of the product is the key factor in successful implementation
and
acceptance
not the perceptions of the development group.
characteristics,
in which
perceptions
analyst
capabilities
the
of
of
technical
the
focus,
to
product
A study
characteristics
and
product were examined in depth would help provide
valuable information for product modification.
able
of
perceptions
end-user
on
based
is
precisely
more
than
the
It would in addition
study,
present
on
be
the
characteristics of the methodologies which are perceive d as their strong
points and which may be expanded on in the future.
One dimension not addressed in this study but
the
wor th
examining
in
future is the history of innovation, adoption and implementation of
these
methodologies.
decision
A
study
longitudinal
which
focussed
on
the
variables and the decision-making process at each phase in the
life cycle of this new product could highlight key indi viduals
in
positions
implementation.
the
in
organization
regard
and
to effectiveness
These findings would suggest, in turn,
ways
key
of
in wh ich
future change efforts should be organized to take advantage of potential
sources
of
support
and
to avoid potential sources of resistance.
found evidence in the current study that such variables as contact
methodology
advocates
and
level
of
supervisor
support
We
with
both made a
126
At what point are
difference in the extent of use of the methodologies.
in the life cycle are these two
whaztt
ways
they
do
the
and
important,
most
in
to the sequence of decisions made in the
relate
organization over time?
factors
It might be
to
also
worthwhile
the
compare
level of acceptance of Methodology 1, which is only about two years old,
with that of Methodology 2 which has been in use for five years or more.
The
and
visibility
willingness to use a
reputation
of
an
related
later,
early
one.
influence
This
and
other
possible
will be of particular interest in this setting in that the
interactions
automated
organization has plans for the release of a series of
one
can
product
tools,
set for each of the methodologies, over the next few years, and the
success of these new products will likely be related to the patterns
of
use of their precursors, the structured methodologies.
Finally, it is recommended that future studies
establish
a
clear
the
broader
policies of the firm in regard to information systems strategy.
Higher
framework
level
for
assessment
management
in the
of
the
Computer
methodologies
Science
within
department,
as
well
management in the end-user departments, have explicit or implicit
for
increasing
productivity in the organization.
studies should be shaped, in part, by these goals.
is moving
toward
end-user
computing
clients to a much greater degree in the
client
future
universe
goals
The design of future
If the
studies
to
organization
should include
be
sampled,
concerns should be defined and used to frame the hypotheses.
the organization is concerned with increasing long run
systems
as
productivity
and
If
in
development, future studies should attempt to identify existing
127
baseline data and develop ways to compare
methodologies
to
systems
systems developed without the methodologies in regard
to well-defined measures.
If the organization aims to improve short run
establish
productivity, a study might be designed to
analysts'
statements
that
of
use
the
currently
as
validity
the
methodologies
time-consuming and decreases productivity
management
the
with
developed
of
is costly and
measured.
If
goals include productivity improvement of any sort, it would
be worthwhile to
examine
the
incentives
and
evaluation
performance
currently in place to see whether they encourage or discourage
criteria
use of the methodologies, assuming that the
methodologies
do
in fact
increase productivity.
As is common with research in organizational
raised
as
many questions as it answered.
the lines suggested here is clear.
change,
this
study
The need for more work along
128
BIBLIOGRAPHY
Michael, "SSA - A Tool for Involving Users in Systems
Anderson, C.
Development", paper presented at the American Association for the
Advancement of Science, Annual Meeting, Toronto, 1981.
in Organizations",
Anderson, David G., "Mobility and Innovation
unpublished dissertation proposal, Stanford University, 1981.
Benjamin, Robert I., "Information
Quarterly, June, 1982.
Technology
in the
MIS
1990's",
Bertan, Jennifer, "Learning Styles and Job Satisfaction among Systems
Analysts", unpublished master's thesis, Sloan School of Management, MIT,
May 1983.
Bigoness, W.J., and Perreault, W.D. Jr., "A Conceptual Paradigm for the
Study of Innovations", Academy of Management Journal, vol.24 (1981).
Blau, Peter M., The Dynamics
Chicago Press, 1963.
Brooks,
F.P.,
The
Mythical
of
Chicago,
Bureaucracy,
Man-Month,
University
Addison-Wesley,
of
Reading,
Massachusetts, 1975.
Campbell, Joh P., "Personnel Training and Development", in Annual Review
of Psychology, Paul Mussen and Mark Rosenzweig, eds., vol. 22, (1971).
and Weick, K.E.,
Campbell, John P., Dunnette, M.D., Lawler, E.E.
Managerial Behavior and Effectiveness, McGraw Hill, New York, 1970.
Crawford, A., "Implementation of Electronic Mail", MIS Quarterly, March,
1982.
the
Firm,
Downs, George W. Jr. and Mohr, L.B., "Conceptual Issues in the
of Innovation", Administrative Science Quarterly, Dec. 1976.
Study
Cyert, Richard M. and March, J.G., A Behavioral
Prentice-Hall, Englewood Cliffs, 1963.
Theory
of
Gibson, Cyrus, "Making the Information Revolution Organizationally
Friendly", Index Systems, Cambridge, Massachusetts, 1983.
Goldstein, David K., "Effects of Structured Development Methodologies on
1330-82,
Job Satisfaction of Programmer-Analysts", Working Paper No.
Sloan School of Management, 1982.
129
-------- , "The Impact of Development Aids on the Systems Development
Process", Conference Proceedings, Human Factors in Computer Systems,
Center for Information Systems Resear ch, Sloan School of Management,
MIT, 1982.
Gremillion, Lee L. and Pyburn, P. "Breaking the
Bottleneck", Harvard Business Review, vol. 61 no.
Change
Hage, Jerald and Aiken, M., Social
Random House, New York, 1970.
Systems Development
2 (1983).
in Complex
Organizations,
-------- and Dewar, R., "Elite Values Versus Organizational Structure in
Predicting Innovation", Administrative Science Quarterly, 1973.
Hammer, Michael, "Strategic Issues in Office Automation: The Innovative
Process", Proceedings, Symposium for Senior Executives, Sloan School of
Management, MIT, Nov. 1981.
Holton, John, "Are the
Datamation, July, 1977.
Johnson, James R.
New
Programming
Techniques
Used?",
Being
"A Working Measure of Productivity", Datamation, vol.
23, 1977.
Keene, Peter, "Editor's Preface", Office:
1, 1982.
Technology and
Kerr, Stephen, "On the Folly of Rewarding A while
Academy of Management Journal, Dec. 1975.
-
II 1
--
Hoping
People,
for
nCc?,
B",
vol.
The
AI.-AC-?,-...:'I
n
M.A.,
A
Strat
unpublished master's thesis, Sloan School of Management, MIT, 1980.
Khalil,
Knight, Kenneth E., "A Descriptive Model of the
Process", Journal of Business, vol. 40 (1967).
Kraft, Philip, Programmers
and
Managers,
Intra-Firm
Springer-Verlag,
Innovation
New
York,
1977.
Leonard-Barton, Dorothy, "Professionals as Information Priests in the
Diffusion of Innoavations", paper presented at the workshop, Consumerism
and Beyond, sponsored by the Harvard Marketing Science Institute, April,
1982.
, "Introducing
Production Innovation into an
Organization:
Structured Methods for Producing Computer Software", paper presented at
the ORSA/TIMS Conference, Chicago, April, 1983.
130
Journal
Locke, E.A., "The Motivational Effect of Knowledge of Results",
of Applied Psychology, vol. 51.
-, Bryan, J.F. and Kendall,, L.M., "Goals and Intention as Mediators of
Monetary Incentives on Behavior", Journal of Applied Psychology, vol.
52.
March, James G., "Footnotes to
Science Quarterly, 1981.
-------
Organizational
Change",
Administrative
and Simon, H., Organizations, John Wiley & Sons, New York, 1958.
Markus, Lynne M., "Power, Politics and MIS
Liaison Symposium, MIT, March 1981.
Industrial
Implementation",
Development
Applications
James,
Martin,
1982.
Cliffs,
Englewood
Prentice-Hall,
without
Proarammers,
McErlean, Donald, Training and Communications and the use of Structured
Systems Methodologies, unpublished master's thesis, Sloan School of
Management, MIT, 1983.
McGowan, Clement L. and Kelly, J.R., Top-Down Structured Programming,
1975, reprinted by Datapro Research
Mason/Charter Publishers, Inc.,
Corporation, Delran, NJ 1976.
McLaughlin, R.A., "That Old Bugaboo, Turnover", Datamation, Oct.,
Mendes, Kathleen S., "Structured Systems Analysis",
Review, MIT, vol. 21 no. 4, Summer, 1980.
Management
Sloan
Menard, Jayne B., "Exxon's Experience with the Michael Jackson
11, no. 3, Winter, 1980.
Data Base, vol.
Resource
Organiz tional
Moch, M.K., "Structure and
Administrative Science Quarterly, vol. 21, 1976.
1977.
Method",
Allocation",
and
Morse, E.V., "Size, Centralization and Organizational
Adoption of Innoavations", American Sociological Review, vol. 42, 1977.
Mohr, L., "Determinants of Innoavation
Political Science Review, vol. 63, 1969.
Nolan, Richard L., "Controling
Business Review, July, 1977.
------- , "Managing the Crisis
Review, March, 1979.
the
Cost
in Data
in Organizations",
of
Data
Processing",
Services",
Harvard
American
Harvard
Business
131
Shetal
Productivity,
Parikh, Girish, How to Measure
Enterprises, Chicago, 1981.
Programmer
Pelz, Donald C. and Andrews, Frank
Productive Climates for Research
1966.
M., Scientists in Organizations:
and Development, John Wiley & Sons,
Pierce, J.L. and Delbecq, A.L., "Organizational Structure, Individual
Attitudes and Innovation", Academic Management Review, vol. 2, 1977.
Porter, L.W. and Lawler, E.E., Managerial
Dorsey, Homewood, IL, 1968.
Attitudes
and
Performance,
Rogers, Everrett, Diffusion of Innovation, Free Press, Glenco, NY, 1962.
Current
Innovation:
Rowe, Lloyd A., and Boise, W.B., "Organizational
34,
vol.
Review,
Administration
Public
Research and Evolving Concepts",
1974.
Strassman, Paul A., "Managing
Business Review, Sept., 1976.
the
Costs
of
------"Strategic Aspects of
Information
1, 1982.
Technology and People, vol.
Information",
Harvard
Management",
Office:
Synnott, William R. and Gruber, W.H., Information Resource
John Wiley & Sons, New York, 1981.
Management,
Thurow, Lester T., conversation with the author, November, 1982.
Yourdan, Edward, Managing the Structured Techniques, Yourdon, Inc.,
York, 1976.
J.,
Holbek,
and
Zaltman, Gerald, Duncan, R.
Organizations, John Wiley & Sons, New York, 1973.
Innovations
New
in
Development
Efforts",
------- "Diffusion of Modern Software Practices", Management
vol. 28, no. 12, 1982.
Sciences,
Zmud, Robert W., #"Management of Large
MIS Quarterly, June, 1980.
Software
i.d.
M.I.T. STUDY OF ANALYST ATTITUDES TOWARDS,
AND USE OF NEW SYSTEfIS DEVELOPMENT TECHNOLOGIES
January 24. 1983
Version 3.0
H
FOR CODING
ONLY
Work Description
PLEASE ANSWER THE FOLLOWING QUESTIONS IN THE WAY
WHICII
BEST
DESCRIBES YOUR JOB RESPONSIBILITIES OVER THE PAST 12 MONTHS,
FROM JANUARY 1982 TO THE PRESENT
1.
What is your current job title?
2.
How long have you held this job?
3.
What projects have you worked on in this position over the past 12 months?
DURATION AND DESCRIBE PROJECT BRIEFLY.
a.
Client:
years
months
INDICATE CLIENT ORGANIZATION AND PROJECT
Duration:
months
Your role in that project (CHECK )NE)
Group Leader
Project Leader
Project Member
Other
Duration:
months
Your role in that project (CHECK ONE)
Brief Description:
b.
Client:
Group Leader
Project Leader
Project Member
Other
-T -
Brief Description:
-n -t
c.
"1"
-i5-
-UT
Client:
Duration:
months
Your role in that project (CIECK ONE)
Group Leader
Project Leader
Project Member
Other
Brief Description:
"-
FOR ADDITIONAL PROJECTS PLEASE USE LAST PAGE OF QUESTIONNAIRE.
***([IMPORTANT NOTE FOR SUPERVISORS AND PROJECT LEADERS:
ACTIVITIES, NOT YOUR PERSONAL ACTIVITIES.]***
4.
-Ef
f
"Iff2
FOR QUESTIONS 4-20 ANSWER IN TERMS OF THE PROJECT TEAMS'
What proportion of the time over the past 12 months have you worked in each of the following functional areas?
engineering
2
finance
logistics
other:
TOTAL
100 %
FOR CODING
ONLY
5.
-g2
-3-'
3TV
3T -i
What proportion of the time have you worked with each systems type?
operational/transactions systems
2
stewardship systems (e.g. reporting)
2
planning/analysis systems (e.g. decision support)
2
other:
2
TOTAL
6.
1002
What proportion of the time have you worked at each project level?
scoping/exploration
systems development/acquisition
major enhancement of existing system
application/user support, including
maintenance
-W
other:
TOTAL
7.
100 2
What proportion of the time have you worked on each of the following tasks? RESPONSES NEED NOT TOTAL 100% IF YOU HAVE
WORKED ON TASKS NOT LISTED HERE. FOR ANY TASKS YOU HAVE NOT WORKED ON. ENTER 0%.
understand business operation
define client needs
51
-5T
-5T
recommend solution based on alternatives
document solution specifications
%
produce logical data base design
analyze role of information (data dependencies) in business and solution
-T
design program to meet system specifications
2
code program with or without aids
2
document program with or without aids
-6W
provide user support
FOR CODING
ONLY
8.
Which three of these tasks do you find most important in their impact on the quality of your work?
PLEASE RANK ORDER. USING I FOR MOST IMPORTANT,
2 FOR SECOND MOST IMPORTANT,
3 FOR THIRD MOST IMPORTANT.
RANK
TASK
understand business operation
define client needs
recommend solution based on alternatives
document solution specifications
7_1
produce logical data base design
72
analyze role of information (data dependencies) in business and solution
design program to meet system specifications
code program with or without aids
document program with or without aids
-77
provide user support
9, 10.
Which three of these tasks do you find most difficultt?
ost satisfying?
PLEASE RANK ORDER. USING 1 FOR MOST DIFFICULT/SATISFYING. 2 FOR SECOND MOST DIFFICULT/SATISFYING
DIFFICULT/SATISFYING.
3 FOR THIRD MOST
RANK
TASK
7-
T7-
understand business operation
define client needs
recommend solution based on alternatives
-T
-ff
-iT
r--
document solution specifications
produce logical data base design
analyze role of information (data dependencies) in business and solution
-b
-T•
design program to meet system specifications
code program with or without aids
2T
document program with our without aids
provide user support
MOST DIFFICULT
MOST SATISFYING
FOR CODING
ONLY
Methodologies and Alternatives (Questions 11-19)
Methodologies
.t"structured methodologies (...
) have been developed for use in various tasks for which you may have
been responsible over the past 12 months. The questions in this section refer to your choice of either the
methodology or an alternative approach for each task you have spent time on.
PLEASE RESPOND TO TIIESE QUESTIONS ONLY FOR TIIETASKS YOU HAVE WORKED ON DURING THE PAST 12 MONTHS (SUPERVISORS:
ANSWER FOR YOUR PROJECT TEAM).
11.
To what extent have you used
. to carry out each task listed below? SEE KEY FOR EXTENT OF USE.
APPROPRIATE NUMBER IN EXTENT OF USE COLUMN.
CIRCLE N/A FOR ANY TASKS YOU HAVE NOT WORKED ON.
12.
When you have not used
., what has been your principal alternative approach? SEE KEY FOR ALTERNATIVE
APPROACH,
CIRCLE APPROPRIATE NUMBER IN ALTERNATIVE APPROACH COLUMN OR WRITE IN YOUR OWN RESPONSE IN BLANKS
PROVIDED.
KEY:
EXTENT OF USE
KEY:
ALTERNATIVE APPROACH
always
frequently
sometimes
infrequently
never
not applicable
my own method
flowcharting
peer discussions
client discussions
prototyping
other structured methodologies
other:
8
other:
NA
not applicable
TASK AND METHODOLOGY
EXTENT OF USE
a)
Understand the business operation (
b)
Define client needs (.
c)
Evaluate alternatives and recommend
solution (
13.
CIRCLE
)
..)
ALTERNATIVE APPROACH
1 2
3 4
5 N/A
1 2
3 4
5
6
7 8
N/A
1
2
3 4
5 N/A
1 2
3 4
5 6
7 8
N/A
1 2
3 4
5
1 2
3 4
5
7 8
N/A
When you used an alternative approach rather than
BRIEFLY IN TIIESPACE BELOW.
N/A
.,,what were your reasons:
-
PLEASE EXPLAIN
--
--
6
--
--
---
-
--
-----
FOR CODING
ONLY
14.
To what extent have you used . ..to carry out each task listed below?
SEE KEY FOR EXTENT OF USE.
CIRCLE APPROPRIATE NUMBER IN EXTENT OF USE COLUMN. 'CIRCLE N/A FOR ANY TASKS YOU HAVE NOT WORKED ON
(SUPERVISORS: ANSWER FOR YOUR PROJECT TEAM).
15.
When you have not used
.. what has been your principal alternative approach?
SEE KEY FOR
ALTERIZATIVE APPROACHI. CIRCLE APPROPRIATE NUMBER IN ALTERNATIVE APPROACH COLUMN OR WRITE IN YOUR OWN
RESPONSE IN BLANKS PROVIDED.
KEY:
EXTENT OF USE
I
2
3
4
5
N/A
KEY:
ALTERNATIVE APPROACH
always
frequently
sometimes
infrequently
never
not applicable (Haven't
done this task)
TASK AND METHODOLOGY
my own method
flowcharting
peer discussions
client discussions
prototyping
other structured methodologies
other:
8
other:
NA
not applicable
EXTENT OF USE
ALTERNATIVE APPROACH
a) Analyze role of information in business
and In solution (;.. )
1 2 3 4 5 N/A
1 2 3 4 5 6 7 8 N/A
b) Produce logical data base design ( ..)
1 2 3 4 5 N/A
1 2 3 4 5 6 7
-W7 16.
When you used an alternative approach rather than -'' . what were your reasons:
PLEASE EXPLAIN
BRIEFLY IN THE SPACE BELOW.
--
----
--
8 N/A
FOR CODING
ONLY
17.
to carry out each task listed below? SEE KEY FOR EXTENT OF USE.
To what extent have you used
CIRCLE N/A FOR ANY TASKS YOU HAVE NOT WORKED ON.
CIRCLE APPROPRIATE NUMBER IN EXTENT OF USE COLUMN.
18,
When you have not used ... what has been
CIRCLE APPROPRIATE
ALTERNATIVE APPROACH.
RESPONSE IN 11E BLANKS PROVIDED.
KEY:
EXTENT OF USE
your principal alternative approach? SEE KEY FOR
NUMBER IN ALTERNATIVE APPROACH COLUMN OR WRITE IN YOUR OWN
KEY:
ALTERNATIVE APPROACH
always
frequently
sometimes
infrequently
never
not applicable
my own method
flowcharting
peer discussions
client discussions
prototyping
other structured methodologies
other:
other:
not applicable
TASK AND METHODOLOGY
-a
--
ALTERNATIVE APPROACH
EXTENT OF USE
a)
Design program to meet system
specifications ( .)
1 2 3 4 5 N/A
1 2
3 4 5
6 7 8 N/A
b)
Code program ( .) (with or without AIDS)
1 2 3 4 5 N/A
1 2 3 4 5
6 7 8 N/A
c)
Document program (_ ) (with or without AIDS)
1 2
N/A
1 2 3 4 5
6 7 8 N/A
M
53-i
3If -6
19. When you used an alternative approach rather than
IN THE SPACE BELOW.
3
4
5
. , what were your reasons:
a)
b)
c)
20. What approach do you use in carrying out support tasks?
PLEASE EXPLAIN BRIEFLY
FOR CODING
ONLY
We are now interested in how you personally perceive your work, your organizational environment and
yourself.
21.
"GCood Job" questions
a.
We are interested in finding out what factors are important to you personally in evaluating your
own work. Please select from the following list the three factors which are the most important
to you in judging your own work as an analyst, and rank order these three. (1 - most important.)
Your own work as an analyst,
completed on time
clearly documented
completed within the budget
easily enhanced
easily run by the user
easily maintained
designed to the user's specifications
meeting standards and guidelines for methodology requirements
accuracy of solution
good communication with user
other:
b.
6
"-7
-E-
Now please rank three factors in the order of importance which you think your supervisor would
assign them In judging the results of your work:
completed on time
clearly documented
completed within the budget
easily enhanced
--fl
"T
easily maintained
easily run by the user
deuigned to the user's specifications
meeting standards and guidelines for methodology requirements
accuracy of solution
good communication with user
other:
FOR CODING
ONLY
c.
Now please rank three factors in the order in which you think your clients would rank them in judging
the results of your work:
completed on time
clearly documented
completed within the budget
-g71
easily enhanced
easily maintained
easily run by the user
designed to the user's specifications
-TG
meeting standards and guidelines for methodology requirements
accuracy of solution
good communication with user
other:
d.
In your opinion, which three of the following factors are the most important to upper level management
in meeting managerial goals for your organization:
completed on time
clearly documented
completed within the budget
easily enhanced
easily maintained
easily run by the user
designed to the user's specifications
meeting standards and guidelines for methodology requirements
accuracy of solution
good communication with user
other:
-8-
O
FOR CODING
ONLY
22.
..
and
.
Whether you have ever used
advantages and disadvantages of the current
have listed below some possible responses.
your thoughts, please write your own in the
or not, we would like to know what you think are the
versions of these systems development technologies. We
If you feel that these responses do not adequately express
space provided on page 10.
a.
For each of the technologies, please rank order the three most important advantages and the three
most important disadvantages by writing the corresponding number in the allotted space on page 10.
b.
Also please indicate your overall rating of each technology in the space provided on page 10
using the following scale.
excellent
very poor
ADVANTAGES
DISADVANTACES
1.
Structured approach
1.
Time consuming to use
2.
Easier operation by user
2.
Restrictive (inhibits creativity)
3.
Faster
3.
Unfamiliar/requires learning time and effort
4.
Improves requirement definition
4.
Not oriented to my application
5.
Provides structured design
5.
Not useful for maintainence
6.
Produces better documentation
6.
Difficult to use
7.
Application more maintainable
7.
Too expensive for client (time or budget)
8.
Easier to create and alter information diagrams
8.
Incompatible with programming language I used
9.
Provides improved communications
9.
Poor hard copy quality (for aids)
10.
Helps me understand clients' business
10.
Hardware equipment (for aids) inaccessible
11.
Good for less experienced analysts
11.
Other
12.
Meets standards and guidelines for
methodology requirements
13.
Other (WRITE RESPONSE IN SPACES ON PACE 10)
FOR CODING
ONLY
ENTER NUMBER FROM KEYS ON PACE 9 OR WRITE IN YOUR OWNRESPONSE.
TECHNOLOGY
ADVANTAGES
DISADVANTAGES
-uT
77
m3
-u
OVERALL RATING (PLEASE CIRCLE)
1
2
3
4
5
OVERALL RATING (PLEASE CIRCLE)
1
2
3
4
5
OVERALL RATING (PLEASE CIRCLE)
1
2
3
4
5
OVERALL RATING (PLEASE CIRCLE)
1
2
3
4
5
OVERALL RATING (PLEASE CIRCLE)
L
2
3
4
5
-37
"76
-T
-n
f7~ 7T
-3-
3-n
"Ti
-g
-g5 --Y
rsT
-10-
FOR CODING
ONLY
Nov we would like to know how skilled you feel you personally are in the use of certain systems development
methods, techniques, and languages and the extent in which these skills have been augmented by TRAINING AND
COMMUNICATION.
23.
How skilled are you in the use of each of the following:
(FILL IN THE BLANK TO THE LEFT OF THE ITEM WITH A NUMBER FROM 1 to 5)
1
2
Extremely Skilled
3
Very Skilled
Moderately Skilled
4
Slightly Skilled
5
Not at all skilled
IMPLEMENTATION TOOLS AND LANGUAGES
1.
FORTRAN
2.
NOMAD,
3.
COBOL
4.
PL-1
FOCUS,
RAMIS OR OTHER FOURTH-GENERATION LANGUAGES
ANALYSIS AND DESIGN TECHNIQUES
W
6.
7.
8.
9.
-U
OTHER (FOR EXAMPLE:
10.
APPLICATION GENERATORS,
MINI-COMPUTERS)
Other, please specify
It1. Other, please specify
-Il-
FOR CODING
ONLY
24.
PLease indicate next to the tool or technique listed below, the number corresponding to the source of
instruction which was most helpful, second most helpful, and third most helpful.
SOURCES OF INSTRUCTION
1.
Consulting with designated local experts on the new tool or technique.
2.
Consulting with other experienced user (informal expert)
3.
Consulting with the Computer Technology Division
4.
Formal Training Sessions
5.
Instruction Manual/Documentation
6.
Self (i.e., Trial and Error)
7.
Consulting with S&C Applications Software Coordinator
8.
Other (specify)
(ENTER NUMBER 1-8)
(ENTER NUMBER 1-8)
Most
Second Most
Helpful
Helpful
0" 717 "1-
W2 -U
-6
-12-
(ENTER NUMBER 1-8)
Third Most
Helpful
FOR CODING
ONLY
29.
Do you know anyone who is a real advocate for (really "sells")
NO
If yeas
.?
YES
How well do you know that person? (CIRCLE NUMBER)
5
not well
at all
very well
30.
Do you know anyone who is a real advocate for (really "sells")
NO
If yes:
How well do you know that person? (CIRCLE NUMBER)
very well
not well
at all
31.
How many times in the past 12 months have you had the following types of contact with the Computer
Technology Division
' ? (CIRCLE THE CATEGORY IN COLUMN "A" WHICH BEST
REPRESENTS THE NUMBER OF TIMES).
32.
For each type of contact you had, please indicate how productive in general you felt that communication
was. (CIRCLE A NUMBER IN COLUMN "B": 1 - very productive; 2 - somewhat productive; 3 - no comment;
4 - somewhat unproductive; 5 - very unproductive.)
(A)
(B)
CIRCLE A CATEGORY
0/1-2/3-4/5 or more times
1
2
3
4
5
technology
You made a suggestion about modifying
0/1-2/3-4/5 or more times
1
2
3
4
5
the training
You made a suggestion about modifying
a technology
e. CTD asked you if you were using one of the
0/1-2/3-4/5 or more times
1
2
3
4
5
0/1-2/3-4/5 or more times
1
2
3
4
5
L
l
2
2
3
3
4
4
5
5
a. Walk-through of new technology
2W
-_f
CIRCLE A NUMBER
b. You requested help in using a new
-7
f.
g.
h.
"ZT
1.
new technologies
Informal conversation with someone in CTD
You attended a CTD program review
Received information from CTD on
methodology updates
other (please describe)
-14-
0/1-2/3-4/5 or more times
0/1-2/3-4/5 or more times
0/1-2/3-4/5 or more times
0/1-2/3-4/5 or more times
0/1-2/3-4/5 or more times
FOR CODING
ONLY
How many times in the Dast 12 months have you had the following types of contact with the
Systems and Computer Senior Staff?
(CIRCLE THE CATEGORY IN COLUMN "A" WHICH BEST
REPRESENTS THE NUMBER OF TIMES).
For each type of contact you had, please indicate how productive in general you felt that communication
was. (CIRCLE A NUMBER IN COLUMN "B"t I - very productive; 2 - somewhat productive; 3 - no comment;
4 - somewhat unproductive; 5 = very unproductive.)
(A)
(B)
CIRCLE A NUMBER
CIRCLE A CATEGORY
Walk-through of new technology
0/1-2/3-4/5 or more times
46
48
50
52
54
56
58
60
You requested help in using a new
technology
You made a suggestion about modifying
the training
You made a suggestion about modifying
a technology
CTD asked you If you were using one of the
new technologies
Informal conversation with someone in CTD
You attended a CTD program review
Received information from CiD on
methodology updates
other (please describe)
6T
-15-
0/1-2/3-4/5
more times
0/1-2/3-4/5
more times
0/1-2/3-4/5
more times
0/1-2/3-4/5
0/1-2/3-4/5
0/1-2/3-4/5
more times
more times
0/1-2/3-4/5
0/1-2/3-4/5
more times
more times
more times
1
2
3
4
5
FOR CODING
ONLY
33.
Please Indicate the person to whom you would go FIRST for day to day consultation about the following
application technologies and indicate how accessible that person is (in terms of their willingness/time to help
and geographic proximity).
I - not Accessible; 2 - fairly Accessible; 3 - reasonably Accessible; 4 - Accessible;
and 5 a extremely Accessible
Computer Technology
Div..
Locally
Designated
Contact
Section
Head
Supervisor
Project
Peer
Leader
Accessibility
(1 - 5)
a. SSA
b. LDA
c. PST
34.
We would like to know what you have heard about the first version of
hardware.
a. Did you use that version of
no
IF YES:
Was that experience:
(CHECK ONE)
. the one on the Tektronix
yes
mostly positive
mostly negative
About how many people whom you know tried that version?
b. (NOTE NUMBER ONLY)
people
c. Have the comments you heard been (CHECK ONE)
35.
mostly positive
fow would ycu rate each technology on the elements listed below (compared to other technologies you are
familiar with), using a scale of 1 to 5, where 1 - excellent and 5 - very poor?
5
excellent
NOTE:
~73 -7F-r
mmT-p 1
-fl-U
mostly negative
very poor
IT IS NOT NECESSARY TO HAVE USED THE TECINOLOGY TO ASSIGN IT A RATING
Documentation
Access to Consulting
Help
---
'-tI
-16-
FOR CODING
ONLY
VALUES IN ORGANIZATIONS
36. Each of the words or phrases listed in the far left column describes a quality or characteristic of
work. From the row of phrases and associated numbers (1 to 7), select one (BY CIRCLING THE NUMBER)
which best describes the attitude of YOUR WORK GROUP towards that quality or characteristic. How is
it valued or emphasized?
Very Highly
Valued/Emphasized
Highly Valued Valued
Somewhat
Valued
Not Much
Valued
Not at all Disliked/
De-emphasized
Valued
a.
Creativity
1
2
3
4
5
6
7
b.
Standardization
1
2
3
4
5
6
7
c.
Craft
(Art; Individual style)
1
2
3
4
5
6
7
d.
Engineering
(Structured Process)
1
2
3
4
5
6
7
e.
Long Range Work Goals
1
2
3
4
5
6
7
f.
Immediate Client Needs
1
2
3
4
5
6
7
g.
Getting the work out
quickly
1
2
3
4
5
6
7
h.
High quality in design
1
2
3
4
5
6
7
i.
Easily maintained, well
documented code
1
2
3
4
5
6
7
T4-
Tr
IT
-17-
FOR CODING
ONLY
37.
Each of the words or phases listed in the far left column describes a quality or characteristic of
work. From the row of phrases and associated numbers (1 to 7), select one (BY CIRCLING THE NUMBER)
which best describes the attitude of the Systems and Computing Management ,
.
toward that quality
or characteristic, as nearly as you can judge from what you know about that organization. How do
they value or emphasize that quality?
Very Highly
Highly Valued Valued
Valued/Emphasized
a.
Creativity
1
b.
Standardization
Craft
c.
2(Art;
Somewhat
Valued
Not Much
Valued
Not at
all Valued
Disliked/
De-emphasized
2
3
4
5
6
7
1
2
3
4
5
6
7
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Individual style)
d.
I(Structured
Engineering
Process)
e.
Long Range Work Coals
1
2
3
4
5
6
7
f.
Immediate Client Needs
1
2
3
4
5
6
7
g.
Getting the work out
quickly
1
2
3
4
5
6
7
)-*
h.
High quality in design
1
2
3
4
5
6
7
tO
1.
Easily maintained, well
code
1
2
3
4
5
6
7
26
Tdocumented
-18-
FOR CODING
ONLY
AGREE/DISAGREE STATEMENTS
38.
Please Indicate you agreement or disagreement with the following statements by selecting a number from
1 to 5, where 1 means strongly agree, 2 = agree, 3 a neutral, 4 a disagree S * strongly disagree.
WRITE SELECTED NUMBER IN BLANK.
w-
1.
No one gets rewarded for effective design and programming around here.
2.
Structured application development (as a concept) is a good idea.
3.
My client users cannot determine whether a piece of code is efficient or not.
4.
Organizational guidelines do not influence my decision to use
5.
Computer Technology Division
that will make my
job easier.
6.
People in Application Development Groups should have major involvement In methodology and tool
7.
The application development technologies developed by Computer Technology Division
scalable to the extent warranted by the nature of the application.
8.
In the next 5-7 years,
9.
Most of my dissatisfactions with my job are due to the nature of the work.
10.
People in Application Development Groups know best what tools their job requires.
35
is actively designing methodologies and tools
3development.
TI--
wilt be obsolete.
UL
O
11.
Structured application development techniques (in general) Increase maintainability of code.
12.
It is not important for my client users to understand how a program works as long as it does the job.
13.
My supervisor (over the past year) would have liked me to use
14.
The guidelines for application development are generally followed by programmer-analysts whom I know.
15.
I am rewarded according to the quality (versus quantity) of product I put out.
16.
My client users want maintainability more than efficiency in code.
17.
Computer Technology Division
18.
The main reason I haven't used the new systems development technologies more is that I don't have time
h
31-
are
..
understands what it takes to do my job.
to.
19.
System development methodologies inhibit my creativity.
20.
No one In the organization besides the Computer Technology Division
cares whether or not I use the new system development technologies.
21. We need to move towards application development as a highly structured, engineering process.
-19-
really
FOR CODING
ONLY
22. If and when I use the ncw system development technologies developed by Computer Technology Division
my client users appreciate my work more.
increases productivity.
I do not need to use structured application development techniques.
33-
I won't be developing applications long enough to make it worthwhile to learn new methodologies
and techniques.
My client users don't really want to know how a program works.
Application development is an art.
My supervisor (over the past year) would have liked me to use
..
LEARNING STYLE INVENTORY
This inventory is designed to assess your method of learning.
As you take the inventory, give a high rank
to those words which best characterize the way you learn and a low rank to the words which are least
characteristic of your learning style.
You may find it hard to choose the words that best describe your learning style because there are no right
or wrong answers. Different characteristics described in the inventory are equally good. The aim of the
inventory is to describe how you learn, not to evaluate your learning ability.
Instructions
There are nine rows of four words listed below. Within each row, rank order each of the four words
assigning a 1 to the word which best characterizes your learning style, a 2 to the word which next best
characterizes your learning style, a 3 to the next most characteristic word, and a 4 to the word which is
least characteristic of you as a learner. Be sure to assign a different rank number to each of the four
words in each row, Please do not make ties.
1.
discriminating
tentative
involved
practical
2.
receptive
relevant
analytical
impartial
3.
feeling
watching
thinking
doing
4.
accepting
risk-taker
evaluative
aware
5.
intuitive
productive
logical
questioning
6.
abstract
observing
concrete
active
pragmatic
6T-
~63- 13W757-1
7-
7--
-61 7-57•-
TV"
-T"
U-
6- TT- U-
7.
presentoriented
reflecting
futureoriented
8.
experience
observation
conceptualization
experimentation
9.
intense
reserved
rational
responsible
-20-
FOR CODING
ONLY
40.
Now please indicate how you. personally feel about the job you have held the last 12 months.
Each of the statements below is something that a person might say about his or her job. You are to
indicate your own personal feelings about your job by marking how much you agree with each of the
statements, based on the following scale:
SA
A
N
D
SD
Strongly Agree
Agree
Neutral
Disagree
Strongly Disagree
a.
FOR EACHSTATEMENT BELOW, PLEASE CIRCLE THE APPROPRIATE RESPONSE
It's hard, on this job, for me to care about whether or not the work
b.
My opinion of myself goes up when I do this job well,
c.
Generally speaking, I am very satisfied with this job.
d.
Most of the things I have to do on this job seem useless or trivial.
e.
I usually know whether or not my work is satisfactory on this job.
f.
I feel a great sense of personal satisfaction on this job.
g.
The work I do on this job is very meaningful to me.
h.
I feel a very high degree of personal responsibility for the work I do
on this job.
i.
I frequently think of quitting this job.
J.
I feel unhappy when I discover that I have performed poorly on this job.
k.
I often have trouble figuring out whether I'm doing well or poorly
on this job.
1.
I feel that I should personally take the credit or blame for the
results of my work on this job.
m.
I am generally satisfied with the kind of work I do on this job.
n.
My own feelings generally are not affected much one way or the other
by how well I do on this job.
o.
Whether this job gets done right is clearly my responsibility.
gets done rignt.
T627IT-f
TT
wu
M--
wb
-21-
.
SA
A
N
D
SA
A
N
D
SA
A
N
D
SA
A
N
D
SD
FOR CODING
ONLY
PLEASE CHECK APPROPRIATE RESPONSE
41.
Sex:
1 Male
Age:
I Under 20
20 - 29
2
30 - 39
3
40 - 49
4
5
50 - 59
6
60 or over
Education:
2 Female
PLEASE CHECK THE HIGHEST LEVEL YOU HAVE ATTAINED
1. High school degree
2. College degree, computer concentration
3j College degree, other concentration
4. Masters degree, computer concentration
5. Hasters degree, other concentration
6. Higher degree
zr ZT- wT- ZZw- Z- "7TOSt- 'T0-
3-1 3-2 37 3ZF 33
6
How long have you worked at this company?
years
months
How long have you worked in the field of systems development?
years
months
How long have you worked in systems development activities at
years
How long have you worked in your current work group?
years
How many members are there in your work group?
members
42.
a.
In general do you see yourself pursuing a managerial or technical career path?
Managerial
b.
Technical
In the future, what type of work do you see yourself doing? Write the appropriate number in
the blank beside each question, or write in your own response.
1.
2.
3.
4.
Systems development/management
Managing a technical group
General management
Other (Please specify):
TYPE OF WORK (enter number)
1 year from now?
53
months
3 years from now?
5 years from now?
-22-