Q:
I! 2.. Yer
{I

L». J. ; e
inflgwinuhufi

a.v,.
.: . ,
. I.

I .14 1.31535...
1.: (DIS.
3" 'l
(313.5,...)f. 5'13...)
. , "r V'LIOfI‘I-ni"x.‘.t.~!’l
; t" u’r!‘ u-Aaflgui
$7.02!...-
4.>|..}..l§ p I. ‘ l
‘ I;
~ -31...
‘ a 1.5.).
l
. . ‘ .
, L... 1‘...
V bin-t"! -‘ .
‘Isv.i.,:s.:= P...
. {V 1" ‘ ‘
‘ .
if;
21.3.... :1
t... 5....
.. .3.
.h. . iii;
.V‘Q EPIC52.§I’-3n.v
1.53er .i!
I,

$5..

”3.
; 1r

. 2.1.3955...
. c:\ .x.

 

.22.. E.
.s.

v

:15 .v
c...:t.. .
a. an >l.l¢I .
. .l. 2..
, tr. .3 31‘
2.. I. . .12... .
Oiltrsxbluxtiwt g: .
itpkulii o

 

‘- If." vl‘ a-
. :Is 1...): I...
. .Itn... a.
In}
’91,}. ..
x .

I
~ .4
.“1I._ .k.

3:913:11 .453: .9.
.rknlhzft; 0.51:...:
a! erw. v.
V ..
..'(.A.. .7 v. .
.l ‘ :41! .lv.
1.. 4uo31o:

.. LE..-
Ir
w. ...i .
1.1.3.
‘ .
32.5.»!
I... 85.3.... . .11: a: ‘
; . ,Lf .

:1..».tvr..tlhl> ‘ i
:. .)e'f.v;zn{ll
1?;‘X:
.,. .1 i
.r.5..fo:..
:1) :x .

 

.9:
u... .91.). ..
1.31:}: 9..
r, #:-

.

 

 

Itinrrli 4‘
. I .5. .y.

1:: A ‘ .I
II;

P (1.: ‘
I)... v»? ).,

MICHIGAN sure u

I Hi 1:21:11 Im“Mimi/W

93 00904 8 78]

 

ll

 

 

 

 

 

 

 

 

 

W

3

This is to certify that the

dissertation entitled

The Multidimensionality of Organizational Change:
Developing and Testing Models of Change in Organizations

presented by

Parshotam Dass

has been accepted towards fulfillment
of the requirements for

Ph,D. degree in Management Policy &

Strategy

////M@L

V Major professor

 

 

Date May 2]., 1993

MSU is an Affirmative Aclirm /I-_‘qual Opportunity Institution 0-12771

 

' LIBRARY W
Michigan State
University

K; J

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Affirmative Action/Equal Opportunity Institution
eWDmG-o.‘

THE MULTIDIMENSIONALITY OF ORGANIZATIONAL CHANGE:
DEVELOPING AND TESTING MODELS OF CHANGE

IN ORGANIZATIONS

By

Parshotam Dass

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Management

1993

Cohei
due [I
Chang
COrpO
COM‘

nuflfid

ABSTRACT
THE MULTIDIMENSIONALITY OF ORGANIZATIONAL CHANGE:
DEVELOPING AND TESTING MODELS OF CHANGE
IN ORGANIZATIONS

By

Parshotam Dass

The research findings on organizational change have not generated a
coherent body of knowledge. This study postulated that this has been partly
due to neglecting distinctions among different dimensions of organizational
change. Using multiwave data of over 700 national and international business
corporations registered on stock-exchanges in the United States from
COMPUSTAT II and Value Line DataFile, this study examined the issue of
multidimensionality of change in organizations. Two variables of
organizational change were studied: change in product diversity from 1978-
1990, and change in international geographic diversity of organizations from
1984—1990. Four dimensions of organizational change were identified:
increment, relatedness, pervasiveness, and direction. These four dimensions of
change generated four dependent variables in the study: degree of dramatic

change, degree of related change, degree of pervasive change, and convergent

ch
dc
tfis
va

lus

ire
cha
fiou
on;
org;
dran
Shhi
dcgn
ChanJ
theh

Organ

change, respectively. Hypotheses regarding the relationships between the four
dependent variables of change and a variety of independent variables were
tested using hierarchical regression and probit analysis. The independent
variables included sources of change depicting attributes of organizational
history, strategic choice, performance, and environment.

The results of this study indicated that the four dimensions of change
were not strongly correlated with one another. Therefore, organizational
change could not be treated as unidimensional. Different dimensions were
found to have different patterns of association with the correlates of
organizational change. For example, in the case of change in product diversity,
organizational performance had a negative relationship with the degree of
dramatic change and a U-shaped relationship with convergent change.
Similarly, environmental munificence had a negative relationship with the
degree of related change whereas it had a positive relationship with convergent
change. The study provided no support for a generic model of change across
the two variables of change: product and international geographic diversity in
organizations.

The primary contribution of this study is that it has laid a foundation for
dimensionalizing organizational change. Future research can consider other
variables, dimensions, and sources of change to build upon these findings to

help researchers evolve a coherent and cumulative body of knowledge on

organizational change.

Copyright by
IMARSflHDTV¥NlILASS

1993

Dedicated to my parents and sister

and to the memory of my brother

proc:

contr

Mike

me [0

deceplll

WOIVes_

ACKNOWLEDGEMENTS

This dissertation represents completion of one phase of a continuing
process of learning. It has been a fascinating experience for me that involved
contributions of many people. First of all, my gratitude is to the members of
my Dissertation Committee, Mike Moch, John Wagner, and Larry Stimpert.
Mike Moch, my mentor, and Chair of the Dissertation Committee, introduced
me to theory development on organizations. He shared with me the challenging
task of developing sophisticated models that could be empirically tested. With
his great sense of humor and patience for dealing with ambiguity, he walked
with me through the jungles. He provided me with the compass and the torch
and made sure I was using them. It was especially important because I (and he
too) was not interested in choosing the places just because they were brightly
lit. Not only did he take care of my professional development during the
course of my program, but he also watched for the rain and the sun over my
head to enable me to reach my full potential.

John Wagner introduced me to a wide range of perspectives and gave me
the independence to choose my own. He taught me that appearances could be
deceptive, and that rigorous inquiry would be my tool for dancing with the

wolves. He equipped me with the eyes of a hawk for my journeys. Larry

vi

 

Su‘
exj

coi

ha‘

b6!

2“:
to
co:
stu
in

int

We

Stimpert helped me sharpen my insights by allowing me to draw upon
experiences and treasures of his recent similar journeys. His know—how and
collegiality have been invaluable to me in my endeavors.

Besides the Dissertation Committee members, other faculty members
have contributed to my rich experience at Michigan State University. It has
been a great privilege to work with Dan Ilgen who prepared me in the science,
art and craft of conducting research. Not only do I appreciate him for his
guidance and expertise but also for his kindness and generosity. I am grateful
to John Hollenbeck for allowing me to benefit from his great research methods
competence. I thank the other faculty members, staff, and fellow doctoral
students in the department for their companionship that made me feel at home
in East Lansing during the ups and downs that are common in the life of
international students.

I would also like to acknowledge the contribution of other people who
were very helpful to me in different ways during my doctoral program.
Therefore I thank Professors Bruce Allen, Richard Hill, and Stanley Stark.
David Mendez answered many challenging questions on quantitative methods.
Manjit Virdi was very supportive in the dissertation and other projects in my
virtual and physical journeys. I also thank Sue Polhamus, Kathy Mullins, and
Lonnie Herman, staff members in the department, who pleasantly facilitated me

all the time. Faye Becky, Business Librarian, and Consultants in the Computer

vii

 

Center
services
Univert
Vidhay

prograi

prepar.
I woul
Harbh
my lh.
Unive

me to

Wife‘
is hai
to 0U
impo
Why

Center at Michigan State University are appreciated for their very professional
services. Thanks are also due to computer consultants Melora Ghoosey of the
University of Michigan, Peter Luan of Arizona State University, and Naneera
Vidhayasirin of the University of Illinois for their aid in computer
programming.

I owe an immense gratitude to my earlier teachers and friends who
prepared me for the challenges of a career in research and teaching. Therefore,
I would like to thank Professors P. N. Chowdhury, Sukhdev Singh Gill,
Harbhajan S. Kehal, and D. R. Singh. I also take this opportunity to express
my thanks for the faculty members, staff, and doctoral students at the
University of Arkansas, Fayetteville, Arkansas, for their support that allowed
me to complete this dissertation in time.

Last and most important of all, there is one person, my best friend and
wife, Veena, whose contributions are so intertwined with those of mine that it
is hard for me to delineate the boundaries. I appreciate her love and dedication
to our life-long relationship. All of these experiences would have been
impossible without the dreams and courage of our other family members, that is

why this effort is dedicated to them.

viii

 

III.

IV,

II.

III.

IV.

TABLE OF CONTENTS

Chapter 1: Introduction ........................... .1
Chapter 2: Review of Literature ....................... 8
A. The Sources of Persistence and Change
in Organizations .............................. 9
1. Organizational Inertia, Stability,
or Persistence .......................... 10
2. Organizational Change and Adaptation ........... 14
3. Organizational Performance ................. 23
B. The Process of Change ........................ 25
C. Integrating the Perspectives .................. 29
D. Gaps in Existing Literature ...................... 34
Chapter 3: Conceptual Framework and Hypotheses .......... 36
A. The Process of Change ........................ 37
1. Dimensions of Organizational Change ........... 40
B. The Dimensions of Change and Their Correlates ........ 47
1. Organizational Size ...................... 48
2. System Coupling ........................ 51
3. Environmental Munificence ................. 53
4. Environmental Dynamism .................. 54
5. Asset Specificity ........................ 55
6. Organizational Risk Taken .................. 57
7. Organizational Resources ................... 59
8. Organizational Performance ................. 61
9. Interactions ........................... 64
Chapter 4: Methods ............................. 69
A. The Sample ............................... 70
B. Variables and Their Measurement .................. 72
1. Dependent Variables ...................... 72
2. Independent Variables ..................... 80

ix

VI.

VII .

C. Statistical Analysis ........................... 86

1. Power Analysis ......................... 86
2. Data Analysis .......................... 87
Chapter 5: Results .............................. 89
A. Matrix of Correlations ......................... 90
B. Factor Analyses ............................. 94
C. Regression and Probit Analyses .................. 100
1. An Overview ......................... 100
2 Degree of Dramatic Change ................ 102
3. Degree of Related Change ................. 115
4. Degree of Pervasive Change ................ 123
5 Convergent Change ..................... 130
Chapter 6: Discussion ........................... 143
A. Summary of Findings ........................ 144
B. Limitations of the Study ....................... 156
C. Contributions of the Study ..................... 162
List of References ............................... 167

 

Table

Table

Table

Table

Table

Table

Table

Table

Table 1

Table 2

Table 3

Table 4

Table 5

Table 6

Table 7

Table 8

Table 9

Table 10

Table 11

LIST OF TABLES

Pearson Correlation Coefficients among Dimensions
of Change in Product Diversity ................... 91

Pearson Correlation Coefficients among Dimensions
of Change in International Diversity ................ 92

Factor Analysis of Measures of Change in Product
Diversity ................................. 95

Factor Analysis of Measures of Change in
International Diversity ......................... 96

Pearson Correlation Coefficients among Independent
Variables of Change in Product Diversity ............. 98

Pearson Correlation Coefficients among Independent
Variables of Change in International Diversity .......... 99

Results of Hierarchical Regression Analysis of
Degree of Dramatic Change in Product Diversity ....... 103

Results of Hierarchical Regression Analysis of
Degree of Dramatic Change in International
Diversity ................................ 105

Results of Hierarchical Regression Analysis of
Degree of Related Change in Product Diversity ........ 116

Results of Hierarchical Regression Analysis of
Degree of Related Change in International Diversity ..... 117

Results of Hierarchical Regression Analysis of
Degree of Pervasive Change in Product Diversity ....... 124

xi

 

 

 

Tablc

Table

Table

Table 12

Table 13

Table 14

Results of Hierarchical Regression Analysis of
Degree of Pervasive Change in International
Diversity ................................ 126

Results of Probit Analysis for Convergent Change in
Product Diversity ........................... 131

Results of Probit Analysis for Convergent Change in
International Diversity ........................ 132

xii

 

 

Figure
Figure
Figure
Figure

Figure
Figure
I:igure
I:lgure
Pigm-e
I:lgllre ]
Figure 1
Figure 1

FiEUre I:

Figure
Figure
Figure
Figure

Figure

Figure

Figure

Figure

Figure

1

2

3

Figure 10

Figure 11

Figure 12

Figure 13

LIST OF FIGURES

Incremental-Dramatic Change .................... 41
Related-Unrelated Change ...................... 43
Piecemeal-Pervasive Change ..................... 45
Convergent-Divergent Change .................... 46

Organizational Size and Degree of Dramatic Change
in Product Diversity ......................... 108

Organizational Size and Degree of Dramatic Change
in International Diversity ...................... 109

Organizational Resources and Degree of Dramatic Change
in International Diversity ...................... 111

Interaction of Performance and Dynamism, and
Degree of Dramatic Change in Product Diversity ....... 114

Performance and Degree of Related Change
in Product Diversity ......................... 119

Organizational Size and Degree of Related Change
in International Diversity ...................... 122

Organizational Size and Degree of Pervasive Change
in Product Diversity ......................... 128

Performance and Convergent Change
in Product Diversity ......................... 134

Performance and Convergent Change
in International Diversity ...................... 136

xiii

 

 

Figm

Figur

Figure 14 Organizational Size and Convergent Change
in International Diversity ...................... 138

Figure 15 Organizational Resources and Convergent Change
in International Diversity ...................... 140

xiv

Chapter 1

Introduction

 

evol
inco
attn'l
iden'

SOUn

Organizational persistence and change are of immense theoretical and
practical significance (Pettigrew, 1987). The question of persistence versus
change is posed in different disciplines in several ways (Van de Ven & Poole,
1988). In organizational studies, it has been a central point for many years
(Romanelli & Tushman, 1986). The research findings have not, however,
evolved into a coherent body of knowledge (Ginsberg, 1988). In this study, the
incoherence of the research findings and their limited usefulness in practice is
attributed to considering organizational change as unidimensional. This study
identifies multiple dimensions of organizational change and relates them to
sources of change.

The objective of this study is to test multidimensionality of organizational
change. Two variables namely, change in product diversity and change in
international diversity of organizations are examined in the study. Sources of
change such as organizational performance, industry environment, and strategic
choice are considered and their relationships with multiple dimensions of
change are tested. It is expected that the dimensions of change will be weakly
correlated with one another. The dimensions of change are likely to have
different correlates, or different patterns of association with their correlates. In
other words, different dimensions are expected to have different models of
change. However, a generic model is expected to apply to change in product

and international diversity of organizations.

2

P
persister
(Roman:
conceptu
Child. 11
1980). (
factor the
Tushman.

Organizati

3

Previous research on organizations has covered three broad sources of
persistence and change: inertia and history, environment, and strategic choice
(Romanelli & Tushman, 1986). These correlates of change have been
conceptualized in a wide range of organizational theories (e. g., Boeker, 1989;
Child, 1972; DiMaggio & Powell, 1983; Hannan & Freeman, 1989; Scherer,
1980). Organizational performance, though relatively understudied, is another
factor that is a cause of change in organizations (Ginsberg, 1988; Romanelli &
Tushman, 1986). Performance can influence subsequent changes in
organizations.

Various models of organizational change have been analyzed in previous
research. Basically, these are models of evolutionary and revolutionary change
(T ushman & Romanelli, 1985; Nelson & Winter, 1982). However, finer
distinctions can be made using piecemeal and pervasive change, and
incremental and dramatic change (Johnson, 1987; Lindblom, 1968; Miller,
1982; Quinn, 1980). Other suggested models involve first-, second-, and third-
order change (Bartunek & Moch, 1987; Moch & Bartunek, 1990; Watzlawick,
Weakland, & Fisch, 1974) and alpha, beta, and gamma change (Golembiewski,
Billingsley, & Yeager, 1976). Moreover, organizations experience convergent
and divergent change (Mintzberg & McHugh, 1985; Pettigrew, 1985, 1987;
Tushman & Romanelli, 1985) and related and unrelated change (Rumelt, 1974;

Palepu, 1985).

(
mQMm
change :
isolated
change a
organiza
Studying
general i
cases res
dfifingufi
Pcrforma
DOSlllVel)
Fombr‘un

fimfingsi

4

Organizational change is most frequently studied in terms of its
magnitude (i.e., more change or less change). Usually researchers analyze
change in a generic form; sometimes they analyze one attribute of it in an
isolated manner. In other words, most studies implicitly treat organizational
change as unidimensional in nature. Few studies consider the possibility that
organizational change may be multidimensional (Miller & Friesen, 1984).
Studying generic change or individual processes of change gives researchers 3
general idea of the phenomenon of change in organizations. However, in these
cases researchers may study different processes of change without
distinguishing them from one another. For example, consider the influence of
performance on organizational change: change has been found to vary
positively, negatively, and curvilinearly with performance (Baum, 1990;
Fombrun & Ginsberg, 1986; Ginsberg, 1988). Consequently, the research
findings in this area cannot be accumulated over time.

However, if change is considered in dimensional terms (e. g., take
direction as a dimension of change in any variable) two processes of change can
be differentiated: (a) convergent change or when an organization makes
changes in the same direction and (b) divergent change or when an organization
alters the direction of change. Using this perspective, it can be hypothesized

that high-performing organizations make more convergent change, whereas

low-performing organizations make more divergent change. Similarly,

diffen
drama
dimen.
unsucc
results.
lead to

l
of Chan,-
differen'
in organ
full unde
studied i]
pr 0Ccsses
Greenwm
Change ra
more dran

5

differentiation between incremental change and piecemeal change and between
dramatic change and pervasive change can be made by considering change in
dimensional terms. A researcher studying generic change in successful and
unsuccessful organizations will come up with insignificant or inconsistent
results. Therefore, distinguishing among dimensions makes more sense and can
lead to resolving inconsistencies in research on organizational change.

Further, the processes of change have been isolated from the dimensions
of change. Previous research has sought to compare the effectiveness of
different strategies of change to find universally successful strategies of change
in organizations (Miller & Friesen, 1982). However, if theorists are to gain a
full understanding of organizational evolution the processes of change must be
studied in relation to their contexts (Pettigrew, 1987). It is likely that different
processes of change grow out of different sources of change (I-Iinings &
Greenwood, 1988), in effect, a "switching mechanism" triggers one kind of
change rather than another. For example, low performance is likely to lead to
more dramatic and pervasive change but low related and convergent change.
Similarly, environmental dynamism is expected to influence degree of dramatic
change, but not other processes of change. Therefore, it is necessary: (a) to
identify different dimensions of change that correspond to different processes of
change in organizations, and distinguish them from one another and (b) to relate

these dimensions to their correlates so as to associate the processes of change to

their .-

an isol
proces
range t
range 13
CXpectel
diversit)
intemati
ln
Organizat
“1656 di r
organizan'
of Organjz
(b) envim,

envimnmel

their sources.

Rather than posing the questions of sources and processes of change in
an isolated manner, I develop and test models of change that relate different
processes of change with their sources. The purpose is to look for middle-
range theories and examine their boundary conditions. The models are middle-
range because they recognize multiple processes of change. However, they are
expected to be generic enough to apply to change in the following kinds of
diversity in organizations: change in product diversity and change in
international diversity.

In this study, I identify the following four dimensions of change in
organizations: increment, relatedness, pervasiveness, and direction of change.
These dimensions are then related to the following sources of change in
organizations: (a) organizational size and system coupling which reflect effects
of organizational inertia and history (Ginsberg & Buchholtz, 1990);

(b) environmental munificence and dynamism which manifest organizational
environments (Dess & Beard, 1984); (c) asset specificity, organizational risk,
and organizational resources (Hill & Hansen, 1991; Huff, Huff, & Thomas,
1989; Finkelstein & Hambrick, 1990), which signify strategic choice; and
(d) financial performance of the organization which represents organizational
performance and feedback.

This study is expected to contribute to a better understanding of the

 

phenon
literatu:
the basi
change

4, resea
Chapter

their iml

7

phenomena of organizational change and evolution in organizations. The
literature is reviewed in Chapter 2 to ground this study in sound research. On
the basis of extant theory in organization studies, a framework of organizational
change is proposed in Chapter 3, where hypotheses are also given. In Chapter
4, research methods used to test the proposed hypotheses are presented.
Chapter 5 offers results of the study. Finally, a summary of the findings and

their implications for future research are drawn in Chapter 6.

Chapter 2

Review of Literature

ln1

 

 

 

 

 

reviewed.
models of
understan.

this study

In this chapter, the literature on sources and processes of change is
reviewed. Also, how the sources and processes of change are integrated into
models of change in organizations is considered. The objectives are to
understand extant theory, to find gaps in literature, and to address these gaps in

this study.

THE SOURCES OF PERSISTENCE AND CHANGE
IN ORGANIZATIONS

Broadly, two perspectives dominate how organizational evolution has
been regarded in organizational studies and related disciplines. The first can be
described as organizational inertia, stability or persistence, and the second can
be termed organizational change and adaptation. These perspectives have
drawn their ideas from fields as diverse as biology and political science, but
neither one is a homogeneous stream of ideas, which may be due to their
diverse origins.

These perspectives are examined to explicate what they have and have
not contributed to our understanding of organizational change. The reviewed
literature has been chosen from the areas closely related to organization theory
and strategic management. The goals of this review were to understand how
the issues have been addressed in the past and to lay the basis for a more

systematic framework for the study of change in organizations.

9

 

Organic

Or
1986). T
survival.
selection
tradition
1977, 19'
organizat
fOundatic
and 0916,

Carroll,

organizat

10
Organizational Inertia, Stability and Persistence

Organizations often have been characterized as "organisms" (Morgan,
1986). This metaphor reflects that organizations have life cycles and need for
survival. Organizational births and deaths have been attributed to natural
selection and to notions such as "the survival of the fittest.” Studies in the
tradition of natural selection, or population ecology (Hannan and Freeman,
1977, 1984) provided the foundations for how this approach is applicable to
organizational studies. This research included studies of organizational
foundations and mortality of American labor unions (Hannan & Freeman, 1988)
and other populations of organizations. Other studies (e. g., Aldrich, 1979;
Carroll, 1988) extended this approach to more diverse populations of
organizations.

Other variants of this approach can be found in organizational
systematics (McKelvey, 1982) and in community ecology (Astley, 1984).

These approaches give preeminence to the forces of environment that are
thought to ”select out" organizational forms that are not well suited to it. It is
proposed in these studies that the forces of history dominate the organizations;
therefore, the organizations keep the same patterns of activities (Romanelli &
Tushman, 1986; Stinchcombe, 1965) that they had at the time of their founding.
Organizations, according to this view, are like species, have their life cycles;

the individual organizations cannot adapt to their environments; they either

 

survive o
successfu
The fit c:
enhanced
the unit I
relate to
time. A.
as well a
Se
haV'e dis:
lhe ecolc
inhibitinl
COrporat.
T.
entry, e)
1982). 1
diScjpum
incl-Cage
attache d
PIESCn-be

want to b

11

survive or die. If they die, other organizations replace them, and these new
successful organizations are expected to have a better fit with the environment.
The fit cannot be significantly enhanced by the organization’s activities; it is
enhanced only by the ecology of the organization’s founding. In these studies,
the unit of analysis is at the population level, and the changes that are discussed
relate to foundings, mortality, niche width, and so on, that take place over
time. Additionally, the influences of history and environments include internal
as well as external sources of inertia in organizations.

Several researchers (e. g., Ginsberg, 1988, Tushman & Romanelli, 1985)
have discussed the influences that lead to inertia in organizations. Apart from
the ecological approaches, other perspectives have also contributed to the
inhibiting role of various factors. These factors could be at the industry,
corporate, business unit, functional, or individual level in the organizations.

The theorists in the field of industrial organization economics analyze the
entry, exit and mobility barriers (Caves & Porter, 1977; Harrigan, 1981; Yip,
1982). According to some, industry concentration, market power, industry
discipline (Porter, 1980, 1985; Scherer, 1980) tend to keep the status-quo and
increase inertia at the strategic group level or industry level. There are costs
attached to all changes within or between industries. Some researchers
prescribe that the companies should follow a consistent strategy if they don’t

want to be stuck in the middle (Porter, 1980).

 

Ort
the impor
times of I
and predic
and effect
1958; Thc
inertial tei
& Dunkei
Additiona
c("l‘lDiexit
reSearch z
inferenCe
behavim.

To
in Strateg
& FrieSet
1985; HQ
their Env
consistEn
institufio‘

ZUCkEr’

12

Organization and management theory abounds in studies that emphasize
the importance of work rationalization, routinization and simplification from the
times of Taylor and Fayol. Such an emphasis calls for stability, consistency,
and predictability for coordination and control in the workplace for efficiency
and effectiveness of organizations (Cyert & March, 1963; March & Simon,
1958; Thompson, 1967; Watson, 1969). Other researchers have attributed
inertial tendencies in organizations to political and ideological structures (Clegg
& Dunkerley, 1980; Edwards, 1979; Marglin, 1974; Pettigrew, 1973).
Additionally, studies from Weber to the present point out that increasing
complexity in organizations decreases innovation and creativity. Streams of
research at Aston, and others (e. g., Blau & Schoenherr, 1971), also support the
inference that increasing bureaucratization tends to decrease entrepreneurial
behavior in organizations, and reinforces inertia and resistance to change.

To be successful, organizations should follow one kind of configuration
in strategy, structure, reward and control systems (Miles & Snow, 1978; Miller
& Friesen, 1980a). Contingency and consistency approaches (Donaldson,
1985; Hofer, 1975) focus on compatibilities of organizations within and outside
their environments. The economic and institutional environments demand
consistencies within organizations and between organizations and their
institutional infrastructure (DiMaggio & Powell, 1983; Meyer & Rowan, 1977;

Zucker, 1983). These consistencies lead to rigidities in organizations that

 

discourage
Greenwoo
committed
(Staw & R
reinforce c
thereby de

Ald
conditions
internal c.
inttrnal c
beCOmi n g
more hor
conditior
commitn

and re 3‘

13
discourage future changes (Ginsberg & Buchholtz, 1990; Hinings &

Greenwood, 1988). At the individual level, executives and managers become
committed to their decisions which results in an escalation of commitment
(Staw & Ross, 1987). All these processes develop their own life and can
reinforce earlier strategies, structures, and other patterns in organizations,
thereby decreasing the incentives or pressures for change (Huff & Huff, 1990).
Aldrich and Auster (1986) have discussed the internal and external
conditions that increase inertia or resistance to change. They enumerated four
internal conditions: the founders’ or entrepreneurs’ retention of control, higher
internal consistency for better coordination and control, power relations
becoming stronger that resist change, and organizational members becoming
more homogeneous in their assumptions and perceptions. The external
conditions that they focused on are: organizational and interorganizational
commitments to continue old practices; legal protections (e. g., patents, tariffs,
and regulations); dependencies on other organizations and their control; and
industry characteristics (Aldrich & Auster, 1986). Other sources of inertia may
include organizational myths, ideologies, or cultures (Clark, 1972; Mitroff &
Kilmann, 1976); organizational learning, and defense routines (Argyris, 1985,
1987). Furthermore, the leaders in the organization may not register
environmental changes that are inconsistent or threatening (Hedberg, Nystrom,

& Starbuck, 1976; Starbuck, Greve, & Hedberg, 1978).

 

All
organizatic
inertia. Tl
patterns, d
that are re
inconsister
case, man:
opportunit
organizati:
organizati
Harman (g

1979; Gla

Organiu

Th
are SeVer
liter-atUre
Or Eaniza.
SOurCeS (
Citojce V'

and “We .

14

All the factors mentioned above lead researchers to believe that
organizational activity patterns will persist over time due to organizational
inertia. Therefore, organizations will largely reflect their earlier activity
patterns, despite changes in the environment and management. In the systems
that are relatively more tightly coupled, changes in a subsystem could lead to
inconsistencies within a system resulting in organizational dysfunction. In this
case, management is likely to delay changes and may wait for a better
opportunity for change. The result is increased continuity and persistence in
organizations. Other factors that contribute to organizational inertia are
organizational size (Aldrich & Auster, 1986; Ginsberg & Buchholtz, 1990;
Hannan & Freeman, 1977; Huff et al., 1989) and system coupling (Aldrich,

1979; Glassman, 1973; Thompson, 1967; Weick, 1976).

Organizational Change and Adaptation

The second perspective is that of change. Within this perspective, there
are several approaches. Organization theory and strategic management
literatures are replete with theories that propose different sources of change in
organizations. Two distinct streams of ideas can be found regarding the
sources of organizational change: the environmental view, and the strategic
choice view. These are referred to as the weak-organization theory perspective

and the strong-organization theory perspective, respectively (Smircich &

 

Stubbart, i

or individu

 

l. The R0
The
organizatio
organizatio
Population
organizatio
an extreme
Similarly,
Studies PFC
i(16213 is th:
YESOUrces.
achieve,“e
This may
the ec0001

\

l 0t}
Lg
( M
dej
Ins
Po

15
Stubbart, 1985). All these forces pressure the organizations (Ginsberg, 1988)

or individuals to change (Huff & Huff, 1990).
l. The Role of Environment

The environment has long been considered a source of variance for
organizations. It is a source of strategic inertia. It can also be a source for
organizational change. This view can be drawn from many theories. The
population ecology view recognizes the influences of environment on
organizations (Aldrich, 1979). In fact, the ecological approaches may represent
an extreme form of environmental determinism (T ushman & Romanelli, 1985).
Similarly, the structural-functional systems, and other theories in organization
studies provide support for environmental influences.‘ The premise of these
ideas is that organizations are dependent on their environments for critical
resources. Because organizations are "rational instruments for goal
achievement," they try to adapt themselves according to their environments.
This may help the organizations to gain needed resources and legitimacy from
the economic, social, and institutional forces (DiMaggio & Powell, 1983) for

their survival and growth.

 

‘ Other theories include contingency theory (Lawrence &
Lorsch, 1967; Burns & Stalker, 1961), neo-contingency ideas
(Miles & Snow, 1978; Snow & Hrebiniak, 1980), resource
dependence approach (Pfeffer & Salancik, 1978), and
institutional factors (Meyer & Rowan, 1977; DiMaggio &
Powell, 1983).

 

Li

1971) ha.
opportuni
(1985, 19
power am
and institt
ofchange:
research b
(Mintzberg
Associates
discuss hm
lndu
Paradigm o

the market .

16

Likewise, the traditional model of strategic management (e.g., Andrews,
1971) has focused on how organizations scan their environments for
opportunities and threats so they can formulate strategies. Fligstein’s studies
(1985, 1987) on organizations that adopt multidivisional structures and shift of
power among various organizational functions historically attest to the economic
and institutional explanations of organization theory. In a like way, the studies
of changes at the Imperial Chemical Industries (ICI) (Pettigrew, 1985), and
research by Miller and Friesen (1980b, 1982), Mintzberg and Associates
(Mintzberg, 1978; Mintzberg & McHugh, 1985), and Nadler, Tushman, and
Associates (Nadler & Tushman, 1989, 1990; Tushman & Anderson, 1986) all
discuss how organizations adapt to their environments.

Industrial organization economics has been based on Mason-Bain’s
paradigm of market structure, conduct, and performance. Under this paradigm,
the market structure is a given. If rational strategic conduct is assumed, all
firms and their managements will analyze the objective environment of the
industry. Therefore, market structure was considered to determine
organizational performance.2 The vast literature in industrial organization

economics testifies to the importance of market structure. It also indicates that

 

2 Though more recently, there have been changes in the
paradigm whereby the influence of conduct on performance
is also taken into consideration. This point is considered in
detail in the following sections on strategic choice.

 

the interi
because (
competiti
on the m:
1980). T
A1:
in their (1
increasing
environml
change in
Tu
organizati

C€lrtnot ali

17

the interindustry and intraindustry structure do not remain static at all times
because of following influences: new entries, exits, conduct, the intensity of
competition, industry life cycles, and performance of the competing companies
on the market structure (Caves & Porter, 1977; Porter, 1980, 1985; Scherer,
1980). Thus, these industry trends demand change in organizations.

Also, studies of national and multinational corporations making changes
in their domestic and global strategies and structures because of the ever-
increasing international competition offer support for the major role that
environment plays. Together, all these studies lead to the conclusion that
change in an organization is a result of the environment.

Tushman and Romanelli (1985) suggested two ways in which
organizational environments may cause changes. First, if the organization
cannot align within itself and with the environment, this problem will lead to
continuous low performance. Therefore, the organization cannot sustain itself
without changing. Second, various spheres of the environment, for example,
economic (input, product, market, labor), social, legal, political, institutional,
and technological may undergo transformation, resulting in the misalignment of
organization-environment relationships. This misalignment will precipitate
changes in the organization so that it again can have a good fit with its
environment. The changes may take place according to the product life cycle,

business cycle, or in stochastic processes (Scherer, 1980).

2. The]
F

industria

 

The indu
paradigrr
firm and
structure.
Schendel.
the firms
Similar m;
This idea
an industr-
The
March, 19.
orgartizatio
ConseClUEfnt

the Obiecu'v

18
2. The Role of Strategic Choice

For a long time, researchers who used the Mason-Bain paradigm of
industrial organization economics, ignored the importance of a firm’s conduct.
The industry was considered to be homogeneous. However, in the revised
paradigm, theorists have given more importance to the conduct (strategy) of the
firm and how conduct may influence the firm’s performance and the market
structure. Studies of strategic groups in different industries (e.g., Hatten,
Schendel, & Cooper, 1978; Harrigan, 1985) have brought the point home. All
the firms in an industry do not behave or change in the same way, or even in a
similar manner. They follow different strategies for performance and change.
This idea leads to the critical question: What are the sources of variation within
an industry?

The theorists of the Carnegie School (March & Simon, 1958; Cyert &
March, 1963) pointed out that there may be "bounded rationality" in
organizations, particularly when ambiguity and uncertainty are high.
Consequently, in such cases, managers cannot know all there is to know about
the objective environment, and they can’t process all the information that is
available. Therefore, they make choices that are based on their selective
perception, cognition, retention, and decision making (March & Olsen, 1976).
There may be other constraints on managers’ "knowing the objective

environment”, that is, if it is actually ”out there”. The tradition of

anthro‘
Variou
1986) 1

color t1

 

and Sin

idea that

the idea:

together
managen
even enat
Such actit
SU‘ategy n
managemé
Significant.
The

(BourgeOis1

studies on t.
Chara“Cristi
have ad ded t
Friedland’ 1e

organizations

19

anthropological studies has always drawn attention to the "native views."
Various metaphors of organizations (e. g., cultures, psychic prisons) (Morgan,
1986) have shown that organizations may have unique characteristics that can
color the members’ perceptions of their environments. For instance, March
and Simon (1958) alluded to the role of "dominant management coalitions” an
idea that was developed by Cyert and March (1963). Child (1972) integrated
the ideas of dominant coalitions and their strategic choices, and he showed how
together they influence the survival or performance of the firms. In addition,
managers may act on the basis of their perceived environment, or they may
even enact their own environment (Smircich & Stubbart, 1985 ; Weick, 1979).
Such action may best be described in terms of Chaffee’s (1985) interpretive
strategy rather than adaptive strategy or linear strategy. Therefore,
management characteristics, perceptions, discretion, and choices are expected to
significantly influence the strategic change in organizations.

The influence of strategic choice has been well recognized by researchers
(Bourgeois, 1984; Miles & Snow, 1978; Schendel & Hofer, 1979). Many
studies on the role of boards of directors; executive leaders; and the
characteristics, orientations, tenure, discretion, and succession of such leaders
have added to the empirical evidence of this perspective. For example, Palmer,
Friedland, Jennings, and Powers (1987) reported that the ownership of

organizations by families and banks influence the organizations’ adoption of the

 

multidivis

economic

 

cxplanatior
organizati<

Hat

20

multidivisional (M-form) organization structure. The findings gave evidence of
economic and political explanations of organizational behavior. This
explanation is consistent with Fligstein’s (1985, 1987) results of testing various
organization theories that provide support for economic and power theories.
Hambrick and Mason (1984) conceptualized that organizations reflected
the choices of top management. They developed ”the upper-echelons
perspective," which pointed out the complexities and problems of studies that
showed the unimportance of chief executive officers (e.g., Lieberson &
O’Conner, 1972). They emphasized that background characteristics of top-
management teams and dominant coalitions partly determined the organization’s
strategic choices and performance outcomes. This perspective has been
supported conceptually (Hambrick & Finkelstein, 1987) and empirically (e. g.,
Chaganti & Sambharya, 1987; Finkelstein & Hambrick, 1990). Also,
Goodstein and Boeker (1991) analyzed the effects that changes in an
organization’s ownership, its board of directors, and its top management have
on the process of strategic change. Similarly, Nadler, Tushman, & Associates
focused on how executive leaders influence strategic changes in organizations
(Nadler & Tushman, 1988, 1990; Tushman, Newman, & Romanelli, 1986;
Virany & Tushman, 1986). Additionally, Day and Lord (1988) reviewed the
literature on influences of executive succession and argued that it can explain

20-45 percent of variance in organizational outcomes. Other researchers

21

analyzed managers’ orientation, specialization, and experience (Gupta &
Govindarajan, 1984; Govindarajan, 1989; Fligstein, 1985, 1987), and personal
characteristics (e.g., Miller, Droge, & Toulouse, 1988). They reported that
these attributes influence strategic choices of managers in organizations.

Research on strategic management in the last two decades illustrates that
strategic choices of managements are reflected in terms of several variables.
These variables are likely to be good candidates for influencing persistence and
change in organizations. For example, management commitment to current
strategy and investments in specific assets augment inertia in organizations
(Huff et al., 1989). Such a tendency inclines organizations to momentum
rather than change. Williamson (1975, 1985) has attached a great deal of
importance to asset specificity in influencing strategic decision-making by
executives in organizations. Similarly other researchers (e. g., Montgomery &
Wernerfelt, 1988; Wernerfelt, 1989) have studied links between nature of assets
and corporate strategies in organizations. It is argued that more specific assets
constrain managements to limited options whereas more general assets allow
managers to take advantage of a wide range of opportunities (Chatterjee &
Wernerfelt, 1991).

The level of organizational resources and slack in organizations has also
been receiving increased attention in strategic management (e. g., Huff et al.,

1989; Lundberg, 1984; Sharfman, Wolf, Chase, & T ansik, 1988; Singh, 1986).

 

A high 1e
whereas it
making ch
organizath
strategic o
Anc

deals with
organizatio
liSk and re
Set/era] res
1982, 1984
have COntri‘
differ”! sti
1n Sh
Strategies, 5
1962; Child
resourcCS, L
organizafim

Changes in (

22

A high level of organizational slack decreases the motivation for change,
whereas it makes change more feasible by providing resources needed for
making change possible (Ginsberg & Buchholtz, 1990). Therefore,
organizational resources are likely to play a critical and a complex role in
strategic organizational changes.

Another very interesting stream of research in strategic management
deals with the relationships of risk-taking, strategies, and performance in
organizations. Since Bowman’s study (1980) of negative relationship between
risk and return, there has been a considerable debate over these relationships.
Several researchers (e.g., Bettis, 1981, 1983; Bettis & Hall, 1982; Bowman,
1982, 1984; Fiegenbaum & Thomas, 1988; Hoskisson, Hitt, & Hill, 1991)
have contributed to the understanding that different levels of risk lead to
different strategies of change in organizations.

In short, the dominant coalitions make strategic choices regarding
strategies, structures, and processes that their organizations follow (Chandler,
1962; Child, 1972). These choices influence the deployment of organizational
resources, use of different types of assets, and decisions about risk taking in
organizations. In turn, these choices and decisions result in different kinds of

changes in organizations (Huff et al., 1989).

 

Organi
I
strategic
in orgar
perform
(Mintzb
High- ar
Strategic
perform;
Pancms
Tl
0f Slack .
have hi g1
t1‘9"] to a
high‘PCrf
DCrcepfiO
I)Utton, 1
low Perfo
in the Tag
organllatii

0r -
ganlzam

23

Organizational Performance

Apart from the influences of origin and history, environment, and
strategic choice, another variable that is also likely to exert influence on change
in organizations is the organizational performance. Most of the studies treat
performance as a dependent variable but its influences are equally important
(Mintzberg, Raisinghani, & Theoret, 1976; Moch, Buchko, & Rubin, 1988).
High- and low-performance organizations may have different influences on
strategic change (Ginsberg, 1988). The above is particularly true when
performance is extremely good or bad (Baum, 1990). The results are different
patterns of organizational evolution.

The influence of organizational performance is also related to the issue
of slack (Sharfman et al., 1988). High-performing organizations are likely to
have higher organizational slack (March & Simon, 1958), which may allow
them to adapt and change (Romanelli & Tushman, 1986). At the same time,
high-performing organizations may protect themselves either to preclude
perception of external change or any need of adaptation (Staw, Sandelands, &
Dutton, 1981). The opposite may be true of organizations that have sustained
low performance. In these cases, management may not be able to defend itself
in the face of pressures and incentives for change. At the same time, such
organizations may not have enough resources to bring about change. Thus, the

organization’s level of performance may be a critical variable in defining the

 

interplay
resources
legitimac
significant
(Ginsberg,
decrease,
Ginsberg,
Fro
exhibited p1
come frOm
environmen
phenomEnO,
organization
in the Same
to One kind
changes in 0
The S
Categofies 01
environment
Sources of C}

in .
Orgaanat‘“

24

interplay of forces of persistence and change. Higher performance may provide
resources for change while lower performance may render the basis for
legitimacy for change in organizations. Previous studies have reported
significant influences of performance; however, the findings lack consistency
(Ginsberg, 1988). That is, organizational performance was found to increase,
decrease, and have a curvilinear effect on change (Baum, 1990; Fombrun &
Ginsberg, 1986).

From the studies reviewed above, it is inferred that organizations
exhibited pressures for inertia as well as for change. These pressures may
come from enduring and changing aspects of the organizations and their
environments. Organizational change should not be regarded as a monocausal
phenomenon. That is, multiple influences may lead to change in different
organizations at different times, in the same organization at different times, or
in the same organization at the same time. In addition, some sources may lead
to one kind of change, whereas other sources may result in other kinds of
changes in organizations.

The studies reviewed above showed that there were at least four broad
categories of factors that contributed to the change in organizations: history,
environment, strategic choice, and performance of an organization. These
sources of change may make varied contributions to different kinds of changes

in organizations. Not only are there varied influences on the change in

25

different organizations, but these organizations may also change in different
ways. An organization may change in different ways during different time
periods, or even in different ways during the same time period. Therefore, the
process of change can be characterized in several diverse ways that may give

rise to different patterns of change in organizations.

THE PROCESS OF CHANGE

Broadly, there are two processes of change in organizations: the
evolutionary and the revolutionary. The evolutionary models draw from the
adaptation literature in strategic and organization theory, industrial organization
economics (Nelson & Winter, 1982), and the contingency approaches cited
previously. In the field of economics, these two processes are reflected in
Marshallian and Shumpeterian analyses. Within Marshallian analysis, economic
evolution takes place in small increments that are gradual, smooth, uniform,
and predictable. In Shumpeterian theory, economic (r)evolution is considered
to be spontaneous, discontinuous, and disharmonious (Awan, 1986). Similarly,
in the organization development literature, different types of organizational

changes have been conceptualized.3

 

3 For example, Golembiewski, Billingsley, and Yeager (1976)
distinguished among alpha, beta, and gamma change in
organizations. Similarly, Argyris and Schon (1978)
conceptualized single loop and double loop learning in

 

h
in orgar
SU‘UCIUI‘I
dramatic
evolutio
changes.
organiza
Was mor

T1
logical (t
(Cyert (g-
is Consist

(Aldrich.
fiiShion,

and Can’t
reSearche.
Miller, TS
equilibrim

\

Org;
OTdC
198‘,
F 1.Sc;

26
Miller and Friesen (1984) analyzed quantum and revolutionary changes

in organizations. Their concept of revolutionary change in organizational
structures described quantum change (as compared with piecemeal change) and
dramatic change (as contrasted with incremental change). Their concept of
evolutionary change characterized the piecemeal and incremental organizational
changes. They found evidence of quantum and revolutionary change in the
organizations they studied. Further, they contended that this type of change
was more successful in the organizations studied.

These models are not simply descriptive; they also have noted that
logical (or disjointed) incrementalism is a prescription for organizational change
(Cyert & March, 1963; Hedberg et al., 1976; Lindblom, 1968). Such an idea
is consistent with the view of organizations as loosely-coupled systems
(Aldrich, 1979; Weick, 1979), where the subunits can change in a disjointed
fashion. However, some variables in an organization may be tightly coupled
and can’t change incrementally (Miller & Friesen, 1984). Slowly and steadily,
researchers have justified metamorphic changes (Hinings & Greenwood, 1988;
Miller, 1982; Mintzberg, 1978; Pettigrew, 1985). An emerging non-

equilibrium theory can help us understand the breakdown of existing structures

 

organizations. Changes may also be characterized as of first
order, second order, and third order (Bartunek & Moch,
1987; Moch & Bartunek, 1990; Watzlawick, Weakland, &
Fisch, 1974).

andthe

1989).

SOUI'CCS

 

The inte
manager
& Dunk.
quuant
organiza
Mintzber
1988). l
per SPCCtii

.ii-TSITI‘KxuiI

27

and the generation of new ones in the form of discontinuous changes (Kiel,
1989).

These changes may be due to external and internal sources. The external
sources include changed environments, technology, and actions of competitors.
The internal circumstances encompass political tensions, power conflicts,
management ideologies, or other forms of strategic choice (Child, 1972; Clegg
& Dunkerley, 1980; D02 & Prahalad, 1987; Pettigrew, 1973, 1987). The ideas
of quantum change are similar to the holistic and gestalt notions of
organizations that are referred to as archetypes (Hambrick, 1984; Miller, 1981;
Mintzberg, 1973, 1978) and organizational tracks (Hinings & Greenwood,
1988). In fact, Miller (1987) has argued that research from the contingency
perspective, particularly multivariate linkages, represent configurations and
justification for quantum change.

Apart from the incremental versus dramatic, and piecemeal versus
pervasive divisions, another dimension of organizational change has been
studied. It is referred to as convergence versus reorientation (T ushman &
Romanelli, 1985), continuity versus discontinuity (Pettigrew, 1985), or
momentum versus reversal (Miller & Friesen, 1984). Accordingly, Miller and
Friesen (1984) analyzed 21 strategic and structural variables in order to
understand organizational adaptation. They contrasted momentum (continuity)

and reversals within these organizations and concluded that momentum was a

 

dominant

Sti
intematio
(Montgon
voluminor
unrelated t
Varadarajz
organizatic
remains in
When IBM
Segment of
Out of its 1)
reTimid and
(Geringer,

1989)_ F0,

28

dominant phenomena, and reversals were relatively rare in organizations.

Still another dimension has been studied in the diversification and
internationalization of organizations. It is the relatedness of change
(Montgomery, 1985; Palepu, 1985; Rumelt, 1974; Vachani, 1991). There is a
voluminous literature on the antecedents and consequences of related and
unrelated diversification (Bettis, 1981; Hill & Hansen, 1991; Ramanujam &
Varadarajan, 1989). In the strategic management literature, when an
organization moves out of its industry segment to another industry segment but
remains in its broad industry, this is called related change. An example is
when IBM which was in mainframe computers entered personal computer
segment of the industry. An unrelated change is when an organization moves
out of its broad industry, e.g., from computers to chemicals. The studies of
related and unrelated change are being extended to the international arena
(Geringer, Beamish, & daCosta, 1989; Kim, 1989; Kim, Hwang, & Burgers,
1989). For example, organizations moving to other countries in the same
geographical region (e. g., continent) are involved in related change, whereas
those moving to a new region are making an unrelated change (Vachani, 1991).

Therefore, the different models of the process of organizational change
involve at least the following four processes of change: incremental versus
dramatic change, piecemeal versus pervasive change, convergent versus

divergent change, and related versus unrelated change.

 

T1
altemativ
Finkelste
gathered
Greiner 8
More rece
Hambrick
strategic c
together .
t0 reconcili
View has it.

Add
environme
2" 2 mam
organizatio
influenCes 0
their man.“
organizatio”;
WETStandin

argued f0r di

29
INTEGRATING THE PERSPECTIVES

The champions of these differing perspectives posed various theories as
alternatives to the dominant theory of the time, at least initially (Hambrick &
Finkelstein, 1987). The followers of each of the streams of research have
gathered theoretical and empirical support for these theories (Boeker, 1989;
Greiner & Bhambri, 1989; Hinings & Greenwood, 1988; Johnson, 1988).

More recently, studies have included multiple perspectives. For example,
Hambrick and Finkelstein (1987, p.370) discussed about population ecology and
strategic choice, and observed, "(T)hese two camps seem to be moving close
together . . . fortunately both theoretical groups have begun to invest energies
to reconcile the two views and to determine the circumstances under which each
view has its greatest explanatory power."

Additionally, Hrebiniak and Joyce (1985) combined the degree of
environmental determinism with the degree of strategic choice and suggested a
2 x 2 matrix that represented four mutually exclusive types of strategic and
organizational changes. Romanelli and Tushman (1986) tried to delineate the
influences of inertia, environment, and strategic choice to empirically find out
their relative contributions. Zammuto (1988) drew a correspondence between
organizational ecology and strategic choice that can help further our
understanding of organizational adaptation in different industries. Baum (1990)

argued for different models of inertia, one of which (i.e., the "fluidity of

30
aging” hypothesis) (Singh, Tucker, & Meinhard, 1988), he asserted is

consistent with the model of quantum changes.

Alternatively, theorists can use the distinctions of level (micro vs. macro)
or other dimensions (e. g., time) to understand the role of different sources and
processes of change (Hinings & Greenwood, 1988; Pettigrew, 1985, 1987).
Greiner (1972) proposed a model of organizational adaptation involving growth
phases interrupted by crisis after each period of growth. Other researchers
(e.g., Miller & Friesen, 1982; Mintzberg, 1978; Mintzberg & McHugh, 1985;
Mintzberg & Waters, 1982) proposed similar ideas regarding incremental and
quantum changes, and continuity and reversals over time in organizations.

The above-mentioned studies revealed that, in general, revolutionary
changes or reversals punctuated the evolutionary changes in organizations.
These studies have focused on dramatic and pervasive changes. However, the
authors do acknowledge the possibilities of other kinds of changes in
organizations (piecemeal incremental changes, pervasive incremental changes,
and so on). Tushman and Romanelli (1985) posed the ideas of convergence
and reorientation and further explicated the punctuated equilibrium model of
strategic and organizational change. Pettigrew (1985, 1987) made a historical
study of organizational changes at Imperial Chemical Industries (ICI) and
validated the punctuated equilibrium model.

The punctuated equilibrium model has been further developed by Gersick

 

(1991) ir
well as 0
deal with
role of st
managem
ecological

Johnson (

political. t
model of r

Organizatit

 

researchers

SCV‘L
involve dift

(1989) disc

  
 
  
 

transilional.

 

leadership 11
R0"lanelli, 1
and Tughma
and anfiCipar.

Chang“, nan-

31
(1991) into a more general framework that may apply to individual behavior as

well as organizational phenomena. The model uses the dimension of time to
deal with stability and change, internal and external sources of change, and the
role of structure and human action. It also conceptualizes the role of executive
management during different time-periods. This model tries to reconcile the
ecological, convergence and reorientation approaches to organizational change.
Johnson (1988) proposed an ’organizational action’ view that combines
political, cognitive, and cultural aspects of management in an incremental
model of change. Similarly Schwenk ( 1989) integrated the cognitive,
organizational, and political factors and offered a framework that can help
researchers in studying strategic change.

Several other studies showed that the organizational evolution may
involve different kinds of changes in organizations. Kleiner and Corrigan
(1989) discussed three types of organizational changes: developmental,
transitional, and transformational. Tushman, Nadler, and Associates discussed
the role of charismatic and institutional, internal/old, and external/new
leadership in frame-bending and frame-breaking changes (T ushman, Newman,
Romanelli, 1986; Nadler & Tushman, 1989; Virany & Tushman, 1986). Nadler
and Tushman (1990) used the dimensions of incremental versus strategic change
and anticipatory versus reactive change to view four types of organizational

changes, namely, tuning, adaptation, reorientation and recreation, in order to

 

understa
Goes (1‘.
mode of
(organiz.
adaptatic
revolutio
incremer
change a
organizat
lives of i
the Puzzli
Th
CondUCt e
per SPCCtis
and GFCer

these idea;

32

understand strategic change in organizations. Similarly, Meyer, Brooks, and
Goes (1990) attempted to resolve the inconsistencies by combining the primary
mode of change (continuous or discontinuous) with the level of change
(organization or industry). This combination yielded four types of change:
adaptation and metamorphosis at the organizational level and evolution and
revolution at industry level. Similarly, Dunphy and Stace (1988) juxtaposed
incremental and large-scale organizational transformation with participatory
change and coercive change; the result was different types of changes in
organizations. A broader framework encompassing various dimensions and
types of changes will help the researchers take stock of the missing pieces of
the puzzle and understand the phenomena of change in organizations.

There have been a number of attempts by organizational researchers to
conduct empirical tests regarding the integration of various theoretical
perspectives. The studies by Mintzberg; Miller and Friesen; Pettigrew; Hinings
and Greenwood referred to previously all point to the validity of integrating
these ideas. A synthesis of various approaches provides a better explanation of
strategic and organizational change (Fligstein, 1985, 1987; Palmer et al.,
1987). Additionally, Ginsberg and Buchholtz (1990) studied corporate
responsiveness to radical change in the health care industry. They revealed that
various perspectives (inertia, adaptive, and institutional) were complementary

rather than competitive in explaining organizational transformation.

 

 

Ham
discretion 1
choice. F11
managemer
and strateg
make stratc'
discretion
& Hambri
and strate}
eVOiUIlQn
COlleaguC-

bem'een C

33
Hambrick and Finkelstein (1987) developed the concept of managerial

discretion to bridge the gap between environmental determinism and strategic
choice. Finkelstein & Hambrick (1990) reported a moderating role of
management discretion on the relationship between top-management’s tenure
and strategic persistence in organizations. High management discretion may
make strategic choice a powerful explanation, whereas a low management
discretion may indicate that strategic choice is a weak explanation (Finkelstein
& Hambrick, 1990). Researchers are now realizing that ”inertia, environment,
and strategic choice may interactively determine courses of organizational
evolution over time" (Romanelli & Tushman, 1986, p. 618). Palmer and
Colleagues (1987) also emphasized the importance of studying the interplay
between different dimensions of organizational behavior. Similarly, Boeker and
Goodstein ( 1991) studied the moderating role of performance between
environment and organizational change. Such analyses can act as bridges to
link various middle-range theories. Hence, it is important to not only study the
main effects of these causes but their interactions as well.

Bedeian (1990) argued that simple models do not capture the complex
parallel processing and bi-directional causality of organizational processes, and
proposed a model of continuous multidirectional interaction of environment and
strategic choice for organizational adaptation. Similarly, Mascarenhas (1989)

argued that combining the population ecology, environmental adaptation, and

34

strategic choice perspectives contributes more to our understanding of
organizational dynamics than these perspectives considered in isolation.

The foregoing discussion demonstrates that researchers are considering
multiple sources of change in organizations, but not the processes of change.
Regarding the processes of change, most of the research work has been based

on disparate research perspectives (Huff et al., 1989).

GAPS IN EXISTING LITERATURE

The earlier studies tried to combine different perspectives of
organizational change. But, despite the conceptual initiatives in this direction,
there was a paucity of empirical research. Most frequently, organizational
change was studied in terms of its magnitude (i.e., more change or less
change). Different processes of change were studied in an isolated manner.
Organizational change was implicitly treated as unidimensional. The
undifferentiated study of change can lead to inconsistencies in research findings.
For example, in the study of performance on change, change was found to have
a positive, negative, and curvilinear effect on change (Baum, 1990; Fombrun &
Ginsberg, 1986; Ginsberg, 1988).

Studies of various types of organizational change have not been
accumulated into a coherent body of knowledge (Ginsberg, 1988). Instead, the

studies have resulted into confusion over the use of terms to describe different

35

processes of change (e. g., piecemeal, concerted, incremental, dramatic,
quantum, revolutionary, and evolutionary, convergence, divergence,
reorientation, continuity, discontinuity, and reversals) (Miller & Friesen,
1980b, 1984). In the literature, two broad models of change are recognized:
the evolutionary and revolutionary. A large variety of processes have been
aggregated into these models without making distinctions among various
processes of change. There is a need to distinguish among different processes
of change and relate them to their correlates that are sources of change (Miller
& Friesen, 1982). The sources and processes of change are isolated from each
other for the purpose of study only. They need to be related to understand an
answer to the question: What determines what kinds of organizational changes
take place under conditions both within and outside of organizations over time
(Hinings & Greenwood, 1988; Pettigrew, 1985, 1987; Van de Ven & Poole,
1988). Such an integration of the sources and processes of change can help
researchers to arrive at middle-range theories of change in organizations.

This study is an attempt to address these points to fill the gaps in the
literature and to recognize the multidimensionality of change. Accordingly, the

conceptual framework and hypotheses are developed in the next chapter.

Chapter 3

Conceptual Framework and Hypotheses

From the preceding review of literature, one can conclude that
organizations change in a variety of ways, that is, through incremental and
dramatic change, related and unrelated change, piecemeal and pervasive
change, and convergent and divergent change. The process of change is
influenced by the internal and external factors of the organization (Pettigrew,
1985, 1987) that can be called sources of organizational change. These
correlates of change can be broadly categorized in terms of organizational

history, environment, strategic choice, and performance.

THE PROCESS OF CHANGE

The process of change can be characterized at different levels of
abstraction. One level is found in the processual framework presented by
Bartunek (1984), Ginsberg (1988), and Huff and Huff (1990). This type of
analysis tries to capture the dynamics of the process of change in organizations.
The studies at this level view change as resulting from two kinds of forces: the
pressures for inertia and the stresses for change. These forces interact and lead
to diverse kinds of change in organizations. Researchers argue that the
pressures for persistence and change may come from the enduring and changing
aspects of various factors at different levels of the organization and its
environment. Change for these analyses may be at individual, organization,

industry, or societal level. The forces for change may have their origins in the

37

38

economic and institutional environment, in organizational characteristics, in
objective assessments and perceptions of the managements, and in
organizational performance. This is a more abstract level of analysis that
studies the underlying process of interplay among different forces influencing
change.

The second level of abstraction of the change process involves various
archetypes of change. This analysis is more concrete than the previous one. It
can be viewed in the holistic frameworks developed by Hinings and Greenwood
(1988) and Laughlin (1991). These models have been described in terms of
various organizational tracks: inertia, reorientation, aborted excursions, and
unresolved excursions (Hinings & Greenwood, 1988). Different tracks involve
various sequences of decoupling and recoupling in organizations.

The third level of analysis involves different ways of characterizing
organizational change using attributes of the process of change. For example,
researchers have studied incremental, dramatic, convergent, and divergent
changes in organizations. This level can be seen in the studies by Mintzberg
and Associates (Mintzberg & McHugh, 1985; Mintzberg & Waters, 1982),
Miller and Friesen (1982), Pettigrew (1985), and Tushman and Associates

(T ushman & Romanelli, 1985; Nadler & Tushman, 1990).‘ This is the level

 

‘ Other studies at this level include Greenwood (1984), Johnson
(1987), Lundberg (1984), Meyer et al. (1990).

39

of the present study. Of interest is a model that can be applied to changes at
the organizational level. The two variables of interest in this study are change
in product and international diversity in organizations.2

From the review of literature, four attributes were identified that can be
used to characterize the process of organizational change in product and
international diversity of organizations. The use of different names for these
attributes has caused much confusion and has concealed the distinctions among
various attributes. Therefore, it is important to clarify the labels and the ideas
underlying each attribute.

In this study, the following attributes of the process of organizational
change are considered: incremental versus dramatic change, related versus
unrelated change, piecemeal versus pervasive change, and convergent versus
divergent change. The criterion used to select the attributes was two-fold: (i)
relevance of the attribute to the two types of changes analyzed in this study,
and (ii) role of the attribute in distinguishing the pattern of change in diversity
of organizations. The four attributes selected in the study were relevant to the
types of changes studied, and they helped distinguish the pattern of change in

organizations. However, there may be other attributes of change that may be

 

2 The phrases international diversity, global diversity, and
geographic diversity are used interchangeably in this study.

4o

pertinent to these or other changes in organizations.3 These dimensions are
expected to apply to similar organizational changes such as transforming
strategy and structure. However, these dimensions may not be entirely
independent of the content of change. Thus, there may be some dimensions
that are more critical to one kind of change than to other kinds of changes.
DIMENSIONS OF ORGANIZATIONAL CHANGE

Increment of change (incremental vs. dramatic change). Incremental-
dramatic change refers to the size of the step of change. In other words, it is
the extent of incrementalism in a process of change. The net change in product
and global diversity, or organizational structure (e.g., adoption of M-form
structure) may come in small increments or in big leaps (Hoskisson &

Galbraith, 1985; Johnson, 1988). These are illustrated in Figure l.

 

3 The purpose of choosing these dimensions is primarily to
demonstrate the multiplicity of dimensions of change and
build models of organizational change. Other attributes
somewhat related to the phenomena of change in organizations
found in the literature include anticipatory versus reactive
change (Nadler & Tushman, 1990); deliberate versus
emergent change (Mintzberg & Waters, 1982); and first-,
second-, or third-order change (Watzlawick et al., 1974;
Bartunek & Moch, 1987; Moch & Bartunek, 1990). All of
these may not be regarded as distinct dimensions of the
process of change.

3003 800230.. 30:..— Omht<o<

 

WI; . “Irv

 

 

 

 

 

 

 

 

 

 

41

 

 

 

 

”8‘89“
08(2081

 

 

 

 

 

 

 

 

 

 

 

Hag-(:0 0.h¢3<¢6 m02<zo J<h2m3u2n¥=

mug—.0 qua—5.5.3.5582:

_ 2.3....

 

 

42
Relatedness of change (related vs. unrelated change). Related-unrelated

change refers to the nature of the activities that the organization is changing. It
relates to the relationship of the activities being considered and the other
activities of the organization. Related versus unrelated change may describe
changes in products, markets, technology, innovations, and so forth. An
organization changing a technology that is somehow related to its other
technologies is said to be involved in a related change (see Figure 2).
Similarly, if the activity being changed has no relationship with the other
activities, it is called an unrelated change. These changes have been clearly
understood in the literature on related and unrelated diversification (Rumelt,
1974; Palepu, 1985). Some ideas of related and unrelated internationalization
also have been proposed (Vachani, 1991). Changing product diversity in
related products (e. g., within industries) or in related countries (within a
homogeneous market) may be regarded as related change. Investment and
divestiture in unrelated products (across industries) or in unrelated markets may
be considered unrelated change.

Pervasiveness of change (piecemeal vs. pervasive change). Piecemeal-
pervasive change refers to the scope of change in the organization (i.e., how
much a part of the organization is involved in change). In other words, it
describes the simultaneity of organizational changes in different variables. The

change in an organization may be brought about in one variable at one time, or

43

 

 

 

mOO0 0.0 :0..- Q 540.....0 34(30

    

NOO0 0.0 :03 N 5403.0 0..

i 08 :0 992.582.

m02<20 by: K

 

 

99.2.0 1828:36823—

n 2.3....

 

 

 

44

in all variables simultaneously (see Figure 3). If a company changes its
strategy, structure, and control systems all at the same time, this is a
simultaneous change that is termed pervasive change. If the changes are made
in one variable at a point of time (i.e., changes in strategy occur when the
structure and control systems remain unchanged) it represents a piecemeal
change. A high correlation over time between the changes in variables will
suggest a more pervasive nature of change; similarly, a low correlation will
represent change that is more piecemeal in nature.

Direction of change (convergent vs. divergent change). Convergent-divergent
change refers to the direction of change. If an organization continues to make
changes in the direction of its earlier posture, this is convergent change (see
Figure 4). For example, an organization that has been globalizing keeps on
increasing its geographic diversity is making convergent change. Its attempts to
divest international operations will be considered divergent change. In an
organization with the history of growth, an increase is convergent change.
Similarly, in the same organization, downsizing exemplifies divergent change.
Organizational expansion and contraction can be called convergent and
divergent changes, respectively, depending on an organization’s previous
pattern of change. Different kinds of convergent and divergent changes are

receiving increasing attention in organizational research (Goodstein & Boeker,

1991).

 

45

 

 

 

 

wa; m3: NSF 5! ml... < h‘
080.0.)3 >8<3 98.08<:0 20.0.)... WZO 9.202(10

 

 

 

 

 

 

 

w02<20 Dmh¢u0200 m02<10 4<w2m0mz

09.2.0 ozmatuA—Aaefiouofi
n 2:3...

 

 

46

 

 

OOO —

QCOp . 0¢<m> 2. “3....

Chop

 

 

w92<2 P2 O¢U>200

   

~02<k0 hzuocmza

 

w02(x0 h! 0¢m>800

 

D->|H¢O-l->

 

 

yuan—.0 .:ou._o>§..:ow..c>=e0

v 2:3...—

 

 

47
THE DINIENSIONS OF CHANGE AND THEIR CORRELATES

Pettigrew (1985, 1987) argued in favor of analyzing the processes of
change in relation to their context in a historical perspective. Studies by
Mintzberg and Associates (e. g., Mintzberg & Waters, 1982), Hinings and
Greenwood (1988), Lundberg (1984), and Laughlin (1991) provide illustrations
of the contextual approach to organizational change.

There are multiple processes of change in organizations. Certain
processes of change are more likely to happen in some contexts and not in
others. For example, quantum change is more likely to occur in tightly coupled
systems rather than loosely coupled systems (Miller & Friesen, 1984).
Therefore, change processes are likely to be related to the content and the
context of change.

Analyzing processes of change and sources of change in isolation
artificially compartmentalizes the reality of change in organizations. It also
makes it difficult to make sense of different parts of the "elephant” of
organizational change, and gives rise to inconsistency of research findings.
The multiple processes of change in organizations depend upon the
characteristics of the organization and the environment. In other words,
different dimensions are expected to have different correlates, or different
patterns of association with their correlates. In essence, each dimension is

likely to have its own model. There is a need to develop models of change in

48

organizations (Miller & Friesen, 1982). Therefore, the interest here is in
middle-range theories, and their boundary conditions, for different dimensions
of organizational change. However, it is proposed that the models of the
dimensions presented here will be generic enough to apply across both changes:
the change in product and international diversity of organizations.

In the following section, hypotheses are developed relating the
dimensions of change to their correlates. The hypotheses will relate to product
and international diversity of organizations. There is a large body of literature
on product and international diversity in organizations, and more recently, the
literatures on change in product and international diversity have been growing
along parallel lines. In fact, some researchers have started integrating different
streams of research in the areas of diversification and globalization (e. g.,
Geringer et al., 1989; Kim, 1989; Grant, Jammine, & Thomas, 1988; Kim et

al., 1989; Vachani, 1991).

Organizational Size

Size is a frequently studied but controversial variable (Gooding &
Wagner, 1985) that may be a source of inertia as well as change. Therefore, it
may have negative as well as positive influence on the ability of organizations
to adapt and change. A stream of ideas from organizational studies asserts that

larger organizations have more difficulty implementing changes (Aldrich &

49
Auster, 1986; Ginsberg & Buchholtz, 1990; Harman & Freeman, 1977; Huff et

al., 1989). This is because larger organizations are more formalized (Blau &
Schoenherr, 1971; Scott, 1987), and thus managers may not notice the need for
change (Ginsberg & Buchholtz, 1990). Large organizations are also less
efficient at least in relative performance (output/input) terms (Gooding &
Wagner, 1985). Large organizations have more complicated structures;
therefore, it is harder for them to change (Huff et al., 1989). These
organizations also are likely to have bigger, powerful, and diverse groups of
stakeholders who are entrenched in the organization (Freeman, 1984). Such a
high amount of power and politics may make it more difficult for organizational
members to build a consensus to change the organization drastically.

Other ideas provide a contrast to the view mentioned above. These
particularly come from the economists who have emphasized economies of
scale (Porter, 1980; Scherer, 1980). Larger organizations have many resources
that may help them to make changes and take risks (Kimberly & Evanisko,
1981; Moch & Morse, 1977). Therefore, they may be in a better position to
pass through transition states (T ushman & Romanelli, 1985).

One can also visualize a threshold for organization (i.e., a minimum
efficient scale up to which there are economies of scale). There may be
diseconomies of scale after an organization has reached a certain level of size

and complexity. Therefore, there is liability of bigness and liability of

50
smallness (Aldrich & Auster, 1986). These liabilities may yield a curvilinear
relationship between size and organizational change.

It is argued that smaller organizations have limited resources to make
dramatic, pervasive, or divergent changes. Bigger organizations may have
developed too much inertia to undertake such changes. However, the
intermediate-sized organization is expected to have access to resources. Such
an organization is probably not too complex to have inertia, and can make more
dramatic, related, pervasive and divergent changes. The relationship of
organizational size and different dimensions of change is hypothesized as
follows:

Hypothesis. Small and big organizations change in a less dramatic

fashion, whereas medium-sized organizations change in a more

dramatic fashion.

Hypothesis. Small and big organizations make less related

change, whereas medium-sized organizations make more related

change.

Hypothesis. Small and big organizations make less pervasive

change, whereas medium-sized organizations make more pervasive

change.

Hypothesis. Small and big organizations have a higher probability

of changing in a convergent fashion, whereas medium-sized

organizations have a lower probability of changing in a convergent

fashion.

New inertial theories propose that inertial tendencies change in a complex

nonmonotonic relationship during the life cycle of an organization (Baum,

51

1990). Therefore, it is expected that even a cubic relationship may exist
between the size and some dimensions of change in organizations. Because
there is not enough theory to support the cubic relationship a priori, a quadratic
relationship is hypothesized, and a cubic relationship will be explored.
Exploring a cubic relationship may be particularly useful to explain if the

results are in contrast to the hypothesized effect.

System Coupling

Interdependence of organizational units influences the degree of strategic
inertia or change (Aldrich, 1979; Lundberg, 1984; Perrow, 1986; Weick,
1976). Systems that are more tightly structured have higher interdependence
and need higher coordination and control that leads to higher inertia (Ginsberg
& Buchholtz, 1990). In such cases, piecemeal changes are likely to be more
disruptive. The organization therefore has an incentive to make changes in
many related variables relatively simultaneously. Systems that are loosely
coupled will be more autonomous and can afford to change without
synchronizing changes in other subsystems. Miller and Friesen (1984)
acknowledged that organizations can change in other than a quantum or
revolutionary fashion. That is, variables that are loosely coupled may allow
organizations to change in a piecemeal fashion. Thus, system coupling is a

well-recognized variable that is likely to help explain pervasive change in

52

organizations. Its relationship with pervasive change is hypothesized as
follows:

Hypothesis. The tighter the system coupling in an organization,
the more pervasive the organizational change.

The streams of research reviewed in Chapter 2 show organizations’
dependence on their environments for critical resources. The critical role of
environments on organizations is well documented in several areas of
organization and management studies.

In the literature on organizational change, Pettigrew (1985, 1987)
emphasized the contextual approach to research. In their models of
organizational change, Lundberg (1984), Hinings & Greenwood (1988),
Mintzberg & McHugh (1985), & Tushman & Romanelli (1985) identified a
well-deserved role for the organizational context. Organizations try to align
themselves with their environments for better performance and achievement of
other goals. In doing so, they may undertake different types of changes in the
organization that may include divergent and unrelated changes. Further,
environments can stimulate organizational change when environments
themselves change.

Organizational environments can be characterized in terms of

munificence and dynamism (Aldrich, 1979; Dess & Beard, 1984; Staw &

53

Szwajkowsky, 1975). These characteristics of environments are likely to

influence different dimensions of change in organizations.

Environmental Munificence

Environmental munificence refers to the capacity of the environment to
support products (Castrogiovanni, 1991; Dess & Beard, 1984). In a munificent
environment, there is more likelihood of convergent change. For example,
organizations tend to diversify into product and market areas that provide
munificent environments. Organizations are likely to divest themselves of the
products and markets that have low capacity to support them, resulting in
divergent change.

An organization is also likely to remain in its own segment of the
industry and grow in that segment when environmental munificence is high. If
the munificence of its industry segment is low, it is more likely to look for
opportunities in other segments of its industry. The result will be more related
change. The effect of environmental munificence is hypothesized as below:

Hypothesis 2.3. The lower the munificence of the industry
segment, the higher the related organizational change.

Hypothesis 4.4. The higher the munificence of the industry
segment, the more the probability of convergent organizational
change.

54

Environmental Dynamism

Environmental dynamism refers to the rate of change and unpredictability
of change in an environment (Dess & Beard, 1984). In a continually changing
and unpredictable environment, an organization can keep its match with the
environment by making incremental changes (Friesen & Miller, 1986;
Ginsberg, 1988; Ginsberg & Abrahamson, 1986). An abruptly changing
environment requires the organizations to make dramatic and discontinuous
changes (Meyer et al., 1990) to be efficient. These changes allow the
organization to minimize its misalignment within itself and with its
environment.

Apart from the above-mentioned economic pressures of the environment
that influence the organizational change, institutional forces also play a part.
The isomorphic, normative, and mimetic forces (DiMaggio & Powell, 1983)
influence a company to undergo organizational change. For example, in
today’s environment, more industries are becoming global. The normative,
isomorphic and mimetic effects of other organizations going global may
influence the orientation of a company. An organization is likely to be inclined
to make changes similar to others in its industry. The strategies that have
already been followed by the competitors provide legitimacy to organizational
actions and moves in that direction. This legitimacy can be taken as a resource

by those who are dissatisfied with the earlier strategy to make dramatic

55

changes. Therefore, it is expected that environments influence the degree of
dramatic change in organizations as follows:

Hypothesis 1.4. The lower the environmental dynamism, the less

dramatic the change; and the higher the environmental dynamism,

the more dramatic the change.

As reviewed in the previous chapter, the strategic choice perspective
focuses on the liberty available to the organizations and their managements
(Bourgeois, 1984; Child, 1972), particularly to the "dominant management
coalitions" (Cyert & March, 1963; March & Simon, 1958) or "upper echelons"
(Hambrick & Mason, 1984). This perspective underscores the importance of
decisions and choices made by the strategic management and the political
processes within the organizations (Murray, 1978). In keeping with the role of
strategic choice, three factors of asset specificity, organizational risk taken, and

organizational resources are considered for relating them to different

dimensions of organizational change.

Asset Specificity

The commitment of an organization’s resources to current strategy (Huff
et al., 1989) has important consequences for change in organizations.
Managers who are highly committed to the status-quo are likely to focus on the

advantages of the old strategy and on the dangers of switching to a new one. A

56

higher commitment of assets specific to the current strategy reduces the
propensity of redeploying assets, and it increases the replacement costs.
Therefore, asset specificity is likely to influence the dimensions of change in
organizations (Montgomery & Wernerfelt, 1988; Williamson, 1975),
irrespective of the level of organizational resources.

High asset specificity creates inertial pressures in the production process.
Such pressures make it difficult for an organization to contemplate alternative
products and markets (Hambrick & MacMillan, 1985). Therefore, the
resources committed to a strategy will increase inertia in the organization.
Highly specific resources are more likely to be useful for a strategy that is
convergent and related to the current strategy, resulting in positive incentives
for continued momentum. The more specific assets are not likely to be
recovered if the strategy is changed (Huff et al., 1989), increasing the
possibility of related and convergent change.

Organizational commitment to a course of action escalates (Staw, 1981)
because of investments in more specific assets. Such commitment is expected
to be high, without regard to the level of success or failure (Staw & Ross,
1987). Assets that are less flexible may be used to make related change (e.g.,
excess physical capacity or intangible assets can be used for a related change)
(Barton, 1988; Bettis, 1981; Chatterjee & Wernerfelt, 1991; Hoskisson & Hitt,

1990). In other words, organizations with highly specific assets are more likely

57

to pursue related and convergent changes. In the scenario of an organization
where the resources are more general, these can be put to a variety of uses.
For example, financial resources are more flexible than physical or intangible
ones (Chatterjee & Wernerfelt, 1991). Therefore, more proportion of financial
resources may allow the organization broader options. Thus, the organization
is not "stuck" with any strategy. It can consider any area for change. The
less-specific assets can be redeployed without considering the relatedness and
direction of their current strategy. Therefore, it is hypothesized that:

Hypothesis. The higher the asset specificity in an organization,

the higher the related change.

Hypothesis. The higher the asset specificity in an organization,

the higher the convergent change.
Organizational Risk Taken

Research on organizational adaptation and change revealed differential
responsiveness of organizations to environmental change (Ginsberg &
Buchholtz, 1990). It is well known in the literature on adoption of innovations
that early adopters and laggards have different characteristics. A characteristic
that influences the process of adoption and change in organizations is the degree
of risk undertaken by the organization.

An organization faces two kinds of risks: the systematic and the

unsystematic. Systematic (or market risk) is the sensitivity of an organization’s

58

returns to the market returns. It includes the operating risk and the financial
risk. Unsystematic risk is risk that is unique or business specific. It may be
due to many things, for example, the possibility of a wildcat strike or entry of a
new competitor with the same product (Bettis, 1983; Van Horne, 1983).
Organizations can manage both the systematic and the unsystematic risk and
add value for stockholders and the company (Bettis, 1983; Hoskisson & Hitt,
1990). Therefore, examination of total risk is appropriate, particularly because
systematic risk and total risk are empirically and theoretically correlated
(Fiegenbaum & Thomas, 1988). Organizations, like individuals, are not solely
risk averse or risk takers; they tend to reduce risks as well as take risks under
different circumstances (Fiegenbaum & Thomas, 1988; Hoskisson, Hitt, & Hill,
1991).

Under the high-risk conditions, organizations make changes to reduce
their risk (Hill & Hansen, 1991) to cope with the already high risk that they
have. The literature in organization studies is abundant with theories that argue
how organizations try to reduce risks and uncertainties (Amit & Wernerfelt,
1990; Burns & Stalker, 1961; Thompson, 1967). Therefore, an organization is
more likely to invest in other segments of its broad industry, rather than
growing only in one segment that will be more risky. Organizations thus
engage in related change that reduces the risk as compared to the risk involved

in expansion in its narrow segment. This type of risk reduction is desirable and

59

adds value for the management and the shareholders (Amit & Wernerfelt,
1990). Therefore, it is hypothesized as follows:

Hypothesis. The higher the risk taken by an organization, the

more the related organizational change.
Organizational Resources

In the literature on organization theory, organizational resources and
slack have received much attention (Bourgois, 1981; Cyert & March, 1963;
Sharfman et al., 1988). These are the resources that an organization has above
its requirements for carrying on its usual activities. In other words, the
resources are considered in relative terms, and not in absolute terms.

On one hand, organizational resources allow the organization to plan and
implement changes. It is recognized that for an organization to change, it
needs some minimal level of resources to successfully attempt implementing the
change in organizations (Hedberg, 1981; Huff et al., 1989; Lundberg, 1984).
An organization that does not have enough resources may be starved of the
essential ingredients of making certain types of changes, at least in a planned
way. Certain organizational activities are necessary for making changes.

These range from scanning the environment to collecting information on
changing consumers, suppliers, and competitors to help in strategic management
for formulating and implementing changes. Organizational resources can

provide support for organizations to adapt and change (Romanelli & Tushman,

60

1986), despite the influence of organizational performance.

The resources may be tangible, intangible, or of financial nature that
allow an organization to accomplish organizational changes. Thus, with
adequate organizational resources, an organization can consider broader change
(Hoskisson & Hitt, 1990). The funds may come from the equity, loans,
retained earnings, and so on. These may be invested in research and
development, advertising, plant and machinery, and so forth for change in
product diversity (Hill & Hansen, 1991) and international diversity of
organizations.

Organizational resources, on the other hand, also build complacency in
the organization and are likely to lower motivation for change (Cyert & March,
1963; Hedberg, 1981). These give rise to inertia and resistance to change
(Starbuck & Hedberg, 1977; Thompson, 1967). Management receives more
legitimacy for the current strategy, eroding the need for unrelated, divergent, or
dramatic change. Both influences are expected to go on in any organization.
Although organizational resources make it easier to carry out change, they
lower the motivation to undertake change (Ginsberg & Buchholtz, 1990).
Therefore, a curvilinear relationship between organizational resources and
different dimensions of change is expected.

It is hypothesized that availability of higher resources will make change

more feasible and, therefore, an organization is more likely to involve in

61

dramatic and divergent change. Similarly, lower organizational slack will
increase motivation for change, increasing the likelihood of dramatic and
divergent change. However, when the organizations have moderate amounts of
resources, they are more likely to make incremental and convergent changes.
The hypotheses are as follows:

Hypothesis. Organizations with extreme levels of organizational

resources will make more dramatic change, whereas organizations

with moderate levels of organizational resources will make less

dramatic change.

Hypothesis. Organizations with moderate levels of organizational

resources are more likely to change in a convergent mode,

whereas organizations with extreme levels of organizational

resources are less likely to change in a convergent mode.
Organizational Performance

Organizational performance influences change in organizations (Boeker
& Goodstein, 1991; Ginsberg & Buchholtz, 1990; Grant, Jammine, & Thomas,
1988). It provides feedback to management and sends signals about the
effectiveness of the organizational strategy (Lundberg, 1984). The perceptions
and interpretations of the earlier patterns of organizational behavior or
processes of change undertaken by the organization are likely to be altered by
performance (Moch & Bartunek, 1990; Weick, 1979).

The performance variable provides feedback to the managers on their

strategies, on organizational structure, and on their management styles. It is

62

expected that low levels of performance will signify stress and crisis in the
organization (Huff & Huff, 1990; Lundberg, 1984). That is, while high
performance protects the organizations and their managements (Staw et al.,
1981), poor performance indicates ineffectiveness of the strategy. Therefore,
under these circumstances, organizations are likely to make divergent and
dramatic changes (Duhaime & Grant, 1984; Montgomery & Thomas, 1988).
For example, crisis of performance can lead organizations to sell off their
business units they are unrelated to their core business (Ravenscraft & Scherer,
1987). Similarly, lower performance is likely to result in lower share price,
increasing chances for a hostile takeover (Hoskisson et al., 1991). An
extremely high level of performance is expected to lead to convergent and
related organizational changes. Therefore, the relationship of performance to
various dimensions of change is hypothesized as below:

Hypothesis. The lower the level of performance, the more
dramatic the organizational change.

Hypothesis. The higher the level of performance, the more related
the organizational change.

Hypothesis. The lower the level of performance, the more
pervasive the organizational change.

Hypothesis. The higher the level of performance, the higher the
probability of convergent organizational change.

Linear relationships of performance with different dimensions of change have

63
been hypothesized. However, some studies have argued that there may be a

curvilinear relationship between performance and extent of change (Hoskisson
& Hitt, 1990). Poor performance may motivate the organization to undertake
increased change, but continuing poor performance may decrease further
change (Grant et al., 1988; Hoskisson & Hitt, 1990; Ravenscraft & Scherer,
1987).

It is argued in the literature that extremely low performance motivates an
organization’s management to undertake "problemistic search" (Baum, 1990).
This search is due to gaps in expected and actual performance that lead to
strategic organizational changes (Cyert & March, 1963; Kiesler & Sproull,
1982; Schendel & Patton, 1976). Similarly, it is contended that an extremely
high level of performance may also have a similar effect of resulting in high
organizational change (Baum, 1990; Child & Kieser, 1981; Dutton & Duncan,
1987; Singh, 1986). Therefore, low and high performance may lead to higher
change as compared to moderate performance (Baum, 1990).

However, Fombrun and Ginsberg (1986) reported the opposite results
(i.e., extreme high or low levels of performance in previous time periods were
related to lower organizational change in subsequent periods rather than
intermediate levels of performance). Lower organizational change in poorly
performing organizations may be attributed to the threat-rigidity effects (Staw et

al., 1981) and adversity leading to conservatism in organizations (Cameron,

64
Kim, & Whetten, 1987; Whetten, 1987). Thus, there is support for a U-shaped

as well as an inverted-U shaped relationship of performance and change in
organizations. Therefore, at this time there can only be speculation that there is
a possibility of a nonlinear relationship of performance with some dimensions
of change. The curvilinearity of relationship will be explored, in case the

findings are not in the hypothesized direction.

Interactions

The studies reviewed in the previous chapter reported interaction effects
of the sources of change on the processes of change. For example, Ginsberg
(1988) noted several studies (Graham & Richards, 1979; Harrigan, 1981;
Oster, 1982) that did not find the main effects for poor performance. Graham
and Richards (1979) reported that performance influenced organizational change
only when there were changes in the general environment (e. g., deregulation of
the railroad industry). Similarly, Harrigan ( 1981) and Oster (1982) argued that
the influence of performance on organizational change depended on the industry
characteristics. Therefore, performance is expected to interact with
environmental dynamism and munificence.

Organizational Performance x Environmental Dynamism

The main effect of environmental dynamism is related to the degree of

dramatic change in organizations. This relationship is likely to vary with the

65

level of organizational performance. A poorly performing organization faced
with a low dynamic environment is likely to keep on adjusting to improve its
alignment and future performance. However, the poorly performing
organization in a fast-changing environment needs to take drastic steps fast.
Therefore, an organization in a more dynamic environment with poor
performance is more likely to make a more dramatic change than would be the
case if it had good performance conditions.

Hypothesis. The positive relationship between environmental

dynamism and the degree of dramatic organizational change will

be amplified to the extent the organization experiences low

performance.

Organizational Performance x Environmental Munificence

As noted previously, under the circumstances of lower environmental
munificence, an organization is expected to make more related change.
However, the effect of environmental munificence on the extent of related
change is expected to be moderated by the level of organizational performance.
A lower performing organization in an industry segment that has lower
munificence is more likely to explore opportunities outside its segment. As a
first alternative, it is expected to look for products in other related industry

segments. Therefore, it is likely to involve itself in much more related change

than would be the case if it were in a scarce environment but had higher

performance.

66

Hypothesis. The negative relationship between environmental
munificence and the extent of related change will be amplified to
the extent that the organization experiences low performance.

In the foregoing sections of this chapter, the relationships between
dimensions of change and their correlates have been hypothesized. The
hypothesized relationships propose that there are multiple dimensions of change,
and that each dimension is depicted by a characteristic model that differentiates
it from other dimensions. Therefore, it will be more convenient to arrange
results according to the dimensions of organizational change. Accordingly, the
hypotheses of the study are enumerated below with respect to each dimension.
Degree of dramatic change. The degree of dramatic change is expected to be
related to organizational performance, size, environmental munificence, and
organizational resources as follows:

Hypothesis la: The lower the level of performance, the more
dramatic the organizational change.

Hypothesis 1b: Small and big organizations change in a less
dramatic fashion, whereas medium-sized organizations change in a
more dramatic fashion.

Hypothesis 1c: The lower the environmental dynamism, the less
dramatic the change; the higher the environmental dynamism, the
more dramatic the change.

Hypothesis 1d: Organizations with extreme levels of
organizational resources will make more dramatic changes,
whereas organizations with moderate levels of organizational
resources will make less dramatic changes.

67

Hypothesis 1e: The positive relationship between environmental
dynamism and the degree of dramatic organizational change will
be amplified to the extent the organization experiences low
performance.
Degree of related change. The model of related change is expected to include
significant relationships with organizational performance, size, environmental

munificence, risk, and asset specificity, as given below:

Hypothesis 2a: The higher the level of performance, the more
related the organizational change.

Hypothesis 2b: Small and big organizations make less related
change, whereas medium-sized organizations make more related

change.

Hypothesis 2c: The lower the munificence of the industry
segment, the higher the related organizational change.

Hypothesis 2d: The higher the risk taken by an organization, the
more the related organizational change.

Hypothesis 2e: The higher the asset specificity in an organization,
the higher the related change.

Hypothesis 2f: The negative relationship between environmental
munificence and the extent of related change will be amplified to
the extent that the organization experiences low performance.
Degree of pervasive change. The correlates of pervasive change are likely to
be organizational performance, size, and system coupling, as hypothesized

below:

Hypothesis 3a: The lower the level of performance, the more
pervasive the organizational change.

Hypothesis 3b: Small and big organizations make less pervasive

68

change, whereas medium-sized organizations make more pervasive
change.

Hypothesis 3c: The tighter the system coupling in an organization,
the more pervasive the organizational change.

Convergent change. Convergent change is hypothesized to be related to
organizational performance, size, environmental munificence, resources, and
asset specificity, as follows:

Hypothesis 4a: The higher the level of performance, the higher
the probability of convergent organizational change.

Hypothesis 4b: Small and big organizations have a higher
probability of changing in a convergent fashion, whereas medium-
sized organizations have a lower probability of changing in a
convergent fashion.

Hypothesis 4c: The higher the munificence of the industry
segment, the more the probability of convergent organizational
change.

Hypothesis 4d: Organizations with moderate levels of
organizational resources are more likely to change in a convergent
mode, whereas organizations with extreme levels of organizational
resources are less likely to change in a convergent mode.
Hypothesis 4e: The higher the asset specificity in an organization,
the higher the convergent change.

The details of research methods used in this study to test the proposed

hypotheses are given in the next chapter.

Chapter 4

Methods

THE SAMPLE

The population of organizations for the study included the companies
registered on the stock exchanges in the United States. The sample included
the companies for which data were available on the business information
segments in COMPUSTAT 11. Data on different variables were available for a
varying number of companies. Data on most variables were available for 731
companies for product diversity and 855 companies for international diversity.
The exceptions were pervasiveness of change and system coupling for which
availability of data was limited. The sample sizes of 731 and 855 for product
and international diversity, respectively, gave statistical power of > .99 for an
overall test of the analysis. These sample sizes provided statistical power of
about .80 for testing the effect of an individual independent variable in multiple
regression (Cohen & Cohen, 1983, pp. 1 16-1 19). Data on pervasive change and
system coupling were available for 431 and 755 companies, respectively.
However, the number of companies for which both pervasiveness of change and
system coupling data were available was reduced to 166 companies.

The companies included in the sample represented a wide diversity on
various dimensions. The industries included ranged from mining to
manufacturing to service. Their SIC codes were from 10 to 87, with SIC code
99 representing the conglomerates. The sample included a very wide range of

small, medium and large organizations. The size of the organizations in terms

70

71
of number of employees varied from 1 to 130,500, with an average of 4,615 .'

The mean sales of the sample were 1.35 billion and the range was from 30.8
million to 12 billion. The companies in the sample had product diversity from
0 to 1.50, with an average of .20. Product diversity index of 0 represented an
undiversified company that has all of its sales in one industry segment.
Examples of such companies are Intel Corp., Michigan Bell, MCI
Communications, and United Parcel Service America Inc. Examples of
diversified companies included in the sample were Revlon, Inc. (index=.55)
and USX Corp. (index=.71).

The average international diversity was .11, with a range of 0 to 1.42.
A 0 index of international diversification implied the company had all of its
sales in the domestic market, for example, such companies are Nordstrom Inc.,
Michigan Bell, and Burlington Northern Railroad Co. The companies with
higher international diversity indexes included Intel Corp. (index=.99), Revlon,

Inc. (index=1.33), and Dow Corning Corp. (index=1.42).

 

‘ It may be mentioned here that there were some companies in
the sample such as cooperatives that had very few employees.
Such companies had other individuals or organizations as
members. Log of employees was taken that helped to
normalize the variable of organizational size, as described
later in the chapter.

72
VARIABLES AND THEIR MEASURENIENT

Dependent Variables

Four dimensions of organizational change were identified: increment,
relatedness, pervasiveness and direction of change. All these dimensions are
expected to be relevant for the two types of diversity analyzed in this study:
change in product and international diversity. Both variables of diversity were
measured along the four dimensions of change.
1. Level of Product Diversity

Change in product diversity over time is product diversification (Grant et
al., 1988) or de-diversification a.uffman & Reed, 1982). The literature on
product diversification offers alternative measurements of product diversity.
Broadly, there are two types of measures. The first, entropy measures
originated in the field of industrial economics (Jacquemin & Berry, 1979) and
have since been improved upon (Palepu, 1985). The second, subjective
measures followed the categorization developed by Rumelt (1974). These two
measures converge and can be considered valid and reliable, though they are
not without their limitations (Montgomery, 1982). These two measures or their
small variants have been extensively used in research (Grant et al., 1988). In
this study, the entropy measures of diversity (Palepu, 1985) were used because
they were objective, precise, continuous and could be reliably computed on a

large number of companies over time. Product diversity was measured as

73

follows:
Total Product Diversity (TPD) ="E,__.,P,ln(1/P,.) ;

where
P, is the share of the ith segment in the total sales of the company.

i = 1 ...... N is the number of industry segments the company
operates in.

The above measure is better than the product count measures because it
considers the number of the segments of a company and the relative importance
of each segment in the company’s total revenues. The total product diversity is
measured based on weighted average of the shares of the segments. Each
segment is given the weight equal to its logarithm of the inverse of its share
(Palepu, 1985).
2. Level of International Diversity

The second variable of organizational change involved change in the
internationalization of the company. The process of measuring international
diversity was very similar to that of product diversity. The measures assesses
geographic diversity of the organization in different countries. With the
internationalization of competition and global marketplace (Porter, 1990),
international strategies are gaining increasing importance. These changes are
taking place because of strategic considerations in organizations. There are
many possible objective measures of internationalization: sales of foreign

subsidiaries/sales of domestic subsidiaries, foreign sales/total sales, foreign

74

production/total production, foreign assets/total assets, and foreign income/total
income. These measures have been suggested and used by many researchers in
the area (Geringer et al., 1989; Grant et al., 1988). The measures mentioned
above have a weakness, that is, they do not consider the dispersion of the
activities of the organization. Vachani (1991) proposed entropy measures of
international diversity that are similar to entropy measures of product diversity.
Organizations may be operating in one homogeneous broad region (e.g.,
Europe), or its activities may be dispersed in several heterogeneous regions
(e.g., Europe, North America, and the Pacific Rim). The homogeneity or
heterogeneity of geographical regions can be defined in several ways. The
criteria may include physical, social and cultural proximity; the level of
economic development (V achani, 1991); or market conditions related to these
characteristics (Cavusgil, 1980). Attempts to define nature of the markets have
been made by Buhner (1987), Grant (1987), and Kim and Colleagues (1989).

Total change in international diversity was calculated like the total
change in product diversity, except that the country groups (e. g., Europe, Asia)
were used in place of two-digit SIC codes, and countries (e. g., the United
Kingdom) were considered instead of four—digit SIC codes. Therefore,

Total International Diversity (TID) =”E,=,P,ln(l/P,.);

where

P is the share of the ith country in the total sales of the company.
i = l ...... N rs the number of countries the company operates in.

75

Dimensions of Change

The four dimensions of change were the dependent variables. The
measures of each dimension for change in product and international diversity
are described below.

1. Increment of Change (Degree of Dramatic Change)

Product Diversity. Increment of change refers to the size of change per
unit of time. Net change in diversity may have happened in small increments
or in big leaps. It was measured by the variance of the annual changes in
product diversity for the period 1978-1990. It was a continuous measure. A
lower variance reflected a lower degree of dramatic change (and a higher
degree of incremental change). Similarly, a higher variance meant that the
change was more dramatic in nature. Annual change was calculated using the
following formula:

Annual change = Change in total product diversity
= (TPDr ' TpDr-t)

International Diversity. Incremental or dramatic internationalization was
measured by the variance of the annual changes in international diversity for the
period 1984-1990. Higher level of variance reflected more dramatic nature of
change in globalization, and the lower variance suggested an incremental
change. This measure was computed in a similar way as the product diversity

measure using the following formula to calculate annual changes.

76

Annual change = Change in total international diversity
= (T ID, - TIDH)
2. Relatedness of Change (Degree of Related Change)

Product Diversity. Relatedness of change refers to the relationship
among the newly added or divested activities and other activities of the
organization. Entropy measures (Palepu, 1985) were used that assess related
diversity of corporate entities. Related diversity is taken in terms of the
business segments (the four-digit SIC code industries) within industry groups
(two-digit SIC code industries) of the company. The measure of related change
in diversity was continuous, and it reflected related change for the period 1978-
1990. The measure of related product diversity (RPD) was calculated as
follows:

Related Product Diversity (RPD) = ”Ej=,RPDjF ;
where
P = the share of the jth group sales in the total sales of the
company. .
RPDJ = Eijian/H) ;

where F, is the share of the segment i of
group j in the total sales of the group

RPDj is the related diversity of the different
segments within an industry group j

The measure of related change was computed as the difference score between

related product diversity over the years using the formula:

77
Degree of annual related change = (RPD, - RPD“)
Total related change = ,,,,{(RPD, - RPD,,,)

where
N = number of years
RPD = Related product diversity

International Diversity. Related change in international diversity was
calculated in a similar way as related change in product diversity. The formula
used was as follows:

Related International Diversity (RID) = ”13)., ,RIDJF' ;

where

F = the share of the jth country group sales in the total sales

of the company.

RID} = 2,&jF,1n(1/F,) ;

where
P} is the share of the country i of country group j in the
total sales of the group

RIDj is the related diversity of the different countries within a
country group j

Degree of annual related change in international diversity was computed using

the following formula:

Annual related change = (RID, - RID”)
Total related change = ”E,=,{(RID, - RID“)
where

N = number of years
RID = Related international diversity

78

3. Pervasiveness of Change (Degree of Pervasive Change)

Pervasiveness of change is defined in terms of functions that
organizational change encompasses in an organization at one time. The
multiple correlation over time among the following three variables was the
measure of the degree of pervasive change: change in capital expenditure,
change in advertising expenditure, and change in research and development
expenditure. These variables are closely related to the change in product and
international diversity of organizations, and they show the degree of
pervasiveness of change. Two measures of capital expenditures were used in
the study. The first was change in net property, plant, and equipment of the
company, and the other was change in gross property, plant, and equipment of
the company. These two measures of capital expenditures gave two measures
of the degree of pervasive change. Both the measures were correlated .74 and
.77 in product diversification and geographic diversification data, respectively.
4. Direction of Change (Convergent-Divergent Change)

Product Diversity. As conceptualized, convergent-divergent change
measures the direction of change. An organization may have increased the total
diversity (diversification), or may have decreased it over time (de-
diversification or reverse-diversification (Luffman & Reed, 1982)). For an
organization that was following a strategy of increasing product diversity, a

further increase in diversity (diversification) reflects convergent change. A

79

decrease in diversity (de-diversification) shows divergent change in the
company. If the organization continues to follow the same strategy, this
reflects convergence, continuity, or momentum (Miller & Friesen, 1984).
Reversal in strategy reflects divergent change.

The direction of change of the organization was measured in the first
half period of change (1978-1984) and the second half period change (1984-
1990). If the change of diversity was in the same direction during both the
periods, it meant the organization made a convergent change. Otherwise, the
change was interpreted as divergent change. This was a discontinuous
measure. A score of l and 0 represented convergent and divergent change,
respectively.

International Diversity. For an organization that was increasing its
globalization previously, an increase in the degree of internationalization
reflected convergence, whereas a decrease meant divergent change in the
organization. This was measured in the same way as product diversity, except
that the years of the two half periods were 1984-1987 and 1987-1990,

respectively.

80
Independent Variables

An organization does not respond to change immediately; it has a longer
planning horizon. It may be hard to know the exact planning horizon of an
organization, and such horizons may vary according to the characteristics of the
organizations and their environments. But, it may be assumed that an
organization normally uses 4-5 year planning period. Similar periods have been
used in other studies (e.g., Finkelstein & Hambrick, 1990; Miller & Friesen,
1984). Therefore, independent variables were computed for the period 1975-
1987, and their relationships were studied with change in product diversity
from 1978-1990. In the case of change in international diversity, the
independent variables were measured for the period 1978-1983, and the
dependent variables were measured for the period 1984—1990. The
measurement of current resources was the exception. Current resources were
measured for the same years as were the dependent variables.

Independent variables in the study were operationalized as follows:

1 . Organization Size

Size is an often-studied variable, and it is measured in many different
ways. In this study, it was measured in terms of total number of employees of
the company. The mean of the number of employees was computed over the
years. Log of the mean number of employees over the years was taken because

it helped normalize the variable (Blau & Schoenherr, 1971). This measure is

81

related with other measures of size (e.g., assets and sales of the company).
Assets and sales were used in other variables; therefore, these measures were
not used for size.
2. System Coupling

System coupling expresses the interdependence of subsystems in a
system, and therefore it was measured using a surrogate measure of
interdependency. It was assumed that companies in which subunits reported
market transactions with one another would be less tightly coupled than
companies in which subunits did not report such transactions. It was expected
that undifferentiated organizational units were interrelated by administrative
mechanisms rather than by markets (Williamson, 1975), and were more tightly
coupled. Intracompany sales increase as units attain more independence
because more intracompany transactions are recorded. Therefore, low
intracompany sales represented high system coupling, and vice versa. The
measure was intracompany sales divided by total sales of the company,
averaged over time. The measure was multiplied by -1 so that higher value of

the measure represented high system coupling.

3. Environmental Munificence
Munificence relates to the capacity of the environment to support the

activities of an organization for growth and stability (Aldrich, 1979; Dess &

82
Beard, 1984; Starbuck, 1976). It is measured essentially using the growth rate

of the industry over the years (Dess & Beard, 1984; Keats & Hitt, 1988). This
variable was measured using the variables of sales at the 4-digit SIC code
industry level (Keats & Hitt, 1988). The measure of growth is derived from
the linear regression equation, and it is represented by the regression
coefficient. It is identical to using the compound rate of interest formula to
compute the rate of growth of the industry (Allen, 1988; Moch, Dass, Rubin,
& Mendez, 1991). Assuming, the constant rate of growth, the formula is a
basic form of the regression equation (Keats & Hitt, 1988). It uses a growth
curve measure derived from multiwave data and offers improvement over two
data point measure (Rogosa, Brandt, & Zimowski, 1982).
Y,=bo+b,+e,;
where Y = Industry sales or industry operating income at time t
b0= Intercept
b1= beta (regression coefficient)
t = year
e = residual
4. Environmental Dynamism
Dess and Beard (1984) combined stability-instability and the turbulence
dimensions (Aldrich, 1979; Thompson, 1967) in the form of a measure of
environmental dynamism. This variable reflects variability in the industry

environment of an organization. Environmental dynamism can be measured by

the dispersion from the regression equation given in the environmental

83
munificence variable (Dess & Beard, 1984; Keats & Hitt, 1988). The exact

measure of environmental dynamism in the study was the standard error (SM) of
the regression slope coefficient in the equation given above (in environmental
munificence).

5. Asset Specificity

Asset specificity reflects organizational commitment to a strategy and the
likelihood of its change. It refers to the assets that are specifically invested in
pursuing a strategy. The specific assets further reinforce managers to continue
the course of action they have undertaken (Staw & Ross, 1987). They do this
because these assets may not be recovered if the strategy is changed (Huff et
al., 1989). The variable of asset specificity can be measured in terms of fixed
assets relative to current or total assets, where higher ratio means higher asset
specificity. However, this measure was not used in the study because a similar
measure was employed for another variable (i.e., organizational resources).

An alternative measure that was used in this study operationalized
specific assets through levels of intangible and financial assets in the
organization. Intangible assets include patents, copyrights, and brand names.
These are more specific assets (Bettis, 1981; Chatterjee & Wernerfelt, 1991;
Hill & Snell, 1988; Montgomery & Hariharan, 1991). However, financial
assets are more flexible (Chatterjee & Wernerfelt, 1991). The measure was the

ratio of intangible assets to financial assets. This measure may vary with the

84

type of industry and therefore was standardized and considered relative to
industry average. The COMPUSTAT II database reports intangible assets and
financial assets of the organizations.

Asset specificity may fluctuate from one year to another. The mean of
asset specificity over the years 1975—1987 reflected average asset specificity of
the organization.

6. Organizational Risk

A measure of total risk was used in the study. The measure was the
variance of company returns in the past. It is an accounting-based measure.
The measure relates to the fluctuations of the past returns of the company over
time (Bettis, 1981; Bettis & Hall, 1982; Bowman, 1980; Christensen &
Montgomery, 1981; Fiegenbaum & Thomas, 1988; Rumelt, 1974).
Organizational risk was measured for the time-period of 1975-1987. High
variability in returns shows higher risk for the organization. This variable was
also standardized with respect to the industry.

7. Organizational Resources

Two measures of resources were used in the study: long-term and short-
term. Following Bourgeois (1981) and Singh (1986), long-term resources
represent unabsorbed slack (excess uncommitted resources) that was measured
by the ratio of equity to debt (Finkelstein & Hambrick, 1990; Hambrick &

D’Aveni, 1988). It reflects resource availability over the long-term that is

85

important for making changes in strategy of the organization. The second
measure was of short-term liquidity (Finkelstein & Hambrick, 1990; Hambrick
& D’Aveni, 1988) that was measured in terms of current ratio (current
assets/current liabilities) (Hill & Hansen, 1991). Because characteristics of
industries differ with respect to the requirement of resources, the measures
were taken relative to industry.
8. Organizational Performance

Organizational performance was measured in terms of financial
performance of the company using an adjusted return on assets (AROA). This
measure adjusts for average industry performance and gives organizational
performance relative to other organizations with similar portfolios. Return on
assets is a measure of organizational performance that managers accept as a
good criterion of organizational performance (Bettis & Mahajan, 1985). It was
computed using the following formula:
AROA = ROA, - (Emu-,ROAF) ;
where ROA, is the return on assets of the organization i

mg, is the proportion of organization’s sales in four-digit industry j

ROAfl is the return on assets of four-digit industry j.
ROA =(Profits after tax + interest)/Total assets

86
STATISTICAL ANALYSIS

Power Analysis

All studies need enough statistical power to get positive results when
they are actually there. The requirement of power may vary from one study to
another, even from one hypothesis to another, depending on the effect size and
Type I error. As a convention, Cohen and Cohen (1983) suggested the
statistical power of .80 and alpha = 0.5. Effect size depends on the variables
of interest. Effect size may be estimated from the past studies, or from the
expectations. The effect size (R2) for an overall test that was of interest in the
present study was estimated to be 0.05 (Robertson, Roberts, & Porras, 1993).
Therefore, following the procedure given by Cohen and Cohen (1983) for
computing statistical power for multivariate regression, the minimum sample
size required for an overall test was 353. There were 700 companies for most
of the analysis, which gives a power of > .99 for the overall test. However,
for testing effect of individual variables, with an effect size (R2) of .01, the
sample size required was about 787 for a power of .80 (alpha = .05).
Therefore, there was a statistical power of about .80 for testing individual
effects for change in product diversity. The statistical power for testing
individual hypotheses for change in international diversity was > .80, because

the sample size was 855.

87
Data Analysis

The sample of organizations included companies from different
industries. Therefore, the company-level variables, both the independent and
dependent variables were standardized with a mean of 0 and standard deviation
of 1 with respect to their 2-digit SIC code level industry to render data
comparable across industries. However, as a result, a number of variables
were negative. The negative numbers were converted to positive numbers by
adding the corresponding minimum number in each variable. Furthermore, in
case of interaction terms, variance of two contributing variables was made
equal (to 1) to make sure two variables contributed almost equally to the
interaction term. As a result of these operations, the variables were all positive
with a standard deviation of 1.

Data were analyzed using hierarchical regression analysis for dramatic,
related, and pervasive change. The dependent variable of direction of change
was discontinuous; therefore, it was analyzed using probit analysis (Aldrich &
Nelson, 1984). An econometric software called SHAZAM (White, Haun, &
Horsman, 1987) was used for the probit analysis. Probit analysis is used as an
alternative to the linear regression analysis because the discontinuous dependent
variable does not meet the requirements of the ordinary least squares solution.
Consequently, the regression estimates though are unbiased, but are not best

(i.e., they do not have the least variance) (Aldrich & Nelson, 1984). In such

88

cases where dependent variables are qualitative, probit analysis is a good
alternative. Probit analysis uses a nonlinear normal distribution similar to
logistic distribution. Therefore, the interpretation of the estimated coefficients
is different. According to LeClere (1992), "(T)he sign of the coefficient gives
the direction of the effect of a change in an independent variable. The relative
magnitude of a coefficient indicates the relative influence that a variable has on
the probability of choice" (p. 771). Slope or first derivative gives the marginal
probability at the mean of the independent variables. But, the problem with the
slope is that it is not invariant to scale. A better measure of the true marginal
probability is the elasticity of probability because elasticities are independent of
the scale. Regarding the interpretation of elasticities, LeClere (1992) noted as

follows:

”A high elasticity (> 1) implies that the probability is very
responsive to changes in the exogenous variables as there is a
greater than proportionate change in the probability relative to the
exogenous variable. A low elasticity ( < 1) implies that the
probability is very unresponsive to changes in the exogenous
variables as there is a less than proportionate change in the
probability relative to the exogenous variable." (p. 772).

The problem of multicollinearity was assessed in the independent
variables and appropriate steps were taken to deal with it (Hill & Hansen,
1991; Kelejian & Oates, 1974). Results of the study are reported and discussed

in the following chapters.

Chapter 5

Results

MATRIX OF CORRELATIONS

The focus of this study was to investigate multidimensionality of
organizational change. Organizational theories have implicitly treated change as
unidimensional. For change to be considered unidimensional, the change
dimensions must be strongly correlated with one another. The null hypothesis
was that the multiple dimensions would be strongly correlated with one another.
The alternate hypothesis was that the multiple dimensions of change would be
weakly correlated with one another.

Tables 1 and 2 exhibit the pearson correlation coefficients among various
dimensions of change in product and international diversity, respectively. As
these tables revealed, most of the correlation coefficients were equal to or less
than .05 in both types of diversity. Specifically, five out of six correlation
coefficients in product diversity, and four out of six correlation coefficients in
international diversity were < = .05. The rest of the correlations ranged from
.09 to .20. These relatively weak correlations among the dimensions of change
rejected the null hypothesis that the dimensions were strongly correlated.

Some correlations among the dimensions, though weak, were significant.
These can be interpreted as representing part of the thinking in the literature.
For example, the correlation coefficient between the degree of dramatic change
and the degree of convergent change was -.20. It suggested that organizations

were slightly inclined to make less convergent (divergent) changes in a dramatic

90

91
Table 1

Pearson Correlation Coefficients among Dimensions of Change

in Product Diversity

 

 

Variables 1 2 3

1. Degree of dramatic change

2. Degree of related change -.05*

3. Degree of pervasive change -.01 -.05

4. Convergent change -.20*** .04* -.02

 

N=2285, except in case of pervasive change, where N=432.

***p <=.001
** p <=.01
* p <=.05

92
Table 2

Pearson Correlation Coefficients among Dimensions of Change

in International Diversity

 

Variables 1 2 3

 

1. Degree of dramatic change

2. Degree of related change .11***
3. Degree of pervasiveness change .00 -.02
4. Convergent change -.09*** -.05** .04

 

N=3340, except pervasiveness of change, where N=457.
*** p < =.001

** p < =.01

* p < =.05

93

fashion. The same seemed to be true in the case of international diversity,
though the correlation coefficient was even weaker. The pervasive change was
least correlated with other dimensions of change in product and international
diversity. These correlations indicated that pervasiveness of change was
relatively more independent of other dimensions of change.

The second point observed from Tables 1 and 2 was that the correlations
of different dimensions of change were not universal in nature. In other words,
the differing correlations rejected a generic model of change in product and
international diversities. The pearson correlation coefficients varied depending
on the type of change. For example, degree of dramatic change and degree of
related change were negatively correlated in the case of product diversity. The
same two dimensions were positively correlated in the case of international
diversity. It may be noted that these correlations were weak and need to be
interpreted cautiously. The pearson correlation of the degree of dramatic
change in product diversity with the degree of dramatic change in international
diversity was .11 (p< .01). The direction of change in two diversities was
negatively related (r = -.08, p = .06). The related changes in two diversities
had correlation of .03 (p = .41) whereas the correlation of pervasive changes
was .37 (p< .00). Most of these correlations were weak that did not support

the expectation of a generic model of change.

94
FACTOR ANALYSES

Table 3 and 4 present the results of factor analyses of five measures of
change in product diversity and international diversity, respectively. Four
factors were derived from the data. Factor analysis using principle components
method was run using factors 1 to 5. Varimax rotation was used in the
analysis. An examination of these results showed that the four-factor solution
provided the most clearly interpretable solution in both types of changes.
Change of product diversity. As Table 3 reveals, the first factor explained 38
percent of the variance, and the second factor contributed 28 percent of the
variance. Third and fourth factors contributed 17 percent and 14 percent of the
variance. The eigen values for the factors 1 to 4 were 1.90, 1.38, .84, and
.70, respectively. Two measures loaded on one factor of pervasive change.
The other three measures loaded on each of the other three factors. The factor
loading on the corresponding factors were .98 to .99. On other factors, the
loadings were less than .15. In solutions involving 1 to 3 factors, measures
loaded on more than one factor with high loadings.

Change in international diversity. As Table 4 displays, factors 1 to 4 had
eigen values of 1.94, 1.19, 1.03, and .77, respectively. The percentages of
variance explained by the four factors were 39, 24, 21, and 15, respectively.

Other results were similar to the change in product diversity.

382 $80 mafia 85988 32582.. .838 c. owcacu ._
882 .02 mafia 3330:. .5582: 3.98 5 352.0 .

 

 

3. 2. mm. mm. 8:393 855., 386..

E. g. w? 8.. 2.? 5mm.
ma. 2. 3.- N. .- EoEwom .3565 25:5 :5 bums—E.

SEE 55.3 35:36 me omSEU .m

:. 8. a .- 8. £9.26 .6 882% a owsao .4

3.- 2 .- 8. 8.522% a 83% 3:5 .0 8§E> .m
.83on 9808 mama: 35.582: 5

8.- mo. 2 .- 3. 89:20 mecca 5:22.50 03232 .N
.23on 25 was: 35.582: 5

2.- 8.. no. mm. mowswco wee—5 cows—2.80 032:2 ..
v 58m”. m 8.8... a 88m". . 8.8... 85302

 

5.83.: 8.69:. :. mug—.0 me 8253:. we mam—«5... .383.

m 2:5.

no

383. 8.80 was: 8588 “88.8.6. 3&8 a. owSEU ..
382 .02 mg”: 8:386 28.582: .538 c. omSEU .

 

 

2. a. 8. 8. 8:33... 855.. 688

R. 8.. 82 8. 83> 5mm—

8. 8.- 8. 8.- 88% .6 882% 5 8526 .m

8.- 8. 8. 8.5.32.6 5 .855 Base .6 8§E> .4
.9550 05 0233 :5 :2on

8. 8. 8. 8.- .83 as? 88% co 856 .m
A.2385 9508 $5: 3:25ng 5

oo. 3.- nor .3. mowfico waoEm cone—“E8 03232 .N
.2388 use was: 38:582.. 5

mo: 5. 8. mm. mowed—.0 macaw coca—oboe 032:2 ..
a 88m". m “88". m “Scam . 88¢”. 8:582

 

5...».on 3:23.53:— 3 09.2.0 .3 @2382 he £93.25. 532“."—

9 ~35...
ea

97

From the foregoing, it can be concluded that these dimensions of change
are not strongly correlated with one another. The change dimensions cannot be
represented in one factor, thereby, rejecting the notion of unidimensionality of
change.

Tables 5 and 6 present correlation coefficients among the independent
variables for change in product and international diversities, respectively. The
tables showed that organizational size, performance, risk, and resources were
correlated with one another to varying degrees in the case of change in product
and international diversities. However, the maximum correlation among the
independent variables was observed between two measures of industry
environment (i.e., environmental munificence and dynamism). The correlation
was very strong (.81) and will cause multicollinearity problem in regression
analyses. Therefore, only the variable of primary interest in the regression
equation is retained out of these two variables, when there is a problem of

multicollinearity.

no.flV a .w

 

 

8." v a 3.

8o." v a 1....

.2. .8. 8.- 8. 8. 8. :. 8.- 8:88 83% .8
:8. 5.3.3.- 8. 8.- 3.8. 8.- 8. .88QO E0..- 83 .w
5.8.. 8. :8. 8. 8. 8. 58:53 .555 s

aim. 1.8. 2.3.2. 8. 3:8.- ofim 5855580 .0

5.8. .........8. :8.- ..ta..- 85.583 5583590 .m

it». .........8.- .8.- exeéE 5555.35. .5

.38.- :8.- 8§E§2 555235. .m

.8. 55583 ~33 .N

52 5585555 ._

m o v N _ §55>

 

mam-83n— 8255 E 8_aa€a> “no—853...: 958a 8535er noun—8.80 =8..qu—

m 035—.

mm

8.nv ..

*

 

 

S." v a 3..

.8." v a 3....

2.. .. 8. 8.- z. 8. 8. on. 8. 8:88 22.8 .8
3....8. :8..- 8.- :8.- 8.- .8.- ..8. .8388. :59 83 .8
film.- ..8. .58.- ...8: :8.- 8. .8588. .855 s

3.8. 3...... ........8. 8.- 3.8.- 88 5:85.580 8

.58. 3:8. 58.. 5.8.. 85.5.8.8 8.2.5.580 .m

:8... _........8.- 8.- 5.585 38:52.28 ...

.........8.- 8.- 8:885: 5.55.5828 .m

5:8. 8.8.8.5 .32 .m

8.... 5:88.580 ..

w h o m 5 m N . 8.8.5

 

5.89.5 3:32.58:— E 833.8.» 3359...: Mae—5. 85.9580 5.38.80 new-:8.—

e $35.
a

100
REGRESSION AND PROBIT ANALYSES

An Overview

Three dependent variables (degree of dramatic change, degree of related
change, and degree of pervasive change) were analyzed using hierarchical
regression analysis. Probit analysis (Aldrich & Nelson, 1984) was applied to
the fourth dependent variable namely, convergent change because it was a
discontinuous variable.
1. Change in Product Diversity

Four models were tested for change in product diversity; one each for
the four dependent variables of dramatic change, related change, pervasive
change, and convergent change. Each model was found to be significant
overall. A canonical correlation analysis of the four dependent variables and
fifteen independent variables1 yielded an F = 1.97 that was significant at p

level of .0001 (df = 60, 572).

 

‘ The independent variables included were organizational size
(linear, quadratic, and cubic), performance (linear and
quadratic), current resources (linear and quadratic), long term
resources (linear and quadratic), organizational risk, asset
specificity, environmental munificence, environmental
dynamism, interaction of performance and munificence, and
interaction of performance and dynamism. It may be noted
that canonical correlation analysis for change in product
diversity included all variables except system coupling. The
addition of system coupling resulted in too few cases to
compute canonical correlation results.

101
2. Change in International Diversity

Similar to the change in product diversity, four models were tested for
change in international diversity. The models for dramatic change, related
change, and convergent change were found to be significant overall. The
model for pervasive change was not significant. Therefore, this model was
dropped from further analysis. Consequently, change in international diversity
was further analyzed and interpreted only for three dependent variables. A
canonical correlation analysis for the three dependent variables and all
independent variables was conducted that was found to be significant at p =
.0001 (F = 6.13, df = 45, 2487).

A canonical correlation analysis involving the overall significant models
of change in product and change in international diversity also resulted in a
statistical significant P value of 1.44 (p = .0004, df = 210, 655). This test
controls Type I error for all the analyses conducted in the study.2 The
following section reports the details of the regression and probit analyses for

change in product and international diversity of organizations.

 

2 This analysis also did not include system coupling variable
due to the missing data problem as mentioned previously.

102
Degree of Dramatic Change

Table 7 presents the change in standardized regression coefficients, R2,
and F values, with p values of hierarchical regression analyses on product
diversity.

Equation 1 contained the control variables. The control and the
hypothesized main effects were included in Equation 2. Among others,
Equation 2 included the linear and quadratic effects of organizational size.3
Cubic effect of organizational size was added in Equation 3.‘ An interaction
term was added in the last step (Equation 4). All the equations on change in
product diversity were found to have a significant overall F at p = .00 (Table
7). The R2 for each of the four equations was .03, .096, .109, and .12,
respectively. Change in R2 was .066 for Equation 2 that included hypothesized

main effects and it was significant (F change = 6.84, p< .001).

 

3 The effects of linear and quadratic components were earlier
added in a sequence and were found to be significant (Cohen
& Cohen, 1983).

As discussed in Chapter 3, nonlinear effects of size were
hypothesized. The quadratic terms of performance, or cubic
components of size were added where the results were not
significant, or when the results were not in the predicted
direction. This was done to explore the nature of the
relationship. There was enough sample size for entering these
polynomials in the study (Cohen & Cohen, 1983).

.2.uv3. ”anv? sqqu: :8.uv?.:

88 33.93 H8 8 33> a .. 685%“ 08 356508 Sumo-awe poem—yang.“ .

 

 

 

 

naaamd .....:...:.: axivwd . owSEU n—
:o. m5. coo. - and
*iiQnH 3.3.095 Ewimqo 33:05.0 n—
thu wK.N_ 25.: Rha .3
Q. 2:. 08. 8. NM
.3. mm .0— . Emmgcan—noocwgotom
.........8.~ :82 886 Ba 583590
3%.2 3.83.. *ssmv. 88:8:an §=oE:E_>=m
an. em. 3. 3:35 mooSomom Eco-r was
on: «Q. omr mooSSom .EB- was
mm. aw. ow. 383m moosofim E250
3.- B.- mwr 38:83“ 22:6
38.n- ......_m.m- 32.. Bang 3% Haggis
$2 .84 :8.- 85 38:35ch
nawad *o_.. .33 .. gauge-tom Ecomuaumcawuo
tam: timmr 1.3m.- 5. 8:85—52 3288:835—
aanfifi ***v~. 323.3. 3.307 bmowmooam Homm<
3.- 8.- 8.- 8. Eu 3833580

n n n in
v .5233 m 5:35 m .8235 H 5933 838:3 83:82.5

 

£235 335.5 E owns—U 9385.5 we 8..qu .5. $9222 :cmmuéhwom .329335 be 338%

h 035.

m2

 

104

Equation 4 included interaction and had a change in R2 of .011 that was
significant (F change = 9.29, p< .001). Apparently, the amounts of R2 in the
above equations seem to be low, but they were found to be reasonable when
they were compared to similar studies. For example, in an equation that
involved similar variables, Wiersema and Bantel (1992)5 found an R2 of .06.

A recent meta-analysis of the 52 planned change studies found all interventions
resulting in an R2 from .0004 to .05 for different dependent variables
(Robertson et al., 1993).

Table 8 presents the results of hierarchical regression analyses for
international diversity. Similar to the analyses in change of product diversity,
four regression equations were analyzed in international diversity. The results
revealed that in the case of international diversification, the main effects
Equation 2 yielded change in R2 = .09 that was significant (F change = 10.44,
p< .001). The change in R2 contribution by the interaction, however, was not
significant. Because the overall equations of product and international
diversities were significant, individual hypotheses regarding the relationship of

independent variables and the degree of dramatic change were tested.

 

5 Wiersema and Bantel (1992) published in Academy 9f
Management JQQmal, 35: 91-121.

.Ofi. "V an ”no. "V Am... 30. "V a: 38. "V at;
83 33.92 28 mm 33> a a ”3282 2.: 356580 =382w2 uoamufiucfim .

 

 

 

 

S. .........8.2 3.3.2 - 08:6 m
8. no. mo. - ~m<
*iifim.w sinuofi .a iiwvd 60. n—
m§.~_ 9%: 3mg: mmw.m .8
:- 2. mo. 8. NM
8. fl Em_Em§D*oo§:_-_otom
.........8.~ .........m~.~ :85 38 3:88:85
9.7 *8. 1.8.- 82:3:an BEES-fig
8.: .........:.~ 2.3.8.: 888 839.2 :5: 83
***mm.Nu timmN- ***3.N- moo-Somew— E-BH. was
2. mo. 5.- 333cm moosomom 82.50
S .- 2.. Sr 82.58% 52.50
:88- :43- .........8. 88$ 35 3:88:85
*8; LS $1.3.- 86 3:83:85
2... no. no. cognac-tom 3.232590
.8: *8. :8; 3.- 858% .32
8.- 8.- 5.- 8.- :2: 88885

n n n ....n
v comma—um m comma—um m 5335 fl .8935 SEE-:3 885865

 

 

£235 3533.25 5 own—EU gag—En .8 vogue: .Su macs—23: game-Eva _339_83= no 338%

w 035.

m2

106

1. Organizational Performance. The relationship of organizational
performance to the degree of dramatic change was hypothesized as follows:

Hypothesis la: The lower the level of performance, the more
dramatic the organizational change.

This hypothesis postulated that organizational performance was negatively
related to the degree of dramatic change. It suggested that organizations were
inclined to make more dramatic change in times of low performance, while
they changed in a less dramatic (incremental) fashion, when performance was
high. The results presented in Table 7 revealed a significant negative linear
relationship of performance and the degree of dramatic change, as
hypothesized. The association of performance with the degree of dramatic
change in international diversity was not significant, as revealed by Table 8.
2. Organizational Size. Organizational size was hypothesized to be related to
the degree of dramatic change as follows:

Hypothesis 1b: Small and big organizations change in a less

dramatic fashion, whereas medium-sized organizations change in a

more dramatic fashion.
It was expected that as organizations grew in size, the size of the step of change
varied in a nonlinear fashion. An inverted U-shaped relationship was expected
between the size and degree of dramatic change in organizations. The results

presented in Table 7 for product diversity and in Table 8 for international

diversity showed a U-shaped relationship that was contrary to the hypothesis.

107

Therefore, the cubic term was added to find out if the relationship was of cubic
nature. The results presented in Equation 3 (Tables 7 and 8), showed
significant relationships of size with change in both kinds of diversities. These
results were graphed in Figure 5 for product diversity and in Figure 6 for
international diversity. The results showed that the small and large
organizations changed in a less dramatic (incremental) manner, whereas the
medium-sized organizations changed in a more dramatic manner, as
hypothesized. However, the results showed that in the case of very large
organizations, change again became more dramatic in nature. The results were
similar in the case of both types of diversities. The curve exhibited a very
interesting and complex nonlinear relationship between size of an organization

and the degree of dramatic change in organizations.

Doyo. of Drmtlc Change

0.16
0.15
0.14
0.13
0.12
0.11
111
0.09
000
0.07
0.06
0.05
0.04
0.03
0.02
0.01

-0.01

’ ' 108
Figure 5
Organizational Size and Degree of Dramatic Change

in Product Diversity

 

 

 

 

 

 

olo'iIo‘zlobldtloislohlobloblobli
ans 015 025 (135 0.45 (155 (155 0.75 ass as
' when»

 

Dow.- ol Oran-tic Ghana.

0.19
0.18
0.17
0.16
0.15
0.14
1113
0.12
0.11
0.1
0.09
0.00
0.07
0.06
0.05
0.04
0.03
0.02
0.01

109
Figure 6

Organizational Size and Degree of Dramatic Change

in International Diversity

 

l
u
u

'1
a

1‘
'- u

L‘

t‘

t‘

74

 

'4

fl

9

 

"P l l 1 1 1 l 1
olmlazloslmlosloslor
ans ms-ozsnasmsassass

lo'nlo‘sli
muse-95

 

110

3. Organizational Resources. Organizational resources were expected to be
related to the degree of dramatic change as follows:
Hypothesis 1c: Organizations with extreme levels of organizational
resources will make more dramatic change, whereas organizations
with moderate levels of organizational resources will make less
dramatic change.
The relationships of organizational resources with product and international
diversity can be seen in Tables 7 and 8, respectively. The above hypothesis
was tested with respect to current resources and long-term resources. The
association of neither of these two variables was found to be significant in the
case of change in product diversity. Similarly, current resources were not found
to significantly related to the degree of dramatic change in international
diversity. However, the relationship of long-term resources with change in
international diversity was significant. Both the linear and quadratic
components were significant, therefore, the relationship was nonlinear, as
hypothesized. It suggested that a low level and a high level of long term
resources were related to more dramatic change, and a moderate level of

resources was related to less dramatic change. It was represented graphically in

Figure 7.

00". o! Dram-Clo Chang.

1 l 1
Figure 7

Organizational Resources and Degree of Dramatic Change

in International Diversity

 

-0.17

-02

l

-o:1—

«is—

-0.6

l

 

 

 

1 1 1 1 1 1 1 1

 

W Rm (Lon Tum)

 

112

4. Environmental Dynamism. Environmental dynamism was hypothesized to
be related to the degree of dramatic change as follows:

Hypothesis 1d: The lower the environmental dynamism, the less

dramatic the change; the higher the environmental dynamism, the

more dramatic the change.
The results presented in Table 7 for change in product diversity and in Table 8
for change in international diversity revealed that dynamism was a significant
correlate of these changes. The relationship was positive in product
diversification, as hypothesized. However, dynamism was found to be
negatively related to change in international diversity. The variables of
organizational risk, asset specificity, and environmental munificence on degree
of dramatic change were used as control variables in the study. The
relationships of organizational risk with degree of dramatic change in product
and international diversities were not significant. Similarly, asset specificity
and environmental munificence did not significantly relate to international
diversity, as expected. However, asset specificity was found to have a
statistically significant positive relationship with the degree of dramatic change
in product diversity of organizations (Table 7).

Environmental munificence though insignificant in the first equation,
gained significance when other variables were added. This showed that the

presence of other factors, particularly environmental dynamism, influenced the

association of munificence and change in product diversity. In international

113

diversification, the presence of both munificence and dynamism caused a
multicollinearity problem. Taking munificence out of the equation made the
coefficient of dynamism significant. However, the sign of the relationship was
opposite to the hypothesized relationship, as mentioned above.

5. Interaction. Organizational performance and environmental dynamism were
expected to interact as given below:

Hypothesis 1e: In an organization with lower performance, higher

environmental dynamism will lead to a less dramatic change as

compared to lower environmental dynamism.
Equation 4 in Tables 7 and 8 shows the interactive effects of organizational
performance and environmental change for product and international diversities,
respectively. The interaction added an R2 of .011 to the degree of dramatic
change in product diversity that was significant (F change = 9.29, p< .001).
These results were graphed in Figure 8. There was no support for the
interaction hypothesis in international diversification.

In a nutshell, the results showed organizational performance, size,
organizational resources, environmental dynamism, and asset specificity were
correlates of dramatic change. At least some of them had different
relationships with change in product diversity and change in international

diversity in organizations.

1 14
Figure 8

Interaction of Performance and Dynamism, and Degree of Dramatic

Change in Product Diversity

 

 

o
:11
q:
CD
(13
GP
eh
GP
1::
1-
1:1

Douro. of Dramatic Chang.
(Thomandcl
61
T

 

 

 

 

0 5 I) 15 20 25 30 35 4O 45
Enviumnhl Dynamism _
D [ow Mam + M Mum

115
Degree of Related Change

The analyses of the degree of related change were similar to the earlier
variable, the degree of dramatic change. The results of these analyses are
presented in Tables 9 and 10 for change in product and international diversities,
respectively. Three equations were used in the regression analysis. Equation 1
included the control variables. Main effects were added in Equation 2,‘
whereas interaction was added in Equation 3. Equation 1 included the variables
of average industry performance and organizational resources. Both the current
and long-term resource variables were included as control variables. The
Equation 2 was significant overall (Table 9, F = 2.02, df = 10,721, p = .03)
and it added an R2 of .03 (F change = 2.22, p< .05) in product diversity.
Similarly, the equation added R2 of .02 that was significant (Table 10, F =
1.90, p< .05) in the case of change in international diversity. The Equation 3
was significant overall, but the interaction was not significant in either of the
two diversities. Because the Equation 2 was overall significant for change in
product and international diversities, testing of the individual hypotheses related

to the degree of related change was undertaken.

 

The linear effects of performance were not found to be significant;
therefore, the quadratic term of performance was added to understand the
relationship, as discussed in Chapter 3 and previously in this chapter.

AZ. "V ax ”m6. NV Am... So. "V a: 38. "V Q31.
“:3 =S.o§ :8 mm 33> a ..
cor—82 2: 358508 guy-Ewe: 3:58:53: .

 

 

8. .3: - 8:96 m
8. 8. - R:
*8.N *NO.N 2 A nu
out: 5.2 5.: :8
8. 8. 8. N:
3.2- ou:ooE:=2*oo::E-5tom
8.- 8.- 83:3 0:: 38:85:85
2. on. 86 3:23:35
3.8.: :3: 8.2:: 8588:»: 3:88:35
86. 3R6. cage-tom 3:332:90
wad *wor 8:35:22 §:oE:o:>:m
8. 8. 5358:: :82
8. 8. :5. 3:88:35
8. oo. 5. moo-580m 85,—. w:3
3. 3. x8. 88:80m Eat—6
8. S. S.- ..5: 93:5 8:83:

n n .2:
m 8:25 N 8:33 : 8:85 8.::::> 38:83::

 

$225 35822.: a 8:26 23:: u: :28: a: 33:3. 5:828: 32883: :: $58:

a 2:3.
2 fl

.2."me HQHVQ... SQHVQ: ”_oo.flVQ..:.z..

3.2 =S-oE :8 mm 02.? a ..

8:88 93 3:055:50 86838 EMF—3:35.: .

 

 

8. *8; - “.335 "—
8. mo. - W3
*3; *3; *3.~ "—
33: 3.: 823 a:
No. 8. 8. NM
36 oo:8¢_:=2*8::§otom
tan? 3%.. 83% Ba 3:83:35
2:. tom. 86 3:83:35
3.7 *8. cognac-tom 3:335:90
and- No.- oo:oo_-::32 35:85:35
3.- 8.- 53:38: :82
8. 8. :2: 3::::::3w5
3.- 3.- 3.- 88:80m 85,—. w:3
8.- 8.- $3.. moosaom Eotso
3. 3. 8. .tom E355 owfio><

n n eh
m 5233 m 8235 fl 5335 83:9; Eugoaoufi

 

haw—35 3:555:85 E own—EU 6353— 3: 8..on :8 xmm_::< Emmanuel _8_5..Eo_= a: 838%

3 03:9

5:

118

1. Organizational Performance. The relationship of organizational
performance and related change was expected as hypothesized below:

Hypothesis 2a: The higher the level of performance, the more
related the organizational change.

Table 10 shows a significant association of organizational performance and the
degree of related change in international diversity. Therefore, the hypothesis
was supported for international diversity. Table 9 exhibits the relationship of
performance on the degree of related change in product diversity. The linear
effect of performance was not significant in product diversification, in contrast
to the hypothesis. The quadratic term was added to test if the relationship was
nonlinear in nature, as some studies reported (e. g., Baum, 1990). The results
produced significant regression coefficients of linear and quadratic terms of
performance (p< .10). These results graphed in Figure 9 suggested a nonlinear
relationship between performance and related change in product diversity that
was not hypothesized. Higher performance led to higher related diversification
after a threshold. Below the threshold, total related change decreased as
performance increased.

Because the relationship of performance and related change in product
diversity was nonlinear, the slope of the curve was graphed. The slope
represented the rate of related change with respect to organizational

performance. The rate of change was equivalent to the first derivative of the

Douro. of Rolatod Chang.

1 19
Figure 9

Performance and Degree of Related Change in Product Diversity

 

4— /
._   /

 

 

 

 

0 n a

.15

-2r

-3-

.4_

-5 1 1 1 1 1 1 1 1 1 1 1
0 010.2 0.3 04 0.5 0.6 07 0.0 09

120

total related change. It revealed a positive relationship between organizational
performance and rate of related change in organizations (see Figure 9).
2. Organizational Size. The association of organizational size and related
change was expected as follows:

Hypothesis 2b: Small and big organizations make less related

change, whereas medium-sized organizations make more related

change.
The hypothesis was supported in international diversification, but it did not get
support in the case of product diversification. Figure 10 shows an inverted U-
shaped curve that provided unequivocal support for the hypothesis for change of
international diversity.
3. Environmental Munificence. The relationship of environmental munificence

was hypothesized as follows:

Hypothesis 2c: The lower the munificence of the industry
segment, the higher the related organizational change.

The hypothesis was supported for change in product diversity, but not the in
case of change in international diversity. The results of change in product
diversity implied that an organization in a declining segment of an industry was
more inclined to search for opportunities in other segments of its broad industry
(2—digit SIC code industry). This would result in more related organizational
change.

Hypotheses concerning organizational risk and asset specificity to related

121

change were not supported in the study. Current and long-term resources were
employed as control variables. The control model was not significant in
product diversification. However, it was significant (Table 10, F = 2.74, df =
3,1992; p = .04) in the case of international diversification. The regression
coefficient of current resources was negatively related to related change in

international diversity (p< .05) as shown in Table 10.

Door-o of Rolatod Change

0.13
0.12
0.11
0.1
0.09
0.00
0.07
0.06
0.05
0.04
0.03

0.01

-0.01
-0.02

122

Figure 10

Organizational Size and Degree of Related Change

in International Diversity

 

 

 

 

 

123
Degree of Pervasive Change

Table 11 reveals the results of regression analysis for product
diversification, whereas Table 12 presents the results for change in international
diversity. Four equations were used in each type of diversity. Equation 1 was
the control model containing current resources and long-term resources. In the
next step, organizational performance was added in Equation 2. Organizational
size was added in Equation 3. All these equations employed the first measure
of the degree of pervasive change. This measure involved using change in
annual net property, plant, and equipment as the measure of capital
expenditures of the company. Equation 4 used the second measure of capital
expenditures that was annual change in gross property, plant, and equipment of
the company in computing pervasive change. Because the data for these
equations came from different sources, system coupling could not be added to
the other equations. Therefore, system coupling was analyzed as a single
variable using least squares regression, rather than in a hierarchical regression
model.

Table 11 displays the regression coefficients, probabilities, R2 and other
details of the results with respect to change in product diversity. The second
regression equation yielded an R2 of .06 and change in R2 = .02, that was only
marginally significant at p = .07. The variable included was organizational

performance. Adding linear and quadratic form of size increased

o_.nv : x unquv : ..

8." v : 2.. :8." v : 2....

$3 #3123 :8 mm a o

3:82 03 3:221:08 3:838 336335 ..
233595 3:98 .8 233:. 2: 3 383 88w 5 omSEo 3:: 8:33.: 33 0330
03:33: 20:3 633:3 3:38 :52? 8 5:33.. 5 E85 £83 _3 5 23:3 385%:

2: .33 8:33:33 3:93 a: 233:. a .3 33.3 3: :_ 033:0 3:: 8338 093:0 933.8: .

 

 

- to: .m 5.3 - 0320 m
- no. mo. - «Md
same .126 x2.m 3.86 m
02; oflfi 33m $1~ .8
no. 8. co. 3. «M
3m. 3:950 883m
*8. 3:33 3% 38:33:90
5%.. 36 35:33:30
S .- x3 .- 8:338:05 35:33:30
:3 .- 5.3. :2 .- 88:80: 8:8. 33
8.- S .- S .- 82:80: 82.50

m n a .3
:3 5:35 m 3:35 m 3:35 fl 3:33....— mo_3::> 33:83::

 

...§w8>5 625.5 E 3:30 ozmatom a: as :8 $9232 33853— 39:33.55 a: 338m

3 035,—.
wfi

125
R2 to .09 (F = 3.19, df = 5,159, p = .01), adding R2 = .03 that was

significant (F change = 3.10, p< .05). The analysis of similar three equations
using the second measure of the degree of pervasive change (i.e., change in
gross property, plant, and equipment of the company) did not reveal any
significant results. The results of Equation 4 (Table 11) using second measure
of pervasive change (change in gross property, plant, and machinery as a
measure of capital expenditures) that included only system coupling were
significant overall at p = .03 (F = 4.92, df = 1,100).

Table 12 reports the results of regression analysis of pervasive change in
international diversity. The analysis was conducted similar to change in
product diversity using four equations as explained above. None of the
equations was significant; therefore, no further analysis was conducted for
change in international diversity of organizations, as noted previously in the
overview.

Because the overall Equations 3 and 4 (Table 11) in product diversity
were significant, the individual hypotheses for change in product diversity were
tested.

1. Organizational Performance. Organizational performance was expected to
be related to the degree of pervasive change as hypothesized below:

Hypothesis 3a: The lower the level of performance, the more
pervasive the organizational change.

0:.uv : : ”Suv : ..
S.uv : .2. :8.nv : ........

:8: 8:03: 5.: m: a o

8:08: 03 3885000 538:8: 838383: ..
0:38:88 3:83 .«0 08808 0:: 8 30:8 808 8 08:80 3:: 8.8808 8B 0330
03820: 0:08: 083:3 8:300 80:3,»: 0: 5:20: 8 8080 .880 :3 8 0883.» 80:88:
05 83 8:88:88 3830 :0 08808 3 8 30:8 :0: 8 0830 33: 88808 0830 03820: .

 

 

- 0m; 5.: - 093:0 m
- 8. 8. - mad
8. 8.: 0:. F. m
8.: 5.: $3 08.: :0
co. 3. 8. 5. NM
5. 3:300 806%
8.. 83cm 08m 35:38:80
um. 08m 35:38:30
.2 .- 8.- 0038580: 35:38:90
mo. 3. ca. 80883: 8:0,: m5:
no: mo: 2 .- 80:58: 80:30

0 n m an
e 5:33: m 8833 a 3:35 : 5:30: 8389 80:08:08:

 

2.538080 .302502: 5 0:030 038.80.: .8 00:80 :0: £8.22: 588.80% 80:80:33: :0 838%

N— 030,—.
cm:

127

The relationships of organizational performance with changes in product
diversity presented in Table 11 (Equation 2) were marginally significant using
the first measure of pervasive change. The second measure did not yield
significant results of performance on pervasive change in product diversity.
The marginally significant results in Table 11 provide a weak support for the
hypothesis that under the conditions of low performance, the organizational
change is more pervasive.
2. Organizational Size. The relationship of organizational size and pervasive
change was hypothesized as follows:

Hypothesis 3b: Small and big organizations make less pervasive

change, whereas medium-sized organizations make more pervasive

change.
Table 11 reveals the results of organizational size on the degree of pervasive
change in product diversity. Organizational size added an R2 of .03 (F change
= 3.10; p = .05) to the degree of pervasive change in product diversity using
first measure (Table 11). However, the regression coefficients were not
significant. The nature of the relationship was also in contrast to the
hypothesis. An inverted-U shaped curve was expected, but the above
relationship represents a U-shaped curve (Figure 11). The U-shaped
relationship implied that small and large organizations made more concerted

change and that medium—sized organizations changed in a piecemeal manner.

Using second measure of pervasive change, there was no significant effect for

Down. at Perv-chr- Change

128

Figure 11
Organizational Size and Degree of Pervasive Change

in Product Diversity

 

§§§

I70 ~

8§§§§§
ITIII

 

 

 

cassassassé
i

 

 

 

 

129

size on pervasive change in product diversity.
3. System Coupling. The association of system coupling was hypothesized as
given below:

Hypothesis 3c: The tighter the system coupling in an organization,
the more pervasive the organizational change.

Table 11 exhibits the relationship of system coupling on pervasive change in
product diversity using second measure of the degree of pervasive change. The
relationship of system coupling on the degree of pervasive change varied with
the measure of pervasive change. Using the change in gross property, plant,
and equipment as capital expenditure of the company in computing the measure
of pervasive change, the hypothesis was supported (Table 11, Equation 4). In
fact, the single variable of system coupling accounted for .05 percent of the
variance in product diversity change. It had a standardized coefficient of .22
significant at p = .03 level (F = 4.92, df = 1,100).

These results need to be interpreted in view of that (i) there were two
variables of pervasiveness of change thereby giving two chances for testing a
relationship; (ii) the system coupling variable could not be entered in one
equation with other variables. This was because there were not enough
companies available with data on system coupling, performance, and other

variables in the study.

l 30
Convergent Change

Tables 13 and 14 report the results of the probit analyses of product and
international diversities, respectively. These analyses were conducted in a
simultaneous fashion rather than in a hierarchical manner. The hypothesized
variables were analyzed using model 1 (Tables 13 and 14). It included the
variables of organizational performance, size, resources, and environmental
munificence.

The likelihood ratio test in probit analysis serves the purpose of overall F
test of ordinary least squares regression. The likelihood ratios for change in
product and international diversities were high. Therefore, both models were
significant. In the case of product diversification, the likelihood ratio was
25.40, with an R2 = .05, and the model made 65 percent predictions correctly
(Table 13, model 2).7 The model of international diversification had a
likelihood ratio of 117.31, and an R2 of .18 (Table 14, model 2). It made 71
percent predictions correctly. Because the overall models were significant, the

results of individual hypotheses on convergent change were tested.

 

Cragg-Uhler R2 was used. It serves the function of R2 of ordinary least
squares regression.

2.uv a x ”me-av a ..

~0."V a a... u2.5.".V a .12..

9 a $5629 5 owcfiuvnbggm .
386580 8355.... .

 

 

8. 3. a. 522-380
3.3 3.5 amok. 23m EOE—8:4
mmdw smmgfl wwd- 5&7 ugmcoo
053 *mm. 333m eunuch—atom Ecoufimfiwuo
94 3.2. 9V.— ........2 . 8=3cm§2 §=oE:§>=m
an; ac.- vmr co.- vogcm 80.583— .th was
2 ._ 2. he; 2. 88.58% Ebb was
mhd- N0.- 2.v- mo.- uouasvm 82:83— 28.50
m“ .5 mu. 2 .w mm. «8583— 2:05.50
3 .- No.- 2 .- No.- Bugvm ofim again—«who
S. 3. mo. 3. ofim Handgun—«moo
8.2:- .32. E. 8. Badges... 3.83330
8. No. 8. 8. £0583 29.2
awn—m .U.m .gmsm ..U.m
N 682 :08: 8:353 302385

 

haw—35 8:8...— 5 3:25 Bow—9:80 .8. 39:52 ~59:— ue 338%

2 035,—.

gm

o_.nv a x HQHV a ..
~O.HV & 8.. Seq-IIV a 3......

Q a 03.2039 s owfinovubsufim .

3565000 093808 03 Km .

 

 

3. 3. t. .3 520.330
3.3: 2...: m3: .3 25 82293
5.3 m;- 33 e;- 32 03 5.80
3. U3. Bed 83. 3:83.830
3 .m- .8: R.“- .3.- 3m 85550.03 382.3530
3.3 .52.- 3.3 33.. 33 2.33.. 8§E§2 35525-3
mom- 8.. $3. .8- Sa- .8.- 8.3% .838.“ E...- 3.3
mm.» 5. 33 .3. was .3. 82883 .58. 3.3
93 3. m: 3.- 83 3. 8.8% .8593 .550
n2- 3.. 83- 3.- 30.n- :..- 82:83 2230
3.3- .3... 3.- 3.- 3.- 3.- 823% 03m 3833530
23 5: e. :. om. 3. 35 0833530
02.- .3: 3.» .8. 8.7 .3.- 85.583 3833530
3. .8. 3. .8. 3. .:. 30588 .32
32m .0.m .033 .0.m 2.2m ..0.m
m 382 N 082 3 032 8§E> 288385

 

haw—35 30050530— £ 09.25 Bow-5.500 00.. 33:23. “30.:— H0 338%

3 030,—.
N2

133

1. Organizational Performance. Organizational performance was hypothesized
to have the following relationship:

Hypothesis 4a: The higher the level of performance, the higher
the probability of convergent organizational change.

Tables 13 and 14 display the relationship of organizational performance and
convergent change. Organizational performance was not significant for change
in product diversity. Therefore, the hypothesis of linear positive effect was not
supported. Because the hypothesis was not supported, and some previous
studies have reported nonlinear relationships, a quadratic term was added to test
the possibility of nonlinear effect. Model 2 (Table 13) and Figure 12 present
the results. Both the linear component and quadratic components of
organizational performance were significant (p < .05), and the elasticities were
high.

Concerning change in international diversity (Table 14, Model 1),
organizational performance was found to be significant (p < .05). However, its
sign was in contrast to the hypothesized effect. A positive linear relationship
was expected, but the results showed a negative one. Because the findings
were unexpected, and earlier studies found nonlinear (and contradictory)
effects, Model 2 was analyzed by adding quadratic term of organizational
performance. The linear component was significant at p < .05, whereas the

quadratic term was significant at p < .001 level. Using the quadratic term as a

Convorgont Change

134
Figure 12

Performance and Convergent Change in Product Diversity

 

:i M
-2_ MW

 

 

 

 

143 I 533 I aialdulnfolzt‘nlziolza'olainl
3.43 .143 “.43 15.43 3.43 23.43 27.43 31.43 35.43
0 Commentary 0 lid-clamp

135

control, the linear term was related to convergent change in a positive
relationship, as hypothesized. The results have been presented in Figure 13.
The results revealed a U-shaped curve in the case of product diversity, but an
inverted U-shaped curve in the case of international diversity.

Because the relationship was nonlinear, the first derivative of the model
was graphed to understand the slope of the curve that represented rate of
convergent change in organizations. The above hypothesis was supported in the
case of rate of convergent change in product diversity. That is, higher
performance led to more probability of convergent change in product diversity
(Table 13, Figure 12). However, the relationship was opposite in case of

international diversification (Table 14, Figure 13).

Convergent Change

136
Figure 13

Performance and Convergent Change in International Diversity

 

 

 

 

 

'3 I I I r I I I
1.43 I 543 I 143 I 1143 I I74: I an I 2543
143 7 43 I143 15.43 343 2:443
WW Mam

137

2. Organizational Size. Organizational size was hypothesized to be related to
convergent change as follows:

Hypothesis 4.2. Small and big organizations have a higher

probability of changing in a convergent fashion, whereas medium-

sized organizations have a lower probability of changing in a

convergent fashion.
A nonlinear relationship was expected between organizational size and
probability of convergent change in organizations. The results presented in
Table 13 for product diversity and in Table 14 for international diversity did
not show support for the hypothesized relationships. As the hypothesis was not
supported, a cubic term was added to explore the nature of relationship between
size and convergent change (Model 3). Even the cubic relationship of size was
not significant for change in product diversity. With respect to international
diversification, there was marginal significance of a cubic relationship (Table
14: Equation 3, and Figure 14). The relationship suggested that the probability
of convergent change in international diversification increased from medium to
large organizations, as expected; however, interestingly, the probability of
convergent change in small organizations was lowest. The results were

marginally significant and unexpected. Nevertheless, they point to possibilities

of a fascinating relationship.

138
Figure 14

Organizational Size and Convergent Change in International Diversity

 

 

Convoraont Change
I
8
I

 

 

 

139

3. Organizational Resources. Organizational resources were hypothesized to
have the following relationship with convergent change:

Hypothesis 4c: Organizations with moderate levels of

organizational resources are more likely to change in a convergent

mode, whereas organizations with extreme levels of organizational

resources are less likely to change in a convergent mode.
The above hypothesis was tested with respect to current resources and long-
term resources. As Table 13 exhibits, neither of these two types of resources
was found to have a significant relationship with change in product diversity.
Similarly, current resources did not have a significant association with
international diversity, as Table 14 reveals. However, the estimated coefficient
of long-term resources on change in international diversity was significant
(Table 14, p< .05). As Table 14 shows, estimated coefficients of both the
linear and quadratic components were significant at p < .05 (Model 2),
providing support for the hypothesis. It showed that a moderate level of long-
term resources was related to a higher probability of convergent change (Figure

15). At the same time, a low level and a high level of long—term resources was

associated with lower probability of convergent change.

Convorgont Change

10

140
Figure 15

Organizational Resources and Convergent Change

in International Diversity

 

 

 

 

 

I53 I 3'43 I 3'43II343II7343I2143I23'43I23I43I3343I
343 7.43 I143 I343 I343 23 343 27.43 343 3343
- MMlflmll-uuhn)

141

4. Environmental Munificence. The relationship of environmental munificence
was hypothesized as follows:

Hypothesis 4d: The higher the munificencc of the industry

segment, the more the probability of convergent organizational

change.
As Table 13 shows, the results supported the hypothesis for change in product
diversity. The results implied that a company in the growth industry was more
likely to change in the direction it had been following earlier. In other words,
a company that was in a growth stage will keep on making convergent changes
(e. g., increasing its product diversity). The hypothesis was not supported in
international diversification. In fact, the opposite results were received. These
results suggested that an organization had higher probability of reversing its
direction of change in international diversification when its major industry’s
munificence was high.
5. Asset Specificity. Asset specificity was hypothesized to have a relationship

with convergent change as follows:

Hypothesis 4c: The higher the asset specificity in an organization,
the higher the convergent change.

A positive relationship between asset specificity and convergent change was
expected. The results presented in Table 14 demonstrate that the above
hypothesis was supported in international diversification. However, there was

no support in change of product diversity.

142

An overview of the results of the study shows that the dimensions of
change identified in the study were not strongly correlated with one another.
This gives credence to the idea of multidimensionality of change in
organizations. Second, various dimensions of change had different models of
change. In other words, the dimensions either had different correlates, or they
had different patterns of association with the correlates. For example, with
respect to change in product diversity, organizational performance had a linear
negative relationship with dramatic and pervasive change, and a U-shaped
relationship with related change and convergent change. Third, the models of
change for product and international diversification were found to be different.
Therefore, it can be concluded that there is support for the major thesis of the
study that organizational change is not unidimensional, but multidimensional in
nature. However, there was no support for a generic model of change that can
apply to different kinds of changes in organizations. The next chapter
summarizes the results of the study and gives its limitations and further areas

for advancing research on the topic.

Chapter 6

Discussion

Most researchers analyze organizational change in unidimensional terms.
The objective of this study was to examine multidimensionality of
organizational change, and to develop and test models of dimensions of
organizational change. It was expected that different dimensions of change
were weakly correlated with one another and had different patterns of
association with the sources of change. This chapter reviews the findings of the
study, discusses its limitations and contributions, and suggests implications for

future research on decision making and change in organizations.

SUMMARY OF THE FINDINGS OF THE STUDY

The first focus of the study was to identify multiple dimensions of
change and to test relationships among them. Four dimensions of change were
identified: increment, relatedness, pervasiveness, and direction of change.
These dimensions were studied with respect to two kinds of strategic
organizational changes: change in product diversity and change in international
diversity. For change to be regarded as unidimensional, the dimensions must
be strongly correlated. There were no strong correlation coefficients among the
identified dimensions. The dimensions of change in the study were found to be
weakly correlated with one another for change in product and international
diversity. Most of the correlations were below .05, and the maximum

correlation was -.20 that was between the increment of change (dramatic

144

145

change) and direction of change (convergent change). Five measures of the
dimensions of organizational change in the study were factor analyzed. There
was no support for one factor of organizational change. In fact, four factors
were derived from these measures for change in product and international
diversities that gave the best interpretable results.

The above results led to the second focus of the study, which was to
identify various correlates of change dimensions. These correlates act as
sources of change in organizations. Dimensions of change involved various
processes of change. Increment of change was associated with incremental to
dramatic change, whereas low to high related change referred to relatedness of
change. Similarly, piecemeal to pervasive change was linked to the dimension
of pervasiveness, and convergent and divergent changes were reflected by the
direction of change.

The following four broad categories of the sources of change were
identified: organizational performance, history, environment, and strategic
choice. These broad categories yielded the correlates of change that were used
as predictor variables in the study. The four dimensions of change were the
dependent variables. The association of independent variables with increment,
relatedness, and pervasiveness of change was analyzed using hierarchical
regression analysis. The direction of change was a discontinuous variable,

therefore, probit analysis was used to test the relationship of independent

146

variables with convergent-divergent change in organizations. Data on 731 and
855 companies registered on the U.S. stock-exchanges were used for analyzing
change in product and international diversities, respectively. The results have
been summarized according to the independent variables in the study.
Organizational Performance. Organizational performance was hypothesized
to be associated with change in all the four dimensions of change in product
and international diversities. Linear effects were hypothesized in the study.

With respect to change in product diversity, organizational performance
was found to be negatively related to the degree of dramatic change, as
hypothesized. There was marginal support for pervasive change in product
diversity using net assets as the measure of capital expenditures. These results
implied that lower performance was associated with higher dramatic and
pervasive change in product diversity. Lower performance is likely to provide
incentives, political leverage, and institutional legitimacy for making dramatic
and pervasive change in product diversity of organizations. Organizational
performance was not significantly related to dramatic and pervasive change in
international diversity.

Organizational performance had a nonlinear relationship with related
change in product diversity of organizations. The curve was U-shaped (Figure
9). The slope of the curve representing rate of related change was positive.

Therefore, higher organizational performance was associated with a higher rate

147

of related change. In the case of international diversification, performance had
a positive linear relationship with related change, as hypothesized. The
relationship of performance was different in product and international diversities
that rejected the possibility of a generic model in the study, contrary to
expectations.

Organizational performance had a significant relationship with the
direction of change in product and international diversities. An interesting
point was that the nature of the relationship between performance and direction
of change in product diversity (Figure 12) was different from the same in
international diversity (Figure 13). The curve of performance and convergent
change was U-shaped in change of product diversity, whereas it was an
inverted-U shape in change in international diversity. The rate of change of
direction was positive in the case of product diversification, whereas it was
negative in international diversification. The negative relationship in
international diversification was not expected. An improvement in performance
may incline the management to increase product diversity. In other words, in a
high-performance organization, managers keep on increasing the product
diversity. A company making product diversification may focus on its current
markets. It is because the company can take advantage of the opportunities in
the domestic market and it may not go to international markets. Therefore,

organizations may choose between increasing product and international diversity

148

simultaneously.

The above-mentioned nonlinear relationships of performance and
dramatic change should be regarded as tentative because these were speculative
in nature. There was not enough theory to hypothesize these relationships.
Previous studies on change have reported inconsistent results of nonlinear
effects. Some studies reported a U-shaped curve, whereas others found an
inverted-U shape relationship. These studies include analysis of threat-rigidity
effects (Staw et al., 1981) and conservatism in low performing organizations
(Cameron et al., 1981; Whetten, 1987). Other studies include Baum (1990),
Fombrun and Ginsberg (1986), and Ginsberg (1988). The findings of this
study provide a basis for testing these relationships in future research using a
multiple dimensions format.

The results of organizational performance showed that concerning change
in product diversity, organizational performance had a negative linear
relationship with dramatic and pervasive change. However, performance had a
U-shaped relationship with related change and convergent change in product
diversity of organizations. In the case of international diversification,
performance was found to have no relationship with dramatic change and
pervasive change. However, it had a positive linear relationship with related
change and an inverted-U shaped relationship with convergent change in

international diversity. The above results showed that organizational

149

performance had dissimilar associations with different dimensions of change
analyzed in this study. This supports the basic thesis of this study that
organizational change was multidimensional in nature and that various
dimensions of organizational change had different models of change.
Organizational Size. Organizational size was found to have an identical
nonlinear relationship with increment of change (dramatic change) in product
diversity (Figure 5) and international diversity (Figure 6), as hypothesized.
Size had a marginally significant nonlinear association with the direction of
change (convergent change) in international diversity when the cubic terms
were entered. However, as expected, the relationship of size to dramatic
change (Figures 5 and 6) was qualitatively different from its relationship to
convergent change (Figure 15) in international diversity. The association of
size with convergent change in product diversity was not significant.
Organizational size was not found to have a significant relationship with
related change in product diversity. However, in the case of international
diversity, the relationship was an inverted-U shape, as expected (Figure 10).
The results revealed that organizational size had an inverted-U shape
relationship with pervasive change in product diversity using measure 1 (Table
11, Figure 11). This was in contrast to the hypothesized effect. The above
results in product diversification implied that small organizations changed in a

pervasive manner and medium-sized organizations changed in a piecemeal

150

manner. These results can be attributed to the ease of small companies to
orchestrate changes because of their characteristics of low formalization,
entrepreneurial culture, and relatively fewer components.

Change, creativity, and innovation decrease as organizations increase
their sizes. The finding that larger companies were more likely to make
pervasive change can be understood based on the ideas expressed in industrial
organization economics (Porter, 1980; Scherer, 1980). These ideas, as
mentioned in Chapter 3, show that larger organizations have more resources in
absolute terms. These resources may help them make changes in many
variables simultaneously and take risks. It may be mentioned that statistical
controls in the regression equation included resources in relative terms (i.e., the
measures of long-term and current resources were ratios in this study). These
organizations have a better ability to raise capital and can invest in capital
expenditures, research and development expenditures, and advertising
expenditures simultaneously. They, therefore, are in a better position to make
pervasive changes. Larger organizations may not have the choice to make
piecemeal changes because they have more complex structures and have
developed interdependencies among different functions and variables. The
relationship of size with pervasive change in international diversity was not
significant.

The above results of organizational size revealed that its relationship with

151

related change and convergent change was not significant in product diversity.
However, size had a nonlinear relationship with dramatic and pervasive change
in product diversity. In the case of international diversity, size was
significantly related to pervasive change, however, it had an inverted-U shaped
relationship with related change. Similarly, size had different types of
nonlinear relationships with dramatic change and convergent change in
international diversity. These results showed that different dimensions of
change had vastly different relationships with organizational size. It supported
the basic thesis of the study that organizational change dimensions were
multiple and distinct from one another. The results, however, rejected the idea
of a generic model of change because most of the relationships of size were
different in the two types of change studied.

System Coupling. System coupling was hypothesized to have a significant
relationship with pervasive change. Its relationship was found to vary
depending on the measures of pervasive change in organizations. System
coupling alone explained variance of .05 in pervasive change when
pervasiveness was measured employing change in net property, plant, and
equipment as a measure of capital expenditures of the company. System
coupling was not hypothesized to significantly relate to other dimensions of
change. Its above-mentioned significance for pervasive change in product

diversity supported the contention that there were different correlates of

152

different dimensions of change.

Environmental Munificence and Dynamism. The environmental variables of
munificence and dynamism were significantly associated with change in product
diversity, as hypothesized. A munificent environment was related to a lower
degree of related change, and a higher probability of convergent change in
product diversity of organizations. Organizations in highly dynamic
environments made more dramatic change in product diversity as compared to
organizations in less dynamic environments that made more incremental
changes. These findings were also consistent with rationale developed by
Ginsberg (1988) on changes in organizations.

In contrast to change in product diversity, environmental variables were
not found to have a positive relationship with change in international diversity.
Environmental munificence had a negative association with convergent change
in international diversity, contrary to the hypothesis. Munificence did not have
a significant relationship with related change in international diversity.
Environmental dynamism was negatively associated with the degree of dramatic
change in international diversity, in contrast to the hypothesis. These results
were attributed to two possibilities: the way these variables were measured in
the study and the change in product and international diversities may be
competitive in nature. The environmental variables were measured with respect

to their 4-digit SIC code industry in the U.S., whereas international diversity

153

referred to diversification in international markets. The opposite results
indicated that an organization had higher probability of reversing its direction of
change in international diversification in view of higher munificence of its
major industry at home. As a post-hoc interpretation of the results, an
organization may reverse its globalization if its major industry in its home
country offers good opportunities. Moreover, a highly dynamic environment at
home is associated with dramatic changes in product diversity, whereas change
in international diversity may be ignored or made incremental. The contrasting
results in product and international diversification are similar to the relationship
of performance and convergent change. These results seem very interesting
and need to be explored in future research.

The results of environmental munificence and dynamism further attest to
the idea that different dimensions of change have different models of change in
organizations. These results support the notion of multidimensionality of
change in organizations and reject the generic model of change.

Strategic Choice Variables. The strategic choice variables of organizational
risk, asset specificity, and organizational resources were analyzed in the study.
Their effect in the study was relatively limited. The relationship of
organizational risk was hypothesized with related change only. It was not
found to be significant. In other dimensions, where it was used as a control

variable, the results also showed no significant association. The results were

154

similar in change of product and international diversity.

The results of asset specificity were similar to those of organizational
risk in the case of related change. A significant relationship of asset specificity
with related change was hypothesized. However, it was not significantly
related to related change in any of the two kinds of diversities studied.
However, asset specificity was found to have a significant relationship with the
degree of dramatic change in product diversity. The results implied that higher
asset specificity led to more dramatic change in product diversity. Specific
assets have low propensity for redeployment in new circumstances. Therefore,
managers need to develop a rationale, a master plan, and political support for
the changes they want to bring about. The organization has to be prepared for
a dramatic move to accept the losses that may be incurred due to change. All
these call for a dramatic change in the organization. The relationship of asset
specificity was not significant in any of the other cases, as expected, except in
convergent change in international diversity. In this case, the relationship was
significant but the size of the effect was negligible (.01).

Organizational resources were hypothesized to be associated with the
degree of dramatic change and convergent change in organizations. It was
interesting to find that current resources did not have a significant relationship
with any dimension in either of the diversities. The only exception was related

change in international diversity that was not hypothesized. Current resources

155

were generally not critical in strategic decisions such as change in product or
international diversity, because these decisions required long-term
commitments. Long-term resources were not significantly related to the
increment and direction of change in product diversity, but they were
marginally significant in pervasive change. However, long-term resources did
have a significant relationship with increment of change and direction of change
in international diversity. The nature of these relationships of long-term
resources with international diversification was as hypothesized.

In short, organizational performance and size were the factors that were
associated with most dimensions of organizational change in product and
international diversities. However, their relationships with different dimensions
were not the same, as expected. Tighter system coupling had a positive linear
relation with the pervasive change in product diversity using change in net
assets as the measure of capital expenditures. With respect to product
diversification, environmental munificence had a negative relation with related
change, but it had a positive relationship with convergent change in
organizations. Environmental dynamism was found to have a positive
relationship with dramatic change. These relationships of environmental
variables with change in product diversity differed from those with the change
in international diversity. Strategic choice variables were found to have limited

association. Organizational risk and asset specificity did not have a significant

156

association with related change. Long—term resources had a significant relation
with increment of change and a marginally significant association with the
direction of change in international diversity. The results of the study led to the
following two conclusions:
1. Different dimensions of change were weakly correlated with one

another. Various dimensions of change had different models of

change in organizations. Therefore, different dimensions need to

be distinguished from one another, and change needs to be treated

as multidimensional.
2. The pattern of association between dimensions and their correlates

also differed with the type of variable involved in change.

Change in product and international diversities had different

models. Therefore, the study rejected the notion of a generic

model of change.

LIMITATIONS OF TIIE STUDY
No research study is without limitations, and this one was no exception.
One has to make choices and compromises while planning and executing any
research study because of trade-offs in research.
First, the study considered only two variables of change, four dimensions

of change relevant to these variables, and a limited number of sources of

157

change as their correlates. Moreover, the considered dimensions may not be
the most critical of all the dimensions of change. The four dimensions were
selected because of their relevance to the two types of change analyzed in the
study. The objective was to test multidimensionality of change and it could be
accomplished using these dimensions. However, future research can consider
other dimensions of change relevant to the types of change studied here.
Further studies can also extend the model of organizational change considering
other sources of change (e. g., management tenure, top-management turnover).
Other variables of change (e. g., change in structure, change in technology, and
other strategic changes) can also be studied in future research. An analysis of
other changes in future research will help researchers to understand models of
change in a variety of organizational changes.

Second, the study was limited with respect to availability of data. Data
on international diversity and product diversity were available for the periods
1984-1990 and 1978-1990, respectively. A planning horizon of 4-5 years was
normally considered in previous studies (Finkelstein & Hambrick, 1990).
Therefore, these data provided a reasonable period of time for analyzing
changes in organizations. However, availability of data for more years would
prove useful in such studies, particularly in case of international diversification.
Data on some variables did provide a real constraint (e.g., in case of pervasive

change and system coupling). Data on these two variables were collected from

158
different databases, COMPUSTAT II, and Value Line DataFile. Individually

these data were available for a sufficient number of companies, but the number
of overlapping companies was relatively small. These ranged from 66 to 165,
depending on the number of other variables included in the regression equation.
This problem constrained the statistical power and, thus, the probability of
finding positive results when present. Future research would benefit from
exploring alternative sources of data. Obviously, insufficient data points were
available for time-series analysis. Time-series analysis can add dynamism to
the study in future research.

Third, research was based on secondary data. Though COMPUSTAT II
was a reliable and appropriate database for the study (Davis & Duhaime, 1992),
these data were not collected by the investigator for this research. Secondary
data may have some known and unknown limitations that may have crept into
the study. There were advantages of data being objective, but it did lack the
perspectives of the managers that can be a rich source of insight, especially in
case of explaining unexpected findings. Following up this study with field
studies of organizations can enrich the understanding of the phenomenon of
organizational change.

The study was descriptive rather than prescriptive in nature. The study
showed different processes of change and their correlates. No attempt was

made in the study to find out the relative success of different strategies of

159

change under the conditions they were occurring. Future research involving
performance as a dependent variable can throw light on this aspect of processes
of change. Then the issue of recommending what kind of change ought to be
made under what conditions both within and outside of organizations can be
addressed.

The study employed most of the commonly used measures of the
variables. Alternative measures of some variables revealed that the findings
depended upon the variables employed. Therefore, it will be interesting to use
alternative measures of the variables to check if the results of the study hold
with other measures. Alternately, one can explain the results by the nature of
the measures used in this study. Some variables such as organizational
performance, environment, and system coupling are themselves
multidimensional in nature. Measures using other dimensions or alternative
conceptualizations can increase confidence in the findings of the study, and/or
improve R2. Consequently, further research can contribute to the development
of a more rigorous model of multidimensionality of organizational change.

This process may also help to generate improved measures that can be used for
advancing other streams of research in organizations.

Though overall F values, t values, R2, and so on, were mostly significant
in the regression equations of dimensions of change, amount of absolute R2

varied from .02 to .18. Considering the data used and organizational level of

160

research, these numbers were found to be reasonable with comparable studies.
It was hard to find studies with comparable independent and dependent
variables. However, a study by Wiersema and Bantel ( 1992) on top-
management team demography published in Aeaglemy ef Management Jeamal
considered a control model with the variables of prior organizational
performance, organizational size, top-management team size, industry growth,
industry performance, and industry concentration on strategic change measured
as absolute change in product diversity. The above equation had an R2 of .06.
Another somewhat related study on effects of top-management team on
innovation in organizations published in Strategie Management Jeurnal found an
R2 of .07 to .10 (Bantel & Jackson, 1989). A recent meta-analysis of 52
planned changes in organizations found effect sizes of all interventions
combined that will provide R2 between the range of .0004 to .05 (Robertson et
al., 1993). Though the results were comparable to other studies, they do pose
the question of explaining rest of the variance in different processes of
organizational change. Individual level, economy level, and other
organizational level variables play a part in the process of organizational
change. Therefore, adding other variables is likely to reduce the unexplained
variance. This is an opportunity for future research that can be very fruitful in
improving the explained variance. Some control variables may also be included

that may test the robustness of the findings of this study and/or explain further

161

variance in dimensions of organizational change. For example, the effect of the
business cycle was not considered in this study that may add to the
understanding of how organizations change under different economic
circumstances.

The lagged effects of independent variables were considered on the
dimensions of change. Therefore, data on most independent variables came
from the period 1975-1987, whereas data for dependent variables came from
the period 1978-1990 in the case of product diversification. Because of the
overlap of the data years, there may be questions about the direction of the
relationship. For international diversity, analyses presented involved data on
independent variables for the period 1978-1984 and on dependent variables for
1984—1990. That is to say, the data that were considered for international
diversity came from nonoverlapping years. Further research exploring other
time lags and other time periods may help in improving the confidence in the
findings of the present study.

Finally, this study considered changes at the organizational level. Future
research can extend the study of multiple dimensions of change to other levels

(e.g., individual, group, or social level).

162
CONTRIBUTIONS OF THE STUDY

Despite the above-mentioned limitations, the present study made
theoretical and methodological contributions. The major contribution of the
study was testing multidimensionality of organizational change. Previous
research on organizational change focused on generic change. Most studies
analyzed one dimension or a part of it at a time. Different terms were often
used for the same dimension or new dimensions were added in the literature
without distinguishing them from one another. The processes of change were
studied without any systematic framework. For example, most often
incremental, piecemeal, or convergent changes were studied without making
distinctions among these. Consequently, research findings on organizational
change have not evolved into a coherent body of knowledge.

This study conceptualized organizational change as a multidimensional
phenomenon. Different processes of change were considered along different
dimensions in the study, and multiple dimensions of two types of change were
identified. For example, it conceptualized the dimensions of increment,
relatedness, pervasiveness, and direction of change that involve different
processes of change. The results of the study supported the view that the
dimensions of change were relatively weakly correlated with one another. This
supported the notion of multidimensionality of change in organizations.

Second, the study considered multiple sources of change.

163

Previous research considered sources of change in a mutually exclusive fashion
that resulted in dilemmas among environment, strategic choice, and historical
inertia. More recent theories have conceptualized complementary relationships
among different sources of change. This study included different sources of
change, including environmental, strategic choice, and historical-inertial types.
Moreover, it included the variable of organizational performance, a relatively
understudied source of change. The study found further evidence of the
complementarity of the sources of change in organizations.

Third, the study related sources of change with dimensions (and
processes) of change. Earlier research that treated sources and processes of
change in an isolated manner gave birth to grand theories of incremental, or
quantum, change. Researchers then sought to compare the effectiveness of
these strategies of change without considering the context of change (Miller &
Friesen, 1982). Taking a multidimensional and contextual approach, different
models of organizational change were developed and tested in this study. The
goal was to develop middle-range variance theories (Mohr, 1982) of
multidimensional organizational change. The study recognized that different
processes of change were occurring under different circumstances. For
example, the study found that medium-sized organizations with low
performance and highly dynamic environments changed their product diversity

in a dramatic fashion. Similarly, it was found that higher related change in

164

product diversity was associated with extreme levels of performance and scarce
environments. Small or large (as compared to medium) organizational size,
lower performance, and tighter system coupling were more likely to lead to
higher pervasive change in product diversity of organizations. Organizations
with small size, extreme levels of performance, and munificent environments
were related to convergent change in product diversity. The regression and
probit analyses showed that various dimensions of change had different models
of change. Therefore, different processes of change can no longer be studied
without distinguishing them.

Fourth, the present study found that different types of change (namely,
product diversity and international diversity) had different models. For
example, the relationships of organizational performance, size, munificence,
and dynamism to increment and direction of change were different under
product diversification from those in international diversification. The two
types of changes analyzed in this study were found to follow different models.
Therefore, the study rejected the expected generic model of change.

Finally, the present research developed new measures for the three
dimensions of change: increment, pervasiveness, and direction of change.
These measures, in fact, were different characteristics of the frequency
distribution of the pattern of change in organizations using multiwave data

(Rogosa et al., 1982). These measures can be used and further refined in

165

analyzing other kinds of changes in organizations. More quantitative measures
that assess other aspects of the frequency distribution of the pattern of change
can prove useful. Pictorial methods can also be used for pattern recognition in
how organizations change over time along multiple dimensions. The measure
of international diversity used 2-digit international geographic codes for related
diversification. This provides a new way of measuring related and unrelated
diversification using entropy measures, which advances conceptualization by
Vachani (1991).

In conclusion, the present study focused on the ideas of
multidimensionality of organizational change. It investigated correlations
among different dimensions of change and associated dimensions to their
correlates to develop and test models of change. The result was a
multidimensional framework of change that helps to address the issue of
cumulating research findings on change in organizations (Ginsberg, 1988).
Therefore, the study has contributed toward developing a coherent body of
knowledge on organizational change. The notion of multidimensionality of
organizational change calls for a review of the literature on organizational
change so that researchers can stop mixing the proverbial oranges and apples of
change research. Clustering different research studies based on the dimension
studied can help to contribute to a developing a systematic body of knowledge

on organizational change. Ultimately, the study contributed to an enhanced

166

understanding of how organizations evolve over time. Future research can
consider other dimensions, alternative sources of change, and different kinds of
changes to extend and build on the models presented in this study. The
conceptual framework can also be extended to other levels of analysis (namely,
individual, group, and social level) to inform theory and practice of related
fields of study. Once researchers understand that organizational change is
multidimensional, they can integrate it with multiple sources of change, and
multiple consequences of change in the form of effectiveness of organizations.
In other words, it unlocks immense potential for understanding the evolution of

organizations.

List of References

List of References

Aldrich, H. 1979. W. Englewood
Cliffs, NJ: Prentice-Hall.

Aldrich, H., & Auster, E. 1986. Even Dwarfs Started Small:
Liabilities of Age and Their Strategic Implications.

Reaearch in Qrganizatienal Behavier, 8: 165-198.

Aldrich, J. H., & Nelson, F. D. 1984. Linear; Prebability, Legit,
é; Prebit Mulels. Beverly Hills, CA: Sage Publications.

Allen, B. T. 1988. Managerial Eeenemies. New York: Harper & Row.

Amit, R., & Wernerfelt, B. 1990. Why do Business Firms Reduce
Business Risk? Aeaglemy ef Management Jearnal, 33: 520-533.

Andrews, K. R. 1971. Th n f r r
Homewood, IL: Dow Jones-Irwin.

Ansoff, H. 1991. Critique of Henry Mintzberg’s ’The Design School:
Reconsidering the Basic Premises of Strategic Management.’

Strategie Management Jeurnal, 12: 449-461.

Argyris, C. 1985. Strategy, Change. and Defensive Reutines.
Cambridge, MA: Ballinger.

Argyris, C. 1987. Review Essay: First and Second Firms in
Managing Strategic Change: The Role of Organizational
Defense Routines. In A. Pettigrew (Ed.), The Management ef
Strategicflagge: 342-351. Oxford: Basil Blackwell.

Argyris, C., & Schon, D. 1978. r niz i nal rnin ° A Th
ef Aetien Eersmetive. Reading, MA: Addison-Wesley.

Astley, W. G. 1984. Toward an Appreciation of Collective
Strategy. Aeademy ef Management Review, 9: 526-535.

167

168

Awan, A. A. 1986. Marshallian and Schumpeterian Theories of
Economic Evolution: Gradualism Versus Punctualism. Atlantie

Eeenemic Jeumal, 14: 37-49.

Bantel, K. A., & Jackson, S. E. 1989. Top Management and
Innovations in Banking: Does the Composition of the Top Team

Make a Difference? WM. 10:
107-124.

Barton, S. L. 1988. Diversification Strategy and Systematic Risk:

Another Look. W, 31: 166-175.

Bartunek, J. M. 1984. Changing Interpretive Schemes an
Organizational Restructuring: The Examples of a Religous
Order. A mini r iv i n rl , 29: 355-372.

Bartunek, J. M., & Moch, M. K. 1987. First Order, Second Order,
and Third Order Change and Organizational Development

Interventions: A Cognitive Approach. learnal ef Applied
Behavieral Seienee, 23: 483-500.

Baum, J. A. 1990. Inertial and Adaptive Patterns in
Organizational Change. In L. Jauch & J. Wall (Eds.), Aeademy

ef Management Beat Pamr Ermeedings: 165-169.

Bedeian, A. G. 1990. Choice and Determinism: A Comment. Strategic
Management Jearnal, 11: 571-573.

Bettis, R. A. 1981. Performance Differences in Related and

Unrelated Diversified Firms. Strategie Management Journal,
2: 379-393.

Bettis, R. A. 1983. Modern Financial Theory, Corporate Strategy

and Public Policy: Three Conditions. Aeademy ef Management
Review, 8: 406-415.

Bettis, R. A., & Hall, W. K. 1982. Diversification Strategy,
Accounting Determined Risk, and Accounting Determined

Return. Aeademy ef Management Jeumal, 25: 254-264.

169

Bettis, R. A., & Mahajan, V. 1985. Risk/Return Performance of
Diversified Firms. Maeagemeatfiejenee, 31: 785-799.

Blau, P., & Schoenherr, R. 1971. The 5131191111: ef Qrganizatiene.
New York: Basic Books.

Boeker, W. 1989. Strategic Change: The Effects of Founding and
History. Aeademy ef Management Jeumal, 32: 489-515.

Boeker, W., & Goodstein, J. 1991. Organization Performance and
Adaptions: Effects of Environment and Performance on Changes

in Board Composition. Aeademy ef Management Jedrnal, 34:
805-826.

Bourgeois, L. J. 1981. On the Measurement of Organizational

Slack. Aeademy ef Management Review, 6: 29-39.

Bourgeois, L. J. 1984. Strategic Management and Determinism.

Academy ef Management Review, 9: 586-596.

Bowman, E. H. 1980. A Risk/Return Paradox for Strategic
Management. Slean Management Review, 21: 27-31.

Bowman, E. H. 1982. Risk Seeking by Troubled Firms. 5.19311
Management Review, 23: 33—42.

Bowman, E. H. 1984. Content Analysis of Annual Reports for Corporate
Strategy. I_n_te;f_aee_s, 14: 61-72.

Buhner, R. 1987. Assessing International Diversification of West

German Corporations. Strategie Management Jearnal, 8: 25-37.

Burns, T., & Stalker, G. 1961. The Management ef Innevatien.
London: Tavistock.

Cameron, K. S., Kim, M. U. & Whetten, D. A. 1987. Organizational
Effects of Decline & Turbulence. Administrative Seienee

Qdagerly, 32: 222-240.

Carroll, G. R. (Ed.) 1988. Eeelegical Mflels et Qrganizatiens.
Cambridge, MA: Ballinger.

170

Castrogiovanni, G. 1991. Environmental Munificence: A Theoretical

Assessment. Aeademy ef Management Review, 16: 542-565.

Caves, R. E., & Porter, M. E. 1977. From Entry Barriers to
Mobility Barriers: Conjectural Decisions and Contrived

Deterrence to New Competition. WM

Eeenemies, 91: 241-261.

Cavusgil, S. T. 1980. On the Internationalization Process of
Firms. W, 8: 273-281.

Chaffee, E. 1985. Three Models of Strategy. Aeademy ef Management
Review, 10: 89-98.

Chaganti, R., & Sambharya, R. 1987. Strategic Orientation and

Characteristics of Upper Management. Strategje Management
Jeurnal, 8: 393-401.

Chandler A D 1962 MW
thelnddstrial Enterprise. Cambridge, MA: MIT Press.

Chatterjee, 8., & Wernerfelt, B. 1991. The Link between Resources
and Type of Diversification: Theory and Evidence. Sttategie

Management Jedmal, 12: 33-48.

Child, J. 1972. Organizational Structure, Environment and
Performance: The Role of Strategic Choice. 5m, 6:
2-21.

Child, J ., & Kieser, A. 1981. Development of Organizations Over
Time. In W. H. Starbuck & P. C. Nystrom (Eds.), Handbeek ef

We, vol. 1: 28-64. Oxford, England: Oxford

University Press.

Christensen, H. K., & Montgomery, C. A. 1981. Corporate Economic
Performance: Diversification Strategy Versus Market

Structure. “WM, 2: 327-343.

Clark, B. R. 1972. The Organizational Saga in Higher Education.
Administrative Seience Quarterly, 17: 178-184.

171

Clegg, S. R., & Dunkerley, D. 1980. W
em. London, England: Routledge and Kegan Paul.

Cohen, J., & Cohen, P. 1983. Applieththiple

Rrin rrelinA if hvi i
Hillsdale: Lawerence Erlbaum Associates Publishers.

Cyert, R., & March, J. 1963. A Behavienal flfheet’y Qt the Finn,
Englewood Cliffs, NJ: Prentice-Hall.

Davis, R., & Duhaime, I. M. 1992. Diversification, Vertical
Integration, and Industry Analysis: New Perspectives and

Measurement. Strategie Management Jenmal, 13: 511-524.

Day, D. V., & Lord, R. G. 1988. Executive Leadership and
Organizational Performance: Suggestions for a New Theory and

Methodology. Jenrnal ef Management, 14: 453-464.

Dess, G. G., & Beard, D. W. 1984. Dimensions of Organizational

Task Environments WHEEL! 29:
52— 73.

DiMaggio, P. J ., & Powell, W. W. 1983. The Iron Cage Revisited:
Institutional Isomorphism and Collective Rationality in

Organizational Fields. Amerieanfimiolegiealm, 48: 147-160.

Donaldson, L. 1985. In Defense ef Qrganizatien Theergy; A Reply tn
the Critjes. Cambridge, England: Cambridge University Press.

Doz, Y., & Prahalad, C. K. 1987. A Process of Strategic
Redirection in Large Complex Firms: The Case of
Multinational Corporations. In A. Pettigrew (Ed.),

Managemenmfsmmgicflange: 63-83- Oxford. England:
Basil Blackwell.

Duhaime, I. D., & Grant, J. H. 1984. Factors Influencing
Divestment decision-Making: Evidence from a Field Study.

WWW-193L113!- 5: 301-318.

172

Dunphy, D. C., & Stace, D. A. 1988. Transformational and Coercive
Strategies for Planned Organizational Change: Beyond the

OD. Model. MW, 9: 317-334.

Dutton, J. E., & Duncan, R. B. 1987. The Creation of Momentum for
Change through the Process of Strategic Issues Diagnosis.

WM, 8: 279-298.
Edwards, R. 1979. Win. New Tork: Basic Books.

Fiegenbaum, A., & Thomas, H. 1988. Attitudes Toward Risk and the
Risk-Retum Paradox: Prospect Theory Explanations. Aeademy

of Management Jeamal, 31: 85-106.

Finkelstein, S., & Hambrick, D. 1990. Top-Management-Team Tenure
and Organizational Outcomes: The Moderating Role of
Managerial Discretion. A mini r iv ci nc

484-503.

Fligstein, N. 1985. The Spread of the Multidivisional Form Among
Large Firms, 1919-1979. Am ri i l i vi w, 50:
377-391.

Fligstien, N. 1987. The Interorganizational Power Struggle: Rise
of Finance Personnel to Top Leadership in Large

Corporations, 1919-1979. Amerieaa Seeielegieal Review, 52:
44-58.

Fombrun, C. J ., & Ginsberg, A. 1986. Enabling and Disenabling

Forces on Resource Development. Prmeedings ef the Deeisiea
Seienms Institute Annual Mating: 1249-1251.

Freeman, E. R. 1984. r i n ' A k h l r
Appreaeh. Marshfield, MA: Pitman Publishing.

Friesen, P. H., & Miller, D. 1986. A Mathematical Model of the

Adaptive Behavior of Organizations. Jeurnal ef Management
Stem, 23: 1-25.

173

Geringer, J. Beamish, P. & daCosta, R. 1989. Diversification
Strategy and Internationalization: Implications for MNE

Performance W 10:109-119

Gersick, C. J. G. 1991. Revolutionary Change Theories: A Multilevel
Exploration of the Punctuated Equilibrium Paradigm, Aeadennlef

Wm, 16: 10-36.

Ginsberg, A. 1988. Measuring and Modelling Changes 1n Strategy.
Theoretical Foundations and Empirical Directions. Stzategie

ManagemenLLQumal 9: 559-575
Ginsberg, A., & Abrahamson, E. 1986. Qgitimizing Sttategie

h n ' fl en E iv i n Ex rn
misnltants. Paper presented at the Annual Meeting of the
Academy of Management, Chicago.

Ginsberg, A., & Buchholtz, A. 1990. Converting to For-Profit
Status: Corporate Responsiveness to Radical Change. Academy

ef Management Jenrnal, 33: 445-477.

Glassman, R. 1973. Persistence and Loose Coupling. Behaxienal
Seienee, 18: 83-98.

Golembiewski, R., Billingsley, K. & Yeager, S. 1976. Measuring
Change and Persistence in Human Affairs: Types of Change

Generated by OD Designs. Jenrnal ef Appligt Behaveria!
Seienee, 12: 133-157.

Gooding, R., & Wagner, J. 1985. A Meta-Analytic Review of the
Relationship Between Size and Performance: The Productivity
and Efficiency of Organizations and their Subunits.

Administrative Seienee Qnagterly, 30: 462-481.

Goodstein, J ., & Boeker, W. 1991. Turbulence at the Top: A New
Perspective on Governance Structure Changes and Strategic

Change. Aeademy ef Management Jenmd, 34: 306-330.

Govindarajan, V. 1989. Implementing Competitive Strategies at the
Business Unit Level: Implications of Matching Managers to

Strategies. WW, 10: 251-269.

174

Graham, K. R., & Richards, M. D. 1979. Relative Performance
Deterioration: Management and Strategic Change in Rail-Based

Holding Companies. Aeademy ef Management Psegeedings:
108-1 12.

Grant, R. M. 1987. Multinationality and Performance Among British
Manufacturing Companies. Jeurnfl ef Internatinnal Business
Sntdies, 18: 79-89.

Grant, R. M., Jammine, A. P. & Thomas, H. 1988. Diversity,
Diversification, and Profitability among British

Manufacturing Companies, 1972-1984. Aeademy ef Management
Jeumal, 31: 771-801.

Greenwood, R. 1984. Instrumental Budgeting: Antecedents of
Change. Jeurnal ef Publie Peliey, 4: 277-306.

Greiner, L. E. 1972. Evolution and Revolution as Organizations

Grow. Harvam Business Review, 50: 37-46.

Greiner, L., & Bhambri, A. 1989. New CEO Intervention and

Dynamics of Delibrate Strategic Change. Strategie Management
Joanna, 10: 67-86.

Gupta, A. K., & Govindarajan, V. 1984. Business Unit Strategy,
Managerial Characteristics, and Business Unit Effectiveness

at Strategy Implementation. Academy ef Management Jnurnal,
27: 25-41.

Hambrick, D. C. 1984. Taxonomic Approaches to Studying Strategy:
Some Conceptual and Methodological Issues. ,leurnal ef

Management, 10: 27-42.

Hambrick, D. C., & D’Aveni, R. 1988. Large Corporate Failures as
Downward Spirals. mini r iv ’
1-23.

Hambrick, D. C., & Finkelstein, S. 1987. Managerial Discretion: A
Bridge Between Polar Views of Organizational Outcomes. In B. M.

Staw & L. L. Cummings (Eds.), Reseaseh in Qrganizatienal
Behavier, vol. 9: 369-406, Greenwich, CN: JAI Press.

175

Hambrick, D. C., & MacMillan, I. C. 1985. Efficiency of Product
R & D in Business Units: The Role of Strategic Context.

Aeademy ef Management Jeurnal, 28: 527-547.

Hambrick, D. C., & Mason, P. 1984. Upper Echelons: The
Organization as a Reflection of its Top Managers. Aeademyef

Management Review, 9: 193-206.

Hannan, M., & Freeman, J. 1977. The Population Ecology of
Organizations. AmerjeanlgunnaLoifieeielegy, 83: 929-964.

Hannan, M., & Freeman, J. 1984. Structural Inertia and

Organization Change. Anseg'ean Seeielegical Review, 49:
149-164.

Hannan, M. ,& Freeman, J. 1988. The Ecology of Organization
Mortality: American Labor Unions, 1836- 1985. Mean

Jeu ma! ef Secinlegy, 94. 25- 52.

Hannan, M., & Freeman, J. 1989. Qrganizatienal Eeelegy.
Cambridge, MA: Harvard University Press.

Harrigan, K. R. 1981. Deterrents to Divestiture. Academy ef
Management lnumal, 31: 306-323.

Harrigan, K. R. 1985. Vertical Integration and Corporate
Strategy. Aeademy nf Management Jeurnal, 28: 397-425.

Hatten, K. J., Schendel, D. E. & Cooper, A.C. 1978. A Strategic
Model of the U.S. Brewing Industry: 1952-1971. Aeademy ef

Management Jeurnal, 21: 592-610.

Hedberg, B. L. 1981. How Organizations Learn and Unlearn. In P. C.

Nystrom & W. H. Starbuck (Eds.), Handbeek ef Qrganizatien
Design, vol. 1: 3-27. Oxford, England: Oxford University Press.

Hedberg, B. L., Nystrom, P. C. & Starbuck, W.H. 1976. Camping on
Seesaws: Prescriptions for a Self-Designing Organization.
A mini tr tiv i nc u rl , 21: 41-65.

176

Hill, C. W., & Hansen, G. S. 1991. A Longitudinal Study of the
Cause and Consequenses of Changes in Diversification in the

U.S. Pharmaceutical Industry 1977-1986. Stsategie Management
Jnurnal, 12: 187-200.

Hill, C. W., & Snell, S. A. 1988. External Control, Corporate
Strategy, and Firm Performance in Research-Intensive

Industries. Wad, 9: 577-590.
Hinings, C. R., & Greenwood, R. 1988. Mynamieuflfitnategie

Change. Oxford, England: Basil Blackwell.

Hofer, C. W. 1975. Toward a Contingency Theory of Business
Strategy. Aeademy nf Management Jnumal, 18: 784-809.

Hoskisson, R. E., & Galbraith, C. S. 1985. The Effect of Quantum
Versus Incremental M-Form Reorganization on Performance: A
Time-Series Exploration of Intervention Dynamics. LnurnaLnf

Management, 11: 55-70.

Hoskisson, R. E., & Hitt, M. A. 1990. Antecedents and Performance
Outcomes of Diversification: A Review and Critique of

Theoretical Perspectives. Jnurnal nf Management, 16:
461-509.

Hoskisson, R. E., Hitt, M. A. & Hill, C.W. 1991. Managerial Risk
Taking in Diversified Firms: An Evolutionary Perspective.

Qrganizatinn Seienee, 3: 296-314.

Hrebiniak, L. G., & Joyce , W. F. 1985. Organizational Adaption,
Strategic Choice, and Environmental Culturalism.

Administrative Seienee Quartesly, 30: 336—349.

Huff, A. S., & Huff, J. O. 1990. Stress and Ineztia in Stnategy
Refnrmulatinn. Working Paper, University of Illinois.

Huff, A. S., Huff, J. O. & Thomas, H. 1989. W

Resnnnse tn Qrganizatinn Stress, Cumulative Inema and
W. Working Paper, University of Illinois.

177

Jacquemin, A. P., & Berry, C. H. 1979. Entropy Measure of
Diversification and Corporate Growth. W

Indusm'al Eennnmies, 27: 359-369.

Johnson. G. 1987. Strategicflaumndthcflanagmflum.
Oxford, England: Basil Blackwell.

Johnson, G. 1988. Rethinking Incrementalism. Stnategi_c_M_anage_1nent
Jnmal, 9: 75-92.

Keats, B. W., & Hitt, M. A. 1988. A Causal Model of Linkages
Among Environmental Dimensions, Macro Organizational

Characteristics, and Performance. Aeademy nf Management
Jnurnal, 31: 570-598.

Kelejian, H., & Oates, W. 1974. lntrnduetinn tn Ecnnnmetries. New
York: Harper & Row.

Kiel, L. 1989. Nonequilibrium Theory and Its Implications for

Public Administration Wm 49:
544-551.

Kiesler, S., & Sproull, L. 1982. Managerial Response to Changing
Environments: Perspectives on Problem Sensing from Social
Cognition. mini r iv i n erl , 27: 548-570.

Kim, W. C. 1989. Developing a Global Diversification Measure.
' , 35: 376-383.

Kim, W. C., Hwang, P. & Burgers, W.P. 1989. Global
Diversification Strategy and Corporate Profit Performance.

Strategie Management Jnurnal, 10: 45-57.

Kimberly, J. R., Evanisko, M. J. 1981. Organizational Innovation:
The Influence of Individual, Organizational, and Contextual
Factors on Hospital Adoption of Technological and

Administrative Innovations. Aeademy nf Management Jnunrtal,
24: 689-713.

178

Kleiner, B., & Corrigan, W. 1989. Understanding Organizational

Change. WWW 10:
25-31.

Laughlin, R. C. 1991. Environmental Disturbances and
Organizational Transactions and Transformations: Some

Alternative Models. Qggamuonjmdjes, 12: 209-232.

Lawrence, P., & Lorsch, J . 1967. WWW
Managing Differentiatinn and Integratinn. Boston, MA:

Graduate School of Business Administration, Harvard
University.

LeClere, M. J. 1992. The Intrepretation of Coefficients in Models
with Qualitative Dependent Variables. Degisinnsficienees, 23:
770-776.

Lieberson, S., & O’Conner, J. F. 1972. Leadership and
Organizational Performance: A Study of Large Corporations.

American Snciolngical Review, 37: 117-130.

Lindblom, C. 1968. The Pnliey Making Prneess. Englewood Cliffs,
NJ: Prentice-Hall.

Luffman, G. A., & Reed, R. 1982. Diversification in British

Industry in the 1970’s. Strategin Management Jnurnfl, 3:
303-314.

Lundberg, C. C. 1984. Strategies for Organizational
Transitioning. In J. R. Kimberly & R. E. Quinn (Eds.), Managing

Qrganizatinnal Transitinns: 60-82. Homewood, IL: Irwin.

March, J. G., & Olsen, J. P. 1976. Ambiguity and Cbnine in
Qrganizatinns. Bergen, Norway: Universitetsforlaget.

March, 1., & Simon, H. 1958. Qrganizatinns. New York: Wiley.

Marglin, S. 1974. What the Bosses Do: The Origins and Functions
of Hierarchy in Capitalist Production. Review nf Radieal

Pnlitiea! Eennnmies, 6: 60-112.

179

Mascarenhas, B. 1989. Strategic Group Dynamics. Aeademy nf
WW, 32: 333-352.

McKelvey, B. 1982. MW. Berkeley, CA:

University of California Press.

Meyer, A. D., Brooks, G. R. & Goes, J. B. 1990. Environmental
Jolts and Industry Revolutions: Organizational Responses to

Discontinuous Change. StnateginManagemenununnal, 11: 93-110.

Meyer, J. W., & Rowan, B. 1977. Institutional Organizations:
Formal Structure as Myth and Ceremony. Ameriean Jnurnal nf

muggy, 83: 340-363.

Miles, R. E., & Snow, C. C. 1978. Qrganizatinnal Stzategy,
Stnueture, and Prncess. St Louis, MO: McGraw-Hill.

Miller, D. 1981. Towards a New Contingency Approach: The Search

for Organizational Gestalts. Mnflbdanagensenmudies,
18: 1-26.

Miller, D. 1982. Evolution and Revolution: A Quantum View of

Structural Change in Organizations. W
m, 19: 131-151.

Miller, D. 1987. The Genesis of Configuration. Aeademy nf
Management Review, 12: 686-701.

Miller, D., Droge, C. & Toulouse, J. 1988. Strategic Process and
Content as Mediators Between Organizational Context and

Structure. Aeademy of Management Jnumal, 31: 544-569.

Miller, D., & Friesen, P. H. 1980a. Archetypes of Organizational
Transition. Administrative Seienee Quagerly, 25: 268-299.

Miller, D., & Friesen, P. H. 1980b. Momentum and Revolution in

Organizational Adaptation. Aeademy nf Management Seieng,
23: 591-614.

180

Miller, D., Friesen, P. H. 1982. Structural Change and
Performance: Quantum Versus Piecemeal-Incremental

Approaches. Aeademy nf Management Jnumal, 25: 867-892.

Miller, D., & Friesen, P. H. 1984. Wm
View. Englewood Cliff, NJ: Prentice-Hall.

Mintzberg, H. 1973. Strategy-Making in Three Modes. Califnmia
Manges-mm, 16: 44—53.

Mintzberg, H. 1978. Patterns in Strategy Formulation. Management
Science, 24: 934-948.

Mintzberg, H. 1990. The Design School. Reconsidering the Basic

Premises of Strategic Management. StzategnMLagement
Jnu rn,al 11:171-195.

Mintzberg, H. 1991. Learning 1, Planning 0: Reply to Igor Ansoff.
Strategie Management Jnurnal, 12: 463-466.

Mintzberg, H., & McHugh, A. 1985. Strategy Formation in an
Adhocracy. Administnative Seienee Quanterly, 30: 160-197.

Mintzberg, H., Raisinghani, D. & Theoret, A. 1976. The Structure
of Unstructured Decision Processes. Administrative Seieng
Quantenly, 21: 246-275.

Mintzberg, H., & Waters, J. 1982. Tracking Strategy in an

Entreprenurial Firm. Aeademy nf Management Jnumm, 253:
465-499.

Mitroff, I. I., & Kilmann, R. 1976. On Organizational Stories. In
R. Kilmann, L. Pondy, & D. Slevin (Eds.) The Management nf
rniz'Din°V1 rin
Imnlemenmtinn: 189-208. New York: American Elsevier.

Moch, M. K., &Bartunek, J. M. 1990. Cr reating Alternative
W. New York. Harper Publishers.

181

Moch, M. K., Buchko, A. & Rubin, P. 1988. A Simulation-Based
Time-Series Policy-Capturing Methodology For studying
Recurring Decision Making in Organizations. In R. L. Cardy, S. M.

Puffer & J. M. Newman (Eds.) Ad vanees in Infnrmation
WW, vol. 3: 221-244, Greenwich, CN: JAI

Press.

Moch, M. K, Dass, P, Rubin, P., & Mendez, D. 1991.

r n1z 1 ii 11 mi Pr ces
Prnblems a Snlutinns fn: Analyzing flfime- Sen'es Data. Paper

presented at the Annual Meeting of the Academy of
Management, San Francisco, CA.

Moch, M. K., & Morse, E. V. 1977. Size, Concentration, and
Organizational Adoption of Innovations. Ameriean

Sneinlngieal Review, 42: 716-725.

Mohr, L. 1982. Explaining Qrganigtinnal Behavinr; The Limits and
Pnssinilities nf Thens'y and Reseaseh. San Francisco:

Jossey-Bass.

Montgomery, C. A. 1982. The Measurement of Firm Diversification:

Some New Empirical Evidence. Aeademymiflfianagesnegllnumal,
25: 299-307.

Montgomery, C. A. 1985. Product-Market Diversification and Market
Power. Aeademy nf Management Jnumal, 28: 789-798.

Montgomery, C. A., & Hariharan, S. 1991. Diversified Expansion by
Large Established Firms. Jnurnal nf Eennnmie Behavinr d;
Qrganizatinn, 15: 71-89.

Montgomery, C. A., & Thomas, A. R. 1988. Divestment - Motives and
Gains. W31, 9: 93-97.

Montgomery, C. A., & Wernerfelt, B. 1988. Diversification,

Ricardian Rents, and Tobin’s q. Randlmmnalnfficnnnmies,
19: 623-632.

Morgan, G. 1986. Images nf Qrganizatinns. Beverly Hills: Sage

Publications.

182

Murray, E. A. 1978. Strategic Choice as a Negotiated Outcome.
Managemticieuce, 24: 969972.

Nadler, D. A., & Tushman, M. L. 1988. What Makes for Magic
Leadership? Ens-tune, 117: 261-262.

Nadler, D. A., & Tushman, M. L. 1989. Organizational
Framebending: Principles for Mananging Re-orientation.

Aeademy nf Management Exeeutive, 3: 194-202.

Nadler, D. A., & Tushman, M. L. 1990. Beyond the Charismatic
Leader: Leadership and Organizational Change. Califnrnia

Management Review, 77-97.

Nelson. R., & Winter. S. 1982. WWW
flange. Cambridge, MA: Belknap Press.

Oster, S. 1982. Intra-Industry Structure and the Case of

Strategic Change WW 64:
376- 383.

Palepu, K. 1985. Diversification Strategy, Profit, Performance

and the Entropy Measure. Str rategie Man nagement Jnum ml,a 6.
239- 255.

Palmer, D. ,,Friedland R. J,e,nnings P. D. ,& Powers, M. E. 1987.
The Economics and Politics of Structure: The Multidivisional
Form and the Large U. S. Corporation. Adnnnistnatiyejeienee
Quanemly, 32. 25-48.

Perrow, C. 1986. Cnmnlex Crganizatinns; A Critieal Essay. Glenoe,

IL: Scott, Foresman & Co.

Pettigrew, A. 1973. fifhe Pnlities nf Qrganizational Denisinn
Making. London, England: Tavistock.

Pettigrew, A. 1985. The Awakening Giant; Cnntinuity and Change in
1C1. Oxford, England: Basil Blackwell.

Pettigrew, A. (Ed.) 1987. The Management nf Strategie Change.
Oxford, England: Basil Blackwell.

183

Pfeffer, J., & Salancik, G. 1978. The Eatennai Cnntrnl nf

Qrganizatinns; A Resnuree Demndenm Eerstmctive. New York:
Harper and Row.

Porter, M. E. 1980. Cnmmtitive Stnategy; Teehninues fnr Analyzing
Industries and Cnmnetitnrs. New York: Free Press.

Porter. M. E. 1985.1ommmmamand
Sustaining Sumrinr Perfnrmanee. New York: Free Press.

Porter, M. E. 1990. The Cnmpetitive Advantage nf Natinns. New

York: Free Press.

Quinn, J. B. 1980. Strategies fnr Change; Lngieal
Inerementalism. Homewood, IL: Irwin.

Ramanujam, V., & Varadarajan, P. 1989. Research on Corporate
Diversification: A Synthesis. r t i M n
10: 523-551.

Ravenscraft, D., & Scherer, R. 1987. Mergers, Sell-nffs, and

Eennnmie Effieieney. Washington, DC: The Brookings
Institution.

Robertson, P. J., Roberts, D. R. & Porras, J. I. 1993. The Impact
of Planned Organizational Change: Results of a Meta-Analysis.

Aeademy nf Management Qrganizatinnal Develnnment and Change
Newsletter: 4—6. Winter.

Rogosa, D., Brandt, D. & Zimowski, M. 1982. A Growth Curve
Approach to the Measurement of Change. Psyehnlngieal
Bulletin, 92: 726-748.

Romanelli, E., & Tushman, M. L. 1986. Inertia, Environments, and
Strategic Choice: A Quasi-Experimental Design for

Comparative-Longitudinal Research. Management Seienee, 5:
608-621.

Rumelt, R. 1974. Strategy, Stmeture, and Eennnmie Perfnr_rnanee.

Boston, MA: Harvard Business School Press.

184

Schendel, D. E. ,& Hofer, C. W. 1979. Str rategie Management; A
MW Boston MA: Little

Brown and Co.

Schendel, D., & Patton, G. 1976. Corporate Turnaround Strategies:
A Study of Profit Decline and Recovery. lnunnflnffienenfl

Management, 3-11.

Scherer, F. 1980. Industrial Market Sti'uetum and Eennnmie
Perfnrmang. Boston, MA: Houghton-Mifflin.

Schwenk, C. R. 1989. Linking Cognitive, Organizational and
Political Factors in Explaining Strategic Change. Jnumal nf

Management Studies, 26: 177-187.

Scott, W. R. 1987. Qrganizatinns; Ratinnal, Natural, and an
Systems. Englewood Cliffs, NJ: Prentice-Hall.

Sharfman, M., Wolf, G., Chase, R., & Tansik, D. 1988.

Antecedents of Organizational Slack. Aeademy nf Management
Review, 13: 601-614.

Singh, J. V. 1986. Performance, Slack, and Risk Taking in

Organizational Decision Making. Aeademy nf Management
Jnumai, 29: 562-585.

Singh, J. V., Tucker, D. & Meinhard, A. 1988. _A_r_e_h’_nl_unt_any

Qrganizatinns Stmeturally Inent? Exnlnring an Assumptinn

in Qrganization Eenlngy. Paper Presented at the Annual Meeting
of the Academy of Management, Anaheim, CA.

Smircich, L., & Stubbart, C. 1985. Strategic Management in an
Enacted World. Aeademy nt Management Review, 10: 724-736.

Snow, C. C., & Hrebiniak, L. G. 1980. Strategy, Distinctive
Competence, and Organizational Performance. Administratiye

Selma/Quarterly, 25: 317-336.

Standard & Poor’s COMPUSTAT 11 Services, Inc., 1986. Business
Infnrmatinn; COMPUSTAT II. New York: Standard & Poor’s.

185

Starbuck, W. H. 1976. Organizations and Their Environments. In M. D.

Dunnette (Ed) WWW
Psychnlngy: 1069- 1123. Chicago, IL: Rand McNally.

Starbuck, W. H., Greve, H. & Hedberg, B. L. 1978. Responding to
Crisis: Theory and the Experience of European Business.

.lnurnal nt Business Administratinn, 9: 111-137.

Starbuck, W. H., & Hedberg, B. L. 1977. Saving an Organization from
a Stagnating Environment. In H. B. Thorelli (Ed.), Su'ategyi
Structure = Performance: 249-258. Bloomington, IN: Indiana
University Press.

Staw, B. M. 1981. The Escalation of Commitment to a Course of

Action. Aeademy nf Management Review, 6: 577-587.

Staw, B. M., & Ross, J. 1987. Behavior in Escalation Situations:
Antecedents, Prototypes, and Solutions. In B. M. Staw & L.

L. Cummings (Eds.), Researeh in Qrganizatinnal Behaving, vol. 9:
39-78. Greenwich, CN: JAI Press.

Staw, B. M., Sandelands, L. E. & Dutton, LE. 1981.
Threat-Rigidity Effects in Organizational Behavior: A
Multilevel Analysis. A ministrativ i nce u erl , 26

501-524.

Staw, B. M., & Szwajkowsky, E. 1975. The Scarcity-Munificence
Component of Organizational Environments and the Commission

of Illegal Acts. Administrative Seienx Quanterly, 20:
345-354.

Stinchcombe, A. 1965. Organizations and Social Structure. In J.

G. March (Ed.), Handhnnk nf Qrganizatinns: 153-193. Chicago,
IL: Rand NcNally.

Thompson, J. 1967. Crganizatinns in Aetinn. New York:
McGraw-Hill.

186

Tushman, M. L., & Anderson, P. 1986. Technological
Discontinuities and Organizational Environments.
A mini r iv i n rl , 31:439-465.

Tushman, M. L., Newman, W. H. & Romanelli, E. 1986. Convergence
and Upheaval: Managing and Unsteady Pace of Organizational

Evolution Wm 29: 29-44

Tushman, M. L., & Romanelli, E. 1985. Organizational Evolution: A
Metamorphosis Model of Convergence and Reorganization.

Administrative Seienee Quanterly, 7: 171-222.

Vachani, S. 1991. Distinguishing between Related and Unrelated

International Geographical Diversification: A Comprehensive
Measure of Global Diversification. Jnunnal nf Internatinnai

Business Studies, 22: 307-322.

Value Line DataFile, 1991. AJlsefiLManuai. New York: Value Line
Publishing Inc.

Van de Ven, A. H., & Poole, M. 1988. Paradoxical Requirements for
a Theory of Organizational Change. In R. Quinn and K.

Cameron (Eds.), Baradnx and Transfnrmation; anard a Thengt

nf Change in Qrganizatinn and Management: 19-67. Cambridge,
MA: Ballinger.

Van Horne, J. C. 1983. Finaneiai Management and Pnliey. 5th ed.
Englewood Cliffs, NJ: Prentice Hall.

Virany, B., & Tushman, M. L. 1986. Top Management Teams and
Corporate Success in an Emerging Industry. W

Business yenturing, 1: 261-274. .

Watson, G. 1969. Resistance to Change. In W. Bennis, K. Benne, &
R. Chin (Eds.), iifhe Planning ni Change: 488-498. New York:

Holt, Reinhart and Wilson.

Watzlawick, P., Weakland, J. H. & Fisch, R. 1974. Change;
Prinil PrlmFrmin PrlmR li.New
York: W.W. Norton.

187

Weick, K. E. 1976. Educational Organizations as Loosely Coupled
Systems. Administrative Seienee Quanteriy, 21: 1-19.

Weick, K. E. 1979. flfhe Sneiai Bsyehnlngy nf Qrganizing. Reading,
MA: Addison-Wesley.

Wernerfelt, B. 1989. From Critical Resources to Corporate Strategy.
inumal nf General Management, 14: 4-12.

Whetten, D. 1987. Organizational Growth and Decline Processes.

Annual Review nf Sneinlngy, 13: 335-358.

White, K. J., Haun, S. A. & Horsman, N. G. 1987. SHAZAM User’s
Referenee Manual. Vancouver, Canada: University of British
Columbia.

Wiersema, M. F ., & Bantel, K. A. 1992. Top Management Team
Demography and Corporate Strategic Change. Aeademy nf

Management Jnumai, 35: 91-121.

Williamson, 0. 1975. M rk 11 Hi r hies: An 1
Anti-tnust Implieatinns. New York: Free Press.

Williamson, 0. 1985. The Eennnmic Institutinns nf Capitalism.

New York: Free Press.

Yip, G. S. 1982. Barriers tn Enti'y. Lexington, MA: D.C. Heath and
Company.

Zammuto, R. F. 1988. Organizational Adaptation: Some Implications
of Organizational Ecology for Strategic Choice. ,lnumal nf

Management Studies, 25: 105-119.

Zucker, L. G. 1983. Organizations as Institutions. In S. B.

Bacharach (Ed.), Researeh in the Sminlngy nf Qrganiaatinns:
Vol. 2: 1-47. Greenwich, CN: JAI Press.

   

MICHIGAN smr: UNIV. LIBRARIES
WIWWWNW“WillVlNlllHlllllMHl
31293009048178