1' r1 E5538

..-o

                   

llllllllllllllllllllllllllllllllllllllll

193 01694 3536

This is to certify that the

dissertation entitled

CONTEXT-STRATEGY-STRUCTURE-PERFORMANCE
IN LOGISTICS: A CONTINGENCY APPROACH

presented by

Steven Ray Clinton

has been accepted towards fulﬁllment
of the requirements for

Ph.D. degreein Business Administration

 

 

Major fes or

23/ 7/ a]

MSU is an Afﬁrmative Action/Equal Opportunity Institution 0- 12771

 

LIBRARY

Mlchlgan State
Unlverslty

 

 

 

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

 

DATE DUE I DATE DUE DATE DUE

 

AUG 2 4 lQLB

;~ 3‘1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1M campus-m4

 

CONTEXT -STRATEGY-STRUCTURE-PERFORMANCE IN LOGISTICS:
A CONTINGENCY APPROACH

By

Steven Ray Clinton

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Marketing and Supply Chain Management

1 997

ABSTRACT

CONTEXT-STRATEGY-STRUCTURE-PERFORMANCE IN LOGISTICS:
A CONTINGENCY APPROACH

By

Steven Ray Clinton

Contingency-based environment - strategy - structure - performance paradigms
have been investigated extensively in the organizational science and strategic
management ﬁelds. The general premise of such research is that performance is
contingent upon a ﬁrm’s environmental, strategic, structural ﬁt. The purpose of this
research was to study such contingent relationships within the context of logistical
operations. The objectives were to develop and examine a context-strategy-structure-
performance model for logistics based on established constructs. This Logistics
Contingency Model was then tested for the existence of organizational relationships
within logistics operations.

North American World Class Logistics Research case ﬁrms were examined to
test the model. In total, data for fifty-seven ﬁrms was available for analysis. In
addition, thirty-nine of the ﬁrms completed a follow-up questionnaire providing
longitudinal strategy and performance data. T-tests and regression analysis were used
in evaluating selected relationships.

Context was represented by (1) environmental dynamism and hostility and
(2) logistics information technology operating systems and planning systems. Strategy

was examined using the Bowersox and Daugherty (1987) typology. Structure was

 

 

 

evaluated using established contructs representing formalization, centralization,

integration, and span of control. Performance was evaluated using perceived measures

of logistics competency relative to competitors. The major ﬁndings of the research are:

1.

Higher levels of environmental dynamism and hostility resulted in higher levels
of formalization. This ﬁnding is contrary to long-established empirical results
in organizational science.

Information technology positively impacts logistics structure. Highly proﬁcient
operating and planning systems extend logistical span of control. The former is
also associated with greater levels of formalization.

Highly dynamic environments stimulate extensive use of integrative committees
within logistics organizations.

Externally-oriented logistics strategy ﬁrms were found to be signiﬁcantly more
likely to use integrative committees in comparison to non-externally—oriented
strategy ﬁrms.

Consistent with the organizational science literature, logistics structure alone did
not impact performance. However, the corollary relationship - that structure
would mediate contextual and strategic inﬂuences on performance - was not

supported in this research.

Copyright by
STEVEN RAY CLINTON

1 997

Dedicated to my departed grandmother, Laura Clinton. Her saying,
“Self-praise is no endorsement.” was never truer.

ACKNOWLEDGMENTS

Throughout my doctoral program I noted the great relief and satisfaction that
other students expressed upon ﬁnishing the process. Nearly every student uttered
something to the effect, “It was all worth it.” I always wondered how they could be
that certain? I was positive that I would feel differently. After all, the tremendous loss
of income, the steadily mounting student loans, the stress and frustration of everything
associated with a doctoral program -- none of that could be so easily erased simply by
ﬁnishing the dissertation and program? Okay - so it appears I was wrong again!
Perched on the exit side of this process, I must say, things look different. And I have a
number of people to thank for helping me to this point.

First and foremost, I would like to thank Dr. Donald Bowersox. Inclusion in
his World Class Logistics research was the single most inﬂuential part of my doctoral
program. It has given me great insight into how one should conduct meaningful
research. Participation in the research provided an opportunity to develop my
academic career through publications and conference attendance. As co-chair of my
dissertation, his frequent admonishments to “get going” provided the necessary “spur-
in-my-saddle.” I leave Michigan State with nothing but the utmost respect for Dr.

Bowersox’s professional achievements and dedication to the ﬁeld of logistics.

vi

My other co-chair, Dr. Cornelia Droge, assisted me greatly throughout the
program. More so than any other faculty member, she witnessed my weaknesses, low
points and anger. Through it all she assisted me with well-reasoned assessments of
situations and encouragement based upon my strengths. The quality of this dissertation
is directly related to her substantial time and effort involved in critiquing draft after
draft. Her thoroughness is appreciated and admired.

I also thank my other committee members: Dr. Roger Calantone and Dr.
Bixby Cooper. As doctoral program advisor, seminar professor, co-author, and
storyteller, Dr. Calantone has had multiple positive inﬂuences on me during the
program. Dr. Cooper’s laid-back style and perceptive comments and questions were
appreciated throughout the World Class Logistics research and this dissertation process.

I would be remiss if I did not also thank Dr. Stanley lFawcett and Dr. Edward
Morash for their generosity in helping me get started in the all-important publishing
side of academia, Dr. David Closs for his excellent example of what good teaching is
all about as well as his aid in publishing, and department chairperson, Dr. Robert
Nason, for treating me fairly in good times and bad.

This research would have been impossible without the voluntary cooperation of
the respondents. Having been in industry, I fully appreciate the demands placed upon
managers and executives. I thank them for taking the time to complete the
questionnaires.

The international connection in my program and this dissertation also deserves
mention. Participation in Nijenrode University’s Summer Scholars Program (1996)

provided an ideal environment for writing and thinking. I sincerely doubt the
vii

dissertation process could have moved along without my Dutch interlude - or my Dutch
connection, Mr. Hans van der Hoop. Hans was extremely helpful during my visits to
Europe. His hospitality, friendliness, and knowledge of logistics are all well-
appreciated.

Although the doctoral experience is highly individualized in many respects --
and difﬁcult to explain to outsiders -- the process is shared with other students, all with
their own idiosyncrasies and foibles. That sharing, in large measure, makes the
experience bearable, stimulating and frequently so absurd it is hilarious. I appreciate
the support and humor shared with Judy Schmitz Whipple, Dave Frayer, Matt Myers,
and the “mahvelous one,” Gary Knight. Stress relief through basketball was also
welcome. I thank all those doctoral students who tolerated my limited hardwood skills.

Additionally, I thank my “look-alike,” Jeff Schmidt, and his wife, Sue, and my
“logistics conference room” cellmate, Robb Frankel, and his wife, Vivienne, for
continually opening their homes and dinner tables to the “single guy.” They gave far
more than I could ever repay in witty conversation. The experiences shared with Robb
and Jeff will forever be embedded in my mind and will generally bring a smile to my
face if not an outright snort of laughter.

I also thank the departmental staff -- Laurie Fitch, Sharon Baker, Cheryl
Lunden, Cindy Seagraves, and Renee Dixon -- for their warm personalities and
willingness to always assist me. The fact that they would endure my antics and pranks
without losing patience was appreciated.

Finally, in closing, I thank my friends and family. My friends from earlier

periods in my life always served as a reminder of where I have come from. Their
viii

 

 

collective vows to never address me as “Doctor” helped ground me throughout this
process and kept my ego in check.

My parents are owed a huge debt of gratitude as they instilled a love of learning
in me at an early age. Their encouragement throughout my life, in both academic and
non-academic efforts, is something every child should be fortunate enough to
experience. I truly regret that my father, Jim, is not alive to see me cross this ﬁnal
academic hurdle. His passing, seven months prior to my entering the doctoral
program, left a huge gap in my life as well as the life of my mother, Caroline. 1 can
never thank her enough for her strength these past ﬁve years. In the face of
tremendous sadness she rarely asked much of me, preferring instead that I maintained
my focus on my program. This dissertation is as much a testament to her resolve as to

my own.

TABLE OF CONTENTS

ITEM PAGE
LIST OF TABLES .......................................................................... xiii
LIST OF FIGURES ......................................................................... xv
CHAPTER ONE
INTRODUCTION ........................................................................... 1
General Status of Environment-Strategy-Structure-Performance Research ......... 5
General Contingency Theory .............................................................. 6
Research Purpose ............................................................................ 9
Research Scope .............................................................................. 10
Organization ................................................................................. 11
CHAPTER TWO
SURVEY OF THE LITERATURE ....................................................... 12
Contingency Theory ........................................................................ 12
The Concept of “ Fit” .................................................................. 13
The Use of Contingency Theory ...................................................... 17
Contingency Variables ..................................................................... 20
Environmental Uncertainty ............................................................ 21
Environmental Heterogeneity ......................................................... 27
Importance of Logistics ............................................................... 29
Information Technology ............................................................... 30
Strategy, Structure, Integration, and Performance: Individual Deﬁnitions
Under the Chow et al. Framework ........................................................ 33
Strategy .................................................................................. 33
Structure ................................................................................. 35
Central ization ....................................................................... 35
Span of Control and Scope ....................................................... 37
Formalization ....................................................................... 38
Integration ............................................................................... 39
Performance ............................................................................. 40
Strategy and Structure ...................................................................... 43
Environment-Strategy-Structure-Performance .......................................... 44
Logistics Contingency Model ............................................................. 48

X

Context ................................................................................... 49

Strategy .................................................................................. 50
Structure ................................................................................. 50
Performance ............................................................................. 51
Summary ..................................................................................... 52
CHAPTER THREE
RESEARCH METHODOLOGY ......................................................... 53
Research Purpose and Objectives ......................................................... 53
Research Questions ......................................................................... 54
Research Hypotheses ....................................................................... 55
Methodology Operationalization .......................................................... 61
Unit of Analysis ........................................................................ 61
Sample Size and Characteristics ...................................................... 61
Statistical Analysis ..................................................................... 63
Rationale of Overall Design ............................................................... 64
CHAPTER FOUR
RESULTS .................................................................................... 65
Assessment of Construct Unidimensionality and Reliability ........................... 65
Context ................................................................................... 67
Environmental Dynamism and Environmental Hostility ....................... 67
Information Technology Operating Systems and Information
Technology Flaming Systems .................................................... 69
Structure ................................................................................. 80
Span of Control .................................................................... 80
Integration ........................................................................... 81
Formal ization ....................................................................... 86
Centralization ....................................................................... 86
Strategy .................................................................................. 89
Performance ............................................................................. 90
Summary of Primary and Secondary Constructs ....................................... 91
Evaluation of Research Questions and Hypotheses .................................... 91
Environmental Dynamism ............................................................. 91
Environmental Hostility ............................................................... 97
Logistics Information Technology: Operating and Planning Systems .......... 102
Mediation Issues ........................................................................ 112
Strategy and Structure ................................................................. 125
Strategy, Structure and Performance ................................................ 129
Summary ..................................................................................... 135

xi

CHAPTER FIVE

CONCLUSIONS ............................................................................ 139
Research Results ............................................................................ 139
Context and Strategy ................................................................... 140
Context and Structure .................................................................. 144
Strategy and Structure ................................................................. 153
Structure and Performance ............................................................ 155
Interactions and Mediation ............................................................ 156
Performance-Strategy Feedback Loop ............................................... 159
Contributions: Academic .................................................................. 162
Contributions: Managerial ................................................................. 165
Limitations ................................................................................... 168
Directions and Propositions for Future Research ....................................... 169
Summary ..................................................................................... 172
APPENDIX A - Workbook Questionnaire .............................................. 175
APPENDIX B - Follow-up Questionnaire .............................................. 198
APPENDIX C - Initial CFA Results - Context Construct ............................. 202
APPENDIX D - Kuder-Richardson Calculations ...................................... 204
APPENDIX E - Research Hypotheses ................................................... 205
BIBLIOGRAPHY ........................................................................... 209

xii

LIST OF TABLES

129.19 Page
1.1 Type of Research Performed ..................................................... 2
2.1 Items Used to Portray Environmental Uncertainty ............................. 23
2.2 Environmental Uncertainty in Logistics Research ............................. 24
2.3 Process/Market/Channel Deﬁnitions ............................................ 34
2.4 Logistics Performance Measures ................................................. 41
4.1 Initial Items in EFA Analysis: Environmental Dynamism and

Environmental Hostility ........................................................... 68
4.2 Factor Loadings: Environmental Dynamism / Environmental Hostility .. 69
4.3 Information Technology Characteristics ........................................ 70
4.4 Factor Loadings: Logistics Operating Systems / Logistics Planning

Systems .............................................................................. 72
4.5 Results of Revised Context Construct Model .................................. 77
4.6 Reliability Estimates: Context’s Secondary Constructs ....................... 79
4.7 Underlying Indicators of Integration Construct ................................ 82
4.8 EFA of Integration Construct .................................................... 83
4.9 Results of Integration Construct Model ......................................... 85
4.10 Activities Under Logistics Responsibility ...................................... 88
4.11 Process/Market/Channel Deﬁnitions ............................................ 89
4.12 Indicators of Performance/Importance .......................................... 90
4.13 Categorical ANOVA: Environmental Dynamism and Strategy .............. 92
4.14 T -Test: Inﬂuence of Environmental Dynamism on Centralization ......... 93
4.15 T-Test: Inﬂuence of Environmental Dynamism on Formalization ......... 94
4.16 T-Test: Inﬂuence of Environmental Dynamism on Integration .............. 95
4.17 T-Test: Inﬂuence of Environmental Dynamism on Span of Control ...... 96
4.18 Categorical ANOVA: Environmental Hostility and Strategy ................ 97
4.19 T—Test: Inﬂuence of Environmental Hostility on Centralization ............. 99
4.20 T -Test: Inﬂuence of Environmental Hostility on Formalization ............. 99
4.21 T—Test: Inﬂuence of Environmental Hostility on Integration ................. 101
4.22 T-Test: Inﬂuence of Environmental Hostility on Span of Control .......... 101
4.23 Categorical ANOVA: Operating Systems and Strategy ...................... 103
4.24 T-Test: Inﬂuence of Operating Systems on Centralization ................... 104
4.25 T-Test: Inﬂuence of Operating Systems on Formalization ................... 104
4.26 T-Test: Inﬂuence of Operating Systems on Integration ....................... 105
4.27 T-Test: Inﬂuence of Operating Systems on Span of Control ................. 106

xiii

4.28
4.29
4.30
4.31
4.32
4.33
4.34

4.35

4.36

4.37

4.38
4.39
4.40
4.41
4.42
4.43
4.44
4.45
4.46
4.47
4.48
4.49

Categorical ANOVA: Planning Systems and Strategy ........................
T -Test: Inﬂuence of Plaming Systems on Centralization ....................
T -Test: Inﬂuence of Planning Systems on Formalization ....................
T -Test: Inﬂuence of Planning Systems on Integration ........................
T-Test: Inﬂuence of Plaming Systems on Span of Control ..................
Interaction Effects of Context on Structure ....................................
Tests of Mediation: Environmental Dynamism (ED)—Structure-
Performance ..........................................................................
Tests of Mediation: Environmental Hostility (EH)-Structure-
Performance ........................................................................
Tests of Mediation: I.T. Operating Systems (ITOP)-Structure-
Performance .........................................................................
Tests of Mediation: I.T. Planning Systems (ITPLAN)-Structure—
Performance ........................................................................

Tests of Mediation: Environmental Dynamism (ED)-Strategy-Structure..

Tests of Mediation: Environmental Hostility (EH)-Strategy-Structure
Tests of Mediation: I.T. Operating Systems (ITOP)-Strategy-Structure
Tests of Mediation: I.T. Planning Systems (ITPLAN)-Strategy-Structure
T -Test: Inﬂuence of Strategy on Centralization ...............................
T -Test: Inﬂuence of Strategy on Formalization ...............................
T—Test: Inﬂuence of Strategy on Integration ...................................
T—Test: Inﬂuence of Strategy on Span of Control ............................
Chi-Square Test of Performance and Change in Strategy .....................
T -Test: Inﬂuence of Structure on Performance ...............................
Tests of Mediation: Strategy-Structure-Performance .........................
Summary of Statistical Tests and Related Research Hypotheses .............

xiv

107
108
109
110
111
112

115

116

117

118
121
122
123
124
125
126
127
128
130
132
134
135

Figure

NNNHt—r—t
wNHri-e

4.2
4.3
4.4
4.5
4.6
4.7
4.8
5.1
5.2
5.3
5.4
5.5

LIST OF FIGURES

Logistics Organization and Performance ........................................

Logistics Contingency Model.

Relationships in a Contingency Model ..........................................

Six Perspectives of Fit .........

Logistics Organization and Performance .......................................

Logistics Contingency Model .
Construct Relationships .......

OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO

0000000000000000000000000000000000000000000000000000

Conﬁrmatory Factor Analysis: Initial Context Construct .....................
Conﬁrmatory Factor Analysis: Revised Context Construct ..................
Conﬁrmatory Factor Analysis: Integration Construct ........................

Mediational Model ..............

Mediational Model: Context-Structure-Performance .........................
Mediational Model: Context-Strategy-Structure ...............................
Mediational Model: Strategy-Structure—Performance .........................

Logistics Contingency Model:
Logistics Contingency Model:
Logistics Contingency Model:
Logistics Contingency Model:
Logistics Contingency Model:

Context-Strategy ..............................
Context-Structure .............................
Strategy-Structure .............................
Structure-Performance .......................
Mediation Paths ....... ' ........................

XV

CD

Pag

112
114
119
133
141
145
153
156
158

CHAPTER ONE - INTRODUCTION

Behavioral research has gained increasing attention in academic logistics
journals. In contrast to the long tradition of logistics study in location theory,
transportation economics, inventory modeling, process integration and total cost
analysis, etc., behavioral logistics research examines the interplay of human actions and
associated work. This “softer” side of logistics research is continually venturing into
new areas, attempting to provide both managers and academicians with a better
understanding of such diverse topics as strategic alliances, information utilization and
customer service. A natural outcome of this research has been the adaptation of
concepts and theories from other disciplines. Critics of behavioral logistics research
argue that for the ﬁeld to advance, these concepts must be woven into a broader
theoretical structure that serves to illuminate and explain logistics practice.

Mentzer and Kahn’s (1995) review of the Journal of Business Logistics
dramatically illustrates this deﬁciency. Table 1.1 reproduces their ﬁndings concerning
research reported in the Journal of Business Logistics from its inception through

Volume 14, Number 1, 1993.

2

Table 1.1
Type of Research Performed

 

Category % of Articles Published in JBL
Normative Literature/Reviews 53.6%
Exploratory Studies 36.2%
Methodology Reviews 6.0%
Hypothesis Testing 4.3%

Nearly 90% of all published articles in the Journal of Business Logistics, widely
regarded as one of the top logistics journals, have been either normative research,
literature reviews or exploratory studies. According to Mentzer and Kahn (1995) this
indicated that “... present logistics research has a large degree of substantive
justiﬁcation, but little subsequent theory development and testing.” In their view,
logistics research -- including behavioral logistics research -- must “ adopt a rigorous
research process that manifests theory development.”

Similarly, Chow, Heaver and Henriksson (1995) found logistics research
lacking a sound theoretical basis. In particular, Chow et al. noted that despite a great
deal of attention to organizational issues in logistics, comparability among studies is
difﬁcult, hampered by the “ variety of ways in which organizational concepts have
been deﬁned and measured.”

In response, Chow et al. proposed a contingency-based environment, strategy,
structure, and performance (ESSP) framework for behavioral logistics research. While
others have previously advocated the use of contingency theory (Persson 1982; Pfohl
and Zéllner 1987), Ch0w et al. contend that their framework could be used in “...
research on the organization of logistics.” A graphical representation of the Chow et

al. framework is reproduced in Figure 1.1.

 

 

 

Variables

| Strategy | | Structure IVA [Integration |——*

Figure 1.1
Logistics Organization and Performance
Source: Logistics and Transportation Review (1995, Number 4)

Contingency ]

 

 

 

 

 

 

 

Performance 1

The framework of Chow et al. is, however, an example of the general
deﬁciency identiﬁed by Mentzer and Kahn (1995). Chow et al. put forth a conceptual
framework but leave “ . .. the theory testing to others.” Nonetheless, the framework is
conceptually rich, with links to organization, management and marketing literature, is
theory-driven, and potentially could initiate a coherent and focused stream of
behavioral logistical research.

Such behavioral research would be of interest to both academics and
practitioners. As organizations strive to compete in a global economy, managers are
increasingly viewing logistics as a competitive core competency. Although logistics
has been widely described -- and its work practices well documented -- the current state
of behavioral logistics research is inadequate to more fully explain the role and value of
logistics within the ﬁrm. Research leading to a clearer understanding of the
relationship between environment, strategy, structure, and performance at the logistics

level would be welcome by both academicians and practitioners.

4
Bearing the challenge of Mentzer and Kahn (1995) in mind, it was the goal of

this research to use the Chow et al. framework as the basis for developing an
alternative logistics contingency model. The alternative conceptualization supplements
the strengths of the Chow et al. framework. Figure 1.2 presents the alternative model.

It is fully discussed in Chapter 2.

Context

.....................

f Environmental ,

j Dynamism j 11
Environmental —:

; Hostility mp

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Logistics H .R. Logistics
. . ' Structure Performance

Logistics I.T. "P .
Operating Systems ‘3 I
. I
.4 Logistics I.T. ‘ ' I 1
2— Planning Systems j i
F ................... . v ‘7 :
Logistics ‘_ _ _ _ _° _3

Strategy

Figure 1.2

Logistics Contingency Model

The application of a contingency approach facilitated the testing of different
variables interactive in the environment-strategy-structure relationship and their
subsequent impact on logistics performance. The focus of this research was logistics at
the strategic business unit level. This level of analysis permitted investigation of
logistics on a task level basis and avoided the more complex interrelations of corporate-
level processes. Examination of the ESSP framework within a logistics perspective
advances logistical theory, establishes a fruitful research stream while simultaneously

offering practical insight to practitioners.

5

GENERAL STATUS OF ENVIRONMENT-STRATEGY-STRUCTURE-
PERFORMANCE (ESSP) RESEARCH

The ESSP paradigm is well-established in the organizational science literature.
It is concisely deﬁned as follows (Galunic and Eisenhardt 1994, p.216):

This paradigm maintains that congruency exists between strategy and

structure. Moreover, where there is alignment between an

organization’s strategy and structure, better performance is likely.

Galunic and Eisenhardt’s review article (1994) indicates that within the
organizational science literature, the ESSP paradigm has been dominated by structural
contingency research and theory. Despite this dominance, Galunic and Eisenhardt
conclude the structural contingency approach suffers from two primary deﬁciencies:
(1) contradictory ﬁndings based on different deﬁnitions and interpretations of strategic
orientations, structural conﬁgurations and performance measures, and (2) empirical
shortcomings caused by “... reliance on bivariate models and single contingencies.”
To overcome these problems, Galunic and Eisenhardt recommend (1994, p.217):

The current paradigm draws an incomplete picture of the fast-paced

change and complex global reality of organizations today. In contrast, a

renewed strategy-structure-performance paradigm embraces systems

thinking, multiple contingencies, more accurate conceptualization of
strategy, structure, and performance and dynamic models. To achieve

this renewal, we suggest future research that is multivariate, multilevel,
inductive, and longitudinal.

Thus, within the organizational science discipline a structural contingency approach to
ESSP is still generally accepted, albeit in need of updating.

Outside the organizational science literature one ﬁnds the ESSP paradigm, or
signiﬁcant components of it, used in such diverse areas as strategic management (Habib

and Victor 1991; McWilliams and Smart 1993), new product development (Cusumano

6

and Nobeoka 1992), management information systems (Jarvenpaa and Ives 1994;
Kambil and Short 1994), and marketing channels (Ruekert, Walker and Roering 1985;
Mohr and Nevin 1990).

In logistics research, Chow et al. (1995) note that relatively few studies have
utilized an ESSP framework (Persson 1982; Pfohl and Zéllner 1987). They further
note that the logistics literature contains few empirical studies linking, in any manner,
the areas of strategy, structure, and performance. The work of Droge and Germain
(1989); Germain (1989), Germain, Droge and Daugherty (1994) has, however,

examined various facets of these areas within logistics.

GENERAL CONTINGENCY THEORY

Although Chandler (1962) and Woodward (1965) are recognized as the earliest
researchers to investigate contingent relationships between strategy, Structure and
performance, the development of general contingency theory in organizational science
is typically attributed to the work of Lawrence and Lorsch (1967) and Thompson
(1967). Building upon this foundation, Lorsch and Morse (1974) and Luthans (1976)
further reﬁned the application of contingency theory in organizational studies. Luthans
and Stewart (1977, p. 183) deﬁne contingency theory as follows:

The contingency approach is generically situational in orientation, but

much more exacting and rigorous. The contingency approach is deﬁned

as identifying and developing functional relationships between

environmental, management, and performance variables.

In their opinion, contingency theory maintains theoretical validity via an open system

while providing a more structured framework consistent with a closed system.

 

7

When applied to research, contingency theory is used to identify the different
responses to common environmental situations. The different responses are attributed
to the differing structures and strategies associated with individual ﬁrms. As a result,
“The contingency approach suggests, therefore, that we can observe wide variations in
effectiveness, but that these variations are not random” (Zeithaml, Varadarajan and
Zeithaml 1987).

These non-random variations can be explained by equiﬁnality. As used within
the organizational science literature, equiﬁnality is deﬁned as “ multiple
organizational conﬁgurations that result in the effective operation of the ﬁrm” (Galunic
and Eisenhardt 1994). Firms can and do choose paths based on ﬁrm-speciﬁc
considerations.

According to Luthans and Stewart (1977), these considerations can be grouped
into primary, secondary, and tertiary system variables. Primary system variables are
grouped into three broad categories: environmental, resource, and management.
Secondary system variables include situation, organization and performance. Tertiary
system variables represent the interactions between primary and secondary system
variables.

Within the Luthans/Stewart framework of primary system variables,
environmental factors are largely beyond the control of the ﬁrm (e.g., federal
legislation). Resource variables typically are characterized as within the control of the
ﬁrm (e. g., human rescurces, raw materials). Management variables comprise the
operating mechanisms used by managers (e.g., written company policies) to make

decisions.

8

At the secondary system level, situation variables are deﬁned by the
interaction of environmental and resource variables” (Luthans and Stewart 1977). The
interaction of managerial and resource variables comprises the secondary system
organizational variables. Environmental and management variables interact to create
performance criteria variables. The conceptualization of Luthans and Stewart’s (1977)
contingency model is reproduced in Figure 1.3.

The interaction of all the variables produces a total system performance, which
“ . represents the actual performance output of the organization as measured by
relevant performance criteria variables” (Luthans and Stewart 1977). From Figure
1.3, it is apparent that numerous relationships are evident, creating a complex model.

In order to utilize the model, researchers have sought to simplify or reduce the number

of relationships examined within any given study.

Environmental

 

SITUATIONAL SYSTEM PERFORMANCE
PERFORMANCE
Resource 7 T Management
ORGANIZATIONAL
Figure 1.3

Relationships in a Contingency Model

9
RESEARCH PURPOSE

As increasing attention is focused on logistics, the need exists to better
understand how logistics strategy and structure contends with contingencies in order to
achieve a satisfactory level of performance. To date, academics working in the
logistics ﬁeld have barely examined the complex relationships of the ESSP paradigm.
Consequently, the purpose of this research was to extend the investigation of whether
logistics structure and performance are contingent upon speciﬁed logistics context‘ and
strategic orientation and to what extent, if any, subsequent performance evaluation
impacts strategy.

The speciﬁc objectives of this research were:

(1) Test empirically if context variables directly or indirectly affected
strategy, structure and performance.

(2) Test if a relationship, possibly reciprocal, existed between structure
and strategy.

(3) Test if there was any support for a feedback mechanism whereby
performance inﬂuenced strategy.

(4) Suggest reﬁnements to the proposed CSSP model in an effort to
develop additional hypotheses and directions for future logistics
research.

 

' “Context” is increasingly being used in place of “environment.” Context signiﬁes inclusion of a
wider scope of variables, not just traditional environmental variables. Nonetheless, historically,
“environment” appears in the literature. Thus, in Chapters One, Two and Three, “environment” and
“ESSP” are used. Once the discussion shifts to the Logistics Contingency Model, the more
appropriate terms, “context” or “CSSP,” are used.

10
RESEARCH SCOPE

The research scope examined the ESSP paradigm from a logistics perspective at
the strategic business unit level. This research was based on the expectation that
organizational science concepts could be integrated into logistics research, yielding
insights into how contingency variables interact with logistics' organizational structure
to achieve or obstruct desired performance levels. This expectation was based on the
numerous studies previously mentioned that have successfully used the ESSP approach
within general management research. It appeared logical that the paradigm could be
used productively in logistics research.

No particular industry, ﬁrm size or channel positioning restrictions were
imposed. Previous logistics research (Bowersox, Daugherty, Droge, Rogers, and
Wardlow 1989; Council of Logistics Management 1995) indicated that logistics
performance was not affected by industry type, ﬁrm size or position in channel of
distribution. This study was limited, however, to manufacturers and merchandisers.
No third party service providers were considered in this research.

The relative lack of ESSP research within logistics created an interesting
paradox. From a logistics perspective, the Chow et al. framework and the alternative
model were untested, suggesting an exploratory approach. At the same time, some of
the constructs and scales had been used in logistics ESSP research. Fortunately, the
existence of the World Class Logistics database offered the potential for initial
investigation of the model through a statistical approach. A follow-up survey, tightly
focused on pertinent constructs in strategy and performance, supplemented the existing

database.

1 1
ORGANIZATION

The remainder of this dissertation consists of four chapters. Chapter Two
reviews ESSP literature. Particular emphasis is given to the literature cited by Chow et
al. as some deﬁnitional and operational issues exist. Support is provided for the
transference of organizational science concepts into the logistics discipline.

Chapter Three details the methodology and research design. The chapter also
includes the speciﬁc research questions and hypotheses investigated in this research.
Chapter Four presents the results derived from the statistical investigation using the
North American World Class Logistics database.

Chapter Five presents the conclusions in terms of theoretical and pragmatic
contributions. Research implications for academics and practitioners are presented.

Future research directions, considerations and hypotheses are also suggested.

CHAPTER TWO - SURVEY OF THE LITERATURE

The following is a selected review of pertinent literature covering contingency
theory, speciﬁc contingency variables identiﬁed by Chow et al. (1995), and individual
and integrated areas of strategy, structure and performance. Given the nature of this
dissertation, the discussion of the strategy, structure and performance areas will begin
with deﬁnitions and constructs as proposed by Chow et al. However, additional
material will be drawn upon to support the development of an alternative model of

context-strategy-structure-performance.

CONTINGENCY THEORY

Contingency theory is rooted in organizational science. It is generally
acknowledged that the works of Burns and Stalker (1961), Chandler (1962), and Cyert
and March (1963) laid the foundation of contingency theory through discussion of
general and open systems theory. Soon thereafter, Lawrence and Lorsch (1967) and
Thompson (1967) introduced the formal contingency approach. As deﬁned by
Lawrence and Lorsch, contingency theory states that organizations should be viewed as
open systems interacting with and affected by the surrounding environment.

In a later and expanded view of contingency theory, Kast and Rosenweig (1973)

state:

12

13

The contingency view seeks to understand the interrelationships
within and among subsystems as well as between the organization
and its environment and to deﬁne patterns of relationships or
conﬁgurations of variables. It emphasizes the multivariate nature
of organizations and attempts to understand how organizations
operate under varying conditions and in speciﬁc circumstances

(page ix).

THE CONCEPT OF “FIT”

Inherent in these relationships is the idea of “fit” -- a central theme of
contingency theory. Fit is generally characterized as the alignment of a ﬁrm’s strategy,
structure and process with its context or environment. Firms experiencing better ﬁt are
expected to have better effectiveness or performance than “ misﬁt” ﬁrms (Drazin and
Van de Ven 1985). Discussion of the concept of ﬁt is important as researchers believe
that the mixed results of the contingency approach are attributable to poor deﬁnition
and operationallzations of fit (Drazin and Van de Ven 1985; Van de Ven and Drazin
1985; Venkatraman 1989). For the purposes of this research, two perspectives of ﬁt
are important. The ﬁrst perspective is ﬁt as a general concept in organizational science
research. The second perspective is ﬁt in terms of an evaluative testing scheme being
appropriately matched to a theoretical position.

Three primary types of fit are prevalent in organization science research (Van
de Ven and Drazin 1985). These are (1) selection, (2) interaction and (3) systems ﬁt.
Each ﬁt “signiﬁcantly alters the essential meaning of a contingency theory” (Drazin
and Van de Ven 1985, p. 515).

Selection ﬁt is the most familiar. It is sometimes referred to as congruency ﬁt,

a reference to earlier literature when the terms ﬁt and congruency were used (T osi and

14
Slocum 1984; Drazin and Van de Ven 1985). Typically, selection ﬁt does not test the
relationship of context and/or environment and strategy and/or structure to
performance. Selection ﬁt can be described and deﬁned as follows (Selto, Renner and
Young 1995, p. 669):

In the selection approach, organizational context drives

organizational design. Fit is deﬁned in terms of predictable

correlations between pairs of organizational variables. Natural

selection predicts that all structure and control variables are

correlated with context since anything less would lead to

extinction in a competitive environment. Managerial selection

predicts correlations between context and only those control and

structural characteristics managed by the organization.

Although widely criticized, selection ﬁt continues to be used in contingency
theory. Drazin and Van de Ven (1985) suggest the selection approach may be most
appropriate in identifying macro-level relations (e.g., formalization and technology)
which can provide insights into micro-level relations (e.g., routinization between
departments).

Interaction ﬁt is also quite familiar to contingency theorists. This ﬁt focuses on
the effect of the interaction of context and/or environment and strategy and/or structure
on performance. Selto et al. (1995, p. 669) describe interaction ﬁt as follows:

The basic concept is that none of context, structure, or control

alone should affect performance; it is the ﬁt among them that

affects performance. In a regression explaining performance,

therefore, an interaction term of context and workgroup structure

(and control) should be signiﬁcant while main effects should not.

Statistically, main effects may be signiﬁcant, of course, but such

results detract from the theory.

Schoonhoven (1981) described the drawbacks of interaction ﬁt but also

demonstrated its potential. Hers remains a minority in terms of success as most

15

researchers in the organizational science area have not been successful in supporting
interaction effects (Drazin and Van de Ven 1985).

Whereas selection and interaction ﬁt are reductionist approaches, systems ﬁt is
holistic (Miller, Gilmour and Van Dierdonck 1981; Drazin and Van de Ven 1985).

Advocates of this approach assert that the understanding of

context-structure performance relationships can only advance by

addressing simultaneously the many contingencies, structural

alternatives, and performance criteria that must be considered

holistically to understand organization design (Drazin and Van de

Ven 1985, p.519).

In order to assess ﬁt in the systems approach, pattern-matching has been
recommended and employed (Drazin and Van der Ven 1985; Selto, Renner and Young
1995). Pattern-matching requires identiﬁcation of a high performing subgroup within
the data. Then, using summary measures of Euclidean distance, other groups are
compared to the high performing group. The further the distance between the high
performing group and the subject group, the greater the misﬁt (i.e., groups plotted
closest to the high group are assumed to have better ﬁt and subsequently it can be tested
whether they also have better performance or effectiveness).

Due to the overall difﬁculty in demonstrating ﬁt in organization studies, Drazin
and Van de Ven (1985) recommend a “triangulation” approach of all three types of ﬁt.
To date, two studies that have applied this approach have had different results. Drazin
and Van de Ven (1985) were able to support the selection and systems approaches but
not the interaction approach. Selto et al. (1995) were only able to support the selection

approach. These results, ironically, lead to the conclusion that within contingency

approaches there is no one “best ﬁt” measure at this time.

 

 

16

The often contradictory results of contingency theory can be disconcerting.
However, Venkatraman (1989) argues that much of the difﬁculty in investigating ﬁt
can be attributed to “a lack of correspondence between the concept and its
mathematical formulation ...” (p. 423). As a result, he offers a second perspective of
the concept of ﬁt.

In Venkatraman’s view, six conceptualizations of ﬁt are available to
researchers. These are: ( 1) ﬁt as moderation, (2) ﬁt as mediation, (3) ﬁt as proﬁle
deviation, (4) ﬁt as matching, (5) ﬁt as covariation, and (6) ﬁt as gestalts. None is
superior to the others per se but each is appropriate in different research designs.

Venkatraman’s classiﬁcatory framework is reproduced in Figure 2.1.

 

'6 Fit as Proﬁle Fit as Gestalts
” 0 Deviation
u: m in
“a e -

.. D 8
Z": 53 z:
.8 h H N
c- o N 5
3‘: . . . . . . > u'
8 E Fit as Mediation Fit as Covariation “5 o

o o—D
%:e E“
h—

o 2 m a 2
0 O = ..i
o o- o =

._ i:
8- 8 c-r Z on:
an s a:
0 v—
c: a 2 . . . .

Fit as Moderation Fit as Matching

 

 

 

E
n
I
'11
F1
2

Criterion -speciﬁc Criterion-free

Choice of anchoring the speciﬁcation of
ﬁt-based relationships

Figure 2.1
Six Perspectives of Fit
Source: Venkatraman (1989) Academy of Management Review

17

The three dimensions indicated in Figure 2.1 serve to guide researchers in
determining what type of ﬁt is appropriate for their theoretical discussion. Each ﬁt is
further tied to appropriate statistical techniques. It is anticipated that within this
research, ﬁt will be assessed using a limited number of variables tied to speciﬁc
criterion variables. Thus, using Venkatraman’s framework, ﬁt as mediation will be
most important. Mediation, with its causal relationships, is generally examined within

a path-analytic approach.

THE USE OF CONTINGENCY THEORY

The incorporation of different variables, conditions and circumstances, permits
the use of the contingency approach in a wide range of research settings. Within
organizational science, the contingency approach has been used in research involving
savings and loans (Jennings and Seaman 1994), manufacturing ﬁrms (Miller 1988), and
surgical operating rooms (Schoonhoven 1981). In each case, the contingency approach
was used to test relationships within the general sphere of environment, strategy,
structure, and performance.

Use of the contingency approach has expanded beyond organizational science.
It is frequently applied in organizational behavior and strategic management research.
In the mid-19805, Ruekert, Walker and Roering (1985) employed the contingency
approach in their study of marketing activities. Since then, Zeithaml, Varadarajan and
Zeithaml (1988) have advocated the use of contingency theory in marketing research
and theory-building. Just as the marketing literature has witnessed the growing use of

this approach, so too has the area of logistics (Pfohl and Zhllner 1987; Droge and

18
Germain 1989; Germain 1989; Kohn, McGinnis and Kesava 1990; Germain, Droge

and Daugherty 1994; Germain and Droge 1995 ; Germain 1996).

Despite this widespread use across different disciplines and research settings,
contingency theory is not without critics. Even a reduced number of contingency
variables can produce a dizzying array of possible combinations (Hofer 1975).
Therefore, contingency theorists have typically used a severely reduced number of
contingency variables. As a result, much of the work within the contingency approach
has used bivariate analysis (e. g., low/high uncertainty versus mechanistic/organic
structure). With the advent of more sophisticated analysis tools, most researchers agree
that contingency approaches need to consider multivariate levels of analysis (Galunic
and Eisenhardt 1994). However, as indicated in the preceding section, Venkatraman
(1989) would argue that contingency theorists need to pay greater attention to
correspondence between theoretical conceptualization and actual empirical testing.

Child (1972) criticized contingency determinism as an inherently weak type of
theoretical construction and not suitable in the contingency approach. This point has
been reiterated in the work of Donaldson (1982; 1987) who argued that the problem is
much more complex. In Donaldson’s view, the change in the environment eventually
leads to “substandard performance which comes from the mismatch of structure and
contingency.” As the ﬁrm strives to eliminate the mismatch, it may try to indirectly
inﬂuence the environment or seek to realign itself in a different environmental context.
The resulting structure ‘need only exceed a minimum threshold in terms of economic

performance.

19

A long-standing tenet of contingency theory, and its rejection of the universalist
approach, is that there is no one “best way” for ﬁrms to organize - even when facing
identical environments. This basic belief is currently captured in the concepts of
adaptation and equiﬁnality. Adaptation “asserts that the elements within the system
adapt to one another to preserve the basic character of the system” while equiﬁnality
“holds that a system can reach the same ﬁnal state from differing initial conditions and
by a variety of paths” (Zeithaml, Varadarajan and Zeithaml 1988).

Schoonhoven (1981) identiﬁed ﬁve problems with contingency theory: (1) lack
of clarity; (2) contingency relations as interactions; (3) functional forms of interaction;
(4) the analytic model used; and (5) assumptions about contingency relationships. The
ﬁrst problem, lack of clarity, has lead to calls for greater reﬁnement of variables and
constructs as a remedy (Schoonhoven 1981; Tosi and Slocum 1984). The second
problem, contingency relations as interactions, also stems from incomplete
speciﬁcation. In Schoonhoven’s ( 1981) words, “Explicit recognition should be given
to the fact that contingency arguments produce interactive propositions.” Her results
(Schoonhoven 1981) indicate that contingency theory can accommodate interactions but
it is the responsibility of the researcher to specify such relationships.

The third problem, functional forms of interaction, concerns the mathematical
function of implied interactions. Schoonhoven argued that a lack of clarity permitted
researchers to use mathematical or statistical techniques that represented different
interactions than described by researchers. Consequently, conﬂicting ﬁndings among
researchers could potentially be traced to this problem. Again, Schoonhoven suggested

several solutions to the problem and demonstrated results supporting her suggestions.

20

Schoonhoven’s position coincides with Venkatramen’s (1989) concept of ﬁt.

Schoonhoven’s fourth criticism, the analytic model used, attacks an underlying
assumption of most contingency research -- linear relationships. This problem can be
addressed to some extent by the use of more sophisticated quantitative methods and less
reliance on bivariate correlations (Gupta 1987; Galunic and Eisenhardt 1994).

Finally, the problem concerning assumptions about contingency relationships
suggests that contingency theory makes an assumption of symmetrical effects (i.e., an
increase in one variable must result in the same increase in the affected variable). Tosi
and Slocum (1984) acknowledge this problem and suggest that it is incumbent upon
each researcher to clearly specify proposed contingency relationships. From their
perspective, this problem is surmountable but it requires a thorough review and analysis
of concepts and methodological procedures each time contingency relationships are

proposed.

CONTINGEN CY VARIABLES

Lawrence and Lorsch (1967) originally envisioned contingency variables to
come from “outside” the organization - typically from the environment or contextual
domain of the ﬁrm. As such, these exogenous variables were considered to be subject
to little control or inﬂuence by ﬁrms. Although environmental variables and their
contingent relationships with strategy, structure and performance continue to be
investigated, a second set of contingency variables has emerged. These variables
represent aspects of a ﬁrm’s strategic position (e. g., vertical integration) and are

directly under a ﬁrm’s control (Hambrick and Lei 1985). Similarly, Ginsberg and

21

Venkatraman (1985) argue that strategy and performance can be important contingency
variables.

This more expansive view of contingent relationships is reﬂected by the ﬁve
contingency variables identiﬁed by Chow et al.: (1) environmental uncertainty,
(2) environmental heterogeneity, (3) the importance of logistics, (4) information
technology, and (5) strategy. The ﬁrst four will be discussed in this section. Strategy
will be discussed in the next section covering the strategy-structure-performance

paradigm.

ENVIRONMENTAL UNCERTAINTY

Within the contingency approach, environmental uncertainty can be traced to the
work of Lawrence and Lorsch (1967) and Duncan (1972). Lawrence and Lorsch
investigated the subenvironments of marketing, manufacturing and research. In each of
these subenvironments three questions were asked concerning: (1) lack of clarity of
information, (2) general uncertainty of causal relations, and (3) long time span of
feedback about results (Lawrence and Lorsch 1967). Duncan (1972) investigated two
dimensions of the environment, simple-complex and static-dynamic.

Subsequent application and testing of these two approaches to environmental
uncertainty indicated reliability and replication problems (T osi, Aldag and Storey 1973;
Downey, Hellriegel and Slocum 1975). Consequently, subenvironment-centered
approaches have generally not been used although aspects of information continue to be
of interest. Additionally, Duncan’s static-dynamic dimension has evolved into a

commonly cited measure of environmental uncertainty.

22

Environmental uncertainty is deﬁned by Chow et al. as “the extent to which
outcomes are unpredictable” (p.300). In their view, environmental uncertainty has two
important dimensions: (1) the ability of the decision-maker to predict the behavior and
expectations of constituent groups, and (2) the range of these behaviors and
expectations. This deﬁnition is fairly consistent with the use of environmental
uncertainty in various disciplines.

A representative example is Miller and Friesen’s (1978) use of dynamism and
hostility as part of their environment construct. The respective deﬁnitions capture the
dimensions favored by Chow et al.

Dynamism in the environment is manifested by the amount and

unpredictability of change in customer tastes, production or

service technologies, and the modes of competition in the ﬁrm’s

principal industries.

Hostility in the environment is evidenced by price, product,

technological, and distribution competition, severe regulatory

restrictions, shortages of labor or raw materials, and

unfavourable demographic trends (e.g., the drying up of

markets).

The deﬁnitions underscore an important point. Items designed to measure the
respective variables can be drawn from numerous areas (i.e., customer tastes,
technologies, price competition, etc.), thereby making it difﬁcult to assemble a
standard list of items applicable to every research design. Table 2.1 further illustrates

this point by indicating the wide range of items used to measure environmental

uncertainty in various organizational and marketing studies.

23

TABLE 2.1
Items Used to Portray Environmental Uncertainty

 

 

 

 

 

 

 

Measure of Environmental Relationship
Source Uncertainty Studied
Technology Type:
Woodward - custom Structure
(1965) - mass production
- continuous process
Hofer (1975) Stag of Product Life Cycle Bus. Strategy
Marketing
N onaka & Certainty-Uncertainty of Organization
Nicosia (1979) Environmental Information Structure
Market Growth Rate
Day (1986) (high versus low) Strategic Choice
Firm’s Competitive Position
(leader versus follower)

 

Source: Adapted from Zeithaml, Varadarajan and Zeithaml (1988)

Given this ﬂexibility, environmental uncertainty remains a frequently employed
contingency variable as it is predicted to both directly and indirectly affect elements of
organization strategy and structure. Its use has spread to the areas of strategic
management (e.g., Miller 1988), marketing (e.g., Ruekert, Walker and Roering 1985)
and logistics (e.g., Kohn, McGinnis and Kesava 1990).

The basic premise is that as environmental uncertainty increases, organizational
structure must adapt to process information related to an unstable environment. A
common proposal reﬂecting this relationship is that a low level of environmental
uncertainty leads to mechanistic structures. These structures are generally considered
to be highly centralized and focused on efﬁciency. Conversely, high levels of
environmental uncertainty lead to organic structures. Organic structures are typically
characterized as decentralized and emphasizing ﬂexible, nonroutine thinking and

solutions (Burns and Stalker 1961).

24

A complete review of environmental uncertainty and its hypothesized impact on
the components of the strategy, structure and performance framework is difﬁcult. As
mentioned previously, and illustrated by Table 2.1, items measuring these relationships
are many and varied. Fortunately, contingency approaches used in logistics research
provide a much narrower scope of relevant work. The remainder of the discussion of
environmental uncertainty focuses on this research. Table 2.2 summarizes the use of

environmental uncertainty items in logistics research.

 

 

Table 2.2
Environmental Uncertainty in Logistics Research
Research Measure of Environmental Relationship Studied
Uncertainty
Germain (1989) Percentage of Manufacturing to Organizational

Customer Order: Structure, Strategic

 

 

- Standardized Planning, Flexibility

- Customized

- Hybrid
Kohn, McGinnis and Based on Miller and Friesen’s Logistics Strategy
Kesava (1990) (1978) dynamism and hostility

constructs

Fawcett and Closs Perceived Globalization Impact: Logistics,
(1993) 5-item measure of Manufacturing,

opportunities/threats created by
economic globalization

Competitive Position

 

Germain, Droge and
Daugherty (1994)

Miller and Drbge (I986) JIT

Environmental Uncertainty Scale

 

 

 

 

World Class Miller and Droge (1986) Unpublished-to-date.

Logistics Research Environmental Uncertainty Scale

Survey - MSU

(1994/95)

Germain and Drbge Miller and Droge (1986) EDI/Technology/
1995) Environmental Uncertainty Scale JIT

Germain (1996) Miller and Drbge (1986) Logistics Imovation

Environmental Uncertainty Scale Adogtion

 

Germain’s (1989) research supported the hypothesis that environmental

uncertainty (i.e., deﬁned as environmental market uncertainty according to percentage

 

25

of manufacturing to customer order) is related to organizational structuring of logistics.
He found that “ manufacturers within the standardized context were found to
consolidate more logistics activities in a single department than customized context
manufacturers” (p.27). This follows the mechanistic/organic rationale of the
organizational science literature.

Using ﬂexibility as a proxy for performance, Germain (1989) obtained mixed
results. His ﬁndings generally support the hypothesis that manufacturers in a
customized context will exhibit greater logistics ﬂexibility. However, certain measures
such as product phase-out and product introduction did not support “the contingency
view that organizations facing environmental uncertainty attempt to remain ﬂexible and
adaptive” (p.23).

The research of Kohn, McGinnis and Kesava (1990) found that logistics strategy
was independent of “dynamism” but not “hostility.” In this study, dynamism was
deﬁned as “the unpredictability of change in customer taste, technology and modes of
competition;” hostility was deﬁned as “level of competition, severity of regulatory
restrictions, shortages and unfavorable demographic trends” (p.27). Thus, if viewed in
terms of the Chow et al. deﬁnition, range of behaviors affects logistics strategy but the
decision-maker is able to adequately predict constituent groups’ actions, thereby
minimizing the impact of those actions.

Fawcett and Closs (1993) found that Perceived Globalization Impact did
inﬂuence performance, logistics and manufacturing. However, as ﬁrms move to global
manufacturing, the increased uncertainty has a greater impact on logistics than

manufacturing. The authors believe this is due in large part to the “additional attention

26
required to design an effective and ﬂexible logistics system” (p. 18).

The remaining studies in Table 2.2 use the Miller and Droge (1986)
environmental uncertainty scale. Given the relative importance of this scale, each of

the items is reproduced below:

Our ﬁrm must rarely change its
marketing practices to keep up
with the market and competitors.

The rate at which products/
services are getting obsolete in
the industry is very slow (e. g.,
basic metal like copper).

Actions of competitors are quite
easy to predict (as in some
primary industries).

Demand and consumer tastes
are fairly easy to forecast (e. g.,
for milk companies).

The production/service technology
is not subject to very much change
and is well established (e. g., in
steel production).

Our ﬁrm must change its marketing
practices extremely frequently.

The rate of obsolescence is
very high as in some
fashion goods.

Actions of competitors are
unpredictable.

Demand and tastes are
almost unpredictable (e. g. ,
high-fashion goods).

The modes of production/
service change often and in a
major way (e.g., advanced
electronic components).

The collective results of the three reported studies support the use of
environmental uncertainty in contingency approaches to logistics research.
Environmental uncertainty predicts J IT selling, integration, performance control and
operations decentralization but not specialization or scheduling decentralization
(Germain, Droge and Daugherty 1994). Germain and Droge (1995) again found
support for the positive association between J IT strategy and environmental uncertainty.
They also identiﬁed a signiﬁcant relationship between environmental uncertainty and

production complexity but not between environmental uncertainty and EDI technology.

27

Finally, environmental uncertainty is a signiﬁcant predictor of radical logistics
innovation, manufacturing operations decentralization and integration, but not

specialization or innovation adaptation decentralization (Germain 1996).

ENVIRONMENTAL HETEROGENEITY

Environmental heterogeneity is closely related to environmental uncertainty.
Chow et al. have chosen to deﬁne it as “the degree of complexity in the ﬁrm’s
environment” (p.301). The degree of complexity is typically reﬂected in number of
products, customers, suppliers, locations, etc. It is noteworthy that Chow et al. would
use the term “complexity” in their deﬁnition as it is a term that appears synonymous
with environmental heterogeneity in the literature.

Duncan’s (1972) work in perceived environmental uncertainty used a simple-
complex dimension in addition to the static-dynamic dimension discussed in the
preceding section. He conceptualized the simple-complex dimension as follows:

The simple part of the simple-complex dimension deals with the

degree to which the factors in the decision unit’s environment are

few in number and are similar to one another in that they are

located in a few components. The complex phase indicates that

the factors in the decision unit’s environment are large in number

(p.315).

At the same time, Child (1972) described “environmental complexity” as “the
heterogeneity of and range of an organization’s activities.” Hage and Dewar (1973)
employed the term “complexity” to mean “the number of different organizational
specialties.” The general premise underlying all of these deﬁnitions is that a more

heterogeneous environment causes greater perceived uncertainty, leading to greater

information-processing requirements (Duncan 1972; Pennings 1975; Tung 1979; Dess

28
and Davis 1984).

Although this premise is intuitively appealing, Duncan (1972) found the static-
dynamic dimension to be considerably more important than the simple-complex
dimension. Researchers working in strategic management have also had similar
ﬁndings, resulting in considerably less usage of environmental heterogeneity as a
contingency variable (Bourgeois 1980, 1985; Lindsay and Rue 1980).

In other studies that have employed environmental heterogeneity, results have
been mixed. Keats and Hitt (1988), writing on the environment-organization interface,
posited a positive relationship between “complexity” and organizational
divisionalization. Their ﬁndings did not support the relationship, although it was
determined that complexity’s primary inﬂuence was on size of ﬁrms. Hambrick and
Lei (1985), working in the area of strategy, attempted to prioritize contingency
variables relevant to business strategy. Among the ten variables selected for analysis
was rate of concentration in an industry. Even though it does not fully represent the
environmental heterogeneity construct, industry concentration can be considered an
element. Hambrick and Lei’s results offered some support for the importance of ﬁrm
concentration. However, of the ten variables, it was classiﬁed in the “secondary
importance group,” the least important variables in Hambrick and Lei’s study.

Miller and Friesen (1978) used “heterogeneity” in their research of strategy
types. They deﬁned it as “the differences in competitive tactics, customer tastes,
product lines, channels of distribution, etc.” (p.922). In this study, heterogeneity was
characterized as an aspect of an overall environment construct. Of the ten archetypes

identiﬁed, heterogeneity is the most important environmental factor in only one and it

29

is the least important in ﬁve.

Within the logistics literature there is little evidence of previous use of
environmental heterogeneity. Kohn, McGinnis and Kesava (1990) did use it in
examining organizational environment and logistics strategy. They used two different
deﬁnitions in their study. The ﬁrst deﬁnition is based on Miller and Friesen’s (1978)
work. Kohn et al. found that when using this deﬁnition, the result was that logistics
strategy is independent of environmental heterogeneity. The second deﬁnition was
based on Dess and Beard’s (1984) use of complexity: “Complexity is deﬁned as the

’3

heterogeneity and range of an organization’s activities. The same result was obtained.
Logistics strategy was independent of complexity. This ﬁnding led Kohn et al. to
conclude that in terms of logistics research, environmental heterogeneity as deﬁned by

Miller and Friesen and complexity as deﬁned by Dess and Beard have similar meanings

but no signiﬁcant relationship to logistics strategy.

IMPORTANCE OF LOGISTICS
The third contingency variable advocated by Chow et al. is the importance of
logistics. They deﬁne the importance of logistics as “the extent to which logistics
activities constitute an important portion of the ﬁrm’s value-adding activities” (p.301).
The underlying justiﬁcation for inclusion of this variable is the relative cost of logistics
within a given ﬁrm. Citing Christopher (1986) and Davis and Brown (1974), Chow et
al. contend that if the ratio of logistics costs to total costs is high, logistics will merit
closer attention. Positioning logistics closer to top management may be one way of

ensuring attention and emphasis.

30

There is no prior use of this exact variable in logistics research. Perceptual
measures of the importance of logistics have been obtained in Michigan State
University’s Leading Edge and World Class Logistics research. Bowersox et al. (1989)
concluded that “reporting to top-level executives is indicative of greater visibility and

recognition” (p.87).

INFORMATION TECHNOLOGY

Information technology has been widely introduced to the ﬁeld of logistics as
evidenced by geo-positioning satellite communications, optical scanner links between
retailers and suppliers, bar coding of packages, computerized picking systems, etc.
The general belief is that information technology facilitates efﬁcient and effective
logistics operations. But from a logistics research perspective, what is information
technology? Unfortunately, Chow et al. do not deﬁne it.

Two studies have examined the impact of adoption of incremental and radical
logistics technology (Germain 1996; Germain, Droge and Daugherty 1994). With the
exception of robotics, their examples of logistics technology are all information-based
applications (e.g., AS/AR systems, handheld data entry devices, order entry software,
freight consolidation software, etc.). The 1994 study identiﬁed three logistics
technology clusters: (1) high cost, revolutionary technology, (2) medium cost, medium
revolutionary technology, and (3) low cost, incremental technology. Additionally,
adoption of the three logistics technology types were signiﬁcantly correlated with a
formalization of measurement construct but not with decentralization. Integration was

related to medium cost, medium revolutionary logistics technology and to span of

31

control.

Germain (1996) deﬁned the three clusters as radical, intermediate and
incremental. His ﬁndings generally support the earlier study. Of special interest to
this research is his work with environmental uncertainty and its relationship to the two
extreme technology clusters. Germain found that environmental uncertainty is
signiﬁcantly related to radical but not incremental innovation. His conclusion is that a
more dynamic environment creates a “future-oriented organization aware of external
change and hence more receptive to original solutions” (p.124). In an EDI-speciﬁc
research, Williams (1994) found that logistics channel EDI-adoption was not
signiﬁcantly related to environmental uncertainty. This is consistent with Germain,
Droge and Daugherty (1994). Germain (1996) did not speciﬁcally test the EDI-
uncertainty relationship as EDI was classiﬁed as an intermediate technology.

In terms of interorganizational relations, Williams (1994) found that channel
power played a signiﬁcant role in EDI-adoption in a logistics channel. She notes that
in many cases EDI-implementation is mandated by the stronger channel partner. This
supports the work of Clemons and Row (1993) who found that interﬁrm coordination
through information technology frequently faces signiﬁcant resistance.

Improving coordination within the distribution channel requires

investment in IT, but it also requires signiﬁcant changes in the

organizational mechanisms used to manage the interactions,

which we term coordination structure. Conceptually, improved

coordination must be viewed not just in terms of players changing

their strategies within an existing game, but in terms of changing

the structure of the game. Changes in coordination structure can

improve overall channel efﬁciency, but can also affect the

bargaining power of the parties. In some cases, one party can

ﬁnd him or herself actually worse off under improved
coordination, due to a loss of bargaining power, thus leading him

32

to resist the change; this can be true even if coordination yields a
net economic surplus to the channel (p.75).

Such resistance can perhaps be explained within logistics by the microsocial
view that posits that technologies ﬁrst alter tasks and skills, only later does
organizational structure modiﬁcation take place (Barley 1990). If so, it is possible that
the impact of information technology has been felt most at the lower levels or “front
lines” of the organization. Given the fact that most logistics tasks are low to medium-
skilled jobs it is entirely possible that resistance is encountered. The electronic
integration view holds that impacts are felt beyond the ﬁrm’s boundaries and
subsequently affect employees of other ﬁrms (Kambil and Short 1994). This
interlocking relationship created by information technology and its effects on lower-
level personnel has not been examined in the logistics literature.

Closs, Goldsby and Clinton (1997) have recently examined information
technology and its inﬂuence on logistics capability. They deﬁne information
technology as the operating and planning systems used by logistics personnel. In each
of these systems they evaluate reliability, timeliness, accuracy, connectivity,
availability, ﬂexibility, and format. Their ﬁndings support the idea that a ﬁrm’s
logistics capability is primarily inﬂuenced by logistics operating timeliness, usage

driven formatting and ﬂexibility.

33

STRATEGY, STRUCTURE, INTEGRATION, AND PERFORMANCE:
INDIVIDUAL DEFINITIONS UNDER THE crrow ET AL. FRAMEWORK

This section reviews the deﬁnitions and literature pertaining to four constructs:
strategy, structure, integration and performance. The following section examines the

relationships of these variables.

STRATEGY

A fundamental problem in researching logistics strategy is that most “grand
strategies” (Miles and Snow 1978; Porter 1980) have been developed in relation to
overall corporate or divisional strategy. The richness of such strategies is usually lost
at the functional level. Consider, for example, Porter’s typology of cost
leadership/differentiation/focus. This is usually applied in logistics research in the
form of a low cost <---> customer service continuum (e.g., World Class Logistics
Research, Michigan State University 1992-1995). While this can provide insights into
a ﬁrm’s logistics orientation, it falls far short of the descriptions and explanations
offered at the corporate level. Furthermore, there is no study that has systematically
related logistics strategy to overall corporate strategy.

Chow et al. recognize the limitations of corporate strategies in logistics
research. While acknowledging the existence of logistics-speciﬁc typologies they
prefer to use the following definition: “Logistics strategy can be deﬁned as a pattern
Of action plans designed for the purpose of achieving logistics goals” (p.296). In the
words of the authors, this deﬁnition “can encompass different typologies” (p.296).

Bowersox et al. (1989) proposed the logistics process/market/channel typology.

Chow et al. discuss this typology, noting its focus on organizational mechanisms. This

34

emphasis is evident in the deﬁnitions provided in Table 2.3.

Table 2.3
Process / Market / Channel Deﬁnitions

 

Process strategy is concerned with managing a broad group of logistics
activities as a value-added chain. Emphasis is on achieving efﬁciency
from managing purchasing, manufacturing, scheduling and physical
distribution as an integrated system.

Market strategy is concerned with managing a limited group of
logistics activities for a multidivision single business unit or across
multiple business units. The logistics organization seeks to make joint
product shipments to common customers for different product groups
and seeks to facilitate sales and logistical coordination by a single-
order invoice.

Channel strategy is concerned with managing logistics activities
performed jointly with dealers and distributors. The strategic
orientation places a great deal of attention on external control.
Signiﬁcant amounts of ﬁnished inventories are typically maintained
forward or downstream in the distribution channel.

Source: Bowersox et al. (1989) Leading Edge Logistics: Competitive
Positioning for the 19905, Oak Brook, IL: Council of
Logistics Management.

 

 

 

 

 

This organizational focus clearly indicates patterns of activities associated with
different logistics strategies. Presumably such organization is intended to facilitate the
achievement of logistics objectives. In that sense, the Bowersox et al. typology is
much more specific than Chow et al.’s broad deﬁnition. In advocating their position,
Chow et al. argue that the Bowersox et al. typology fails to incorporate trade-offs.
However, since the release of the Chow et al. framework, additional empirical support
for the process/market/channel typology has been obtained (Clinton and Closs 1997).
Factor analysis has established clear evidence of speciﬁc tendencies and practices
within each of the individual strategies. The varying intensity of several factors

common to all three strategies suggests that trade-offs might be identiﬁable. Based on

35

this new development it would seem worthwhile to pursue further research into the

typology.

STRUCTURE

Structure is a multidimensional construct. Unlike strategy, however, the
dimensions are better deﬁned. Chow et al. indicate that ﬁve properties of structure are
most important in logistics research: centralization, formalization, span of control,
scope, and logistics integration. In developing the deﬁnitions for each of these
properties, Chow et al. have relied upon the organizational sciences literature. The
ﬁrst four properties are discussed under structure. Integration will be discussed
separately, highlighting its position in the Chow et al. framework.

Central ization

 

In the view of Chow et al., centralization is deﬁned along two dimensions.
Centralization is “the extent to which the power to make logistics decisions is
concentrated in the organization” and “the hierarchical distance between logistics
decision-makers and senior executives who make more ‘global’ decisions on an
organization-wide scale” (p.288). The former dimension is more closely related to
organizational science while the latter dimension has been discussed only in the
logistics literature in relation to centralization.

Centralization involves the locus of authority to make decisions in

organizations. If, for instance, the power to make decisions is

exercised by one or relatively few individuals, the structure is

considered centralized. The minimum degree of centralization

(decentralization) would exist in an organization if decision-

making authority were exercised equally by every member of the

organization. Degree of centralization, then, refers to the
dispersion of decision-making authority throughout the

36

organization (Dalton, Todor, Spendolini, Fielding and Porter
1980 p.58-59).

The preceding is a representative deﬁnition of centralization/ decentralization in
the organizational science literature. The underlying rationale for including
centralization in contingency models is that it can result in better overall coordination
and control, leading to greater organization performance (Ruekert et al. 1985). But
this is believed to hold true only in stable environments. Highly volatile, uncertain
environments force decentralization as decision-makers are unable to cope with the
jumbled or erratic information ﬂow (Pennings 1975 ; Mintzberg 1979; Tosi and Slocum
1984).

The second dimension, hierarchical distance between logistics and senior
executives, is not routinely investigated in organizational theory literature -- even using
other functional areas. The “logistical excellence” studies conducted at Michigan State
University (i.e., Leading Edge; World Class Logistics) have been at the forefront of
this dimension. However, there is no evidence that researchers associated with these
projects have advanced the idea that hierarchical distance is a distinct dimension of
centralization. In any event, Bowersox et al. (1989) have found that closer proximity
or easier access to top management by logistics managers is “indicative of greater
visibility and recognition” (p.83).

There is, however, yet another use of centralization within the logistics
literature. Germain (1989) has used the number of activities under logistics
departmental line control to indicate centralization. This control “was assumed to

represent the extent to which logistics activities were centrally consolidated within a

37
single department (p. 24). ”

This deﬁnition presents a dilemma. It is logical to equate the consolidation of
activities within a single department as representing centralization. After all, decision-
making authority would then be concentrated in one department and not shared across-
functional or divisional boundaries. However, measuring the level of centralization
based on line control is contrary to the organizational science meaning based on number
of people wielding decision-making authority. In the particular study used by
Germain, respondents provided an indication of whether the activity was (a) part of the
formal control of logistics, and then if it was, (b) did line or staff have responsibility
for the activity? Using the organizational science position, one is compelled to think
that perhaps staff control (i.e., concentration of decision-making authority in fewer
parties) over the activities is more consistent with the concept of centralization. Line
personnel are typically more numerous than staff. Thus, if line personnel are entrusted
with the responsibility one could perhaps argue that this is consolidation of the
activities but decentralization of decision-making authority.

Span of Control and Scope

Chow et al. note that span of control has evolved into two closely related
meanings - range of activities under a manager versus number of people under a
manager. Based on Van Fleet and Bedeian’s (1977) review of the concept, the number
of subordinates under a manager is the more traditional measure. Chow et al. adopt
this position and deﬁne span of control as the number of subordinates who report to a

single superior.

38

Span of control has been of interest to researchers primarily because of its
relationship to effectiveness. It is generally theorized that supervisors can only
effectively manage a limited number of subordinates. Of interest to this research is the
ﬁnding that information technology can “extend” the span of control within
organizations (Germain, DrOge and Daugherty 1994).

The concept of “scope” is deﬁned by Chow et al. as “the degree to which
logistics activities are grouped together in the same organization or organizational sub-
unit” (i.e., see preceding discussion of centralization). This could be of interest as it
will illustrate whether ﬁrms deploy logistics responsibility from a single, concentrated
unit or disperse it across-functional areas such as manufacturing, purchasing or
marketing. The latter approach is frequently advocated by proponents of integrated
logistics management. The concept itself, however, has little prior application in the
literature as it has previously come under the deﬁnition of span of functional activity
control.

Formal ization

 

Formalization has been extensively studied in the organizational science
literature. Its consistent meaning within that body of work has enabled Chow et al. to
apply a similar meaning in their framework: Formalization is the degree to which
goals, rules, policies and procedures for logistics activities are precisely and explicitly
formulated (p.289). The general premise is that stable environments promote structures
high in formalization whereas dynamic, uncertain environments are associated with
lower levels of formalization. When the appropriate matching of environment and

formalization occurs, ﬁrms tend to be more efﬁcient than under mismatched

 

39

circumstances. Researchers have generally found that centralization and high levels of
formalization are associated. DrOge and Germain (1994) have evidence of this in

logistics.

INTEGRATION

Integration has usually been deﬁned as a structural property in the
organizational sciences and logistics literatures (Parsons 1960; Miller and DrOge 1986;
Miller 1987; DrOge and Germain 1989). Germain (1996) has succinctly deﬁned
integration as “lateral communications that span divisions, functions, or departments”
(p.118). Within logistics research, integration is typically depicted as a dimension of
organizational structure impacted by environmental uncertainty, context and strategy
(Germain, DrOge and Daugherty 1994; Germain 1996).

This view of integration contrasts with the recommendation of Chow et al. that
integration be considered as an intermediate outcome variable, a product of
organization structure. They state:

Integration is an outcome that can be associated with the state of

collaboration at one of three levels: two or more logistics

activities within a ﬁrm; the logistics process and other corporate

activities within the ﬁrm (such as marketing); participants in the

supply chain. In sum, integration is the degree to which logistics

tasks and activities within the ﬁrm and across the supply chain

are managed in a coordinated fashion (p.291).

Thus, although Chow et al. generally deﬁne integration in the same manner as
previous researchers, they position it as the key variable between structure and

performance. Although not explicitly stated, this reﬂects a belief that the systems

approach to logistics thinking (i.e., integrated logistics) requires acknowledged

 

-Aiig'. ’- '3 ‘

40

integration antecedent to successful logistics performance. However, a recent stream of
research into the integrated logistics concept and information systems does not support
this view of integration as an outcome (Gustin, Stank and Daugherty 1994; Stank,
Daugherty and Gustin 1994; Gustin, Daugherty and Stank 1995). The authors note that
“the integrated distribution concept creates an organizational structure ...” (Gustin,

Daugherty and Stank 1995, p.2).

PERFORMANCE

For the majority of their framework, Chow et al. draw upon the organizational
theory literature. They feel no compuncture to do so in the area of performance. The
logistics literature is overﬂowing with measures of performance - so many in fact that
Chow et al. do not even attempt a deﬁnition. This omission is defensible when one
considers the numerous ways in which logistics performance can be measured. But that
is the key. Any measure used should “directly evaluate the performance of the
logistics system” (DrOge and Germain 1989, p.85). In that respect, many of the global
measures (e.g., market share, corporate proﬁtability) used in the strategy-structure-
performance paradigm literature are inappropriate for this research focus. In order to
properly assess performance in this study, the performance construct must reﬂect the
task level evaluation consistent with the other model variables. Table 2.4 presents a
summary of logistics performance measures used in ﬁve leading logistics journals

between 1982 and 1992 (Chow, Heaver and Henriksson 1994).

41

Table 2.4
Logistics Performance Measures: 1982-1992

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source Measure Deﬁnition of Performance
“ Distribution effectiveness”: adequacy,
Rhea & Shrock (1987) Soft consistency, timeliness, initiative,
responsiveness
“Operating Performance” of speciﬁc
Perry (1988) Soft logistics dimensions
Bowersox et al. (1989) Soft Common attributes index
Firm’s ability to accommodate product
DrOge & Germain (1989) Soft introduction, phaseout, recall and
customization of service
Logistical performance: length of order
Gassenheirner, Sterling & Soft cycle time, meeting delivery dates. ﬁll
Robicheaux (1989) rate, advance notice on shipping delays,
accuracy, etc.
Firm’s ability to accommodate supply
disruption, production schedule changes,
Germain (1989) Soft product introduction, phase-out. etc. and
customization of service
Yavas. Luqmani & Quraeshi Efﬁciency measures compared to
(1989) Soft purchasing sophistication
Cooper, Browne & Peters Performance indicators of logistics
119904 Soft efﬁciency
Fawcett (1990) Soft Carrier performance
Read & Miller (1990) Soft Quality
Clarke (1991) Soft Productivity
Data envelopment analysis (DEA):
Clarke & Gourdin (1991) Hard efﬁciency/productivity
“Competitive position” of ﬁrm/proﬁt
Fawcettj1991) Soft center on mmwm measures
Gomes & Mentzer (1991) Hard Proﬁtability
Harrington, Lambert & Vendor performance: lead-time, lead-
Christopher (1991) Hard time variability, ﬁll rate, etc.
Kleinsorge, Schary &
Tamer (1991) Hard DEA - efﬁciency
“ Distribution service performance”/ 1-
Marr (1991) Soft item performance measure
Daugherty, Sabath & Rogers Firm’s ability to accommodate special
(1992) Soft requests
Daugherty, Stank & Rogers
(1992) Soft Service capabilities of ﬁrm
Diewert & Smith (1992) Hard Total factor productivity
Fawcett & Vellenga (1992)
Soft Carrier performance
Fawcett & Closs (1993) Soft “Competitive position” of ﬁrm/proﬁt

center on 5 perceptual measures

Adapted from: Chow, Heaver and Henriksson (1994)

 

 

42
Table 2.4 illustrates three key points. First, considering that Chow et al. (1994)

looked at published logistics research from 1982 - 1992, it is apparent that research into
logistics performance is a relatively recent logistics research phenomenon. Signiﬁcant
numbers of articles did not appear until the late 19805.

Second, perceptual or “soft” measures have dominated the research efforts.
There are several notable exceptions that used “hard” or Objective performance data.
However, Chow et al. (1994) note the difﬁculty that most researchers have in obtaining
proprietary performance data broken down to functional or departmental levels.
Finally, a perusal of the list of measures perhaps explains why, in their framework
discussion, Chow et al. do not offer a deﬁnition of logistics performance. Few
researchers have used the same measure to investigate logistics performance!

The interest in logistics performance has not slackened after 1992. In 1995
alone, academics were reporting on the relationship of ﬁrm responsiveness and
performance (Daugherty, Ellinger and Rogers 1995), comparisons of domestic and
international distributors’ performance (Cabaniss 1995), impact of information
technology on performance measurement (Bowersox and Daugherty 1995), and
performance characteristics of world-class logistics ﬁrms (Council of Logistics
Management 1995). Most of these initiatives used soft, perceptual measures of
logistics performance. And, as evidenced by the titles, performance was measured in a
variety of contexts.

Although this might be disconcerting to some researchers, it reﬂects the
multidimensional character of performance. What is important is that the performance

measures can be justiﬁed as realistic evaluations of the performance of the logistics

43
system (DrOge and Germain 1989). Chow et al. believe there is a place for both

“hard” and “soft” measures of logistics performance. What appears to be lacking is
reliable measures and scale development of the various dimensions (i.e., customer
service, productivity measures, cost considerations, etc.) of performance itself (Dunn,

Seaker and Waller 1994).

STRATEGY AND STRUCTURE

A long-standing belief and dominant paradigm in organizational theory holds
that structure follows strategy (Chandler 1962). Numerous researchers in the
organizational ﬁeld support this position (e.g., Rumelt 1974; Lewis and Fandt 1989;
Ansoff 1991). Amburgey and Dacin (1994) note that this relationship has also been
posited in a variety of other theoretical schools (e.g., design school of strategic
management (Mintzberg 1990); organizational ecology (Harman and Freeman 1984)).

A contrary view was hypothesized by Bower (1970) and Hall and Salas (1980).
They argued that structure must partly determine strategy. In their view, the ﬁrm
develops an initial strategy based on environmental conditions and ﬁrm resources. A
structure develops to support the strategy. Thereafter, however, the structure that is in
place limits the strategic alternatives available to the ﬁrm. This is believed to be
particularly true in the case of incremental shifts in strategy. Only when severe
disruptive changes in strategy occur must structure once again follow. Thus, a
reciprocal relationship exists between strategy and structure.

Amburgey and Dacin (1994) posited that “the link from strategy to structure is

based on efﬁciency and effectiveness, and the link from structure to strategy is based

 

 

44

on the evolution of managerial cognitions and skills” (p.1432). Results from their
study support Chandler’s position, structure follows strategy:

Our results indicate that reciprocity is not the same as equality;

changes in strategy are more likely to produce changes in

structure than the reverse, and changes in strategy produce

changes in structure more quickly than the latter produce changes

in strategy (p. 1448).

In contrast, Keats and Hitt (1988) were unable to support the structure-follows-
strategy link. They found that structure affects strategic choice, with the latter being
limited by the type of information gathered. However, an important distinction
between the Amburgey and Dacin study and the Keats and Hitt work is that temporal
considerations were built into the Amburgey and Dacin design. Keats and Hitt note
this limitation in their research.

In the logistics literature there is no guidance about reciprocity in the strategy-
structure relationship. An interesting consideration is that most of the work in the
organizational theory and strategic management level is done at an overall corporate or
strategic business unit level. There is little written about a strategy-structure linkage at

the functional level. Given the uncertainty, Chow et al. have indicated they believe in

a reciprocal relationship at the logistics level.

ENVIRONMENT-STRATEGYSTRUCTURE-PERFORMANCE
The environment-strategy-structure-performance paradigm has been criticized
for being implemented in research designs by single contingency/bivariate approaches
(Galunic and Eisenhardt 1994). Galunic and Eisenhardt cite the example of a single

contingency -- strategy -- requiring a speciﬁc structural arrangement to obtain desired

45

performance. They argue that this interaction of strategy and structure does not capture
the complexity of either the contingency variable or the structural adjustment.

Studies that comprehensively cover the entire paradigm are relatively few in
number. Their results are generally not directly comparable due to differences in
performance measures, industry settings, etc. At best, mixed results have been
obtained on the various relationships within the paradigm.

But the underlying reasoning of the ESSP paradigm continues to attract interest.
It states that where "ﬁt" exists between environment and strategy and between strategy
and structure, better performance should result. Chow et al. advocate use of the
paradigm at the logistics level to develop and facilitate comparability among studies.
Although no logistics study to date has tested the entire paradigm, components have
been examined. For example, environmental/strategy (McGinnis and Kohn 1993),
environmental/strategy/structure (Germain and DrOge 1995; Germain, DrOge, and
Daugherty 1994), environmental/structure (Germain, DrOge, and Daugherty 1994), and
structure/performance (DrOge and Germain 1989) linkages have appeared in the
logistics literature. In each case, organizational theory concepts have been integrated
into the respective studies without any apparent problems. This is an important
consideration as researchers should be concerned whether theoretical developments in
other ﬁelds can be imported wholesale into a different ﬁeld. The results of the
aforementioned studies have generally supported the ﬁndings in related organizational
theory and strategic management literature, further buttressing the use of a
contingency-based strategy, structure, and performance approach to logistics research.

If that is true, a brief review of comprehensive studies in organizational theory and

46

strategic management should provide some insights. into what might be reasonably
expected in the logistics area.

Prescott (1986) found that strategy and performance were related and that
environment modiﬁed the strength but not the form of the relationship. Davis and
Schul (1993) found a particularly interesting relationship between strategy and
performance when examining resource sharing. Resource sharing was deﬁned by “the
extent to which a business unit shares functional activities like logistics with other
business units” (p. 185). Davis and Schul found that such sharing positively inﬂuences
performance in ﬁrms pursuing an overall corporate strategy of lowest cost. No such
relationship was found for ﬁrms pursing a differentiation strategy.

Miller’s (1988) research produced several results with potential signiﬁcance to
logistics research. Using Porter’s strategies, a number of hypotheses were developed
regarding the environment and structure -- particularly integrative mechanisms -- and
performance. The relationship of Porter’s strategy to environment was consistent with
previous literature (i.e., differentiation is associated with high environmental
uncertainty, low cost with stable environments). The structural property of integrative
devices received support and was found most strongly in uncertain environments using
the differentiation strategy. Together, these results point to the more successful
differentiated ﬁrms employing these tactics than the unsuccessful ﬁrms. It again
underscores the issue of overall corporate or business unit strategy and its inﬂuence, if
any, on logistics strategy and structure.

In a unique approach, Keats and Hitt (1988) used three different models to test

variables of interest from the organizational science and strategic management

47

literatures. The external control model found environmental instability to have a
signiﬁcant impact on organizational structure, consistent with other ﬁndings mentioned
previously. The strategic management model found that performance was actually
negatively impacted by ﬁrms trying to reduce uncertainty by curtailing their
diversiﬁcation efforts. This supports Miller and Friesen’s (1978) ﬁnding that suggests
some ﬁrms intentionally pursue uncertain environments and still maintain high levels of
performance. This again raises an interesting question from the logistics perspective:
What is the appropriate strategy-structure ﬁt in response to an organization pursuing
market opportunities in continuously uncertain environments? The third model used by
Keats and Hitt is the inertia model. Its primary ﬁnding of interest to this research is
that strategy follows structure.

The ESSP paradigm has been utilized in examining service organizations as well
as manufacturing or retailing industries. Jennings and Seaman (1994) studied the
savings and loan industry. They speciﬁcally looked at ﬁrms’ adaptive characteristics in
relation to the Miles and Snow (1978) typology, then compared ﬁnancial performance.
Their research supports the idea of a ﬁt between strategy and structure leading to better
ﬁrm performance.

The preceding discussion is not intended to be exhaustive. It is intended to
illustrate that: (1) complete examination of the ESSP paradigm in one study is still
infrequent, (2) ﬁndings from the entire paradigm do support narrower studies in the
organizational theory and strategic management literature that only focus on selected
elements and relationships within the paradigm, and (3) studies reported in the logistics

literature are consistent in the methods and ﬁndings of organizational theory and

48

strategic management, thereby supporting the use of the paradigm in logistics research.

LOGISTICS CONTINGENCY MODEL
Chow et al. ’5 general conceptual framework is reproduced in Figure 2.2. As
discussed in the preceding section, much of the framework is based upon organizational
science research. However, certain variables and relationships are not adequately
deﬁned or are subject to different interpretations when viewed from a logistics
perspective. These concerns suggest that a competing framework may be better suited
to conduct research in logistics using the environment-strategy-structure-performance

paradigm .

THE ESSP PARADIGM - CHOW ET AL.

Contingency
Variables

- / \
F... m p

 

 

 

 

 

 

Performance |

 

 

Integration l'—‘
Figure 2.2

Logistics Organization and Performance

An alternative framework, termed Logistics Contingency Model, is presented in
Figure 2.3. As indicated in Figure 2.3, the Logistics Contingency Model contains all
of the major components traditionally used in the organizational science literature. The
details of each major component are brieﬂy described in this section. Research

hypotheses are addressed in Chapter Three.

49

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

- - -CPPWX! _. .....
; Environmental "'
7—4; Dynamism : l ‘
_ Environmental q ;
: Hostility "P Logistics H.R. _’ Logistics
. Logistics I.T. 41> Structure Performance
Operating Systems I
' I
.~ Logistics I.T. f‘ ' :
§~ Planning Systems ' :
F ----------- e -------- ‘2.
g Logistics <_ _ _ _ _ _ _|
1 Strategy
)
Figure 2.3

Logistics Contingency Model

CONTEXT

Although “environment” was originally used in the ESSP paradigm, context is
used here for two reasons. First, the term context has a broader meaning, extending
beyond traditional “environmental” variables. Given the two information technology
variables, context more accurately depicts the nature of the relationships under
investigation. Second, in much of the work done in this area in logistics, context is
more commonly used (e. g., Germain 1996; Germain and DrOge 1994). It is hoped that
a standard terminology will reduce confusion in continued research.

Four contingent contextual variables are contained in the model. They are:
(1) environmental dynamism, (2) environmental hostility, (3) logistics information
technology operating systems, and (4) logistics information technology planning
systems. The ﬁrst two are well known in the organizational science literature and were

discussed earlier under “environmental uncertainty.” However, to avoid any confusion

50

with the proliferation of environmental uncertainty characteristics, they are precisely
identiﬁed in the model as dynamism and hostility. In this sense, they retain the
meanings as developed in the organizational science literature. Their inclusion
provides an opportunity to examine whether environmental variables affect functional
level components in the same manner as they have affected corporate level components
in other studies. In addition, it is hoped they can be compared to other studies
conducted in logistics.

Chow et al. (1995) suggest that information technology has become critical to
the functioning of logistics. They cite Schary and Coakley (1991) in noting that the
nature of logistics organization is being changed by information technology.
Presumably, strategic considerations are also inﬂuenced by the deployment of
information technology. For these reasons, two aspects of logistics information
technology -- operating (i.e., tactical) and planning systems -- are included in the

Logistics Contingency Model.

STRATEGY

The Bowersox and Daugherty logistics strategy typology is used in the model.
The strategies of Process, Market and Channel have received empirical support
(Clinton and Closs 1997). Their organizational emphasis is consistent with the ESSP

framework and represent logistics strategies - not overall corporate strategy.

STRUCTURE
Four dimensions of structure are used in the Logistics Contingency Model:

integration, formalization, centralization, and span of control. Chow et al.’s

51

conceptualization of integration as an outcome variable is rejected. Although their
argument is interesting, the long history of integration as a structural property and the
recent investigations of researchers in the area of integrated logistics (e.g., Stank,
Daugherty and Gustin 1994) are more persuasive. Formalization and centralization are
well-known structural properties and have received support in logistics research (e.g.,
DrOge and Germain 1989; Germain, DrOge and Daugherty 1994). The traditional
deﬁnition of span of control -- number of subordinates under a superior -- is used in

this research.

PERFORMANCE

As indicated earlier in this chapter, performance has been measured in many
different ways in the organizational science literature as well as in logistics. The more
“global” measures (e.g., market share, proﬁtability) typically used in the
organizational science literature are inappropriate when investigating logistics
performance. For that reason, logistics-speciﬁc measures were used. A “soft”
construct based on perceptual measures of logistics performance (e.g., delivery
dependability, order ﬁll capacity) was developed and used.

An additional feature of the model includes a feedback loop from performance
to strategy. Ginsberg and Venkatraman (1985) advocated this path in their model of a
systems model of contingency theory-based strategic research. They suggested that
performance can be considered a contingency variable that then affects the range of a

ﬁrm’s strategy responses.

52
This is consistent with the World Class Logistics model. This model, depicted

as a circle or wheel, represents the dynamic nature of a business. As ﬁrms make
adjustments to changes (i.e., contingent situations) they strive to measure their
performance in relation to the changes. But the act of performance measurement then
causes further adjustments and assessments. The Logistics Contingency Model ’3

feedback loop captures this dynamic nature of the business environment.

SUMMARY

This chapter introduced relevant sources in contingency theory and the variables
speciﬁed by Chow et al. in their framework. Where appropriate, other variables were
introduced. An overview of the ESSP framework was provided and an alternative
model, the Logistics Contingency Model, was introduced. The various items
comprising the major components of the model were brieﬂy described. This review
and model provide a basis from which research questions and hypotheses can be framed
in a manner consistent with a contingency approach to the ESSP paradigm in logistics

research.

CHAPTER THREE - RESEARCH METHODOLOGY

This chapter describes research methodology. It contains three sections. First,
the research purpose and objectives are presented. The second section then details the
speciﬁc research questions. The third section addresses the Operationalization of the
methodology. This particular section discusses the unit of analysis, sample selection,

interview protocol, data collection, and analyzes.

RESEARCH PURPOSE AND OBJECTIVES

The purpose of this research was to investigate if logistics structure and
performance were contingent upon speciﬁed logistics context and strategic orientation
and to what extent, if any, that subsequent performance evaluation impacted strategy.
The research procedure consisted of two related approaches. The ﬁrst approach
utilized the database generated from the World Class Logistics research. A sample of
approximately sixty (60) American and Canadian manufacturers and retailers was used
to explore relationships hypothesized by the proposed Logistics Contingency Model.
The second approach used additional survey data generated from a select group of
approximately forty (40) World Class Logistics research ﬁrms. This second database
permitted time series analysis of the dynamic nature of the hypothesized performance

-- > strategy link.

53

54

The speciﬁc objectives of this research were:

(1)

(2)

(3)

(4)

Test empirically if context variables directly or indirectly affected
strategy, structure and performance.

Test if a relationship, possibly reciprocal, existed between structure
and strategy.

Test if there was any support for a feedback mechanism whereby
performance inﬂuenced strategy.

Suggest reﬁnements to the proposed CSSP model in an effort to

develop additional hypotheses and directions for future logistics
research.

RESEARCH QUESTIONS

In accordance with the preceding objectives and review of the literature, the

following ﬁfteen research questions were addressed:

(1)
(2)
(3)
(4)
(5)

(6)

(7)

(8)

(9)
(10)

What is the relationship of environmental dynamism to logistics strategy?
What is the relationship of environmental dynamism to logistics structure?
What is the relationship of environmental hostility to logistics strategy?
What is the relationship of environmental hostility to logistics structure?

What is the relationship of logistics information technology operating
systems to logistics strategy?

What is the relationship of logistics information technology operating
systems to logistics structure?

What is the relationship of logistics information technology planning
systems to logistics strategy?

What is the relationship of logistics information technology planning
systems to logistics structure?

What is the relationship of context to logistics strategy?

What is the relationship of context to logistics structure?

(11)

(12)

(13)

(14)

(15)

55

What indirect relationships, if any, exist between the contextual variables,
strategy, and/or structure and performance?

Is there evidence of a reciprocal relationship between strategy and
structure?

What is the direct relationship, if any, between logistics structure and
performance?

What is the relationship of logistics strategy on performance as mediated
by structure?

Does the level of logistics performance affect logistics strategy?

RESEARCH HYPOTHESES

Environmental dynamism has routinely been found to be signiﬁcantly related to

strategy and structure (Miller and Friesen 1978). Generally-speaking, the

organizational science and strategic management literatures have found high levels of

perceived environmental dynamism to be linked to decentralized, less formalized,

“organic” structures and differentiated (i.e., Porter) or prospector (i.e., Miles and

Snow) strategies. It was expected that comparable ﬁndings would be evident at the

logistics level. This research used the environmental dynamism measure developed by

Miller and DrOge (1986).

Hypothesis 1a: Under conditions of high environmental

dynamism, ﬁrms will choose more externally-oriented logistics
strategies (i. e., market or channel) than under conditions of low
environmental dynamism.

Hypothesis 1b: Under conditions of high environmental
dynamism, ﬁrms will use more decentralized structures than
under conditions of low environmental dynamism.

Hypothesis I c: Under conditions of high environmental
dynamism, ﬁrms will use less formalized structures than under
conditions of low environmental dynamism.

56

Hypothesis Id: Under conditions of high environmental
dynamism, ﬁrms will use more integrated structures than under
conditions of low environmental dynamism.

Hypothesis 1e: Under conditions of high environmental

dynamism, ﬁrms will use narrower spans of control than under

conditions of low environmental dynamism.

Environmental hostility is the second dimension of Miller and Friesen’s (1978)
environment construct. It focuses on the range of behaviors and has been found to
affect a ﬁrm’s structure.

Hypothesis 2a: Under conditions of high environmental

hostility, ﬁrms will choose more externally-oriented logistics

strategies (i. e. , market and channel) than under conditions of low

environmental hostility.

Hypothesis 2b: Under conditions of high environmental

hostility, ﬁrms will use more decentralized structures than under

conditions of low environmental hostility.

Hypothesis 2c: Under conditions of high environmental

hostility, ﬁrms will use less formalized structures than under

conditions of low environmental hostility.

Hypothesis 2d: Under conditions of high environmental

hostility, ﬁrms will use more integrated structures than under

conditions of low environmental hostility.

Hypothesis 2e: Under conditions of high environmental

hostility, ﬁrms will use narrower spans of control than under

conditions of low environmental hostility.

Information technology has only recently been examined in the logistics
literature. As indicated previously in Chapter Two, there is little guidance in the
technology literature concerning information technology speciﬁcally. For the purposes

of this research, information technology was deﬁned rather broadly to encompass both

logistics operating and planning systems in the manner used by Closs et al. (1997).

57

Hypothesis 3a: Under conditions of highly capable logistics
information technology operating systems, ﬁrms will choose more
externally-oriented logistics strategies (i. e. , market and channel)
than under conditions of less capable logistics information
technology operating systems.

Hypothesis 3b: Under conditions of highly capable logistics
information technology operating systems, ﬁrms will use
decentralized structures than under conditions of less capable
logistics information technology operating systems.

Hypothesis 3c: Under conditions of highly capable logistics
information technology operating systems, ﬁrms will use more
formalized structures than under conditions of less capable
logistics information technology operating systems.

Hypothesis 3d: Under conditions of highly capable logistics
information technology operating systems, ﬁmis will use more
integrated structures than under conditions of less capable
logistics information technology operating systems.

Hypothesis 3e: Under conditions of highly capable logistics
information technology operating systems, ﬁrms will use broader
spans of control than under conditions of less capable logistics
information technology operating systems.

Hypothesis 3f: Under conditions of highly capable logistics
information technology planning systems, ﬁrms will choose more
externally-oriented logistics strategies (i. e., market and channel)
than under conditions of less capable logistics information
technology planning systems.

Hypothesis 3g: Under conditions of highly capable logistics
information technology planning systems, ﬁrms will use more
decentralized structures than under conditions of less capable
logistics information technology planning systems.

Hypothesis 3h: Under conditions of highly capable logistics
information technology planning systems, ﬁrms will use more
formalized structures than under conditions of less capable
logistics information technology planning systems.

58

Hypothesis 3i: Under conditions of highly capable logistics

information technology planning systems, ﬁrms will use more

integrated structures than under conditions of less capable

logistics information technology planning systems.

Hypothesis 3j: Under conditions of highly capable logistics

information technology planning systems, ﬁrms will use broader

spans of control than under conditions of less capable logistics

information technology planning systems.

In addition to the individual main effects of each contextual variable on strategy
and structure, the interaction effects of an overall “context” effect were considered.
This addressed the issue of whether ﬁt between all of the contextual variables was the
primary determinant of, respectively, strategy and structure (Venkatraman 1989).

Hypothesis 3k: With respect to strategy, there will be a

signiﬁcant interaction ejfect between the levels of environmental

dynamism, environmental hostility, information technology

operating systems, and information technology planning systems.

Hypothesis 31: With respect to structure, there will be a

signiﬁcant interaction ejfect between the levels of environmental

dynamism, environmental hostility, information technology

operating systems, and information technology planning systems.

Indirect effects are studied within the ESSP paradigm. In the strategic
management literature the focus is usually on the environment-strategy-performance ﬁt
while the organizational theory literature has a tendency to look at environment-
structure-performance. The strategy, structure, and performance paradigm combines
the two approaches. Environmental dynamism is again the most frequently used
“environmental” variable. Mixed results have been obtained in previous research
concerning its impact on performance via strategy. Environmental complexity has

fallen from use due to the inability of earlier researchers to establish causal

relationships. The two remaining contextual contingency variables, logistics

59

information technology operating and planning systems, have no precedent in this area.

Hypothesis 4a: Structure is a signiﬁcant mediator of the
environmental dynamism -- > performance relationship.

Hypothesis 4b: Structure is a signiﬁcant mediator of the
environmental hostility -- > performance relationship.

Hypothesis 4c: Structure is a signiﬁcant mediator of the
logistics information technology operating systems -- >
performance relationship.

Hypothesis 4d: Structure is a signiﬁcant mediator of the
logistics information technology planning systems -- >
performance relationship.

Hypothesis 4e: Strategy is a signiﬁcant mediator of the
environmental dynamism -- > structure relationship.

Hypothesis 4f: Strategy is a signiﬁcant mediator of the
environmental hostility -- > structure relationship.

Hypothesis 4g: Strategy is a significant mediator of the
logistics information technology operating systems -- > structure
relationship.
Hypothesis 4h: Strategy is a signiﬁcant mediator of the
logistics information technology planning systems -- > structure
relationship.

Chow et al. conceptualized the strategy and structure relationship as reciprocal
in nature. The discussion in Chapter Two indicated some support for this position
(e.g., Hall and Salas 1980; Miller 1987) though many authors adhere to Chandler’s
(1962) position that structure follows strategy. It is generally agreed, however, that a
ﬁt between strategy and structure is indispensable to high levels of performance. The
following hypotheses focused on the logistics strategy-logistics structure ﬁt speciﬁcally

as well as its overall performance. In addition, the contingent relationship whereby

performance inﬂuenced strategy was considered (Ginsberg and Venkatraman 1985).

structure and performance.

support, particularly of a structure-performance association.

60

Hypothesis 5a: Externally-oriented logistics strategy (i. e.,
market or channel) is associated with less centralized structure
than internally-oriented logistics strategy (i. e. , process).

Hypothesis 5b: Etternally-oriented logistics strategy (i. e. ,
market or channel) is associated with less formalized structure
than internally-oriented logistics strategy (i. e. , process).

Hypothesis 5c: Externally-oriented logistics strategy (i. e. ,
market or channel) is associated with more integrated structure
than internally-oriented logistics strategy (i. e., process).

Hypothesis 5d: Externally-oriented logistics strategy (i. e. ,
market or channel) is associated with broader spans of control
structure than internally-oriented logistics strategy (i. e. , process).

Hypothesis 5e: Under conditions of satisfactory
performance, less strategic changes will occur than under
conditions of unsatisfactory performance.

Finally, there was the question of a relationship between logistics strategy,

logistics area has empirically supported the existence of such a relationship (Drt’ige and

Germain 1989; Germain 1989).

Hypothesis 6a: Logistics structure, by itself, has no
signiﬁcant relationship to performance.

Hypothesis 6b: Logistics structure will be a signiﬁcant
mediator of the logistics strategy -- > performance relationship.

Dalton et al. (1980) noted that there was little empirical

Subsequent work in the

61
METHODOLOGY OPERATIONALIZATION

This section describes how this dissertation’s research methodology was
conducted. The relevant unit of analysis is identiﬁed and described, the sample size
and sample characteristics are discussed, and proposed data collection and analyses are

reviewed.

UNIT OF ANALYSIS

For the purposes of this research, the unit of analysis was the ﬁrm’s logistics
operations. Encompassed in this unit of analysis was the ﬁrm’s perceived view of the
external logistics environment and its internal logistics planning, control, and
measurement mechanisms. It was anticipated that examination of the environment and
mechanisms would reveal the pattern of relationships within the strategy, structure, and

performance framework.

SAMPLE SIZE AND CHARACTERISTICS

All ﬁrms involved in this research were drawn from the World Class Logistics
research database. Speciﬁcally, the ﬁrms must have participated as a case study or
been involved as a fax respondent completing relevant sections of the case study
workbook. Additionally, only manufacturing and retailing ﬁrms were included in this
analysis. As a result, the actual sample size for the main body of this analysis was
forty-eight (48) case study ﬁrms and nine (9) fax respondents, an effective sample of
fifty-seven (57) ﬁrms. This ﬁrst sample was used to statistically test the Logistics

Contingency Model.

62

The time-series section of this research contains a smaller sample. Due to
original respondents changing or vacating positions, ﬁrm acquisitions, etc., only ﬁfty
(50) ﬁrms were eligible for the follow-up questionnaire associated with the time-series
analysis. Of the ﬁfty ﬁrms, usable responses were received from thirty-nine (39)
companies.

This smaller sample was used to closely examine the hypothesized relationship
of performance impacting strategy (Ginsberg and Venkatraman 1985). Participating
ﬁrms were asked to complete a limited set of questions originally asked in the World
Class Logistics workbook. This matched set of responses at different time intervals
facilitated a time-series analysis, thereby allowing investigation of the dynamic nature
of the Logistics Contingency Model.

Other than the aforementioned restriction limiting analysis to manufacturing and
retailing ﬁrms, there were two other characteristics of note. First, all of the case study
ﬁrms were identiﬁed by expert opinion as leading practitioners of logistics. Thus, they
were not randomly chosen. Also, given that these ﬁrms were generally considered
among the top ten percent logistics performers they were not representative of the
general population. Additionally, the case study ﬁrms and fax respondents were all
members of the Council of Logistics Management, presumably indicating a greater
professional awareness and interest in logistics than a random sample of manufacturing
and retailing ﬁrms. This potential bias was recognized. However, given the intent of
the research - to identify whether certain relationships exist in logistics - one was more
likely to ﬁnd such relationships in highly motivated logistics organizations than in the

general population.

63

The second characteristic was ﬁrm size. Although there was no explicit
measure built into the research to control for ﬁrm size, a cursory review of the ﬁrms
involved identiﬁed most of them as Fortune 500 ﬁrms. Therefore, they would be
generally characterized as large ﬁrms.

Firms were drawn from numerous industries. There was no single dominant
industry represented in this research. Previous research in logistics has routinely found
that there was no discernible difference between ﬁrms from different industries in terms

of their overall logistics practices and capabilities.

STATISTICAL ANALYSIS

The statistical analyses performed in this research were based on data gained
from the original case study workbook and subsequent replication of a limited number
of questions in a follow-up survey. The case study workbook was a twenty-four page
document divided into seven sections: (1) Background, (2) Strategy, (3) Performance
Competencies, (4) Organization, (5) Performance Measurement, (6) Information
Technology, and (7) Alliances. It is reproduced in its entirety in Appendix I. This
research drew upon all sections except for ‘Alliances’. Speciﬁc questionnaire items are
discussed in relationship to the results presented in Chapter Four.

The research questions and related hypotheses were presented earlier. These
hypotheses were investigated using t-tests, chi-square, analysis of variance, and
regression analysis. Regression analysis was used to test for the existence of signiﬁcant
mediational relationships between the variables and constructs. Given a mixture of new

and established measures, CL = .10 was used to test for statistical signiﬁcance.

64
RATIONALE OF OVERALL DESIGN

The overall design of this research was driven by two facts: (1) in the area of
organizational science and strategic management, a considerable contingency-based
strategy, structure, and performance literature exists, and (2) most of the literature is
not based on task or function-based research.

The ﬁrst fact permitted insights into the relationships conceptualized by the
Logistics Contingency Model. By taking advantage of prior work, this research was
able to generate hypotheses at an early stage. The World Class Logistics database
contained a majority of the variables used in the framework. This permitted statistical
analysis with the intent of evaluating the efﬁcacy of transferring organizational science
and corporate-level concepts to logistics research with a much more functional
perspective. Previous work by DrOge, Germain, and Daugherty (1994) strongly
suggested that this transferal would work.

It was fully expected that the statistical testing would lead to more questions
than answers. Such was the expected result of this design. In that sense, this research
served to both validate the logic of the general model and identify its strengths and

weaknesses when applied to logistics.

CHAPTER FOUR - RESULTS

This chapter presents the research results. Discussion of the results is reserved
for Chapter Five. Chapter Four is composed of the following sections: (1) assessment
of construct unidimensionality and reliability; (2) evaluation of research questions and

hypotheses; and (3) a brief summary of the results.

ASSESSMENT OF CONSTRUCT UNIDIMENSIONALITY AND RELIABILITY

A major focus of this research concerns the use of latent constructs in logistics
research. A limited amount of logistics research has used latent constructs (e.g., DrOge
and Germain 1989; Germain 1989, 1996; Germain, DrOge and Daugherty 1994). In
this particular research the relationship of latent constructs and item indicators is
illustrated in Figure 4.1. For the purposes of this research, the focus is on the
secondary constructs (as indicated in Figure 4.1) and their associated items. To avoid
confusion, unless otherwise noted, any reference to constructs throughout the remainder
of this chapter means secondary constructs.

In order to use latent constructs in logistics research, it is necessary to establish
the unidimensionality and reliability of such constructs. This section details the results
of a multistep process used in the evaluation of this research’s context and structure

COI'lStI'llCtS .

65

66
Theoretigal Structure

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Questionnaire Items
anary secondary — Questionnaire Items
Construct Constructs

Questionnaire Items

Exam 1;
' ° Questionnaire Item
Structure Centralization ‘ .
Formalization Questionnaire Item
Integration Questionnaire Item
Span of Control Questionnaire Item
Figure 4. 1

Construct Relationships

Following the guidelines of Gerbing and Anderson (1988), the various
constructs were examined using a three-step procedure. First, exploratory factor
analysis (EFA) was conducted. EFA is typically used to determine the underlying
structure of data in exploratory phases of research. Although the organizational science
and logistics literatures .provide sufﬁcient theoretical basis for some of this research’s
constructs, others are exploratory or relatively new to logistics applications. For
consistency, EFA was used in this research to further substantiate existing theoretical
constructs or to aid in the development of new constructs.

The second step in the Gerbing and Anderson paradigm calls for conﬁrmatory
factor analysis (CF A). This step allows for each item to be simultaneously evaluated,
thereby directly addressing unidimensional and discriminant measurement. The third

step, assessment of construct reliability, is determined through coefﬁcient alpha (i.e.,

67
Cronbach’s alpha). Results of the three-step procedure for each construct group are

presented next.

CONTEXT
Environmental Dynamism and Environmental Hostility

As discussed in Chapter Two, environmental dynamism and environmental
hostility are two related concepts that have been used extensively in organizational
science research. Their usage in logistics research is limited (e.g., Kohn et al.
1990).Table 4.1 reproduces the “background” section of the workbook in which
respondents answered questions concerning environmental factors. Several of the items
(e.g., #1, #2, #3) have been frequently used to measure elements of environmental
uncertainty. Therefore, it was anticipated that factors representing environmental
dynamism and environmental hostility would emerge from a series of EFAs.

Using the common criteria of a .40 loading requirement, minimum eigenvalues
of one, and examination of scree plots, successive EFAs were examined. The number
of items was reduced to eight. Items 4, 5 and 10 were eliminated. Table 4.2 contains
the results of the ﬁnal EFA performed on the remaining eight items. Factor loadings

greater than .40 are indicated in bold.

68

Table 4.1
Initial Items in EFA Analysis: Environmental Dynamism and Environmental Hostility

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Question
Item (Left Anchor) 1 2 3 4 5 (Right Anchor)
Minimal time, effort, resources and Much time, effort, resources and
l managerial attention are required managerial attention are required to
to keep up with major competitors. keep up with major competitors.
When decisions are made within my When decisions are made within my
2 ﬁrm, competitor reaction or retaliation ﬁrm, competitor reaction or
is not an important consideration. retaliation i_s an important
consideration.
3 Our major competitors are not Our major competitors a_r_e
particularly aggressive. ﬁercely aggressive.
4 Actions of competitors are E Actions of competitors are difﬁcult
to predict. to predict.
A minimal amount of time is spent A mt deal of time is spent
5 analyzing major competitors’ analyzing major competitors’
strategies and actions. strategies and actions.
Our ﬁrm rarely changes marketing Our ﬁrm f_r_eguently changes its
6 practices to keep up with competitors. marketing practices to keep up with
competitors.
Our ﬁrm must rarely chapgg its Our ﬁrm must change its logistics
7 logistics practices to keep up with practices extremely frequently
competitors. to keep up with competitors.
8 Supplier capabilities change at a very Supplier capabilities change at a very
slow rate. rapid rate.
The rate at which products/services The rate of product/service
9 are becoming obsolete in the obsolescence in the industry is
industry is vegy slow. veg high (e. g., some fashion
goods).
10 Demand for logistics services is Demand for logistics services is
@y to forecast. difﬁcult to forecast.
Production and/or service The modes of production and/or
11 technology is stable and well service change often and in it
established. major way.

 

 

69

Table 4. 2
Factor Loadings: Environmental Dynamism / Environmental Hostility

 

 

Item Factor 1 Factor 2
3 0.72546 0.09191
1 0. 62977 0.01975
2 0.62960 0. 11730
8 -0. 10729 0.54826
11 0. 18382 0. 54826
9 -0. 04002 0. 50827
7 0.26734 0.50435
6 0.25548 0. 48584

 

Variance explained by each factor:
Factor 1: 1.5029
Factor 2: 1.3377

 

Rotation Method: Varimax

 

 

 

Factor 1 represents environmental hostility. As discussed in Chapter Two, each
of the three items comprising environmental hostility concerns the level of competition.
Factor 2 represents the amount and unpredictability of change - commonly
characterized as environmental dynamism. This EFA result is consistent with

organizational science literature.

Infgrmatipn Technology Operating Systems and Infgrmatign Technolggy Planing
Systems

Information technology operating systems concern the day-to-day tasks
performed within a logistics network, whereas the planning systems are concerned with
forecasting and longer range management issues. Therefore, it was expected that the
characteristics associated with each system would differ. The characteristics are

detailed in Table 4.3.

70

Table 4.3
Information Technology Characteristics

 

Using the following scale, respondents rated the level of each characteristic in terms of
their ﬁrm’s logistics operating and planning systems.

Low12345High

 

(l) TIMELINESS

 

(2) ACCURACY

 

(3) AVAILABILITY

 

(4) EXCEPTION BASIS FORMATTED

 

(5) FORMATTED TO FACILITATE USAGE

 

(6) INFORMATION SHARING (Deﬁned as willingness to share common
information across functions within the ﬁrm.)

 

(7) FLEXIBILITY (Deﬁned as ability to adapt processes and capabilities to speciﬁc
customer segment requirements.)

 

(8) INTERNAL CONNECTIVITY (Deﬁned as ability to effectively exchange
information across managerial areas within our ﬁrm.)

 

 

(9) EXTERNAL CONNECTIVITY (Deﬁned as ability to effectively exchange
information with next destination customers and/or suppliers.)

 

Closs, Goldsby and Clinton (1997) have found that these characteristics explain
different capabilities within logistics operating and planning systems. Building upon
their ﬁndings, the nine characteristics were subjected to an EFA conducted
simultaneously across both logistics operating systems and logistics planning systems
(i.e., an initial total of eighteen items). A series of EFAs resulted in three factors,
each meeting the criterion of a minimum eigenvalue of one. These factors are shown
in Table 4.4. Factor loadings greater than .40 are indicated in hold.

The ﬁrst factor is composed of information technology characteristics associated

with logistics planning systems (LPS). Again, logistics planning systems within the

 

71

questiomaire were described as forecasting, inventory management and distribution
requirements planning. As might be expected in a planning environment, Factor 1’s
associated characteristics facilitate the accurate and timely dissemination of plaming
reports in a manner or format which facilitates their usage. Additionally, the planning
system appears to encourage sharing of this information beyond the boundaries of the
responding ﬁrms (i.e., external connectivity). However, external comectivity also
cross-loads on the third factor.

Factor 2 is represented by items from logistics operating systems (LOS).
Flexibility of the information technology operating system is a key requirement if ﬁrms
are to be responsive to immediate operating concerns and conditions. Similarly, having
the information available and being able to exchange this information internally and
externally are key items loading on Factor 2. From a shorter term operational
perspective, the items contained in Factor 2 are logical. Again, it should be noted that
external connectivity cross-loads on Factor 2.

As indicated by the preceding discussion, external connectivity cross-loaded on
Factors 1 and 2. However, external connectivity of planning and operating systems
were the only items loading at .40 or above on Factor 3. Given that the respective
cross-loadings (i.e., 0.53894 on LPS; 0.55175 on LOS) were quite strong - and that
these two items were the only items meeting the loading criterion on Factor 3 - a
decision was made to proceed using only Factors 1 and 2. Clearly external
comectivity was an important item. However, it was still captured in Factors 1 and 2

with no loss of information concerning any other variable in Factor 3.

72
Table 4.4

Factor Loadings: Logistics Operating Systems / Logistics Planning Systems

 

Item Factpr 1 Fmpr 2 Facth 3
Formatted for Usage (LPS) 0.83829 0.09795 0.14130
Information Sharing (LPS) 0.77393 0.23953 0.03968
Accuracy (LPS) 0.76429 0.01909 0.24020
Availability (LPS) 0.74524 0.34165 0.07845
Timeliness (LPS) 0.73953 0.25549 0.27967
External Connectivity (LPS) 0. 53894 0.22186 0.67819
Flexibility (LOS) 0.17643 0.74591 0.03505
Internal Connectivity (LOS) 0.10864 0.73446 0.31598
Availability (LOS) 0.22142 0.69216 0.17473
External Comectivity (LOS) 0.14450 0.55175 0.68120

 

Variance explained by each factor:
Factor 1: 3.3914
Factor 2: 2.1779
Factor 3: 1.2192

 

Rotation Method: Varimax

 

 

 

Based on these EFA results, there is tentative support for the four secondary
constructs -- environmental dynamism, environmental hostility, logistics information
technology operating systems, and logistics information technology planning systems.
The next step in the process involved assessment via conﬁrmatory factor analysis
(CFA).

Within this research, CFA was designed to test the discriminant validity
between the four secondary constructs as well as the unidimensionality of each
secondary construct. The graphical representation of the overall context construct and

the afﬁliated secondary constructs is provided in Figure 4.2.

73
Figure 4.2 is depicted in EQS nomenclature as that program was used to test the

relationships. Parameter estimates for the model were calculated from the covariances

of the eighteen indicator variables. The variances of the common factors (i.e., F1-F4)

were ﬁxed at 1.0, thereby standardizing the unit variance of the latent factors. The

remaining paths were set free to be estimated. As noted by Bentler (1992) this is a

common approach in CFA. The program was run using the generalized least squares

(GLS) method using a sample of 46 cases. The corresponding indicators (i.e., V1-

V18) were as follows:

V1:

V2:

V3:

V4:

V5:

V6:

V7:

V8:

V9:

V10:

V11:

V12:

Much time, effort, resources and managerial attention are required to keep up
with major competitors. (Right-hand anchor of 1-5 scale.)

When decisions are made within my ﬁrm, competitor reaction or retaliation jg
an important consideration. (Right-hand anchor of 1-5 scale.)

Our major competitors ar_e ﬁercely aggressive. (Right-hand anchor of 1—5 scale.)

Our ﬁrm frequently changes its marketing practices to keep up with
competitors. (Right-hand anchor of 1-5 scale.)

Our ﬁrm must change its logistics practices extremely frequently to keep up
with competitors. (Right-hand anchor of 1-5 scale.)

Supplier capabilities change at a rapid rate. (Right-hand anchor of 1-5 scale.)

The rate of product/service obsolescence in the industry is vei_’y high. (Right-
hand anchor of 1-5 scale.)

The modes of production and/or service change often and in a major way.
(Right-hand anchor of 1-5 scale.)

 

Availability (LOS) - (Measured Low/High on 1 — 5 Scale)
Flexibility (LOS) - (Measured Low/High on 1 - 5 Scale)
Internal Connectivity (LOS) - (Measured Low/High on 1 - 5 Scale)

External Connectivity (LOS) - (Measured Low/High on 1 - 5 Scale)

74
V13: Timeliness (LPS) - (Measured Low/High on 1 - 5 Scale)

V14: Accuracy (LPS) - (Measured Low/High on 1 - 5 Scale)

V15: Availability (LPS) - (Measured Low/High on 1 - 5 Scale)

V16: Formatted to Facilitate Usage (LPS) - (Measured Low/High on 1 - 5 Scale)
V17: Information Sharing (LPS) - (Measured Low/High on 1 - 5 Scale)

V18: External Connectivity (LPS) - (Measured Low/High on 1 - 5 Scale)

Complete results of this initial CFA analysis are presented in Appendix C. The
results indicated that several variables should not be included in the secondary
constructs. Based on these results, several revised models were tested, each dropping
one individual indicator successively. Indicators V2, V9 and V12 were ultimately
deleted from the original overall context construct. As a check, several other variables
that appeared troublesome in the initial model were dropped. However, their exclusion
negatively affected the subsequent results. The resulting overall context construct is
presented in Figure 4.3. Table 4.4 provides detailed results from the revised context
construct model.

No convergence problems existed in the revised context model. It converged in
thirty iterations, yielding the best ﬁtting measurement equations for the model. The
average off-diagonal standardized residual was an acceptable 0.0880. A plot (not
shown) of the standardized residuals graphically provides evidence of a reasonably
symmetrical distribution. As indicated in Table 4.4, the major ﬁt indices (i.e., chi-
square, normed ﬁt index, etc.) are well above acceptable thresholds. These
characteristics, in conjunction with the other results presented in Table 4.4, provide

ample support for the revised context construct.

75

 

   

 

 

     
 

 

 

  

 

 

 

 

 

 

 

 

   

 

  
 

 

 
 

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

E1 _’ V1
FI
E2 _" V2 Environ.
Hostility
E3 _’ V3
E4 —" V4
E5 _’ V5
F2
E6 -—> V6 Environ.
Dynamism
E7 "" V7
E8 —’ V8
E9 _’ V9
E10_’ V10 F3
Info. Tech.
Ell—' V11 LOS
E12—’ V12
E13—' V13
E14—" V14
EIS—" V15 F4
Info. Tech.
El6_' V16 LPS
El7_’ V17
EI8_' V18 ‘
Figure 4.2

Conﬁrmatory Factor Analysis: Initial Context Construct

 

76

 

 

 

   

 

     
   

 

 

 

 

 

 

 

     
   

 

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

E1 _' V1
Fl
Environ.
Hostility
E3 —’ V3
E4 _’ V4
E5 —" V5
F2
E6 —* V6 Environ.
Dynamism
E7 —" V7
E8 """"' V8
E10 —* V10 [:3
Info. Tech.
E11 __'" V11 LOS
E13 ——> V13
E14 ——9 V14
E15 —" V15 F4
Info. Tech.
E16 ‘—" V16 LPS
E17 —_" V17
E18 —" V18
Figure 4. 3

Conﬁrmatory Factor Analysis: Revised Context Construct

 

Results of Revised Context Construct Model

CONSTRUCT LOADINGS (t values)

Environmental Hostility

 

 

GOODNESS-OF-FIT MEASURES

 

 

 

V1 .538
(4.447)
V3 .579
(5.363)
Environmental Dynamism
V4 .602
(3.987)
V5 .496
(3.049)
V6 .333
(2.874)
V7 .292
(2.003)
V8 .318
(2.190)
Information Technology LOS
V10 .475
(2.818)
V11 .569
(3.194)
Information Technolggy LPS
V13 .976
(8.032)
V14 .813
(5.446)
V15 .932
(6.339)
V16 .824
(6.792)
V17 .808
(6.225)
V18 900

(5:921)

Chi-square ()6) 76.12
Degrees of Freedom 84
Signiﬁcance Level 0.71
Bentler-Bonett Normed

Fit Index (NFI) 0.982
Bentler Bonett Nonnormed

Fit Index (NNFI) 1.002
Comparative Fit Index 1.000

78
Table 4.5 (continued)

Results of Revised Context Construct Model

Standardized Residual Matrix

E

y;

y_6_

yy

V13
V14
V15
V16
V17
V18

0.077
-0.006
0.042
0.034
0.034

0.223
-0.006
0.068
0.027

0.127
0.009
0.016

0.192
-0.113

0.292

79

The revised model provides support for both the unidimensionality and
discriminatory value of the four constructs. Each construct is considered distinct for
the purposes of further evaluation within this research. Of course, it should be noted
that two of the secondary constructs are now measured by only two items.

Table 4. 6
Reliability Estimates: Context’s Secondary Constructs

 

Environmental Hostility
Standardized Cronbach’s Alpha: .75

Environmental Dynamism
Standardized Cronbach’s Alpha: .68

Correlation With Total

V4 0.3928
V5 0.4627
V6 0.4321
V7 0.4528
V8 0.4708

Information Technology - Logistics Operating Systems
Standardized Cronbach’s Alpha: .74

Information Technology - Logistics Planning Systems
Standardized Cronbach’s Alpha: .91

Correlation With Total

V13 0.8141
V14 0.6954
V15 0.7195
V16 0.7788
V17 0.7515

 

 

V18 0.7193

 

80

Finally, the revised secondary context constructs were evaluated for reliability.
The Cronbach’s alpha and item-to-total correlations (where applicable) are provided in
Table 4.5. With the exception of environmental dynamism (.68), the other reliability
measures are well in excess of the generally accepted .70 threshold.

The overall results of the exploratory factor analysis, conﬁrmatory factor
analysis and reliability estimates provide strong support for the revised context
construct. Therefore, it is reasonable to proceed using an overall context variable
composed of environmental hostility, environmental dynamism, information technology
logistics operating systems, and information technology logistics planning systems. For

the remainder of this document any reference to context assumes the revised measure.

STRUCTURE

The four secondary constructs used in the overall structure construct are span of
control, integration, formalization, and centralization. As discussed in Chapter 2, each
of these secondary constructs has been used extensively in organizational science
literature and to varying degrees in logistics research.

Span of Control

 

Span of control is a single item measure in this research. Respondents were
asked: “How many individuals report directly to each of the following executives (do
not include secretaries or assistants)?” The list of executives included: (1) chief
executive of the business unit; (2) senior manufacturing executive; (3) senior
distribution/logistics executive; and (4) senior sales (or marketing) executive. Only the

responses concerning the number of individuals under the senior distribution/logistics

81

executive are used. Previous research has used summated scales based on responses to
each category of executive (e. g., Germain, DrOge, and Daugherty 1994). That
approach is not used in this research as the focus is limited to logistics organization and
does not deal with questions of a broader scope.

Similarly, levels within a ﬁrm are often closely linked with span of control.
However, levels are more appropriately considered part of complexity. For that
reason, even though the information was available, it was not used in this research.
Integration

A number of questions concerning teaming and cross job coordination were
asked in the mail survey. The same or similar questions and/or scales have previously
been used in logistics research (Germain, DrOge, and Daugherty 1994). Their results
indicated two operational measures of integration: (1) integrative committees, and (2)
integrative mechanisms. This set of questions was factor analyzed. A series of EFAs
resulted in a two factor model. The individual items and their respective scales are
presented in Table 4.7. The ﬁnal rotated EFA is presented in Table 4.8.

It must be noted, however, that in this case the second factor did not meet the
minimum eigenvalue requirement. In fact, its eigenvalue of 0.45 is well below the
generally accepted cutoff of 1.0. However, it was decided to retain the second factor
based on the following: ( 1) the scree test (not shown) suggests the factor should be
retained; (2) the items composing the second factor have been previously used in
logistics research; and (3) the subsequent CFA would provide a further test of the

factor’s suitability.

82

Table 4.7
Underlying Indicators of Integration Construct

 

Item I
In my ﬁrm, interdepartmental committees are set up to allow departments to engage in
joint decision-making.
Scale
Strongly Disagree 1 2 3 4 5 Strongly Agree

 

Item 2
In my ﬁrm, task forces or temporary groups are set up to facilitate interdepartmental
collaboration on a speciﬁc project.
' Scale
Strongly Disagree 1 2 3 4 5 Strongly Agree

 

Item 3
In my ﬁrm, liaison personnel exist whose speciﬁc job it is to coordinate the efforts of
several departments for purposes of a project.
Scale
Strongly Disagree 1 2 3 4 5 Strongly Agree

 

Item 4

In my ﬁrm, cross-functional teams make decisions concerning distribution or logistics
strategy.

 

Scale
Strongly Disagree 1 2 3 4 5 Strongly Agree

 

 

Item 5

In my ﬁrm, cross-functional teams make decisions concerning marketiig or sales
strategy.

 

Scale
Strongly Disagree 1 2 3 4 5 Strongly Agree

 

Consistent with the work of Germain, Drbge, and Daugherty (1994), the ﬁrst
factor represents items relating to project or committee groups and the second factor
represents cross-functional team activity and strategic decision-making. However, a

smaller number of measures emerge in this research. Unlike Germain et al. this

 

research includes both manufacturers and retailers. Retailers did not answer two of the
possible questions in the overall set.

considered as a manufacturer/retailer-consistent construct was desired.

83

Table 4. 8

EFA of Integration Construct

For that reason, those questions were not

 

Item 1
Item 2
Item 3
Item 4
Item 5

Faggr 1
0.69650
0.56494
0. 45693
0. 16727
0. 19796

£39112

0.09207
0. 23 193
0.26982
0. 64454
0.61596

 

Variance explained by each factor:
Factor 1: 1.0802
Factor 2: 0.9299

 

 

Rotation Method: Varimax

 

It appears that the distinguishing feature of the two factors is temporal duration.
The ﬁrst factor implies integrative mechanisms that come together periodically as a
group, then return to their respective tasks. The second factor suggests a team that is
together on a daily basis, continually working as a group to foster integrative processes.

In order to fully investigate this subtle difference, a CFA was conducted. The CFA

model is presented in Figure 4.4.

As was the case with the context CFA, the proposed integration construct was
examined using EQS. Vl through V5 correspond to Items 1-5 shown in Table 4.8.

Factor 1 represents the short duration committee or project groups whereas Factor 2

represents the integrated teams working together on a daily basis.

 

84

 

 

 

 

 

 

 

 

 

 

 

 

El —’ V1
E2 "" V2 7* FACTOR]
E3 —’ V3
E4 _’ V4
E5 “" V5

 

 

 

Figure 4.4
Conﬁrmatory Factor Analysis: Integration Construct

The results of the integration construct CFA are presented in Table 4.9. As
indicated in Table 4.9, the major ﬁt indices (i.e., chi-square, normed ﬁt index, etc.)
are well above acceptable thresholds. The factor loadings for each equation are
signiﬁcant.

No convergence problems existed in the model. It converged in twenty
iterations with no appreciable improvement after seventeen iterations. The average off-
diagonal standardized residual was an acceptable 0.0812. The standardized residual
matrix in Table 4.9 indicates that the individual residuals are reasonably small. A plot
(not shown) of the standardized residuals graphically provides evidence of a slightly
leftward skewed distribution. These characteristics, in conjunction with the results
presented in Table 4.9, provide ample support for the two dimensional integration

COI‘I SITUCI.

85

Table 4.9
Results of Integration Construct Model

CONSTRUCT LOADINGS (t values)

Factor 1

V1

V2

V3

Factor 2

V4

V5

Cronbach’s alpha was calculated for both integration factors.

V1
V2
V3
V4
V5

.884
(4.536)

.549
(3.939)

.683
(3.281)

.649
(3.057)

.800
(3.297)

GOODNESS-OF-FIT MEASURES

 

Chi-square ()8) 6.976
Degrees of Freedom 4

Signiﬁcance Level 0.137
Bentler-Bonett Normed
Fit Index (NFI) 0.979
Bentler-Bonett Nonnormed

Fit Index (NNFI) 0.977
Comparative Fit Index 0.991

Standardized Residual Matrix

0.033

—0.011
-0.007
-0.119
-0.091

y;

0.057
-0.107
0.056
0.096

v_3 X4 X2
0.105
0.167 0.166
0.053 0.106 0.085

Factor 1 had a

reliability estimate of .64. Factor 2 had a reliability estimate of .69. Although these

levels were above Nunnally’s (1967) acceptable range of .50 - .60 for exploratory

measures the measures cannot be considered entirely exploratory. They were quite new

in logistics research but have been used in organizational science research. If judged

by the more stringent .70 level commonly used for established measures, the reliability

86

of these integration factors would be considered somewhat marginal.
Formalization

Seven binary (i.e., yes/no) measures were used to measure formalization.
Respondents were asked if logistics had a separate mission statement. Additionally,
respondents were asked to identify if their ﬁrm had a formal benchmarking initiative in
each of the following six areas: (1) asset management; (2) logistics cost; (3)
productivity; (4) customer service (i.e., internal measures); (5) logistics quality; and
(6) customer expectations concerning customer service (i.e., external measures). Each
response was assigned a value of either zero for “ no” or one for “ yes.”

The summated scale was then treated as a continuous scale with a range of zero
to seven. Higher summed totals represented higher levels of formalization. The
reliability of this summated scale was evaluated using the Kuder-Richardson 20 (KR-
20) test of internal consistency. The summated formalization measure had a KR 20
score of .74 (Calculations are provided in Appendix D). K‘R 20 scores in excess of .50
are generally considered indicative of reliable measures (Kaplan and Saccuzzo 1989).

Central ization

 

Centralization is typically described as the vertical locus of decision-making
authority. As brieﬂy described in Chapter Two, the organizational science literature
generally equates centralization with concentrated decision-making authority (i.e.,
authority residing with one or few people in an organization). Closely related to this
concept is the idea that if decision-making is made at corporate headquarters, then the
organization is centralized. If decision-making authority is diffused to ﬁeld locations,

then the organization is considered decentralized.

87

Within the logistics literature, the usage of centralization is not as clear-cut.
Germain (1989), in his study of product standardization and logistics, used line
responsibility to represent centralization: “The sum out of 15 (line responsibility only)
reporting to logistics was assumed to represent the extent to which logistics activities
were centrally consolidated within a single department (p. 24).” This presents a
problem. If a ﬁrm were considered centralized on the basis of “presence” of
responsibility but that responsibility was for the most part at the staff level (i.e., not
centralized according to Germain’s (1989) usage), is the ﬁrm’s logistics centralized or
decentralized? Organizational science researchers would likely argue that if line
personnel control the responsibility then decentralization exists. If, however, staff are
in a position of responsibility and decision-making, then centralization results.

For the purposes of this research, the question of centralization was handled in
the following manner: Consolidation of an activity in a single department was
considered centralization. No differentiation was made based on line or staff
responsibility. The rationale was that once an activity was positioned under the
jurisdiction of a single logistics department, fewer people overall throughout the entire
organization would have decision-making authority over the activity.

Within the questionnaire, respondents were asked to indicate whether a number
of activities (Table 4.10) were part of the formal responsibility of logistics.
Respondents were asked to indicate the number of years each activity had been under
logistics control. If the activity was not part of the formal responsibility of logistics,

respondents were instructed to indicate zero years.

88

Consequently, these responses were recoded as “0 / 1.” A zero represented
that the activity was not part of logistics’ responsibility. A one indicated that the
activity was under logistics’ control - irrespective of the number of years. The sixteen
measures were then summed. Given the binary “0 / 1” scoring pattern, the summated
scale could range from 0 to 16. Again, a KR 20 test was used to assess the measure’s
reliability. The overall KR 20 score for the sixteen items was .88. This was above
acceptable limits.

Table 4.10
Activities Under Logistics Responsibility

 

Sales Forecasting
Sourcing/Purchasiﬁ

Inbound Transportation

Finished Goods Inventory Management
Intra—Company Transportation
Finished Goods Warehousing

Order Processing

Customer Service

Outbound Transportation

Lgistics Systems Planning

Facilities Desiggi

Materials Handling

Logistics Administration

International Logistics

Capital Equipment Procurement

Data Processing for Distribution Apps.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

89
STRATEGY

The Bowersox-Daugherty logistic strategy typology (1987) is used in this
research to operationalize strategy. Clinton and Closs (1997) have investigated the
validity of the typology using data from an earlier phase of this research. Their
ﬁndings support the typology and identify the underlying elements of each strategy
(i.e., process, market and chamel). Additionally, their work involved a much broader
sample of American and Canadian ﬁrms than the present subset. Based on their
ﬁndings, no further construct validation is performed within this research framework.
The deﬁnitions are repeated in Table 4.11.

Table 4.11
Process / Market / Channel Deﬁnitions

 

Process strategy is concerned with managing a broad group of logistics
activities as a value-added chain. Emphasis is on achieving efﬁciency
from managing purchasing, manufacturing, scheduling and physical
distribution as an integrated system.

Market strategy is concerned with managing a limited group of
logistics activities for a multidivision single business unit or across
multiple business units. The logistics organization seeks to make joint
product shipments to common customers for different product groups
and seeks to facilitate sales and logistical coordination by a single-
order invoice.

Channel strategy is concerned with managing logistics activities
performed jointly with dealers and distributors. The strategic
orientation places a great deal of attention on external control.
Signiﬁcant amounts of ﬁnished inventories are typically maintained
forward or downstream in the distribution channel.

Source: Bowersox et al. (1989) Leading Edge Logistics: Competitive
Positioning for the I990s, Oak Brook, IL: Council of
Logistics Management.

 

 

 

 

 

PERFORMANCE

In the survey instrument, respondents were asked a series of questions
concerning thirty-two (32) performance measures (Appendix 1 - Section Three -
Relative Performance Competencies). Respondents provided perceptual measures of
importance (Least Important = 1; Most Important = 5) and performance (Worse than
Competitors = 1; Better than Competitors = 5). Based on previous studies and case
study responses, eight measures were selected to represent the performance construct.
These measures are shown in Table 4.12.

Table 4.12
Indicators of Performance / Importance

 

 

 

 

 

 

 

 

 

 

PERFORMANCE COMPETENCY MEAN
Advanced Shipment Notiﬁcation 3.73
Delivery Dependability 4.75
Delivery Speed 4.15
Delivery Time Flexibility 4.19
Low Logistics Cost 4.17
Order Fill Capacity 4.65
Order Flexibility 3.67
Responsiveness to Key Customers 4.45

 

 

 

As indicated by the mean scores in Table 4.14, each performance competency is
considered to be important. In fact, delivery dependability, order ﬁll capacity and
responsiveness to key customers were rated the three most important performance
competencies of the thirty-two original measures. The overall reliability of this
performance construct is .62. As this construct is somewhat exploratory this reliability

is acceptable. The summated score of the eight measures is used in all of the statistical

91

analyzes that follow.

SUMMARY OF PRIMARY AND SECONDARY CONSTRUCTS
The preceding section of Chapter 4 has critically examined and explained the
primary and secondary constructs detailed in the Logistics Contingency Model. It has
been demonstrated that the constructs are acceptable in terms of empirical research.
Using these constructs, the remainder of this chapter is devoted to results pertaining to

the research questions and the associated research hypotheses.

EVALUATION OF RESEARCH QUESTIONS AND HYPOTHESES
This research addressed ﬁfteen research questions. In most cases there are
multiple research hypotheses attached to each question. The following section presents

the statistical approach and results pertaining to each research hypothesis.

ENVIRONMENTAL DYNAMISM

Environmental dynamism was evaluated on a summated scale as discussed
earlier. The sum was then divided by the number of items to return to a 1 - 5 scale
with the right-hand anchor representing higher levels of dynamism. The mean
summated response for environmental dynamism was 3.21, indicating a slightly
dynamic average environment. Two groups were created from this continuous
variable. Respondents answering at the mean or below were considered to be
experiencing comparatively low levels of environmental dynamism. Respondents
indicating a response above the mean of 3.21 were considered to be experiencing high

levels of environmental dynamism. This two group categorical variable was then used

92

in the statistical analysis of environmental dynamism and strategy and structure.

Hypothesis 1a: Under conditions of high environmental

dynamism, ﬁmis will choose more externally-oriented logistics

strategies (i. e., market or channel) than under conditions of low

environmental dynamism.

Due to small sample sizes, market and channel strategies were combined into
one group representing external-oriented logistics strategies. Therefore, the two levels
of environmental dynamism were examined in terms of process (i.e., internally-
oriented) and externally-oriented strategies. Given the presence of two categorical
variables, the categorical ANOVA program (CATMOD) in SAS was used in this
analysis. Table 4.13 displays the results.

Table 4.13

Categorical ANOVA: Environmental Dynamism and Strategy

Levels: 2
. . 2

N=52
Prob: 0.4910

 

Table 4.13 indicates that the result was not signiﬁcant and that H1(a) was not
supported. It had been expected that higher levels of environmental dynamism would
encourage ﬁrms to gravitate to externally-oriented strategies. Such strategies and their
associated information exchange would - theoretically - counteract the uncertainty
created by the dynamic environment.

A one-tailed t test was used as a cross-check. Using strategy as the categorical

variable, an uncategorized environmental dynamism was examined. As with the

93
CATMOD procedure, the results were not statistically signiﬁcant (p> .20). Process

ﬁrms reported a mean level of environmental dynamism of 3.26 and externally-oriented
ﬁrms reported a mean of 3.10.

Hypothesis 1b: Under conditions of high environmental

dynamism, ﬁrms will use more decentralized structures than

under conditions of low environmental dynamism.

A one-tailed t test (a = .10) was performed comparing the two levels of
environmental dynamism to ﬁrms’ levels of centralization. The results are shown in
Table 4.14. The summated centralization scale was arranged in such a manner that a

higher mean score represented greater centralization of activities in the logistics

department.

 

 

Table 4.14
T -Test: Inﬂuence of Environmental Dynamism (E.D.) on Centralization
CENTRALIZATION
N Mpgn STD t P
LOW ED. 31 10. 58 4.08 0.07 .47
HIGH ED. 26 10.50 4.41 0.07 .47

 

 

 

The results were not signiﬁcant. Respondents in the low environmental
dynamism group reported, on average, 10.58 activities out of sixteen (16) being
consolidated in the logistics department. The mean was 10.50 for respondents
operating under conditions of a high environmental dynamism. Consequently, H1(b)

was not supported.

94
Hypothesis 1c: Under conditions of high environmental
dynamism, ﬁrms will use less formalized structures than under
conditions of low environmental dynamism.
A one-tailed t test (01 = .10) was performed comparing the two levels of
environmental dynamism to ﬁrms’ levels of formalization. The results are shown in

Table 4.15. The summated formalization scale was arranged in such a manner that a

higher mean score represented greater formalization

 

 

Table 4. 15
T -Test: Inﬂuence of Environmental Dynamism (E.D.) on Formalization
FORMALIZATION
N Mpan STD t P
LOW ED. 30 2.87 1.87 -l.46 .07
HIGH ED. 25 3.72 2.35 -1.50 .07

 

 

 

As indicated in Table 4.15, the inﬂuence of environmental dynamism on
formalization was not supported as the difference was opposite the hypothesized
direction. Although this was initially surprising, there is a plausible explanation.
There is an emerging thought that greater levels of logistics formalization lead to
greater levels of logistics ﬂexibility. Flexibility, in turn, leads to more logistically
responsive ﬁrms. Therefore, given this sample of generally above-average logistics
performers, it is plausible that this result is not surprising and may merit further
investigation. Perhaps in their quest for ﬂexibility, more competent logistics
performers choose formalization regardless of the dynamic environmental state. In any

event, H1(c) was not supported.

95
Hypothesis 1 d: Under conditions of high environmental

aynamism, ﬁrms will use more integrated structures than under
conditions of low environmental dynamism.

A one-tailed t test (OL = .10) was performed comparing the two levels of
environmental dynamism to ﬁrms’ levels of integration. The results are shown in
Table 4.16. The summated integration scale was arranged in such a manner that a
higher mean score represented greater integration.

Table 4. 16
T-Test: Inﬂuence of Environmental Dynamism (EU) on Integration

 

 

INTEGRATION (Committees)

N Megn STD t P
LOW ED. 29 3. 52 0.73 -1.36 .09
HIGH ED. 23 3.86 0.95 -1.40 .08

INTEGRATION (Cross-functional)

LOW ED. 28 3.32 0.98 -0.89 .19
HIGH ED. 23 3.56 0.96 -0.89 .19

 

 

 

As indicated in Table 4.16, the inﬂuence of environmental dynamism on the
two dimensions of integration was mixed. Integration through committees was
signiﬁcantly different whereas cross-functional integration was not. Under conditions
of high environmental dynamism, ﬁrms were significantly more likely to use
committees as an integrative mechanism. In contrast, ﬁrms operating under high levels
of environmental dynamism report, on average, a cross-functional integration mean of
3.56. Firms in less dynamic environments report a mean of 3.32 for cross-functional

integration. Thus, H1(d) is partially supported.

96
Hypothesis 1e: Under conditions of high environmental

dynamism, ﬁrms will use narrower spans of control than under

conditions of low environmental dynamism.

A one-tailed t test (01 = .10) was performed comparing the two levels of
environmental dynamism to ﬁrms’ spans of control. The results are shown in Table
4.17. For span of control, respondents were simply asked to indicate the number of
people under their control. Higher mean scores represented a wider span of control.

Table 4. 17
T -Test: Inﬂuence of Environmental Dynamism (E.D.) on Span of Control

 

SPAN OF CONTROL

 

 

 

N Megn STD t P
LOW ED. 25 6.40 2.68 0.06 .48
HIGH ED. 20 6.35 2.74 0.06 .48

 

As indicated in Table 4.17, the inﬂuence of environmental dynamism on span of
control was non-signiﬁcant and H1(e) was not supported. This result may reﬂect the
impact of information technology and downsizing in American industry. Span of
control may be independent of environmental forces due to the monitoring and
coordinating ability of information technology linkages. Additionally, perhaps the
stripping away of layers in organizations is forcing span of control to increase
regardless of the environmental state, thereby lessening the diﬁ’erences that may have

appeared in the past.

97
ENVIRONMENTAL HOSTILITY

Environmental hostility was evaluated on a summated scale as discussed earlier.
The sum was then divided by the number of items to return to a 1 - 5 scale with the
right-hand anchor representing higher levels of hostility. Respondents answering at the
mean or below were considered to be experiencing comparatively low levels of
environmental hostility. Respondents indicating a response above the mean of 4.28
were considered to be experiencing high levels of environmental hostility. This two
group categorical variable was then used in the statistical analysis of environmental
hostility and strategy and structure.

Hypothesis 2a: Under conditions of high environmental

hostility, ﬁrms will choose more externally-oriented logistics

strategies (i. e., market and channel) than under conditions of low

environmental hostility.

As was the case with environmental dynamism and strategy, the relationship
between environmental hostility and strategy was examined through categorical
ANOVA (CATMOD). The categories are the two levels of environmental hostility and
two levels of strategy (i.e., process versus externally-oriented). Results are shown in

Table 4.18.

Table 4.18
Categorical ANOVA: Environmental Hostility and Strategy

Levels: 2

. 2
N=52
Prob: 0.2209

 

98

Table 4.18 indicates that the result was not signiﬁcant and that H2(a) was not
supported. As mentioned earlier, it had been expected that higher levels of
environmental hostility would encourage ﬁrms to gravitate to externallycriented
strategies in order to reduce uncertainty created by the hostile environment. The
greater level of communication and cooperation of externally-oriented strategies was
expected to serve as a counterbalance to the uncertainties in the environment.

A one-tailed t test was used as a cross-check. Using strategy as the categorical
variable, an uncategorized environmental hostility was examined. As with the
CATMOD procedure, the results were not statistically signiﬁcant (p>.32). Process
ﬁrms reported a mean level of environmental dynamism of 4.28 and externally-oriented
ﬁrms reported a mean of 4.36.

Hypothesis 2b: Under conditions of high environmental

hostility, ﬁrms will use more decentralized structures than under

conditions of low environmental hostility.

A one-tailed t test (or = .10) was performed comparing the two levels of
environmental hostility to ﬁrms’ levels of centralization. The results are shown in
Table 4.19. The summated centralization scale was arranged in such a manner that a
higher mean score represented greater centralization of activities in the logistics

department.

99

Table 4.19
T -Test: Inﬂuence of Environmental Hostility (E.H.) on Centralization

 

 

CENTRALIZATION
N Mean STD t P
LOW RH. 25 10.12 5.08 -0.64 .26
HIGH EH. 32 10.86 3.41 -0.67 .25

 

 

 

The results in Table 4.19 indicate that there was not a signiﬁcant difference
between the low environmental hostility and high environmental hostility groups. The
former reported an average of 10.12 activities under the responsibility of logistics
whereas the latter reported a mean of 10.86 activities. Consequently, H2(b) was not
supported.

Hypothesis 2c: Under conditions of high environmental

hostility, ﬁrms will use less formalized structures than under

conditions of low environmental hostility.

A one-tailed t test (01 = .10) was performed comparing the two levels of

environmental hostility to ﬁrms’ levels of formalization. The results are shown in

 

 

Table 4. 20.
Table 4.20
T -Test: Inﬂuence of Environmental Hostility (ER) on Formalization
FORMALIZATION
N Mean STD t P
LOW E.H. 23 2.74 2.16 -1.53 .07
HIGH EB. 32 3.63 2.06 -1.54 .06

 

 

 

100

As indicated in Table 4.20, the inﬂuence of environmental hostility on
formalization was not supported. Similar to the environmental dynamism/formalization
outcome, the results are opposite the hypothesis. Firms indicating high levels of
environmental hostility reported a mean of 3.63 for formalization. Firms in less hostile
situations reported a mean of 2.74. The same potential explanation holds true: It is
possible that formalization is so important to ﬂexibility - and ﬂexibility so important in
dynamic and hostile environments - that contrary to the long—established organizational
science view, formalization may be more relevant to adverse logistics environments
rather than the opposite. In any event, H2(c) was not supported.

Hypothesis 2d: Under conditions of high environmental

hostility, ﬁrms will use more integrated structures than under

conditions of low environmental hostility.

A one-tailed t test (bi = .10) was performed comparing the two levels of
environmental hostility to ﬁrms’ levels of integration. The results are shown in Table
4.21. The two dimensional integration construct was arranged in such a manner that a
higher mean score represented greater integration.

As indicated in Table 4.21, the inﬂuence of environmental hostility on the two
dimensions of integration was not signiﬁcant. Within the high environmental hostility
group, the mean use of integrative committees was 3.77. The low environmental
hostility group reported a mean of 3.52 for usage of integrative committees. Both
groups reported less usage of cross-functional integration, with a mean of 3.42 for high
environmental dynamism and 3.45 for low environmental dynamism. H2(d) was not

supported.

101

Table 4.21
T-Test: Inﬂuence of Environmental Hostility (E.H.) on Integration

 

INTEGRATION (Committees)

 

N Mean STD t P
LOW EH. 21 3.52 0.76 -1.08 .14
HIGH EH. 31 3.77 0.89 -1.05 .15

INTEGRATION (Cross-functional)

LOW E.H. 20 3.45 1.02 0.11 .46
HIGH E.H. 31 3.42 0.95 - 0.11 .46

 

 

 

Hypothesis 2e: Under conditions of high environmental

hostility, ﬁrms will use narrower spans of control than under

conditions of low environmental hostility.

A one-tailed t test (<1 = .10) was performed comparing the two levels of
environmental dynamism to ﬁrms’ spans of control. The results are shown in Table
4.22. For span of control, respondents were simply asked to indicate the number of
people under their control. Higher mean scores represented a wider span of control.

Table 4. 22
T -Test: Inﬂuence of Environmental Hostility (E.H.) on Span of Control

 

SPAN OF CONTROL

 

N Mean STD t P
LOW EH. 16 5.81 1.42 ' -1.28 .10
HIGH EH. 29 6.69 3.14 -105 .15

 

 

 

As indicated in Table 4.22, the inﬂuence of environmental hostility on span of

control was not signiﬁcant and H2(c) was not supported. Respondents facing a more

102

hostile environment reported a mean of 6.69 employees under the senior logistics
executive’s control. Respondents operating in a less hostile environment reported, on
average, 5.81 employees under the senior logistics executive’s control.
LOGISTICS INFORMATION TECHNOLOGY: OPERATING AND PLANNING
SYSTEMS

Logistics information technology was evaluated in terms of operating and
planning systems. Both constructs represent a summated scale as discussed earlier.
The sum was then divided by the number of items to return to a 1 - 5 scale with the
right-hand anchor representing greater capability. Respondents answering at the mean
or below were considered to have less capable systems. For operating systems,
respondents indicating a response above the mean of 3.41 were considered to have
highly capable operating systems whereas respondents at or below this mean were
considered to have less capable operating systems. Similarly, for plaming systems the
dividing point was a mean of 3.06. These two group categorical variables were then
used in the statistical analysis of logistics information technology and strategy and
structure. Operating systems results are presented ﬁrst, followed by planning systems
results.

Hypothesis 3a: Under conditions of highly capable logistics

information technology operating systems, ﬁrms will choose more

externally-oriented logistics strategies (i. e., market and channel)

than under conditions of less capable logistics information

technology operating systems.

Consistent with previously reported analyses involving context variables and

strategy. the relationship between logistics information technology operating systems

and strategy was examined through categorical ANOVA (CATMOD). The categories

103

are two levels of operating systems and two levels of strategy (i.e., process versus
externally-oriented). Results are shown in Table 4.23.
Table 4.23
Categorical ANOVA: Operating Systems and Strategy

levels: 2
' : 2

N=52
Prob: 0.11874

 

Table 4.23 indicates that the result was not signiﬁcant and that H3(a) was not
supported. Contrary to the hypothesis, there is no discernible evidence that ﬁrms with
highly capable operating systems are signiﬁcantly more engaged in externally-oriented
logistics strategies.

As with the previously reported tests involving CATMOD, a one—tailed t test
was performed as a cross-check. Strategy was the categorical variable and the
operating systems variable was measured on its 1-5 scale. The result was not
signiﬁcant (p> .34). Externally-oriented strategy respondents reported a logistics
operating system mean capability of 3.47. Process strategy respondents reported a
mean of 3.35.

Hypothesis 3b: Under conditions of highly capable logistics

information technology operating systems, ﬁrms will use more

decentralized structures than under conditions of less capable

logistics information technology operating systems.

A one-tailed t test (or = .10) was performed comparing the two levels of

logistics information technology operating systems to ﬁrms’ levels of centralization. A

104

higher mean represented greater centralization of activities in the logistics department.

The results are shown in Table 4.24.

 

 

Table 4.24
T -Test: Inﬂuence of Operating Systems (0. S.) on Centralization
CENTRALIZATION
N Mean STD t P
LOW OS. 29 10.48 4.35 -0. ll .46
HIGH 0. S. 28 10.61 4.11 -0.11 .46

 

 

 

As indicated in Table 4.24, H3(b) is not supported. Respondents reporting a

lower capability operating system had, on average, 10.48 activities under the

responsibility of logistics. The other group, higher capability operating systems,

reported a mean of 10.61 activities under the responsibility of logistics.

Hypothesis 3c: Under conditions of highly capable logistics

information technology operating systems, ﬁrms will use more
formalized structures than under conditions of less capable

logistics information technology operating systems.

A one-tailed t test (CL = . 10) was performed comparing the two levels of

operating systems to ﬁrms’ levels of formalization. The results, shown in Table 4.25,

are significant and H(3)c is supported.

 

 

 

Table 4. 25
T -Test: Inﬂuence of Operating Systems (0. S.) on Formalization
FORMALIZATION
N Mean STD t P
LOW 08 28 2.75 1.86 -1.82 .04
HIGH 0. S. 27 3.78 2.29 -1.83 .04

 

 

105
According to the results in Table 4.25, firms with more highly capable

operating systems exhibit greater levels of formalization. This was expected as ﬁrms
with highly capable information systems usually have extensive documentation and
procedures related to those systems. In addition, the information inputs are typically
quite speciﬁc, thereby necessitating strict compliance by personnel. Such practices or
requirements can then be expected to appear in other areas of the ﬁrm.

Hypothesis 3d: Under conditions of highly capable logistics

information technology operating systems, ﬁrms will use more

integrated structures than under conditions of less capable

logistics information technology operating systems.

A one-tailed t test (01 = .10) was performed comparing the two levels of
logistics infOrmation technology operating systems to ﬁrms’ levels of integration. The
results are shown in Table 4.26.

Table 4.26
T -Test: Inﬂuence of Operating Systems (0.8.) on Integration

 

INTEGRATION (Committees)

 

N Megn STD t P
LOW O. s. 26 3.72 0.92 0.38 .35
HIGH 0.5. 26 3.63 0.77 0.38 .35

INTEGRATION (Cross-functional)

LOW O. S. 25 3.36 0.96 -0.51 .31
HIGH 0. S. 26 3.50 0.99 -0.51 .31

 

 

 

As indicated in Table 4.26, the inﬂuence of operating systems on the two

dimensions of integration was not signiﬁcant and H3(d) was not supported. The

106

underlying rationale, that better capability would foster easier integration and would
therefore be exploited, was not supported. Firms with highly capable operating
systems report means of 3.63 and 3. 50, respectively, regarding the usage of integration
committees and cross-functional teams. Firms with less capable operating systems
report means of 3. 72 and 3.36 on the two integration measures.

Hypothesis 3e: Under conditions of highly capable logistics

information technology operating systems, ﬁrms will use broader

spans of control than under conditions of less capable logistics

information technology operating systems.

A one-tailed t test (01 = .10) was performed comparing the two levels of
logistics information technology operating systems to ﬁrms’ spans of control. The
results are shown in Table 4.27. For span of control, respondents were simply asked
to indicate the number of people under their control. Higher mean scores represented a
wider span of control.

Table 4. 27
T -Test: Inﬂuence of Operating Systems (0. S.) on Span of Control

 

SPAN OF CONTROL

 

N Mean STD t P
LOW 08. 20 5.75 1.83 -1.51 .07
HIGH 0.8 25 6.88 3.14 -1.43 .08

 

 

 

As indicated in Table 4.27, the inﬂuence of operating systems on span of
control was signiﬁcant and H3(e) was supported. F irms with higher operating systems
capability indicated a greater number of personnel under the direction of the senior

logistics executive (6.88) compared to ﬁrms reporting less capable operating systems

107

(5.75). This supports the general belief that information technology enables managers
to maintain or increase their span of control.

Hypothesis 3f: Under conditions of highly capable logistics

information technology planning systems, ﬁrms will choose more

externally-oriented logistics strategies (i .e. , market and channel)

than under conditions of less capable logistics information

technology planning systems.

The relationship between logistics information technology planning systems and
strategy was examined through categorical ANOVA (CATMOD). The categories are
two levels of planning systems and two levels of strategy (i.e., process versus
externally-oriented). Results are shown in Table 4.28.

Table 4.28
Categorical ANOVA: Planning Systems and Strategy

rise Levels: 2
lations: 2

N =52
Prob: 0.2080

 

Table 4.28 indicates that the result was not signiﬁcant and that H3(f) was not
supported. Firms with highly capable planning systems did not appear likely to choose
one strategic orientation over another based on the results of Table 4.28.

As with the previously reported tests involving CATMOD, a one-tail t test was
performed. Strategy was the categorical variable and the planning systems variable was
measured on its 1-5 scale. The result was marginally non-signiﬁcant (p> .10).
Externally-oriented strategy respondents reported a mean planning system capability of

3.32. Process strategy respondents reported a mean of 2.94.

108
Hypothesis 3g: Under conditions of highly capable logistics
information technology planning systems, ﬁrms will use more
decentralized structures than under conditions of less capable
logistics information technology planning systems.
A one-tailed t test (<1 = .10) was performed comparing the two levels of
logistics information technology planning systems to ﬁrms’ levels of centralization.

The summated measure is such that higher mean scores represent greater centralization

of activities in a logistics department. The results are shown in Table 4.29.

 

 

Table 4.29
T -Test: Inﬂuence of Planning Systems (P. S.) on Centralization
CENTRALIZATION
N Mean STD t P
LOW PS. 34 10.03 4.31 -l.15 .13
HIGH PS. 23 11.30 3.99 -1.13 .13

 

 

 

As indicated in Table 4.29, H3(g) was not supported. There was no signiﬁcant
difference between firms exhibiting different levels of planning system capability. The
lower capability planning system group reported 10.03 activities (out of sixteen) under
the responsibility of logistics. The higher capability planning system group reported
11.30 activities under the responsibility of logistics.

Hypothesis 3h: Under conditions of highly capable logistics

information technology planning systems, ﬁrms will use more

formalized structures than under conditions of less capable
logistics information technology planning systems.

A one-tailed t test (01 = .10) was performed comparing the two levels of

planning systems to ﬁrms’ levels of formalization. The results are indicated in Table

109

4.30. The summated formalization scale was arranged in such a manner that a higher

mean score represented greater formalization.

 

 

Table 4.30
T—Test: Inﬂuence of Planning Systems (P. S.) on Formalization
FORMALIZATION
N Mean STD t P
LOW PS. 33 3.00 2.09 -1.08 .14
HIGH PS. 22 3.64 2.17 -1.09 .14

 

 

 

Firms with highly capable planning systems reported a formalization mean of
3.64. Firms with less capable planning systems reported a formalization mean of 3.00.
However, the results were not statistically signiﬁcant and H3(e) is not supported.

Hypothesis 3i: Under conditions of highly capable logistics

information technology planning systems, ﬁrms will use more

integrated structures than under conditions of less capable

logistics information technology planning systems.

A one-tailed t test (<1 = .10) was performed comparing the two levels of
logistics information technology planning systems to ﬁrms’ levels of integration. The
results are shown in Table 4.31. The two dimensional integration construct was
arranged in such a manner that a higher mean score represented greater integration.

As indicated in Table 4.31, the inﬂuence of planning systems on the two
dimensions of integration was not signiﬁcant and H3(i) is not supported. What is

rather curious about the results of Table 4.31 is how closely matched the two levels of

planning systems appear to be. In this particular case, the reported means from the two

110

groups are remarkably similar for both integrative measures.

Table 4. 31
T -Test: Inﬂuence of Planning Systems (P. S.) on Integration

 

INTEGRATION (Committees)

 

N Mean STD t P
LOW P. S. 30 3.67 0.87 -0.06 .47
HIGH P. S. 22 3.68 0.82 -0.06 .47

INTEGRATION (Cross-functional)

LOW PS. 29 3.41 1.07 -0.15 .44
HIGH PS. 22 3.45 0.84 -0.15 .44

 

 

 

Hypothesis 3]: Under conditions of highly capable logistics

information technology planning systems, ﬁrms will use broader

spans of control than under conditions of less capable logistics

information technology planning systems.

A one-tailed t test (01 = .10) was performed comparing the two levels of
logistics information technology planning systems to ﬁrms’ spans of control. The
results are shown in Table 4.32. For span of control, respondents were simply asked
to indicate the number of people under their control. Higher mean scores represented a
wider span of control.

As indicated in Table 4.32, the inﬂuence of planning systems on span of control
was signiﬁcant and H3(j) was supported. As hypothesized, higher capability in the
area of planning systems appears to support broader span of control. This lends further

support to the idea that information technology enhances managers’ ability to monitor

greater numbers of personnel.

111

Table 4. 32
T -Test: Inﬂuence of Planning Systems (P. S.) on Span of Control

 

SPAN OF CONTROL

 

N Mean STD t P
LOW PS. 24 5.54 2.54 -2.35 .01
HIGH P. S. 21 7.33 2.56 -2.35 .01

 

 

 

In addition to the main effects described above there is the question of an
overall interaction effect. Do the four context variables - environmental dynamism,
environmental hostility, logistics information technology operating systems, and
logistics information planning systems - interact in any manner that affects strategic
orientation or elements of structure? These interaction effects were tested using the
general linear models (GLM) procedure in SAS.

Hypothesis 3k: With respect to strategy, there will be a

signiﬁcant interaction eﬂect between the levels of environmental

dynamism, environmental hostility, information technology

operating systems, and information technology planning systems.

Using two levels of strategy (i.e. , internally focused or processed and externally
focused or market/ channel) and two levels for each of the context variables, the result
was not signiﬁcant. An F value of 1.60 was obtained with a p-value of .12. The
conclusion was that a signiﬁcant interaction effect between context and strategy does
not exist in the Logistics Contingency Model and H3(k) was not supported.

Hypothesis 3]: With respect to structure, there will be a

significant interaction effect between the levels of environmental

dynamism, environmental hostility, information technology
operating systems, and information technology planning systems.

112

Again, the GLM procedure was used to assess the effect of the interaction of
context variables on each of the elements of structure. The results are summarized in
Table 4.33. As indicated, none of the interaction effects are signiﬁcant at the 0.10
level. Therefore, the premise that context variables interact and serve as predictors of
elements of structure is rejected. H3(l) is not supported.

Table 4. 33
Interaction Effects of Context on Structure

ED*EH*ITOS*ITPS

 

MEDIATION ISSUES

Within the general CSSP framework, there is ample opportunity to study
potential mediational effects. The general mediational model is depicted in Figure 4.5.
As illustrated, there are two causal paths to the outcome variable, with paths a and b

representing the mediation path.

 

Mediator
a b
Independent , Outcome Variable
Variable
c
Figure 4. 5

Mediational Model

113
As described by Baron and Kemy (1986), tests of mediation can be conducted

through a series of regression models. Three equations must be estimated. These
equations correspond to paths (1, b and c indicated in Figure 4.5. Baron and Kenny
state:

These three regression equations provide the tests of the linkages of the

mediational model. To establish mediation, the following conditions

must hold: First, the independent variable must affect the mediator in

the ﬁrst equation; second, the independent variable must be shown to

affect the dependent variable in the second equation; and third, the

mediator must affect the dependent variable in the third equation. If

these conditions all hold in the predicted direction, then the effect of the

independent variable on the dependent variable must be less in the third

equation than in the second. Perfect mediation holds if the independent

variable has no effect when the mediator is controlled (p. 1 177).

Similar estimates are provided for in structural equation modeling (e.g.,
LISREL). However, given the limited sample size of this research and the typical
requirements of fully developed structural equation models, the Baron and Kenny
approach was used in this research.

Hypotheses 4a-d evaluated the mediational effects of structure on the context-
performance relationship. Figure 4.6 provides a corresponding mediational model of
the hypothesized relationships. Each speciﬁc path was evaluated (e.g., environmental
hostility --> span of control --> performance). The context, structure and
performance variables were treated as continuous variables for estimation purposes

(i.e., unlike some of the previous statistical approaches, no levels or classes were

utilized). Tables 4.34-37 provide the results of the tests.

114

Hypothesis 4a: Structure is a signiﬁcant mediator of the
environmental dynamism --> performance relationship.

Hypothesis 4b: Structure is a significant mediator of the
environmental hostility -> performance relationship.

Hypothesis 4c: Structure is a signiﬁcant mediator of the
logistics information technology operating systems ->
performance relationship.

Hypothesis 4d: Structure is a signiﬁcant mediator of the
logistics information technology planning systems ->
performance relationship.

Formalizati on
Decentralization
Integration
Span of Control

Envir. Dynamism
Envir. Hostility
I.T. Op. Systems '
I.T. Plan Systems

Performance

 

Figifre 4. 6
Mediational Model: Context-Structure-Performance

The results indicate that H4(a)-(d) are not supported. There were no
mediational effects present in this model. This led to the rejection of the premise that
the impact of context effects on performance would be mediated through structure
variables. Although mediation was the main area of interest within this set of
hypotheses, an interesting ﬁnding did emerge. Information technology planning
systems did have a signiﬁcant main effect on performance. This suggests that even
though context is ﬁxed in the short run, organizations with performance concerns may

potentially alter their information technology planning systems to impact performance.

115

 

 

 

 

 

 

Table 4.34
Tests of Mediation: Environmental Dynamism (ED)—Structure-Performance

PATH t—VALUE PROB.
Env. Dynamism --> Formalization 1.048 0.2991
Env. Dynamism --> Performance 0.064 0.9491
Env. Dynamism -— > Formalization -- > Perf.

Env. Dynamism 0.180 0.8581

Formalization -0.523 0.6033

 

 

 

Conclusion: No direct effect of ED on performance nor is there an indirect effect of
ED throu h formalization on erformance

 

 

 

 

 

 

Env Dynamism -- > Centralization —0. 037 0. 9708
Env. Dynamism --> Performance 0.064 0.9491
Env. Dynamism -- > Centralization --> Perf.
Env. Dynamism 0.206 0.8377
Central ization -1.362 0. 1798

 

 

Conclusion: No direct effect of ED on performance nor is there an indirect effect of
ED throuh centralization on .1 erformance . .

       

 

 

 

 

 

Env Dynamism —- > Integration (Committees) 1.989 0. 0522
Env. Dynamism --> Performance 0.064 0.9491
Env. Dynamism -- > Integration -- > Perf.
Env. Dynamism -0.323 0.7481
Integration 1.577 0. 1217

 

 

   

Env Dynamism --> Integrationl(Teams)

Conclusion: No direct effect of ED on performance nor is there an indirect effect of
.. ED: throuh interation (committees) on . -.

...0...2356--

 

 

 

 

 

Env. Dynamism --> Performance 0.9491
Env. Dynamism -- > Integration -- > Perf.
Env. Dynamism -0.230 0.8191
Integration 1.629 0. l 103

 

 

    

 

.. ED throuh interation teams) on

I erformance

Conclusion: NO direct effect of ED on performance nor is there an indirect effect of

 

 

 

 

 

Env Dynamism -- > Span of Control -0. 292 0. 7721
Env. Dynamism --> Performance 0.064 0.9491
Env. Dynamism -- > Span of Control -- > Perf.
Env. Dynamism 0.730 0.4696
Span of Control 0.571 0.5709

 

 

 

Conclusion: No direct effect of ED on performance nor is there an indirect effect of
ED through span of control on performance.

 

 

116

Table 4.35
Tests of Mediation: Environmental Hostility (EH)-Structure—Performance

 

 

 

 

 

 

PATH t-VALUE PROB.
Env. Hostility --> Formalization 0.929 0.3569
Env. Hostility --> Performance -1.183 0.2429
Env. Hostility -- > Formalization -- > Perf.
Env. HOSIIILY -1.139 0.2605
Formalization -0. 419 0.6768

 

 

 

Conclusion: No direct effect of EH on performance nor is there an indirect effect of
EH throu h formalization on erformance

            

 

 

 

 

Env. Hostility -- > Centralization 0. 951 0. 3458
Env. Hostility --> Performance -1.183 0.2429
Env. Hostility -- > Centralization -- > Perf.
Env. Hostility -1.333 0.1891
Centralization -1.492 0.1424

 

 

 

Conclusion: No direct effect of EH on performance nor is there an indirect effect of
EHmthrou hcentralization on erformance ' ‘

       

 

 

 

 

Env Hostility --> Integration (Committees) 1. 359 0 1803
Env. Hostility --> Performance -1.183 0.2429
Env. Hostility -- > Integation -- > Perf.
Env. Hostility -1.672 0.1012
Integration 1.959 0.0561

 

 

 

Conclusion: No direct effect of EH on performance but there is an indirect effect of
EH-throuxh interation committees) on erformance. .,..

      

      
   

 

 

 

 

Env Hostility --> Integration (Leams) -0. 288 0.7743
Env. Hostim --> Performance -1.183 0.2429
Env. Hostility -- > Integration -- > Perf.
Env. Hostility -1.097 0.2784
Integration 1.545 0. 1293

 

 

 

Conclusion: No direct effect of EH on performance nor is there an indirect effect of
EH throu h interation teams) on erformance

-
.....................................................................................................................................................................................................
-'. :~:/-"' ‘E-II"'~Z-f\'-12;‘21I'\“'ZZZ’.'.'.-.~I;/"""'.11‘.Z2:2:ZZZZI‘\>ZAL-l~$i-'z'.’3”?1 ......

............................................................. \.,;‘:‘;';‘;';':'~‘.‘.;' «$3535.59, " .-.’.- j“ ./ _’;;_.- ' W%\ém%

      
 

          
 

 
    

 

 

 

 

Env HOStIlLL" > Span of Control 0.905 0.3703
Env. Hostility -- > Performance -1.183 0.2429
Env. Hostility -- > Span of Control -- > Perf.
Env. Hostility -1.275 0.2097
Span of Control 0.704 0.4857

 

 

 

Conclusion: No direct effect of EH on performance nor is there an indirect effect of
EH through span of control on Jerformance.

 

 

 

117
Table 4.36

Tests of Mediation: I.T. Operating Systems (ITOP)-Structure-Performance

 

PATH

 

 

 

 

 

 

 

t-VALUE PROB.

ITOP -- > Formalization 1.551 0.1272

ITOP -- > Performance 1.521 0.1356
ITOP -- > Formalization -- > Perf.

ITOP 1.589 0.1195

Formalization -0.544 0.5895

 

   

Conclusion: No direct effect of ITOP on performance nor is there an indirect effect
of ITOP throu“ h formalization on _. rformance ...1

 

 

 

 

 

 

ITOP -- > Centralization -0. 009 0. 9932
ITOP -- > Performance 1.521 0.1356
ITOP -- > Centralization -- > Perf.
ITOP 1.410 0.1658
Centralization -0.752 0.4561

 

of ITOP throu - h centralization on

        
 

. .. .‘ [.174 '7l7'» '. I‘, _ _ ,1, _ :27 . . .5, I .;.j.;.:.:»;.;.33.1.5:.;.;.:.;,;.:.;.'...g...,.;.;.;.:.'..I .:.:.;-‘_.:.:.f-:-\:.j.:.:.:I:.:.:.'.;“.‘l.:.:.:.:.:.f.;-: ~ . .-.;.~Z:7;'.;7:.;.;.'\ .3: 2. - .-,;' .;.;.' .................................

 

   

Conclusion: No direct effect of ITOP on performance nor is there an indirect effect
. erformance

- s .\§.;.:I~.:.- . 1 . . .- . :1‘.'-.-. - 4v ‘ ~ ~ _.-.-.-.:.'.'.' z, .....

 

 

 

 

 

 

ITOP -- > Integration (Committees) -0. 727 0. 4710
ITOP -- > Performance 1.521 0.1356
ITOP -- > Integration -- > Perf.
ITOP 1.609 0. 1 151
Integration 1.523 0. 1353

 

 

. o
.........................................................................................................
‘..I;.7 -.--, ‘ / -' ----.'-- :::::::::::::

    

Conclusion: No direct effect of ITOP on performance nor is there an indirect effect

of ITOP throuh interation (committees) on erformance

 

 

 

 

 

 

ITOP -- > Integration (Teams) 0.958 0. 3431
ITOP -- > Performance 1.521 0.1356
ITOP -- > Integration -- > Perf.
ITOP 1.369 0. 1786
Integration 1 .288 0.205 1

 

   

 

1. '. I_.' 5.3.2. ..................................................

Conclusion: No direct effect of ITOP on performance nor is there an indirect effect
of ITOP through integration (teams) on

  
 

  

 

 

 

 

 

 

ITOP -- > Span of Control 0. 2819
ITOP -- > Performance 1.521 0.1356
ITOP -- > Span of Control -- > Perf.
ITOP 1.100 0.2781
Span of Control 0.214 0.8317

 

 

Conclusion: No direct effect of ITOP on performance nor is there an indirect effect
of ITOP through span of control on performance.

 

 

118

Table 4.37
Tests of Mediation: I.T. Planning Systems (ITPLAN)-Structure-Performance

 

 

 

 

 

 

PATH t-VALUE PROB.
ITPLAN -- > Formalization 1.527 0.1335
ITPLAN -- > Performance 2.496 0.0167
ITPLAN -- > Formalization -- > Perf.
ITPLAN 2.568 0.0141
Formalization -0.701 0.4876

 

 

 

  

MW\IEICVWQM\>AVN\ “-21.0“ IN

 

      

W“ “W WWW/If

   

Conclusion: There is a signiﬁcant direct effect between ITPLAN and performance
with no mediatin inﬂuence throu ' h formalization

-
”"3“"! \W/IYI MW 'o'mvmx WI) ~ W: r,- .- -_- 7.. 7.1., ' ‘1 . .‘(r.‘.‘.‘¢-§._\\\\A.~

 

 

 

 

 

ITPLAN --> Centralization -0. 126 0.9006
ITPLAN --> Performance 2.496 0.0167
ITPLAN --> Centralization --> Perf.
ITPLAN 2.358 0.0234
Centralization -0.645 0.5227

 

 

 

   

 

 

   

Conclusion: There is a signiﬁcant direct effect between ITPLAN and performance
with no mediatin. inﬂuence throuh centralization.

 

 
 

mmwmmmmmm meumx \Wwwvlu

   

"WWW

 

 

 

 

ITPLAN --> Integration (Committees) 0.940 0. 3522
ITPLAN -- > Performance 2.496 0.0167
ITPLAN -- > Integration -- > Perf.
ITPLAN 2.302 0.0266
Integration 1 . 181 0.2447

 

 

 

   

MW‘ mm“ \Wﬂm\w W Wﬂf

 

Conclusion: There is a signiﬁcant direct effect between ITPLAN and performance
With no mediatin inﬂuence throu h inte ration (committees)

   

 

 

 

 

IT PLAN --> Integration (Teams) 0.315 0. 7540
ITPLAN -- > Performance 2.496 0.0167
ITPLAN --> Integration --> Perf.
ITPLAN 2.425 0.0200
Integration 0.782 0.4387

 

 

 

 
    

Conclusion: There is a signiﬁcant direct effect between ITPLAN and performance
.With.._’39 mediatinwmﬂuence throuh inte~ ration teams).

-
‘W WWWWWIX’lyM-t -wr rﬁ')’ le-i 0WWhX< ~50 WW WWI- m‘\\\\\wo:f/Xl

  
    

  

 

 

 

 

ITPLAN -- > Span of Control 3.249 0.0023
ITPLAN -- > Performance 2.496 0.0167
ITPLAN --> Span of Control -- > Perf.
ITPLAN 1.497 0. 1430
Span of Control 0.561 0.5782

 

 

 

 

Conclusion: There is a significant direct effect between ITPLAN and performance
With no mediating inﬂuence through span of control.

 

 

119
Hypotheses 4e-h evaluate the mediational effects of strategy on the context-

structure relationship. Figure 4.7 provides a corresponding mediational model of the
hypothesized relationships. Each speciﬁc path was evaluated (e.g., environmental
hostility --> strategy --> span of control). The context and structure performance
variables were treated as continuous variables for estimation purposes. Strategy was
treated as a dummy variable with internally and externally-oriented strategies. Tables
4. 37-40 provide the results of the tests.

Hypothesis 4e: Strategy is a significant mediator of the

environmental dynamism --> structure relationship.

Hypothesis 4f Strategy is a signiﬁcant mediator of the
environmental hostility --> structure relationship.

Hypothesis 4g: Strategy is a signiﬁcant mediator of the
logistics information technology operating systems -> structure
relationship.

Hypothesis 4h: Strategy is a signiﬁcant mediator of the
logistics information technology planning systems -> structure

 

relationship.
Strategy
a h
Envir. Dynamism Formalization
Envir. Hostility Decentralization
I.T. Op. Systems ’ Integration
1.1". Plan Systems c Span of Control
Figure 4. 7

Mediational Model: Context-Strategy Structure

120
The results of Tables 4.38-41 indicate that H4e-h were not supported. No

mediating effects were present in the model in terms of context-suategy-structure
arrangements. Therefore, the hypotheses suggesting a mediational role for strategy in
terms of context and structure must be rejected. There were two signiﬁcant main
effects. However, these were expected based on the results of the earlier t tests. These
main effects were: (1) environmental dynamism --> integration (committees); and

(2) information technology planning systems ~- > span of control.

121

 

 

 

 

 

 

Table 4.38

Tests of Mediation: Environmental Dynamism (ED)-Strategy-Structure
PATH t-VALUE PROB.
Env. Dynamism --> Strategy 0.823 0.4143
Env. Dynamism -—> Formalization 1.048 0.2991
Env. Dynamism -- > Strategy -- > Formalization

Env. Dynamism 1.458 0.1514

Strategy 0.708 0.4822

 

 

 

Conclusion: No direct effect of ED on formalization nor is there an indirect effect of
on formalrzatron _

ED throu h strate

      

Env Dynamlsm--> Strategy

...0 .823

. _O 41.43

 

   

  

 

 

 

 

 

Env. Dynamism --> Centralization -0.037 0.9708
Env. Dynamism --> Strategy --> Centralization
Env. Dynamism 0.973 0.3352
Strategy -0.637 0.5271

 

 

Conclusion: No direct effect of ED on centralization nor is there an indirect effect of
”ED throu h strate

on central1zat1on

 

 

 

 

 

..-:-:: M24":- . . . .. -:.:.:.:.:.:.::‘ ----- -:->.:-. '44:! -:'~-: WWWAﬂW
Env Dynamtsm --> Strategy 0. 823 0.4143
Env. Dynamism --> Integration (CommitteesL 1.989 0.0522
Env. Dynamism -- > Strategy -- > IntegLation

Env. Dynamism 2.178 0.0343
Strategy -1.525 0. 1336

 

 

 

Conclusion: There is a signiﬁcant direct effect between ED and integration

(committees) w1th no med1at1n inﬂuence throu h strate; .
5W... ... .,. .: . mamas: saw/ﬂats“

   

Env Dynamrsmi--> Strategy

”1);;4’351;WW§<$%£\ r._\ .wﬁ$::::_::“."

 

 

 

 

 

0.823 0.4143

Env. Dynamism -—> Integration (Teams) 1.201 0.2356
Env. Dynamism -—> Strategy -- > Integration

Env. Dynamism 1.443 0.1555

Strat_egL -1 .793 0.0793

 

 

Conclusion: No direct effect of ED on integration (teams) nor is there an indirect
on mte ration .

““6323 _ ,

effect of ED throu h strate

.........................................................................................
.;.;.’ ‘ ; :. ;. ;., - ' V .’.-.-y~.~.-.'.-.'.~.\A - ..' ‘ ..' ‘ ~. -. - .-. ' ' ~ (2.1.4.75. . . '7; . :.;._ ..._.:.;.;.;\.;.;.;.;.;.;.:.;.;.;.'.: ' ‘52: ','-.".-.~.'.'.'T-‘.-I-. -.-~" ............................................................

Env Dynamism --> Strategy ...................................

 

.0 4.143

 

 

 

 

 

Env. Dynamism —- > Span of Control -0.292 0.7721
Env. Dynamism -- > Strategy -- > Span of Control
Env. Dynamism -0.l90 0.8503
Strategy -1.319 0.1944

 

 

 

Conclusion: No direct effect of ED on span of control nor is there an indirect effect

of ED through strategy on span of control.

 

 

122
Table 4.39

Tests of Mediation: Environmental Hostility (EH)-Strategy-Structure

 

 

 

 

 

 

 

PATH t—VALUE PROB.
Env. Hostility --> Stratggy -0.405 0.6875
Env. Hostilig --> Formalization 0.929 0.3569
Env. Hostility --> Strategy --> Formalization
Env. Hostility 0.635 0.5285
Strategy 0.902 0.3712

 

 

”Env Host111ty --> Strategy

Conclusion. No direct effect of EH on formalization nor is there an indirect effect of
_ EH throu h strate on formal1zat1on

 

41465

4 o. 6875” H

 

 

 

 

 

Env. Hostility --> Centralization 0.951 0.3458
Env. Hostility -- > Strategy -- > Centralization
Env. Hostility -0.431 0.6683
Strategy 0548 0.5864

 

 

 

    

mszH throu : h strate

Env Host111y «5 Strategy“

on central1zat1on

Conclusion: No direct effect of EH on centralization nor is there an indirect effect of

' v. ... . ..’.-..A»‘ -:c... 5‘
'.. /'/. ”:.f.\\::€.:." $15.9};- :.:\.;3:.;.:.:.:.;.;§§.:

0 6875 ..

 

 

 

 

 

Env. Hostility --> Integration (Committees) 1.359 0.1803
Env. Hostility -- > Strategy -— > Integration
Env. Hostility 1.295 0.2015
Strategy -1. 169 0.2481

 

 

1nd1rect effect of EH throu h strate

 

Env Host111ty --> StrategL

Conclusion: No direct effect of EH on integration (committees) nor is there an
on inte; ration (committees) .. __

-0405

 

 

 

 

 

 

0. 6875

Env. Hostility -- > Integration (Teams) -0.288 0.7743
Env. Hostilityu> Strategy --> Inggration

Env. Hostility -0.342 0.7335

Strategy -1.602 0.1157

 

 

Env Host1l1ty --> Strategy

Conclusion: No direct effect of EH on integration (teams) nor is there an indirect
effect of EH throu h strate . '

    

 

 

 

 

 

Env. Hostility -- > Span of Control 0.905 0.3703
Env. Hostility -- > Strategy -- > Span of Control
Env. Hostility 0.772 0.4445
Strategy -1.256 0.2159

 

 

 

Conclusion: No direct effect of EH on span of control nor is there an indirect effect
of EH through strategy on span of control.

 

 

123

 

 

 

 

 

 

Table 4.40

Tests of Mediation: I.T. Operating Systems (ITOP)-Strategy-Structure
PATH t-VALUE PROB.
ITOP -- > SUM -0.395 0.6947
ITOP -- > Formal ization 1.551 0.1272
ITOP -- > Strategy -- > Formalization

ITOP 1.526 0. 1340

Strategy 0.817 0.4182

 

 

 

       

.. ....__._°f.....1T0P._“"0“.h strate ._

Conclusion: No direct effect of ITOP on formalization nor is there an indirect effect
on formal1zat1on ’ ..

 

 

 

 

 

 

ITOP -- > Strategy -0 395 0. 6947
ITOP -- > Centralization -0.009 0.9932
ITOP -- > Strategy -- > Centralization
ITOP -0.667 0.5082
Strategy -0.305 0.7615

 

      

H-011“)? throu ' h strate

Conclusion: No direct effect of ITOP on centralization nor is there an indirect effect
on central 1zat1on

     

 

 

 

 

 

 

ITOP -- > Strategy ~0. 395 0. 6947
ITOP -- > Integration (Committees) —0.727 0.4710
ITOP -- > Strategy -- > Integration
ITOP —0.802 0.4267
Strategy -1.241 0.2209

 

indirect effect of ITOP throu h strate

5:23:323243;233:224;:-:1:-':1:'.~;-:7:121:~:-":¢:5:3:-:-: .A:-’:7:'I..‘ i':5:2'i'?“‘?:5:‘~5:1:1:$32171:2:i:1:1:"1:i:?:3:1:3:1:1::. ‘5:4221:1312;'.‘<¢:1:1:1:-:-:-.7:~: :I:§;§:323:§:L:3:~.-A.~:::-:§:§:§:§:§;§:§:~:- ‘vz»:3:1:1:I:1:l:1:3:'-:i:111:3:1:#:121:1:3:‘;:g:3_:§:§:§:§:3:355?»S‘11‘3-313‘3111'413’f5'-".. ;:~:i:5_:i:1:1:5:~:'<1‘&".‘:‘:‘:': :‘z‘:‘:‘;&i~l'2~i-i+3544“‘1" -I:l'1:?-?:1;i'i'1'1'2"'.4345:2559. . 0””? ~_:\‘.-:-‘.§1:1:1:?:3:1;1 :::::

Conclusion: No direct effect of ITOP on integration (committees) nor is there an

on inte ration committees).

..........

.........................

 

 

 

 

 

 

ITOP -- > Strategy -0.395 0.6947
ITOP -- > Integration (Teams) 0.958 0.3431
ITOP -- > Strategy -- > Integration
ITOP 0.891 0.3778
Strategy -1.511 0.1380

 

effect of ITOP throu - h strate

       

.3353181'775'3'E'E‘t‘5'3'5'5';'L".2115- .I.{5-3:;.-.-.-}.-:-;353:'?:3$:'4’7...‘I,"~.135'3333‘1‘?5'35???”7252552.:32'; {fa- .f.3:"/'f_' ‘ ‘ ‘. ‘ .1-31.3125}.33E}.5.15:”>32:23:311212121:2;T:':3.-‘< ’ >Sz7;3:i:1:5:3:?:‘-:‘-:1:1;i:3.\:‘~:‘.-:2:3:?:I:i:1:i:3:i:5:3:1:3:'-:1:3$‘":'- :';';‘:‘;':':':3:','1;‘,;‘"it""z'x;935'1'3‘3'3‘l‘r‘v“ """"""""""""""""""""""""""""

Conclusion: No direct effect of ITOP on integration (teams) nor is there an indirect
on inte ration (teams)

6" #3223113” edgy ram}.

 

 

 

 

 

 

ITOP —- > Strategy -0. 395 0. 6947
ITOP -- > Span of Control 1.090 0.2819
ITOP -- > Strategy -- > Span of Control
ITOP 1.018 0.3145
Stratggy -1.640 0. 1086

 

 

Conclusion: No direct effect of ITOP on span of control nor is there an indirect
effect of ITOP through strategy on span of control.

 

 

124

 

 

 

 

 

 

Table 4.41
Tests of Mediation: I.T. Planning Systems (ITPLAN)-Strategy—Structure

PATH t-VALUE PROB.
ITPLAN --> Strategy -1.315 0.1952
ITPLAN -- > Formalization 1.527 0.1335
ITPLAN -- > Strategy -- > Formalization

ITPLAN 1.478 0. 1468

Strategy 1. 170 0.2483

 

 

 

Conclusion: No direct effect of ITPLAN on formalization nor is there an indirect

 

_ effect of ITPLANchrou . hbstrate. _

on formalization ‘ _ . A

 

 

 

 

 

ITPLAN --> Strategy -1.315 0 1952
ITPLAN --> Centralization -0.126 0.9006
ITPLAN -- > Strategy -- > Centralization
ITPLAN —0.885 0.3813
Strategy -0.350 0.7283

 

 

Conclusion: No direct effect of ITPLAN on centralization nor is there an indirect

.............

,. effect of ITPLAN throu h strate

4;""x “"" I:3212...:ZZI:IZ .......................

on centralization

 

 

 

 

 

ITPLAN --> Strategy -1.315 0. 1952
ITPLAN -- > Integration (Committees) 0.940 0.3522
ITPLAN -- > Strategy -- > Integration
ITPLAN 0.755 0.4541
Strategy -0.843 0.4040

 

 

 

Conclusion: No direct effect of ITPLAN on integration (committees) nor is there an
indirect effect of ITPLAN through strategy on integration

 

 

 

 

 

 

(committees) W W p _
ITPLAN --> Strategy -1.315 0.1952
ITPLAN -- > Integation (Teams) 0.315 0.7540
ITPLAN --> Strategy -- > Integration
ITPLAN 0.087 0.9309
Strategy -1. 199 0.2372

 

 

 

Conclusion. No direct effect of ITPLAN on integration (teams) nor is there an

1nd1rect effect of ITPLAN throu h strate

........................................................................................................................................

................................

on inte _ration (teams).

I . ;.. , . . \\ ~.\\\)\\, , I/\\

ITPLAN --> Strategy

 

 

 

 

-1.315 0.1952

ITPLAN -- > Span of Control 3.249 0.0023
ITPLAN -- > Strategy -- > Span of Control

ITPLAN 2.843 0.0071

Stratgy -1.622 0. 1 128

 

 

 

 

Conclusion: There is a signiﬁcant direct effect between IT PLAN and span of control

with no mediating inﬂuence through strategy.

 

 

125
STRATEGY AND STRUCTURE

The preceding section examined the relationship between strategy and structure
as part of mediational models. This section specifically isolates the strategy-structure
relationship. It is examined through a series of t tests, dividing strategy into internal
and external strategies (i.e., process versus market and channel). Structure variables
continue to be used in the same manner as previously reported in this research.

Hypothesis 5a: Externally-oriented logistics strategy (i. e., market

or channel) is associated with less centralized structure than internally-

oriented logistics strategy (i. e., process).

A one-tailed t test (CL = .10) was performed comparing the two levels of
strategy to centralization. The results are shown in Table 4.42. The summated
centralization scale was arranged in such a manner that a higher mean score represented

greater centralization.

 

 

Table 4.42
T ~Test: Inﬂuence of Strategy on Centralization
CENTRALIZATION
N Mean STD t P
INTERNAL 34 11.41 2.20 0.53 .30
EXTERNAL 18 11.83 3.55 0.46 .33

 

 

 

The results in Table 4.42 indicate that there is no signiﬁcant difference in terms
of centralization between the two strategy groups. H5(a) is not supported. Within this
sample, the internally-oriented strategy group reported a mean of 11.41 activities

consolidated under the responsibility of logistics. Externally-oriented strategy

126

respondents reported a mean of 11.83 activities consolidated under the responsibility of
logistics.

Hypothesis 5b: Externally-oriented logistics strategy (i. e.,

market or channel) is associated with less formalized structure

than internally-oriented logistics strategy (i. e., process).

A one—tailed t test ((1 = .10) was performed comparing the two levels of
strategy to formalization. The results are shown in Table 4.43. The summated

formalization scale was arranged in such a manner that a higher mean score represented

greater formalization.

 

 

Table 4.43
T -Test: Inﬂuence of Strategy on Formalization
FORMALIZATION
N Mean STD t P
INTERNAL 34 3.59 2.06 -0.86 .20
EXTERNAL 18 3.06 2.15 -0.87 .19

 

 

 

As indicated in Table 4.43, the result was not signiﬁcant. H5(b) is not
supported. The group using an internally-oriented strategy reported a mean

formalization measure of 3. 59. The externally-oriented strategy group reported a mean

of 3.06.

127

Hypothesis 5c: Externally-oriented logistics strategy (i. e.,

market or channel) is associated with more integrated structure

than internally-oriented logistics strategy (i. e., process).

A one-tailed t test (01 = .10) was performed comparing the two levels of
strategy to ﬁrms’ levels of integration. The results are shown in Table 4.44. The
summated integration scale was arranged in such a manner that a higher mean score
represented greater integration.

Table 4.44
T -Test: Inﬂuence of Strategy on Integration

 

INTEGRATION (Committees)

 

N Mean STD t P
INTERNAL 34 3.56 0.87 1.24 .11
EXTERNAL 18 3.87 0.78 1.28 .10

INTEGRATION (Cross-functional)

INTERNAL 33 3.27 1.05 1.61 .06
EXTERNAL 18 3. 72 0.75 1.77 .04

 

 

 

As indicated in Table 4.44, the inﬂuence of strategy on the two dimensions of
integration was mixed. Integration through cross-functional teams was significant
whereas committee-driven integration was not. The results show that externally-
oriented strategy has a higher level of cross-functional team integration compared to
internally-focused strategy. As for committee-driven integration, externally-oriented
strategy reported a mean of 3.72 while internally-oriented firms reported a mean of

3.27. Thus, H5(c) is partially supported.

128
Hypothesis 5d: Externally-oriented logistics strategy (i. e.,

market or channel) is associated with broader spans of control

structure than internally-oriented logistics strategy (i. e., process).

A one-tailed t test (a = .10) was performed comparing the two strategy groups
to firms’ spans of control. The results are shown in Table 4.45. For span of control,
respondents were simply asked to indicate the number of people under control of the
senior logistics executive. Higher mean scores represented a wider span of control.

Table 4.45
T-Test: Inﬂuence of Strategy on Span of Control

 

SPAN OF CONTROL

 

N Meain STD t P
INTERNAL 30 6.00 2.52 1.35 .09
EXTERNAL 15 7.13 2.90 1.29 .10

 

 

 

As indicated in Table 4.45, the inﬂuence of strategy on span of control was
significant and H5(d) was supported. As hypothesized, externally-oriented strategy
favors broader span of control. On average, the externally-oriented ﬁrms report more
than one full person under the senior logistics executive as compared to the intemally-
focused strategy. This wider span of control is most likely necessary as ﬁrms extend

linkages beyond their own boundaries and interface with channel partners.

129
STRATEGY, STRUCTURE AND PERFORMANCE

This ﬁnal section of speciﬁc results focuses on the relationships between
strategy, structure and performance as hypothesized in the Logistics Contingency
Model. The investigation of the strategy-performance-strategy feedback loop
necessitated collection of time-series data. This was accomplished by sending a highly
focused questionnaire to the respondents comprising the database used thus far in this
research. This follow-up questionnaire solicited responses to replications of strategy
and performance items used in the initial questionnaire (i.e., “Workbook” as reprinted
in Appendix A). The follow-up questionnaire is reproduced in Appendix B.

The initial data was collected during the ﬁrst half of 1994. The follow-up
sample was collected during the period of March-May 1997. Therefore, the
approximate time lag throughout the sample is three years. Of course, as a result of
this lag, certain respondents were no longer at the same position and responses could
not be obtained. Consequently, the follow-up sample is smaller.

Hypothesis 5e: Under conditions of satisfactory

performance, less strategic changes will occur than under

conditions of unsatisfactory performance.

A review of the merged data reveals that selected logistics strategies do change.
There were thirty-nine (39) usable responses after the data was merged. Of these
thirty-nine ﬁrms, twenty-one (21) or 54% of the ﬁrms, had changed strategy during the
last three years. Within this group, there was one clear pattern: Of the initial thirty-
nine ﬁrms, ﬁfteen (15) indicated an externally-oriented strategy in 1994. Eighty

percent of these ﬁrms (i.e., 12/ 15) reported using a different strategy in 1997. Eight

130

of the changing ﬁrms changed focus and went to the internally-focused process
strategy. The remaining four ﬁrms simply changed to a different externally-oriented
strategy.

But what of the total sample? Table 4.46 presents the results of a chi-square
test that examined 1994 performance and strategic change between 1994 and 1997. For
the purposes of this test, performance was simply divided into two groups, those at or
below the mean of 3.78 from the thirty-nine (39) ﬁrms and those above the mean.
Performance above the mean was considered satisfactory, performance at or below the

mean was considered non-satisfactory.

 

 

 

 

 

Table 4.46
Chi-Square Test of Performance and Change in Strategy
No Change in Change in Strategy
Strategy from 1994 from 1994 to 1997 TOTAL
to 1997

Performance At or

Below Mean (1994) 12 9 21
Performance Above

Mean (1994) 6 12 18
TOTAL 18 21 39

 

 

 

 

The results do not support H5(e). At or = .10, the chi-square value of 2.21 was
not significant (p=.137). There is no statistical evidence supporting the linkage

between performance and subsequent change in logistics strategy.

 

131

Hypothesis 6a: Logistics structure, by itself, has no
signiﬁcant relationship to performance.

A series of two-tailed t tests (or = .10) were performed using two levels of each
structure variable. The mean of each structure variable was determined. Values at or
below the mean were coded as “low” and values above the mean were coded as
“high” in a given structure variable. The results are presented in Table 4.47.

There were no signiﬁcant differences between the different levels of the
structure variables and their relationship to performance. Thus, H6(a) was supported
as it was hypothesized that independent of the other model variables, structure would
not impact performance. This is consistent with the idea of “ﬁt” and the alignment of
a ﬁrm’s strategy and structure with its context in order to produce desired performance.

In addition to the individual t tests, a general linear model was tested. It
checked for interaction effects of a combined structure variable on performance. The

resulting F value was 1.25 (p=0.29). This provides further support for H6(a).

132

Table 4.47
T -Test: Inﬂuence of Structure on Performance

 

 

 

 

 

 

 

 

 

 

 

 

 

 

VARIABLE N MEAN STD t P

Low Formalization 26 3.75 0.3807 0.57 0.57
High Formalization 23 3.67 0.5882 0.58 0.56
Low Centralization 20 3.76 0.4033 0.64 0.53
High Centralization 29 3.68 0.5392 0.60 0.55
Low Integ. (Committee) 26 3.65 0.5203 -0.82 0.42
High Integ. (Committee) 23 3.77 0.4470 -0.81 0.42
Low Integ. (Teams) 19 3.64 0.4368 -0.87 0.39
High Integ. (Teams) 30 3.76 0.5158 -0.84 0.40
Low Span of Control 33 3.64 0.4275 -1.28 0.21
High Span of Control 16 3.85 0.5777 -1.42’ 0.16

Hypothesis 6b: Logistics structure will be a signiﬁcant

mediator of the logistics strategy -- > performance relationship.

This ﬁnal research hypothesis probes the relationship between logistics strategy,
the elements of structure and the outcome as represented by performance. The
relationships are pictured in Figure 4.8. As with the previous mediational models
presented in this research, Baron and Kenny’s (1986) approach was used. Three
regression equations were estimated for each relationship. The results are presented in

Table 4.48.

133

Formalization
Decentralization
Integration
Span of Control

Strategy , Performance

C

Figure 4. 8
Mediational Model: Strategy-Structure-Performance
Contrary to the hypothesized mediation effects, there was no evidence of any
signiﬁcant effects. Of course, in the absence of a main effect between strategy and
performance the only possible outcome could have been complete mediation by the
elements of structure. As Table 4.48 shows, no such mediation occurred in the

Logistics Contingency Model. Therefore, H6(b) is not supported.

134

Table 4.48
Tests of Mediation: Strategy-Structure-Performance

 

 

 

 

 

 

 

 

PATH t-VALUE PROB.
Strategy-- > Formalization 0.873 .3869
Strategy-- > Performance -0.245 .8074
Strategy-- > Formalization -- > Perf.
Strategy -0. 179 .6425
Formal ization -0. 179 .85 89

 

   

effect of strate-

   

  
  

Conclusion. No direct effect of strategy on performance nor is there an indirect

throu - h formalizatron .99 erformance

    

 

 

 

 

 

 

Strategy--> Centralization -0. 528 .5996
Strateg -- > Performance -0.245 .8074
Strategy-- > Centralization -- > Perf.
Strategy -0.381 .7048
Central ization -1. 380 . 1742

 

   

.. ”effect. of strate

Strategy-- > Integration (Commlttees)

  
 

-1" 237

Conclusion. No direct effect of strategy on performance nor is there an indirect

throu . h centrallzauon on erformance

 

 

 

 

 

 

Strategy-- > Performance -0.245 .8074
Strate --> Integration --> Perf.
Strategy 0. 105 .9167
Integration 1 .527 . 1337

 

effect of strate; throu

      

  

Conclusion: No direct effect of strategy on performance nor is there an indirect

h interation (commlttees) on erformance . .

   

 

 

 

 

 

 

Stratgyn Integration (Teams) -1. 607 .1145
Strategy-— > Performance -0.245 .8074
Strategy--> Integration --> Perf.
Strategy 0.084 .9334
Integration 1.595 . l 177

 

  

   

I erformance

Conclusion: No direct effect of strategy on performance nor is there an indirect
- throu ~ h . 1nte~ rat1on (teams _on”

 

 

 

 

 

 

Strategy-- > Span of Control -1.353 .1831
Strategy-- > Performance -0.245 .8074
Strategt > Span of Control -- > Perf.
Strategy -0. 185 .8539
Span of Control 0.457 .6504

 

 

Conclusion: No direct effect of strategy on performance nor is there an indirect
effect of strategy through span of control on performance.

 

 

I 3 5
SUMMARY

This chapter has presented the results of the statistical tests performed in relation
to the research questions (RQ) and related research hypotheses. Table 4.49 presents a
condensed summary of these results. For ease of reference, the research hypotheses are
repeated in Appendix C.

Table 4.49
Summary of Statistical Tests and Related Research Hypotheses

 

RQ #1: What is the relationship of environmental dynamism to logistics

strategy?
Ressargh Hypothesis Tsst Result
Hla Categorical ANOVA Not Supported

 

RQ #2: What is the relationship of environmental dynamism to logistics

structure?
Hlb T Test Not Supported
ch T Test Contradicted
Hld T Test Partially Supported
Hle T Test Not Supported

 

RQ #3: What is the relationship of environmental hostility to logistics
strategy?

H2a Categorical AN OVA Not Supported

 

 

 

136
Table 4.49 (con’t)

Summary of Statistical Tests and Related Research Hypotheses

 

 

 

 

 

RQ #4: What is the relationship of environmental hostility to logistics
structure?
Researeh Hypothesis Teg Result
H2b T Test Not Supported
H2c T Test Contradicted
H2d T Test Not Supported
H2e T Test Not Supported
RQ #5: What is the relationship of logistics information technology
operating systems to logistics strategy?
H3a Categorical ANOVA Not Supported
RQ #6: What is the relationship of logistics information technology
operating systems to logistics structure?
H3b T Test Not Supported
H3c T Test Supported
H3d T Test Not Supported
H3c T Test Supported
RQ #7: What is the relationship of logistics information technology

planning systems to logistics strategy?

H3f Categorical AN OVA Not Supported

 

 

137
Table 4.49 (con’t)

Summary of Statistical Tests and Related Research Hypotheses

 

RQ #8: What is the relationship of logistics information technology

planning systems to logistics structure?

 

Resegteh Hyputhesis Test Result
H3g T Test Not Supported
H3h T Test Not Supported
H31 T Test Not Supported
H3j T Test Supported
RQ #9: What is the relationship of context to logistics strategy?

H3k

GLM

Not Supported

 

RQ #10:

What is the relationship of context to logistics structure?

H31

GLM

Not Supported

 

 

RQ#11:

What indirect relationships, if any, exist between the contextual
variables, strategy, and/ or structure and performance?

H4a
H4b
H4c
H4d
H4e
H4f
H4g
H4h

Regression
Regression
Regression
Regression
Regression
Regression
Regression
Regression

Not Supported
Not Supported
Not Supported
Not Supported
Not Supported
Not Supported
Not Supported
Not Supported

 

 

 

138

Table 4.49 (con’t)
Summary of Statistical Tests and Related Research Hypotheses

 

RQ #12: Is there evidence of a reciprocal relationship between strategy

and structure?
R r h H th si lest _ResulL
HSa T Test Not Supported
H5b T Test Not Supported
H5c T Test Partially Supported
H5d T Test Supported

 

RQ #13: What is the direct relationship, if any, between logistics
structure and performance?

H6a T Tests Supported

 

RQ #14: What is the relationship of logistics strategy on performance as
mediated by structure?

H6b Regression Not Supported

 

 

RQ #15: Does the level of logistics performance affect logistics strategy?

HSe Chi-Square Not Supported

 

 

CHAPTER FIVE - CONCLUSIONS

This ﬁnal chapter is divided into six sections. The ﬁrst section summarizes
research results. The results are grouped and discussed according to the major
relationships of the Logistics Contingency Model. These relationships are: (1) Context
- Strategy; (2) Context - Structure; (3) Strategy - Structure; (4) Structure —
Performance; (5) Performance - Strategy; and (6) Mediational Paths between Context -
Strategy - Structure, Context - Structure - Performance, and Strategy - Structure -
Performance.

The second and third sections examine, respectively, the academic and
managerial contributions of this research. The fourth section addresses limitations of
this research. The ﬁfth section, propositions and directions for future research,
suggests ways in which the Logistics Contingency Model can be used in continuing

research. Finally, a brief summary ends the chapter.

RESEARCH RESULTS
Four major constructs were used in this research: (1) Context, (2) Strategy,
(3) Structure, and (4) Performance. Each is brieﬂy described.
Context is increasingly being used in contingency research in lieu of
environment. In both cases, the variables are ﬁxed in the short-term. However,
whereas environment is typically limited to variables outside the

control of the ﬁrm, context is more expansive. Not only does context usually
139

 

 

140

incorporate environment variables, it also includes variables under the direct control of
the ﬁrm (e.g., information technology). In this sense, context represents the ﬁrm’s
short-term situation as determined by environmental considerations and previous ﬁrm
decisions. Beyond the short-term, it is expected that the ﬁrm adjusts those variables
under its control in order to improve its operational situation.

Strategy was deﬁned in this research according to the Bowersox and Daugherty
(1987) typology. These classiﬁcations were process, market and channel. Within this
speciﬁc research design, process strategy was considered an internally-focused strategy
while market and channel strategies were considered externally-oriented strategy.

Structure in this research was investigated from a human resource perspective.
Consequently, formalization, centralization, integration, and span of control were all
considered in terms of their relationships to personnel and their deployment.

Performance, as discussed in Chapter Two, is measured in numerous ways in
the logistics literature. This research used self-reporting perceptual measures.
Respondents compared themselves to their competition in terms of speciﬁc logistics

measures .

CONTEXT AND STRATEGY

Contingency theory posits that strategy is designed to cope with a ﬁrm’s
context. For example, if a ﬁrm faces a highly variable environment, strategy should
account for that variability. Within this research design, the basic premise was that
higher or lower levels of context variables would account for differences between

internally-oriented and externally-oriented strategies. Figure 5.1 highlights the

141

relationship being discussed.

..... .-.--C,9!1_t¢x.t_.-.- -

 

Environmental .
—- Dynamism I l

 

 

 

.__ Environmental .
—{3 Hostility “P" Logistics H.R.
.4.»

 

 

Logistics
Structure Performance

 

 

 

 

 

 

 

‘4 Logistics LT. .
1 Operating Systems I T

 

 

 

---.-l-.- - -

.. Logistics LT.
r- Planning Systems

 

 

 

 

 

_______________________ 1!

Logistics 4* _____ _
Strategy

Figure 5.1
Logistics Contingency Model: Context-Strategy

 

 

«mummwmm

 

 

 

 

 

 

 

 

 

 

 

It was hypothesized that less dynamic and hostile environments would be more
conducive to internally-oriented strategy. In the absence of environmental uncertainty,
ﬁrms would focus on the efﬁciency of internal operations. External linkages would not
be crucial as environmental monitoring would not be as important as in a highly
uncertain environment.

In contrast, highly dynamic and hostile environments should encourage ﬁrms to
utilize externally—oriented strategy. By building links beyond their ﬁrm’s boundaries,
ﬁrms might reduce uncertainty through information exchange, co-location of
employees, etc. Such efforts should then serve as a buffer between a ﬁrm’s fortunes
and the volatility of the business environment.

Hypotheses 1(a) and 2(a) reﬂected this general relationship of environmental

dynamism and hostility and strategic choice. However, the results of this research did

142

not support the premise that environmental dynamism and hostility impact logistics
strategic orientation. The internally and externally-oriented strategies were not
signiﬁcantly different from each other on either environmental dimension.

The non-signiﬁcant result between dynamism and strategy is consistent with the
ﬁndings of Kohn et al. (1990) in which they found no association. However, this
research’s non-signiﬁcant result between hostility and strategy contrasts with Kohn et
al. They reported a signiﬁcant positive association between hostility and strategy.

This raises several considerations. First, it is possible that logistics strategy is
formed independently of traditional environmental variables. Much of the work in
environmental dynamism and hostility in organizational science has focused on ﬁrm-
level conditions. Severity of market competition or rate of change in a given industry
may impact corporate strategy but have considerably less inﬂuence on logistics
strategy. Environmental dynamism and hostility in this research were used and
measured in the traditional sense. This approach may not have adequately captured an
environmental-strategic link at the logistics departmental level. Perhaps product type,
distribution network or other such factors are much more important in logistics strategic
orientation.

Second, logistics strategy, as deﬁned in this research, may not adequately
capture strategic intent in the same manner as organizational science research and its
focus on Porter’s or Miles and Snow’s strategic typologies. As Chow et al. (1995)
pointed out, the Bowersox and Daugherty typology is heavily based on organizational
characteristics. This stands in rather stark contrast to a low cost or differentiation

strategy based on manufacturing strategy. Further work and research in logistics

 

143

strategy is necessary to better understand the similarities and differences between
corporate-level and departmental or task-driven strategies.

Closely related to the ﬁrst consideration is a third possible explanation -- the
nature of this research’s sample. The majority of this sample was composed of highly
regarded logistics organizations. It is entirely possible that this select group, having
mastered and elevated many logistics practices, determines logistics strategy orientation
on substantially different criteria than most ﬁrms. Environmental considerations,
though not ignored, may play a relatively minor role.

The other two context variables, information technology operating systems and
planning systems, were examined in hypotheses 3(a) and (f). The general premise was
that highly capable information systems afford ﬁrms the opportunity to interface with
other ﬁrms -- both operationally and strategically. Therefore, the expectation was that
the more highly capable ﬁrms would favor externally-oriented strategies. This position
was not supported. However, it should be noted that operating systems narrowly
missed being signiﬁcant (p=.11). This is encouraging for a new construct.

But if information technology is as pervasive as academic and managerial
articles would lead one to believe, why does it not clearly differentiate between the
strategic orientations? The answer may lie in the mean responses obtained in this
research. Operating on a 1-5 scale, with ﬁve representing high levels of capability, the
self-reports indicated means of 3.47 (externally-oriented) and 3.35 (internally-oriented)
for operating systems capability and 3.32 and 2.94 for planning systems. It is clear
that the groups do not consider themselves as highly capable on either information

technology measure.

144

Even in highly rated logistics organizations strategic deployment of information
technology is still in an early stage of development. It is one thing to automate manual
procedures, generate shipping documents and collect data for status reports. It is an
entirely different approach to link vendors and customers electronically, willingly share
proprietary information and design an information system to support a strategic
orientation.

This may also explain why the operating systems were close to being a
signiﬁcant differentiator while planning systems were not. The emphasis may currently
be on operational issues as that information is more likely to be less proprietary and
deliver more immediate impact for the investment. Although no statistical inference
should be drawn from the mean scores for the respective strategies, their reported
capabilities may provide a hint of the future. One can speculate that externally-oriented
strategy ﬁrms will continue to push development of outside electronic linkages to
support their strategy.

In the ﬁnal analysis, however, context did not have any statistically signiﬁcant
association with strategy in this research. Nonetheless, these results do provide a point
of comparison with Kohn et al. and future research within the framework of the

Logistics Contingency Model.

CONTEXT AND STRUCTURE
A signiﬁcant amount of research has been conducted into the context-structure
relationship. Previous research in organizational science and logistics has demonstrated

that context does inﬂuence organizational structure. In this research, four common

145

elements of structure were used - centralization, formalization, integration, and span of

control. Figure 5.2 highlights the relationship.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

........ C9n.t.ex.t- - -_
Environmental ‘ ‘
Dynamism : I
Environmental
Hostility . Logistics H.R. 4 Logistics
Logistics LT. 5» Structure Performance
Operating Systems j t
. . 1
I .. Logistics I.T. _j i
357 Planning Systems j ‘
...................... 11 2
3; Logistics 4— _ _ __ _ __ f
1: Strategy
Figure 5 2

Logistics Contingency Model: Context-Structure

Environmental dynamism was shown to have no signiﬁcant impact on
centralization and cross-functional integration. These results were not consistent with
the prevailing beliefs of organizational science. The logistics work of Germain, Droge,
and Daugherty (1994) and Germain and Droge (1995) found environmental uncertainty
predicted integration and operations decentralization but not scheduling
decentralization. Thus, the results of the present research are at odds with some of
these previous ﬁndings. However, it should be noted that the environmental
uncertainty construct used in the former efforts is more expansive than environmental
dynamism. Direct comparisons should therefore be made with caution.

As hypothesized, span of control and integration through committees were

predicted by environmental dynamism. Firms experiencing higher levels of

146

environmental dynamism reported narrower spans of control and greater use of
integrative committees. This is consistent with the organizational science literature
which states that as dynamism increases, ﬁrms take action to mitigate against the
increased volatility accompanying the increased dynamism. Under such conditions,
managers are only able to effectively manage smaller numbers of subordinates and
integrative measures are used to increase information ﬂow and thereby offset
uncertainty.

This result raises a question concerning integration. Why are integrative
committees signiﬁcant but not cross-functional teams? The answer may be tied to
strategy. In addition to the temporal consideration of the integration dimensions
mentioned earlier in this dissertation, there is also the reason committees are formed.
The questionnaire links committees to speciﬁc projects and cross-functional teams to
strategic decision-making. Environmental dynamism had no association to logistics
strategy. It would appear reasonable that cross-functional teams related to strategy
would also not have an association. It is possible that environmental dynamism
periodically creates a perceived need to deal with a particular situation. In these
situations the shorter duration committees are brought together for a particular project.
Rather than being part of an overall strategic resolution, perhaps committees represent
an “adjustment” to changes brought on by environmental dynamism.

The remaining dyad in this group - environmental dynamism-formalization -
produced the most surprising result. Formalization was predicted to be higher when
environmental dynamism was lower. The organizational science perspective underlying

this premise is that the “quieter” the environment, the easier it is to have set rules and

147
procedures. Not only was hypothesis 1(c) rejected, it was contradicted. Firms

identiﬁed as experiencing high levels of environmental dynamism reported mean levels
of formalization of 3.72. The low environmental dynamism group reported a mean of
2.87. The contradictory ﬁnding and relatively high levels of formalization reported by
the high environmental dynamism group were unexpected. It was expected that the low
environmental dynamism group would indicate greater formalization as per the
organizational science literature.

This result is consistent with another emerging theory in logistics. Research at
Michigan State University (1995) has demonstrated that routinization of tasks leads to
increased ﬂexibility. Flexibility is considered crucial to good logistics performance.

The ﬁndings of the present research regarding formalization advance the
“routinization leads to ﬂexibility” concept. Routinization is built upon the capabilities
of standardization, simpliﬁcation and discipline (World Class Logistics, p.207).
Underlying those capabilities - particularly standardization and discipline - is
formalization. Whereas routinization typically focuses on the actual work practices in
the logistics system, formalization can be thought of as the essential building block of
the related capabilities. In order to achieve standardization, simpliﬁcation and
discipline, ﬁrms must have well-understood operating rules and processes. To the
extent that these rules and processes are clearly documented and closely adhered to,
they represent formalization.

For this sample of highly competent logistics performers, highly dynamic
environments pose a potential threat - loss of ﬂexibility. In such an environment there

exists greater opportunity for problems to occur simply because of dynamic and

148

ﬂuctuating changes. To combat this and retain maximum ﬂexibility, formalization is
instituted. Procedures are standardized and simpliﬁed - thereby reducing the decision-
making options regarding the routine, repetitive logistics tasks. Adherence to these
procedures (i.e., discipline) ensures a similarity of outcomes that customers can
anticipate and appreciate. As argued in World Class Logistics, this approach reduces
the time managerial talent devotes to routine requirements and instead can “take
advantage of and accommodate unexpected events (p. 207). ”

In summary, the contradictory ﬁnding refutes traditional organizational science
beliefs concerning dynamism and formalization. But the ﬁnding does support the
emerging tenets of superior logistics performance. It is, therefore, reasonable to
assume that formalization underlies logistical routinization and contributes to the goal
of maintaining or enhancing ﬂexibility in the logistics system. This represents an
important step in understanding the counter-intuitive position that formalization -- >
routinization does not equal rigidity. It represents competitive advantage at the
logistics level.

A second plausible explanation exists concerning the dynamism-formalization
ﬁnding. This concerns the formalization construct used in this research. Composed of
seven benchmarking measures and a question concerning existence of a mission
statement, the construct does follow the normally accepted deﬁnition of formalization
(i.e., presence of written rules and procedures). But upon reﬂection, one might pose
the question: Which ﬁrm is most likely to benchmark, the one in the less volatile
environment or the more volatile environment? Although not statistically signiﬁcant,

the results of this research indicate that ﬁrms in more volatile environments are more

149

likely to benchmark. To combat an uncertain environment, it is advantageous to
benchmark. Therefore, the outcome is that ﬁrms in highly dynamic environments
benchmark more than ﬁrms in less dynamic environments. This is consistent with the
results of this research.

Turning to environmental hostility and structure, the same contradictory ﬁnding
exists between hostility and formalization as between dynamism and formalization.
Firms in more hostile environments reported a mean score of 3.63 versus 2.74 for
ﬁrms operating in less hostile environments. This result closely parallels the
dynamism-formalization outcome. In all likelihood, the same explanations hold true in
the hostility-formalization analysis. In an effort to avoid redundancy, those
explanations are not repeated here.

The other environmental hostility-structure relationships (i.e., centralization,
integration and span of control) were not supported. It was expected that ﬁrms
experiencing low levels of environmental hostility would be characterized by greater
centralization, less integration and broader span of control. There is little guidance
offered in the logistics literature concerning this outcome. Previously mentioned
studies incorporating environmental uncertainty do not break out the hostility
component separately. Kohn et al. (1990) did not assess the impact of environmental
hostility on logistics structure.

The results suggest that logistics structure is unaffected by environmental
hostility. The traditional view of hostility is the intensity of competitors’ actions. But
those actions are generally described in terms of market share, product line, etc. It

may be that this form of hostility is not perceived as relevant to the structuring of a

—————' H—A—A' _cuq... ‘-.. ...-.34. L- MA. ’u ' M-..+,_..-._, . .

 

150

logistics department or organization.

It should be noted that integration through committees and span of control
approached the signiﬁcance level of .10. The results may be more reﬂective of sample
size and homogeneity than theoretical deﬁciencies. As such, they offer support for the
continued use of these constructs and posited relationships in subsequent research
utilizing the Logistics Contingency Model.

The information technology component of this research was new in terms of
logistics contingency research. Unlike environmental dynamism and hostility, the ﬁrm
exercises direct control over information technology operating and planning systems.
Fixed in the short-term, such systems can be improved upon over a period of time.

With no precedent to serve as a guide in this area, exploratory positions were
staked out according to general logic concerning information technology. It was
expected that information technology would serve as an integrative mechanism of the
ﬁrm. Electronic linkages would increase span of control by increasing supervisory
monitoring ability, provide integration through the transfer of information, decentralize
decision-making as information increased in content and accessibility, and be
accompanied by greater formalization as written procedures and rules dictated the “new
rules of the game.” Firms with more capable systems would exhibit more of these
characteristics than less capable ﬁrms.

These associations were examined for operating systems through hypotheses
3(b)-(e). The results supported two of the four hypotheses. Firms with highly capable
operating systems were signiﬁcantly more likely to have greater formalization and span

of control. In fact, ﬁrms with highly capable operating systems indicated a span of

 

151

control (6.88) that is, on average, more than one full person greater than the less
capable ﬁrms (5.75).

If viewed from a resource allocation perspective, highly capable operating
systems are worthwhile as they extend managerial input and control over a larger
number of employees. As logistics operations are asked to “do more with less,” this is
an important consideration. It suggests that if ﬁrms are going to make investments in
information technology, there are potential pay-offs for going beyond mere adequacy.

The result for formalization was particularly strong (at a signiﬁcance level of
.04): The mean formalization score for ﬁrms with highly capable operating systems
was 3.78, a full point higher than the other group at 2.75. What might explain this
difference?

One explanation is that the construct, though rigorously evaluated, does not
capture the important dimensions of operating systems. That explanation is rejected.
Not only did the construct hold up under critical examination, a cross-check indicated
that it also incorporated most of the measures deemed important by the sample. Absent
that explanation, one is left to conclude that less capable ﬁrms appear unwilling or
unable to include documentary support and procedures for their information technology
operating systems.

As for centralization and integration, operating systems have little impact. Of
the two constructs, the lack of association with integration is more surprising. It was
expected that information technology operating system capability would be closely
identiﬁed with the shorter duration committee work. Again, it may be the promise of

information technology has yet to be fully realized and harnessed in logistics.

152
Information technology planning systems (hypotheses 3(f)—(i)) had only one

signiﬁcant relationship - span of control. It was, however, the strongest relationship
(p=.01) uncovered in this research. Centralization, formalization and integration were
not predicted by planning systems.

One explanation for greater spans in ﬁrms with highly capable planning systems
(7.33 versus 5.54) is the type of work performed by the planning systems. Deﬁned in
this research as forecasting, inventory management and distribution resource planning,
such systems would be heavily computerized. Monitoring these activities would
therefore seem to be easier. It is also reasonable to assume that the personnel
associated with planning are fairly autonomous and require less oversight, thereby
allowing managers to supervise more of them compared to other personnel. Any
system that is less capable would therefore reduce the ability of management to
remotely monitor these subordinates, effectively reducing span of control.

The other results in the area of planning systems and structure may be due to the
constructs. While the activities associated with planning systems could easily be
conceived as candidates for centralization and formalization, the centralization and
formalization constructs used in the research are quite different. There is little reason
to expect that forecasting and inventory management would predict whether or not such
activities would be consolidated in a logistics department. Such planning activities are
unlikely to predict whether or not a ﬁrm benchmarks (i.e., formalization) or forms
integrative committees 0r cross-functional teams. In hindsight, it is probable that the
constructs were a bit too far removed from the original intent of the individual

hypotheses.

“I"

153
STRATEGY AND STRUCTURE

A long-standing tenet of organizational science is that structure follows strategy.

 

Although mixed results have been widely reported - and some have argued for
acceptance of a reciprocal relationship - structure follows strategy is the prevailing

belief. There is no published research that uses a pure logistics strategy as the strategy

 

construct in terms of predicting structure. Figure 5.3 highlights this relationship.

This research has identiﬁed signiﬁcant relationships between strategy and cross-

 

functional integration and strategy and span of control. Non-signiﬁcant relationships
exist between strategy and centralization, formalization and integrative committees.

The strategy—structure relationships were investigated through hypotheses 5(a)-(d).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Context
E n Vi r on m e n t a l lP.‘.‘u‘:'"-VII/ILV‘A‘J‘
Dynamism t
Environmental I I
HOSUWY W Logistics H.R. ...... Logistics
4 E Logistics I.T. ...... Structure Performance
i E -------- 3:3 Operating Systems t
. ‘
Logistics LT. WT :
Planning Systems I
L l
2"; Logistics * “““““ r
3'; Strategy
Figure 5.3

Logistics Contingency Model: Strategy-Structure

As hypothesized, externally-oriented strategy ﬁrms reported greater numbers of
personnel (7.13) under the senior logistics executive than internally-oriented strategy

ﬁrms (6.00). The externally-oriented strategies are focused on an “extended ﬁrm”

154

view. Under this perspective, span of control is a necessary coordinating mechanism.
The decision to extend the ﬁrm beyond its traditional boundaries challenges typical
command and control practices. Tight control is not possible or perhaps even
warranted. With less concern about monitoring, a natural offshoot is greater span of
control.

As for cross-functional integration, externally-oriented strategies are predicated
on cross-functional effectiveness. It logically follows that externally-oriented strategy
ﬁrms would encourage and embrace the formation of cross—functional teams within
their organization. These teams are much more important to ﬁrm and partner success
than to the internally-oriented strategy ﬁrms.

It may also be plausible to speculate that span of control and cross-functional
integration are highly complementary and necessary in externally-oriented ﬁrms. As
integrative supply chain strategies are put in place by progressive boundary-spanning
ﬁrms, cross-functional teams must be used to interface with teams from other channel
members. This “stretching” across both internal and external boundaries forces
participating ﬁrms to rethink span of control and accept greater spans. Afterall, in
many cases the cross-functional teams are charged with the responsibility of facilitating
the supply chain approach. As they accept greater responsibility, it is reasonable that
they require less supervisory direction and oversight. The managerial talent formerly
expended on these individuals can be redeployed and extended to other personnel.

Centralization and formalization were not predicted by strategy. It was
hypothesized that externally-oriented strategy would favor less centralization and less

formalization. The rationale behind this position was that in dealing with extended

--r

155
enterprises, decision-making would have to be pushed away from a centralized

authority and that the give and take of such a relationship might not be amenable to
high levels of formalization. The fact that this position must be rejected in the present
research suggests that as of yet there is no consensus on the organizational structure of
the extended ﬁrm. It may also be related to the rather short duration of some
externally-oriented strategies. As reported in Chapter Four, eighty percent of external
strategists changed strategies in the last three years. Under such conditions of change it

may be difﬁcult to identify the appropriate level of centralization and formalization.

STRUCTURE AND PERFORMANCE

Within the organizational science literature, structure and performance linkages
have received less attention than other areas of the CSSP paradigm. In the logistics
literature there exists some evidence of structure impacting logistics performance (e. g.,
Droge and Germain 1989). But by and large, there is little empirical support in this
area. However, within the frameworks of context-structure-performance and strategy-
structure-performance, the general premise is that structure mediates the relationship.
Based on that idea, hypothesis 6(a) predicted that in this research there would be no
signiﬁcant relationship between the structure variables and performance. Figure 5.4
highlights the relationship.

This hypothesis was supported. The results are consistent with organizational
science theory. Speciﬁcally, neither formalization, centralization, integration, nor span
of control had a signiﬁcant, direct association with performance. In this research, the

conclusion is that logistics organizational structure and its relationship to speciﬁc

 

156

logistics performance mirrors the corporate structure-corporate performance

relationship reported in organizational science literature.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Context
Dynamism E
Environmental I I
HOS‘ili‘Y ”‘4'” Logistics H.R. I Logistics
. . f
E 2 Log1st1cs I.T. ”g” Structure Per ormance
. ........ _Ii’.‘ Operating Systems g :
Logistics I.T. I I
Planning Systems :
, t
'2'; Logistics * g __ __ g g f
i: Strategy
Figure 5 .4

Logistics Contingency Model: Structure-Performance

Of course, this is only part of the structure-performance story. Organizational
science theory holds that the reason no direct relationship exists between structure and
performance is due to the fact that structure serves as a mediational variable. This role

is discussed next.

INTERACTIONS AND MEDIATION

At the heart of contingency theory is the idea of ﬁt. Earlier in this research the
concept of ﬁt was discussed in terms of alignment. That is, if a ﬁrm has an alignment
or ﬁt between context, strategy and structure then better performance should result.
This is not to say, however, that there is a single best ﬁt for a given set of conditions.

Equiﬁnality posits that different ﬁt combinations can produce satisfactory performance.

 

157

This idea of ﬁt as alignment necessitates consideration of interaction and
mediational effects. Performance outcomes are contingent upon this interplay of
variables and effects. Within this research, speciﬁc interaction and mediational paths
were isolated and tested. Unfortunately, no statistically signiﬁcant relationships were
found to support the various paths of the Logistics Contingency Model.

It was hypothesized that the context variables would interact to predict strategy
(H3(k)) and the different structure variables (H3(l)). The lack of support for these
hypotheses leads to the conclusion that signiﬁcant interaction effects are not present in
the model. It should be stressed that, in general, very few interaction effects have been
demonstrated in logistics research. It remains an elusive target.

Three mediational models were examined: (1) Context --> mediated by
Structure --> Performance; (2) Context --> mediated by Strategy --> Structure; and
(3) Strategy -- > mediated by Structure -- > Performance. These models represent,
respectively, hypotheses 4(a)—(d), 4(e)-(h), and 6(b). Figure 5.5 highlights the three
mediational paths.

There are several possible reasons mediation effects were not observed in the
research. Certainly one explanation is that they simply did not exist. If so, logistics
researchers need to identify why organizational science constructs and theory do not
adequately capture logistics relationships. A second possibility relates to one of the
chief criticisms of contingency theory - poor speciﬁcation of the variables. Some of
the constructs - speciﬁcally information technology operating and planning systems and
strategy - have never been used in this type of research. Given that they represent ﬁfty

percent of context and the entire strategy construct, problems within the constructs

it
u

158

would pose a substantial obstacle in the examination of mediation effects. A third
possibility is that the limited sample size created effect size problems. It is possible

that small and medium effects could not be detected.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Context
Environmental "
Dynamism (1)
H Environmental - (1)
“05‘“"3’ Logistics H.R. -> Logistics
. . Structure ._, Performance
+= Logist1cs I.T.
Operating Systems ‘1 (3)
Logistics I.T. " 2 3
Planning Systems ( ) ( )
> Logistics
(2) Strategy
Figure 5 .5

Logistics Contingency Model: Mediation Paths

Although statistically signiﬁcant results were not obtained, there was some
evidence supporting mediation. For example, the direct effect of environmental
hostility on performance resulted in a t-value of -1. 183 and p = .24. However, when
mediated by centralization, environmental hostility had a higher t-value of -1.333 and p
= .18. Even more compelling was the hostility-integration (committees) relationship.
In this case, the t-value improved from -1.183 to -1.672 and the p-value from .24 to
.10. In this case it was evident there was an indirect effect of environmental hostility

through integrative committees on performance.

159

These were not isolated examples. Throughout the mediational relationships,
there was frequently improvement when moving from direct to indirect effects. Of
course, the opposite was also true. But overall, the results offer some encouragement
that future testing might reveal statistically signiﬁcant relationships which support the

mediational hypotheses.

PERFORMANCE-STRATEGY FEEDBACK LOOP

Certainly one of the most intuitively appealing ideas in organizational science
and management research is the idea that poor performance will lead to an adjustment
in strategy. Despite what most would claim to be the obviousness of this assertion, it
has been impossible to empirically demonstrate. This research collected longitudinal
data in an effort to examine this feedback loop in the Logistics Contingency Model.

As discussed in Chapter Four, over half of the respondent ﬁrms have changed
their logistics strategic orientation during a three-year time period. Firms characterized
as having above average performance were more likely to change strategies than those
ﬁrms indicating subpar performance. It is difﬁcult in those cases to argue that poorer
relative performance led to strategic reevaluation and change. This is particularly true
given that the survey questions asked respondents to assess their performance relative to
competitors.

Using the available data, a comparison was made between 1994 strategic choice
and associated perceived performance and 1997 strategy. A chi-square test was
performed to determine if a relationship existed between 1994 strategy-performance

and subsequent 1997 strategy. The test failed to substantiate any statistically

 

160

meaningful relationship. The performance --> logistics strategy results obtained in
this research failed to offer any empirical justiﬁcation for the feedback loop included in
the Logistics Contingency Model.

The results indicated that ﬁrms are much more content to remain in an
internally-oriented strategy than an externally-oriented strategy. Of those externally-
oriented ﬁrms changing strategies, a majority shifted to an internally-oriented strategy.
What then caused both the strategic inertia and change present in this research?

Several factors may be impacting strategic choice. First, resistance to change
may explain why many ﬁrms have not changed strategies. Given the adjustments that
logistics has made in the wake of deregulation and increased competition, ﬁrms may be
content with the status quo. This may be particularly true in the case of internally-
oriented ﬁrms. There would be no outside party pushing them to improve, compared
to the extended channel arrangements associated with the externally-oriented strategy
ﬁrms. Closely related to that is performance and benchmarking. Earlier examination
of data by World Class Logistics revealed a tendency of non-benchmarking ﬁrms to
consistently overrate their performance. The same phenomenon occurred within this
sample. These ﬁrms do not recognize their comparatively poor performance and
therefore feel no need to change.

In terms of those ﬁrms changing strategic orientation, it is important to note two
key points: (1) A majority of ﬁrms making a change were above average performers,
and (2) The majority of ﬁrms had indicated an externally-oriented strategy in 1994 with
many reverting to an internally-focused strategy in 1997. In terms of the ﬁrst point,

these dynamic ﬁrms are not satisﬁed with their performance. They recognize the need

161

to safeguard against complacency and continually seek strategic renewal. The push for
improvement drives their strategic change. As for the second group, information
technology may not have supported their strategic initiatives. Although no longitudinal
information technology data was gathered, an externally-oriented logistics strategy
requires numerous external interfaces. Accurate and up-to-date information is needed
to execute the interfaces. Given the rather harsh assessments concerning operating and
planning systems, it is probable that the information technology systems simply could
not deliver the necessary information in a timely and accurate fashion. Although trade
magazines trumpet the success stories of integrated logistics information systems and
extended enterprises, the mere adequacy of operating and planning systems capability
indicated by this research’s respondents may be the norm rather than the exception.
Faced by this realization, ﬁrms may be regrouping under an internally-oriented
strategy.

It is also possible that logistics finds itself in a “pendulum” mode. If customer
service is emphasized perhaps an externally-oriented strategy gains favor. But if
provision of that customer service is accompanied by too-high costs, the decision may
be made to rein in costs through an internal focus. Therefore, this “pendulum” swing

may explain the shifts in strategy.

CONTRIBUTIONS: ACADEMIC
This research contributes to the body of logistics knowledge in several ways.
First, although not considered in the discussion portion of this chapter, the construct

validation process is crucial in theory development. This research has demonstrated

'1"

 

 

162

application of a method that greatly increases the validity of constructs. This answers
one of the challenges put forth by Mentzer and Kahn (1995) and should encourage
other researchers in the ﬁeld of logistics to consider such approaches. At the same
time, it has called into question the applicability of using formalization in the traditional
organizational science sense. This research demonstrates that logistics characterizes the
formalization-environmental dynamism and hostility relationship in an entirely different
manner than organizational science.

Second, the construct validation process has introduced information technology
operating and planning systems variables to the ﬁeld. Their demonstrated importance
should encourage other researchers to push forward in the investigation of information
technology and its critical role in predicting logistics structure. Given the current
interest and resources devoted to information technology, this is an important ﬁnding.

Third, the Logistics Contingency Model breathes life into - and supersedes - the
Chow et al. framework. It demonstrates one conﬁguration through which contingent
relationships may be examined in logistics. The fact that no signiﬁcant mediational or
interaction effects were found does not invalidate the model. After all, this was the
model’s baptism, not its burial. The fact that it has been used with some success
should encourage other researchers to move forward in this area.

This leads to a closely related fourth contribution, theory testing. Mentzer and
Kahn (1995) lamented the relative lack of theory testing in logistics. The Logistics
Contingency Model provides one vehicle for theory testing. Already it has

demonstrated some interesting results.

163

The ﬁfth contribution is empirical support for a number of relationships. The

following relationships were substantiated:

Higher levels of environmental dynamism and hostility result in greater
formalization. This ﬁnding is contrary to empirical results in organizational
science. In addition, it provides underlying support for the belief that
routinization leads to ﬂexibility, with formalization being a key component
of routinization.

Highly dynamic environments produce signiﬁcant usage of integrative
committees within logistics organizations.

Highly capable operating systems extend span of control and are associated
with higher levels of formalization. Highly capable planning systems also
extend span of control. Other results concerning operating and planning
systems suggest that substantial room for improvement still exists in the area
of information technology.

Externally-oriented logistics strategy ﬁrms were signiﬁcantly more likely to
use integrative committees compared to non-externally-oriented strategy
ﬁrms. Neither strategic orientation could be linked to signiﬁcantly better
performance.

Logistics structure did not impact performance. However, the corollary
relationship - that structure would mediate contextual and strategic

inﬂuences on performance - was not supported.

164

The sixth and ﬁnal major academic contribution of this research is the holistic
nature of the Logistics Contingency Model. Logistics research has typically focused on
single or limited numbers of relationships. Comprehensive models deﬁning logistical
operations are rare. Although much of the preceding discussion has been atomistic in
nature, the Logistics Contingency Model provides a holistic view of the interrelatedness
of logistics operations. In so doing, the overall results raise an issue that is easily
ignored or missed when conducting highly focused research. The issue is: Are
constructs developed in other disciplines transferable to logistics research?

The number of non-signiﬁcant ﬁndings calls into question the transferability of
constructs from other ﬁelds. Although the general meaning of the constructs may be
useful, results of this research deﬁnitely suggest that the transfer of knowledge from
one ﬁeld may be inappropriate when used in a logistics context. The use of borrowed
constructs and verbatim application in logistics research may create more obstacles than
opportunities in conducting meaningful logistics research.

If the theory testing challenge of Mentzer and Kahn (1995) is to be properly
addressed, it should not be built upon simple substitution of borrowed theory.
Nowhere is this more evident in this research than in the results concerning
formalization. Whereas a lengthy and impressive body of organizational science
literature posits a distinct environmental uncertainty and formalization relationship, the
results of this research resoundingly reject that relationship. This does not discredit the
aforementioned literature, it merely points out the potential problems encountered in
borrowing constructs and applying them in a logistics setting. This is not to suggest

that every imported construct is inappropriate. Some constructs such as span of control

 

165

are relatively straightforward and need not be laboriously scrutinized. But in general,
this research highlights the need for further deve10pment of constructs and theory
speciﬁcally rooted in logistics thought and practices, based on holistic relationships

posited in a comprehensive model.

CONTRIBUTIONS: MANAGERIAL

The theoretical challenge of Mentzer and Kahn (1995) was prominently
discussed earlier in this dissertation. While that challenge is necessary, most
academicians studying logistics would agree that ultimately research should have
relevancy to practitioners. The Logistics Contingency Model was designed with these
dual aims in mind. This section discusses the managerial implications of this research.
The following section discusses academic contributions.

The results of this research contribute seven key managerial insights. First, this
research demonstrates the impact of highly capable information technology systems on
logistics. Span of control can be extended through both highly capable information
technology operating and planning systems. Similar to other functional areas within
ﬁrms, information technology extends the reach of logistics executives and managers.
The result has been a decrease in middle management. This research suggests that
logistics is traveling the same road, implementing highly capable information
technology systems and extending span of control. As logistics is asked to do more
with a level or decreased budget, information technology - if properly deployed - may
offer logistics managers the opportunity to decrease personnel while not sacriﬁcing

command and control responsibility.

 

166

Second, despite the positive impact of information technology on span of
control, other anticipated relationships did not materialize. This was due to
underdeveloped information technology capabilities. Respondents reported, on
average, that their logistics information technology operating and planning systems
were not highly capable. Lacking higher capability, most ﬁrms have not been able to
differentiate themselves from their competitors in structuring logistics operations.
Those ﬁrms that do achieve higher capability in logistics information technology will
enjoy a competitive advantage as they become more integrative and formalized.

This mention of formalization leads to a third point. As discussed earlier in this
chapter, in this research higher levels of formalization were associated with more
dynamic and hostile environments. It was speculated that this relationship
demonstrated a means of achieving ﬂexibility - a highly desired operational attribute in
an uncertain environment. This possibility, that formalization may ultimately lead to
ﬂexibility, should encourage managers to examine their written processes and
procedures. While it is fashionable to speak of “empowerment,” it may be that
empowerment is more beneﬁcial in exceptional circumstances than in standard,
repetitive situations.

Fourth, managers should note the absence of most hypothesized signiﬁcant
relationships between environmental dynamism and hostility - strategy and structure.
Considering that most firms in the sample were highly regarded for their logistics
prowess, the absolute absence of concern for environmental dynamism and hostility in
fashioning strategy or structure (other than formalization) might be a clue to other

ﬁrms. They may be concerned about the wrong things if they are expending signiﬁcant

 

167

resources on environmental monitoring.

The ﬁfth managerial contribution of this research concerns strategy. Externally-
oriented strategies do typically have broader span of control. This suggests that ﬁrms
are successful in implementing a management style that extends span of control
internally and externally. Closely tied to this extended ﬁrm is an internal mechanism
designed to foster integration. In this research, ﬁrms with externally-oriented strategies
led the way in the use of integrative cross-functional teams. Although this research
does not demonstrate that increased span of control and greater integration make the
strategy successful, it does clearly differentiate externally-oriented strategy from
internally-oriented strategy. Firms contemplating a change in strategy should expect to
see operational positives in those two areas.

Sixth, strategy is dynamic and equiﬁnality is supported. Over half the ﬁrms
responding to the follow-up questionnaire reported strategic change from 1994 to 1997.
Many of the ﬁrms opting for change had above-average performance in 1994. Lack of
success did not force change upon them - it was deliberate. In addition, there was no
discernible pattern to high performance levels. Both internally and externally-oriented
strategies produced high performance. This was true in 1994 and 1997. The key
appears to be that ﬁrms accurately identify the strategy appropriate to their operational
situation. Internally-oriented strategy continues to be the predominant strategy type.
There was evidence of many 1994 externally-oriented strategy ﬁrms shifting in 1997 to
an internally-oriented strategy while the reverse rarely occurred. Further research is
necessary as it is unclear why ﬁrms choose not to remain in an externally-oriented

strategy.

 

168

Finally, in the area of performance, practitioners are urged to benchmark. The
World Class Logistics Research revealed that non-benchmarking ﬁrms consistently
overrated their own performance in comparison to competitors. This research further
substantiates that ﬁnding. The non-benchmarking ﬁrms again rated their performance
(in 1994 and 1997) better than their competitors. An aggregated measure of the
perceived competencies indicated a different result. Non-benchmarking ﬁrms typically
scored below average on the aggregated measure. Also, these ﬁrms were the most
static in terms of logistics strategy. Few showed any change in strategic choice
between 1994 and 1997, obviously deluded by their own observations. Although there
was no statistical evidence that benchmarking led to high performance, it does appear

lack of benchmarking impeded serious self-assessment.

LIMITATIONS

As is the case with any research, this effort has several limitations. First, and
foremost, is the size and homogeneity of the sample. The sample size restricted the
application of certain statistical approaches. A larger sample may have resulted in
different conclusions. The homogeneity of the sample is an issue. For the most part,
participating ﬁrms represented highly adept logistics practitioners. Also, as members
of the Council of Logistics Management they were more likely to be attuned to logistics
issues and practices. These characteristics limit the generalizability of the results.
Ironically, the homogeneity of the sample may have also limited the number of
signiﬁcant results that might otherwise have been obtained from a randomly drawn

sample.

169

A second limitation is the Operationalization of the variables. Constructs
imported from different literatures have been, to varying degrees, modiﬁed for logistics
research. There is a concern that these constructs may not adequately represent
logistics thought and practice.

The third limitation is that any model or research must be limited in scope in
order to make it tractable. This effort was no different. The number of possible
variables was intentionally limited. A different selection or combination of variables

might have produced different results and/or insights.

DIRECTIONS AND PROPOSITIONS FOR FUTURE RESEARCH

The most immediate and apparent direction for future research is to obtain a
larger, random sample. Before any changes are made to the Logistics Contingency
Model it should be evaluated in terms of a larger sample. This would effectively
broaden the range of ﬁrms beyond top performers.

A larger sample would afford the opportunity to use a methodology comparable
to that used in the limited number of published studies to date. Speciﬁcally, assuming
an adequate sample size, the model would be evaluated using structural equation
modeling. The simultaneous evaluation procedure is better suited for testing the entire
model than the approach used in this research.

Following the evaluation through a larger sample, and any necessary
adjustments to the model, a number of directions are open. The most obvious is the
use of different variables in the model. Size has frequently been demonstrated to have

an impact on organization structure. It should be included in future model use.

170

Environmental dynamism and hostility might be collapsed into an environmental
uncertainty construct. Pfohl and Zollner (1987) suggest using type of product in
further research. Certainly there are many more variables to draw upon.

The information technology variables should be reﬁned and possibly expanded.
Any number of other context variables might be introduced into the model. However,
as a cautionary note, inclusion of any variable should be demonstrated on a theoretical
basis. Proper steps should be taken to ensure the validity and reliability of any new
constructs or variables.

There is also the idea of integration as an outcome variable as proposed by
Chow et al. (1995). The Logistics Contingency Model is ﬂexible enough that one
could easily remove integration from structure and place it between structure and
performance. The results from this research have demonstrated the signiﬁcance of the
integration construct. Perhaps similar to the contradictory ﬁndings associated with
formalization, one might discover that integration does have a different role within
logistics.

Beyond these fairly immediate suggestions, the model should be used in an
international setting. With the growing interest in global supply chain management, the
Logistics Contingency Model is potentially well suited and adaptable enough to use in
testing international applications. Of course, any attempts to do so should be
accompanied by rigorous and thorough construct evaluation. Cultural differences could
easily lead to false ﬁndings if the proper preparatory work is not ﬁrst performed in the

international setting.

 

 

 

171

These future research suggestions should be accompanied by consideration of a
number of propositions emerging from this research. These are as follows:

(1) Environmental dynamism and hostility should be recast in a more
appropriate logistics context. As transferred from organizational science literature
there is considerable question, based on results of this research, as to whether or not
they adequately capture the intent of dynamism and hostility at the logistics level.

(2) Similarly, the use of some structure components from organizational
science should be reexamined and considered from a logistics perspective.
Speciﬁcally, further work should be conducted in the area of formalization and the
closely related area of routinization. Centralization should be investigated to determine
how best to deﬁne it in terms of logistics.

(3) Logistics strategy merits updating. As more is learned about strategy,
more precise dimensions should be identiﬁed. A cost versus customer service focus is
too limiting and the Bowersox and Daugherty typology should be reexamined within
the context of ten years of additional research. Also, factors driving strategic change
deserve attention. This research revealed substantial strategic movement among ﬁrms
but it did not identify the major inﬂuences behind the change.

(4) Performance dimensions pertinent to logistics should be rigorously
investigated. In all likelihood hard performance data will remain difﬁcult to obtain.
Therefore, logistics researchers must determine perceptual measures of logistics
performance that maintain relevancy to logistics and not broader corporate—level
performance. If performance dimensions can be deﬁned with greater accuracy, then it

becomes much easier to determine if ﬁrms have “ﬁt” in their system. It is reasonable

 

172

to believe that different ﬁrms employ different strategies due to different performance
criteria. But the current state of logistics performance constructs makes researching
“ﬁt” a difﬁcult proposition as ﬁrms are measured on relatively crude performance

COIlStI'UCtS .

SUMMARY

At the heart of the CSSP paradigm is the concept of “ﬁt.” According to the
organizational science research it is hypothesized that it is the alignment of a ﬁrm’s
strategy and structure in accordance with its context that will determine performance.
Thus, it is not a single variable that produces a successful outcome but the interaction
of many. It was expected that this research might demonstrate elements of ﬁt through
interaction and mediation tests. The fact that none could be statistically supported
raises a host of questions about sample characteristics, construct deﬁnition,
hypothesized relationships, and the relevance of ﬁt with respect to logistics. This
research represents an early step in this investigative journey of ﬁt and logistics. The
Logistics Contingency Model offers a uniﬁed and coherent framework to continue the
investigation.

Results from this research provided limited support for long-standing
organizational science relationships. For example, highly dynamic environments did
result in ﬁrms using more integrative measures. But more frequently, the results did
not support the hypothesized relationships. For example, the relatively long history of
environmental dynamism and hostility did not predict centralization or formalization of

logistical activities.

 

173

In fact, in terms of formalization the hypothesized relationship was
contradicted. Firms in highly dynamic and hostile environments reported higher levels
of formalization compared to ﬁrms operating in less dynamic and hostile environments.
This surprising result indicates that logistics organizations react differently to
uncertainty than corporations as a whole. In order to cope with dynamic and hostile
environments, logistics organizations formalize procedures whenever possible. This
ﬁnding can further be viewed in the context of the emerging belief that “routinization
leads to ﬂexibility.” Formalization can be thought of as an underlying element of
routinization - providing the mechanism to create routinization through formalization’s
impact on standardization and simplication of procedures.

The investigation of an information technology dimension was fruitful.
Operating and planning systems have impacts on span of control, integrative
mechanisms and formalization. Such ﬁndings are important as future researchers
contemplate contingency designs.

With the exception of the work of Kohn et al. (1990), this is the ﬁrst research
to demonstrate the use of logistics strategy in a CSSP framework. The lack of any
positive ﬁndings associated with logistics strategy does not invalidate its use. It simply
raises the question of how to pursue this issue in the future. Perhaps further reﬁnement
is necessary in this area.

The use of longitudinal data represents an important move forward in logistics
research. The inability to demonstrate support for the purported performance -- >
strategy feedback loop places ﬁndings of this research in the company of others who

have found evidence of this connection elusive. This research did indicate that logistics

 

174

strategy is a dynamic concept within ﬁrms. Over ﬁfty percent of the ﬁrms involved in
the follow-up questionnaire had changed strategic direction over the observation period.
Clearly a relationship exists that merits continued research.

Future research in logistics through a CSSP contingency approach offers
promising potential. The ﬂexibility of the Logistics Contingency Model permits highly
focused examinations into parts of the model or full-blown analysis of the entire
network of relationships. Perhaps most importantly, it provides a point of departure
for logistics researchers wishing to build the theoretical basis of logistics within a
framework whereby results should be comparable. As logistics gains increasing stature
in the boardroom, a stream of research emanating from the Logistics Contingency

Model may potentially offer important direction to academics and practitioners alike.

APPENDICES

.-_. ...

APPENDIX A

WORKBOOK QUESTIONNAIRE

Section One - Background

Please circle the number closest to the competitive situation your company faces.

Minimal time, effort, resources
and managerial attention are
required to keep up with major
competitors.

When decisions are made within
my ﬁrm, possible competitor
reaction or retaliation is_not an
important consideration.

Our major competitors are not
particularly aggressive.

Actions of competitors are easy
to predict.

A minimal amount of time is
spent analyzing major
competitors' strategies and
actions.

Our ﬁrm rarer changes
marketing practices to keep up
with competitors.

Our ﬁrm must rarely change its
logistics practices to keep up
with competitors.

 

Supplier capabilities change at a
very slow rate.

1

2345

175

_M_uc_h time, effort, resources
and managerial attention are
required to keep up with
major competitors.

When decisions are made
within my ﬁrm, possible
competitor reaction or
retaliation i_s an important
consideration.

Our major competitors g
ﬁercely aggressive.

Actions of competitors are
difﬁcult to predict.

A great deal of time is spent
analyzing major competitors'
strategies and actions.

Our ﬁrm frequently changes
its marketing practices to
keep up with competitors.

Our ﬁrm must change its
logistics practices extremely
frequently to keep up with
competitors.

 

Supplier capabilities change
at a rapid rate.

 

 

176

The rate at which products/ 1 2 3 4 5 The rate of product/service

services are becoming obsolete obsolescence in the industry

in the industry is ve1_'y slow. is ve_r_y high (e. g., some
fashion goods).

Demand for logistics services is 1 2 3 4 5 Demand for logistics

m to forecast. services is difﬁcult to
forecast.

Production/service technology is 1 2 3 4 5 The modes of production/
stable and well established. service change often and in a
major way.

 

R_egion Speciﬁc Information
Indicate by a check mark if your ﬁrm manufactures or distributes/sells or sources in
each of the following regions.

Manufacturing Distribution/Sales Sourcing
Africa
Central/Eastern Europe
Mainland Asia
North America
Paciﬁc Rim
South America
Western Europe

 

 

For the regions you checked above, indicate the percent of your ﬁrm's sales in each of
the following geographic regions.

(I)
E.
0
U:

Africa

Central/Eastern Europe
Mainland Asia

North America

Paciﬁc Rim

South America
Western Europe

 

 

177

The following questions concern trends in the expansion or contraction through growth,
acquisition or diversiﬁcation of manufacturing and logistics facilities in various areas of
the world.

1. Please check the categories which describe trends in manufacturing facilities
in each geographic area:

 

a. Europe _ Expansion _ Contraction _ No Change

b. North America _ Expansion _ Contraction _ No Change
(includes Canada and Mexico)

c. Paciﬁc Rim _ Expansion _ Contraction _ No Change
(includes Australia)

2. Please check the categories which describe trends in lggistics facilities in each
geographic area:

 

a. Europe _ Expansion _ Contraction _ No Change

b. North America _ Expansion _ Contraction __ No Change
(includes Canada and Mexico)

c. Paciﬁc Rim __ Expansion _ Contraction _ No Change
(includes Australia)

178
Section Two - Logistics Strategy

Logistics has a separate mission statement. Yes No

If available, please attach a copy of your logistics mission statement.

Logistics strategy includes a priority to reduce:

a.

Using the following scale, what is the primary emphasis of your logistics strategy?

The number of logistics facilities
Strongly Disagree 1 2 3 4 5 Strongly Agree

The number of product/material suppliers
Strongly Disagree 1 2 3 4 5 Strongly Agree

The number of logistics service providers
Strongly Disagree 1 2 3 4 5 Strongly Agree

The number of marginal customers (retailers need not answer)
Strongly Disagree 1 2 3 4 5 Strongly Agree

The number of products or UPC's
Strongly Disagree 1 2 3 4 5 Strongly Agree

Overall complexity of logistics operation
Strongly Disagree 1 2 3 4 5 Strongly Agree

Lowest Highest

Total Cost 1 2 3 4 5 Customer Service

 

179

Please check the statement that most accurately describes your primary logistics
strategy.

 

A process-based strategy is concerned with managing a
broad group of logistics activities as a value-added chain.
PROCESS _ Emphasis is on achieving efﬁciency from managing
purchasing, manufacturing, scheduling and physical
distribution as an integrated system.

 

 

 

 

A market-based strategy is concerned with managing a
limited group of logistics activities for a multidivisional
MARKET _ single business or across multiple business units. The
logistics organization seeks to make joint product
shipments to common customers for different product
groups and seeks to facilitate sales and logistical
coordination by a single order-invoice. Often the senior
sales and logistics executives report to the same manager.

 

 

 

 

A channel-based strategy is concerned with managing
logistics activities performed jointly with dealers and
CHANNEL _ distributors. The strategic orientation places a great deal
of attention on external control. Signiﬁcant amounts of
ﬁnished inventories are typically maintained forward or
downstream in the distribution channel.

 

 

 

OTHER _ If your strategy does not ﬁt into one of the above, please
describe it below.

 

 

 

 

 

 

 

180
Section Three - Relative Performance Competencies

This part of the questionnaire requires three answers for each question. fig, please
indicate the degree of importance you attach to each listed competency. The ﬁve point
scale below should be used to assess the degree of importance of each competency.
Although all of the competencies listed may be important to your ﬁrm's success, please
carefully assess each with respect to the contribution it makes to the achievement of
overall business success.

Five point scale for importance rating

Least important 1 2 3 4 5 Most Important

Second, please specify the performance of your ﬁrm in relation to its major competitors
for the past year (1993) for each competency. A ﬁve point scale is provided below for
you to use in assessing the performance of your ﬁrm with respect to each competency.
Please specify the appropriate response by choosing a single scale value.

Five point scale for importance rating

Worse than Competitors 1 2 3 4 5 Better than Competitors

Lastly, please indicate the approximate percent responsibility of each managerial area
for each competency. Respond 0% if an area has no responsibility. (The total need
not equal 100%)

Response codes for functional areas:
Manufacturing (Mfg): example, Mfg = 30%
Distribution/Logistics (Log): example, Log = 40%
Marketing/Sales (Mkt): example, Mkt = 20%

For retailers, where appropriate please respond to "customer” questions as your next
logistical destination (could be a retail store serviced by a company-owned distribution
center).
1. Product Flexibililty (Customization): The ability to handle difﬁcult,
nonstandard orders to meet special customer speciﬁcations and to manufacture
products characterized by numerous features, options, size and/or colors.

Importance _ Performance _ Mfg=_% Log=_% Mkt=_%

2. Volume Flexibility: The ability to rapidly modify production capacity so as to
accelerate or decelerate production in response to changes in customer demand.

lmportance_ Performance_ Mfg=_% Log=_% Mkt=_%

 

 

 

10.

11.

181

Process Flexibility: The ability to supply low quantities of product efﬁciently
so that product mix changes are easily accommodated.

Importance_ Performance_ Mfg=_% Log=_% Mkt=_%

Low Logistics Cost: The ability to achieve the lowest total cost of logistics
through efﬁcient operations, technology and/or scale economies.

Importance _ Performance _ Mfg=_% Log=_% Mkt=_%

. Delivery Speed: The ability to reduce the time between order receipt and

customer delivery to as close to zero as possible.
Importance _ Performance _ Mfg=_% Log=_% Mkt=_%

Delivery Dependability: The ability to meet quoted or anticipated delivery dates
and quantities on a consistent basis.

Importance _ Performance _ Mfg=_% Log=_% Mkt=_%

Problem Avoidance: The ability to proactively seek solutions to logistics
problems before they occur.

Importance_ Performance_ Mfg=_% Log=_% Mkt=_%

Problem and Complaint Resolution: The ability to quickly solve logistically-
related customer problems and complaints.

Importance _ Performance _ Mfg=_% Log =_% Mkt=_%

Responsiveness to Key Customers: The ability to respond to the needs and
wants of key customers.

Importance _ Performance _ Mfg=_% Log =_% Mkt=_%
Fill Capacity: The ability to provide desired quantities on a consistent basis.
Importance _ Performance _ Mfg=_% Log=_% Mkt=_%

Value-Added Service: The ability to perform services that add value for the
customer during the actual sales process.

Importance_ Performance_ Mfg=_% Log=_% Mkt=_%

 

 

12.

13.

14.

15.

16.

17.

18.

19.

20.

182

Widespread Distribution Coverage: The ability to effectively provide
widespread distribution coverage.

Importance_ Performance_ Mfg=_% Log=_% Mkt=_%

Selective Distribution Coverage: The ability to effectively target selective or
exclusive customers.

Importance_ Performance_ Mfg=_% Log=_% Mkt=_%

Customer Service Flexibility: The ability to accommodate special customer
service requests.

Importance_ Performance_ Mfg=_% Log=_% Mkt=_%

Product Introduction: The ability to accommodate new product introductions
(rollouts to market).

Importance _ Performance _ Mfg=_% Log=_% Mkt=_%
Product Phase Out: The ability to facilitate old product phase out.
Importance _ Performance _ Mfg=_% Log=_% Mkt=_%

Disruption in Supply: The ability to accommodate supply disruption in a
manner that does not adversely affect customers.

Importance __ Performance _ Mfg=_% Log=_% Mkt=_%
Product Recall: The ability to accommodate product recalls.
Importance _ Performance __ Mfg=_% Log=_% Mkt=_%

Product Flexibility During Logistics: The ability to handle product
modiﬁcations while in the logistics system.

Importance_ Performance_ Mfg=_% Log=_% Mkt=_%

Location Flexibility: The ability to service customers from alternative
warehouse locations.

Importance_ Performance_ Mfg=_% Log=_% Mkt=_%

 

21.

22.

23.

24.

25.

26.

27.

28.

29.

183

Reverse Logistics Timing: The ability to perform reverse logistics operations in
a timely manner.

Importance_ Performance_ Mfg=_% Log=_% Mkt=_%

Differentiation: The ability to differentiate logistical service offerings from that
offered by competitors.

Importance __ Performance _ Mfg=_% Log=_% Mkt=_%
Product Innovation: The ability to continuously add new products or variations.
Importance _ Performance _ Mfg =_% Log =_% Mkt=__ %

Order Flexibility: The ability to modify order size, volume or items during
logistics operation.

Importance_ Performance_ Mfg=_% Log=_% Mkt=_%

Delivery Time Flexibility: The ability to accommodate delivery times for
speciﬁc customers.

Importance _ Performance _ Mfg=_% Log=_% Mkt=_%
Expedited Delivery: The ability to expedite shipments or partial shipments.
Importance _ Performance _ Mfg =_% Log=_% Mkt=_%

Advanced Notiﬁcation: The ability to notify customers in advance of delivery
delays or product shortages.

Importance _ Performance __ Mfg=_% Log=_% Mkt=_%

Advanced Delivery Notiﬁcation: The ability to notify customers in advance of
delivery when products will arrive.

Importance_ Performance_ Mfg=_% Log=_% Mkt=_%

Substitution Flexibility: The ability to substitute product or service offerings in
the event of a delay or stockout (versus backorder or line cancellation).

Importance_ Performance_ Mfg=_% Log=_% Mkt=_%

184

30. Innovativeness: The ability to create innovative logistical solutions for speciﬁc
situations, emergencies or customers.
Importance _ Performance _ Mfg=_% Log=_% Mkt=_%
31. Operational Simpliﬁcation: The ability to simplify the overall logistical
process.
Importance _ Performance _ Mfg=_% Log=_% Mkt=_%
32. Operational Standardization: The ability to provide a consistent approach to
performing key logistics work.
Importance _ Performance _ Mfg=_% Log =_% Mkt=_%
Section Four - Organization
§_tru_ctu_r£
1. How many levels are there in your ﬁrm's overall organization? That is, count
the levels along the longest line from direct worker to senior executive. (Please
include the extremes.)
_ Levels for business unit (Firm) __ Levels for distribution/logistics
_ Levels for manufacturing _ Levels for sales (or marketing)
2. How many individuals report directly to each of the following executives (do
not include secretaries or assistants)?
_ Chief executive of the business unit
_ Senior distribution/logistics executive
_ Senior manufacturing executive
__ Senior sales (or marketing) executive
3. In our ﬁrm, an organizational chart is given to:

_ No one _ Chief executive only _ Top two or three executives only
_ Chief executive and most division or department heads
__ Chief executive and gﬂ division or department heads

 

 

185

4. If the following activities are typically part of the formal responsibility of
logistics, as appropriate indicate how long they have been part of logistics and
the nature of responsibility (line or staff). If the activities are not part of
logistics, respond with zero years and indicate if they are likely to be added in
the future.

 

 

 

Length of time as Nature of Likely to be

part of logistics responsibility added in future
(insws) 121.9 gait 16.9: m

a. Sales forecasting ...... [ ] [ ] [ ] [ ]
b. Production planning [ ] [ ] [ ] [ ]
c. Sourcing/purchasing [ ] [ ] [ ] l ]
d. Production planning [ ] [ ] [ ] [ ]

e. Raw materials/work in

process inventory mgmt. _ I ] I I l I l I
f. Finished goods inventory

management .................. _ [ I [ ] [ ] [ ]
g. Intra-company

transportation ............ _ [ I [ ] [ ] [ ]
h. Finished goods

warehousing ............ __ [ ] [ ] [ I [ ]
i. Order processing ....... _ [ ] [ ] [ ] [ l
j. Customer Service ...... _ [ ] [ ] [ ] [ ]
k. Outbound transportation __ [ ] [ ] [ ] [ ]
1. Logistics systems planning_ [ ] [ ] [ ] [ ]
m. Facilities design ............. _ [ ] [ ] [ ] [ l
11. Materials handling ........ _ [ ] [ ] [ ] [ ]
0. Logistics administration _ [ ] [ ] [ ] [ ]
p. International Logistics __ [ ] [ ] [ ] l I
q. Capital equipment

procurement ............ _ [ ] [ ] [ ] [ ]

r. Data processing for
distribution applications

_ [I [I [ll]

 

186

5. In some companies the senior logistics executive manages activities that are not
typically part of a logistics organization. For example, some logistics
organizations have responsibility for data processing, real estate, dealer services
and/or facilities. Does your logistical organization have responsibility for such
non-typical activities? If so, please list in the space below:

 

 

 

 

 

 

 

 

 

6. Highly centralized reﬂects single decision-making authority, usually the ﬁrm's
corporate headquarters. Highly decentralized reﬂects ﬂexibility in decision-
making among individual business units.

 

a. Our logistical planning process is:
Highly Highly
Centralized 1 2 3 4 5 Decentralized

b. Management of our ﬁrm's daily lggtstical activities are:
Highly Highly
Centralized 1 2 3 4 5 Decentralized

Dimensions of Change

 

1. In my ﬁrm, quality programs are used extensively to achieve continuous
improvement.

Strongly Disagree 1 2 3 4 5 Strongly Agree

 

2. In my ﬁrm, logistics quality programs have been very successful.

Strongly Disagree 1 2 3 4 5 Strongly Agree

3. My ﬁrm has undergone extensive logistics process reengineering during the past
ﬁve years.

Strongly Disagree 1 2 3 4 5 Strongly Agree

187
4. The logistics network of my ﬁrm is substantially different than ﬁve years ago.

Strongly Disagree 1 2 3 4 5 Strongly Agree

5. Over the last ﬁve years, my ﬁrm has made a serious effort to empower decision
making among employees.

Strongly Disagree 1 2 3 4 5 Strongly Agree

6. In the last ﬁve years, my ﬁrm has signiﬁcantly centralized decision making.

Strongly Disagree 1 2 3 4 5 Strongly Agree

Customer Contact (Retailers should answer these questions considering their gt}
consumers.)

 

1. The following levels within the logistics organization have direct customer

access.

YE £9
a. Chief logistics ofﬁcer .......................... [ ] [ ]
b. Director ............................................ [ ] [ ]
c. Department manager ........................... [ ] [ ]
d. Sub-department manager .................... [ ] [ ]
e. First-level supervisor .......................... [ ] [ ]

Please indicate whether you agree or disagree with the following statements.

2. In my ﬁrm, customer complaints are systematically collected and analyzed.

Strongly Disagree 1 2 3 4 5 Strongly Agree

3. In my ﬁrm, customer complaints/suggestions are used to improve performance.

Strongly Disagree 1 2 3 4 5 Strongly Agree
4. In my ﬁrm, customer complaints/suggestions are used to determine pay and/or
bonuses.
Strongly Disagree 1 2 3 4 5 Strongly Agree

5. Employees at lower levels in my organization are empowered to make customer
service decisions on the spot.

Strongly Disagree 1 2 3 4 5 Strongly Agree

 

188

Teaming and Cross Job Coordination

In assuring the compatibility among decisions made in one area (e. g., logistics) with
those in other areas (e. g., marketing/sales) certain integrative mechanisms may or may
not be used. Please indicate the extent to which the following are used by answering
these questions.

1.

In my ﬁrm, interdepartmental committees are set up to allow departments to
engage in joint decision-making.

Strongly Disagree 1 2 3 4 5 Strongly Agree

. In my ﬁrm, task forces or temporary groups are set up to facilitate

interdepartmental collaboration.

Strongly Disagree 1 2 3 4 5 Strongly Agree

. In my ﬁrm, liaison personnel exist whose speciﬁc job it is to coordinate the

efforts of several departments for purposes of a project.
Strongly Disagree 1 2 3 4 5 Strongly Agree

In my ﬁrm, cross-functional teams make decisions concerning manufacturing
strategy.
Strongly Disagree 1 2 3 4 5 Strongly Agree

 

. In my ﬁrm, cross-functional teams make decisions concerning distribution or

 

logistics strategy.

 

Strongly Disagree 1 2 3 4 5 Strongly Agree

In my ﬁrm, cross-functional teams make decisions concerning marketing or
sales strategy.

 

Strongly Disagree 1 2 3 4 5 Strongly Agree

Please indicate on the following scale the degree of integration between daily
logistical activities and marketing.

Not Highly Integrated 1 2 3 4 5 Highly Integrated

Please indicate on the following scale the degree of integration between daily
logistical activities and manufacturing.

Not Highly Integrated 1 2 3 4 5 Highly Integrated

 

 

189

Trends

1. Please indicate whether each trend listed below has increased or decreased. Use
the following scale:

 

 

1 = Has increased substantially 4 = Has decreased somewhat
2 = Has increased somewhat 5 = Has decreased substantially
3 = Has remained constant n.a. = Not Applicable
Has Increased Has Decreased
substantiam substantially
a. Global sourcing .................................... 1 2 3 4 5 n.a.

b. Percent ’of total inventory held by suppliers 1 2 3 4 5 n.a.

c. Percent of total inventory at manufacturing

facilities ................................................ 1 2 3 4 5 n.a.
(1. Percent of total inventory at ﬁeld

distribution centers ............................... 1 2 3 4 5 n.a.
e. Percent of total inventory with next

destination customers ............................ 1 2 3 4 5 n.a.
f. Percent of inbound shipments using JIT 1 2 3 4 5 n.a.
g. Percent of outbound shipments using

Quick Response (QR) ............................ 1 2 3 4 5 n.a.
h. Percent of outbound shipments utilizing

Continuous Replenishment (CR) ............ 1 2 3 4 5 n.a.

i. Percent of inbound shipments utilizing
cross-docking ...................................... 1 2 3 4 5 n.a.

j. Percent of outbound shipments utilizing
cross-docking ...................................... 1 2 3 4 5 n.a.

k. The use of time—based strategies in
logistics .............................................. l 2 3 4 5 n.a.

l. Postponement strategies ......................... 1 2 3 4 5 n.a.

 

190

Section Five - Performance Measurement

Below is a list of logistics performance measures. Please indicate if information is
available to perform each measurement. Also, on a scale of 1 to 5, indicate the
importance of these measurements in monitoring operations or identifying problems. ﬂ"
you don't use it, circle n.a. on the scale.

 

 

 

 

 

Information

Available?
Measurement yg m Unimportant - Important
ASSET MANAGEMENT
Inventory turns ............................. [ ] I ] 1 2 3 4 5 n.a.
Inventory levels, number of days supply [ ] [ ] 1 2 3 4 5 n.a.
Obsolete inventory ....................... [ ] [ I 1 2 3 4 5 n.a.
Return on net assets ....................... [ ] [ j 1 2 3 4 5 n.a.
Return on investment ................... . [ ] [ ] 1 2 3 4 5 n.a.
Inventory classiﬁcation (A,B,C) ...... [ ] [ j 1 2 3 4 5 n.a.
COST (Logistics Cost Only)
Total cost .................................... l ] [ ] 1 2 3 4 5 n.a.
Cost per unit ............................... [ ] [ I 1 2 3 4 5 n.a.
Cost as apercentage of sales ........... l ] [ ] 1 2 3 4 5 n.a.
Inbound freight ............................. I ] [ ] 1 2 3 4 5 n.a.
Outbound freight .......................... [ ] [ ] 1 2 3 4 5 n.a.
Administrative ............................. I ] [ ] 1 2 3 4 5 n.a.
Warehouse order processing ........... [ I [ ] 1 2 3 4 5 n.a.
Direct labor .................................. [ ] [ ] 1 2 3 4 5 n.a.
Comparison of actual versus budget [ ] [ ] 1 2 3 4 5 n.a.
Cost trend analysis ....................... [ ] [ ] 1 2 3 4 5 n.a.
Direct product proﬁtability ............ [ ] [ ] 1 2 3 4 5 n.a.
Customer or customer segment
proﬁtability ................................. [ ] [ ] 1 2 3 4 5 n.a.
Inventory carrying ......................... [ ] [ ] 1 2 3 4 5 n.a.

Costof returned goods .................. [] [] 1 2 3 4 5 n.a.

 

COST (Logistics Cost Only)
Cost of damage

Cost of service failure ...................
Cost of backorders ........................
PRODUCTIVITY

Units shipped per employee ...........
Units per labor dollar ...................
Orders per sales representative ......
Comparison to historical standard
Goal programs .............................
Productivity index ........................
Equipment downtime ...................
Order entry productivity ................
Warehouse labor productivity .........
Transportation labor productivity
CUSTOMER SERVICE

Fill rate .......................................
Stockouts .....................................
Shipping errors ............................
On time delivery ...........................
Backorders ...................................
Cycle time ..................................
Delivery consistency ......................
Response time to inquiries ..............
Response accuracy ........................
Complete orders ...........................
Customer complaints .....................
Sales force complaints ...................
Overall reliability ..........................

Overall satisfaction .......................

I]
[l

l]
[I
[I

[I
[I
l]
l]
[l
[I
[I
[I
[I
[I

I]
I]
[I
[I
[l
[I
l]
[l
[I
[I
[I
[I
l]
[I

H

p—au—au—ap—s

N

NNNNNNNNNN

NNNNNNNNNNNNNN

wuuuwwuwwu

wwwuuwwwwuwwwu

.b

k‘k-FnhA-b'kchh-h

&A&b#&&&#&#&A-§

MMMMMMMMU’IM

MMMMMMMMMMMMMM

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

 

 

LOGISTICS @JALITY

Damage frequency ........................ [ ]
Order entry accuracy ................... [ ]
Picking/shipping accuracy ............ [ ]
Document/ invoicing accuracy ......... [ ]
Information availability ................ [ ]
Number of credit claims ................. [ ]
Number of customer returns ........... [ ]

Please indicate if you formally benchmark each of the following performance areas
relative to competitors. Please indicate your performance relative to major

competitors.

Formally

Benchmark?
YE E
Asset Management .................. [ ] [ ]
Logistics cost ......................... [ ] [ ]
Productivity ............................ [ ] [ ]
Customer Service .................... [ ] [ ]
Logistics quality ..................... [ ] [ ]

[I
[I
l]
l]
l]
[I
l]

1
1

1
1
1
1
I

NNNNNNN

2

2
2
2
2

WWMMMWW

Much
Worse than

#hA-ﬁhhh

mumwm

MMMMMMM

Do you formally measure customer expectations concerning customer service?

Yes No

4

##-§&

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

Much
Better Than
Competition - Competition

5

5
5
5
5

If you use unique performance measures aimed at controlling some integrated aspects
of logistical operations, please describe them in the space below:

 

 

 

 

 

 

 

 

193
Section Six - Information Technology

Please indicate whether you disagree or agree with the following statements.

1.

My ﬁrm's logistics information systems capability is better today than ﬁve years
ago.

Strongly Disagree 1 2 3 4 5 Strongly Agree

. My ﬁrm's current logistics information systems are satisfactory in terms of

meeting our requirements.

Strongly Disagree 1 2 3 4 5 Strongly Agree

. Relative to other areas within my ﬁrm, logistics' share of information system

resources has increased over the last ﬁve years.

Strongly Disagree 1 2 3 4 5 Strongly Agree

. Within my ﬁrm, the percentage of transactions completed using EDI has

increased over the last ﬁve years.

Strongly Disagree 1 2 3 4 5 Strongly Agree

. My ﬁrm utilizes industry standards rather than proprietary standards for:

a. The majority of our EDI transmissions.
Strongly Disagree 1 2 3 4 5 Strongly Agree
b. The majority of our bar codes.

Strongly Disagree 1 2 3 4 5 Strongly Agree

My ﬁrm uses bar coding:
a. with gay customers

Strongly Disagree 1 2 3 4 5 Strongly Agree
b. with _all customers

Strongly Disagree 1 2 3 4 5 Strongly Agree
c. with key suppliers

Strongly Disagree 1 2 3 4 5 Strongly Agree
d. with a_ll suppliers

Strongly Disagree 1 2 3 4 5 Strongly Agree

IO.

11.

194

My ﬁrm uses EDI applications:

a.

with _k_ey customers

Strongly Disagree 1 2 3 4 5 Strongly Agree
with _a_ll customers

Strongly Disagree 1 2 3 4 5 Strongly Agree
with E1 suppliers

Strongly Disagree 1 2 3 4 5 Strongly Agree
with a_ll suppliers

Strongly Disagree 1 2 3 4 5 Strongly Agree

My ﬁrm uses Real Time Communications (e. g., Satellite, Direct Connect):

a.

with key customers

Strongly Disagree 1 2 3 4 5 Strongly Agree
with Q customers

Strongly Disagree 1 2 3 4 5 Strongly Agree
with Qy suppliers

Strongly Disagree 1 2 3 4 5 Strongly Agree
with a_ll suppliers

Strongly Disagree 1 2 3 4 5 Strongly Agree

My ﬁrm views the following as essential to increase our competitiveness:

a.

Bar Code Technologies
Strongly Disagree 1 2 3 4 5 Strongly Agree

. EDI Applications

Strongly Disagree 1 2 3 4 5 Strongly Agree
Real Time Communication Systems

Strongly Disagree 1 2 3 4 5 Strongly Agree

My ﬁrm's primary rationale for enhanced logistics information systems is
service improvement.

My ﬁrm's primary rationale for improved logistics information systems is cost

Strongly Disagree 1 2 3 4 5 Strongly Agree

reduction.

Strongly Disagree 1 2 3 4 5 Strongly Agree

 

12.

13.

14.

15.

16.

17.

195

Our ﬁrm has or is currently completing substantial development or replacement
of our logistics operating systems. (Including order entry, order processing,
warehousing and transportation applications.)

Strongly Disagree 1 2 3 4 5 Strongly Agree

Our ﬁrm plans substantial development or replacement of our logistics operating
systems within the next 3 to 5 years.

Strongly Disagree 1 2 3 4 5 Strongly Agree

 

"L

Our ﬁrm has or is currently completing substantial development or replacement
of our logistics planning systems. (Including forecasting, inventory,
distribution, requirements planning and logistics decision support systems.)

Strongly Disagree 1 2 3 4 5 Strongly Agree

Our ﬁrm plans substantial development or replacement of our logistics planning
systems within the next 3 to 5 years.

Strongly Disagree 1 2 3 4 5 Strongly Agree

Our logistics operating systems are substantially (over 60%) purchased from
outside software suppliers or developers.

Strongly Disagree 1 2 3 4 5 Strongly Agree

 

Our logistics planning systems are substantially (over 60%) purchased from
outside software suppliers or developers.

Strongly Disagree 1 2 3 4 5 Strongly Agree

 

196

Systems Evaluation

This section requests an evaluation of your current logistics information system
capabilities. You are asked to evaluate overall adequacy of current logistics operating
and logistics planning systems. Logistics operating systems include such applications as
order entry, order processing, warehousing and transportation. Logistics planning
systems include such applications as forecasting, inventory management and distribution
requirements planning. The following characteristic deﬁnitions are offered to improve
understanding of the question.

Information Sharing - Willingness to share common information across functions within
the ﬁrm.

Flexibility - Ability to adapt processes and capabilities to speciﬁc customer segment
requirements.

Internal Connectivity - Ability to effectively exchange information across managerial
areas within our ﬁrm.

External Connectivity - Ability to effectively exchange information with next
destination customers and/ or suppliers.

1. Using the following scale, please indicate the characteristics of information
related to logistics operating and planning systems.

Low12345High

 

Logistics Logistics

Operating Planning

Systems Systems
Timeliness l 2 3 4 5 l 2 3 4 5
Accuracy l 2 3 4 5 l 2 3 4 5
Availability 1 2 3 4 5 1 2 3 4 5
Exception basis
formatted l 2 3 4 5 1 2 3 4 5
Formatted to
facilitate usage 1 2 3 4 5 1 2 3 4 5
Information sharing .. 1 2 3 4 5 1 2 3 4 S
Flexibility 1 2 3 4 5 1 2 3 4 5
Internal Connectivity 1 2 3 4 5 1 2 3 4 5
External Connectivity l 2 3 4 5 1 2 3 4 5

 

 

 

197
2. Using the scale below, please indicate your ﬁrm's willingness to share the
following types of information with your suppliers and customers.

Very Unlikely 1 2 3 4 5 Very Likely

Next Destination
Suppliers Custptnﬁs____

Inventory Status Inventory Status

Shipment Release Status Shipment Release Status

Forecasts _ Forecasts _
Production Schedules _ Production Schedules

POS (Point of Sale) _ POS (Point of Sale) __
New Product _ New Product _
Marketing/Promotion _ Marketing/Promotion _

 

3. Using the following scale, please indicate the ease with which your ﬁrm is able
to exchange each of the types of information with suppliers and next destination
customers. Ease is deﬁned in terms of speed of information transfer and
responsiveness, not in technological capability.

Difﬁcult to Share 1 2 3 4 5 Easy to Share

Next Destination
Suppliers Custumers

Inventory Status

 

Inventory Status

Shipment Release Status Shipment Release Status

Forecasts _ Forecasts _
Production Schedules _ Production Schedules

POS (Point of Sale) _ POS (Point of Sale) _
New Product __ New Product _
Marketing/Promotion _ Marketing/Promotion __

 

Author's note: The questionnaire had a seventh section entitled Alliances.
However, this dissertation did not use any questions from that section.
Therefore, that section is not reproduced in this appendix.

 

 

APPENDIX B

 

1.

APPENDIX B

FOLLOW-UP QUESTIONNAIRE

 

ALL RESPONSES WILL BE CONFIDENTIAL AND WILL BE
CODED TO SUPPRESS COMPANY AND RESPONDENT
IDENTIFICATION

 

 

 

Please check the statement that most accurately describes your primary logistics
strategy.

 

A process-based strategy is concerned with managing a
broad group of logistics activities as a value-added chain.
PROCESS __ Emphasis is on achieving efﬁciency from managing
purchasing, manufacturing, scheduling and physical
distribution as an integrated system.

 

 

A market-based strategy is concerned with managing a
limited group of logistics activities for a multidivisional
MARKET _ single business or across multiple business units. The
logistics organization seeks to make joint product
shipments to common customers for different product
groups and seeks to facilitate sales and logistical
coordination by a single order-invoice. Often the senior
sales and logistics executives report to the same manager.

 

 

A channel-based strategy is concerned with managing
logistics activities performed jointly with dealers and
CHANNEL __ distributors. The strategic orientation places a great deal
of attention on external control. Signiﬁcant amounts of
ﬁnished inventories are typically maintained forward or
downstream in the distribution channel.

 

 

OTHER _ If your strategy does not ﬁt into one of the above, please
describe it below.

 

 

 

 

 

 

 

198

199

. Using the following scale, what is the primary emphasis of your logistics strategy?

Lowest Highest
Total Cost 1 2 3 4 5 Customer Service

. During the past two years my ﬁrm has increased its organizational commitment to a
more comprehensive integrated supply chain.

Strongly Disagree 1 2 3 4 5 Strongly Agree

. An executive title, using the words "Supply Chain" is being used in my ﬁrm.
Yes No

. Logistics Strategy

For the six subparts of Question #5 please use the following scale. Indicate your
response in the space provided.

Strongly Disagree 1 2 3 4 5 Strongly Agree

Logistics strategy includes a priority to reduce:

. The number of logistics facilities.

a
b. The number of product/material suppliers.

. The number of logistics service suppliers.

c
d. The number of marginal customers (retailers need not answer).

(D

. The number of products or UPCs.

:‘h

Overall complexity of logistics operation.

 

200

6. Relative Performance

Please specify the performance of your ﬁrm in relation to its major competitors for
the past year (1996) for each indicated competency. The following ﬁve point scale
is provided for assessing the relative performance of your ﬁrm.

Worse than Competitors 1 2 3 4 5 Better than Competitors

a. Product Flexibility (Customization): The ability to handle difﬁcult,
nonstandard orders to meet special customer speciﬁcations and to
manufacture products characterized by numerous features, options, size
and/or colors.

b. Low Logistics Cost: The ability to achieve the lowest total cost of
logistics through efﬁcient operations, technology and/or scale
economies.

c. Delivery Speed: The ability to reduce the time between order receipt
and customer delivery to as close to zero as possible.

d. Delivery Dependability: The ability to meet quoted or anticipated
delivery dates and quantities on a consistent basis.

e. Responsiveness to Key Customers: The ability to respond to the needs
and wants of key customers.

f. Order Fill Capacity: The ability to provide desired quantities on a
consistent basis.

_ g. Order Flexibility: The ability to modify order size, volume or
composition during logistics operation.

h. Delivery Time Flexibility: The ability to accommodate delivery times
for speciﬁc customers.

i. Advanced Shipment Notiﬁcation: The ability to notify customers in
advance of delivery when products will arrive.

 

201
7. Benchmarkirtg
Please indicate if you formally benchmark each of the following performance areas

relative to competitors. Please indicate your performance relative to major
competitors.

 

Formally Much Much
Benchmark? Worse than Better Than
yg Q Competition - Competition
Asset Management .................. [ ] [ ] 1 2 3 4 5
Logistics cost ......................... [ j [ ] 1 2 3 4 5
Productivity ............................ [ ] [ ] 1 2 3 4 5
Customer Service .................... [ ] [ ] 1 2 3 4 5
Logistics quality ..................... [ ] [ ] 1 2 3 4 5

THANK YOU FOR YOUR PARTICIPATION!

To receive an analysis of the information collected in this update please complete the
following information.

Name:

 

Company:

 

Mailing Address:

 

 

 

 

 

 

APPENDIX C

 

APPENDIX C

INITIAL CONFIRMATORY FACTOR RESULTS: CONTEXT

CONSTRUCT LOADINGS (t values)

Environmental Hostility

 

 

V1 .259
(2.351)
V2 .058
(0.428)
V3 .564
(7.439)
Environmental Dynamism
V4 .334
(2.434)
V5 .269
(1 .762)
V6 .467
(5.607)
V7 .541
(3.879)
V8 .3 14
(2. 182)
V9 -.097
(-.748)

Information Technology LOS

 

 

V10 .224
(1.557)
V11 .377
(2.705)
V12 .611
(5.450)
Information Technolggy LPS
V13 .862
(8.046)
V14 .740
(5.103)
V15 .599
(4.260)
V16 .766
(6.503)
V17 .758
(6.159)
V18 .705
(6.427)

‘-

GOODNESS-OF—FIT MEASURES

Chi-square ()8) 129.43
Degrees of Freedom 129
Signiﬁcance Level 0.4728
Bentler-Bonett Normed

Fit Index (NFI) 0.993
Bentler-Bonett Nonnormed

Fit Index (NNFI) 1.000
Comparative Fit Index 1.000

202

 

203

Standardized Residual Matrix

3!

y_5_

V_6

y7_

1.320
0.415
0.349
0.550
0.242
0.041
1.033

0.482
0.165
0.282
0.144
0.138
0.291

0.265
0.306
0.151
0.137
0.257

0.791
0.302
0.385
0.456

0.216
0.094
0.221

0.344
0.050

0.971

 

[In

 

APPENDIX D

APPENDIX D

KUDER-RICHARDSON CALCULATIONS

UL?)

Where KRZO = the reliability estimate (R)
N = the number of items
S2 = the variance
p = the proportion of respondents answering “ yes”
q = the proportion of respondents answering “ no”
2 pq = the sum of the products of p times q for each item on the questionnaire

 

KRZO = R=

Source: Psychological Testing - Principles, Applications, and Issues
Robert M. Kaplan and Dennis P. Saccuzzo (1989)

Centralizati on

16
Km... (——17'61' 3'138 )= .8766

16-1 1761

 

Formalization

7 .
KR”: (M79 )= .7362
7-1 4 55

 

204

 

APPENDIX E

APPENDIX E

RESEARCH HYPOTHESES

Hypothesis 1a: Under conditions of high environmental dynamism, ﬁrms will choose
more externally-oriented logistics strategies (i .e. , market or channel) than under
conditions of low environmental dynamism.

Hypothesis 1 b: Under conditions of high environmental dynamism, ﬁmts will use
decentralized structures than under conditions of low environmental dynamism.

Hypothesis 1c: Under conditions of high environmental dynamism, ﬁrms will use less
formalized structures than under conditions of low environmental dynamism.

Hypothesis 1 d: Under conditions of high environmental dynamism, ﬁrms will use
more integrated structures than under conditions of low environmental dynamism.

Hypothesis 1e: Under conditions of high environmental dynamism, ﬁrms will use
narrower spans of control than under conditions of low environmental dynamism.

Hypothesis 2a: Under conditions of high environmental hostility, ﬁrms will choose
more externally-oriented logistics strategies (i .e., market and channel) than under
conditions of low environmental hostility.

Hypothesis 2b: Under conditions of high environmental hostility, ﬁrms will use
decentralized structures than under conditions of low environmental hostility.

Hypothesis 2c: Under conditions of high environmental hostility, ﬁrms will use less
formalized structures than under conditions of low environmental hostility.

Hypothesis 2d: Under conditions of high environmental hostility, ﬁrms will use more
integrated structures than under conditions of low environmental hostility.

Hypothesis 2e: Under conditions of high environmental hostility, ﬁrms will use
narrower spans of control than under conditions of low environmental hostility.

Hypothesis 3a: Under conditions of highly capable logistics information technology
operating systems, ﬁrms will choose more externally-oriented logistics strategies (i. e. ,
market and channel) than under conditions of less capable logistics information
technology operating systems.

205

206

Hypothesis 3b: Under conditions of highly capable logistics information technology
operating systems, ﬁrms will use decentralized structures than under conditions of less
capable logistics information technology Operating systems.

Hypothesis 3c: Under conditions of highly capable logistics information technology
operating systems, ﬁrms will use more formalized structures than under conditions of
less capable logistics information technology operating systems.

Hypothesis 3d: Under conditions of highly capable logistics information technology
operating systems, ﬁrms will use more integrated structures than under conditions of
less capable logistics information technology operating systems.

Hypothesis 3e: Under conditions of highly capable logistics information technology
operating systems, ﬁrms will use broader spans of control than under conditions of less
capable logistics information technology operating systems.

Hypothesis 3f: Under conditions of highly capable logistics information technology
planning systems, ﬁrms will choose more extemally—oriented logistics strategies (i. e.,
market and channel) than under conditions of less capable logistics information
technology planning systems.

Hypothesis 3g: Under conditions of highly capable logistics information technology
planning systems, ﬁrms will use decentralized structures than under conditions of less
capable logistics information technology planning systems.

Hypothesis 3h: Under conditions of highly capable logistics information technology
planning systems, ﬁrms will use more formalized structures than under conditions of
less capable logistics information technology planning systems.

Hypothesis 3i. Under conditions of highly capable logistics information technology
planning systems, ﬁrms will use more integrated structures than under conditions of less
capable logistics information technology planning systems.

Hypothesis 3j: Under conditions of highly capable logistics information technology
planning systems, ﬁrms will use broader spans of control than under conditions of less
capable logistics information technology planning systems.

Hypothesis 3k: With respect to strategy, there will be a signiﬁcant interaction eﬁ'ect
between the levels of environmental dynamism, environmental hostility, information
technology operating systems, and information technology planning systems.

Hypothesis 31: With respect to structure, there will be a signiﬁcant interaction eject
between the levels of environmental dynamism, environmental hostility, information
technology operating systems, and information technology planning systems.

 

 

207

Hypothesis 4a: Structure is a signiﬁcant mediator of the environmental dynamism
-- > performance relationship.

Hypothesis 4b: Structure is a significant mediator of the environmental hostility -- >
performance relationship.

Hypothesis 4c: Structure is a signiﬁcant mediator of the logistics information
technology operating systems -- > performance relationship.

Hypothesis 4d: Structure is a signiﬁcant mediator of the logistics information
technology planning systems -- > performance relationship.

Hypothesis 4e: Strategy is a signiﬁcant mediator of the environmental dynamism -- >
structure relationship.

Hypothesis 4f: Strategy is a signiﬁcant mediator of the environmental hostility -- >
structure relationship.

Hypothesis 4g: Strategy is a signiﬁcant mediator of the logistics information
technology operating systems -- > structure relationship.

Hypothesis 4h: Strategy is a signiﬁcant mediator of the logistics information
technology planning systems -- > structure relationship.

Hypothesis 5a: Externally-oriented logistics strategy (i. e., market or channel) is
associated with less centralized structure than internally-oriented logistics strategy
(i.e., process).

Hypothesis 5b: Externally-oriented logistics strategy (i. e. , market or channel) is
associated with less formalized structure than internally-oriented logistics strategy (i. e. ,
process).

Hypothesis 5c: Externally-oriented logistics strategy (i. e., market or channel) is
associated with more integrated structure than internally-oriented logistics strategy
(i.e., process).

Hypothesis 5d: Externally-oriented logistics strategy (i. e., market or channel) is
associated with broader spans of control structure than internally-oriented logistics
strategy (i. e. , process).

Hypothesis 5 e: Under conditions of satisfactory performance, less strategic changes
will occur than under conditions of unsatisfactory performance.

Hypothesis 6a: Logistics structure, by itself, has no signiﬁcant relationship to
performance.

EL

 

208

Hypothesis 6b: Logistics structure will be a signiﬁcant mediator of the logistics
strategy -- > performance relationship.

BIBLIOGRAPHY

 

BIBLIOGRAPHY

Amburgey, Terry L. and Tina Dacin (1994). “As the Left Foot Follows the Right?
The Dynamics of Strategic and Structural Change,” Academy of Management
Journal 37(6), 1427-1452.

Ansoff, HI. (1991). “Critique of Henry Mintzberg's “The Design School:

Reconsidering the Basic Premises of Strategic Management , Strategic
Management Journal 12, 463-466.

Barley, Stephen R. (1990). “The Alignment of Technology and Structure Through
Roles and Networks,” Administrative Science Quarterly 35(March), 61-103.

Baron, Reuben M. and David A. Kenny (1986). “The Moderator-Mediator Variable
Distinction in Social Psychological Research: Conceptual, Strategic, and

Statistical Considerations,” Journal of Personality and Social Psychology
51(6), 1173-1182.

Bentler, Peter M. (1992). EQS: Structural Equations Program Manual. Los Angeles,
CA: BMDP Statistical Software, Inc.

Bourgeois, L..l., Ill (1980). “Strategy and Environment: A Conceptual Integration,”
Academy of Management Review 5, 25-39.

Bourgeois, L.J., III (1985). “Strategic Goals, Perceived Uncertainty, and Economic

Performance in Volatile Environments,” Academy of Management Journal
28(3), 548-573.

Bower, LL. (1970). Managing the Resource Allocation Process: A Study of Corporate
Planning and Investment. Boston, MA: Harvard Business School Press.

Bowersox, Donald J. and Patricia J. Daugherty (1987). “Emerging Patterns of
Logistical Organization,” Journal of Business Logistics 8(1), 46-60.

Bowersox, Donald J. and Patricia J. Daugherty (1995). “Logistics Paradigms: The
Impact of Information Technology,” Journal of Business Logistics 16(1), 65-81.

209

210

Bowersox, Donald J ., Patricia J. Daugherty, Cornelia L. Droge, Dale S. Rogers and
Daniel L. Wardlow (1989). Leading Edge Logistics: Competitive Positioning
for the 1990s. Oak Brook, IL: Council of Logistics Management.

Burns, Tom and GM. Stalker (1961). The Management of Innovation. London:
Tavistock Publications.

Cabaniss, Roy F. (1995). “Comparing Domestic and International Distributors’
Performance: A Manufacturer’s Perspective,” International Journal of
Physical Distribution & Logistics Management 25(6), 41-54.

Chandler, Alfred (1962). Strategy and Structure: Chapters in the History of Industrial
Enterprise. Cambridge: The M.I.T. Press.

Child, John (1972). “Organizational Structure, Environment and Performance: The
Role of Strategic Choice,” Sociology 6, 1-22.

Chow, Garland, Trevor D. Heaver and Lennart E. Henriksson (1994). “Logistics
Performance: Deﬁnition and Measurement,” International Journal of Physical
Distribution and Logistics Management 24(1), 17-28.

Chow, Garland, Trevor D. Heaver and Lennart E. Henriksson (1995). “Strategy,
Structure and Performance: A Framework for Logistics Research,” Logistics
and Transportation Review 31(4), 285-307.

Christopher, Martin (1986). “Implementing Logistics Strategy,” International Journal
of Physical Distribution and Materials Management 16(1), 52-62.

Clarke, R.L. (1991). “The Measurement of Physical Distribution Productivity: South
Carolina, A Case in Point,” Transportation Journal 31(1), 14-21.

Clarke, R.L. and K.N. Gourdin (1991). “Measuring the Efﬁciency of the Logistics
Process,” Journal of Business Logistics 12(2), 17-33.

Clinton, Steven R. and David J. Closs (1997). “Logistics Strategy: Does It Exist?”
Journal of Business Logistics 18(1), 19-44.

Closs, David J., Thomas J. Goldsby and Steven R. Clinton (1997). “Information
Technology Inﬂuences on World Class Logistics Capability,” International
Journal of Physical Distribution and Logistics Management 27(1), 4-17.

Clemons, Eric K. and Michael C. Row (1993). “Limits to Interﬁrm Coordination
Through Information Technology: Results of a Field Study in Consumer
Package Goods Distribution,” Journal of Management Information Systems
10(1), 73-95.

 

211

Cooper, J.C., M. Browne and M. Peters (1990). “Logistics Performance in Europe:
The Challenge of 1992,” International Journal of Logistics Management 1(1),
28—34.

Council of Logistics Management (1995), World Class Logistics: The Challenge of
Managing Continuous Change. Oak Brook, IL: Council of Logistics
Management.

Cusumano, Michael A. and Kentaro Nobeoka (1992). “Strategy, Structure and
Performance in Product Development: Observations From the Auto Industry,”
Research Policy 21(3), 265-293.

Cyert, R.M. and J.C. March (1963). A Behavioral Theory of the Firm. Englewood
Cliffs, NJ: Prentice-Hall.

Dalton, Dan R., William D. Todor, Michael J. Spendolini, Gordon J. Fielding and
Lyman W. Porter (1980). “ Organization Structure and Performance: A Critical
Review,” Academy of Management Review 5(1), 49-64.

Daugherty, Patricia J., Alexander E. Ellinger and Dale S. Rogers (1995). “Information

Accessibility: Customer Responsiveness and Enhanced Performance,”
International Journal of Physical Distribution & Logistics Management 25(1),
4-17.

Daugherty, Patricia J., Robert E. Sabath and Dale S. Rogers (1992). “Competitive

Advantage Through Customer Responsiveness,” Logistics & Transportation
Review 28(3), 257-271.

Daugherty, Patricia 1., Ted P. Stank and Dale S. Rogers (1992). “The Impact of
Formalization of Warehousing Firms,” International Journal of Logistics
Management 3(2), 49-61.

Davis, GM. and SW. Brown (1974). Logistics Management. Lexington, MA:
Lexington Books.

Davis, Peter S. and Patrick L. Schul (1993). “Addressing the Contingent Effects of
Business Unit Strategic Orientation on Relationships Between Organizational

Context and Business Unit Performance,” Journal of Business Research 27,
183-200.

Day, George S. (1986). Analysis for Strategic Market Decisions. St. Paul, MN: West
Publishing.

Dess, Gregory G. and Donald W. Beard (1984). “Dimensions of Organizational Task
Environments,” Administrative Science Quarterly 29(March), 52-73.

212

Dess, Gregory G. and Peter S. Davis (1984). “Porter's (1980) Generic Strategies as
Determinants of Strategic Group Membership and Organizational
Performance,” Academy of Management Journal 27(3), 467-488.

Diewert, W.E. and A.M. Smith (1992). Productivity Measurement for a Distribution
Firm (unpublished manuscript), University of British Columbia, Department of
Economics.

 

Donaldson, Lex (1982). “Comments on 'Contingency and Choice in Organization
Theory',” Organization Studies 3(1), 65-72.

Donaldson, Lex (1987). “Strategy and Structural Adjustment to Regain Fit and

Performance: In Defence of Contingency Theory,” Journal of Management
Studies 24(1), 1-24.

Downey, H., D. Hellriegel and J. Slocum (1975). “Environmental Uncertainty: The
Construct and Its Application,” Administrative Science Quarterly 20, 613-629.

Drazin, Robert and Andrew H. Van de Ven (1985). “Alternative Forms of Fit in

Contingency Theory,” Administrative Science Quarterly 30(December), 514-
539.

Drbge, Cornelia and Richard Germain (1989). “The Impact of the Centralized
Structuring of Logistics Activities on Span of Control, Formalization and
Performance,” Journal of the Academy of Marketing Science 17(1), 83-89.

Duncan, RB. (1972). “Characteristics of Organizational Environments and Perceived
Environmental Uncertainty,” Administrative Science Quarterly 17(3
September), 313-327.

Fawcett, Stanley E. (1990). “Logistics and Manufacturing Issues in Maquiladora
Operations,” International Journal of Physical Distribution & Logistics
Management 20(4), 13-21.

Fawcett, Stanley E. (1991). “The Status and Impact of Logistics Issues in the Success
of Co-production Via Maquiladoras,” International Journal of Logistics
Management 2(2), 30-41.

Fawcett, Stanley E. and David J. Closs (1993). “Coordinated Global Manufacturing,
The Logistics/Manufacturing Interaction, and Firm Performance,” Journal of
Business Logistics 14(1), 1-25.

Fawcett, Stanley E. and David B. Vellenga (1992). “Transportation Characteristics
and Performance in Maquiladora Operations,” Transportation Journal 31(4),
5-16.

 

 

 

 

213

Galunic, D. Charles and Kathleen M. Eisenhardt (1994). “Renewing the Strategy-
Structure-Performance Paradigm, ” Research in Organizational Behavior 16,
215-255.

Gassenheimer, J .B., Jay U. Sterling and Robert A. Robicheaux (1989). “Long-Term
Channel Member Relationships,” International Journal of Physical Distribution
& Logistics Management 19(12), 15-28.

Gerbing, David W. and James C. Anderson (1988). “An Updated Paradigm for Scale
Development Incorporating Unidimensionality and Its Assessment,” Journal of
Marketing Research 25 (May), 186-192.

Germain, Richard (1989). “Output Standardisation and Logistical Strategy, Structure
and Performance, ” International Journal of Physical Distribution & Materials
Management 19(1), 21-29.

Germain, Richard (1996). “The Role of Context and Structure in Radical and

Incremental Logistics Innovation Adoption,” Journal of Business Research
35(1), 117-127.

Germain, Richard and Cornelia Drbge (1995). “Just-in-Time and Context,”
International Journal of Physical Distribution & Logistics Management 25(1),
18-33.

Germain, Richard, Cornelia Drbge and Patricia J. Daugherty (1994). “The Effect of
Just-in-Time Selling on Organizational Structure: An Empirical Investigation,”
Journal of Marketing Research 31(4), 471-483.

Ginsberg, Ari and N. Venkatraman (1985). “Contingency Perspectives of
Organizational Strategy: A Critical Review of the Empirical Research,”
Academy of Management Review 19(3), 421-434.

Gomes, Roger and John T. Mentzer ( 1991). “The Inﬂuence of Just-In-Time Systems
on Distribution Channel Performance in the Presence of Environmental
Uncertainty,” Transportation Journal 30(4), 36-48.

Gupta, A.K. (1987). “SBU Relations, Corporate-SBU Relations, and SBU
Effectiveness in Strategy Implementation,” Academy of Management Journal
30, 477-500.

Gustin, Craig M., Patricia J. Daugherty and Theodore P. Stank (1995). “The Effects
of Information Availability on Logistics Integration,” Journal of Business
Logistics 16(1), 1-21.

 

214

Gustin, Craig M., Theodore P. Stank and Patricia J. Daugherty (1994).
“Computerization: Supporting Integration,” International Journal of Physical
Distribution & Logistics Management 24(1), 11-16.

Habib, Mohammed M. and Bart Victor (1991). “Strategy, Structure, and Performance
of US. Manufacturing and Service MNCs: A Comparative Analysis,”
Strategic Management Journal 12(8), 589-606.

Hage, J. and R. Dewar (1973). “ Elite Values Versus Organizational Structure in
Predicting Innovations, ” Administrative Science Quarterly 18, 279-290.

Hall, DJ. and M.A. Saias (1980). “Strategy Follows Structurel,” Strategic
Management Journal 1, 149-163.

Hambrick, Donald C. and David Lei (1985). “Toward an Empirical Prioritization of
Contingency Variables for Business Strategy,” Academy of Management
Journal 28(4), 763-788.

Hannan, M.T. and J. Freeman (1984). “Structural Inertia and Organizational
Change,” American Sociological Review 49, 149-164.

Harrington, T.C., Douglas M. Lambert and Martin Christopher (1991). “A
Methodology for Measuring Vendor Performance,” Journal of Business
Logistics 12(1), 83-104.

Hofer, Charles W. (1975). “Toward a Contingency Theory of Business Strategy,”
Academy of Management Journal 18(4), 784-810.

Jarvenpaa, Sirkka L. and Blake Ives (1994). “The Global Network Organization of the
Future: Information Management Opportunities and Challenges,” Journal of
Management Information Systems 10(4), 25-57.

Jennings, Daniel F. and Samuel L. Seaman (1994). “High and Low Levels of
Organizational Adaptation: An Empirical Analysis of Strategy, Structure, and
Performance,” Strategic Management Journal 15(6), 459-475.

Kambil, Ajit and James E. Short (1994). “Electronic Integration and Business Network
Redesign: A Roles-Linkage Perspective,” Journal of Management Information
Systems 10(4), 59-83.

Kaplan, Robert M. and Dennis P. Saccuzzo (1989). Psychological Testing:
Principles, Applications, and Issues. Paciﬁc Grove, CA: Brooks/Cole
Publishing Company.

Kast, Fremont E. and James E. Rosenzweig (1973). Contingency Views of
Organization and Management. Chicago: Science Research Associates.

 

 

 

215

Keats, Barbara W. and Michael A. Hitt (1988). “A Causal Model of Linkages Among
Environmental Dimensions, Macro Organizational Characteristics, and
Performance,” Academy of Management Journal 31(3), 570-598.

Kleinsorge, I.K., Philip B. Schary and R. Tanner (1991). “The Shipper-Carrier
Partnership: A New Tool for Performance Evaluation,” Journal of Business
Logistics 12(2), 35-57.

Kohn, J.W., M.A. McGinnis and PK. Kesava (1990). “Organizational Environment
and Logistics Strategy: An Empirical Study,” International Journal of Physical
Distribution and Logistics Management 20(2), 22-30.

Lawrence, PR. and J. Lorsch (1967). Organization and Environment. Boston, MA:
Harvard University Graduate School of Business Administration.

Lewis, RS. and P.M. Fandt (1989). The Strategy-Structure Fit in Multinational
Corporations: A Revised Model. Academy of Management Conference,
Washington, DC, Academy of Management.

Lindsay, W.M. and L.W. Rue (1980). “Impact of the Organization Environment on
the Long-Range Planning Process: A Contingency View,” Academy of
Management Journal 23, 385-404.

Lorsch, J.W. and 1.]. Morse (1974). Organizations and Their Members: A
Contingency Approach. New York: Harper & Row.

Luthans, Fred (1976). Introduction to Management: A Contingency Approach. New
York: McGraw-Hill.

Luthans, Fred and Todd 1. Stewart (1977). “A General Contingency Theory of
Management,” Academy of Management Review 2(2), 181-195.

Marr, NE. (1991). “Management Sophistication and Service Performance,”
International Journal of Physical Distribution & Logistics Management 21(4),
32-41.

McGinnis, Michael A. and Jonathon W. Kohn (1993). “Logistics Strategy,
Organizational Environment, and Time Competitiveness,” Journal of Business
Logistics 14(2), 1-23.

McWilliams, Abagail and Dennis L. Smart (1993). “Efﬁciency v. Structure-Conduct-
Performance: Implications for Strategy Research and Practice,” Journal of
Management 19(1), 63-78.

Mentzer, John T. and Kenneth B. Kahn (1995). “A Framework of Logistics
Research,” Journal of Business Logistics 16(1), 231-249.

 

216

Miles, Raymond E. and Charles C. Snow (1978). Organizational Strategy, Structure,
and Process. New York, NY: McGraw-Hill.

Miller, Danny (1987). “Strategy Making and Structure: Analysis and Implications for
Performance,” Academy of Management Journal 30(1), 7-32.

Miller, Danny (1988). “Relating Porter's Business Strategies to Environment and
Structure: Analysis and Performance Implications,” Academy of Management
Journal 31(2), 280-380.

Miller, Danny and Cornelia Drbge (1986). “Psychological and Traditional
Determinants of Structure,” Administrative Science Quarterly 31(December),
539-560.

Miller, D. and P. Friesen (1978). “Archetypes of Organizational Transition,”
Administrative Science Quarterly 24, 921-933.

Miller, Jeffrey G., Peter Gilmour and Ronald Van Dierdonck (1981). “Organising for
Materials Management, ” International Journal of Operations & Production
Management 2(1), 38-51.

Mintzberg, Henry (1979). “An Emerging Strategy of "Direct” Research,”
Administrative Science Quarterly 24(4), 582-289.

Mintzberg, Henry (1990). “The Design School: Reconsidering the Basic Premises of
Strategic Management,” Strategic Management Journal 11, 171-195.

Mohr, Jakki and John R. Nevin (1990). “Communication Strategies in Marketing
Channels: A Theoretical Perspective,” Journal of Marketing 54(October), 36-
51.

Nonaka, I. and EM. Nicosia ( 1979). “Marketing Management, Its Environment, and
Information Processing: A Problem of Organizational Design,” Journal of
Business Research 4(December), 277-300.

Nunnally, Jum C. (1967). Psychometric Theory. New York, NY: McGraw-Hill Book
Company.

Parsons, T. (1960). Structure and Process in Modern Societies. Glencoe, IL: Free
Press.

Pennings, Johannes M. (1975). “The Relevance of the Structural-Contingency Model
of Organizational Effectiveness,” Administrative Science Quarterly 20, 393-
410.

217

Persson, Goran (1982). “Organisation Design Strategies for Business Logistics,”
International Journal of Physical Distribution & Materials Management 12(3),
27-36.

Perry, J .H. (1988). “Firm Behavior and Operating Performance in Just-In-Time
Logistics Channels,” Journal of Business Logistics 9(1), 19-33.

Pfohl, Hans-Christian and Werner Zbllner (1987). “Organisation for Logistics: The
Contingency Approach,” International Journal of Physical Distribution &
Materials Management 17(1), 3-16.

Porter, Michael E. (1980). Competitive Strategy. New York, NY: Free Press.

Prescott, John E. (1986). “Environments as Moderators of the Relationship Between
Strategy and Performance,” Academy of Management Journal 29(2), 329-346.

Read, W.F. and MS Miller (1990). “The State of Quality in Logistics,”
International Journal of Physical Distribution & Logistics Management 21(6),
32-47.

Rhea, MJ. and D.L. Shrock (1987). “Measuring the Effectiveness of Physical
Distribution Customer Service Programs,” Journal of Business Logistics 8( 1),
31-45.

Ruekert, Robert W., Orville C. Walker, Jr. and Kenneth J. Roering (1985). “The
Organization of Marketing Activities: A Contingency Theory of Structure and
Performance,” Journal of Marketing 49(W inter), 13-25.

Rumelt, R. (1974). Strategy, Structure, and Economic Performance. Cambridge, MA:
Harvard University Press.

Schary, P. and J. Coakley (1991). “Logistics Organization and the Information
System, ” International Journal of Logistics Management 2(2), 22-29.

Schoonhoven, Claudia Bird (1981). “Problems With Contingency Theory: Testing
Assumptions Hidden Within the Language of Contingency ”Theory”,”
Administrative Science Quarterly 26(September), 349-377.

Selto, Frank, Celia J. Renner and S. Mark Young (1995). “Assessing the
Organizational Fit of a Just-In-Time Manufacturing System: Testing Selection,
Interaction and Systems Models of Contingency Theory,” Accounting,
Organizations and Society 20(7/ 8), 665-684.

Stank, Theodore P., Patricia J. Daugherty and Craig M. Gustin (1994).
“Organizational Structure: Inﬂuence on Logistics Integration, Costs, and

Information System Performance,” The International Journal of Logistics
Management 5(2), 41-52.

 

218
Thompson, James D. (1967). Organizations in Action. New York: McGraw-Hill.

Tosi, Henry, Ramon Aldag and Robert Storey (1973). “On the Measurement of the
Environment: An Assessment of the Lawrence and Lorsch Environmental
Uncertainty Questionnaire,” Administrative Science Quarterly 18, 27-36.

Tosi, Henry L., Jr. and John W. Slocum, Jr. (1984). “Contingency Theory: Some
Suggested Directions,” Journal of Management 10(1), 9-26.

Tung, Rosalie L. (1979). “Dimensions of Organizational Environments: An
Exploratory Study of Their Impact on Organization Structure,” Academy of
Management Journal 22, 672-693.

Van de Ven, Andrew and Robert Drazin (1985). “The Concept of Fit in Contingency
Theory,” Research in Organizational Behavior 7, 333-365.

Van Fleet, David D. and Arthur G. Bedeian (1977). “A History of the Span of
Management,” Academy of Management Review 2(3), 356-372.

Venkatraman, N. (1989). “The Concept of Fit in Strategy Research: Toward Verbal
and Statistical Correspondence,” Academy of Management Review 14(3), 423-
444.

Williams, Lisa R. (1994). “Understanding Distribution Channels: An
Interorganizational Study of EDI Adoption,” Journal of Business Logistics
15(2), 173-203.

Woodward, Joan (1965). Industrial Organization: Theory and Practice. Oxford:
Oxford University Press.

Yavas, U., M. Luqmani and Z.A. Quraeshi (1989). “Purchasing Efﬁcacy in an
Arabian Gulf Country,” International Journal of Physical Distribution &
Logistics Management 19(4), 20-25.

Zeithaml, Valarie A., P. Rajan Varadarajan and Carl P. Zeithaml (1987). “The
Contingency Approach: Its Foundations and Relevance to Theory Building and
Research in Marketing,” European Journal of Marketing 22(7), 37-64.

 

nrcuran STATE UNIV. LIBRoRIEs
IIHIWIIII"WWII"Hllllllllllil”IIIHIIIIHIIIHI
31293016943535