Close
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Organizational learning disabilities: An international perspective
(USC Thesis Other)
Organizational learning disabilities: An international perspective
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI
films the text directly from the original or copy submitted. Thus, some
thesis and dissertation copies are in typewriter face, while others may
be from any type of computer printer.
H ie quality of this reproduction is dependent upon the quality of the
copy submitted. Broken or indistinct print, colored or poor quality
illustrations and photographs, print bleedthrough, substandard margins,
and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete
manuscript and there are missing pages, these will be noted. Also, if
unauthorized copyright material had to be removed, a note will indicate
the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand comer and
continuing from left to right in equal sections with small overlaps. Each
original is also photographed in one exposure and is included in
reduced form at the back of the book.
Photographs included in the original manuscript have been reproduced
xerographicaliy in this copy. Higher quality 6 " x 9 " black and white
photographic prints are available for any photographs or illustrations
appearing in this copy for an additional charge. Contact UMI directly
to order.
A Bel! & Howell Information Company
300 North Zeeb Road. Ann Arbor. M l 48106-1346 USA
313/761-4700 800/521-0600
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ORGANIZATIONAL LEARNING DISABILITIES:
AN INTERNATIONAL PERSPECTIVE
by
Stephen William Nason
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Business Administration)
December 19994
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
UMI Number: 9601036
UMI Microform 9601036
Copyright 1995, by UMI Company. All rights reserved.
This microform edition is protected against unauthorized
copying under Title 17, United States Code.
UMI
300 North Zeeb Road
Ann Arbor, MI 48103
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
UNIVERSITY OF SOUTHERN CALIFORNIA
THE GRADUATE SCHOOL
UNIVERSITY PARK
LOS ANGELES, CALIFORNIA 90007
This dissertation, written by
.
under the direction of h.i§... Dissertation
Committee, and approved by all its members,
has been presented to and accepted by The
Graduate School, in partial fulfillment of re
quirements for the degree of
DOCTOR OF PHILOSOPHY
/
Dean of Graduate Studies
Date December. 13 A. . 1994.
DISSERTATION COMMITTEE
Chairperson
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Though there is great interest in organizational learning
among both academicians and practitioners, the literature remains
a confusing mixture of theories largely unsubstantiated by
empirical research. The goal of this research is to help clarify
organizational learning by focusing on one important aspect,
organizational learning disabilities (i.e., barriers that can
inhibit organizational learning). Specifically, this
dissertation has taken a theoretical approach to examining
organizational learning disabilities by addressing three
fundamental research questions: (1) • What are organizational
learning disabilities? (2) Does the presence of the learning
disabilities reduce the organizational outcomes of innovativeness
and competitiveness? and (3) What antecedent variables are
related to the learning disabilities? The antecedent variables
considered are country culture, organizational culture, HRM,
leadership, and environmental change. Specific hypotheses
between the variables are developed and tested empirically.
A theoretical framework linking the learning disabilities to
their antecedents and outcomes is developed to examine these
questions. This framework involves a four stage model of
organizational learning (Discovery, Invention, Production, and
Generalization), and several organizational learning disabilities
are identified within each stage. Measures that operationalize
the organizational learning disabilities are developed and a
large portion of this dissertation deals with their construct
development and validation.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
A multi-method research strategy involving large scale
surveys, archival sources, and case study ethnographies was
employed; allowing for both depth and breadth in the analysis.
The survey resulted in approximately 2000 responses from 253
usable organizations located primarily in Asia, Europe, and North
America. A key informant approach was utilized and all data was
aggregated to the business unit level. Measurement issues and
the hypotheses are assessed using the LISREL method of structural
equation modeling.
Results indicate that the presence of organizational
learning disabilities reduces organizational innovativeness, and
through innovativeness, performance. Of particular importance
are disabilities that (1) block assessments of the environment,
(2) reduce the variety of perspectives in the firm, (3) prevent
actions from being taken, and (4)limit learnings to a single
unit. The disabilities are also influenced by national culture,
organizational culture, HRM practices, and to a lesser extent,
leadership. The implications for research and practice are
discussed.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
For
Stephanie, Ellie, and Loki;
Who Kept Me Sane
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
ACKNOWLEDGMENTS
I have received a great deal of emotional and
intellectual support while writing my dissertation. First,
I would like to thank my fellow doctoral students— Sally
Baack, Mindy Kirby, Mark Kizilos, and Bill Snyder— for their
invaluable contributions. I learned more from them than any
other single source.
I also need to thank my committee members— Mary Ann Von
Glinow, Gretchen Spreitzer, Arvind Bhambri, and Janet Fulk—
for their encouragement, tolerance, and intellectual rigor.
I am particularly grateful to my chair, Mary Ann Von Glinow,
who became a friend as well as mentor to me. Most of all,
of course, I would like to thank them for passing me.
I would like to thank all the managers who donated
their valuable time by filling out questionnaires and
suffering through interviews. Finally, I owe a great deal
to my parents, who tried their best and should not be held
accountable for the results.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
TABLE OF CONTENTS
I. INTRODUCTION....................................... 1
Statement of the problem...................... 2
Organization of the Proposal.................. 7
II. LITERATURE REVIEW AND CONSTRUCT DEFINITION OF
ORGANIZATIONAL LEARNING............................. 8
Organizational Learning Literature Review.......8
Organizational Learning Disabilities...........16
Chapter Summary............................... 31
III. THEORETICAL FRAMEWORK...............................32
Outcomes......................................33
Antecedents of Organizational Learning..........37
IV. ME THODOLOGY........................................ 59
Survey Research Data.......................... 60
Analyses......................................73
Case Studies..................................77
V. RESULTS AND DISCUSSION. ..........................81
Construct Validation of the Organizational
Learning Disabilities....................... 81
Relationships between Organizational Outcomes
and the Organizational Learning Disabilities.116
Relationships between the Learning
Disabilities and Antecedents................126
Chapter Summary..............................136
VI. CASE STUDIES........................... 139
Samsung .Electronics Co....................... 139
Health Care Systems.......................... 160
Additional Survey Validation..................166
Chapter Summary..............................167
iv
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
VII. CONCLUSIONS.......................................168
Contributions to Organizational Theory........169
Contributions to Practice.................... 175
Directions for Future Research................176
BIBLIOGRAPHY...........................................181
APPENDICES............................................193
v
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Figure
LIST OF FIGURES
2.1 Four Stage Model of Organizational Learning..... 10
2.2 Organizational Learning Disabilities Model...... 17
.,3.1 Antecedents and Outcomes of Org. Learning....... 32
3.2 Org. Learning Outcomes, General Hypotheses...... 35
3.3 Org. Learning Outcomes, Specific Hypotheses..... 3 6
4.1 Learning Disabilities Factor Model.............. 75
5.1a Discovery Disabilities Factor Model............. 87
5.1b Invention Disabilities Factor Model............. 88
5.1c Production Disabilities Factor Model............ 89
5.Id Generalization Disabilities Factor Model........ 90
5.2 Secondary Factor Model of the Organizational
Learning Disabilities........................ 107
5.3 LISREL Results: Learning Disabilities with
Innovativeness and Competitiveness........... 121
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
LIST OF TABLES
Table
2.1 Review of Organizational Learning Models........ 12
2.2 Organizational Learning Disabilities............ 18
4.1 Univaria.te Stastics, Antecedents and Outcomes... 68
5.1 Univariate Stastics, Learning Disabilities...... 82
5.2a LISREL Analyses for the Discovery Learning
Disabilities................................. 91
5.2b LISREL Analyses for the Invention Learning
Disabilities.................................. 92
5.2c LISREL Analyses for the Production Learning
Disabilities.................................. 93
5.2d LISREL Analyses for the Generalization Learning
Disabilities...................... 94
5.3 Comparitive Fit Index Assessments............... 97
5.4 Global Assessments of Fit for the Full
Confirmatory Analysis Model.................. 102
5.5 Internal Assessments for the Full Organizational
Learning Disabilities Model.................. 103
5.6 Global Assessments of fit for the Second Order
Confirmatory Factor Analysis Model........... 109
5.7 Internal Assessments for the Second Order
Confirmatory Factor Analysis Model........... Ill
5.8 Correlations between Learning Disability Scales.. 115
5.9 LISREL Path Results: Learning Disabilities and
Organizational Outcomes...................... 118
vi
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
5.10 Correlations between Learning Disabilities and
Antecedents................................. 128
5.11 Regression Analyses: Learning Disabilities on
Antecedents................................. 130
5.12 Regression Analyses: Summary of Results........ 131
v i i i
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
CHAPTER I
INTRODUCTION
An organization’s ability to learn faster than
competitors may be the only sustainable
competitive advantage [De Geus, 1988].
The "old ways" will not lead to success in the
future. New knowledge, skills, and capabilities
are essential to success. Continual improvement
and learning are fundamental for success in
leaders — and in all employees for that matter.
Continuous and constant questioning of the status
quo is a fundamental requisite for continuous
improvement. All employees must be willing to
question why things are being done the way they
are. Open minded reviews of EVERY aspect of an
organization is necessary for success. [Rich
Teerlink, CEO of Harley Davidson, in a speech to
all managers (cited in Ulrich, Von Glinow, Jick,
Young, & Nason, 1993)].
Organizational learning has recently become an
important topic in both the academic and practitioner
literatures. Heightened global competition has resulted in
an increasingly competitive, complex, and turbulent business
environment where technologies, markets, and competitive
conditions are changing rapidly and unpredictably (Bennis
and Slater, 1969; Drucker, 1993; Emery and Trist, 1973;
Peters, 1987 & 1992; Schon, 1983). Firms that once
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
prospered are now unable to keep up. In this world, an
organization's ability to change or redesign itself
continuously, to function as a "learning organization,1 1 is
necessary for survival (Wick, 1993, p. 4). Some researchers
(e.g. De Geus, 1988; Stata, 1989) even claim that this
ability "may be the only source of competitive advantage" in
the future. Wick (1993, p. 19) puts it even more bluntly by
stating that organizations will either "(l)earn or die."
Unfortunately, it is not clear exactly what organizational
learning is or how it affects performance. My dissertation
addresses these issues by first operationalizing some core
aspects of organizational learning and then developing and
empirically examining a theoretical framework that
incorporates antecedents of organizational learning and
performance outcomes.
Statement of the Problem
The literature on organizational learning has made
significant contributions, yet it suffers from several
serious shortcomings. First, there has been almost no
empirical research on organizational learning (Daft and
Huber, 1987; Simonin and Helleloid, 1993a). Most of the
studies in this area are either "armchair" theoretical,
based on computer simulations, or derived from consulting
style case studies. What little empirical research exists
2
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
is mostly of a case study format that focuses on only one or
a few firms (Simonin and Helleloid, 1993a).
Second, this problem is compounded by the fact that
organizational learning often seems to be "all things to all
people"— an accumulation of every currently fashionable
business practice lumped together with little theoretical
rationale; e.g. total quality, empowerment, JIT, strong
organizational culture, cross functional teams, etc.
(Ulrich, Von Glinow, Jick, Yeung, & Nason, 1993). This
results in a lack of conceptual clarity about the underlying
processes and structures that affect learning in
organizations, and how they can be improved. Peters (1992)
describes the literature on organizational learning as
"maddeningly abstract and vague." Different definitions of
organizational learning abound, yet few to none of these
definitions have been operationalized and empirically
validated.
Third, there has been too much emphasis on the factors
that promote organizational learning and not enough work on
the factors which block it. Several researchers (e.g.
Argyris and Schon, 1978; Nadler, Gerstein, Shaw, &
Associates, 1992; Marcik & Watkins, 1990; Senge, 1990;
Snyder & Cummings, 1992) claim that the presence or absence
of these barriers, or what I'm calling learning
disabilities, ultimately determines the effectiveness of
organizational ^learning. Fourth, there has been almost no
3
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
systematic empirical research attempting to link
organizational learning to performance. Fifth/ the
antecedents to organizational learning (such as
organizational culture, national culture, the HRM system,
and leadership) have not been systematically examined. And
sixth, despite the fact that heightened international
competition helped spark the current interest in
organizational learning and that many of the exemplars of
organizational learning are not American firms, there has
been little analysis of organizational learning from an
international view point (Nonaka 1988, 1990, & 1991 are
prominent exceptions).
The goal in this research is to make a contribution in
each of these areas. First and foremost I heed Daft and
Huber's (1987) call to develop measures that operationalize
basic organizational learning concepts. Organizational
learning is difficult to operationalize as the literature
"in large part remains murky, confused, and difficult to
penetrate" (Garvin, 1993). Even if the literature was less
fragmented, organizational learning theory is too large and
complex to be parsimoniously operationalized in a single
study. Thus, I focus on the more specific and easily
defined area of organizational learning disabilities— i.e.
the barriers which can inhibit organizational learning.
Therefore, a large portion of this dissertation deals with
the construct development and validation of core aspects of
4
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
organizational learning organizational learning disabil
ities .
There is also a need for systematic hypothesis
development and testing (Simonin and Helleloid, 1993a). The
use of organizational learning disabilities allows me not
only to tangibly operationalize organizational learning but
also to empirically examine relationships with antecedents
and outcomes of organizational learning. For example, even
though most researchers claim that organizational learning
leads to increased organizational innovativeness, and
eventually performance, there has been almost no systematic
empirical research that attempts to demonstrate this
relationship. Thus, the second major contribution of this
research is to empirically assess the relationship between
some characteristics of organizational learning (the
learning disabilities) and their outcomes. This is
accomplished through a multi-method research strategy that
utilizes large scale surveys, archival sources, and case
study ethnographies; allowing for both depth and breadth in
the analysis.
The third major contribution involves the hypothesis
development and testing of the organizational learning
disabilities with important antecedent variables. The
antecedent variables considered include international
culture, organizational culture, leadership, the HRM system,
and environmental change. As there has been relatively
5
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
little specific hypothesis development with antecedents to
organizational learning in the literature, this section is
more exploratory than the others.
Finally, organizational learning is particularly
important for firms operating in an international
environment (Simonin and Helleloid, 1993b). The vast array
of different political systems, industry structures,
organizational forms, financial networks, degrees of
customer sophistication, and strategies (Porter, 1990), as
well as cultural differences in the behavior of workers and
consumers (Adler, 1992; Hofstede; 1980) provide
multinational corporations (MNCs) with an almost infinite
set of both opportunities and challenges (Simonin and
Helleloed, 1993b) . MNCs must learn how to use these
differences to their best advantage without becoming
overwhelmed by the complexity or novelty of the various
factors. Such MNCs raise the level of competition in every
market they enter, forcing domestic firms to improve or lose
ground (Porter, 1990). For these reasons organizational
learning is examined within an international context, with
data from firms that are located primarily in Europe, Asia,
and North America.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Organization of the Dissertation
Chapter II provides a brief review and synthesis of the
literature on organizational learning. Four stages of
organizational learning are discussed and several
organizational learning disabilities are identified within
each stage. Chapter III broadens the theoretical framework
by exploring the relationship between organizational
learning with antecedent and outcome variables. Specific
hypotheses to capture the essence of these relationships are
developed. Chapter IV describes the research design and
methodology used to (1) assess the construct validity and
reliability of the learning disabilities and (2) test the
theoretical model. Chapter V presents the results of the
analyses and discusses their implications. Chapter VI
concludes with a discussion of the implications for theory
and practice and directions for future research.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
CHAPTER II
LITERATURE REVIEW AND CONSTRUCT DEFINITION
OF ORGANIZATIONAL LEARNING
The purpose of this chapter is to anchor organizational
learning disabilities in a theoretical framework and then to
operationalize the core concepts. The literature on
organizational learning is reviewed and synthesized to
develop a construct definition of the organizational
learning disabilities. Issues regarding construct validity
and reliability are also discussed.
Organizational Learning Literature Review
This chapter presents a model that integrates two
current conceptualizations of organizational learning. One
perspective emphasizes interpersonal communication (Argyris
and Schon 1978) while the other focuses more on broader
organizational processes and structures (March and Simon
1958, Huber 1991) . The model describes four stages through
which learning occurs, and common organizational learning
disabilities are identified for each stage of the learning
8
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
process. The relevant literature is discussed to provide
context and theoretical support for the model.
Model of Organizational Learning
In the most general sense, organizational learning
refers to an organization's capacity to continuously change
or redesign itself in an effort to meet shifting
environmental characteristics (Garvin, 1993; Fiol and Lyles,
1985; March and Simon, 1958). Beyond this very general
definition, few researchers agree on exactly what
organizational learning entails or build on the previous
literature in their theory construction (Huber, 1991). In
fact, Huber (1991) notes that building on previous
literature is one of the most pressing needs of
organizational learning researchers. The model of
organizational learning that has perhaps the most widespread
acceptance was originally developed by Dewey (1933) and
applied to organization level phenomena by Argyris and Schon
(1978). This model views organizational learning as
consisting of four interrelated processes: discovery,
invention, production, and generalization.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Figure 2.1
Organizational Learning Model
Generalization Production Invention Discovery
Set Goals. Analyze Gap Implement Document
Solutions
Discover * Develop Diffuse
Gaps Solutions
A ssess
The four stages of this model include the (1) Discovery of
gaps between the current situation and the organization's
objectives, (2) the analysis of the gap and Invention of
solutions, (3) the Production of solutions in real world
actions or procedures, and (4) the Generalization of
learning throughout the organization.
All four stages must be successfully accomplished for
the organization as a whole to successfully learn (Argyris &
Schon, 1978. Discovery involves the detection of potential
problems or opportunities in the environment, such as
through environmental scanning or benchmarking. A firm can
also discover problems or opportunities within its own
production system, e.g. through continuous improvement. How
the firm addresses the problems or opportunities occurs in
Invention. Organizations can invent solutions through a
process of systematic experimentation, duplication,
10
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
innovation, or specialization (Ulrich et al. 1993).
Solutions are useless unless they are effectively
implemented, which occurs in Production. Solutions must be
translated into new or modified standard operating
procedures, organizational structures, or reward systems.
Yet, even the implementation of successful new procedures is
not enough to guarantee learning at the organizational
level. For this to occur the learning must be spread to all
relevant areas of the organization. Not only must the
learning be spread from the individual level to the
organizational level (Snyder and Cummings, 1992), but it
must be shared across organizational boundaries— such as
over time, up the hierarchy, between departments, and across
geographic locations (Ulrich et al. 1993). Although these
processes are presented in the model as occurring
sequentially, in practice they often occur simultaneously
(Argyris & Schon, 1978, Marsick & Watkins, 1992).
This model of discovery, invention, production, and
generalization has several advantages. It is simple,
intuitive, and much of the organizational learning
literature fits well within its framework. For example
Weick (1979) discusses three processes of organizational
learning: enactment, selection, and retention. Enactment
involves creating a variety of ideas and is analogous to
discovery. Selection is the process of choosing and
implementing the -ideas from enactment, and closely
11
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
corresponds to invention and production. Retention is
embedding the ideas in the organization so that they endure
and is similar to generalization. Table 2.1 summarizes the
fit between some prominent organization learning research
and the four stage model of discovery, invention,
production, and generalization.
Table 2.1
Review of Orgaanizational Learning Models
Dewy
Argyris &
Schon
Discovery
gap between
expectations and
reality
Invention
Analysis &
develop
solutions
Production
Implement
solutions
Generalizat.
document,
diffuse, &
evaluate
Daft & Huber Information
Acquisition
Information
Distribution
Huber Knowledge—
Acquisition
Knowledge—
Acquisition
Interpretatio.
Information
Distribution
Org. Memory
Weick Enactment Selection Selection Retention
Nevis Acquisition Sharing
Generalizat.
Slocum &
Dilloway
Acquisition Institutional
ization
Application
Review of
Experience
Senge Creative
Tension
Build Shared
Vision
Team
Learning
Systems
Thinking
1 2
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
There are several characteristics of organizational
learning that are particularly important to consider.
Organizational vs. Individual Learning
Organizational learning is not the same as individual
learning. Individual learning is necessary, but not
sufficient for organizational learning to occur. Individual
learning is "organizational" to the extent that it is
communicable and shared publicly by organizational members
and is related to the transformation process of the
organization (Duncan and Weiss, 1979). Organizations
develop learning systems that influence both immediate
members and are also transmitted to others through the
organization’s history and norms (Fiol and Lyles, 1985) and
its routines, procedures, and organizational culture (Levitt
and March, 1988; Walsh, 1991). When organizational learning
is retained in these ways it survives the turnover of
individuals (Fiol and Lyles, 1985; Hedberg, 1981; Levitt and
March, 1988) .
Organizational learning is not merely the sum of all
the individual learning either (Fiol and Lyles, 1985;
Hedberg, 1981). Organizational routines and culture may
change: in response to learning and may affect individual
behavior and organizational functioning without employees
having learned or being conscious that the organization has
13
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
learned (Fiol and Lyles, 1985; Levitt and March, 1988;
Schein, 1985).
The dynamics involved with learning at an
organizational level are very different from issues involved
with how individuals learn, so it is inappropriate to borrow
concepts from the literature on individual learning without
a specific theoretical rationale. Unfortunately
organizational learning often seems to be discussed as if it
were .merely individual level learning on a larger scale,
what Rousseau (1985) calls the ecological cross level
fallacy. This is particularly problematic as many of the
terms organizational learning researchers use are borrowed
from psychology and education. For example, organizational
memory is quite different from an individual memory, or the
sum of all employees' memories. An organization's memory is
embodied in the organization's culture, rules, procedures,
and structure as well as in individuals' memory (Hedberg,
1981; Walsh, 1991). Care must be taken to keep differences
in levels of analysis clear (Rousseau, 1985), and this is
particularly important when reading this dissertation. I
often discuss organizations in this study as if they are
acting or thinking (e.g. I might say that organizations
assess the environment and develop solutions), but I do this
to avoid unwieldy writing and do not intend to
anthropomorphize.
14
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
First and Second Order Learning
Organizational learning can focus on current goals
without questioning basic assumptions (first-order or single
loop learning) or involve changing the goals and underlying
values (second-order or double loop learning). First-order
learning involves improving the organization's ability to
achieve known objectives while working within the current
system and culture (Argyris and Schon, 1978; Bateson, 1972;
Fiol and Lyles, 1985; Snyder and Cummings, 1992). This
adaptive (Cyert and March, 1963; Shrivasta, 1983) or
exploitative (March, 1991) learning involves the refinement
of existing processes in order to adapt to the environment
or maximize efficiency and produces organizations that learn
merely how to survive in relatively stable times (Senge,
1990) .
Second-order learning reevaluates the nature of
objectives as well as the fundamental values and assumptions
underlying them (Argyris and Schon, 1978; Bateson, 1972;
Snyder and Cummings, 1992). Second order learning involves
changing the organizational culture while first order
learning is learning without significant change in the
firm's basic assumptions (Argyris, 1991). Senge (1990)
calls this "real" learning while Fiol and Lyles call it
higher learning. March refers to it as the "exploration" of
new possibilities with a focus on innovation,
15
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
experimentation, and flexibility. Hammer and Champy's
(1993) radical reengineering and reinventing the structure,
processes, and culture of organizations is a good example of
second order learning.
Organizational Learning Disabilities
It is no accident that most organizations learn
poorly. The way they are designed and managed,
the way people’s jobs are defined, and, most
importantly, the way we have all been taught to
think and interact create fundamental learning
disabilities [Senge, 1990: 18].
The framework presented above indicates that successful
organizational learning involves an ability to discover
problems or opportunities, invent solutions to them, produce
those solutions into real actions, and generalize the
results to all relevant areas. However, several researchers
argue that the overall success or failure of organizational
learning is primarily determined by the presence or absence
of learning disabilities which prevent learning from
occurring (Argyris, 1990; Nadler, et al., 1992; Marcik &
Watkins, 1990; Senge, 1990, Wick, 1992). Unfortunately,
such learning disabilities are prevalent throughout most
organizations (Argyris, 1990; Hedberg, 1981; Levitt & March,
1988; Senge, 1990) .
.Argyris and Schon's (1978) four stage model of
organizational learning provides a parsimonious framework
16
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
for examining the organizational learning disabilities
(Snyder and Cummings, 1992) . Through a search of the
relevant literature, three or four learning disabilities
were identified within each stage of the organizational
learning model. The model with its associated learning
disabilities is presented in figure 2.2.
Figure 2.2
Organizational Learning Model
with Organizational Learning Disabilities
Generalization Production Discovery Invention
Blindness
Internal
Blindness
Projection
Competency
Traps
Requisite
Variety
Simple-
m indedness
Schizophrenia
Paralysis
Hyperactivity
Alien Hand
Amnesia
Superstitious
Learning
Diffusion
Deficit
The disabilities are summarized in Table 2.2 on the next
page and are described in more detail in the following
section.
The organizational learning disabilities are
intentionally named to evoke physical and psychological
disabilities as organizational learning analogously is
limited both by structural features of the organization
and
1 7
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Figure 2.2
Organizational Learning Disabilities
DISCOVERY
Blindness obstructions in organizational search or scanning processes
Internal
Blindness
lack of internal performance monitoring
Competency
Traps
past successes instill a false sense of confidence that leaves the
organization for future changes
Projection organization's biases distort perceptions
INVENTION
Requisite Variety lack of variety in skills, information, ideas, & values
Simplemind
edness
deficiencies in analysis and solution generation
Schizophrenia lack of coordination between multiple competing perspectives
within
the organization
PRODUCTION
Paralysis inability to implement new actions or procedures
Hyperactive the organization is constantly adding new procedures without
taking
care to integrate them or reflect on the results
Alien Hand a disconnection between organization intentions and actions
GENERALIZATION
Amnesia inability to monitor experiences and document results
Superstitious - -
Learning
inability to interpret accurately the meanings of experience
Diffusion Deficit Learning is not shared to all relevant parts of the organization
1 8
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
by the social-psychological characteristics of its members.
Even though many of the names of the disabilities are
borrowed from the fields of education and psychology, the
disabilities refer to organizational level processes, not
individual ones.
Discovery
Discovery processes involve identifying potential
problems or opportunities by contrasting the organization's
goals with the current situation (Argyris and Schon, 1978).
The resulting gap between the actual and desired states of
the organization, or between the current reality and vision,
produces a "creative tension" (Senge, 1990) which can
"trigger" the learning process (Hedberg, 1981) . The process
of discovery often results in the identification of
performance gaps that "trigger" the stages of invention and
production. Learning disabilities that typically occur
during discovery are called external blindness (Snyder and
Cummings, 1992), internal blindness, competency traps, and
projection.
External Blindness. This disability refers to an
inability to accurately perceive the organization's
environment (e.g. through poor scanning processes). An
organization's ability to accurately perceive its current
19
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
environment and to anticipate changes is crucial for success
(Duncan and Weiss, 1979; Dutton, Fahey, & Narayanan, 1983;
Hickson, 1987; Huber, 1991). Furthermore, organizations are
often unaware of the extent to which they actually enact
their environment (Cangelosi and Dill, 1965; Smircich and
Stubbart, 1985; Weick, 1979).
Internal Blindness. This disability refers to barriers
in scanning the organization's internal environment. An
understanding of the internal functioning of an organization
and its internal boundaries greatly facilitates learning
(Hedberg, 1981). Internal blindness can also be thought of
as being un-self-aware or unreflective. The fact that most
managers are not aware of what their subordinate's job
really entails an example of this disability.
Competency Traps.
Managing success is a tough job. There’s a very-
fine line between self-confidence and arrogance.
Success often breeds both, along with reluctance
to change. The bureaucracy builds up. The people
start to believe they're invulnerable. Before
they know it, the world changes and they've got to
react [Jack Welch, interview in Fortune 1993].
A history of past successes may retard organizational
learning because "success breeds somnolence" (Hedberg et
al., 1976). Sitkin (1992) argues that a high degree and
consistency of past successes leads to less attention paid
20
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
to discrepant information, greater complacency about
experimenting, and greater risk aversion within the
organization. Therefore, past successes can actually lead
to a reduction in the ability of the organization to learn,
what Levitt and March (1988) call competency traps. Large
organizations can be especially susceptible, because they
are accustomed to getting their way (e.g., American auto and
steel firms in the 1970s).
Organizations may also fall into competency traps in
which they successfully adjust to small changes but are not
prepared for categorical changes that may occur after small
changes have accumulated (Levitt and March, 1988). Over the
last 20 years General Electric's consumer electronics
business made a series of "correct" short term decisions
that when taken together left it unable to compete in the
new more competitive environment (Pucik, 1988). GE was
forced to withdraw from consumer electronics. Organizations
also may frequently improve current, inferior procedures
instead of learning new, superior ones (Levitt and March,
1988). For these reasons the ability of organizations to
unlearn and to reframe the lessons from past successes to
fit the changing environment is crucial for success
(Hedberg, 1981; Lyles, 1987).
Projection. This learning disability occurs when
organizational members have a distorted perception of the
21
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
firm itself or the environment (Snyder and Cummings, 1992).
Organizational politics or financial interests may result in
distorted perceptions of the current situation. A
disproportionately powerful department or group of
stakeholders may also affect perceptions of the environment
(Hickson, 1987). For example, an organization dominated by
individuals with finance backgrounds may underestimate the
human factor. Particularly ambiguous or anxiety producing
situations often result in a denial among organizational
members that a significant problem exists (Schein, 1985) .
Finally, "goal erosion," where aspiration levels are
adjusted downward in the face of difficult objectives can
also result in projection (Hedberg, 1981).
Invention
Invention entails both the analysis of problems
surfaced in the discovery stage and the invention of
solutions to address them. This involves determining the
causes of the problem, developing possible solutions to it,
and choosing one of the solutions (Argyris and Schon, 1978).
One method of accomplishing this at the organizational level
is through building a shared vision (Bennis and Nanus, 1985;
Senge, 1991). The vision can range from an idealized
picture of the future to a solution to a specific problem
(Mohrmari and Cummings, 1989). Learning disabilities that
22
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
occur in the invention stage are simplemindedness, requisite
variety, and schizophrenia.
Simplemindedness. This disability refers to an overly
simplistic approach to problem analysis and solution
generation (Snyder and Cummings, 1992). Simplemindedness
occurs when simple heuristics or "rules of thumb” are
applied to complex situations without careful analysis
(Senge, 1990; Weick, 1979). There is a tendency in
organizations to rely on simple answers to complex
questions, to overemphasize one cause among the many, and to
fail to consider the organization as a complex system with
multiple feedback loops (Argyris, 1990; Senge, 1990). This
can result in conceptual maps and cultures that reinforce
simplistic approaches and fail to question assumptions
(Argyris and Schon, 1978; Senge, 1990).
Requisite Variety. Requisite variety refers to the
total variety of skills, ideas, and values present in an
organization (Hedberg et al., 1976, Weick, 1979). While the
sheer quantity of information available throughout the
organization is important, it is more important that
information comes from different sources and perspectives
(Weick, 1979). Huber (1991) argues that the greater the
number of different interpretations that are developed
within an organization the greater the organizational
23
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
learning, and this is particularly true the more complex the
environment is (Hedberg et al-, 1976).
The disabilities of simplemindedness and requisite
variety are related yet distinct. Simplemindedness refers
to how well information is analyzed, but says nothing about
how much information is present- Requisite variety deals
with the variety of information and variety of different
perspectives available in the firm, but says nothing about
how well the information is analyzed. In simplest terms,
Simplemindedness concerns the quality of analysis while
requisite variety concerns the quantity of information
present.
One counterintuitive yet interesting aspect of
organizational learning is the claim that a regular
incidence of small failures actually promotes learning and
performance (Sitkin, 1992) . The reason for this is that
constant success fosters restricted search, complacency,
risk aversion, and homogeneity while frequent small failures
provide the variety necessary for learning to occur. Sitkin
(1992) defines intelligent failures as containing five
aspects: "(1) they result from thoughtfully planned actions
that (2) have uncertain outcomes, (3) are of modest scale,
(4) are executed and responded to with alacrity, and (5)
take place in a domain familiar enough to permit effective
learning."
24
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Schizophrenia. Organizational schizophrenia occurs
when different units or departments operate independently
without much coordination (Bateson, 1972; Snyder and
Cummings, 1992) . In order for learning to occur at the
organizational level the learning that takes place in a
department or group must be transferred to other areas of
the organization (Crossan, 1991; Duncan and Weiss, 1979).
When' there is little coordination between various
departments any learning tends to remain localized and thus
does not become organizational. Rigid hierarchical
structures and a high degree of politic behavior tend to
retard information transfer across organizational boundaries
and so often result in schizophrenia.
Production
The competitive difference is not in deciding what
to do, but in how to do it. Execution becomes
paramount [Lawrence Bossidy, CEO of Allied Signal,
interview in Fortune, 1993].
Production is the stage of organizational learning
where the solutions from Invention are implemented into
concrete actions, procedures, or structural changes (Argyris
and Schon, 1978). In Production the idealistic vision is
translated into action and real, tangible change. Learning
disabilities in Production include paralysis, hyperactivity,
and "alien-hand syndrome."
‘ . . 25
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Paralysis. Paralysis refers to an inability of the
organization to take action or implement new procedures
(Snyder and Cummings, 1992) . Paralysis can occur when an
organization holds on to the "tired and proven ways of doing
things" (Argyris, 1990) long after their usefulness has
expired. Hornstein (1986) describes this as the "rule of
repeated action," when "[i]n doubt, do what you did
yesterday. If it isn't working do it twice as hard, twice
as fast, and twice as carefully." Great ideas and
solutions may abound, but if these solutions can not be
translated into action there will be little change. Reward
systems that reinforce "old" behaviors will hinder change
(Kerr, 1988) . Rigid rules, procedures, role expectations,
or organizational cultures can all result in an inability of
the organization to take new actions (Hedberg, 1981; March
and Olsen, 1975). Conflicting goals, a weak vision, and
apathy may also retard action (Weick, 1979). Organizations
where failure is heavily punished are prone to inaction as
employees are hesitant to take risks or make any change for
fear of the consequences (Sitkin, 1992).
Hyperactivity is the learning disability in which an
organization is constantly adapting to some new procedure or
other in -a non-serious way or without taking care to
integrate and reflect on the effect of one procedure before
moving on to the next. This can produce a "fad of the week"
2 6
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
or "program du jour" phenomena where employees ignore new
programs, confident that in time they will fade away like
all the rest. Senge (1990) refers to this as the "illusion
of taking charge," where seemingly proactive behaviors are
really an aggressive'set of reactive approaches.
One example of this disability occurred in a hospital
located in eastern Washington that I studied. In July, 1994
the administrator in charge became interested in Total
Quality. They hired consultants to come in and all
administrative personnel sat through endless hours of
meetings on Quality. One month later, just as the managers
were beginning to become familiar with Quality concepts, the
head administrator became interested in self managed teams.
More consultants were hired and an attempt was made to
integrate self managed teams with Total Quality. The
following month corporate headquarters informed the hospital
that it was going to switch away from "traditional medicine"
to a focus on "community health." A month later it was
decided that the organization was to become a "Learning
Organization" along the lines of Senge’s (1990) The Fifth
Discipline. It is not surprising that in this situation the
employees became rapidly "burned out." They went to
required meetings, said the right buzz words, and never made
any changes.
2 7
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
"Alien-Hand Syndrome" refers to the situation where
organizational actions are in conflict with the
organization's goals and objectives (Snyder and Cummings,
1992). The medical condition of "alien-hand syndrome"
occurs when the neural fibers between the two hemispheres of
the brain are severed. This results in the patient's left
and right hands operating independently of each other, often
with out conscious control (Heilman, 1985). An analogous
situation occurs in organizations when the organization or a
department operates in a way that seems contradictory to the
organization's intentions. One of the most common cases of
this disability occurs when organizational reward systems
promote behavior that is contrary to organizational goals
(e.g. basing compensation of salespersons on number of sales
when the organization is trying to focus on quality and long
term relationships) . Weak control mechanisms can also
result in departments operating in their own, not the
organization's, interests (Levitt and March, 1988).
Generalization
History is more or less bunk! [Henry Ford]
Those who cannot remember the past are condemned
to repeat it! [George Santayana]
Generalization refers to the process by which
organizational experience and results are evaluated, encoded
into the organizational memory, and diffused to all relevant
28
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
areas of the organization (Argyris and Schon, 1978).
Encoding into the organizational memory can occur via new
routines, policies, and procedures (Levitt and March, 1988),
conceptual maps (Argyris and Schon, 1978), behavioral norms,
organizational culture (Walsh and Ungson, 1991), as well as
paper documentation and computer data. Diffusion of the
learning throughout the organization occurs through the
communication ' system, computer network, training,
socialization, and cross-functional teams. Learning
disabilities associated with generalization include amnesia,
superstitious learning, and diffusion deficit.
Amnesia. Amnesia refers to an inability to remember
past learnings. Walsh and Ungson (1991) discuss six
locations of organizational memory: individuals,
organizational culture, routines and procedures,
organizational structure, ecology, and archives.
Deficiencies in any of these processes can contribute to
amnesia. The organization may not document or record
results, it may have difficulty translating learnings into
new routines, or turnover may reduce the knowledge located
in individual employees. This can result in the
organization facing similar events over and over, yet having
to learn how to cope with each one anew (Eisenhardt, 1989;
Walsh and Ungson, 1991) . Even if the firm encodes learnings
into the organizational memory, the ability to retrieve
29
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
information is just as important (Walsh and Ungson, 1991).
For example, serious access to the computer network may be
limited to only top level managers (Levitt and March, 1988) .
Superstitious Learning. Superstitious learning
involves the inability to interpret accurately the meaning
of experience, i.e. connections between actions and outcomes
are misspecified (Levitt & March, 1988; Snyder and Cummings,
1992). This is often caused in turbulent environments or
where there is a great deal of ambiguity (March and Olsen,
1975; Weick, 1979). Such ambiguity can result in the
relationship between an organization's actions and the
environment to be unclear (Hedberg, 1981). Causal
relationships are perceived to exist without much support
and without a consideration of other explanations,
particularly strong positive or negative environments
(Levitt and March, 1988).
Diffusion Deficit . In order for learning to occur at
the organizational level, it is critical that the learning
at one unit be diffused throughout the organization (Huber,
1991, Nonaka, 1991) . The corresponding disability develops
when learning occurs in an individual or unit of the
organization but is never spread to other portions of the
firm. This may lead to "audience learning" where employees
or departments are consigned to "audience" status because
30
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
they can watch but not influence actions on the organization
level (Hedberg, 1981). Political fiefdoms and power
dynamics can restrict the diffusion of ideas and procedures
across the organization (Pfeffer, 1981) as can rigid pyramid
shaped organizational structures (Mintzberg, 1979).
Chapter Summary
This chapter presented and discussed a four stage model
of organizational learning. Thirteen organizational
learning disabilities were developed and placed within the
organizational learning model. In Chapter III the
antecedents and outcomes of the learning disabilities are
discussed and specific hypotheses are developed.
3 1
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
CHAPTER III
THEORETICAL FRAMEWORK
In this chapter, I place the model of organizational
learning that was developed in Chapter II within a framework
that includes both antecedents and outcomes of
organizational learning (see Figure 3.1).
Figure 3.1
Antecedents and Outcomes of Organizational learning
f A
Antecedents
r ---- \
Org. Learning
Disabilities
Natl. Culture
Org. Culture
Leadership
HRM
V Env. Change J
Discovery
Invention
Production
Generalization
v J
Outcomes
Innovativeness
Competitiveness
The antecedents posited to have an effect on organizational
learning include (1) international culture, (2)
organizational culture, (3) leadership, (4) HRM system, and
32
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
(5) degree of environmental change- The right side of the
model links organizational learning with the outcomes of
organizational innovativeness and competitiveness.
Specific hypotheses relating these factors to organizational
learning are discussed in the following sections. When
taken together the antecedent, outcomes, and organizational
learning variables form a partial nomological network which
helps further validate the construct and measurement of
organizational learning.
Outcomes
Organizational Learning and Competitiveness
The outcome of organizational learning is not just the
ability of the organization to successfully adapt to a
changing environment (Cyert and March, 1963; Fiol and Lyles,
1985, Shrivasta, 1983), but also the proactive ability to
anticipate and prepare for shifts in the environment
(Argyris, 1992; Huber, 1991; Levitt and March, 1988) and
even to create or reconceptualize the environment and/or the
organization itself (Argyris, 1992; Senge, 1990).
Organizational adaptation is largely reactive and captures
only one aspect of organizational learning. Organizational
innovativeness though, entails not only the potential to
adapt- to changes but also the potential for heading in new
33
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
directions (Miles, 1982) . Thus organizational
innovativeness is used here as the proximate outcome of
organizational learning.
Organizational learning does not necessarily or
directly lead to increased organizational performance
(Huber, 1991). For example, in a stable environment with an
organization that fits the environment well, organizational
. . *
learning would be unnecessary. Of course, there are few
stable business environments today, and an organization's
ability to continuously innovate and change is necessary for
success (Scho.n, 1983, Peters, 1987 & 1992; Quinn, 1992) .
The point here is that the direct outcome of organizational
learning is not performance but innovativeness (Huber,
1991), and it is innovativeness that leads to increased
performance (Miles, 1982).
The organizational learning disabilities prevent
organizational learning from occurring and so directly
reduce organizational innovativeness (Argyris, 1989; Huber,
1991; Levitt and March, 1988; Senge, 1990) and reduce
organizational competitiveness by acting through
innovativeness. So organizational innovativeness is a
mediating variable between the learning disabilities and
competitiveness. Therefore, I hypothesize that:
Hypothesis 1: Organizational learning disabilities
reduce organizational innovativeness and through
innovativeness, organizational competitiveness.
34
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Hypothesis 1 is diagramatically presented in figures
3.2 and 3.3. Figure 3.2 presents the specific hypotheses
between the organizational learning disabilities with
innovativeness and competitiveness.
Figure 3.2
Learning Stages Hypotheses
Discovery
Disabilities
Discovery
Disabilities
Discovery
Disabilities
Discovery
Disabilities
Discovery
Disabilities
Discovery
Disabilities
Discovery
Disabilities
Figure 3.3 groups the 13 disabilities into four categories
according to the corresponding stage of the organizational
learning model and presents the relationship between each
stage and the organizational outcomes.
3 5
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Figure 3.2
Learning Disabilities Hypotheses
Org. Size
External
Blindness
Internal
Blindness
Projection
Requisite
Variety
Simpleminded
Schizophrenia
►"^Competitiveness Innovativeness
Paralysis
Hyperactivity
Alien Hand
Amnesia
Diffusion
Superstitious L
3 6
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
One caveat is in order here. Most definitions of
organizational learning emphasize that successful learning
entails an organization meeting its goals, and these goals
may mot be performance related (e.g. Argyris and Schon,
1978, Argyris, 1992) . For example a central goal for one of
the health care systems used as a case study in this
dissertation was to improve the level of community health.
Therefore, I am making the assumption in hypothesis 1 that
one of the primary goals of the organizations in this study
is to improve their level of competitiveness.
Antecedents of Organizational Learning
There has been very little systematic empirical
research examining the relationship between organizational
learning and its antecedent variables (Huber, 1991) and even
less research has examined the role of these antecedents on
organizational learning disabilities. One objective of this
section is to strengthen our understanding of these links
and present some exploratory hypotheses. The antecedents
considered here include (1) international culture, (2)
organizational culture, (3) leadership, (4) HRM system, and
(5) degree of environmental change. In general, these
factors are hypothesized to create an environment where the
organizational learning disabilities are more or less likely
to occur. I am not positing a deterministic relationship
37
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
where the antecedent conditions cause the learning
disabilities, as the literature is not well enough developed
for that level of specificity. I do hope to show an
association between the antecedents and the disabilities and
draw preliminary inferences from the analysis.
National Culture
Differences in national culture are an important
determinant of the functioning of organizations in general
(Adler, 1990; Hofstede, 1980, 1991) and on organizational
learning in particular (Lyles, 1988, Parkhe, 1991). Despite
this fact, there has been very little research on
organizational learning from an international perspective.
The exception is a growing body of work on the effect of
international joint ventures on organizational learning
(e.g., Bartlett and Ghoshal, 1987; Lyles, 1988; Parkhe,
1991; Pucik, 1988; Simonin and Helleloid, 1993a & b).
The most widely accepted measure of national culture
was developed by Hofstede (1980a). He describes four
dimensions along which cultures vary: uncertainty avoidance,
power distance, individualism— collectivism, and
masculinity— femininity.
Uncertainty avoidance "indicates the extent to which a
society feels threatened by uncertain and ambiguous
38
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
situations and tries to avoid these situations" (Hofstede,
1980b). In high uncertainty avoidance countries (such as in
East Asia) a great deal of emphasis is placed on
establishing rigid, formal rules, not tolerating deviant
ideas and .behaviors, and believing in absolute truths
(Hofstede, 1980b). Low uncertainty avoidance societies (the
US. is the lowest) are willing to take more risks, try new
approaches to solving problems, and feel less constrained to
rigidly adhere to social norms (Hofstede, 1980a, 1991; Shane
et al, 1993) .
Uncertainty avoidance has a complex relationship with
the organizational learning disabilities. Since high
uncertainty avoidance cultures emphasize formal rules and do
not tolerate deviant ideas it might seem that high
uncertainty avoidance would be associated with the presence
of the learning disabilities. However, high uncertainty
avoidance also leads to stress being placed on the reduction
of ambiguity by searching for a structure that can make the
seemingly ambiguous events become clearly interpretable and
predictable (Hofstede, 1980).
Sullivan and Nonaka (1985) argue that the "typical"
form of organizational learning in Japanese firms occurs
through top managers engaging in variety amplification, that
is deliberately creating uncertainty and providing too much
information for middle managers to easily handle. This
ambiguity makes thei^e managers extremely uncomfortable and
39
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
they strive greatly reduce the level of uncertainty.
Sullivan and Nonaka argue that this process of variety
reduction fosters creativity, adaptability, and
organizational effectiveness. So in a way, senior managers
are creating a strength out of what would otherwise be
thought of as a detriment to organizational learning. They
are violating the "rules" of their own culture with the
knowledge that middle managers will not tolerate the created
ambiguity and in the process of reducing the ambiguity the
organization'learns (Sullivan and Nonaka, 1986).
Hofstede (1980) points out that very low uncertainty
avoidance often results in employees being so easy going
that they accept too much ambiguity. In high uncertainty
avoidance cultures individuals tend to work harder at
reducing perceived ambiguity by searching for a framework
that makes the situation understandable and predictable
(Hofstede 1980, Sullivan and Nonaka, 1986). Organizations
in countries with high uncertainty avoidance would be
expected to search the environment more than countries with
lower uncertainty avoidance and thus have fewer of the
blindness and projection disabilities. Such high
uncertainty avoidance cultures would also be expected to
analyze more carefully (simplemindedness) and pay close
attention to insuring that the relationship between causes
and effects was clearly specified (superstitious learning).
Thus, the overall effect of high uncertainty avoidance
40
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
should be a reduction in the organizational learning
disabilities.
Hypothesis 2: Uncertainty avoidance will have a
negative relationship with the organizational
learning disabilities.
Power Distance "indicates the extent to which a society
accepts the fact that power in institutions and
organizations is distributed unequally" (Hofstede, 1980b).
In high power distance countries (e.g. France and Latin
areas) subordinates are not considered equal with superiors,
and this inequality is perceived as both natural and
necessary for the smooth functioning of the organization.
Thus, organizations should be hierarchical and centralized
(Hofstede, 1991). People in high power distance countries
tend to adhere to the organizational hierarchy, believe in
centralizing decision making in superiors, and think that
participative management does not work (Hofstede, 1991,
Shane, 1993). Organizations in low power distance countries
tend to be more egalitarian and collaborative.
High power distance tends to heighten
organizational hierarchy and the power of superiors.
This leads to employees filtering information to give
it a positive spin, since superiors often blame the
messenger. Such filtering would be expected to lead to
superiors developing a less accurate assessment of the
41
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
environment (blindness), being unaware of the way
things really function in their organization (internal
blindness), and receiving a restricted variety of
information (requisite variety). The adherence to the
hierarchy• by organizational members would likely
retard change as well. For these reasons I hypothesize
that high power distance cultures will have more of all
the organizational learning disabilities. Thus
Hypothesis 3: Power distance will be positively
associated with the organizational learning
disabilities.
Individualism-Collectivism "Individualism implies a
loosely knit social framework in which people are supposed
to take care of themselves and of their immediate families
only, while collectivism is characterized by a tight social
framework in which people distinguish between in-groups and
out-groups; they expect their in-group (relatives, clan,
organization) to look after them, and in exchange for this
they feel they owe absolute loyalty to it" (Hofstede,
1980b). According to Hofstede employees in individualistic
countries perceive their relationship to organizations
through a calculative, cost-benefit analysis. Employees in
collectivist countries feel a much stronger moral component
in their relationship to their firm. According to this
conceptualization, individualism and collectivism are
opposite ends of the same scale.
' 42
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
The relationship between the individualism dimension
and organizational learning is complex. Kedia and Bhagat
(1988) state that individualistic countries tend to be more
creative in developing innovations, indicating that
individualism should be positively associated with the
invention stage of organizational learning. While
collective countries have not had as much success developing
innovations, they have had arguably more success
implementing innovations developed by others. This is
particularly true of the Five Tigers of Asia— Japan, South
Korea, Taiwan, Hong Kong, and Singapore (Kedia and Bhagat,
1986). Collectivist countries are much more concerned with
the seamless integration of organizational units and
processes and so tend to rely more on cross-functional teams
(Shane et al., 1993). This emphasis on cross-functional
teams for implementing process innovations and the focus on
the whole organization help explain why collective cultures
are often better implementers. Thus, I hypothesize that
individualism will result in less of the invention
disabilities (simplemindedness, requisite variety, and
schizophrenia) while it will result in more of the
production disabilities (paralysis, hyperactivity, and
alien-hand).
Thus I hypothesize that:
Hypothesis 4a: Individualism will be negatively
associated with the invention disabilities.
43
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Hypothesis 4b: Individualism will be positively
associated with the production disabilities.
The contrasting effects of individualism on the
different disabilities brings up an important point. I am
not hypothesizing that any one type of culture or country
will be better at organizational learning than another. In
fact, often a cultural characteristic will promote some
organizational learning disabilities while retarding others.
Masculinity— Femininity The fourth cultural dimension
that Hofstede discusses has the unfortunate label of
masculinity. The concept really has more to do with
materialism (which Hofstede associates with a masculine
outlook) verses an emphasis on quality of life (which
Hofstede associates with a feminine perspective).1
Masculine cultures, like Japan, are characterized by an
emphasis on assertiveness, ambition, and the acquisition of
money and objects. In masculine cultures performance is
what counts and you live to work (Hofstede, 1986) . Feminine
cultures, such as in Scandinavia, focus on quality of life,
caring, people, and service (Hofstede, 1986). Here the
emphasis is on working to live, not living to work.
1Hofstede believes that an emphasis on material wealth and machismo are
part of the same construct. I do not believe the items he uses to
measure this construct have anything to do with a sense of machismo, so
when I discuss this concept I am referring only to the quantity verses
quality of life components, I use the label "Masculinity and
Femininity" with some discomfort and do so only because these are the
labels that are commonly used in the literature,.
44
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
It is difficult to predict the effect of masculinity on
the learning disabilities. Masculine cultures emphasize
economic growth as a chief priority and so firms in
masculine countries emphasize a strict adherence to
"objective” performance criteria (Kedia and Bhagat, 1988).
Such organizations would be expected to have little
tolerance for stability if it got in the way of performance.
- -
As organizational learning is an effective way of fostering
performance in today's turbulent environments, organizations
in masculine cultures might be more likely to adopt
organizational learning practices.
However, conflicts in masculine cultures tend to be
resolved through a competitive "good fight" (Hofstede,
1980a). This often results in battles for resources and
turf, and it can hinder the flow of ideas across the
organization. Feminine cultures emphasize cooperation,
consensus, equality, and solidarity, and conflicts are
resolved through compromise and negotiation (Hofstede,
1980a). This fosters an exchange of ideas, discussion, and
implementation of decisions. But feminine cultures tend to
be concerned primarily with quality of life and job
satisfaction issues, not performance (Hofstede, 1986).
Given these conflicting tendencies, I offer no
hypotheses concerning masculinity. Instead, I keep
masculinity in the analysis for exploratory purposes and
because Hofstede (1980a) claims that all four cultural
45
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
dimensions should be considered in order to get an adequate
picture of the effect of culture.
Organizational Culture
Organizational culture is the underlying pattern of
basic assumptions, values, beliefs, and meanings which shape
employees cognitions and behaviors (Denison, 1990; Schein,
1985a; Smircich, 1983). Organizational culture has a
powerful influence on organizational learning. Wilkins and
Dyer (1988) describe organizational culture as socially
acquired and shared knowledge that is embedded in specific
and general organizational frames of reference. Schein
(1985b) relates culture even more specifically to
organizational learning through his description as:
(T)he culture of any group or social unit is ...
the total of the collective or shared learning of
that unit as it develops its capacity to survive
in its external environment and to manage its own
internal affairs. Culture is the solution to
external and internal problems that has worked
consistently for a group and that is therefore
taught to new members as the correct way to
perceive, think about, and feel in relation to
those problems. Such solutions eventually come to
be assumptions about the nature of reality ...
then they come to be taken for granted and,
finally, drop out of awareness.
Not only is organizational culture a result of shared
learning, but it also shapes future behavior (Denison, 1990;
Quinn and McGrath, 1985, Schein, 1985a). Geertz (1973: 44)
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
points out that culture often acts as "a set of control
mechanisms — plans, recipes, rules, instructions — for the
governing of behavior."
Quinn and his colleagues (Quinn and Rohbraugh, 1983;
Quinn and McGrath, 1985, Quinn, 1988, Quinn and Spreitzer,
1991) have developed a framework for describing four types
of organizational cultures: clans, adhocracies, hierarchies,
and markets. The four culture types are based on a
competing values perspective along two dimensions: tight or
loose control and internal or external focus. Tight
controls involve an emphasis on centralization and
integration while loose controls are more compatible with
decentralization and flexibility. With an internal focus
there is more concern with the inner workings of the
organization, while with an external focus there is greater
emphasis on looking outside the organization at the overall
competitive position.
A hierarchy culture is characterized by tight controls
and an internal focus. The emphasis is on stability,
control, predictability, security, and rules. I suggest
that such a bureaucratic culture is unlikely to promote
organizational learning. The internal focus often results
in littl.e or no scanning (blindness) and the combination of
internal focus and tight controls is likely to restrict the
variety of information available (requisite variety)
resulting in simplistic analysis and solution generation
47
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
(siirtplemindedness) . Furthermore, too much of a focus on
rules and stability can result in an aversion to new
procedures (paralysis) and exaggerated confidence in the
benefits pf business as usual (competency traps). For all
these reasons I hypothesize that hierarchy cultures will be
positively associated with the organizational learning
disabilities.
Hypothesis 5a: Hierarchy cultures will have a
positive association with the organizational
learning disabilities.
Adhocracy cultures are polar opposites to hierarchy
cultures on the competing values scales. They are
characterized by a loose, decentralized control and an
external focus. Quinn describes this as an innovative
values system that emphasizes adaptability, growth,
flexibility, and risk. Such an external focus with flexible
controls is likely to have effective scanning (blindness),
be open to divergent information (requisite variety), use
this information to make effective analyses
(simplemindedness), and be willing to try new procedures
(paralysis). For these reasons I hypothesize that
adhocracies will have few of the organizational learning
disabilities.
Hypothesis 5b: Adhocracy cultures will be
negatively associated with the organizational
learning disabilities.
48
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
A clan culture is characterized by a loose,
decentralized control with an internal focus. Clan cultures
emphasize human resource values such as morale, involvement,
cohesion, and openness. On the one hand clan cultures
foster "active" employees that are likely to be committed to
the organization and flexible (Spreitzer, 1992). These
qualities are necessary for effective implementation of
ideas (production), sharing of information (requisite
variety), and internal assessment of the organization
(internal blindness). On the other hand, the internal focus
may result . in poor environmental scanning (blindness).
Overall though, I hypothesize that clan cultures will have
few of the organizational learning disabilities.
Hypothesis 5c: Clan cultures will be negatively
associated with the organizational learning
disabilities.
Market cultures have tight controls with an external
focus. This promotes a rational value system with a focus
on output, production, efficiency, productivity, and
objectivity. The external focus is likely to make these
firms particularly aware of their environment and so limit
blindness while the tight controls might restrict
innovativeness. I suggest that this will be compensated for
as the combination of tight controls with the focus on the
external environment, profits, and rationality is likely to
prevent biases from distorting perceptions (projection),
49
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
insure efficient coordination (schizophrenia), and promote
change once old procedures no longer positively affect the
bottom line (paralysis). For these reasons I believe that a
market culture will be negatively associated with the
learning disabilities. Thus, I hypothesize that:
Hypothesis 5d: Market cultures will be negatively
associated with the organizational learning
disabilities, but to a lesser extent than
adhocracy and clan cultures.
In the real, messy world it is unlikely that any one
type of organizational culture will perfectly promote all
aspects of organizational learning, especially when the
culture types are as general as Quinns' competing values
typology. For example the cultural type that is
hypothesized to have few of the learning disabilities, an
adhocracy, may be especially poor at monitoring its internal
processes (internal blindness) as it has an external focus
on the competing values dimensions.
Leadership
My job is to listen to, search for, think of, and
spread ideas, to expose people to good ideas and
role models. I'm almost a maitre d', getting the
crowd to come sit at this table: "Enjoy the food
here. Try it. See if it tastes good" [Jack
Welch].
While leadership is a crucial component of
organizational learning, learning organizations require a
new sort of leadership (Senge, 1990; Wick, 1993) . In fact
50
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
the traditional authoritarian style leaders may actually
retard organizational learning (Stata, 1989; Senge, 1990;
Wick, 1993) . What is needed is a new breed of leader that
acts more like a designer, steward, teacher, visionary
(Senge, 1990; Wick, 1992), and even maitre d’. Based on
the change and leadership literatures (particularly Kanter,
Stein, and Jick, 1992) Ulrich and his colleges (e.g. Ulrich
et al, 1993) have identified seven processes that are
important in learning organizations. The processes are
creating a shared need for change, shaping a vision, leading
the change, mobilizing commitment to the change, using
systems and structures, monitoring progress, and embedding
the change.
The first leadership process involves creating a shared
need for change. Organizations and individuals tend to
resist change unless convinced that it is necessary (Schein,
1985; Sashkin, 1988; Senge 1990). The leader must assess
the environment and show that it presents a challenge or
opportunity that the organization must to respond to (Conger
and Kanungo, 1988: Eisenhardt, 1989; Tichy and Devana,
198 6). Creating a shared need for change is particularly
relevant to the organizational learning disabilities in the
discovery stage. Discovery entails the organization
becoming aware that there is a problem or opportunity in the
environment. One way that this awareness may develop is by
51
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
the leader making organizational members feel the need for
change and open their eyes.
Once there is a felt need for change, leaders should
provide a road map or plan of action for dealing with the
change, and this is accomplished through forming and
articulating a vision (Bennis and Nanus, 1985; Conger, 1989;
Tichy and Devana, 1986; Westley and Mintzberg, 1987). The
vision should present a set of ideal goals and solutions
that not only satisfy employees concerns about the change,
but the vision should do this in a way that creates
enthusiasm and meaning (Bennis and Nanus, 1985; Conger and
Kanungo, 1988; Sashkin, 1988). A leader's shaping a vision
is one way in which the invention stage of organizational
learning may be accomplished.
It is vitally important to institutionalize the change
or learning into actual, tangible behaviors and structures
(Conger, 1989; Kotter, 1985; Schein, 1985; Nadler and
Tushman, 1989). The third, fourth, fifth, and sixth
leadership processes reflect this importance. "Leading the
change" emphasizes the importance of having a leader that
continues to make the change effort a primary concern over
the whole life cycle of the change effort. As employees pay
more attention to what leaders do than to what they say,
continual concern (Bennis and Nanus, 1985) and role modeling
by the leader are important (Conger, 1989; Schein, 1985).
"Mobilizing commitment" to the change continues to play an
52
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
important role. Most change efforts start well then fizzle
out (Mohrman and Cummings, 1989), so the leader must work to
keep the change effort going (Bass, 1988; Bennis and Nanus,
1985) through such activities as maintaining coalitions,
problem solving, meetings and other "mundane" actions
(Nadler and Tushman, 1989). "Using systems and structures"
involves developing an administrative apparatus and
organizational structures to support the change (Bass, 1989,
Schein, 1985). "Monitoring progress" is concerned with
making sure the progress is real and sharing the results.
All these leadership processes which help institutionalize
change act to reduce or counter the learning disabilities in
the production stage of organizational learning.
The final leadership process, embedding the change,
helps accomplish the generalization phase of the
organizational learning model. Embedding the change means
ensuring that the learning is not forgotten (amnesia) and is
shared throughout the organization (diffusion deficit).
Without the embedding of change, not only will the change
process fail, but there can be no learning (Huber, 1991;
Walsh and Ungson, 1991) . Two of the most important ways
that the process of embedding can occur are through the
leader using symbols, language, and rites to embed the
change into the organizational culture (Bass, 1988; Schein,
1985) and by empowering employees to become leaders of the
53
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
change in their own rights (Bennis and Nanus, 1985; Westley
and Mintzberg, 1987).
This discussion leads to the following hypotheses:
Hypothesis 6: The leadership behaviors will be
negatively associated with the organizational
learning disabilities.
Hypothesis 6a: The leadership process of creating
a shared need will be most strongly associated
with a reduction of the learning disabilities in
the discovery phase of the organizational
learning model.
Hypothesis 6b: The leadership process of shaping a
vision will be most strongly associated with a
reduction of the learning disabilities in the
invention phase of the organizational learning
model.
Hypothesis 6c: The leadership processes of
leading the change, mobilizing commitment to the
change, using systems and structures, and
monitoring progress will be most strongly
associated with a reduction of the learning
disabilities in the production phase of the
organizational learning model.
Hypothesis 6d: The leadership process of embedding
the change will be most strongly associated with a
reduction of the learning disabilities in the
generalization phase of the organizational
learning model.
Human Resource Management
Human resource management practices are key with either
promoting or hindering organizational learning (Pucik, 1988,
Ulrich and Lake, 1990) and many learning disabilities can be
traced directly to ineffective HRM practices (Argyris, 1989;
Marcik & Watkins, 1990). Effective staffing policies in
54
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
learning organizations focus on selecting individuals that
are able and willing to learn, have a positive attitude
toward failure, show interpersonal flexibility, and fit well
with the company's values (Ulrich and Lake, 1990). Hedberg
(1981) argues that selection policies should identify
individuals with a high tolerance for ambiguity and a "lust
for experimenting." Many job candidates would find working
under these conditions extremely stressful so it is even
more important to give realistic job previews (Mohrman and
Cummings, 1989).
People basically do what they are rewarded and
appraised for (Kerr, 1988) so reward systems in learning
organizations must promote behaviors that foster learning
(Ulrich and Lake, 1990). Reward systems should tolerate
failures and support risk taking (Hedberg, 1981; Sitkin,
1992), stress skill and performance based pay (Mohrman and
Cummings, 1989; Ulrich and Lake, 1990), be equitable and use
some form of profit sharing (Ulrich and Lake, 1990), be
flexible (Hedberg, 1981) and be frequent and visible
(Cummings, 1989). Traditional reward systems that reward
for mere attendance (e.g. salary), success, seniority, and
hierarchical position tend to retard organizational learning
(Mohrman and Cummings, 1989; Sitkin, 1992; Ulrich and Lake,
1990) .
One challenge in developing reward systems to foster
learning is that the benefits from organizational learning
55
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
develop over a long time span while the costs are immediate
(Pucik, 1988). As most reward systems are based on short
term criteria they often retard organizational learning
(Pucik, 1988). This presents the challenge to learning
organizations in that they need to both implement less
traditional, more innovative rewards and base them on a
longer 'time horizon.
Training and development have obvious implications for
organizational learning. Mazda is typical of the Japanese
auto transplants in that every new employee undergoes 3
weeks of formal training (on such subjects as continuous
improvement, interpersonal skills, and Mazda's culture), 5-7
weeks of on-the-job training, 3-4 more weeks of close
supervision, additional training throughout his or her
career, and job rotations every 6-9 months (Galbraith,
1989). At traditional US. auto plants new workers are
trained for less than an hour (Galbraith, 1989). It should
come as no surprise that Mazda learns better than General
Motors. Training programs that promote learning focus on
increasing skills and knowledge, confidence, and flexibility
(Mohrman and Cummings, 1989; Ulrich and Lake, 1990) .
Pucik (1988) points out that information is often seen
as a source of power by managers and not as a resource to be
shared. Unfortunately, reward systems often implicitly
foster this attitude^ Managers that do not share
information can become seen as experts and are rewarded with
56
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
promotions, raises, or autonomy. As free and open
communication is necessary for organisational learning to be
successful (Argyris and Schon, 1978} care should be taken to
prevent such bottlenecks.
It is beyond the scope of this research to assess all
the myriad of way that various HRM practices can affect
organizational learning. However, it seems clear that an
ineffective HRM system will promote the learning
disabilities (Argyris, 1989; Marcik & Watkins, 1990). Thus,
I hypothesize that:
Hypothesis 7: Ineffective HRM practices will be
positively associated with the organizational
learning disabilities.
Environmental Change
In an Economy where the only certainty is
uncertainty, the one source of competitive
advantage is in knowledge. When markets shift,
technologies proliferate, competitors multiply,
and products become obsolete almost overnight,
successful companies are those that consistently
create new knowledge, disseminate it widely
throughout the organization and embody it quickly
in new technologies and products [Nonaka, 1991].
Just about all organizations are now facing an
environment of unprecedented change and turbulence (Bennis
and Slater, 1969; Drucker, 1993; Peters, 1987). Previously
regulated industries, such as transportation, communication,
and utilities, are suddenly forced to compete in a
competitive environment while manufacturing firms must
57
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
compete against foreign businesses from countries with lower
wages, fewer taxes, longer working hours, and lower costs
(Mohrman and Cummings, 1989. Today's turbulent environments
are characterized by rapid technology cycles, increased
competition, greater customer expectations, both new and
constricting markets, changing demographics, and new laws
and regulations (Mohrman and Cummings, 1989; Schon, 1983:
Ulrich et al., 1993). In such rapidly changing environments
it is critical for the organization to be able to assess the
environment and adapt quickly. Organizations that are
unable to do such learning will not survive (De Geus, 1988;
Stata, 1989, Wick, 1993). So in rapidly changing
environments the organizations that are able to survive will
be the ones with fewest learning disabilities.
Thus, I hypothesize that:
Hypothesis 9: Environments characterized by rapid
change will have a negative association with the
organizational learning disabilities.
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
CHAPTER IV
METHODOLOGY
This chapter describes the methodology used to develop
the organizational learning disabilities presented in
Chapter II and the theoretical framework presented in
Chapter III. Due to the complexity of the concept of
organizational learning and lack of empirical research on
the topic I use a multimethod approach that includes
widespread survey distribution, case studies, and archival
data. The survey data provides breadth and generalizablity
while the results from the case analyses help validate the
questionnaire results and add depth to the analyses. The
sample, procedures, measures, and analyses are described
below, first for the questionnaire data and then for the
case analyses.
5 9
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Survey Research Design
Sample and Procedures
A key informant approach was used collect the data. A
key informant methodology is appropriate when the research
topic "is such that complete or in-depth information cannot
be expected from representative survey respondents" (Kumar,
Stern, & Anderson, 1993). While survey respondents describe
"their personal feelings, opinions, and behaviors," key
informants "generalize about patterns of behavior after
summarizing either observed or expected organizational
relations" (Seidler, 1974). Respondents give information
about themselves as individuals, while informants give
information about organizational properties (Anderson,
1987) . Key informants are chosen because they are
knowledgeable about the research questions (Kumar, et al.,
1974) . Such an approach is particularly appropriate for
"quantitative, large scale, empirical" research when
archival data is not available (Kumar, et al., 1974) and
when the unit of analysis is at the organizational level
(Anderson, 1987).
Five surveys were sent to 2,000 "contacts" (mostly
high level HR executives) using this key informant
methodology. Informants were at firms located throughout
the world, but primarily in North America, Asia, and Europe.
60
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
The names and addresses of these executives were developed
from mailing lists from three sources: the IBEAR program
of the Graduate school of Business at the University of
Southern California, the Executive Education Center at the
University of Michigan, and the International Consortium on
Executive Development Research (ICEDR).
Only relatively senior managers and executives are
likely to have a a' good overall sense of the organizational
characteristics of interest in this study, especially since
questionnaire items covered a diversity of organizational
processes (e.g. organizational scanning, HRM system
effectiveness, financial competitiveness). For this reason
senior managers and department heads were chosen for
informants. The mailing lists used covered most major
international and domestic firms across most major
industries.
Each contact was asked to fill out one questionnaire
and give enclosed copies to four other executives in
different departments at the same organization. Though a
total of 10,000 surveys were sent, I estimate that about 40%
were actually distributed. Approximately 1,532 surveys
were returned from a total of 460 businesses. This
represents a response rate of approximately 38%.
Additional surveys were collected from each of the case
studies, 50 from Samsung and over 200 from 6 health care
organizations, representing a response rate of close to
61
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
100%. Due to the different procedures the surveys from the
cases were analyzed separately and were not combined with
the data from the mailed surveys. The cases will be
discussed in greater detail in a following section.
Characteristics of the data are presented in Appendix
2. North American firms made up 75% of the respondents.
While this was not ideal, there were still almost 400
responses from outside North America. To my knowledge this
is both the largest empirical study of organizational
learning, both in general and with an international sample.
The sample was comprised of primarily large firms with the
average firm size being 40,738. As Table 4.2 in Appendix 2
shows though, the distribution of firm size was relatively
even from under 1,000 to over 100,000. As might be expected
with large organizations, the average age of businesses was
high, 69 years (see table 4.3). Once again there was a
relatively even age distribution from 1 to over 150 years.
An attempt was made to include organizations from a
wide range of major industry types to insure that the
findings are generalizable. Miscellaneous manufacturing
firms made up the single largest group (298 responses).
There was also good representation from chemical/pharmacy,
electronics/computer, utilities, retail, finance, and
securities industries with a range of between 111 and 185
responses. Smaller samples from petroleum, automobiles, and
aircraft (39, 36, and 17 respectively) were returned. A
62
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
complete list of all the organizations that participated is
presented in Appendix 3.
The average age of the informants was 45.7 years, and
the average .tenure with the company was 15.8 years. The
respondents came mainly from middle and upper level
management (21% were managers, 39% were directors of
managers, and 28% were general managers). The age, tenure,
and level in the organization indicated that the informants
should have a solid knowledge of the business and its
history and so fill the role of key informants well.
Finally, 35% of the respondents were from HR/Personnel
departments with the rest fairly evenly distributed across
all other major functional specialties. This was felt to be
the ideal distribution for the survey as the HR managers
would have the greatest overall knowledge about internal
working of the business while managers from other
specialties would add valuable functional specific
information.
Unit of- Analysis
As organizational learning, by definition, is a macro
level phenomenon the appropriate level of analysis was at
the business level. Business was defined as the primary
organizational unit within which the informant works, and
all data was aggregated to the business level. There were
63
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
between 1-37 respondents from each business, but businesses
with less than 3 respondents were not included in the
analyses. This left 268 usable businesses.
While multiple informants should be used to increase
reliability and validity (Bagozzi, Yi, & Philips, 1991,
Seidler, 1974), perceptual agreement between informants
within each business should be shown before items are
aggregated to the organizational level (Anderson, 1987;
James 1982). The appropriateness of the aggregations was
assessed through two methods: inter-rater reliabilities and
one-way ANOVAs. Inter-rater reliabilities were calculated
for each business to assess the internal agreement among
raters. Fifteen business out of 268 had inter-rater
reliabilities of less than .65 and were not included in the
analyses. The remaining sample of 253 had reliabilities
ranging from .65 to .98. One-way ANOVAs were also run on
all remaining respondents to make sure that the within group
variance was significantly smaller than the between group
variance, which was shown to be the case.
Measures
Whenever possible established scales from the
literature were used. In cases where no established
measures existed, new measures were developed based on the
64
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
literature. The individual measures are discussed below and
are presented in Appendix 1.
Organizational Learning Disabilities. Separate
measures were developed for each of the organizational
learning disability scales. I could not find many survey
measures for any aspects of organizational learning2, much
less for the disabilities, so it was necessary to develop
new items for the learning disabilities. I discuss this
development in some detail due to the theoretical centrality
of the learning disability constructs.
The first step was the theoretical development of the
disabilities from the literature, which was described in
Chapter II. Initially 10 to 20 five-point Likert type items
were generated for each disability. Roughly a quarter of
the items were negatively worded to minimize common response
bias.
The first pretest was conducted with a sample of 31
business undergraduates, 12 of which were not US. citizens.
Based on these results items were modified, deleted, or
added. Another set of undergraduates, with similar
characteristics, was used for a second pretest. Refinements
were made and a third pretest was conducted with an
executive MBA class (n=26) . By this time most of the
2I was able to find a few surveys being developed by consultants that
had items relating to organizational learning. Without exception they
were all in the early stages of development, a-theoretical, and paid
little attention to psychometric properties.
65
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
difficulties had been worked out, and the goal was to obtain
as parsimonious a set of items as possible.
A fourth pretest was conducted at the Los Angeles
branch of a large Taiwanese MNC. The sample contained 12
Taiwanese and 13 American managers. After the surveys were
completed I was afforded the opportunity to discuss the
items in detail with the respondents. This proved
invaluable as I was able to identify exactly what the
questions meant to each respondent, where confusion existed,
and what seemed repetitious. Based on this and previous
pretests the survey began to take its current form with 3-4
items for each disability.
A final set of pretests were conducted with the health
care organizations. The head of the HR department at each
of the six health care organizations, 2 consultants, a
fellow researcher, and myself went through the survey item
by item over several days of on site meetings. The last
refining of questions was completed. A few weeks later the
final survey was tested with 27 managers at one of the
hospitals in Ohio. Twelve interviews lasting from 45 to 90
minutes were conducted afterwards to strengthen the
reliability and validity of the constructs. Preliminary
psychometric'statistics were assessed with this admittedly
small sample and all cronbach alphas but one were greater
than .65.
66
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
The final items are included in Appendix 1 and all the
data presented in this dissertation are from these final
items and the case analyses. The combination of the
successive pretests, interviews, and input from managers at
learning organizations, consultants specializing in
organizational learning, and fellow researchers help insure
the psychometric quality of the instrument.
I use a LISREL confirmatory factor analysis on the
final sample; to*' assess construct, convergent and
discriminate validities. Due to the theoretical
centrality 'of the organizational learning disability
measures to the rest of the dissertation, the construct
validation of these measures will be examined in the next
chapter.
Other Measures
Each of the other sets of measures was taken from
established scales in the literature or archival data. The
univariate statistics are presented in Table 4.1 and the
actual items are presented in Appendix 2.
6 7
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
TABLE 4.1
Univarite Statistics
Antecedent Scales
Scale Mean Std Dev Minimum Maximum Alpha
Org. Culture
Burocracy 4.10 .81 2.00 6.13 .73
Clan 4.81 .83 2.38 6.63 .92
Adhocracy 4.78 .77 2.53 6.33 .64
Market 5. 63 .57 3.63 6.83 .81
Leadership
Shared Need 3.43 .51 2.10 4.50 .67
Shape Vision 3.36 .57 1.69 4.50 .89
Lead Change 3.51 .49 2.15 4.63 .88
Commitment 3.06 .48 1.63 4.20 .85
Syst. & Str. 3.32 .46 1.87 4.22 .93
Monitor 3.44 .50 1.92 4.69 .82
Make Last 3.04 .51 1.60 4.13 .88
HRM
Comm 3.04 .53 1.58 4.83 .88
Development 2.98 .52 1.40 4.25 .83
PA 3.12 .58 1.60 4.50 .89
Rewards 3.27 .54 1.33 4.56 .75
Staff 3.32 .41 2.20 4.17 .63
Train 3.04 .54 1.44 4.50 .86
Env. Change
Env. Change 3.49 .34 2.71 4.40 .71
The national culture values are taken from Hofstede
(1991) . The actual numerical values calculated by Hofstede
are used as the measures on the four dimensions of national
culture. Each business unit was given culture scores based
on the country that the business unit (not the corporate
headquarters) was .^located in.
Hofstede's scores were used for several reasons.
While there is some controversy around the validity and
68
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
reliability of his items (e.g. Dorfman and Howell, 1991;
Roberts and Boyacigiller, 1984), they have had their
psychometric properties examined (and supported) far more
extensively than any other cross-cultural study (Hambrick &
Brandon, 1988; Kogut & Singh, 1988). The measures were
developed in the largest cross-cultural study ever
undertaken, both in terms of number of countries and
respondents (Hofstede, 1980; Hoppe, 1990; Shane, 1993)
Finally, Hofstede's measures are the only items with any
widespread acceptance in the international literature
(Shane, 1993). The final advantage of using culture scores
developed in a separate study is that it prevents common
method bias between the independent and dependent variables.
The items used to assess organizational culture come
from the Competing Values Model developed by Quinn and his
colleges (Quinn and Rohbraugh, 1983; Quinn and McGrath
1985; Quinn, 1988). Their psychometric properties have been
assessed and were shown to be strong (Quinn and Spreitzer,
1991) . The scales consist of between three and four Likert
type items (see Appendix 2 for the actual items) , and the
univariate statistics are shown in Table 4.1. Cronbach
alphas ranged from a low of .64 with adhocracy to a high of
.92 with clan. The organizational culture types of
bureaucracy (alpha = .73) and market (alpha = .81) fell
between the extremes.
69
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Given questionnaire length limitations, it was not
feasible to assess all the ways in which HRM practices
affect organizational learning. However, the hypotheses
concerned with HRM posited that, in general, an ineffective
HRM system will promote the organizational learning
disabilities. Items that assess the overall effectiveness
of the major HRM functions were available and succinct
enough to be practical. The items on human resource
management practices were developed by Ulrich and his
colleges (Brockbank, Ulrich, and Yeung, 1989; Ulrich, Yeung,
and Brockbank 1992) and have been used extensively in the
literature. The human resource management scales were
concerned with the resources devoted to HR functions and
their ability to promote change. There were 3 or 4 Likert
type items for each scale— performance appraisal (alpha =
.89), rewards (alpha = .75), communication (alpha = .88),
organizational design (alpha = .83), staffing (alpha = .63),
and development (alpha = .8 6). Univariate statistics for
each of these scales are provided in table 4.1, and the
specific items used to create the scales are presented in
Appendix 2.
The leadership questions were developed by Ulrich at.
al. (1993) to assess an organization's capacity for change.
There are between 4 and 9 Likert type items per scale. The
cronbach alpha's ranged between .82 to .93, except for the
scale "creating a shared need" which had a cronbach alpha of
70
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
.67. Univariate statistics for these scales are displayed
in table 4.1.
The items designed to assess the environment were
developed by Ulrich and his colleges (Brockbank, Ulrich, and
Yeung, 1989; Ulrich, Yeung, and Brockbank 1992) as well.
The items ask about the growth or decline of the market, the
predictability ' ' o f environmental changes, longevity of
customer relationships (4 items) market concentration, and
technological interdependence. The cronbach alpha for this
scale is .71 and the items are provided in Appendix 2.
Outcome Measures
The competitiveness measures were taken from Ulrich et
al. (1993) . They compared the performance of the business
to major competitors on 11 characteristics (e.g. customer
relations, production capacity, research and development,
distribution channels, etc.). Financial competitiveness was
assessed by two questions comparing financial performance to
relevant competitors. While objective performance measures
are always preferred, recent research has shown that
perceived measures can be a reasonable substitute (Dess &
Robertson, 1984; Venkatraman & Ramunujam, 1987). Yeung
(1990) demonstrated that the measures used in this study
have both strong reliabilities and correlations with other
measures of .financial performance. His results showed that,
' * 71
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
at the organizational level, the competitiveness measure had
a strong correlation (.45) with a composite factor of
traditional performance variables (ROA, ROI, and
earnings/share) provided by CompuStat. The competitiveness
scale in this study had a cronbach alpha of .84 and the
univariate statistics are presented in table 4.1.
The other outcome variable, organizational
innovativeness, was also taken from Ulrich et. al. (1993)
and Yeung (1990). It was measured by six items: one item on
"percent of sales based on new products introduced in the
last three years" and five Likert type items asking the
informant about innovativeness compared to the major
competitor. The univariate statistics are shown in table
4.1 and the scale had a cronbach alpha of .86. The actual
items used for both innovativeness and competitiveness are
presented in Appendix 3.
Common Method Bias
A major concern with survey research data is common
method, or "same source" data, bias. In order to minimize
the bias of having the same respondents provide data on both
the dependent and independent variables different sources of
data were used. The data were split such that different
respondents within each organization provided information on
the dependent variables and independent variables. The most
72
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
senior level respondent was used for the dependent variables
as they were most likely to have an accurate assessment of
the outcome variables. If two or more respondents were at
the same level the choice was made by functional specialty:
general management, planning, finance, or marketing (in that
order). The other respondents were aggregated to form the
independent variables. This approach helps to mitigate the
problem of common method bias.
Analyses
Construct Validation of the Organizational Learning
Disabilities
The first of the three central goals in this study was
to develop a rigorous measure of the organizational learning
disabilities. A LISREL confirmatory factor analysis was
conducted to assess the validity and reliability of the 13
disabilities (Bagozzi, Yi, & Philips, 1991; Joreskog &
Sorbom, 1989). LISREL confirmatory factor analysis had
several advantages over exploratory factor analysis; it
provided measures on how well the model fits the data,
provided assessments of validity and reliability, and
accounted for measurement error (James, Mulaik, and Brett,
1982; Joreskog and Sorbom, 1989). Scales for each of
the disabilities were then constructed for use in the
73
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
hypotheses testing. A second order confirmatory factor
analysis was also conducted to insure that the
organizational learning disabilities fit into the four stage
theoretical model as hypothesized (Joreskog & Sorbom, 1989).
The model is presented in Figure 4.1a&b on the following
pages.
Relationship between the Learning Disabilities, Innovation,
and Competitiveness.
The second, and central, objective in this study was to
assess the relationship between the organizational learning
disabilities, organizational innovativeness, and
organizational competitiveness. A LISREL path model was
used to assess these relationships that were hypothesized in
Chapter III and presented in Figure 3.1. Two separate
LISREL path models were run. The first included all 13 of
the organizational learning disabilities. The second model
grouped the 13 disabilities into the four stages of the
learning model (discovery, invention, production, and
generalization). The relationships between each of the four
stages and the endogenous variables were then examined.
This was done to get a more global and concise picture of
the effect of the disabilities. Organization size and age
were included as control variables.
74
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Figure 4.1a
Learning Disabilities Factor Model
Blindness
Int. Blindness
Projection
Simpieminded
Requisite Var.
Schizophrenia
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Figure 4.1b
Learning Disabilities Factor Model
Hyperactive
Alien Hand
Diffusion Deficit
Superstitious L
Amnesia
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Relationship Between Antecedents and the Learning
Disabilities
The hypothesized relationships between the antecedents
and organizational learning were assessed using
correlational analyses and OLS regression analyses.
Separate regression analyses were run for each of the 13
learning • disabilities and their groupings into the four
global stages. Control variables such as firm size and age
were included in the regression equations as well.
Case Studies
This dissertation uses several different research
methods in an attempt to utilize the strengths of both
quantitative and qualitative methodologies. Two case studies
were completed in addition to the survey results. The cases
allow a richer in-depth analysis into the functioning of
organizational learning, and an opportunity to explore
unanticipated findings. These cases also serve as an
additional check on the reliability and validity of the
survey items and help ensure that the surveys are being
interpreted correctly.
7 7
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Samsung Electronics Co.
The first case was of Samsung Electronics Co. in Korea.
This site was chosen for a variety of reasons. First,
despite continued success, they have recently undergone a
radical organization wide restructuring in an attempt to
become a "world class" learning organization. The magnitude
of the change is truly impressive. It involves
decentralizing, empowering, and a complete reengineering of
their plants and assembly lines. Second, Samsung learned
how to learn from the best US and Japanese firms, which
raises intriguing cross cultural and HRM issues. Third,
aspects of the change at Samsung run counter to some of the
hypotheses in this study (e.g. Korean culture is
characterized by a high power distance score yet Samsung
appears to have few of the learning disabilities) . I was
intrigued about how this could occur and whether it was the
exception that proves the rule. Fourth, I was graciously
given access to nearly all levels of the organization
including top level executives and was allowed to observe
the functioning of several plants. In summary, the case
study at Samsung allowed me to explore organizational
learning in depth and from a non-US. system.
The Samsung case study involved interviews,
observations, survey data, and archival data. I spent two
weeks in Korea for this study and had one to four hour long
78
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
semi-structured interviews with approximately 20
individuals. Interviewees included assembly line workers,
plan managers, lower and middle managers, senior managers,
and vice presidents. I spent time in two SEC plants (one
produced microwaves and the other electronics) at Suwon and
at the corporate headquarters in Seoul. In addition to the
interviews and observations nearly 50 of the organizational
learning surveys were completed and returned. Finally,
archival data from company documents, financial statements,
and the literature were collect as well.
Health Care Systems
The second case is on a consortium of six health care
systems which agreed to both partially fund and participate
in the study. Each health care system was made up of
several hospitals and/or HMOs or other forms of insurance.
This was an interesting sample as each of the health care
systems was attempting to implement organizational learning,
but to very different degrees. A few were attempting total
organizational change in the hope of becoming a true
learning organization, while others were limiting the
redesign to a specific department. There was also a wide
range of performance among the various plants with many
considered "world class" and others considered inefficient.
Each of the health care networks had identified a control
79
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
group that is not participating in the changes and these
will be used for comparison purposes. These factors make
this an ideal natural experiment.
The seven health care networks are located in
Minneapolis, Seattle, San Jose, Columbus, Portland, and San
Antonio. Surveys were distributed to each of the
participants. Several additional items were added (e.g.
foster community health as a performance outcome) to the
original survey to make it relevant to the hospitals. All
of the networks have completed and returned the surveys for
a total of about 400. Between two and five days of
interviews were completed at five of the organizations.
Respondents came from all levels of the organization and
several different locations within each hospital network.
Archival data on performance, organizational structures, and
the environment were gathered.
8 0
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
CHAPTER V
RESULTS AND DISCUSSION
In this chapter the construct validation of the
organizational learning disabilities is presented and
discussed. The hypotheses regarding the learning
disabilities, antecedents, and outcomes are then examined.
Construct Validation of the
Organizational Learning Disabilities
Confirmatory factor analyses using LISREL were
conducted to assess the convergent validity, discriminate
validity, and reliability of the organizational learning
disabilities. A second order confirmatory factor analysis
was then run to assess the relationship between the learning
disabilities and the four stage theoretical model.
The univariate statistics for the learning disabilities
are presented in Table 5.1. Each of the scales has a mean
that is- close to 3.0 (the midpoint of the scale). Even
though aggregating data tends to reduce the range and
variance of results, the scales have minimum and maximum
81
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
TABLE 5.1
Univatite Stistics
Organizational Learning Disabilities
Variable Mean Std Dev Minimum Maximum Alpha
Discovery
Ext. Blind 2.94 .48 1.40 4.20 .70
Int. Blind 2.77 .47 1.50 4.12 .69
Comp. Traps 2.48 .54 1.00 4.10 .35
Projection 2.94 .48 1.83 4.22 . 66
Invention
Simpleminded 3.03 .49 2.00 4.33 .78
Req. Variety 2.78 .53 1.32 4.28 .74
Schizophrenia 3.00 .61 1.00 4.62 .78
Production
Paralysis 2.92 .47 1.56 4.44 .68
Hyperactive 2.48 .57 1.17 4.10 .73
Alien Hand 2.76 .53 1.33 4.22 . 63
Generalization
Amnesia 2.96 .52 1.33 4.33 .70
Diffusion 3.08 .57 1.00 4.80 .82
Superstition 2.62 .50 1.33 4.17 .87
scores that cover nearly the complete range of 1 to 5. The
full use of the range of possible values indicates that the
strength of the learning disabilities varies significantly
among organizations. Since the individual items are
aggregated both to scales and to the business level, the use
of nearly the full range of values provides preliminary
evidence that the items are adequately measuring
organizational characteristics.
8 2
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
One serious problem becomes apparent from examining the
scale reliabilities in Table 5.1. The cronbach alpha for
competency traps is only .35, clearly unacceptable. To help
determine if any one item is causing the poor reliability,
the cronbach alphas were run for each pair of items in the
three item scale. There was no improvement when any one
item was dropped. Since any relationships between this
scale and other variables would be uninterpretable,
competency traps is excluded from the hypotheses testing.
One other scale is potentially problematic: alien hand
(alpha = .63). These scales will be examined and discussed
more thoroughly in the confirmatory factor analysis section.
All other cronbach alphas were greater than .65.
Correlations between the 34 learning disability items
are presented in Appendix 4. Nearly all correlations were
significant at the .05 level, but the highest correlations
occurred between items from the same scale. This provides
additional preliminary evidence of scale validities.
The principal assessment of convergent validity,
discriminate validity, and reliability is provided through
the use of confirmatory factor analyses.
Confirmatory Factor Analyses of the Learning Disabilities
LISREL confirmatory factor analysis was conducted to
assess the validity and reliability of the 13 disabilities
83
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
(Bagozzi, Yi, & Philips, 1991; Joreskog & Sorbom, 1989).
LISREL confirmatory factor analysis has several advantages
over exploratory factor analysis; it measures how well the
model fits the data, assesses validities and reliability,
and accounts for measurement error (James, Mulaik, and
Brett, 1982; Joreskog and Sorbom, 1989).
Two issues need to be considered when dealing with
LISREL measurement models that have many variables (in this
case 34); complexity and sample size. LISREL models with
many free and _fixed parameters often have both poor overall
fit statistics and ambiguous clues for the reasons behind
the poor fit (Herting & Costner, 1985) . Even if a complex
model is properly specified, the complexity itself can
result in convoluted matrix equations that are unsolvable
(Herting & Costner, 1985; Joreskog and Sorbom, 1989; Niehoff
& Moorman, 1993; Sullivan, 1985) . A solution to this
problem and method for getting less biased results is to
simplify the model into theoretically appropriate portions
(Hayduk, 1989; Herting & Costner, 1985) . Once the core
portions of the model are adequately specified, they can be
combined into a more complex "gestalt."
I take this approach by first presenting the
confirmatory factor analyses for each of the four stages of
the learning model separately. For example, I examine the
four "discovery” learning disabilities (external blindness,
internal blindness, competency traps, and projection) alone.
84
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Three additional confirmatory factor analyses are run for
the disabilities in the stages of invention, production, and
generalization. I then present the results for the factor
analyses of all 13 disabilities together. The final model,
shown in figure 4.1, is a second order factor analysis that
specifies the items to the disabilities and then the
disabilities into one of the four learning stages.
The second issue mentioned above involves sample size.
The traditional rule of thumb is that the sample size should
be at least 5 times the number of free parameters in the
factor model, and this condition is satisfied in all but the
full secondary factor analysis model. The 5:1 criteria,
however, assumes an individual level of analyses and not an
ecological (organizational) level. Relatively small sample
sizes are prone to unstable factor structures as
fluctuations in single items (whether due to errors or
randomness) have a disproportionate influence on the overall
results (Johnson & Wichern, 1988). Hofstede et. al. (1990)
explains that the necessity of having a large ratio of
subjects to variables...
does not apply to factor analyses of ecological
data, in which each case is based on the mean of a
number of individual scores; such means are
extremely stable. The stability of the factor
structure for ecological matrices does not depend
on the number of aggregate cases but on the number
of independent individuals who contributed to each
case [1990: 299]".
85
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
According to this argument the stability of the factor in
this study is not based on the sample size of 253
organizations but on 1,532 informants. While a strict
adherence to this statement may be a little excessive, the
fact that the analysis is based on aggregated date does
provide additional confidence in the validity of the
results.
All statistics reported in this research are based on a
sample size of 253. Hayduk (1989: 168) argues that even
when the actual sample size is much larger, it is often
advisable to base the LISREL output on an N of about 200.
This is because with large samples even minute differences
are statistically significant but have little practical
significance. A sample of approximately 200 provides the
best balance between statistical and practical significance.
All first order factor analyses underwent an oblique
rotation, because, though the disabilities are hypothesized
to be distinct, I am not claiming that there is no relation
between them. Each model is assessed through various global
fit statistics, error terms, and individual factor loadings.
Confirmatory Factor Analyses of Each Stage.
The confirmatory factor analyses for each of the stages
of the learning model are depicted in Figures 5.1a-d, and
the results are presented in Tables 5.2a-d.
• 86
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Figure 5.1a
Discovery Disabilities Factor Model
x.
‘ 1
10*
ir
x.
‘ 10
x.
‘ 11
Blindness
Int. Blindness
Comp. Traps
Projection
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission
Figure 5.1b
Invention Disabilities Factor Model
x.
i
x
4
X,
5
Simpleminded.
Requisite Var.
Schizophrenia
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Figure 5.1c
Production Disabilities Factor Model
x.
‘ 1
x
4
Paralysis
Hyperactive
Alien Hand
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Figure 5.1 d
Generalization Disabilities Factor Model
.5“
X
4
X,
5
Diffusion Deficit
Superstitious L.
Almnesia
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Table 5.2a
LISKEL Analysis for Discovery Disabilities
Organizational Learning Disabilities LISREL Cronbach
Loadings Alpha
Ext Blindness . 7 0
scanning 1.00
find new markets .71
perceive env. threats, opportunities .85
Int. Blindness .69
scanning internal processes 1.00
. discover info about job 1.22
Comp. Traps .35
little need for change .23
"if it ain't broke, don't fix it" 1.00
create small false alarms .01
Pro j ecti on .66
politics influence actions .64
info must be politically acceptable .92
employees directly exposed to env. 1.00
PHI Matrix
BLI INT COMP PRJ
BLI .621
INT .318 .437
COMP .078 .078 .099
PRJ .242 .272 .061 .221
CHI-SQUARE WITH 38 DEGREES OF FREEDOM = 60.70 (P= .01)
GOODNESS OF FIT INDEX = .96
ADJUSTED GOODNESS OF FIT INDEX = .93
ROOT MEAN' SQUARE RESIDUAL = .047
91
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Table 5.2b
LISREL Analysis for Inventionion Disabilities
Organizational Learning Disabilities LISREL Cronbach
Loadings Alpha
Simplemindedness .74
determin causes of performance 1.00
adressing performance shortfalls 1.15
wide variety of solutions generated 1.05
Requesite Variety .78
everyone has more info than required 1.00
diversity in ideas and people .91
failures punished .92
Schizophrenia .74
coordination between mult, perspectives 1.00
depts. function independtly 1.00
PHI Matrix
_________ SIMP________REQ_______SCH
SIMP .557
REQ .436 .605
SCH .351 .431 .643
CHI-SQUARE WITH 17 DEGREES OF FREEDOM = 50.95 (P = .00)
GOODNESS OF FIT INDEX = .96
ADJUSTED GOODNESS OF FIT INDEX = .91
ROOT MEAN SQUARE RESIDUAL = .043
92
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Table 5.2c
LISREL Analysis for Production Disabilities
Organizational Learning Disabilities LISREL Cronbach
Loadings Alpha
Paralysis .68
over analyze before implimenting 1.00
employees participate in own work .55
trouble implementing procedures 1.00
Hyperactivity .7 3
implement w/o^ thought 1.00
procedures come and go w/o change 1.10
Alien Hand ' ' -63
actions not consistent with goals 1.00
difficult to change direction .67
same mistakes .97
PHI Matrix
__________ PAR_______ HYP_______ ALI
PAR .483
HYP .393 .528
ALI .501 .501 .609
CHI-SQUARE WITH 17 DEGREES OF FREEDOM = 35.62 (P = .01)
GOODNESS OF FIT INDEX = .97
ADJUSTED GOODNESS OF FIT INDEX = .93
ROOT MEAN SQUARE RESIDUAL = .039
93
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Table 5.2d
LISREL Analysis for Generalization Disabilities
Organizational Learning Disabilities LISREL Cronbach
Loadings Alpha
Amnesia .70
loose ideas if not in regular use 1.00
documenting & encoding .91
Superstitious L. .82
rationale for performance 1.00
actions not affect to performance .92
feedback .55
Diffusion Deficit .87
learning spread throughout organization 1.00
learning not transmitted upwards 1.14
PHI Matrix
__________ AMN_______ SUP_______DIF
AMN .401
SUP .490 .748
DIF .444 .537 .682
CHI-SQUARE WITH 11 DEGREES OF FREEDOM = 15.61 (P = .16)
GOODNESS OF FIT INDEX = .98
ADJUSTED GOODNESS OF FIT INDEX = .96
ROOT MEAN SQUARE RESIDUAL = .036
94
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Global Measures of Fit. The goodness of fit index
(GFI) and adjusted goodness of fit index (AGFI) assess how
well the hypothesized model fits the actual data. A
rigorous rule-of-thumb is that the GFI and AGFI should be
close together and greater than the 0.9, though several
researchers (e.g. Gudykunst, Yang, & Nishida, 1987; Gupta,
Dirsmith, & Fogarty, 1994) suggest that .8 is more
practical. Each of the four factor models has GFI and AGFI
indexes- of greater than .9. The AGFI indexes for each model
are: discovery = .93, invention = .91, production = .93, and
generalization = .96.
The root mean square residual (RMSR) is another measure
of overall fit. It is a global measure of the magnitude of
the residuals, and values below .05 indicate that there is
not much unexplained variance in the model. The values for
each of the models was less than .05, once again indicating
a good fit. Q-Q plots provide assessments of whether the
residuals are normally distributed and vary in magnitude
more than expected. The Q-Q plots did not indicate any
significant problems.
The chi-square goodness of fit statistic is a third
measure of the overall fit of the model. Chi-squares that
have a significance of greater than .05 are considered not
to be significantly different than the actual data (note
that here we are looking for non-significance). However,
the chi-square statistic has several problems. At both
95
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
large and small sample sizes it is inaccurate (Bollen, 1990;
McDonald & Marsh, 1990), and it is extremely sensitive to
minor variations in multivariate normality (Bentler, 198 9;
Hayduk, 1989). Jaros, Jermier, Koehler, & Sincich (1994)
argue that given the problems with the chi-square test, .01
is a more appropriate rule-of-thumb for assessing difference
than .05. While three of the models here had chi-squares
that were significant at greater than .01, only the factor
analysis for generalization was greater than .05.
One way of correcting for sample size bias in the chi-
square test is to calculate the ratio of the chi-
square/degrees of freedom (Joreskog & Sorbom, 1989). Models
are considered to fit well when this ratio is less than 5/1
(Hayduk, 1989; Wheaton, Muthen, Alwin, & Summers, 1977).
The ratio in the four models are the following: discovery =
1.60/1, invention = 2.99/1, production = 2.16/1, and
generalization = 1.42/1. Each of these ratios is well under
the 5/1 criteria, indicating an acceptable fit.
Unfortunately the chi-square/df ratio test is subject to
many of the same problems as examining the chi-square alone.
Comparative Fit Index. The comparative fit index (CFI)
addresses the problems inherent in the chi-square test,
normed fit index, and nonnormed fit index (Bentler, 1990;
McDonald • & Marsh, 1990). It provides a more rigorous
assessment of overall model fit and remains unbiased in both
96
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
large and small samples (Bentler, 1990). The CFI assesses
the goodness of fit in relation to the fit of a more
restrictive model (Bentler, 1990).
In other words, the CFI compares the hypothesized model to a
reasonable alternative model and tells us whether the
hypothesized model is superior.
The choice of the "reasonable alternative" is important
as any model can look good if compared to a poor
alternative. A common choice for a baseline model is the
absolute null model that hypothesizes no relationships
between the items and the factors (Bentler & Bonnet, 1980;
Jeros et. al., 1994). A much more rigorous test is to
compare the hypothesized model to a plausible alternative
(Bentler, 1990; Bentler and Bonnet, 1980) . A CFI score
greater than .9 indicates that the hypothesized model is an
improvement over the alternative. The only difference
between the alternative models for each of the four factor
analyses and the hypothesized model is that the factor
rotations for the alternative models are orthogonal instead
of oblique. The results are presented in Table 5.3.
9 7
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Table 5.3
Comparative Fit Index for
Four Sets of Confirmatory Factor Analyses
Chi Square____ df CFI
Discovery 61 38
Null 734 51 .97
Orthogonal 359 36 .93
Invention 51 17
Null 2476 33 .99
Orthogonal 350 20 .90
Production 36 17
Null 700 25 .97
Orthogonal 336 20 .94
Generalization 16 11
Null 775 18 .99
Orthogonal 295 14 .98
the CFI are above .9, indicating a good fit,
is true even when compared to the strong alternative model
(the orthogonal confirmatory factor analysis). The fact
that the hypothesized model is a significant improvement
over another reasonable model, as well as the null, is a
strong indication of the overall fit of the models.
Summary of Global Fit Statistics. Taken as a whole,
the various methods of assessing the fit of the overall
model to the data (GFI, AGFI, Chi-square/df, and CFI)
indicate that the four factor analyses are strong models.
Assessment of the residuals (RMSR, Q-Q plots) also indicates
a good fit. However, a strong overall model does not mean
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
that every individual parameter is correctly specified. The
fit of the individual items is considered next.
Assessment of Internal Fit
In this section I consider the loading of each item on
its hypothesized factor. Following the procedure advocated
by Hayduk (1989), one item loading for each factor was fixed
at 1.0. This does not affect the measurement quality but
provides scale equivalence (Hayduk, 1989; Herding, 1985).
Generally, the most theoretically robust item is set to 1.0,
and the closer the remaining loadings are to 1.0 the
stronger the factor. With only a few exceptions the
loadings are close to 1.0 and were all significant at the
.05 level (see Tables 5.2a-d).
A conspicuous misspecification occurs with the
disability of "competency traps” (see Table 5.2a). One of
the loadings is .01 while the other is .23. This is clearly
unacceptable. The modification indices indicated that none
of the poorly fitting items cross loaded on any of the other
factors. The Q-Q plot indicates that the residuals are
normally distributed while the standardized residuals show
that the residuals from poor loadings are well within two
standard deviations from the norm. None of the diagnostics
provide much help in spotting a potential source of the
problem (e.g. cross loading items, mismodeled covariances,
99
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
structural disturbances, omitted reciprocal effects, coding
errors, etc.). The poor loadings seem to be a result of the
items simply not having a direct affect on the competency
trap factor. For this reason, as well as the low cronbach
alpha, the competency trap learning disability was excluded
from the remaining analysis.
Two other items have weak lambda loadings; employee
participation -(.55) in Paralysis and "feedback" (.55) in
Superstitious learning. While neither one is weak enough to
be' dismissed outright, they need to be considered when
interpreting the results. The other diagnostics for these
two items appear acceptable.
With only one major exception, none of the items cross
loaded significantly on any of the other factors. Even in
cases where the factors were theoretically related (e.g.
blindness and projection) hypothesized items did not cross
load. This indicates good discriminate validity.
The exception mentioned above involves two items from
the disability of simplemindedness which also load on
requisite variety. The two cross loading items are
"generating a variety of solutions" and "addressing
performance shortfalls." While these items do significantly
load on both factors, their magnitude is small. When the
factor analysis was rerun allowing the cross loading, the
overall fit of the model did not significantly improve. Yet
the fact that the simplemindedness items do not cleanly load
100
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
only on simplemindedness does restrict its discriminate
validity.
Summary of Internal Fit Statistics. The assessment of
the internal fit statistics (lambda loadings, modification
indices, and standardized residuals) indicates that the
items do accurately measure characteristics of the
corresponding factor with relatively little error. I will
now present the results of the full factor analysis that
contains all 12 organizational learning disabilities.
Confirmatory Factor Analyses of the Full Model.
The full factor analysis model for the organizational
learning disabilities is depicted in Figure 4.1. This model
is a combination of the four previously discussed models of
discovery, invention, production, and generalization. Its
characteristics are discussed below.
Global Assessment of Fit. The global assessments of
fit for the full confirmatory factor analysis are presented
in Table 5.4.
1 0 1
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Table 5.4
Global Assessments of Fit for
Full Confirmatory Factor Analyses
CHI-SQUARE WITH 220 DEGREES OF FREEDOM = 343.70 (P = .00)
GOODNESS OF FIT INDEX = .91
ADJUSTED GOODNESS OF FIT INDEX = .86
ROOT MEAN SQUARE RESIDUAL = .041
COMPARATIVE FIT INDEX
________________Chi Square_____df_______CFI
Full Model 344 222
Null 3408 275 .96
Orthogonal 2048 289 .93
As expected, the added complexity of the model has caused a
slight decrease in the overall fit. The GFI (.91) and AGFI
(.86) exceed or approach the traditional rule-of-thumb for a
good fit. The chi-square/df. ratio is 1.56/1 and the RMSR
is less than .05. The Q-Q plot indicates that the residuals
are evenly distributed with few excessive values. Finally,
the CFI with both the null (.96) and orthogonal (.93)
competing models are well above .9. Overall, these
statistics indicate that the full model, despite its
complexity, fits the data rather well.
Assessment of Internal Fit. Statistics indicating the
fit of individual parameters in the model are shown in table
5.5.
102
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Table 5.5
LISREL Confirmatory Factor Analysis
for the Organizational Learning Disabilities
Organizational Learning Disabilities LISREL Cronbach
Loadings Alpha
DISCOVERY
Ext Blindness
scanning
find new markets
perceive env. threats, opportunities
1.00
.75
.87
.70
Int. Blindness
scanning internal processes
discover info about job
1.00
1.23
.69
Projection
politics influence actions
info must be politically acceptable
employees directly exposed to env.
.79
.96
1.00
.66
INVENTION
Requisite Variety
everyone has more info than required
diversity in ideas and people
failures punished
1.00
1.01
1.03
.78
Schizophrenia
coordination between mult, perspectives
depts. function independently
1.00
.92
.74
PRODUCTION
Paralysis
over analyze before implementing
employees participate in own work
trouble implementing procedures
1.00
.57
.99
.68
Hyperactivity
implement w/o thought
procedures come and go w/o change
1.00
1.11
.73
Alien Hand
actions not consistent with goals
difficult to change direction
same mistakes
1.00
.69
.99
.63
—•
103
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Table 5.5 (continued)
LISREL Confirmatory Factor Analysis
for the Organizational Learning Disabilities
Organizational Learning Disabilities LISREL Cronbach
Loadings Alpha
GENERALIZATION
Amnesia
loose ideas if not in regular use
documenting & encoding
1.00
.91
.70
Superstitious L.
rationale for performance
actions not affect to performance
feedback
1.00
.97
.56
.82
Diffusion Deficit
learning spread throughout organization
learning not transmitted upwards
1.00
1.06
.87
PHI Matrix
PAR HYP ALI AMN SUP DIFF
PAR .496
HYP .398 .525
ALI .487 .508 .566
AMN .398 .402 .484 .435
SUP .483 .470 .565 .481 .724
DIFF .463 .404 .514 .480 .535 .731
BLI .248 .252 .365 .324 .321 .339
INT .298 .299 .362 .346 .370 .457
PROJ .375 .336 .403 .354 .466 .478
REQ .347 .273 .374 .351 .425 .427
SCH .465 .402 .538 .407 .473 .547
BLI INT PROJ REQ SCH
BLI .592
INT .312 .434
PROJ .351 .404 .477
REQ .361 -.434 .458 .474
SCH .250 .402 .429 .409 .698
1 0 4
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
All factor loadings are close to 1.00 and are significant at
the .05 level. The non-indexed factor loadings range from
.55 to .91, indicating that the measures are strong.
Out of over 600 possible cross loadings, the
modification indices and partial derivatives indicate that
there are only 5 nonhypothesized loadings which show
significance. This is well within what would be expected
due to random chance, and the significant cross loadings
were only marginally significant. The fact that nearly all
items loaded only on their hypothesized factor, especially
in such a complex model, provides strong support for the
discriminat validity of the learning disabilities.
The standardized residuals indicate no significant
problems with outliers. None of the residuals were
extremely large, and only 3.2 percent exceeded two in
absolute value. In a random distribution 5 percent of
residuals are expected to exceed an absolute value of two.
The small number of large residuals indicate both that the
items fit their factor well, and that the model as a whole
fits well.
Summary for Full Model. Virtually every global and
internal goodness of fit measure exceeds or approaches the
appropriate standards. Measurement error does not appear to
be a significant problem. The results indicate both
construct and discriminate validity.
105
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Second Order Confirmatory Factor Analysis
As described in chapter II, the organizational
learning disabilities are hypothesized to group according
the four stages of the organizational learning model and are
depicted in Figure 5.2 on the following pages. In the
second order factor analysis, the first order factors have
the same parameters as in the primary factor analysis,
except ■ that the covariances among the factors are
hypothesized to be explained by the presence of four second
order factors. That is, the covariance among the learning
disabilities is hypothesized to be explained by the four
higher order factors of Discovery, Invention, Production,
and Generalization. Thus, the first rotation is no longer
oblique, but orthogonal. The second order factors are
hypothesized to be independent, so an orthogonal rotation
was used for them as well.
Global Assessment of Fit. Any second order factor
analysis is constrained by the fit of the first order
components. So even if the second order components
perfectly explain the covariance among the first order
factors, the fit will never be greater than that of the
first order (Burke, Borucki, & Hurley, 1992; Marsh &
Hocevar, 19.85) . It is expected that the goodness of fit
106
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
10
1 1
12
13
14
15
16
17
18
19
y i
Figure 5.2a
S eco n d a ry Factor Model
'10
'11
14
y .
B lindness
Int B lindness
D iscovery
C om p. Traps
Projection
Sim plem inded.
Invention R equisite Var.
Schizophrenia
107
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Figure 5.2b
S econ d ary Factor Model
20
2 1
22
23
24
25
26
26
27
28
29
30
31
32
33
34
20
22
24
7 26
7 26
28
30
32
7 33
7 34
Paralysis
H yperactive Production
* 1 10
Alien Hand
Diffusion Deficit
T1 12
Superstitious L. G eneralization
13
A m nesia
108
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
statistics presented in Table 5.5 should slightly decline in
the secondary analysis, but only small reductions indicate a
good fit.
Marsh (1987; Burke et. al., 1992) have developed a more
rigorous measure, the target coefficient (TC), to assess the
fit of the secondary factor model. The TC assesses the
ability -of a second order factor model to capture the
covariation in the first order factors. Burke et. al.
(1992) define the target coefficient as (FU-T)/(FU-F), where
FU = ..chi-square of the first factor null model, T is chi-
square for the second order factor model, and F is the chi-
square for the first order factor model. Values range from
0-1, and a score greater than .9 is considered to indicate a
well fitting model.
The global statistics for the secondary factor analysis
are presented in Table 5.6.
Table 5.6
Global Assessments of Fit for
Secondary Factor Analyses
CHI-SQUARE WITH 258 DEGREES OF FREEDOM = 457.48 (P = .00)
GOODNESS OF FIT INDEX = .88
ADJUSTED GOODNESS OF FIT INDEX = .85
ROOT MEAN SQUARE RESIDUAL = .05
TARGET COEFFICIENT = .96
COMPARATIVE FIT INDEX
' ______ Chi Square____ df_______ CFI
Full Model • - 457 258
Null * 3478 283 .94
109
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
All the global statistics meet or approach the accepted
criteria for a model that fits the data well, and there is
only a slight reduction in the statistics from the first
order mode. In addition, the TC = .96, indicating that the
secondary model explains the covariance in the first order
model. All this indicates that at a global level, the
second order model performs very well. I now turn to the
internal statistics.
Internal Assessment of the Second Order Model. The
internal statistics for the LISREL second order confirmatory
factor model are presented in Table 5.7 on the following
pages. The LISREL loadings for the first order portion of
the model do not change significantly in this model. They
are all close to 1.00 and significant at the .05 level. As
the other statistics for the first order portion of the
model do not differ substantially from the previous model
(Table 5.4 and 5.5) I will limit the discussion to the
second order characteristics.
The loadings for the secondary portion (gamma values)
are not referenced to one, so should be evaluated along
traditional lines. Only one of the loadings is below .6
(external blindness = .47), and even .47 is relatively high.
All loadings are significant at the .05 level and residuals
are within excepted parameters.
110
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Table 5.7
LISREL Second Order Factor Analysis
for the Organizational Learning Disabilities
Organizational Learning Disabilities LISREL Cronbach
Loadings Alpha
DISCOVERY
Ext Blindness
scanning
find new markets
perceive env. threats, opportunities
1.00
.71
.83
.70
Int. Blindness
scanning internal processes
discover info about job
1.00
1.29
.69
Projection
politics influence actions
info must be politically acceptable
employees directly exposed to env.
.79
. 98
1.00
. 66
INVENTION
Requisite Variety
everyone has more info than required
diversity in ideas and people
failures punished
1.00
.88
.91
.78
Schizophrenia
coordination between mult, perspectives
depts. function independently
1.00
.97
.74
PRODUCTION
Paralysis
over analyze before implementing
employees participate in own work
trouble implementing procedures
1.00
.53
1.03
.68
Hyperactivity
implement w/o thought
procedures come and go w/o change
1.00
1.23
.73
Alien Hand
actions not consistent with goals
difficult to change direction
same mistakes
1.00
.64
1.00
.63
111
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Table 5.7 (continued)
LISREL Confirmatory Factor Analysis
for the Organizational Learning Disabilities
Organizational Learning Disabilities LISREL Cronbach
Loadings Alpha
GENERALIZATION
Amnesia .7 0
loose ideas if not in regular use 1.00
documenting & encoding .81
Superstitious L. .82
rationale for performance 1.00
actions not affect to performance .94
feedback .55
Diffusion Deficit .87
learning spread throughout organization 1.00
learning not transmitted upwards 1.08
SECOND ORDER FACTORS LISREL
Loadings
(Gamma)
Discovery
External Blindness .47
Internal Blindness .61
Projection .66
Invention
Requisite Variety .71
Schizophrenia .62
Production
Paralysis .63
Hyperactivity .63
Alien Hand .78
Generalization
Amnesia .65
Superstitious Learning .75
Diffusion Deficit .73
1 1 2
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
For the most part, none of the learning disabilities
cross load on any stage of the learning model, except their
own. The only exception involves requisite variety and
schizophrenia, which load significantly on both the stages
of production and generalization. While the loadings are
significant, they are small and the overall fit of the model
is not improved when these pathways are freed.
The LISREL diagnostics do not provide an obvious
interpretation for these crossloadings. However, there is a
plausible theoretical explanation, involving the central
role that information variety plays in organizational
learning. Requisite variety refers to the amount of
information available in the organization and schizophrenia
refers to the coordination of diverse perspectives. While a
variety of information and perspectives is important for
generating solutions, Invention, it can also affect the
implementation of solutions, Production, and the overall
assessment of the results, Generalization, (Huber and Daft,
1985; Daft, Bettenhausen, & Tyler, 1993). For example, an
insufficient variety of information considered in decision
making would lead to poor decisions. Poor decisions would
be difficult to implement, and it would often be unclear as
to why the implementation failed. This scenario suggests
that requisite variety and schizophrenia have indirect
effects on "'the learning stages of Production and
Generalization. ' This possibility will be more fully
113
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
explored in the analysis of the disabilities' effect on the
organizational outcomes.
Summary of Second Order Model. Overall, the secondary
confirmatory factor analysis worked very well, especially
given its complexity and large number of variables. This
suggests that the four stage theoretical model of discovery,
invention, production, and generalization is an appropriate
framework for considering the learning disabilities.
Summary of Construct Validity
This section has confirmed the construct validity of
the organizational learning disabilities to a substantial
degree. The convergent and discriminat validity of the
learning disabilities was shown through several LISREL
confirmatory factor analyses. In fact, given the large
number of variables, the model worked surprisingly well. A
second order factor analysis provided evidence that the four
stage model of organizational learning is an appropriate
framework for the learning disabilities. The most
significant problem with the measurement models involved the
disability of . competency traps, which was so problematic
that it was excluded from the remainder of the analysis.
Other local problems in the model were discussed, and none
were seen as being substantial.
114
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Scale Development
There are few measures of any aspect of organizational
learning in the literature, so the development of a valid
and reliable instrument to measure the learning disabilities
is an important contribution. The development of these
scales was also necessary for testing hypotheses between the
learning disabilities and their outcomes and antecedents.
The univariate statistics for the scales are provided in
Table 5.1 and the correlations in Table 5.8. The
correlations range from .22 to .68, and indicate that the
learning disabilities are distinct yet related.
Table 5.8
L e a r n i n g D i s a b i l i t i e s S e a l e C o r r e l a t i o n s
ALI AMN BLIND COMP DIF HYP
ALI 1.0000 .5462** .4408** .7508** .6210** .6350**
AMN .5462** 1.0000 .3725** .5499** .5926** .5262**
BLIND .4408** .3725** 1.0000 .4551** .3818** .2948**
COMP .7508** .5499** .4551** 1.0000 .5856** .5661**
DIF .6210** .5926** .3818** .5856** 1.0000 .5210**
HYP .6350** .5262** .2948** .5661** .5210** 1.0000
INTER .5316** .4664** .4249** .5320** .6201** .4342**
PAR .6491** .4486** .3200** .6343** .6113** .5232**
PROJ .5682** .4008** .4122** .5745** .6054** .4924**
REQV .5460** .4364** .4204** .5741** .5744** .3740**
SCHZ .6221** .4825** .2469** .5532** .6283** .5022**
SIMP .5466** .4507** .4921** .5160** .5873** .4516**
SUP .6469** .5917** .3949** .6483** .6524** .5878**
1 1 5
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
T a b l e 5 . 8 ( c o n t i n u e d )
L e a r n i n g D i s a b i l i t i e s S c a l e C o r r e l a t i o n s
INTER PAR PROJ REQV SCHZ SIMP
ALI .5316** .6491** .5682** .5460** .6221** .5466**
AMN .4664** .4486** .4008** .4364** .4825** .4507**
BLIND .4249** .3200** .4122** .4204** .2469** .4921**
COMP .5320** .6343** .5745** .5741** .5532** .5160**
DIF .6201** .6113** .6054** .5744** .6283** .5873**
HYP .4342** .5232** .4924** .3740** .5022** .4516**
INTER 1.0000 .5516** .5942** .7155** .5265** .5592**
PAR .5516** 1.0000 .6419** .6465** .5936** .5661**
PROJ .5942** .6419** 1.0000 .6829** .5568** .6220**
REQV .7155** .6465** .6829** 1.0000 .5368** .6108**
SCHZ .5265** .5936** .5568** .5368** 1.0000 .4585**
SIMP .5592** .5661** .6220** .6108** .4585** 1.0000
SUP .5936** .6496** .6464** .6260** .5425** .6613**
* - Signif. LE .05 ** - Signif. LE .01 (2-tailed)
The hypotheses between the organizational learning
disabilities and their outcomes are now considered.
Relationships Between Organizational Outcomes
and the Organizational Learning Disabilities
Chapter III developed a set of hypotheses relating the
learning disabilities to organizational innovativeness and
competitiveness. The learning disabilities are hypothesized
to directly reduce organizational innovativeness while
innovativeness affects organizational competitiveness. The
learning disabilities (with the exception of superstitious
learning) were not hypothesized to directly affect
competitiveness. Thus innovativeness is a mediating
variable between the learning disabilities and
116
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
competitiveness. These hypotheses are diagramatically
represented in Figures 3.2 and 3.3. The outcome variables,
perceived innovativeness and perceived competitiveness were
discussed in Chapter IV. In order to reduce common method
bias, the data were split such that different informants
within each organization provided information on the
dependent variables and independent variables. The
hypotheses are explored in this section through the use of a
LISREL model.
The results from the LISREL path analysis are presented
in Table 5.9 on the following page. Because of potential
confounding effects, organizational size was included as a
control variable.
The global LISREL statistics indicate that the
hypothesized model fits the actual data well. All
statistics meet or approach traditional acceptance criteria.
The GFI and AGFI are .99 and .88 respectively. The AGFI is
a little low but, given the number of variables in the model
and the split sample, it is acceptable. The chi-square/df.
ratio (1.99/1) and CFI (.91) both indicate acceptable fits.
The overall magnitude of the residuals is low (RMSR =.01),
again suggesting a good fit. The residuals are normally
distributed and none of the individual residuals are large.
While the overall model appears adequate, several of
the individual learning disabilities did not show
significant results. Out of the sixteen hypothesized
117
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Table 5.9
LISREL Path Results:
Learning Disabilities with Outcomes
Exogenous Variables Innovative Competitive
Organizational Size -.20* .14*
Discovery
External Blindness -.21*
Internal Blindness -.02
Projection -. 04
Invention
Requisite Variety -.24*
Schizophrenia .19*
Simplemindedness .07
Production
Paralysis -.18*
Hyperactivi ty -.04
Alien Hand -.01
Generalization
Amnesia -.05
Diffusion Deficit -.14*
Superstitious L. -.18* -.19*
Innovativenes s .85*
Values shown are LISREL maximum likelihood estimates
* = significant at < .05
CHI-SQUARE WITH 10 DEGREES OF FREEDOM =19.85 (P = .031)
GOODNESS OF FIT INDEX = .990
ADJUSTED GOODNESS OF FIT INDEX = .880
ROOT MEAN SQUARE RESIDUAL = .010
COMPARATIVE FIT INDEX (Null) = .91
TOTAL COEFFICIENT OF DETERMINATION = .348
118
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
pathways, nine were significant, six were not significant,
and one was significant with the opposite sign than
predicted. As expected, external blindness, insufficient
requisite variety, paralysis, diffusion deficit, and
superstitious learning all reduce organizational
innovativeness and, through innovativeness, organizational
performance. Innovativeness acts as a mediating variable
for these disabilities and positively promotes
competitiveness.
Disabilities that failed to show significance on the
hypothesized pathways include internal blindness,
projection, simplemindedness, hyperactivity, and alien hand.
Organizational size, paradoxically, appeared to reduce
innovativeness while promoting competitiveness.
Schizophrenia was significant, but in the opposite
direction than predicted. Instead of schizophrenia reducing
innovativeness it appears to promote perceived
innovativeness. I discuss this finding in some detail in
the following section.
The modification indices indicate that none of the non-
hypothesized pathways would have been significant if they
had been hypothesized. In other words, none of the learning
disabilities have a direct effect on competitiveness (except
for superstitious learning which was hypothesized to affect
both outcomes).
119
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Organization innovativeness is a strong predictor of
organizational competitiveness. However, caution needs to
be exercised when interpreting this result. Though the
sample was split such that different informants responded to
the disability items and the outcome items, the scores for
the two outcomes of innovativeness and competitiveness were
taken from the same informants. Thus common method bias is
likely to have inflated the relationship between these two
variables. Still, the beta is so strong (.85, p < .01) that
it is unlikely to be primarily the result of common method
bias. In order to test this proposition the model was rerun
with different informants providing the competitiveness
items than the rest of the sample- The results from this
model indicate that innovativeness is still a powerful
predictor of competitiveness (beta = .59. p < .01).
Discussion of the Relationships between the Organizational
Learning Disabilities, Innovativeness, and Performance
A summary of the results are diagramatically
represented in Figure 5.3 on the following page. As
predicted, the presence of most learning disabilities does
reduce an organization's ability to innovate, and innovation
does lead to competitiveness. However, there are several
results that were unexpected.
120
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Figure 5.3
Learning Disabilities Results
Org. Size
External
Blindness
Internal
Blindness
+.14
- . 2 0
Projection
-.21
Requisite
Variety
-.24
Simpleminded
Schizophrenia +.19
+.85
►"^Competitiveness Innovativeness
Paralysis
Hyperactivity
Alien Hand -.14
-.18
-.19
Amnesia
Diffusion
Superstitious L
1 2 1
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Failure of Simplemindedness to Predict Innovativeness.
One of the more provocative findings involves the fact that
simplemindedness does not reduce organizational
innovativeness. Many researchers view the ability to
analyze and develop solutions to organizational problems as
the most important component of organizational learning
(e.g. Daft & Huber, 1987; Fiol & Lyles, 1985; Huber, 1991;
Senge, 1990; Slocum & Dilloway, 1991). Simplemindedness
refers to the organizational disability that prevents
effective analysis, so it affects what many argue is the
core of the learning organization. Yet, these results
indicate that barriers to this process do not reduce
innovativeness.
The findings regarding requisite variety help explain
this unexpected finding. Requisite variety and
simplemindedness are related, but distinct concepts.
Requisite variety refers to the quantity of diverse
information and perspectives available in the organization,
but says nothing about how well that information is
analyzed. Essentially, requisite variety deals with
quantity of information while simplemindedness addresses
quality of analysis.
Requisite variety has the strongest impact on
innovativeness of all the learning disabilities. These
results indicate that different perspectives are more
important in developing and implementing solutions to
122
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
organizational problems than the ability to analyze limited
information well. Several researchers have come to similar
conclusions. Nonaka, (1988, 1990, 1991) argues that the
process of variety amplification and then reduction is at
the heart of Japanese approaches to organizational learning.
Sitkin (1992) found that firms which punish failures have
restricted perspectives on cause and effect relationships.
An acceptance of "intelligent" failures results in a much
greater variety of information concerning these cause and
effect relationships. Both authors suggest that
sophisticated analytical systems working with limited
information, or information from only one paradigm, will not
produce effective results. The case studies, reported in
the next chapter, also highlight this conclusion.
Schizophrenia Promoting, not Hindering Innovativeness.
A second surprising finding is that schizophrenia actually
promotes organizational innovativeness. I had hypothesized
that it would hinder innovativeness. Schizophrenia refers
to a lack of coordination between multiple competing
perspectives, or departments, in the organization. The
results indicate that this lack of cohesiveness in
organizations promotes innovativeness rather than hinders
it.
Once again, a consideration of the importance of
requisite variety helps explain these contrary findings.
123
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Organizations that are characterized by schizophrenia, as it
is operationalized here, would have departments or subunits
that operate fairly independently. Organizations with many
independent departments are likely to have a much greater
diversity of information and perspectives than firms with
tight structures (Mintzberg, 1979; Weick, 1979). Thus
schizophrenia promotes requisite variety, and by increasing
requisite variety it increases innovativeness.
These results suggest that for organizational learning
to occur, it is not necessarily important that every
organizational unit be perfectly integrated with the rest of
the organization. What is important is that whatever
learning occurs in one unit is transferred to all other
relevant areas of the organization (Crossan, 1991; Duncan
and Weiss, 1979). Organizational learning does not suffer
if departments act autonomously, as long as the learnings in
the sub-unit are transferred to other organizational units
(see Adler, 1990, for an opposing perspective) . Thus
schizophrenia is not a problem in and of itself; the key is
whether the learning can be diffused throughout the
organization. Diffusion deficit is the disability that
affects this process of spreading learning throughout the
•organization, and the results of this study do indicate that
diffusion deficit hinders innovativeness.
1 2 4
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
No Relationship between several of the Disabilities and
Innovativeness. As mentioned previously, six of the
learning disabilities (internal blindness, projection,
simplemindedness, hyperactivity, alien hand, and amnesia)
did not show a significant relationship with organizational
innovativeness.
The results suggest two possible explanations for this,
both having to do with measurement. First, these scales
had among the lowest cronbach reliabilities and most
problems with the confirmatory factor analysis. The four
lowest alphas were among this group. Second, while none of
the residuals were greater than two standard deviations away
from expected, the five largest individual residuals were
among these six scales. These points suggest that these six
disabilities were less effectively measured than the others.
Weak measures tend to reduce whatever relationships exist
between the dependent and independent variables.
The other possibility, of course, is that the model is
misspecified. While there is support in the literature
linking the concepts of internal blindness, projection,
hyperactivity, alien hand, and amnesia to the learning
outcomes, most of this support has not been empirically
based. This empirical study -indicates that these
relationships are problematic at best, and it may be that
the theory in these cases does not fit the empirical
125
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
reality. Further empirical research needs to be conducted
to assess these relationships in more detail.
Summary of the Relationship between the Organizational
Learning Disabilities and the Outcomes.
The- learning disabilities of external blindness,
requisite variety, paralysis, diffusion deficit, and
superstitious learning were found to reduce organizational
innovativeness. The organizational outcome of perceived
innovativeness did lead to organizational competitiveness.
Schizophrenia had a significant relationship with
innovativeness, but in the opposite than predicted
direction. The learning disabilities of internal blindness,
projection, simplemindedness, hyperactivity, alien hand, and
amnesia were found to have no significant relationships to
innovativeness. Possible reasons for the non-significant
findings were presented and discussed.
I now turn to the relationship between the
organizational learning disabilities and its antecedents.
Relationship between the Learning Disabilities
and its Antecedents
Hypotheses regarding the relationships between the
Organizational learning disabilities and antecedents were
126
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
developed in Chapter III. Here both regression and
correlation analyses are used to examine these hypotheses.
Due to the relatively scarce literature that specifically
addresses these hypotheses, this section is treated as
exploratory.
Correlational Analyses
In order to simplify the analyses and presentation in
this section, the twelve learning disabilities are
aggregated to the four stages of the learning model. The
dependent variables in this section are now the discovery
disabilities, invention disabilities, production
disabilities, and generalization disabilities. The results
from the LISREL secondary confirmatory factor analyses
indicate that such aggregation is appropriate. I did
regress each of the twelve learning disabilities on all the
antecedents and the pattern of results is the same.
Table 5.11 provides correlations between the learning
disabilities and the antecedent variables. As hypothesized,
the correlations indicate relationships in the hypothesized
directions. The major exception is the national culture
dimension of power distance, which has no significant
correlations. Overall, the national culture correlations
are the weakest, which is consistent with the fact that
national culture and the learning disabilities' values came
127
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
TABLE 5.11
Correlations between Learning Disabilities and Antecedents
Disc.______Innv. Prod. Gener.
Firm age (In) .0959 .1000 .0767 .0183
Size (In) .0882 .1459* .2751** .1249*
National Culture
Individualism .1845** .1824** .1779** .1247*
Masculinity .2358** .2729** .2119** .2185**
Power Distance -..0496 -.0331 -.0212 .0092
Uncert. Avoid. -.1585* -.1380* -.0657 -
.1632**
Org. Culture '
Adhocracy -.6698** -.6349** -.4408** -. 4664**
Bureaucracy .2047** .1863** .0964 .0931
Clan ■ -.7305** -.7608** -.5770** -.5931**
Market -.6169** -.6158** -.5277**
-
.5399**
Leadership
Create Need -.4719** -.4748** -.3509** -.4559**
Shape Vision -.6621** -.7149** -.6146** -.6283**
Lead Change -.7189** -.7534** -.6450**
-
.6766**
Mobalize -.6586** -.7080** -.5523** -
.5648**
Systems/Struct -.7167** -.7908** -.6531**
-
.6584**
Monitor Comm -.5547** -.6089** -.4814** -.5507**
Make Last -.7100** -.7484** -.5989**
-
.6530**
HR System
Communication -.6133** -.6331** -. 4546**
-
.5501**
Org. Design -.6012** -.6478** -.5131** -.5811**
PA -.4782** -.5061** -.4260**
-
.5233**
Rewards -.3624** -.3833** -.2742**
-
.3344**
Staffing -.5112** -.5359** -.4801** -.5147**
Training -.5596** -.5185** -.3804** —.5354**
Env. Change -.2760** -.2405** -.0952 -
.1380*
* significant at-.05 level
** significant at-.01 level
1 2 8
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
from completely different data sources. These correlations
suggest modest support for the hypotheses. However, zero
order correlations provide a crude analysis at best.
Regression analysis is necessary to ascertain the true
relationships among the variables.
Multicollinearity between the independent variables was
checked through the calculation of Variance Inflation
Factors (Neter, Wasserman, & Kunter, 1985; Myers, 1990).
Variance inflation factor (VIF) values greater than 10
'indicate possible bias due to multicollinearity (Meyers,
1990) . None of the VIF exceeded 10 with this data,
indicating that multicollinearity is not a major issue.
Regression Analyses
A regression analysis was run for each of the four sets
of organizational learning disabilities. The results are
presented in Table 5.11 and summarized in Table 5.12. Each
regression equation had a significant F-statistic that
explained between 57 and 74 percent of the overall variance.
These high adjusted R2s indicate that many of the important
antecedent variables for the learning disabilities were
included in this study.
In general, each category of antecedent variables
(national culture, organizational culture, leadership, HRM,
and environmental change) are important predictors of the
129
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
TABLE 5.11
Regression Analyses: Learning Disabilities on Antecedents
VARIABLE DISCOV INVENT. PRODUCT. GENERAL.
Age (In)
Size (In)
National Culture
Individualism
Masculinity
Power Distance
Uncertaint Av.
Org. Culture
Adhocracy
Bureaucracy
Clan
Market
Leadership
Create Need
Shape Vision
Leading Change
Mobilize Commit.
Systems & Struct
Monitor Commit.
Making it Last
HR System
Communication
Org. Design
Performance App.
Rewards
Staffing
Training
Environment Change
R2
Adjusted R2
-.03 -.04
-.01 .01
.001 .003+
.001 .003+
-.001 .001
.000 -.003+
-.12* .01
-.02 .00
-.07* -.14**
-.08* -.05
.03 -.11*
-.06 -.11*
-.12* -.16*
-.05 -.05
-.05 -.18*
-.02 -.08
.05 .07
-.04 .00
.01 -.03
.01 -.01
.03 -.01
-.12** -.13*
-.14*** -.07+
.05 .02
.70 .77
.67 .74
20.7*** 29.2***
-.08 -.17*
.03** .03*
.003+ .004*
.003+ .001
.003 .006*
-.002 -.001
.11* .10
.03 -.03
-.11* -.05
-.10+ -.04
-.13* .03
-.19** -.17**
-.27** -.29**
.08 .14+
-.15 .02
-.01 -.02
.02 -.08
.10+ .00
-.10+ -.07
-.00 -.04
.03 .04
-.19** -.18**
-.05 —. 21* * *
.14* .16*
.61 .64
.57 .59
13.9*** 15.5***
Values shown are the unstandardized regression coefficients
+ p < .10 * p < .05 ** p < .01 ** p < .001
1 3 0
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
TABLE 5.13
Summary of Results
Regression Analyses: Learning Disabilities on Antecedents
** **
** **
VARIABLE____________DISCOV. INVENT. PROD. GENERAL.
Age (In) *
Size (In) ** *
National Culture
Individualism + + *
Masculinity + +
Power Distance *
Uncertainty Avoid. +
Org-. Culture
Adhocracy * *
Bureaucracy
Clan * *** *
Market * +
Leadership
Create Shared Need * *
Shape Vision *
Leading Change * *
Mobilize Commitment +
Systems/Structures *
Monitor Commitment
Making it Last
HR System
Communication +
Organization Design +
Performance Appr.
Rewards
Staffing ** * ** **
Training *** + ***
Environment Change * *
R2 .70 .77 .61 .64
Adjusted R^ .67 .74 .57 .59
F____________________ 20.7*** 29.2*** 13.9*** 15.5***
Values shown are the unstandardized regression coefficients
+ p < .10 * p < .05 ** p < .01 ** p < .001
1 3 1
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
organizational learning disabilities. Roughly half of the
specific variables had significant relationships with the
disabilities, and the significant associations are spread
evenly throughout the categories of antecedent variables and
types of learning disabilities. The discovery disabilities
had the fewest significant relationships, and none with
national culture. The production disabilities had the
greatest number of significant relationships.
In general these results provide modest support for the
hypotheses developed in Chapter III. Each of the hypotheses
are now discussed.
National Culture
As hypothesized, uncertainty avoidance has a negative
association with the invention disabilities. The
relationships with the other disabilities is in the
hypothesized direction, but does not reach significance. It
is not surprising that the strongest relationship was with
the invention disabilities as Sullivan and Nonaka (1985)
argue that the reduction of uncertainty is particularly
important in generating a wide variety of ideas and
developing creative solutions.
Individualism is associated with the production
disabilities, confirming hypothesis 4b. However, contrary
to hypothesis 4a, it was also found to be a predictor of the
132
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
invention disabilities. Many researchers have found that
individualist countries are particularly good at developing
innovative solutions to problems (Kedia & Bhagat, 1986,
Shane et. al. 1993). This was probably true a decade or so
ago, but Stata (1989) argues that in today’s complex
business environment problems and strategies need to be
addressed by diverse, cross-functional teams in order to be
effectively handled. Individualist cultures have more
difficulty working in such teams (Hofstede, 1991), so
conceivably they would have more of the invention
disabilities.
As hypothesized, power distance was found to promote
the generalization disabilities. This is expected as high
power distance cultures tend towards rigid hierarchies
(Hofstede, 1991) that have more diffusion problems than
looser structures (Mintzberg, 1979; Pfeffer, 1981). Power
distance, however, did not have significant relationships
with the other disabilities.
No specific hypotheses were offered concerning
masculinity. These results suggest that high masculine
cultures with a focus on competition will have more of the
invention and production disabilities. Or, looking at it
from the other perspective, feminine cultures with their
emphasis on cooperation, consensus, and solidarity
(Hofstede, 19?D will have less of the invention and
production disabilities.
133
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Overall, some modest support was found for the national
culture hypotheses, but unexpected and insignificant results
occurred as well.
Organizational Culture
The results concerning organizational culture tend to
support the hypotheses. Clan, adhocracy, and market
cultures had less of the learning disabilities, and clan
cultures had the strongest association with a reduction in
the disabilities. Bureaucracy cultures did not have any
significant relationships with the learning disabilities.
The only surprising finding is that adhocracy cultures had
more of the production disabilities. This makes sense in
hindsight, as adhocracies are loose, decentralized, and
often slightly chaotic (Quinn, 1988), and such
characteristics would be expected to hinder the effective
implementation of new organizational procedures.
Leadership
As predicted, the leadership characteristics of
creating a shared need, shaping a vision, and leading change
were found to be associated with a reduction in the learning
disabilities. The other leadership characteristics were not
found to have much of an impact on the disabilities. Thus
134
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
only moderate support was found for the leadership
hypotheses overall.
HRM System
Effective training and staffing policies had strong
associations with a reduction in nearly all the learning
disabilities. As expected, effective organizational design
policies were found to be associated with reduced production
disabilities. None of the other HRM variables has
significant relationships with the disabilities. It is
particularly surprising that effective reward systems did
not have any significant relationships. One possible
explanation for this is that the HRM items were very general
and did not ask questions specifically directed towards
organizational learning.
Environmental Change
Environmental change was found to have significant
relationships with the production and generalization
disabilities, but in the opposite direction than
hypothesized. I had hypothesized that environments
characterized by rapid change would have less of the
learning disabilities. It appears that this is true only
for successful firms. For many organizations a rapidly
135
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
changing environment appears to exacerbate the learning
disabilities.
Summary of the Relationship between the Organizational
Learning Disabilities and the Antecedents.
Modest support was found for the hypotheses relating
the organizational learning disabilities to the antecedent
variables. National culture, organizational culture,
leadership, HRM systems, and environmental change were all
significant predictors of the learning disabilities at the
broad level.- The specifics varied quite a bit, and only
about 40% of the hypotheses were supported in the predicted
directions. Given the exploratory nature of these
hypotheses, this is a significant contribution. However, it
is clear that more specific, empirical research needs to be
done in each of the areas to pin down the precise
relationships.
Chapter Summary
This chapter presented and- discussed the results of the
construct validation of the organizational learning
disabilities and the hypotheses relating the disabilities to
organizational outcomes and antecedents. Strong support was
provided for the measurement model of the twelve learning
136
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
disabilities and their relationship to the four stage model
of organizational learning. A LISREL path model relating
the learning disabilities to organizational innovativeness
and competitiveness also appeared strong, even though not
all of the specific pathways were significant. The fact
that both LISREL models worked as well as they did, given
their high level of complexity, is a good indication of
their robustness. The exploratory analyses relating the
antecedents to the learning disabilities found more mixed
support.
In general, national culture, organizational culture,
leadership, HRM systems, and environmental change were found
to be significantly associated with the learning
disabilities and explained between 57 and 74 percent of the
variance in each equation. However many of the specific
hypothesized relationships were not supported. For example,
no significant relationships were found with bureaucratic
organizational cultures, compensation, performance
appraisal, and leaders monitoring commitment to change.
Given the only modest findings in predicting the specific
relationships with the antecedent variables, further
research is needed.
These findings suggest ’ overall support for the
construction of a nomological network of the organizational
learning disabilities, the-ir antecedents, and their
outcomes. The following chapter discusses two case studies
137
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
in order to give some context and depth to the results
presented up to this point.
1 3 8
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
CHAPTER VI
CASE STUDIES OF SAMSUNG ELECTRONICS CO.
AND THE HEALTH CARE FORUM
This chapter has two goals. The first is to add depth
and context to the previous analysis by examining two cases
in more detail. The cases provide a useful contrast as
Samsung Electronic Co. (SEC) was found to have successfully
combated the organizational learning disabilities while the
health care organizations were less successful in this
regard. The second goal is to provide additional validity
to the learning disability constructs by examining them
through a different methodological approach. The health
care organizations were particularly useful in this regard
as I was able to interview roughly 50 respondents
immediately after they had completed the questionnaires.
Samsung Electronics Co.
Prior to 1979 Samsung Electronics Co. (SEC) did not
make microwave ovens. Ten years later it had one fifth of
the US market. Despite this impressive success, in 1991 SEC
139
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
chose to completely change its organizational structure and
production system. Why was such radical change seen as
necessary? The answer has to do with what it takes to
successfully learn when entering a mature industry. This
case will examine these issues through a focus on the
implementation of Flexible Manufacturing Systems (EMS) at
the Samsung microwave plant in Suwon, South Korea.
In 'order to provide context for the actions taken by
SEC I will first briefly discuss some relevant aspects of
the Korean economy and of the history of the Samsung
conglomerate (Chaebol) and of SEC. The focus of this case
is on the changes taken by SEC in 1991 to become a learning
organization. I present the challenges Samsung faced and
how it developed solutions to the problems, with a focus on
just one aspect of the solution— the implementation of FMS.
Issues such as how SEC learned about FMS technology, learned
how to implement FMS, and built support mechanisms to
reinforce FMS are dealt with in some detail. Finally, I
relate this case to the concepts of the organizational
learning disabilities that were developed in the previous
chapters and briefly discuss some implications of the case.
Background
South Korea is often touted as an economic miracle.
Starting nearly from scratch after the Korean War, Korea
140
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
became the twelfth largest trading nation by 1988. During
most of the 1980s South Korea enjoyed double digit growth
which has recently dropped to a still enviable 8.4%. There
are four primary reasons for this success. (1) After the
Korean War the Korean government identified target
industries and .target companies and provided these firms
with special financial and legal incentives. The government
set high performance standards for these firms and held each
chairman personally responsible for performance. Typically
these firms were large, highly diversified companies
referred to as chaebols, many of which were dominated by one
central family. (2) Korea has a highly educated, skilled,
and dedicated work force that has the longest average
workweek in the world (54.5 hours per week) and until
recently had extremely low annual pay. This combination of
long work for low pay enabled the Korean chaebols to produce
products at a lower cost than their competitors. (3) The
entrepreneurial skills of the chaebol founders, chaebol
managers, and Korean businessmen in general was very high.
(4) There was a sense of mission at all levels in Korean
society that Korea must enter the forefront of the
industrialized nations as well as a willingness to sacrifice
the present in order to reach their goal in the future.
Each of these factors contributed to the success of Samsung
as well as the Korean economic miracle.
1 4 1
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Samsung. Samsung was founded in 1938 by Lee Byung-
Chull as a small trading company. The Korean War left Mr.
Lee with almost no resources to rebuild his company,
however, in 1951 with savings donated by employees he
started anew. Lee was very good at spotting business
opportunities in a wide range of endeavors and expanded
Samsung into whatever industries presented such
opportunities. The result was that Samsung became a large
conglomerate of about 30 profitable, but often unrelated,
businesses. Among the many industries in which Samsung has
a significant presence, are consumer electronics,
semiconductors, computers, telecommunications, shipbuilding,
petrochemicals, construction, paper manufacturing,
aerospace, sugar, wool, insurance, and department stores.
Samsung is on the cutting edge of many high technology
fields, including DRAM semiconductor chips, robotics, and
biotechnology. While profitable and growing, this hodge
podge of unrelated companies became increasingly difficult
to coordinate over the years. In 1991 Samsung underwent a
thorough reorganization to provide synergy and focus in
three areas— electronics (SEC), engineering, and chemicals.
SEC. Samsung Electronics Co. was founded in 1969 and
has grown at a rate of about 60 percent a year. The
reorganization of Samsung focused SEC into four primary
business divisions— consumer electronics, semiconductors,
142
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
telecommunications, and computers. By 1991 it had 7 billion
USD sales and 46,000 employees. While SEC began producing
microwave ovens only in 1979, it now has captured over a
fifth of both the US. and world market, though most of its
microwaves are sold under other brand names, such as General
Electric. SEC is also the world’s third largest DRAM chip
maker.
Yet.at the height of this remarkable success, and while
still very profitable, SEC chose to discard its efficient
production system and rebuild it literally anew. The new
lines were designed around FMS and entailed not merely new
physical structures but a complete reorganization of how it
conducted business, as well as a mind set change on the part
of the employees: they had to embrace change and learn how
to build capability from it.
The SEC microwave plant I studied is part of an
enormous industrial complex in the town of Suwon, about one
hour south of Seoul. The Suwon complex occupies 370 acres
and has 14,000 employees. It produces not only microwave
ovens but TVs, VCRs, radios, stereos, video monitors,
cameras, dishwashers, heaters, microchips, and glass for TV
picture tubes; and these are only a few of their many
products. Many supposedly Japanese high quality products
are actually built here, such as Sony Walkmans, Seiko
watches, and Minolta Cameras. While large, the various
plants of the Suwon electrical complex are fully integrated.
143
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
The Change to FMS
The Problem (Discovery) . Since SEC's microwave plant
was so profitable, it would have been very easy to sit back
and rest on its laurels without paying much attention to the
environment (blindness). Yet, some customers were asking
for product orders that were either too unique or small for
SEC to profitably fill. SEC already made all General
Electric microwave ovens and both sides were pleased with
the relationship, but in order to reduce inventory costs, GE
asked f o r . . . small batches of some models to be delivered on a
just-in-time basis. Such orders were beyond SEC's abilities
and, despite their ability to fill large orders, this
partial failure was seen as a failure of the whole system.
One of SEC's guiding values had always been to meet customer
needs, so the failure to be able to do this in even a few
instances caused great concern. And try as they might,
teams could not come up with a way to meet such future
orders profitably within the constraints of their existing
system. This strong set of guiding values focused towards
customer needs resulted in SEC being hyper-perceptive
towards customers. Thus these values effectively prevent
the disability of blindness from occurring.
144
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Developing a Solution (Invention) . The first step was
analyzing the situation and identifying the problems. One
problem could be attributed to the success of the Korean
economy. Korean wages had risen significantly in the 1980s
so that Korea was no longer considered a low wage country,
at least when compared to other Asian countries like China
and Malaysia. In fact both Chinese and Malaysian companies
were following the Korean example of translating their low
labor costs into inexpensive entry level products.
Criticism of the special incentives provided to Korean
chaebols by the government were gaining strength in Korean
society, and the government responded by removing many of
them. A rising Won also made Korean products more expensive
overseas. For all these reasons, SEC and other chaebols
were struggling to keep costs down.
SEC had also matured from low cost, entry level
products to more of a focus on quality and reliability. It
realized that while they found it difficult to compete on
cost anymore, it did not have the cutting edge technological
and production systems of the high quality leaders. The
production system at the microwave plant was based on
traditional assembly line principles and required large
production runs to be profitable. Major reasons for its
past success were the hard work and low pay of its work
force. SEC could no longer base its competitiveness on cheap
labor and yet was not as modern as the Japanese competition.
145
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
The old production system based on low costs, which no
longer existed, did not have the flexibility and quality to
compete with the best of the Japanese. Analysis indicated
that these trends and their resulting problems for SEC would
continue and worsen in the future.
What to do about it? In 1990 Chairman Lee, the son of
Lee Byung-Chull, fostered a total commitment program for
dealing with the problem by constantly exhorting employees
to ignore their large profits and think instead as if they
were in the midst of a financial crisis. SEC made the
analysis of the problem its main priority. Competitor
practices were quickly analyzed. Teams studied various
academic theories and applied them to SEC's situation. It
became clear that SEC's most successful competitors were
Japanese firms that were organized around EMS. FMS, along
with the related programs of just-in-time management, batch
production, and total quality seemed the perfect solution.
It would allow SEC to meet the varied needs of its customers
by producing small batches of unique products on the same
line, quickly and profitably. So SEC decided to restructure
along FMS lines.
Unfortunately, making the decision was easier than
implementing the processes needed to achieve it. FMS was
totally foreign to Korean businesses, and SEC was no
exception. Not only did SEC have no experience with FMS, but
it's structure was antithetical to many of the basic
14 6
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
precepts of FMS. Further compounding this problem was the
fact that many of the top Japanese firms treated FMS as
their source of competitive advantage and were not open to
sharing it's secrets.
SEC had a good foundation for beginning this learning
due to its aggressive educational programs. Throughout the
1980s SEC provided many managers with two to three months of
intensive foreign language training, primarily in English
and Japanese. These managers often received grants to study
for advanced degrees at universities in the US., Japan, and
Europe, with SEC picking up expenses for tuition, family
living expenses, and salary. This program had two important
results: first, many managers and supervisors were familiar
with the latest academic theories involving FMS and second,
the language proficiency allowed SEC to immediately send
managers to study the production systems of foreign
companies.
SEC took several approaches to learning more about FMS.
A good deal of the learning was from books and articles,
both in the popular press and academic literature. The
educational programs and the high number of Ph.D.s at SEC
contributed to the ability of SEC managers to do this
research. The planners read everything they could and
determined that they should focus on American theories and
Japanese practices.
147
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Joint ventures were undertaken with exemplary Japanese
and US. firms. In such cases it was to the partners'
benefit to make the joint venture run as effectively as
possible, and so, to transfer FMS know-how to SEC. Also,
SEC was able to trade some of its products for core
capabilities of leading firms. For example, SEC exchanged
DRAM chips to Toshiba for EMS know-how.
Due to its buying power, SEC was able to make large
purchases from Japanese firms and require the transfer of
FMS expertise and know-how as part of the sale. This often
involved sending SEC managers to the Japanese plants to
study their system. SEC has such a wide range of products
that many Japanese firms were willing to provide FMS
expertise to establish a relationship with such a large
potential customer.
SEC was able to circumvent the large Japanese f irms'
resistance to share their production technology, by
fostering strategic alliances with foreign small to mid size
firms that had the capabilities SEC wanted to learn. Joint
projects with these smaller firms had many advantages over
the larger firms. Small and mid sized companies were less
resistant to share state of the art information, especially
if they had financial difficulties where SEC could help out.
Often the smaller firms were more cutting edge than larger
ones and also had a great deal of knowledge about the
practices in the larger firms as well.
148
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
This section makes it clear that SEC went to great
lengths to prevent insufficient requisite variety. To
summarize/ SEC acquired diverse knowledge about EMS through
a variety of mechanisms, including analyzing competitors;
study abroad in Europe, Japan, and the US.; "academic"
research; and the transfer of knowledge from exemplars of
EMS through joint ventures, buying, and trading. The fact
that learning was acquired from three continents,
competitors, and academic theories insured that not only a
wealth of information was gathered, but that many different
paradigms were considered as well. In all, SEC appears to
be an exemplar concerning requisite variety.
Another way that SEC limited insufficient requisite
variety was through tolerating failures. They came to the
realization that when failures were punished it stifled
creativity and risk taking. The lenient treatment of
failure combined with generous rewards for success resulted
in a strong "go getter" attitude among employees. Managers,
R&D researchers, and engineers want "impossible" projects:
or projects that others think are impossible. Many
employees attributed the success of SEC developing products,
when no one thought it could be done, to this enthusiasm for
great challenges and the "can do" spirit of Korean workers.
Failures, when they do occur, are treated as making a
positive contribution in that they provide an opportunity to
149
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
learn. The failure is thoroughly analyzed to determine why
they failed so that future failures can be prevented.
The next step was to take this knowledge and implement
it.
Implementing SMS (Production). The learning acquired
from the Japanese was integrated with the research from the
popular and academic literature and used to form a rough set
of general guidelines. While the learning to date was
important it was problematic in several ways. The knowledge
was acquired from many different sources and was based on
many different products— ranging from telecommunication
equipment to computer chips. It was not perfectly clear how
to apply this knowledge to other products. Their
conclusions were largely untested within SEC itself so it
was unclear how the unique aspects of SEC would affect the
process.
As the microwave plant was the catalyst for beginning
the exploration into EMS, it was decided to implement FMS
there first, work out the bugs, and only then apply FMS to
other product lines. SEC was very aware of the difficulty
in implementing large scale changes and of resistance to
such change, what I have referred to as paralysis. To
insure the success of the program, a total commitment was
made to what SEC calls the 3 Ms. Whatever money, manpower,
. • 150
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
and machinery that was needed, was provided without
question.
The job of translating this general knowledge of FMS to
SEC's unique conditions, and for microwave production, was
given to groups of cross functional and cross level teams.
They developed specific plans for building FMS and were
given carte blanche to build the new line. Work proceeded
around the clock. The building of the new FMS microwave
line and reorganization of the plant occurred primarily
through a system of trial and error. A prototype line was
built, its performance analyzed, modifications were made
with further analysis, and so on.
Top management commitment also helped reduce Alien-Hand
Syndrome (when actions and intentions are disconnected).
The commitment was so pervasive with such resources backing
it, that it would have been difficult for actors not to know
what the organizational intentions were.
The methods used to acquire the knowledge were also
used to help implement it. For example, through joint
ventures and buying clout managers, engineers, and
supervisors from Japanese FMS plants were brought in to
analyze the early attempts. SEC also hired foreign
technical assistants (primarily retired Japanese and
American managers and engineers) to provide assistance with
the implementation. With their practical knowledge these
outside experts were able to offer guidance with
151
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
particularly difficult problems that might otherwise have
resulted in paralysis.
Even though the initial efforts met with only limited
success, top management continued to provide it's full
support. Eventually the problems were ironed out. From
start to finish, the restructuring to FMS took only 6 months
and was completed in early 1992. This commitment of top
management is also a crucial aspect in avoiding paralysis.
Despite the effort, the initial attempts at implementing FMS
were not totally successful. Had top management withdrawn
its backing, the whole effort would likely have stalled
there.
The new system was radically different from the old.
The old production line of 605.7 meters was reduced to only
64.9 meters. The lead time for each product under the old
system was 181.5 minutes; now it is only 25.3 minutes with
FMS. In the course of a day each line is switched over to
produce 4 or 5 different products, all made possible by the
new FMS.
Paralysis was also reduced by the fact that employees
work at stations they design themselves. When switch overs
occur it is the workers who breakdown and rebuild their
station. This differs substantially from the typical
American practice of changeovers where management decides on
the change, industrial engineers redesign the line, other
engineers "build" the new line, and finally workers are
152
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
trained for (or often just placed on) the new line. At SEC
most of these steps have been eliminated by having the line
workers themselves participate in the redesign analysis and
actually build the new stations themselves. This eliminates
the need for training an employee to work on someone else's
design as the SEC worker is already the expert on her own
station. Since workers have participated in the design they
are unlikely to resist the changes. The result is an
organization that is leaner, more dynamic, and filled with
experts at every step of the way.
Spreading EMS Throughout SEC (Generalization). As was
mentioned previously, FMS was initially implemented only in
the microwave plant. Now that it has proven successful the
lessons are being analyzed and spread to other areas of SEC.
Much of the same process that the microwave plant went
through of conceptually learning about FMS and then figuring
out how to apply it to specific product lines is being
repeated by other plants. It is much easier now though as
they learned how to do this: the knowledge is now internal
to SEC and microwave plant changes provide a road map for
other plants. Employees that were involved in the
changeover are now leading teams in other plants with other
product lines, thus helping to minimize diffusion deficit.
All learning and problems are almost instantly
documented, thus reducing Amnesia. For example, every
153
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
worker can shut down the whole line if a problem surfaces.
At the head of each line are 4 clocks and four different
colored lights that resemble police sirens. Each light and
associated clock record the reason and time the line shuts
down. If a worker notices a quality problem she shuts down
the line by pulling one of the hanging levers that is
located periodically along the line. For a quality problem
the red light flashes and a chime sounds. The light and
chime can be heard all over the plant and it signals the
appropriate supervisor to come.’ The quality clock records
the length of time the line was down due to this problem.
Other sets of lights/clocks/chimes are for equipment
failures, lack of materials, and temporary line shutdowns.
So the monitors display every type of problem so the
appropriate supervisor can handle it, and a record of all
types of problems is automatically made.
All employees can stop the line and not only are they
encouraged to do so, they are rewarded for it. It is
considered part of their job to search and discover
problems. If not enough problems are discovered it is
assumed that the system is not working properly. Most
problems are solved quickly by the worker or the worker and
supervisor together. More substantial problems are solved
by teams of workers and managers and the first hour or two
of each shift is devoted to such team based problem solving.
154
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Such rapid documentation and problem solving also
reduce superstitious learning, which involves an inability
to draw correct cause and effect relationships. Since
feedback and accurate information is generated
instantaneously, cause and effect relationships are much
clearer than they otherwise would be.
Relationship of the Antecedents to the Learning Process at
SEC.
The culture of SEC was equally important. Many
employees at all levels said "if we don't change we'll die."
This remarkable attitude is based on a conscious guidance of
the firm's culture by Chairman Lee. SEC uses "corporate
culture research camps" to educate workers about the
corporate culture and why it is important. Workers go to
this camp for several days at a time to learn and develop
cultural and management skills. These workers then return
to the plant and become teachers to other employees. The
corporate culture research camp was instrumental in
explaining the importance of change to SEC employees and
fostering the "if we don't change we'll die" mentality.
This absolute belief in the importance and acceptance of
radical change was certainly a major factor in the
acceptance of FMS by the work-force.
155
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
The corporate culture research camps serve another
purpose as well. Due to the fact that both SEC and its
culture are continually changing, the camp helps determine
what the appropriate culture should be. This helps provide
compatibility between SEC's culture and its strategy and
structure. By all accounts the culture camps help create a
strong "Clan" type organizational culture.
The education and training systems at SEC are also
important supports for their EMS. I have already discussed
their university educational programs, which were provided
to over 1000 employees in 1992. Even more extensive are
their training and vocational programs. A typical training
regimen for new engineers includes one month of lectures on
the company, 3 months of general skill or technological
training related to their jobs, and 9 months of on-the-job-
training (Koh, 1992). SEC also has a recently completed 200
million USD training center which provides its researchers
and engineers with training in the latest technological
advancements. Perhaps most important of all is the
continuous on-the-job training. Since the production line
is continually changing and problems are constantly being
surfaced employees need to be constantly learning. Some of
this learning arises through the vocational programs, but
more of it occurs from working with fellow employees on
teams and job rotations. In fact, the supervisors primary
role is.to facilitate learning for the employees.
156
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
The theme of continuous learning and education is
further emphasized by the view at SEC that the ideal leader
is first and foremost a teacher. The emphasis at all levels
is on sharing one's expertise and skills. The sharing of
expertise can be to one's subordinates, colleagues, or
superiors; one-on-one; or in groups. This is why SEC is
organized into groups, to promote such sharing. There are
45,000 total employees and 5,548 eight to nine person
groups, so everyone is in at least one group. Employees and
managers at SEC described their ideal leader in the
following ways— teacher, coach, cheerleader, gate keeper,
someone to lead us and give us training, and a facilitator.
They totally reject the idea of the "traditional leader"— a
superior who gives orders, a take charge autocrat, a
dominant paternalistic figure. The brilliant "army general"
is not wanted at SEC.
Trust was emphasized by both management and line
workers as a crucial component of the success of FMS. The
lifetime employment and continuous training allowed
employees to accept changes with out fear of job loss or job
obsolescence. Because employees were paid a very high
salary (by Korean standards) and not piece rate they had no
fear that change will depress their short term earnings and
the high salary fostered commitment. SEC also provides large
employee benefits' like paying for the schooling of worker's
children,, providing a dormitory for workers who live outside
157
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Seoul, supporting recreational clubs, providing a present at
employee birthdays and at holidays, etc. All these programs
build up a great deal of loyalty and trust in the work
force. This trust means that the employees will agree to
changes because they have faith in SEC and faith that
whatever happens, SEC will treat them well.
FMS could not work at SEC without empowerment. Even
top managers feel the most important aspects of decisions
are made by lower management and workers. In fact many
decisions are actually started by lower level employees.
When a lower level manager or worker has an idea they are
encouraged to form a project team on their own to develop
it. The team then reports the recommendations to their
superior. The recommendation is either implemented or
passed up the hierarchy for approval. Only very rarely will
mid or upper management veto an idea and most ideas are at
least tried out experimentally.
FMS is also supported by a combination top down— bottom
up leadership style. The process operates both upward and
downward simultaneously. Not only does the top make general
decisions that are clarified and operationalized as they
move down the organization but often ideas are generated and
implemented at the bottom. The ideas or results then move
up the organization for approval and distribution throughout
the firm. The role of the progressively higher levels of
management is to choose between existing modifications,
158
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
rather than make new changes. If changes are made by top
management, the process starts all over again. SEC is
currently working on flattening its organizational
structure. Slim organizations are quicker because as there
are fewer layers of management for decisions and information
to pass through, and they are more effective as decisions
are made closer to where they will be implemented.
While this may sound like a laundry list of different
programs and practices they are all interconnected and
dependent upon each other. For example, the flexible
manufacturing system requires highly versatile and skilled
workers who are able to spot problems and are empowered to
sdlve them. The extensive training and education programs
provide the skills employees need to do this. Empowerment
only works to the benefit of the firm if the semi-autonomous
workers are using their freedom to pursue organizational
goals. The strong SEC culture and trust of management helps
assure that the individual and organizational actions and
goals are aligned toward a common end. Top management
support of money, manpower, and machinery for change also
provide the necessary material conditions for FMS.
Summary of SEC Case
In order to have a learning organization the firm must
successfully discover problems, invent reasonable solutions
159
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
to the problems, produce workable action from the solutions,
and disseminate the results through-out the firm so that the
learning occurs at the firm level. A failure of any one of
these processes will result in incomplete learning. The
presence of any of the organizational learning disabilities
will block learning as well. SEC successfully moved through
each of the stages of organizational learning and did not
have any of the disabilities, which explains its overall
success in becoming a learning organization. In fact, all
employees that I spoke with seemed to have an clear idea of
what these disabilities were and spent a great deal of
effort trying to insure that they did not develop at SEC.
While SEC was found to have few of the learning
disabilities, this is not true of the next case examined.
Health Care Systems
The second case is based on a consortium of six health
care systems (located in Minneapolis, Seattle, San Jose,
Columbus, Portland, and San Antonio). Each health care
system was made up of several hospitals, HMOs, and/or other
forms of insurance, and each was attempting to implement
organizational learning. These cases were conducted roughly
a year into their change process. In order to provide
depth to the statistical analysis I will discuss three of
these organizations in greater detail. Due to issues of
160
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
confidentiality I will refer to the organizations only by
their city.
Columbus Health
Columbus Health did a very good job of Discovery and
Invention but had serious problems with the Production phase
of organizational learning, and particularly with paralysis.
A new CEO joined the organization about a year before this
case was conducted. He was convinced that major changes in
the health care environment necessitated significant changes
in how Columbus Health operated. The new CEO was successful
in communicating this felt need to the rest of the
organization. Teams of managers engaged in benchmarking,
environmental scanning, and customer needs assessments,
including visits to "best in the class" organizations.
These teams presented their findings to the rest of the
organization.
The CEO read and became very enthusiastic about the
seminal book on organization learning, The Fifth Discipline
by Peter Senge. He bought 500 copies of the book and it
became required reading for every manager in the
organization. He also hired Senge's consulting firm to set
up a two week off site experiential learning university that
every manager was required to attend. They learned systems
thinking and spent the two weeks engaged in "Microworlds"
■ ' ' 161
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
computer simulations. At the end of this session managers
returned extremely enthusiastic about reengineering the
organization into a "lean, mean, exciting machine."
Despite this initial enthusiasm, when I visited the
organization six months later nothing had changed. The CEO
was despondent. He told me that "I did everything right and
nothing happened." Managers claimed that there was an
"invisible force" that caused all the best ideas to fade.
Many spoke of the impossibility of altering the system's
"inertia."
This represents a classic case of paralysis, an
inability to implement new ideas or solutions. The primary
causes of the paralysis is that no effort was made to change
the tangible systems of the organization. The reward system
was not modified at all, so employees were still rewarded
for "old" behaviors. The structure of the organization was
not altered. They wanted to be lean and flexible, yet still
had a typical pyramid structure with dozens of layers of
management.
This case highlights the importance of real world
actions with organizational learning. Cognitive changes are
not sufficient for learning to occur at the organizational
level. There must be changes in the systems, structures,
and procedures of the organization. This example also shows
that the presence of one disability can block organizational
learning. _
162
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Seattle Health
Seattle Health was successful in the Discovery stage of
organizational learning, but had much more trouble with
Invention. While they did not do much benchmarking of
other health care organizations, they were adept nonetheless
at assessing their local environment. Each local operation
(located in major population centers throughout Washington
state) assessed its environment through a "community team."
The community ' teams were made up of ten to fifteen
individuals from the local community and from major
stakeholder groups. For example, a team typically consisted
of 5 individuals who were patients within the last year, a
doctor, a primary care nurse, a representative from the
business community, and an administrative staff manager.
These teams were responsible for telling management what the
community wanted in terms of health care. While team
members were unpaid, they took their position very seriously
and felt that they were making a positive contribution to
the community. Management took the teams seriously as well,
and relied on them as their primary source of information.
While this has resulted in an excellent ability with
Discovery, the other learning processes are not as
effective. The community teams tell management what they
would like to see and then the senior branch manager decides
163
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
how to provide it. So effectively, there is only one person
that engages in the Invention processes. This tends to
greatly reduce the variety of ideas and perspectives
considered, and the invention process for the whole branch
is limited by the cognitive and time constraints of a single
individual. The culture that has developed in this
organization considers the branch manager as the only
legitimate source for developing solutions or new
procedures. One branch manager even jokingly remarked that
the community team had "the gall" to suggest a new
procedure, "like they have any idea how to run a hospital."
Once a senior manager developed a solution to the
community teams’ "problem" the branches were fairly
effective in implementing it. But the solutions tended to
be of limited effectiveness and so were usually discarded
after a trial period. This cycle reinforced the need in
people's mind for having a really "sharp" branch head as it
was so difficult to develop solutions that worked.
While the stages of Discovery and Production seemed to
operate fairly effectively, this organization was prone to
the learning disabilities of insufficient requisite variety
and simplemindedness. They were not ineffective at
organizational learning, but they were not nearly as
effective as they could have been.
164
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
San Jose Health
San Jose Health was very strong in the first three
stages of organizational learning, but was weak in the final
stage of Generalization. They had a charismatic CEO and a
strong top management team which mobilized the organization
around learning. They kept close touch with the latest
market changes through actively benchmarking health care
exemplars and forming cooperative learning systems with
these exemplars. They instituted a "participative
restructuring" process that was adept at both developing and
implementing new ideas and procedures.
Generalization was where they had the most difficulty.
They did not keep a record of their solutions and changes
{amnesia), which resulted in a great deal of learning being
continually lost. The only records are in people's heads,
so turnover caused problems with continuity. They also had
no formal structures for diffusing learning across the
organization (diffusion deficit) . For example, one local
hospital attempted to implement a capitated care system and
had a great deal of difficulty. Six months later the head
administrator found out that their cross town affiliate had
implemented a similar system successfully, but these results
had not been diffused throughout the organization.
1 6 5
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Summary of the Health Care Systems
While the health care organizations were effective with
some aspects of organizational learning, they each suffered
from one or more organizational learning disabilities- The
presence of even one disability was enough to substantially
reduce the overall learning.
Additional Survey Validation
From 20 to 200 surveys were collected from each of the
health care systems and approximately 20 interviews were
conducted with survey respondents immediately following
their completion of the survey. Unlike the main survey
sample, a key informant approach was not used with the
health care organizations. Instead a widespread sampling
from all levels and departments in the organization was
conducted. Managers and executives were enthusiastic about
the survey after its completion and reported little
difficulty answering questions. In fact they claimed that
taking the survey was a positive experience as it
highlighted important aspects of learning that their
organization should be doing.
Lower level employees, such as secretaries, nurses, and
support personnel did not report such a positive experience.
Many found the survey "frustrating," "tedious," and
166
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
difficult to answer. This was not an unexpected finding as
the survey was designed for upper managers with a broad
perspective on the organization. It does indicate that the
survey is not appropriate for widespread distributions.
Using the interview data, Samsung Electronics and each
health care organization were ranked on the presence and
magnitude of the organizational learning disabilities. The
surveys from each organization were then aggregated and each
organization was given a disability score based on the
questionnaire. In each instance the survey scores matched
the scores developed from the interviews and observations.
This provides some additional evidence for the validity and
reliability of the survey.
Chapter Summary
Two case studies were discussed. Samsung Electronics
Co. was found to have successfully eliminated the
disabilities and thus was successful at organizational
learning. The health care systems all had at least one of
the disabilities and were not successful in implementing
organizational learning. The cases provided additional
support for the surveys and a few problems were discussed.
1 6 7
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
CHAPTER VII
CONCLUSIONS
Though -there is great interest in organizational
learning among both academicians and practitioners, the
literature remains a confusing mixture of theories
unsubstantiated by empirical research (Hubbell, 1991;
Peters, 1992) . The goal of this research has been to help
clarify organizational learning by focusing on one important
aspect, organizational learning disabilities. Specifically,
this dissertation has taken a theoretical approach to
examining the learning disabilities by addressing three
fundamental research questions: (1) What are organizational
learning disabilities? (2) Does the presence of the learning
disabilities reduce the organizational outcomes of
innovativeness and competitiveness? and (3) What antecedent
variables are related to the learning disabilities? A
theoretical framework linking the learning disabilities to
their antecedents and outcomes was developed to examine
these questions. This framework was tested on a sample of
253 organizations located primarily in North America, Asia,
and Europe.
168
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
This chapter discusses the contributions of my
dissertation to both theory and practice, and suggests new
directions for future research.
Contributions to Theory
Construct Definition and Validation of the Learning
Disabilities.
One factor limiting the advancement of organizational
learning is that the literature "in large part remains
murky, confused, and difficult to penetrate" (Garvin, 1993).
A reason for this confusion is that few organizational
learning researchers build upon previous theory (Hubbell,
1991) . To address this issue an attempt was made to ground
all concepts in the relevant theoretical literature. The
basic model of organizational learning used in this
dissertation was adapted from Argyris and Schon (1979), and
has widespread acceptance in the literature. Within each
stage of the model several learning disabilities were
identified from the literature and integrated into the
overall framework.
The first major goal of this dissertation was to
operationalize and measure the organizational learning
disabilities. Such operationalization has not been
attempted in the literature, even though Daft and Hubbell
169
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
(1987) claim that this is one of the most important
challenges facing organizational learning researchers.
Chapter V presented the results of LISREL confirmatory
factor analyses, which provided evidence for the reliability
and construct validity of the organizational learning
disabilities. A second order confirmatory factor analysis
indicated that the higher order four stage learning model
was an appropriate framework for the twelve organizational
learning disabilities.
Overall, strong evidence was provided for the
reliability and construct validity of the organizational
learning disabilities. First order and second order
confirmatory factor analyses indicated that twelve of the
original thirteen learning disabilities were effectively
operationalized and measured by the survey items. The
exception was "competency traps" which did not show adequate
measurement properties. This construct validation is an
important contribution to theory building and is necessary
for moving the field of organizational learning forward.
At a more general level, this research indicates that
it is possible to operationalize and measure organizational
learning concepts for large scale empirical research. Given
the complexity of the variables though, it is even more
important than usual not to take methodological short cuts.
The fact that the specific items went through 10 pretests
with interviews after each test was necessary for insuring
170
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
that the final items would have adequate psychometric
properties. Without the pretests it is unlikely that the
items would factored together or that informants would have
understood them.
Another methodological finding concerns the importance
of a multimethod research strategy. Organizational learning
is so complex, and still relatively poorly understood, that
a combination of qualitative and quantitative approaches is
necessary to fully encompass it. Either approach alone is
incomplete. For example, qualitative case studies have
problems with generalizability while quantitative measures
are unlikely to be able to measure important aspects of
organizational (e.g. whether changes in deep assumptions
occurred).
Relationship to Organizational Outcomes
Simonin and Helleloid (1993) cite the need for
systematic hypothesis testing as one of the most pressing
concerns for the field of organizational learning. Thus,
the second major contribution of this dissertation has been
to empirically assess the relationship between the
organizational learning disabilities and important
organizational outcomes. The outcomes considered were
organizational innovativeness and competitiveness.
171
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
The findings indicated that the presence of the
organizational learning disabilities reduced innovativeness
and competitiveness. More specifically, the disabilities
were found to have a direct causal link to organizational
innovativeness. Innovativeness acted in a mediating
capacity between the organizational learning disabilities
and competitiveness, so there was no direct effect between
the disabili-ties and competitiveness (with the exception of
superstitious learning which, as hypothesized, was found to
have a direct effect).
The learning disabilities which were found to be the
most important were external blindness, requisite variety,
schizophrenia, paralysis, diffusion deficit, and
superstitious learning. Several of these points merit
special attention.
Requisite variety had the single strongest relationship
with organizational innovativeness, despite the fact that it
is rarely discussed in the organizational learning
literature. This point is even more striking when the
weakness of simplemindedness is considered. Most
researchers focus on the importance of sophisticated
analytical procedures in organizations as the prime
determinant of effective organizational learning. Yet these
results indicate that such analytical systems are relatively
unimportant. What is more important is the availability of
multiple perspectives, values, and information. In other
172
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
words, at the organizational level, having a diversity of
information is more important than analytical ability.
That organizational schizophrenia was found to actually
promote innovativeness, counter to my hypothesis, is also
related to requisite variety. Departments or organizational
units that are not tightly controlled and are free to pursue
their objectives using different approaches result in
increasing' the variety available in organizations. It
becomes clear which approaches work best and which do not.
The important consideration then, is not that everyone is
doing the same thing but that the learning from any given
unit is diffused to other units. Thus, it is not surprising
that an inability to diffuse such learnings, diffusion
deficit, was found to reduce innovativeness.
The results, particularly with regard to paralysis,
also highlight the importance of implementation and action.
Coming up with solutions is not enough. Solutions must be
implemented into changed procedures, structures, reward
systems, and so on. These results indicate that researchers
who focus primarily on analysis and solution generation as
the core of organizational learning should give equal weight
to how the solutions are implemented.
The results also provide support for the basic model of
Argyris and Schon (1979). Learning disabilities within each
of the four stages of discovery, invention, production, and
generalization were found to be significantly related to
173
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
organizational innovativeness. This indicates that each
stage of the model is important for effective organizational
learning to occur. If any one stage were missing it is
unlikely that the organization could effectively learn.
Relationships to Antecedents
The third contribution involves the development and
testing of hypotheses relating the organizational learning
disabilities to important antecedents. The relationships
were theoretically discussed in Chapter IV and empirically
examined in Chapter V. The antecedent variables considered
include national culture, organizational culture,
leadership, HRM, and environmental change. While various
researchers suggest that one or the other of these
antecedent variables have an important effect on the
learning disabilities, I am not aware of any systematic
empirical testing of the relationships.
The findings suggest that, in general, each of these
sets of variables have a significant effect on the learning
disabilities. However, there was a good deal of variation
among the specific relationships. For example, some aspects
of leadership and HRM showed significant relationships with
each of the learning disabilities while other aspects were
not significant with any of the disabilities. Another
surprising ' finding was that environmental change was
174
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
positively related to the learning disabilities, not
negatively related as hypothesized. Given the exploratory
nature of these hypotheses the findings offer modest support
for the relationships.
Contributions to Practice
A greater understanding of the organizational learning
disabilities also has implications for practice. Executives
and managers attempting to implement organizational learning
principles often become frustrated with the "maddeningly
abstract and vague" literature (Peters, 1992). This
dissertation makes a contribution by helping to clarify
specific relationships among antecedents, outcomes, and
organizational learning disabilities.
Another practical implication of this research is the
finding that the presence of only one or two learning
disabilities is enough to severely retard learning at the
organizational level. The Columbus Health case is quite
typical in this regard. The CEO was able to successfully
implement many aspects of organizational learning, only to
see the whole system fail due to some "invisible force."
This dissertation clarifies those forces so that they can be
identified and combated. Samsung Electronics' shift towards
organizational learning was successful largely because
employees at all levels of the organization seemed to have a
175
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
strong sense of these learning disabilities, and took
vigorous actions to address them.
The SEC case also highlights the fact that, in order to
successfully become a learning organization a firm must do a
great many things effectively. This requires a near
complete commitment both of organizational resources and
from employees at all levels of the organization. The
health care organizations show that a less than total effort
results in failure.
Four specific recommendations for insuring the success
of organizational learning appear particularly important.
(1) An inability to effectively assess the environment and
customers can prevent learning from even beginning. (2) A
variety of information and perspectives may be more
important than sophisticated analytical procedures with
limited information. (3) The ability to take action or
change procedures is more important than the great ideas.
(4) If the learning is not diffused to all relevant areas of
the organization the learning will be lost.
Directions for Future Research
Future research could build upon this study by
developing additional outcome variables, refining the
antecedent variables, broadening the sample, and exploring
the unanticipated findings in greater detail. These
176
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
additional directions for future research are discussed
below.
Additional Outcome Variables
While innovativeness and competitiveness are important
ultimate outcomes of the organizational learning
disabilities, there are more proximate outcomes that may
■mediate this relationship. For example, if organizational
learning were occurring the immediate outcomes would include
changes in employee's knowledge (Duncan & Weiss, 1978),
values and assumptions (Argyris & Schon, 1979) , skills and
behaviors (Fiol & Lyles, 1985) and the systems and
structures of the organization (Levitt & March). Prahalad
and Hamel (1990) argue that these core competencies
(knowledge, skills, values, and systems) result from the
"collective learning of the organization" and so are the
direct outcomes of learning. Future research should
operationalize these constructs and examine the
relationships between them and the learning disabilities.
Refine the Antecedent Variables
Greater refinement and elaboration of the antecedent
variables would also make a significant contribution. For
example this study examined four types of organizational
177
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
culture: clan, market, bureaucracy, and adhocracy. While
useful, this typology simplifies a very complex topic. An
examination of organizational culture in much more depth,
perhaps through in-depth ethnographic analyses, would likely
highlight specific relationships with the learning
disabilities.
The HRM variables were also very general. The survey
was concerned with effective or ineffective HRM systems and
did not go into more detail. An examination of specific HRM
practices (such as pay for performance, monitoring, and
different hiring criteria) would provide a better
understanding of the dynamics involved with the learning
disabilities.
While Hofstede's national culture values are the only
measures that have anything approaching wide spread
acceptance in the literature, they have also been the target
of a great deal of criticism (Dorfman & Howell, 1991;
Roberts & Boyacigiller, 1984). A more theoretically and
methodologically rigorous set of measures would provide more
confidence in any demonstrated relationships.
In summary, the relationships between the learning
disabilities and any set of the antecedent variables could
be examined in much finer detail.
1 7 8
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Broaden the Sample
While this study is based on the largest empirical
sample that I am aware of, it could still be improved. One
problem is that almost 70 percent of the organizations
involved were from the United States and Canada. Greater
effort should be made to expand the number of firms from
outside North America. The sample is particularly weak in
Africa and' South America.
Explore the Unanticipated Findings in Greater Detail
Several of the unanticipated findings merit further
analysis. The learning disability competency traps has a
significant amount of theoretical support behind it, yet the
measures for it in this study proved unacceptable. Better
measures could be developed and the relationships with the
antecedents and outcomes tested.
The results indicating that there is no relationship
between simplemindedness and the outcomes are particularly
noteworthy. This finding is controversial as it contradicts
much of the popular wisdom, and its implications are rather
profound. For both these reasons it is important to
reexamine this finding with different samples.
Finally, since this is the first empirical study of its
kind, all conclusions must be treated as preliminary.
179
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Additional research is needed to confirm the relationships
and further refine them.
1 8 0
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
REFERENCES
Adler, N. 1990. International dimensions of organizational
behavior. PWS-Kent Publishing: Boston, MA.
Adler, P. 1990. Shared learning. Management Science,
36(8): 938-957.
Anderson, J. 1987. An approach for confirmatory
measurement and structural equation modeling of
organizational properties. Management Science. 33(4): 525-
541.
Argyris, C. 1990. Overcoming organizational defenses:
facilitating organizational learning. Needham Heights, MA:
Allyn and Bacon.
Argyris, C., 1992. On organizational learning. Blackwell
Publishers: Cambridge, MA.
Argyris, C., & Schon, D.A. 1978. Organizational learning:
a theory of action perspective. Reading, MA: Addison-Wesley
Bagozzi, R., Yi, Y., & Philips, L. 1991. Assessing
construct validity in organizational research.
Administrative Science Quarterly. 36:421-458.
Bartlett, C. & Goshal, S. 1987. Managing across borders:
New strategic requirements. Sloan Management Review.
28 (4): 7-17.
Bateson, G. 1972. Steps to an ecology of mind. New York:
Ballentine Books.
Bennis, W., & Slater, P. 1969. The temporary society. New
York: Harper Press.
Bennis, Warren and Burt Nanus. 1985. Leaders: the
strategies for taking charge. New York: Harper and Row.
Bentler, P. 1990. Comparitive fit indexes in structural
equation modeling. Psychological Bulletin, 107(2): 256-259.
Bentler, P., & Bonett, D 1980. Siggnificance tests and
goodness of fit in the analysis of covarience structures.
Psychological Bulletin, 88(4): 588-606.
181
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Block, P. 1988. The empowered manager. San Francisco:
Jossey-Bass Publishers
Bollen, K. 1990. Overall fit in covariance structure
models: Two types of sample size effects. Psychological
Bulletin, 107(2): 256-259.
Brockbank, W., Ulrich, D., & Yeung, A. 1989 Human resource
practices and their strategic context. Working Paper.
Bruner, J.S., Goodnow, J.J., & Austin, G.A. 1956. A study
of thinking. New York: Wiley
Brunsson, N. 1985. The irrational organization. New York:
Wiley.
Burke, M., Borucki, C., & Hurley, A. 1992.
Reconceptualizing psychological climate in a retail service
environment: A multiple stakeholder perspective. Journal of
Applied Psychology, 77(5): 717-729.
Cangelosi, V.E., & Dill, W.R. 1965. Organizational
learning: observations toward a theory. Administrative
Sciences Quarterly, 10: 175-203.
Conger, Jay. 1989. The charismatic leader: behind the
mystique of exceptional leadership. San Francisco: Jossey-
Bass .
Conger, J, R. Kanungo and Associates (eds.). 1988.
Charismatic leadership: The elusive factors in
organizational effectiveness” San Francisco: Jossey-Bass.
Crossan, M., 1991. Organizational learning: A socio-
cognitive model of strategic management. Unpublished
doctoral dissertation, University of Western Ontario.
Cummings, T.G., & Huse, E. 1989. Organization development
and change, fourth edition. St. Paul: West Publishers.
Cyert, R. & March, J. 1963. A behavioral theory of the
firm. New Jersey: Englewood Cliffs.
Daft, R., Bettenhausen, K., & Tyler, B. 1993. Implications
of top managers communication choices for strategic
decisions. In G. Huber & W. Glick (Eds.) Organizational
change and redesigh: Ideas and insights for improving
performance, New York: Oxford University Press.
182
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Daft, R.L. & Huber, G.P. 1987. How organizations learn: A
communication frameework. Research in the sociology of
organizations. 5: 1-36.
Davis, L. & Taylor, J. 1979. Design of jobs. Santa Monica:
Goodyear Press.
De Geus, A. 1988. Planning as learning. Harvard Business
Review. March-April: 70-74.
Denison, D. 1990. Corporate culture and organizational
effectiveness. New York: John Wiley and Sons.
Dess, G. & Robinson, R. 1984. Measuring organizational
performance in the absence of objective measures: The case
of the privately-held firm and conglomerate business unit.
Strategic Management Journal. 5:265-273.
Dewey, J. 1933. How we think. Boston: D.C. Heath and
Company.
Dorfman, P. and Howell, J. 1988. "Dimensions of national
culture and effective leadership patterns: Hofstede
revisited." Advances in International Comparative
Management Vol 3: JAI Press.
Drucker, P. 1993. The post-capitalist society. New York:
Harper Business.
Dunbar, R.L., Dutton, J.M., & Torbert, W.R. 1982. Crossing
mother: ideological constraints on organizational
improvements. Journal of Management Studies, 19: 91-108.
Duncan, R. & Weiss, A. 1979. Organizational learning:
implications for organizational design. In L.L. Cummings &
Barry Staw [Eds.], Research in Organizational Behavior, 1:
75-123. Greenwich, Conn.: JAI Press.
Dutton, J.E., Fahey, L., & Narayanan, V.K. 1983. Toward
understanding strategic issue diagnosis. Strategic
Management Journal, 4; 317-323.
Eisenhardt, K.M. 1989. Making fast strategic decisions in
high velocity environments. Academy of Management Review,
32: 543-576.
Emery, F., & Trist, E. 1973. Towards a social ecology. New
York: Plenum.
183
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Fiol, M.C., & Lyles, M.A. 1985. Organizational learning.
Academy of Management Review, 10: 803-813.
Galbraith, J. 1977. Organization design. Reading, MA:
Addison-Wesley Publishing Company.
Galbraith, J. 1989. "Japanese transplants." CEO
Publication C 90-20 (182). University of Southern
California.
Garvin, D.A., 1993.- "Building a learning organization."
Harvard Business Review, July/August.
Geertz, C. 1973. The interpretation of cultures. New
York: Basic Books, Inc.
Gudykunst, W., Yang, S., & Nishida, T. 1987. ■ Cultural
differences in self-consciousness and self-monitoring.
Communication Research, 14(1): 7-34.
Gupta, P., Dirsmith, M., & Fogarty, T. 1994. Coordination
and control in a government agency: Contingency and
institutional theory perspectives on GAO audits. Academy of
Management Journal, 39(2): 264-284.
Hambrick, D. & Brandon, G. 1988. Executive values. In D.
Hambrick (Ed.), The executive effect: concepts and methods
for studying top managers. Greenwich, CT: JAI Press.
Hayduk, L. 1987. Structural equation modeling with LISREL:
Essentials and applications. Baltimore: The Johns Hopkins
University Press.
Hedberg, B.L. 1981. How organizations learn and unlearn.
In Paul Nystrom & William Starbuck [Eds.], Handbook of
organization design. 1: 3-27 Oxford: Oxford University
Press.
Hedberg, B.L., Nystrom, P.C., & Starbuck, W.H. 1976.
Camping on seesaws: prescriptions for a self-designing
organization. Administrative Science Quarterly, 21: 41-65.
Heilman, K. 1985. Clinical neuropsychology. New York:
Oxford University Press.
Herting, J. & Costner, H. 1985. Respecification in
multiple indicator models. In H. Blalock (Ed.) Causal
Models in the Social Sciences. Hawthorn, NY: Aldine de
Gruyter.
184
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Hickson, D.J. 1987. Decision-making at the top of
organizations. Annual Review of Sociology, 13: 165-192.
Hofstede, G. 1980a. Culture's consequences: International
differences in work related values. Beverly Hills: Sage
Publications.
Hofstede, G. 1980b. Motivation, leadership, and
organization: Do American theories apply abroad?
Organizational Dynamics, Summer: 42-68.
Hofstede, G. 1984. The cultural relativity of
organizational practices and theory. Journal of
International Business Studies, Fall: 75-90.
Hofstede, G. 1986. The cultural relativity of the quality
of life concept. Academy of Management Review, 9(3): 389-
398.
Hofstede, Geert and Michael H. Bond. 1988. The Confucius
connection: From cultural roots to economic growth.
Organizational Dynamics, 5-21.
Hofstede, G. 1991. Cultures and organizations: software of
the mind. London: McGraw-Hill.
Hofstede, G., B. Neuijen, D. Ohayv, and G. Sanders. 1990.
Measuring Organizational Cultures: A Qualitative and
Quantitative Study across Twenty Cases. Administrative
Science Quarterly, 35: 286-316.
Hoppe, M. 1990. A Replication of Hofstede's Measures.
Unpublished Doctoral Dissertation, University of North
Carolina, Chapel Hill.
Hornstein, H. 1986. Managerial courage. New York: John
Wiley.
Huber, G.P. 1991. Organizational learning: The
contributing processes and the literature's. Organizational
Science, 2: 88-115.
Hunt, J., 1981. Applying American behavioral science: Some
cross cultural problems. Organizational Dynamics, Summer:
55-62.
185
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
James, L. 1982. Aggregation bias in estimates of
perceptual agreement. Journal of Applied Psychology. 62(2):
219-229.
James, L., Mulaik, S., & Brett, M., 1982. Causal analysis:
Assumptions, models, and data. Beverley Kills, CA: Sage.
Jeros, S., Jermier, J., Koehler, J., & Sincich, T. 1994.
Effects of continuance, affective, and moral commitment on
the withdraw process: An evaluation of eight structural
equation models. Academy of Management Journal, 36(5): 951-
995.
Johnson, R. & Wichern, D. 1988. Applied multivariate
statistical analysis. Englewood Cliffs, NJ: Prentice Hall.
Joreskog, K. &.Sorbom, D., 1989. LISREL 7: A guide to the
program and applications. ed. Chicago, IL: SPSS, Inc.
Kahneman, D., Slovik, P., Tversky, A. 1982. Judgment under
uncertainty: heuristics and biases. Cambridge: Cambridge
University Press.
Keaia, B. & Bhagat, R. 1988. Cultural constraints on
transfer of technology across nations: Implications for
research in international and comparative management.
Academy of Management Review. 13(4): 559-571.
Kerr, S. 1988. "Some characteristics and consequences of
organizational reward." In D. Schoorman and B. Schneider
(eds.) Facilitating Work Effectiveness Lexington.
Kogut, B. & Singh, H. 1988. The effect of national culture
on choice of entry mode. Journal of International Business
Studies. 19: 411-430.
Koter, J. 1985. The leadership factor. New York: The Free
Press.
Kumar, N., Stern, L., & Anderson, J. 1993. Conducting
interorganizational research using key informants. Academy
of Management Journal. 36(6): 1633-1651.
Leonard-Burton, D. 1993. The factory as a learning
laboratory. Sloan Management Review. Fall: 23-38.
Levitt, B. & March, J.G. 1988. Organizational learning.
Annual Review of Sociology, 14: 319-340.
186
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Lyles, M. 1988. Learning among joint venture sophisticated
firms. Management International Review. 28 (special
issue): 85-98.
March, J.G. 1991. Exploration and exploitation in
organizational learning. Organization Science, 2: 71-88.
March, J.G., & Olsen, J.P. 1975. The uncertainty of the
past: organizational learning under ambiguity. European
Journal of Political Research, 3: 147-171.
March, J. G. & Simon, H.A. 1958. Organizations. New York:
John Wiley & Sons.
Marsh, H. 1987. The hierarchical structure of self-concept
and the application of hierarchical confirmatory factor
analysis. Journal of Education Management, 24: 17-40.
Marsh, H., & Hocever, D. 1985. Application of confirmatory
factor analysis to the study of self-concept: First- and
higher-order factor models and their invariance across
groups. Psychological Bulletin, 97(4): 562-582.
McDonald, R. & Marsh, H. 1990. Choosing a multivariate
model: Noncentrality and goodness of fit. Psychological
Bulletin, 107(2): 256-259.
Marsick, J. & Watkins, K. 1990. Informal and incidental
learning in the workplace. London: Routledge.
Miles, R. 1982. Coffin nails and corporate strategies.
Englewood Cliffs: Prentice Hall, Inc.
Mintzberg, H. 1979. The structuring of organizationa: A
synthesis of research. Englewood Cliffs, NJ: Prentice-Hall
Mintzberg, H., Raisinghani, D. & Theoret, A. 1976. The
structure of unstructured decision processes.
Administrative Science Quarterly, 21: 246-275.
Mitroff, I.I. & Betz, F. 1972. Dialectical decision
theory: A meta-theory of decision-making. Management
Science, 19: 11-24.
Mohrman, S.A. & Cummings, T.G. 1989. Self-designing
organizations. Reading, MA: Addison Wesley Publishing
Company.■
187
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Morgan, G. 1986. Images of organization. Newbury Park,
CA: Sage Press.
Myers, R. 1986. Classical modern regression with
applications. Boston: PWS-Kent.
Nadler, G., Gerstein, M., Shaw, R., & Associates. 1992.
Organizational architecture: Designs for changing
organizations. San Francisco, Jossey Bass.
Nadler, D. & Tushman, M. 1989. Leadership for
organizational change. In Mohrman, A., Mohrman, S.,
Ledford, G., Cummings, T., Lawler, E., & Associates (eds.),
Large Scale Organizational Change. San Francisco: Jossey-
Bass.
Neter, D., Wasserman, W., & Kunter, M. 1985. Applied
linear statistical models: Regression, analysis of varience,
and experimental designs. Homewood, IL: Irwin.
Niehoff, B. & Moorman, R. 1993. Justice as a mediator of
the relationahip between methods of monitoring and
organizational citizenship behavior. Academy of Management
Journal 36(3): 527-556.
Nisbett, A. & Ross, J. 1980. Human inference: strategies
and shortcomings of social judgment. Englewood Cliffs:
Prentice-Hall
Nonaka, I. 1988. Creating organizational order out of
chaos: Self-renewal in japanese firms. California
Management Review. Spring: 57-73.
Nonaka, I. 1990. Redundant, overlapping organization: A
japanese approach to managing the innovation process.
California Management Review. Spring: 27-38.
Nonaka, I. 1991. The knowledge-creating company. Harvard
Business Review. Nov-Dee: 96-104.
Ouchi, W.G. 1980. Markets, bureaucracies, and clans.
Administrative Science Quarterly, 25: 129-141.
Parkhe, A. 1991. Interfirm diversity, organizational
learning, and longevity in global strategic alliances.
Journal of International Business Studies. 22(44): 579-601.
Peters, T. 1987. Thriving on chaos. New York: Knoph.
188
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Peters, T. 1992. Liberation Management. New York: Knoph.
Pfeffer, J. 1981. Power in organizations. Cambridge, MA:
Ballinger Publishing Company.
Porter, M. 1990. The competitive advantage of nations.
New York: The Free Press.
Prahalad, C. & Hamel, G., 1990. The core competencies of
the organization." Harvard Business Review, 68: 79-91
Pucik, V. 1988. Strategic alliances, organizational
learning, and competitive advantage: The HRM agenda. Human
Resource Management. 27(1): 77-93
Quinn, R. 1988. • • Beyond rational management: Mastering the
paradoxes and competing demands of high performance. San
Francisco: Jossey-Bass Publishers.
Quinn, R. 1992. Intelligent enterprise. New York: The
Free Press.
Quinn, R. and Rohbraugh, J. 1983. A spatial model of
effectiveness criteria: Towards a competing values approach
to organizational analysis. Management Science, 29: 363-
377.
Quinn, R. and McGrath, M. 1985. The transformation of
organizational cultures: A competing values perspective. In
P. Frost , et al. (eds.) Organizational culture. Beverley
Hills: Sage.
Quinn, R. and Spreitzer, G. 1991. The psychometrics of the
competing values culture instrument and an analysis of the
impact of organizational culture on the quality of life. In
R. Woodman and W. Pasmore (eds.) Research in organizational
change and development. Greenwich, CT: JAI Press.
Roberts, K. and Boyacigiller, N. 1984. "Cross-national
organizational research: The grasp of the blind men."
Research in Organizational Behavior, 6: 423-475
Rousseau, D. 1985. Issues of level in organizational
research: Multi-level and cross-level perspectives.
Research in Organizational Behavior, 7: 1-37.
Sashkin, M. 1988. The visionary leader. In Conger, J., &
Kanungo, R. (eds.), Charismatic leadership: The elusive
189
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
factor in organizational effectiveness. San Francisco:
Jossey-Bass.
Schein, E. 1985a. Organizational culture and leadership,
San Francisco: Jossey-Bass Publishers.
Schein, E. 1985b. How culture forms, develops, and
changes, in Kilman, Ralph H. et al., eds. Gaining Control of
the Corporate Culture, London: Jossey-Bass Publishers.
Schein, E., 1988. Organizational culture and leadership: A
dynamic view, San Francisco, Jossey Bass.
Schon, D.A. 1983. The reflective practitioner. San
Francisco: Jossey-Bass.
Seidler, J. 1974. On using informants: A technique for
collecting quantitative data and controlling for measurement
error in organizational analysis. American Sociologic
Review. 39: 816-831.
Senge, P.M. 1990. The fifth discipline. New York:
Doubleday.
Shane, S., Venkatraman, S., & Macmillan, I., 1993.
Cultural differences in innovation championing strategies.
Paper presented at the 1993 Academy of International
Business Conference, October 21-24, Maui, Hawaii.
Shrivastava, P. 1983. A typology of organizational
learning systems. Journal of Management Studies, 20: 7-28.
Simonin, B. & Helleloid, D. 1993a. Do organizations learn?
An empirical test of organizational learning in strategic
alliances. 1993 Academy of Management Best Paper
Proceedings.
Simonin, B. & Helleloid, D. 1993b. An empirical
investigation of double loop learning in international
collaborators. Paper presented at the 1993 Academy of
International Business Conference, October 21-24, Maui,
Hawaii.
Sitkin, S. 1992. Learning through failure: the strategy of
small losses. Research in Organizational Behavior, 14: 231-
266.
.
190
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Slocum, J. & Dilloway, C. 1991 Discussion goup report: The
learning organization. Working paper, Southern Methodist
University, School of Management.
Smircich, L. 1983. Concepts of culture and organizational
analysis. Administrative Science Quarterly, 28:339-358.
Smircich, L. & Stubbart, C. 1985. Strategic management in
an enacted world. Academy of Management Review, 10: 724-
736.
Smith, P., and Peterson. 1988. Leadership, organizations,
and culture, London: Sage.
Snyder, W. & Cummings, T. 1992 Organizational learning
disabilities: Guidelines for organizational development
interventions. Presented at The Western Academy of
Management Meetings, Spokane.
Spreitzer, G. 1992. When organizational dare: The Dynamics
of individual empowerment in the workplace. Unpublished
Doctoral Dissertation
Stata, R. 1989. Organizational learning: The key to
management innovation. Sloan Management Review, Spring: 63-
74.
Staw, B.M. & Ross, J. 1978. Commitment to a policy
decision: a multi-theoretical perspective. Administrative
Science Quarterly, 23: 40-64.
Sullivan, J. 1985. Multiple indicators and complex causal
models. In H. Blalock (Ed.) Causal Models in the Social
Sciences♦ Hawthorn, NY: Aldine de Gruyter.
Sullivan, J. and Nonaka, I. 198 6. The application of
organizational learning theory to Japanese and American
management. Journal of International Business Studies. 4:
127-147.
Tichy, Noel and M. Devanna. 198 6. The transformational
leader. New York: Wiley.
Thompson, J. 1967. Organizations in action. New York: Me
Graw-Hill.
Ulrich, D. 1992. Organizational capability: Competitive
advantage Through human resources." 15:1-15.
191
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Ulrich, D. 1992. Customer management organization: The
advent of the management pioneer. Working paper. 1-31.
Ulrich, D., Halbrook, R., Meder, D., Stuchlik, M., & Thorpe,
S., 1991. Employee and customer attachment: Synergies for
competitive advantage. Human Resource Planning. 14(2): 89-
102. '
Ulrich, D. & Lake, D. 1991. Organizational capability:
Creating competitive Advantage. Academy of Management
Executive. 5(1):77-91.
Ulrich, D. & Lake, D., 1990. Organizational capability:
Competing from the inside/out. New York: Wiley.
Ulrich, D., Yeung, A., & Brockbank, W., 1992 Human
resources in the 1990's: Forging personal competencies into
functional capability. Working Paper.
Ulrich, D., Von Glinow, M.A., Jick, T. Yeung, A., Nason, S.,
1993. Learning organization, culture change, and
competitiveness: How managers can build learning capability.
Manuscript presented to The International Consortium for
Executive development and Research (ICEDR).
Venkatraman, N. & Ramunujam, V., 1987. Measurement of
business economic performance: An examination of method
convergence. Journal of Management. 13: 109-122.
Walsh, J.P. & Ungson, G.R. 1991. Organizational Memory.
Academy of Management Review. 16: 57-91.
Wheaton, B., Muthen, B., Alwin, D., & Summers, G. 1977.
Assessing reliability and stability in panel models. In D.
Heise (Ed.), Sociological Methodology. San Francisco:
Jossey-Bass.
Weick, K.E. 1979. The social psychology of organizing.
New York: Random House.
Westley, Frances and Henry Mintzberg. 1989. Visionary
leadership and strategic management. Strategic Management
Journal 10: 17-32.
Wilkins, A. and W. Dyer Jr. 1988. Toward culturally
sensitive theories of cultural change. Academy of
Management Review 13(4): 522-533.
192
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
APPENDIX 1
QUESTIONAIRE ITEMS
Discovery
Blindness
We are good at scanning the external environment for
opportunities and potential problems.
We are active in finding/creating new markets.
We are good at perceiving long term environmental threats,
opportunities.
Internal Blindness
We are good at scanning the internal workings and processes
of this organization for areas which may be improved upon.
We work to insure that all employees can recognize or
discover the information that they need to perform their
job.
Competency Traps
In general, our work procedures have functioned well sc we
see little need of pervasive change.
"If it ain't broke, don't fix it" would represent the
general attitude here pretty well.
We intentionally create small "false alarms," "crises," etc.
in the work place to stimulate improvements or problem
solving.
Projection
Political considerations of important stakeholders influence
the actions of our organization.
We try to uncover and communicate all relevant facts, not
just those that are politically acceptable.
We work to insure that employees are directly exposed to the
variation and complexity of the environment.
1 9 3
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Invention
Simplemindedness
We are good at determining the causes of unexpectedly low
and high levels of performance.
We are good at deciding on plans of action to redress
performance shortfalls.
We are good at generating and evaluating a wide variety of
alternative solutions to performance shortfalls.
Requisite Variety
We try to insure that all employees have access to more
information than the minimum required to perform their job.
We are good at generating and evaluating a wide variety of
alternative solutions to performance shortfalls.
We encourage diversity in people and ideas within our
organization.
Intelligent risk taking which results in failure is not
punished.
S chi z ophreni a
There is a lack of coordination between multiple, competing
perspectives at this organization.
Groups and Departments in this organization function fairly
independently without much integration.
Production
Paralysis
We tend to over analyze problems before implementing ideas—
and thus miss opportunities for improvement.
Employees have a significant say in how their work is done.
We have trouble implementing new procedures.
194
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Hyperactivi ty
We seem to always be implementing new changes and technology
without giving serious thought to the implications of each
change.
New procedures come and go without making much of a
difference in the organization.
Alien Hand
We have had trouble insuring that our actions are consistent
with organizational goals.
In this firm once a given course of action gains momentum it
is difficult to change direction.
We often make the same mistakes over and over.
Generalization
Amnesia
Good ideas tend to disappear if not in regular use.
Our experiences are not documented or encoded in procedures
here.
Superstitious Learning
The rational we give for our performance outcomes often has
little to do with the actual causes of these outcomes.
The actions we take to impact performance often have little
effect on the performance outcomes.
We work to continually decrease the time between employee
actions and feedback on those actions.
Diffusion Deficit
Learning in one group or unit is not spread throughout the
organization.
Even though individuals here learn, this learning is not
transmitted up to the organizational level.
195
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Appendix 2:
Distribution of Respondents by Geography
REGION FREQUENCY
North American 1014
Latin America 21
Western Europe 85
Eastern Europe 9
Middle East 5
Asia 200
Africa 4
Australia/New Zealand 19
TOTAL RESPONSES 1359
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Appendix 2:
Distribution of Sample by Years of Business Establishment
0 -1 0 years
90
11 - 20 years
118
21 - 20 years 102
31 - 40 years 103
41 - 50 years 104
51 - 60 years 87
61 - 70 years 91
71 - 80 years 85
81 - 90 years 102
91 - 100 years 133
101 - 150 years 247
over 150 years 54
missing 43
TOTAL 1359
MEAN AGE OF FIRM 69 years
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Appendix 2:
Distribution of Sample by Size
(Number of Employees)
1 - 1000
283
1001 - 2000 110
2001 - 4000 129
4001 - 6000 105
6001 - 8000 53
8001 - 10000 59
10001 - 15000 72
15001 - 25000 94
25001 - 35000 94
35001 - 50000 93
50001 - 100000 116
over 100000 119
missing 32
TOTAL RESPONDENTS 1359
MEAN FIRM SIZE 40,738 employees
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Appendix 2:
Distribution of Sample b;y Industry
Miscellaneous manufacturing
298
Petroleum, gas, coal
39
Chemical, pharmacy 127
Electronics, computer 144
Automobiles 36
Aircraft 17
Utilities 145
Wholesaling, retailing 111
Finance 185
Services 167
Missing 90
TOTAL RESPONDENTS . 1359
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Appendix 2:
Distribution by Age of Respondent
20 - 30 years old
29
31-4 0
282
41 -50
706
51 - 60
296
over 60 25
missing 21
TOTAL 1359
MEAN AGE OF RESPONDENT 45.7 years old
Appendix 2:
Distribution of Tenure of Respondents
less than 1 year 69
2 - 5 years 195
6-10 years 219
11 - 15 years 229
16 - 20 years 223
21 - 25 years 186
over 25 years 238
TOTAL 1359
MEAN TENURE OF RESPONDENT 15.8 years
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Appendix 2:
Distribution of Role of Respondents
Individual contributor
137
Manager of individual contributor
290
Director of managers
529
General manager
378
Missing
25
TOTAL 1359
Appendix 2:
Distribution by Functional Specialty of Respondent
Finance/accounting 89
General management 272
HR/Personnel 473
Manufacturing/Production 110
Mirketing/Sales 156
Planning 48
Research & Development 57
Others 103
Missing 51
TOTAL 1359
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
1
Appendix 3: Number of Respondents by ICEDR Firm Membership
CORPORATE MEMBERSHIP RESPONSES
(by 3/30/93)
Alcatel 7
AT&T 19
Burroughs Wellcome 3
Daimler-Benz 0
Eastman Kodak 17
Eisai Company 5
Exxon 9
FIAT Group 1
Fidelity Investments 2
Ford Motor Company 18
Heineken International 3
Hewlett Packard 2
Hoffmann-La Roche 4
IBM Corporation 9
L’OREAL 2
Merck & Company 4
National Australia Bank 8
National Westminster Bank 3
Nippon Mining Company 1
Philip Morris Companies 4
Ssangyong Business Group 20
Westpac Banking Corporation 6
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Appenxix 3: Businesses Participating in Research
(382 total companies)
ABB Power T&D Company Inc.
Abbot Laboratories
Aetna Institute for Corporate Education
Agfa-Gevaert
AGT
Alcalis de Colombia
Alcatel
ALCO Aluminio S. A.
ALCO Health Services Corp.
ALCOA
Alexander Hamilton Life Insurance
Algonquin Gas Transmission Company
Allen Group PR
Allied-Signal
Allstate Insurance
Altron Inc.
Amdahl Company
American Express
American Capital Management & Research
American President Companies Ltd.
Amoco Chemical Company
AMP Incorporated
Amperex Electronic Corporation
Analog Devices
Analysis & Technology
Anderson Consulting
Applied Power
ARA Holding
Arrow-Molded Plastics
Asea Brown Boveri
Asoc Venexolano-Americana de Amistad
ASTRA-CALVE
AT&T
Atlas Consolidated Mining Development
Austrian Industries
Avco Lycoming/Textron
Avon Products
Banc Consolidado
Bangkok Weaving Mills
Bangkok Bank
Bank of Yokohama
Bank of Boston
Bank of Thailand
Bank of Montreal
Bank of Ireland
Barclays Bank PLC
Battelle
Bausch & Lomb
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Baxter Healthcare
Bell Labs
Bell South
Ben & Jerry’s Homemade
Best Western International
Bethlehem Steel
Beverly Enterprises
Biologic Systems
BMW
Boeing
Boise Cascade
Borden
Bose
Brecker & Merry man
Bristol Myers
Bull
Bundy
Burlington Northern Railroad
Burroughs Wellcome
Butterworth Hospital
C. Iioh
Calgon Vestal Laboratories
Camp Dresser & McKee
Canada Post
Canadian General Insurance Group
Canadian Utilities Limited
Cargill
Carnation Company
Caterpillar Inc.
Cathay Pacific Airways
Central Life Assurance
Central Hudson Gas Electric
Centro Analisi Strategiche
Champion International
Chartered Industries of Singapore
Chase Manhattan
China Steel
Chubb & Sons
Cigarette Manufacturers
CIGNA
Citibank
Citizens Gas and Coke Utility
Citytrust Banking
Cobe Laboratories
Coca Cola
Colgate Palmolive
Comerica Inc.
Consumers Power
Continental AG/Werk Limmer
Coors Brewing
Coraza Corporation
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Coming
Coulter Electronics
CPC International
Cyclops
Cyprus Minerals
D.C. Health
DANA Corporation
Dayton Hudson
DDB Needham Worldwide
Deere & Company
Delta
Den Danske
Dharmala Group
Diamond Shamrock
Digital Equipment
Dow Chemical
Dragados Y Construcciones
Dresser Industries
Dresser-Rand
Du Pont
Dim & Bradstreet
Duracell International
Eastman Kodak
Eisai
ELEC-TEC
Eli Lilly
Employee Assistance
Enron
Equitable Financial
Ernst & Whitney
Executive Forum
Exxon
Farm Credit Corporation of America
Farm Credit Bank
Federal Express
Federal Reserve Bank of Philadelphia
Federal Sign
Federal Mogul
Fiat
Fidelity Investments
Finance Corporation of Thailand
Firestone Tire
First of America, Southeast Michigan
First Source Bank
Fleming Companies
FMC
Ford
Foremost
Frequency & Time Systems
Gen-Probe
Gencorp Polymer Products
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
General Motors
General Electric
GENMIN
Genuine Parts
Gnaden Huetten Memorial Hospital
Goodyear Tire & Rubber
GrowMark
Grupo ICI de Mexico
Grupo Cere S.A. de C.V.
Grupo Bufete Industrial
GTE
Hallmark Cards
Hamilton Standard
Harley-Davidson
Haworth
Heineken International
Herman Miller
Hewlett Packard •
Hitachi Ltd.
Hoechest Celanese Corporation
Hoechst Japan
Hofftnann-La Roche
Holiday Inn
Holiday Corporation
Honeywell
Hoteles Quinta Real
Hungarian Telecommunications
Hussey Seating Company
Hyde Group
Hyundai Business Group
ICI (Malaysia) Holding
IDS Financial Services
Imatran Volma Oy
Inchcape Pacific
Indal Limited
Industrial Bank of Japan
Ingersoll-Rand
Institute Manajemen Prasetiya Mulya
Institute for International Economics
Intermedics
Intermountain Health Care
Internal Revenue Service
International Paper
International Business Machines
International Search Associates
Japan External Trade Organization
John Deere
Johnson & Johnson
Jossey-Bass
Joyo Bank
Kaiser Aluminum & Chemical
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Kansai Economic Federation
Kellogg Company
Korea Computer
Korea Stock Exchange
Kraft General Foods
Kroger
L J. Minor Corporation
L’Oreal
Lafarge Canada
Lancaster General Hospital
Levi Strauss
Lcctite Corporation
Long Island Lighting
Loral Infrared & Imaging
Lotte Group
Lufkin Industries
Mahanager Tel. Nigam
Manila Electric
Manulife Financial
Manville Institute
Marathon Oil
Marion Merrell Down
Marriott
Masco
McDonnell Douglas
MCI
Mead
Merck
Merck, Sharpe & Dohme
Metropolitan Life
Mexinox
Michcon
Milk Marketing
Mitsubishi
Mobil
Monsanto
Morgan Guaranty Trust
Motorola
NAC Reinsurance
Nacronal Finsnclierm
NASA
National Australia Bank
National Westminster
NCR
Nebraska Public Power
NEC
Neste OY
New Jersey Bell
Newport News Shipbuilding
Nihon Keizai Shimbun
Nikko Petrochemicals
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Nissan Motors
Nitto Agency
Nomura
Nordata Vestadata
Norrell
North Carolina Memorial Hospital
Northern Telecom
Northwest Natural Gas
Norton Company
Nova Corporation
Nuclear Fuel Industries
Nuclear Support Services
Nutrasweet
Nynex
NYSEG . . .
NZR
Ocean Spray Cranberries
ODR
Oki Electric
Omron
Outboard Marine Corporation
Paccar Incorporated
Pacific Gas & Electric
Pacific Bell
Parke-Davis
Payless Shoes
PD SoedarpoPepsi Cola
Perini
Permodalan Naxional Berhad
Pfizer
PhilAm Life Insurance
Philip Morris
Philippine Long Distance Telephone
Pinkerton Tobacco
Pitney Bowes
Pittsburgh National Bank
Potomac Electric Power
PPG Industries
Price Waterhouse
Prince
Promus
PT Swadhama Suta Data
PT Pembangunan Jaya
PT Hasfarm Dian Konsultan
PT Astra International
PT Humpuss
PT Muladaya Adipratama
PT Caltex Pacific Indonesia
PT Federal Motor
Public Service Company
Rhodia
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Rolls Royce PLC
Royal Bank
Ryder System
Safety Harbor Spa & Fitness Center
San Miguel
Sann Aung Imaing
Saudi Arabian Oil
Security Pacific Company
Seiko Epson
Serigraph
Shell
Shimizu
Shin Caterpillar Mitsubishi
Sibson
Siemens
Signetics
Silicon graphics
Singapore Airline
Skandia Insurance
Sony Corporation
Southam
Southern California Edison
Southern Indiana Gas & Electric
Spandoz
Sprague Semiconductor
SPX Group
Srithai Marketing
SSangyyong Group of Companies
Sterile Design
Sterling Jersey
Stolle Corporation
Stratus Computer
Sumitomo
Supermarkets General
SWIFTSA
Swiss Bank
Taiwan Power
Taiwan Cement
Tandem Computers
Tenneco
Texaco
Thomaston Mills
TN Soon & Company
Tokai Lippo Bank
Tokyo Electric Power
Toppan Printing
Toronto Dominion Bank
Toyota Motor
Trak International
Travelers Express
TRW
209
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Unibanco-Uniao De Bancos Btasileiros
Union Carbide
Unisys
United Telephone Company
United Brands Company
United Microlectronics
United Parcel Service
United States Army
UNUM Corporation
Upjohn
US Postal Service
USI/IBM Indonesia
Valmet
Volkswagen
Vulcan Material
Wang Laboratories
Wargo & Company
Warner-Lambert
Welch’s
Wellcome Foundation
Westinghouse
Westpac Banking
Weyerhaeuser
Whaman
Whirlpool
Yukong
Yunbo Industrial
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
A p p e n d ix 4
I t e m C o r r e l a t i o n s
DIS.01 DIS.02 DIS.03 DIS.04 DIS.05 DIS.O
DIS.01 1.0000 .4535** .5099** .2884** .3508** .0740
DIS.02 .4535** 1.0000 .3360** .1972** .3044** -.0049
DIS.03 .5099** .3360** 1.0000 .3404** .2741** .0787
DIS.04 .2884** .1972** .3404** 1.0000 .5084** .1987*
DIS.05 .3508** .3044** .2741** .5084** 1.0000 .0958
DIS.06 .0740 -.0049 .0787 .1987** .0958 1.0000
DIS.07 .2986** .2318** .2631** .3615** .3678** .4810*
DIS.08 .0501 -.1291* .0774 .0881 .0699 -.0053
DIS.09 .2190** .2441** .2091** .2745** .3647** .0354
DIS.10 .3201** .1474* .2549** .3759** .4728** .1805*
DIS.11 .4175** .1873** .3374** .3844** .5467** .1536*
DIS.12 .3400** .3132** .3018** .3858** .4116** .0354
DIS.13 .3540** - .3073** .2721** .4342** .3882** .0203
DIS.14 .3324** .2239** .2166** .3823** .6897** .0459
DIS.15 .4310** .3343** .4012** .4259** .4771** .0014
DIS.16 .3166** .3720** .2820** .4298** .5845** .1678*
DIS.17 .2715** .2258** .2500** .4159** .5585** .1591*
DIS.18 .2234** .1364* .2169** .4041** .4295** -.0458
DIS.19 .2162** .1121 .1807** .3751** .4470** .0508
DIS.20 .1735** .1819** .1077 .2813** .2961** .0810
DIS.21 .2342** .1656** .1921** .3355** .5773** .2226*
DIS.22 .3385** .1745** .1722** .2773** .3709** .0779
DIS.23 .2836** .1018 .2229** .2656** .3165** .0050
DIS.24 .2954** .1864** .2122** .3375** .3964** .1423*
DIS.25 .3221** .2436** .3242** .3847** .4597** .0474
DIS.26 .2077** .2676** .1873** .3165** .2947** .1153
DIS.27 .3994** .2457** .3251** .3763** .3860** -.0167
DIS.28 .3378** .3498** .2353** .3076** .4116** .0518
DIS.29 .2477** .1290* .2248** .2763** .3512** -.0184
DIS.30 .3439** .2512** .3118** .4457** .4386** .1248*
DIS.31 .2629** .2086** .2355** .3591** .3458** .1848*
DIS.32 .2371** .2856** .1257* .3710** .5160** .1375*
DIS.33 .3127** .2318** .3188** .4638** .5348** .0549
DIS.34 .3000** .2301** .3409** .4829** .5493** .1101
* - Signif. LE .05 ** - Signif. LE .01 (2-•tailed)
. " is printed if a coefficient cannot be computed
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
- - Correlation Coefficients - -
DIS.07 DIS.08 DIS.09 DIS.10 DIS.ll DIS.12
DIS.01 .2986** -.0501 .2190** .3201** .4175** .3400*
DIS.02 .2318** -.1291* .2441** .1474* .1873** .3132*
DIS.03 .2631** .0774 .2091** .2549** .3374** .3018*
DIS.04 .3615** .0881 .2745** .3759** .3844** .3858*
DIS.05 .3678** .0699 .3647** .4728** .5467** .4116*
DIS.06 .4810** -.0053 .0354 .1805** .1536* .0354
DIS.07 1.0000 .0286 .2310** .3960** .4266** .2812*
DIS.08 .0286 1.0000 .1534* .1348* -.0111 -.0707
DIS.09 .2310** .1534* 1.0000 .3578** .3245** .4209*
DIS.10 .3960** .1348* .3578** 1.0000 .5125** .3725*
DIS.ll .4266** -.0111 .3245** .5125** 1.0000 .4231*
DIS.12 .2812** -.0707 .4209** .3725** .4231** 1.0000
DIS.13 .2776** .0286 .4308** .4125** .4669** .6789*
DIS.14 .3711** -.0410 .3837** .4750** .5869** .4926*
DIS.15 .3131** -.0919 .3531** .3409** .5402** .5568*
DIS.16 .4033** -.0029 .2881** .3987** .4504** .3662*
DIS.17 .3894** .1041 .4687** .4896** .4563** .3217*
DIS.18 .2992** .0673 .4281** .4117** .3244** .3607*
DIS.19 .3255** .0376 .3536** .3949** .4057** .2993*
DIS.20 .3403** .0039 .4437** .3304** .2450** .2811*
DIS.21 .3834** -.0339 .2718** .4272** .5044** .3032*
DIS.22 .3585** .0320 .3427** .3642** .3923** .3603*
DIS.23 .1685** .1231* .3217** .2779** .3339** .3315*
DIS.24 .3747** .1984** .3407** .3920** .3260** .2998*
DIS.25 .3219** .0470 .4045** .3859** .3532** .3985*
DIS.26 .2851** -.0083 .3041** .2704** .2694** .2597*
DIS.27 .3126** .0624 .3949** .3768** .3610** .4281*
DIS.28 .3444** .0052 .3366** .3608** .3227** .3958*
DIS.29 .2207** .1039 .1273* .2309** .2419** .3180*
DIS.30 .3467** .1665** .3692** .5086** .4112** .4893*
DIS.31 .2893** .1510* .3415** .4656** .3677** .4409*
DIS.32 .3740** -.0350 .3547** .3744** .4650** .4398*
DIS.33 .3232** .0864 .3799** .3963** .4942** .4370*
DIS.34 .4066** .0493 .4206** .4533** .5194** .5221*
* - Signif. LE .05
★ ★ _
Signif. LE .01 (2-■tailed)
. " is printed if a coefficient cannot be computed
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
— ■ Correlation Coefficients -
DIS.13 DIS.14 DIS.15 DIS.16 DIS.17 DIS.18
DIS.01 .3540** .3324** .4310** .3166** .2715** .2234+
DIS.02 .3073** .2239** .3343** .3720** .2258** .1364+
DIS.03 .'2721** .2166** .4012** .2820** .2500** .2169*
DIS.04 .4342** .3823** .4259** .4298** .4159** .4041*
DIS.05 .3882** .6897** .4771** .5845** .5585** .4295*
DIS.06 .0203 .0459 .0014 .1678** .1591** -.0458
DIS.07 .2776** .3711** .3131** .4033** .3894** .2992+
DIS.08 .0286 -.0410 -.0919 -.0029 .1041 .0673
DIS.09 .4308** .3837** .3531** .2881** .4687** .4281*
DIS.10 .4125** .4750** .3409** .3987** .4896+* .4117*
DIS.ll .4669** .5869** .5402** .4504** .4563** .3244+
DIS.12 .6789** .4926** .5568** .3662** .3217** .3607*
DIS.13 1.0000 .5060** .6688** .4058** .3801** .4014*
DIS.14 .5060** 1.0000 .5903** .5148** .5423** .3988*
DIS.15 .6688** .5903** 1.0000 .5022** .4054** .4062*
DIS.16 .4058** .5148** .5022** 1.0000 .5530** .4021*
DIS.17 .3801** .5423** .4054** .5530** 1.0000 .3890+
DIS.18 .4014** .3988** .4062** .4021** .3890*+ 1.0000
DIS.19 .3607** .3965** .3295** .4412** .4189*+ .6288*
DIS.20 .4520** .3136** .3553** .3100** .3419+* .4485*
DIS.21 .3231** .5609** .4325** .5708** .5691*+ .2739*
DIS.22 .4034** .3571** .3760** .2833** .3299+* .4159+
DIS.23 .3925** .2761** .3662** .2218*+ .2249** .4036+
DIS.24 .3579** .2924** .3295** .2817++ .3213++ .4225*
DIS.25 .4877** .4179** .4286** .3944+* .3205** .5784*
DIS.26 .2346** .2950** .2722** .3615++ .3807** .3583+
DIS.27 .4301** .4428** .4376** .3513+* .3600+* .5493*
DIS.28 .3459** .3870** .3440** .3929+* .3979** .3982+
DIS.29 .2728** .2440** .3022** .2557** .1821** .3743*
DIS.30 .5445** .4256** .4621** .4499++ .4426++ .4725*
DIS.31 .5501** .3986** .3934** .3656** .3912++ .4161*
DIS.32 .4198** .4505** .4224** .4320** .4148++ .3647*
DIS.33 .4439** .4634** .4774** .4182** .4249** .5400*
DIS.34 .5217** .5226** .4812** .4469** .4394** .5581+
* - Signif. LE .05 +* - signif. LE .01 (2-tailed)
" . " is printed if a coefficient cannot be computed
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
- Correlation Coefficients
DIS.19 DIS.20 DIS.21 DIS.22 DIS.23 DIS.24
DIS.01 .2162*+ .1735++ .2342++ .3385*+ .2836++ .2954++
DIS.02 .1121 .1819++ .1656++ .1745*+ .1018 .1864+*
DIS.03 .1807++ .1077 .1921+* .1722+* .2229++ .2122++
DIS.04 .3751++ .2813++ .3355++ .2773++ .2656++ .3375**
DIS.05 .4470*+ .2961++ .5773++ .3709++ .3165++ .3964+*
DIS.06 .0508 .0810 .2226++ .0779 .0050 .1423+
DIS.07 .3255+* .3403** .3834++ .3585+* .1685*+ .3747++
DIS.08 .0376 .0039 -.0339 .0320 .1231+ .1984+*
DIS.09 .3536*+ .4437++ .2718+* .3427+* .3217++ .3407+*
DIS.10 .3949** .3304+* .4272++ .3642+* .2779** .3920+*
DIS.ll .4057*+ .2450+* .5044*+ .3923++ .3339*+ .3260++
DIS.12 .2993** .2811*+ .3032*+ .3603++ .3315** .2998**
DIS.13 .3607+* .4520** .3231** .4034** .3925++ .3579**
DIS.14 .3965+* .3136++ .5609*+ .3571** .2762++ .2924*+
DIS.15 .3295*+ .3553** .4325** .3760** .3662++ .3295*+
DIS.16 .4412** .3100++ .5708** .2833++ .2218+* .2817*+
DIS.17 .4189++ .3419+* .5691++ .3299*+ .2249*+ .3213++
DIS.18 .6288+* .4485++ .2739+* .4159*+ .4036** .4225+*
DIS.19 1.0000 .4338++ .3602++ .4323*+ .3872** .3852**
DIS.20 .4338+* 1.0000 .2641++ .4729** .3104++ .4324**
DIS.21 .3602++ .2641+* 1.0000 .2299+* .1591** .2099**
DIS.22 .4323*+ .4729** .2299++ 1.0000 .4762+* .4458++
DIS.23 .3872*+ .3104++ .1591++ .4762*+ 1.0000 .5589*+
DIS.24 .3852+* .4324+* .2099+* .4458** .5589** 1.0000
DIS.25 .4339*+ .4832+* .2678*+ .4974++ .4910** .5663**
DIS.26 .2926*+ .3555++ .3272** .2990++ .2422** .2658+*
DIS.27 .4450+* .4667** .2710++ .4693++ .5393** .5523*+
DIS.28 .3950*+ .3511** .3345*+ .4047++ .3696** .4723**
DIS.29 .3102** .2387** .0650 .3324+* .3317** .4023*+
DIS.30 .4277++ .4549*+ .3237++ .4357*+ .4407*+ .5027++
DIS.31 .3567*+ .4739** .2357** .4326++ .4417*+ .5137**
DIS.32 .3273+* .3603++ .3927+* .3654** .2674** .3546**
DIS.33 .5336** .4210** .3701+* .4435+* .3630*+ .4333+*
DIS.34 .4990** .4481** .3955** .4774+* .4253*+ .4992**
* - Signif. LE .05 ** - signif. LE .01 (2-tailed)
" . " is printed if a coefficient cannot be computed
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Correlation Coefficients
DIS.25 DIS.26 DIS.27 DIS.28 DIS.29 DIS.30
DIS.01 .3221** .2077** .3994** .3378** .2477** .3439*
DIS.02 .2436** .2676** .2457** .3498** .1290* .2512*
DIS.03 .3242** .1873** .3251** .2353** .2248** .3118*
DIS.04 .3847** .3165** .3763** .3076** .2763** .4457*
DIS.05 .4597** .2947** .3860** .4116** .3512** .4386*
DIS.06 .0474 .1153 -.0167 .0518 -.0184 .1248*
DIS.07 .3219** .2851** .3126** .3444** .2207** .3467*
DIS.08 .0470 -.0083 .0624 .0052 .1039 .1665*
DIS.09 .4045** .3041** .3949** .3366** .1273* .3692*
DIS.10 .3859** .2704** .3768** .3608** .2309** .5086*
DIS.ll .3532** .2694** .3610** .3227** .2419** .4112*
DIS.12 .3985** .2597** .4281** .3958** .3180** .4893*
DIS.13 .4877** .2346** .4301** .3459** .2728** .5445*
DIS.14 .4179** .2950** .4428** .3870** .2440** .4256*
DIS.15 .4286** .2722** .4376** .3440** .3022** .4621*
DIS.16 .3944** .3615**,, „.3513** .3929** .2557** .4499*
DIS.17 .3205** .3807** .3600** .3979** .1821** .4426*
DIS.18 .5784** .3583** .5493** .3982** .3743** .4725*
DIS.19 .4339** .2926** .4450** .3950** .3102** .4277*
DIS.20 .4832** .3555** .4667** .3511** .2387** .4549*
DIS.21 .2678** .3272** .2710** .3345** .0650 .3237*
DIS.22 .4974** .2990** .4693** .4047** .3324** .4357*
DIS.23 .4910** .2422** .5393** .3696** .3317** .4407*
DIS.24 .5663** .2658** .5523** .4723** .4023** .5027*
DIS.25 1.0000 .4281** .5684** .4657** .3468** .5327*
DIS.26 .4281** 1.0000 .3950** .3497** .1208 .3082*
DIS.27 .5684** .3950** 1.0000 .5481** .3994** .5742*
DIS.28 .4657** .3497** .5481** 1.0000 .3873** .5163*
DIS.29 .3468** .1208 .3994** .3873** 1.0000 .4477*
DIS.30 .5327** .3082** .5742** .5163** .4477** 1.0000
DIS.31 .5524** .2690** .5410** .4361** .3920** .6902*
DIS.32 .3212** .1975** .4023** .3289** .2846** .4003*
DIS.33 .4847** .3307** .5340** .4759** .4194** .5032*
DIS.34 .5536** .3441** .5241** .5246** .4658** .5822*
* - Signif. LE .05 ** - Signif. LE .01 (2-tailed)
" . ” is printed if a coefficient cannot be computed
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
— —
Correlation Coefficients
DIS.31 DIS.32 DIS.33 DIS.34
DIS.01 .2629** .2371** .3127** .3000**
DIS.02 .2086** .2856** .2318** .2301**
DIS.03 .2355** .1257* .3188** .3409**
DIS.04 .3591** .3710** .4638** .4829**
DIS.05 .3458** .5160** .5348** .5493**
DIS.06 .1848** .1375* .0549 .1101
DIS.07 .2893** .3740** .3232** .4066**
DIS.08 .1510* -.0350 .0864 .0493
DIS.09 .3415** .3547** .3799** .4206**
DIS.10 .4656**. .3744** .3963** .4533**
DIS.ll .3677** .4650** .4942** .5194**
DIS.12 .4409** .4398** .4370** .5221**
DIS.13 .5501** .4198** .4439** .5217**
DIS.14 .3986** .4505** .4634** .5226**
DIS.15 .3934** .4224** .4774** .4812**
DIS.16 .3656**- .4320** .4182** .4469**
DIS.17 .3912** .4148** .4249** .4394**
DIS.18 .4161** .3647** .5400** .5581**
DIS.19 .3567** .3273** .5336** .4990**
DIS.20 .4739** .3603** .4210** .4481**
DIS.21 .2357** .3927** .3701** .3955**
DIS.22 .4326** .3654** .4435** .4774**
DIS.23 .4417** .2674** .3630** .4253**
DIS.24 .5137** .3546** .4333** .4992**
DIS.25 .5524** .3212** .4847** .5536**
DIS.26 .2690** .1975** .3307** .3441**
DIS.27 .5410** .4023** .5340** .5241**
DIS.28 .4361** .3289** .4759** .5246**
DIS.29 .3920** .2846** .4194** .4658**
DIS.30 .6902** .4003** .5032** .5822**
DIS.31 1.0000 .3365** .4552** .5292**
DIS.32 .3365** 1.0000 .4331** .4639**
DIS.33 .4552** .4331** 1.0000 .7707**
DIS.34 .5292** .4639** .7707** 1.0000
* - Signif. LE .05 ** - Signif. LE .01 (2-tailed)
" . " is printed if a coefficient cannot be computed
216
R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Abstract (if available)
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Organization learning and performance: An exploration of the linkages between organization learning, knowledge, and performance
PDF
Exploring the links among organizational ownership, mission, and commitment: A comparison of social services organizations
PDF
Impact of leader behavior on job satisfaction and employee perceptions of customer service: A communication competence approach
PDF
Toward a strategic reward systems perspective
PDF
Toward a theory of large-scale organizational change: Administrative reform in the state of Missouri.
PDF
Behavioral outcomes of a change from a traditional office to a virtual office: A longitudinal analysis
PDF
The role of personal values in the explanation of salespersons' performance, satisfaction and propensity to quit
PDF
The effects of feedback valence, evaluation process, and participant characteristics upon target attitudes toward job performance feedback: Implications for acceptance of feedback and subsequent ...
PDF
The relationship between coping behavior, personality and stress in performance appraisal
PDF
The adaptation of equivocal IT as an occasion for organizational design
PDF
The effects of psychological type and interpersonal psychological distance on selling-effectiveness in a service industry
PDF
Psychological type as a framework for assessing organizational change: An investigation of interactions between Air Force leadership levels
PDF
Organizational commitment and layoff survivor syndrome: The relationship between communication and downsizing
PDF
The effects of organizational commitment and expected outcomes on the motivation to share discretionary information in a collaborative database: Communication dilemmas and other serious games
PDF
A study in risk management policy: A comparison between public and private organizations
PDF
Impact of accounting information systems on decision processes
PDF
Information asymmetry and internal monitoring: Which blockholders monitor managers more effectively?
PDF
Leading change to create the future: A comparative case study of NASA and Xerox.
PDF
Knowledge management for new product development meetings: The roles of information technology in shared knowledge creation
PDF
The role of cognition and emotion regulation in conflict: A study of the impact on organizational and virtual teams
Asset Metadata
Creator
Nason, Stephen William
(author)
Core Title
Organizational learning disabilities: An international perspective
Degree
Doctor of Philosophy
Degree Program
Business Administration
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
business administration, management,OAI-PMH Harvest,Psychology, industrial
Language
English
Contributor
Digitized by ProQuest
(provenance)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c17-465744
Unique identifier
UC11352066
Identifier
9601036.pdf (filename),usctheses-c17-465744 (legacy record id)
Legacy Identifier
9601036-0.pdf
Dmrecord
465744
Document Type
Dissertation
Rights
Nason, Stephen William
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA
Tags
business administration, management