R Y
Stat:

OVERDUE FINES ARE 25¢ PER DAY
PER ITEM

Return to book drop to remove
this checkout from your record.

 

I‘\
" (I-
'00 ‘1 ~.
\,

\

 

 

 

© Copyright by

RODNEY LEWIS LOWMAN

1979

DIMENSIONS OF CONSULTING:
DEVELOPMENT AND TEST OF A CLINICAL-ORGANIZATIONAL
PSYCHOLOGICAL MODEL OF CONSULTATION

By

Rodney Lewis Lowman

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Psychology
1979

ABSTRACT

DIMENSIONS OF CONSULTING: DEVELOPMENT AND TEST OF A
CLINICAL-ORGANIZATIONAL PSYCHOLOGICAL MODEL OF CONSULTATION

By

Rodney Lewis Lowman

This study develops a theoretical model of organizational consul-
tation integrating clinical and organizational psychological (COP)
perspectives. The model has two primary foci: accurate awareness of
situational reality, and appropriate response to the awareness. It
includes four stages: 1) Awareness and identification; 2) Confronta—
tion; 3) Consolidation; and 4) Integration. Each of these phases is
further elaborated in relation to the organizational setting, and then
operationalized into specific consultation variables. The model is then
tested against perceptions of experienced clients (EXPCLIs) (N = 56),
and inexperienced clients (INEXPCLIs) (N = 5a), of their interactions
with a senior organization development (OD) practitioner, using both
critical incidents (CIs) and questionnaire method010gies.

Results of the study demonstrate construct validity for the essen-
tial components of the revised COP Model, using both methodological
approaches. Analysis of the "objective" questionnaire data shows
support (in both EXPCLI and INEXPCLI samples) for the following
variables: creating an atmosphere of psychological safety; helping
clients confront perceived situational reality; focusing on the needs
of the organization; conceptualizing and redefining problems; identify-
ing discrepancies in perceptions and assumptions; focusing on external

reality issues; determining whether there are compelling reasons for

Rodney Lewis Lowman

change; assessing what would be required if change were to occur;
assuring that solutions are relevant to underlying problems; and imple-
menting and evaluating solutions. The CI data supports a slightly
different subset of the original variables. The CIs also extend the
originally defined variables by providing subdimensions for several

of the a priori COP variables, as well as suggesting three additional
consultant activities not proposed by the original model. Although the
model is multidimensional, the underlying factor structure suggests

two essential components of consultation: a problem conceptualization
and redefinition role, and a benevolent, information-eliciting
dimension.

EXPCLIs differ from INEXPCLIs in their perceptions of the consul-
tant's activities. Specifically, EXPCLIs rated the consultant more
highly than INEXPCLIs on every dimension, and generally exhibited
less variability in their assessments. These differences did not
exactly follow the predicted sequential model. Furthermore, for

both groups, the presence of "halo,' or a general factor, obscured
the nature of relationships among the variables.

The study also defines and operationalizes process variables.
In some cases, these variables received indirect validation from the
CI data, whereas the results from the objective measures of the process
variables were more difficult to interpret. Preliminary data suggests
the primary importance of the consultant's ability to work with the
XélEfi aspects of problems. Helpful and nonhelpful consultant activities
differ in that helpful behaviors are more likely to be cited as

direct applications of the COP Model's "content" dimensions, while

Rodney Lewis Lowman

nonhelpful behaviors are more likely to be related to the manner
of implementation of the content variables.

Methodologically, the "objective" questionnaire data and the CIs
provide different perspectives. The former permitted statistical
comparisons, assessment of group differences, and comparison against a
criterion, while the latter provided rich, detailed clinical data
regarding the consultant's helpful and nonhelpful activities. The
study extensively analyzes both types of data.

This research also discusses the implications of these findings
for the practice of OD consultation, and for the training of organi~
zational interventionists. It suggests that the preparation of
competent practitioners in this area may be much more difficult than
presently conceived. Finally, the limitations and omissions of the
study are reviewed, as well as the advantages and disadvantages of

the methodologies employed.

ACKNOWLEDGMENTS

I owe thanks to many special people.

Whatever expertise I have as a professional psychologist is due
in large part to my training at Michigan State University and the
Texas Research Institute of Mental Sciences. Staff members at each
of these institutions provided me with invaluable opportunities to
develop both professionally and personally, and for these, I am
grateful.

My committee members each contributed special expertise to the
completion of this research. Don Grummon, who was very influential
in my early clinical training, provided relevant clinical psychological
perspectives and a background in psychotherapy research. His influence
on my development as a clinical psychologist is reflected throughout
the study. Neal Schmitt was extremely thorough in working with me
on the statistical analyses. He was quietly encouraging and unusually
generous with his time and energy, despite extensive commitments
elsewhere. He has also been a good friend. Jack Wakeley was especially
helpful in the conceptual stages of the research. His ability to
go straight to the core of the research issues, and to remind me of
the boundaries and limitations of the study was very helpful. I
also found him to be very supportive of the idea of this study, even
when others had their doubts.

My deepest debt of gratitude goes to Carl Frost, who has been a
major influence throughout my graduate training. His patient tutorage
has had permanent influence on my thinking and feeling. This study

itself demonstrates Dr. Frost's finest qualities. He courageously

iii

allowed me to use his clients as the subject matter for the research.
He facilitated entree to organizations for me, and provided immense
intellectual stimulation and curiosity throughout. He has been an
outstanding example to me of the committed intellectual, who believes in
(and practices) the application of knowledge to real-life situations.
He has provided me an inordinate number of professional and personal
growth opportunities, and was encouraging and helping in so many ways.
The Opportunity to have studied with Carl Frost is a life-long debt,
and this brief thanks cannot possibly do justice to his contributions.

Several other people also deserve thanks. Elaine Bish0p
graciously helped coordinate the research. My absence from the univer—
sity on internship and employment placed special burdens on her in
this task, and her efficiency and enthusiasm, despite these obstacles,
is gratefully acknowledged. Tom Mitchell generously helped me through-
out the study, not the least of which was his friendship. I also
appreciate the help of Richard Howard and the industrial/organiza-
tional psychology graduate students in the early stages of the project.
Roger Halley helped coordinate the financial and administrative chores.
The Scanlon Plan Associates not only sponsored this research, but
also provided me with laboratory experiences that were extremely
important in my professional development as an organizational psychol-
ogist. Also, the principal partners at Milford House Associates were
very generous throughout my stays in East Lansing.

Finally, an unexpected bonus of my clinical psychology internship
year in Houston was meeting my wife-to—be, Linda. She has been a

strong help in the editing of this dissertation (despite her own

iv

demanding schedule), but also has been the one who has made it all
worth doing. It is to Linda that I dedicate this volume, to the

pleasures of our time together, and to the anticipation of things to

come .

TABLE OF CONTENTS

LIST OF TABLES. O O O O O O O 0 O O O O O O 0
CHAPTER I INTRODUCTION. . . . . . . . . . .

Some Value Issues . . . . . . . . . . .

Some Critical Organization Development Issues.

Focus on the Present Research. . . . . .

CHAPTER II

Models of Organizational Intervention. . . . . . . .
A Clinical-Organizational Psychological (COP) Model.

of Organizational Effectiveness and Change. .
Operationalization of the PrOposed Model . . . .

CHAPTER III METHODOLOGY . . . . . . . . . .
Data Sources . . . . . . . . . . . . . .
Questionnaire Administration . . . . . .
DataAnaly81Soooooooooooooo

CHAPTER IV RESULTS AND DISCUSSION . . . . .
Characteristics of the Samples . .
HypomeSiSI o o o o o o o o 0
Analysis of the Questionnaire Data
Hypothesis 1A. . . . . .
Hypotheses 2; 2A . . .
Hypothesis 3 . . . . .
HYPOtheSiS 3A. 0 o o o
Hypotheses u, 5, and 6

CHAPTER V SUMMARY AND CONCLUSIONS . . . . .
APPENDIX A CRITICAL INCIDENT QUESTION FORMAT
APPENDIX B COVER AND FOLLOWUP LETTERS . . .
APPENDIX C DIMENSIONS OF CONSULTING SCALES
APPENDIX D GLOSSARY OF ABBREVIATIONS . . .

LIST OF RMENCE O O O O O O I O 0 O O I O 0

vi

MODEL DEVEIOPMENT AND REVIEW OF THE LITERATURE

Page

'vii

160
169
180
181
187
195
196

Table

\) O\\n two N

(I)

10

11
12

13

14

15
16

17

18
19

LIST OF TABLES

Summary of Operationalized COP Model Consultation
DimenSionSo00000000000000...out.co

Distribution and Return of Questionnaires to EXPCLIs . . .
Distribution and Return of Questionnaires to INEXPCLIs . .
Reasons for "Returned, Unuseable" Questionnaires . . . . .
Selected Characteristics of the Samples. . . . . . . . . .
Summary of Coded Critical Incidents . . . . . . . . . . .

Summary of Coded Critical Incidents, Classified by Help-
ful and Nonhelpful Dimensions. . . . . . . . . . . . . . .

Reliabilities of Original and Revised COP Scales, EXPCLIs.
Inter-scale Correlations, EXPCLI Sample. . . . . . . . . .

Partialled Inter-scale Correlation Matrix, EXPCLI
Sadnple I I I I I I I I I I I I I I I I I I I I I I I I I I

ReVised COP ”Mel, EXPCLI salnple o o o o o o o o o o o o o
Inter-Scale Correlations, INEXPCLI Sample. . . . . . . . .

Partialled Inter-scale Correlation Matrix, INEXPCLI
sample I I I I I I I I I I I I I I I I I I I I I I I I I I

Revised COP Model, INEXPCLI Sample . . . . . . . . . . . .
Comparison of Revised COP Model, EXPCLIs and INEXPCLIs . .

Breakdown of Items with High Loadings on Each Group's
GeneralFactor......................

Analysis of Critical Incidents Not Classifiable According
totheCOPModel.....................

Means and Standard Deviations of DOC Scales. . . . . . . .

Univariate and Multivariate ANOVAs to Test for Differences
BetweenExmLIaJldINExmLIGrouPSo o o a o o o o o o o 0

vii

88
89

95
122

12h

126
128
131

133
134
136

137

140
1A3

144

Table Page

20 Discriminant Analysis to Test for Differences Between

EXELI and INEXELI Groups I I I I I I I I I I I I I I I I 114‘5
21 Criterion Items and Reliability Estimates. . . . . . . . . 148
22 Multiple Regression Analyses of Revised COP Model Scales

on the Criterion of Client Ratings of Consultant Effect-
lveneSS, EXELIS o I o 0 o o o o o o I o I o o o o o o o o 114'9

23 Combined Scale Analyses, EXPCLIs . . . . . . . . . . . . . 151

24 Correlations Between Revised COP Model Variables and the
Cri teri on ’ EXPCLIS I I I I I I I I I I I I I I I I I I I I 1 52

25 Multiple Regression Analyses of Revised COP Model Scales

on the Criterion of Client Ratings of Consultant

Effectiveness, INEXPCLIs . . . . . . . . . . . . . . . . . 154
26 Analysis of Combined Scales, INEXPCLIs . . . . . . . . . . 155

27 Correlations Between Revised COP Model Variables and the
Cri terion ’ INEXELIS I I I I I I I I I I I I I I I I I I I 156

28 Multiple Regression Analysis Using Selected Variables as
Covariates........................158

29 Multiple Regression Analysis Using Selected Variables as
covariateS’ INEIXRLIS I I I I I I I I I I I I I I I I I I 159

3O Reliabilities and Interscale Correlations on Process
variables, WLIS I I I I I I I I I I I I I I I I I I I I 161

31 Reliabilities and Interscale Correlations on Process
VariableS,INEXPCLIS...................163

32 Comparison of EXPCLI and INEXPCLI Samples on the Process
variables I I I I I I I I I I I I I I I I I I I I I I I I 1&

33 Correlations Between Process Variables and Criterion . . . 167

viii

CHAPTER I

INTRODUCTION

Some Value Issues

For a person, an organization, or a nation, periodically re-
evaluating the core values to which it is committed in theory, but
which in practice often get misplaced in the minutiae of everyday life,
is extremely important. Gardner (1964) has written a persuasive
statement of the inevitable tendency toward "dry rot" and obsoles-
cence, and the continual need for built-in mechanisms for renewal,
with commitment to meaningful values as essential for effectiveness.
The humanistic tradition of psychology, its concern for understanding
and positively influencing human behavior, is one such core value
that is part of the Science's legacy, but which is frequently lost in
the shuffle of concern for scientific rigor.

Psychology has yet to resolve or integrate its seemingly perenn-
ial pull into the often contradictory directions of research rigor
in methodology, on the one hand, and a humanistic tradition of concern
for improvement of the human condition and for the relief of suffering,
on the other. Maslow (1957) noted that the regard for humanity with
which many students are initially attracted to psychology is typically
snuffed out by the demands to study less "fuzzy, unscientific, tender—
minded and mystical areas of study." Nor is it uncommon for students'
initial enthusiasm and desire to make important contributions to
significant human problems to be lost between beginning training and

its completion. (See, for example, Sarason, Sarason, and Cowden,

1975; Sarason, 1977).

Indeed, the history of psychology has seemed to reflect the ex—
tremes of over-concern with methodology at the expense of the phenomenon
being studied, or well motivated, but often a— or anti—scientific
attempts to "help" others. Neither of these extremes has been very
effective in the quest for an understanding and influence of human
behavior. It would seem more appropriate to ask the right questions,
even with primitive tools, than to get sophisticated answers to
inappropriate or meaningless issues. On the other hand, premature
interventions, however well intentioned, without adequate theoretical
foundations, can be destructive. In any case, the dilemma of psychol-
ogy's identity as a science in the face of pulls to mimic the
scientific methods of the natural sciences, or to devote more attention
to issues concerned with clinical (i.e., practice) areas, remains
unresolved.

One result of these counterposed forces has been the separation
of psychology into increasingly discrete subdisciplines, each with
circumscribed areas of study. Experimental, clinical, developmental,
social, and industrial/organizational psychologies have resulted, with
increasing specialization and separatism. Each area has its own
journals, researchers, students; cross—fertilization of ideas and
interaction or conjoint training among different sub—areas seems con—
tinually to decline. Some of psychology's branches (e.g., experimental
and industrial) have become known for their insistence on rigor in
experimental design and quantitative methods. Others (e.g., clinical)

have emphasized somewhat less the quantitative aspects and have also

relied heavily on methods that employ intuition and feelings; approach—
es which many scientists have found anathema. The overall result has
been a trend toward less and less synthesis.

Even in clinical psychology, which has perhaps been most con-
sistently committed to values of concern for the dignity and worth
of the individual human being (e.g., Revised Ethical Standards for
Psychologists, 1977), and to applying the understanding of human behavior
to the treatment and prevention of human suffering, the scientist/human-
ist dilemma is anything but resolved. It is relatively difficult for
clinical psychologists (whether as diagnosticians, researchers, or
therapists) to totally ignore the core values to which the field is
devoted, since, like medicine, clinical psychology is constantly face-
to-face with the realities of its subject matter. The pain of an
attempted—suicide patient, the self-destructiveness of an autistic
child, or the bizarre hallucinations of a schizophrenic are hard to
deny; the clinician's mandate to help relieve suffering and actualize
potential is clear.

Even so, there are those clinicians (e.g., Meehl, 1973) who call
for greater integration of science with practice, and others who
suggest that the traditional research methods are simply unfit for
the complexities of a practice-based discipline such as psychotherapy.
Certainly there are few areas where research and practice have
pursued such disparate courses, probably to the detriment of both
(e.g., Bergin and Strupp, 1972). After a history of reliance on
a medical, almost Virchowian, model of care and prevention of mental

disorders, some clinicians began to reject this approach, suggesting

that the "relationship" and human-to-human authenticity were the
essential elements of psychological treatment (e.g., Rogers, 1957).
More recently, a flurry of criticism has been directed at the tendency
toward anti—intellectualism that has resulted since this "palace

revolt, and the need for clinical psychology to become intellectually
revitalized (Meehl, 1973; Shakow, 1976; Strupp, 1976a; Albee, 1977).
Certainly the scientist/humanist debate in clinical psychology remains
very much alive.

It can be argued, then, that psychology in general, and its
subspecialties in particular, are overdue for a reassessment of core
values: what should be their fundamental purpose and direction?
Industrial/organizational (I/O) psychology is a second case in point.
For many years, industrial psychology has had a relatively poor press.
In several universities, the specialty is moving out of psychology
departments altogether, either being discontinued or re-located in
schools of business (Cummings, 1976). Among some psychologists, I/O
psychology is dismissed as being too atheoretical and applied; others
criticize it for its perceived liaison with management at the expense
of workers and labor unions (e.g., Baritz, 1960). More substantive,
internal, challenges have been provided by Argyris (1976b) who calls
for a re-evaluation of I/O psychOIOgy's traditional reliance on
individual differences as its primary subject matter, prediction as its
ultimate goal, and its use of implicit rather than explicit theory
about human behavior.

With the increased concern for "relevancy" of recent decades

(e.g., Miller, 1969), some areas of I/O psychology have moved from

the description and prediction of behavior in organizational settings,
to the attempt to change working conditions, organizational structure,
technology, policies, etc. An emerging field, Organization Develop—
ment (0D), has been the result. OD is concerned with the application
of behavioral science to the more effective functioning of organiza-
tions and of the peOple within organizations. The relationship of

OD to traditional organizational psychology is analagous to that between
the study of abnormal psychology and the practice of psychotherapy:
the one is concerned with the theoretical understanding of certain
psychological phenomena; the other, with the application of that
knowledge toward behavioral change, and secondarily, to the generation
of new knowledge as well.

The controversy about OD among behavioral scientists, particularly
among traditional industrial and social psychologists, reflects in
microcosm a larger dilemma, a core value controversy, that pervades
much of the field: should psychology be concerned only with the
generation of knowledge about how man behaves, or should its focus
extend to changing behavior and situations that are undesired? If
the goals of the discipline are more than prediction, then a different,
broader, role set is demanded of the Psychologist than simply labor-
atory and theoretical research. If the psychologist is to be a
"change agent" as well as a discoverer of new knowledge (not that the
two are incompatible), then a clearly articulated normative position
is needed, and an explicitly stated, overtly argued, value assessment
of the goals to which change efforts should be directed. The idea

that organizational psychologists, for example, should be activists,

intervening into systems to create and study change, is a fundamental
value of OD, yet certainly a controversial one. Similarly, the recent
(1978) Houston symposium, "Psychology and Society: In Search of

Symbiosis,‘ well expressed the continuing controversy over psychOIOgy's

activist role.

Some Critical Organization Development Issues

If OD's value and interventionist orientations are controversial,
so is its very nature. Many definitions have been offered for this
emerging discipline (e.g., Beckhard, I969; Bennis, 1969; French and
Bell, 1973) but most include some reference to organizational health
and effectiveness as goals to which the field is committed. Regrett-
ably, the definition of such concepts as "health" and "effectiveness"
is typically left at a state of generalization that is scientifically
unacceptable, though as general abstractions, these concepts under-
score the value orientations of the field.

OD has pursued many different techniques over its short history
in its attempts to change organizational environments. Its historical
roots are found in the fertile ground developed by Kurt Lewin and
his colleagues, especially their sensitivity training work that
began in 1946, and in the extensive survey research work at the
University of Michigan (Huse, 1975). There has, since OD's beginnings,
been no dearth of suggested organizational interventions and purported
panaceas to improve the quality of working life (Lowman, 1978).

This very large collection of activities and approaches encompasses
great diversity (sometimes even creativity)——team building, behavior

modification, transactional analysis, flexitime, Managerial Grid, and

job enrichment are only a few. (Beer, 1976, provides a review of many
of these techniques.) Unfortunately, little attention has been

directed to the underlying dimensions which might account for the
successes and failures of particular techniques (Alderfer, 1976).

Not only have few comprehensive change theories been espoused, but

there have been relatively few "mini theories" that could systematically
explain and direct practice. This dilemma and continuing struggle
between theoretical, pragmatic, and existential perspectives in OD

is colorfully highlighted by Friedlander (1976).

The acceptance among behavioral scientists of "Organization
Development" as a sub-discipline has perhaps been too rapid and
uncritical. Kahn (1974) forcefully takes the OD literature to task
for its excessive use of scientifically undefined concepts all hidden
under the same hole-filled umbrella of "OD." As he puts it:

Organization development is not a concept, at least not in
the scientific sense of the word: it is not precisely defined;
it is not reducible to specific, uniform, observable behaviors;
it does not have a prescribed and verifiable place in a network
of logically related concepts, a theory....the literature of
organizational development is...tied too closely to the labels

in which the varied services of organizational development are

packaged and marketed. (Kahn, 1974, p. 491)

It should also be noted, however, that the same situation exists
with psychotherapy. The term is used to refer to a broad variety of
relationships designed to help lessen clients' suffering or to more
fully maximize their human potential and self—understanding. Generally,
such activities occur in a specific locus and have predictable

characteristics. However, the interventions made by a given psycho—

therapist may vary enormously depending on the therapist's organizing

concepts, training, personality, etc. "Psychotherapy," in short, like

"Organization Development, is an unscientific rubric.

But despite a wide variety of approaches to the conduct of
psychotherapy, there is some evidence to indicate unifying dimensions
that underlie its diversity of "schools" or specific intervention
strategies. Some early research by Fiedler (1950, 1951) suggested that
there were more similarities among experienced practitioners of
widely differing "schools" of therapy than between inexperienced and
experienced practitioners of the same theoretical persuasion. Rogers
(e.g., 1957, 1961) provided essentially a reconceptualization of
effective psychotherapy under a broader level of abstraction, suggest-
ing that certain conditions were necessary and sufficient for any
relationship to be therapeutic. More recently, Karon (1976), in
dealing with the psychotherapy of schizophrenics, notes that the appar-
ent diversity associated with varying approaches to treatment of this
severely disturbed population are often more specious than real. As
he puts it: "You will find as [various theoreticians] talk about
their patients and what they do with them, that the less abstractly
and the more concretely they talk, the more similarity there really
is among therapists."

The point is that "OD" is of limited usefulness for understanding
what interventions are effective, in which organizations, and to
what ends. It is useful only as a general collective term that
probably has more to say of the values and general areas of concern

of its practitioners, than as a focus for systematic research. To

define OD other than in general terms seems impossible, and more,

unnecessary. We do not insist on "scientific rigor" in defining the
field of medical treatment or clinical psychology, but we do in
defining a specific disease or treatment.

But, while the generality of OD as a rubric for the practice of
applied behavioral science in organizational settings may be accept-
able, the relatively low levels of rigor and ingenuity that have thus
far been applied to the research of specific OD interventions or to
the development of OD theory is not. There are several implications
of this assessment. First, higher levels of abstraction or theory,
on the one hand, and more specific concretization of practice, on the
other, are needed. Those studies that have used "survey feedback,"
or "Managerial Grid," or "external consultation" as independent varia—
bles need a greater specificity of exactly what was done, by whom, in
what manner, with what groups, etc., if there is to be adequate
generalizability. There is little evidence that all consultation done
under the same treatment label is alike. Secondly, more abstraction
is required in these studies to look at variables that have greater
psychological and organizational significance-—concepts like roles,
power, and boundaries are examples that seldom receive adequate
empirical attention.

An illustration of this point is provided by Bower's study (1973),
which compares various organizational interventions for their efficacy,
adopting essentially a comparison-of-training types evaluation model.
However, his treatment of such techniques as "survey feedback" or
"T-group" as independent variables begs the questions of what accounts

for any obtained differential effectiveness in the approaches (Ferry,

10

1973). What variables, in other words, explain the results? As Paul
and Bernstein put it: "The goal of treatment research is to find out
what treatment, by whom, is most effective for this individual with
what specific problem under which set of circumstances?" (Paul and
Bernstein, 1973, p. 25)

By extension, if variable identification is what is missing from
much of the OD literature, the training evaluation model as presently
conceived is not likely to fill in the needed information. Evaluation
is not the same thing as research and cannot give the same results
(Sommer, 1977). The assessment that something happened in an inter-
vention cannot, without additional design twists, get at the essence
of the "black box" mechanisms that created any observed change. It
may be, in other words, that an OD practitioner is successful in spite
of himself (or his theories), that it is not his espoused method or
concepts that produce change but something perhaps entirely out of
his awareness or theorizing. Yalom (1975) notes, for instance, that
in group psychotherapy, clients often see seemingly minor incidents
as being of crucial importance or turning points in their "cure".

If one relied simply on the therapist's "school" or espoused treatment
philosophy (e.g., group therapy versus analytic therapy) a very biased
perception of what is helpful or critical might emerge.

It has become fashionable in OD circles to highlight the fact
that the practice of OD is presently far ahead of its theoretical
base (e.g., Argyris, 1971; Alderfer, 1976). What Meehl said of
psychotherapy might be used to express this viewpoint: "Someone has
defined psychotherapy as "the art of applying a science which does

not yet exist." (Meehl, 1973, p. 117) Future theory building in

11

OD will benefit from the pragmatic findings to date, but it is unlike-
ly that substantive contributions to 0D theory will result from more
of the same type of research that now characterizes the field. The
application of s0phisticated research tools to a "science" that is
still in the process of identifying fundamental variables seems pre—
mature, obscuring the nature of the work at hand. What is needed is
theory construction that is integrated with the clinical practice of
OD, its phenomenology and findings about "what works" and why.

Singer expresses this viewpoint well regarding psychotherapy: "Only
when theory becomes secondary, only when theory construction becomes an
g posteriori rather than an g priori endeavor, is there any chance of
therapeutic hearing, any chance of therapeutic success...." (Singer
1970. p. 389)

The other side of the coin, of course, is that organizational
theorists and researchers have been of limited help in providing
behavioral science knowledge that addresses the questions raised by
organizational change efforts. With few exceptions, researchers have
investigated organizational phenomena that assume static, rather than
dynamic, states, and therefore are not able to lend much understanding
to OD interventions (Friedlander and Brown, 1974). Theorists who have
written about organizational change have often relied on analogies
from other change efforts rather than studying specifically organi-
zational change. An example is the concept of "resistance to change,"
which permeates the OD literature, but with limited substantive
integration.

The problem is further complicated by the reliance on "first

12

order" change concepts to attack the problem of creating and testing
the methods of organizational intervention. First and second order
change are concepts suggested by Watzlawick, Weakland, and Fisch
(1974), who provide an innovative theoretical perspective of the
concept of change and its facilitation. First order change occurs
when some within-system shift occurs; in second order change, the
system itself is modified. The latter inevitably involves a dis-
continuity or logical jump. An example of the distinction is provided
by Zubin and Spring's (1977) concept of "vulnerability" as a common
denominator underlying a wide diversity of previously espoused
explanations for the etiology of schizophrenia. They attempt to
demonstrate that each of several explanatory theories (ecological,
neurophysiological, developmental, etc.) explain only a small part

of the total pathological variance. They propose the idea that
"vulnerability" is a second order concept that, potentially, can inte-
grate the findings of each of the six previously dominant theories of
schizophrenogenesis.

This distinction between first and second order concepts is ad0pt-
ed to OD by Golembiewski, Billingsley, and Yeager (1976) who expand
change types into Alpha, Beta, and Gamma change. Alpha and Beta change
refer essentially to first order changes, except that in Beta, unlike
Alpha, the change involves a change in the units of measurement. Gamma
change is second order change. As the authors indicate, the adoption
of methodologies appropriate to first order change to systems in
which second order change has occurred cannot successfully identify

what has happened and by what means.

13

Just as psychology is increasingly questioning the relevance of
traditional scientific models for the study of human behavior, so OD
will not benefit from an unthinking application of traditional method—
ologies. Polanyi (1958) criticized the social science's fairly
mechanical adoption of natural science's methodologies to their
research tasks. Argyris (1968) has vigorously challenged the use of
rigorous research methods that emphasize subject naIvetétand experi-
menter control as essentials for discovering and understanding human
behavior.

In brief, the nascent discipline of OD requires a greater identi-
fication and testing of intervention variables, more focus on specific-
ally group and organizational phenomena, and a more imaginative research

methodology.

Focus of the Present Research

 

It is perhaps not surprising, given the inadequacy of contemporary
OD theorizing and the failure to adequately abstract from experience,
that there is little theoretical-~and still less empirical-~work
available on the tasks and characteristics of the effective organiza-
tional consultant. Experiential data suggest that effective organiza-
tional consultation, like effective psychotherapy, can be an art form
that is difficult to teach, even more difficult to learn, but when
expertly done, a creative, impactful experience. Despite a burgeon-
ing number of OD practitioners (Burke, 1976), however, there is little
agreement about what OD consultants should be trained to do. Nor is

there evidence, other than prescriptive, that organizations require

14

consultants to accomplish the objectives implicit in CD, or can do
it by themselves. Least of all are there solid guideposts for the
novice seeking to master a complex discipline.

The present research grew out of my increasing disillusionment
with the literature and practice of OD, the "salesmanship" of many
practitioners, and the atheoretical perspective of others. My own
graduate school education had included, in addition to extensive clin-
ical psychology training, considerable exposure to the work of Dr. Carl
Frost, who seemed to practice a different sort of consultation than
that often encountered. His own education as a clinical psychologist,
along with his early work with Douglas McGregor and the MIT group
doubtless influenced his "humanitarian" style of interacting with
people in work settings. The clear positive responsiveness of a wide
variety of clientele to what he had to offer, together with his own
limited theoretical writings (to date) on his consultation procedures
and style, suggested to me the need to study his activities more care-
fully and systematically.

My attempt in this research was not just to study the clinical
practice of CD by a senior practitioner, but also to test the feasi—
bility of a research methodology that might capture the essence of an
intervention and build theoretically upon that phenomenonological
understanding. The purpose of this research, then, is to deve10p,
from OD practice, a conceptual model to help integrate some of the
OD literature, to explore some dimensions of at least one consultant's
work, and to provide preliminary empirical evidence about the experience

of OD consultation from the client's perspective. Much has been

15

hypothesized about what is critical in consultation from the consult—
ant's or researcher's view, but almost no literature explores what the
client sees as desirable, necessary, or helpful.

The present study is exploratory in nature. Its aim is not to
suggest the definitive answer to organizational consultation; it is
only a first, and somewhat primitive, beginning. The aim of this
research is less to provide final answers than to discover the right

questions.

CHAPTER II

MODEL DEVELOPMENT AND REVIEW OF THE LITERATURE

Models of Organization Intervention

Few theoretical models of intervention directed toward increased
organizational effectiveness, and the role of the consultant in this
process, have yet been espoused. One of the original and most per-
sistent models of the process of change is Lewin's (1951) three—step
model of unfreezing, moving, and freezing to account for how change
can take place. This has been widely adopted as a heuristic device in
the OD literature, and has also been used (after Lewin) as the base for
a force-field analysis of restraining and facilitating forces for
change, though most often without adequate specification of the com-
ponent forces (e.g., Benne and Birnbaum, 1969; G. Lippitt, 1969).
Schein (1969) also uses Lewin's scheme, and does succeed in operation-
alizing it, but primarily toward the kind of "deep" change that occurs
in psychotherapy and other major, individually focused, change
attempts. Generally, those who have employed Lewin's model have not
adequately operationalized or tested it, and as Kahn (1974) notes:
"The Lewinian schema...remains not only unelaborated and untested,
but really unused."

But while Lewin's model of changing provides a useful point of

departure, it fails to specify, conceptually, what is required for

"unfreezing, moving,‘ and "freezing," to occur, and thus is primarily
a prescriptive model. It does not specify whether all three steps
are essential to all permanent change efforts, the procedures by

which "movement" takes place, and whether different approaches are

16

17

required if the targets of change are individuals, groups, or organi-
zations. Thus, while Lewin's concept of "force field analysis" is a
potentially valuable theory of change, its lack of specificity and of
empirical testing make it, even today, fundamentally incomplete. Its
continued widespread.use as a model of change efforts does suggest
that the theory provides a model which many practitioners have found
useful in field experience (or perhaps, less charitably, that post-
Lewinian theoreticians have not been very innovative).

A major effort was made by Lippitt, Watson, and Westley (1958) to
operationalize concepts of change to the varied tasks of working with
individuals, groups, organizations, and even communities. These
authors attempted to develop a generalized theory of change that would
have widespread applicability. Although their focus was primarily
observational/experiential, rather than systematic or rigorous research,
their contribution to a theory of planned change is substantial.

Specifically, Lippitt et. al. present and operationalize a five-
phase model of planned change which integrates and expands Lewin's
initial model:

1.) DevelOpment of a need for change ("unfreezing")

2.) Establishment of a change relationship

3.) Working toward change ("moving")

4.) Generalization and stabilization of change ("freezing")

5.) Achieving a terminal relationship

Phase 3, "moving," is subsequently subdivided into three subtasks:

3A.) The change problem is identified and clarified

3B.) Alternative possibilities for change are examined; change

goals or intentions of action are established

18

3C.) Change efforts are attempted in the "reality situation"

This attempt to extend Lewin's model succeeded in operationalizing
it and demonstrating its relevance to a broad range of change efforts.
Regrettably, their revised model of change has yet to be put to any
kind of systematic empirical test. While Lippitt, Watson, and Westley
do point to some of the intrinsic methodological difficulties, noting
that theirs is not a neatly sequential model of change, they do
not suggest measurement solutions. They note that more than one phase
is usually Operating simultaneously. As they put it: "Most change
processes probably proceed by a kind of cyclic motion, starting over
and over again as one set of problems is solved and a new set encounter-
ed; hence the different phases become mixed up and the final objective
may be achieved by a process which seems rather muddled to the
observer who is looking for a clearcut developmental sequence" (Lippitt
et. al., 1959, p. 130).

As for the specific components of this model, the attempt to make
the phases of "moving" more concrete focuses on some important dimen—
sions of the change process, but is incomplete. The means by which the
client system's problems are clarified (step 3A) are inadequately
conceptualized and specified. It is probably true, as they assert,
that the generation and analysis of information is necessary, but
the nature of the information obtained, and the manner in which it
is procured, may be equally critical. It would appear that several
additional foci are required before moving into step 3B, the examination
of goals and paths to goals, and 3C, changing intentions into actual

change efforts. Although the authors do attempt to elaborate the

19

change agent's tasks in each phase, their tendency to try to make as
general a theory as possible (in terms of its breadth of application)
ends up without an integrated theory of the necessary and sufficient
conditions of change.

Another, probably more important, but less easily articulated
concern with this model is its focus on change as the primary target
of interventionist activity. This approach implies that systematic
efforts are to be directed to the creation of change. Perhaps this
is a mis-focus, as Argyris (1970) has suggested, and that change, if it
occurs, is the by-product of less direct efforts. In the psychoanalytic
approach to psychotherapy, for example, the desired end product is
of less centrality in therapy than the analysis of what gets in the
way, i.e., so—called resistance. Thus, analytic therapy is sometimes
conceptualized as the process of working through resistance rather
than as the process of induction of change (e.g., Greenson, 1967).
Perhaps, then, in Lippitt et al.'s schema, the focus might more
effectively be directed to a conceptualization of problem resolution
(whether through the means of information collection, analysis of
problem solving styles, or whatever), and less on the process of planning
change per se. Or again, the Watzlawick, Weakland, and Fisch
approach to change (1974) recognizes that change induction is often
the result of paradoxical interventions that may never identify "change"
as a specific goal.

A third major attempt to theorize the process of organizational
intervention has been provided by Chris Argyris. Few theorists or

practitioners have been as conscientious as Argyris in attempting to

20

provide an integrated, in-depth understanding of the theory and pro-
cess of organizational consultation. He postulates three fundamental
consultation tasks (1970):

1.) Generation of valid and useful information;

2.) Free and informed choice;

3.) Internal commitment to the choices made.

This first of these activities is assumed as axiomatic. Valid
information is defined as "that which describes the factors, plus
their interrelationships, that create the problem for the client
system" (Argyris, 1970, p. 17). Usefulness entails the generation of
valid information about variables that are manipulatable by the clients,
and which are such that, if manipulated, effective change will follow.
As such, this factor is similar to Lippitt et al.‘s phase 3A,
identification and clarification of the problem. It differs in its
further specification of the validity and usefulness dimensions.

Free and informed choice, Argyris' second intervention activity,
requires the client to be aware of his choices and to make them
voluntarily. This forces the client system to make its own decisions
and therefore to retain responsibility for its own directions. Argyris
views this task as central, in part, because of its compatability
with the desired end state of system self—responsibility. "A choice
is free," Argyris states, "to the extent its members can make their
selection of a course of action with minimal internal defensiveness,
can define the path (or paths) by which the intended consequence is
to be achieved; can relate the choice to their central needs; and

can build into their choices a realistic and challenging level of

21

aspiration" (1970, p. 19). Obviously, this conceptualization
encompasses a number of assumptions and is an extensive elaboration
Of "free choice" beyond its usual parameters. Argyris does go on

to explain that free choice without an accurate analysis of the
situation (as Opposed to choices made on the basis of the biases or
defenses Of the decision makers) will only lead to increased defen-
siveness, psychological failure, thus creating even more distorted
perceptions and more censored information by the system. While

such an argwment has intuitive plausibility, and is grounded (though
not very rigorously) in the psychOlOgy Of aspiration and Of perception
studies,, no systematic empirical test has yet been reported.

Finally, Argyris argues that the consultant must help the Client
develop internal commitment to the choices made. This is consistent
with the self-responsibility idea encompassed in the second step.
Argyris maintains that internal commitment gives a high probability
that it will be long—lasting, maintain itself over difficult conditions
such as stress, and that it will allow the continual re-examination Of
the choices made (since the commitment has been based presumably on
valid information).

As a theoretical approach to organizational intervention, Argyris'
model is internally consistent and is loosely supported by some Of the
social psychological research findings. Its basic contention is
that truth, i.e., undistorted views Of reality, leads to effective
problem resolution. Taken in broad perspective, the model has the
capacity to integrate much of the less theoretically-based OD litera-

ture. For example, a number of approaches are simply structural

22

methods Of eliciting information from an organization.

The difficulties Of the model seem primarily to be Of Omission
(in addition to the important fact that it has not yet been empiri-
cally validated as a model of intervention). In general, the transla-
tion of Argyris' model into specific consultation activities seems
to favor a process view Of consultation. The interventionist helps
a client system discover its own perceptual distortions, primarily
(though the model does not demand it) by a modified sensitivity group
approach. As the barriers to effective identification Of data are
lifted, the idea is that choices are made on the new information then
available. There is more emphasis, by implication, on the affective,
and perhaps value dimensions of consultation, but seemingly less on
the cognitive. While the implicit values espoused by this approach
are consistent with Argyris' concepts Of interpersonal competence
(Argyris, 1962), it is unclear how his consultation approach ties this
focus into the broader, Often external, demands placed upon the
organization. Presumably, Argyris' contention would be that the
generation of valid and useful information requires the focus on all
information relevant to a problem; but whether an essentially process-
consultation approach is the primary, or even the best, way to achieve
this end is questionable. Finally, Argyris does not clearly integrate
his approach with a consistent theory Of changing, i.e., theory of
how change takes place.

Despite these issues and difficulties with his model, Argyris has
provided a first (if incomplete) step toward a comprehensive theory

Of organizational intervention. The integration Of his model with

23

the broader OD and psychological literature remains unfinished. (His
promised second volume on intervention theory has yet to appear.)

A fourth attempt at systematic integration has been put forth
by Alderfer (1976), who uses the concept of organizational boundaries
and his "ERG" (existence, relatedness, and growth) model of human
needs as central organizing devices to understand change processes in
organizations. His approach views the desired end goal of CD efforts
as the creation of a state of optimally permeable boundaries sustained
by "mutual relationships". He proceeds to organize the various
change attempts around the type of need to which they appeal. Alderfer
concludes that the primary tasks of the organizational change agent
"...involve identifying, understanding and changing the nature of
boundaries and relationships of individuals, groups, and the organiza-
tion as a whole." (1976, p. 1632) These tasks are directed to the
optimal establishment of relevant boundaries, opening closed boun-
daries, and movement of relationships to greater mutuality (Alderfer,
1976; Alderfer and Brown, 1975).

While it is true that greater boundary permeability is likely to
result in more information exchange, there is no assurance that this
information is relevant or valid. Nor are more permeable boundaries
necessarily valued ends in themselves, but rather important only to
the extent that they further such ends as organizational health and
effectiveness. Although Alderfer hypothesizes an inverted U—shaped
functional relationship between "system vitality" and boundary
permeability, he does not provide convincing evidence that the latter

is necessary or sufficient for attaining the former. While his theory

21.

is interesting, and he makes innovative use of an organizing device

to look at studies of organizational change, Alderfer does not make

a solid case that factors other than those reducible to boundary
issues are not equally important. In short, his key concepts do not
seem adequately comprehensive to explain the complexities of organiza-
tional change. Finally, some of his "implications" from his (untested)
theory do not follow very clearly. For example, one suggests that
permanent change in systems is most likely to occur and be maintained
if the changes are programmed with diagnosis, action, and evaluation
phases, with both internal and external involvement. While this may
be very valuable advice, its ties to Alderfer's theory are less than
obvious.

Data-Based Empirical Studies

 

Few OD theories to date have tested-—rigorously or otherwise--
their theoretical proposals with actual data—based research. Gordon
Lippitt (1959) did administer a short series of questions to some
clients of a national Organization with which he was involved in an
OD project, but the questions asked were so rudimentary as to provide
little help in generalizing beyond the immediate experience. For
example, in attempting to extract information regarding the nature
of resistances encountered in the process of change, the question
asks: "What kinds of resistances did your group develop during the
consultation process?" It is unlikely that clients will be able to
respond to such inquiries in a way that would provide substantive
information about the client-consultant relationship.

Researchers at the Institute for Social Research (ISR) are

25

responsible for two major comparative studies of OD interventions.
Bowers (1973) reports on the Michigan Inter-Company Longitudinal
Study involving 23 organizations and some 14,000 respondents. Organ-
izational interventions used as independent variables included Survey
Feedback, Interpersonal Process Consu tation, Task Process Consulta-
tion, and Laboratory Training. Data handback (return of data with

no systematic follow-up) and No Treatment conditions served as compar—
ison groups.

As already discussed in Chapter I, such "treatments" are unsatis-
factory descriptions for the purposes of real understanding, since
they imply a consistency to an intervention approach that is highly
unlikely and which does not give adequate specification Of exactly what
went on during the "treatment." It is true that many of the treat-
ment interventions are described in some detail, and that they may
effectively describe consultants' "theories of action"; whether they
also describe their "theories in use" (to borrow Argyris' terminology)
is questionable. In other words, that a consultant sees particular
ideas as being important in principle does not mean that, in actual
practice of the profession, the principles are actually used. It
is also true that the practitioner is Often the worst evaluator of his
or her own behavior, due to proximity to the situation and the oper-
ation of perceptual defenses.

Nor does Bowers' research make clear whether the various practi-
tioners were equally skilled in their specialties; obviously a test
of a novice "process consultant" against a senior "survey feedback"

practitioner is an inadequate comparison of the methods. It is

26

also the case that the measures of consultation are only described
from the researcher's view, with no attempt to collect data regarding
clients' perceptions of the consultants' activities. Such measures would
provide badly needed "manipulation checks" of whether Bowers' assess-
ment of the consultation activities were corroborated by the recipients
of the services. Finally, while it is entirely possible that survey
feedback (the mainstay of the ISR Center where the research was based)
is indeed superior to other types Of intervention, as the study
concludes, there is a strong need for replication by others affiliated
with institutions not so committed to this technique. As Kahn notes:
"...a sociologist of knowledge might express some lurking skepticism
that Survey Feedback has been discovered by its proponents to be the
most effective form of organizational deveIOpment" (1974, p. 493).

It may also be that the criteria measures in Bowers' study are
biased toward showing a relative efficacy for Survey Feedback that is
more an artifact of the means of measuring the criteria. While Bowers
provides numerous reasons why this possible bias ought to have been
minimal, he also acknowledges the more substantive concern about
errors of Omission. Are the other intervention approaches designed
to change organizations on the dimensions tapped by the ISR question—
naire? If not, it would not be surprising that the differences were in
Survey Feedback's favor. Furthermore, it appears that, except for
Survey Feedback, the other interventions were directed almost exclu-
sively to "capstone groups," (i.e., tOp management), not to the
entire organization. To use organizational criteria rather than group-

based criteria, seems questionable.

27

Some elaboration of Bowers' study was provided by Franklin (1976),
who did a comparative study of "successful" and "unsuccessful" 0D
change efforts in 25 companies (11 "successful"; 14 "unsuccessful"),
involved in the Michigan longitudinal study. His criterion for
"success" was change on 16 survey indices designed to assess organiza—
tional climate, leadership practices, and satisfaction. The change
effort was rated successful if a statistically significant number of
dimensions changed in the more favorable direction. The independent
variables covered a variety Of categories: the organization's environ-
ment, characteristics of the organization itself, initial contact
procedure, data gathering activities, etc. Here we are concerned pri-
marily with the consultative variables. In this study, as in the
Bowers' research, the dimensions of consulting are left too general
to be Of much help in the present research. Specifically, in terms
of "types of skills" of external change agents, Franklin identifies
presence or absence of "interpersonal, structural analySis, persuasion,
laboratory training, T—group, perceptual confrontation, diagnosis,
and political savvy," and two types of change agent style: "catalyst"
and "transducer." NO significant discriminators were found between
successful and unsuccessful organizations on the external change
agent characteristics or on change agent style.

While the Franklin study provides some data regarding a large
number of hypotheses about 0D efforts, it has relatively little to
Offer toward understanding what the effective change agent actually
does. This was not the primary focus of the study, however. Never-

theless, the difficulty created by inadequate specificity of the

28

consultant's activities is illustrated by the case of "internal change
agent dimensions," which the Franklin study reports discriminated
between successful and unsuccessful interventions. One of the four
sub-categories was the presence in the successful OD consultants of
"assessment and prescriptive skills." What constitutes such skills,
and how they were employed, are inadequately specified by the research.

A third empirical attempt to measure the efficacy of 0D efforts
was Frohman's dissertation research (1970), which compared interventions
in three comparable semi—autonomous units of the same organization.
This interesting study manipulated the gmpnnp of external consultant
intervention, but it too had little articulation of the exact nature
of what it was that the consultant actually did, and no data were
collected from the perspective of the client. (Although Frohman
reports that detailed notes were taken by the interventionist after
every visit by the consultant to the firm, these were not employed
as independent variables in the study itself, nor even systematically
categorized.)

A fourth study (Tichy, 1974), also based on dissertation research,
explored various types of change agents, categorized on the basis of
consultants' responses to questionnaires. From a wide variety of

"change agents" (11: 91), four categories of intervention were isolated:

"outside pressure, organization development, people change

technology, and "analysis for the top." Like the previously dis-
cussed studies, however, Tichy's does not give much attention to

reporting what the change agents actually do. He relies on rather

29

general self-descriptions. His "analysis for the tOp" category,
for example, included 15 respondents who used operations research and
systems analysis, six who did "policy studies" and seven who were
"'eclectic' analytic consultants (business school professors)." 0r
again, the "organization development" category included "those who
were associated with two major nonprofit institutes and who referred
to themselves as OD consultants." Thus, the categorization of these
change agents is open to question. There appears to be high within-
group variation in change agent style and approach. Unfortunately,
the grouping of the consultants was used for most of Tichy's analyses,
making the findings that emerged of little relevance here.

Finally, Kolb and Boyatzis (1971) conducted a study to look at

various aspects of the "helping process,’ using MIT management
students as subjects. Students, almost all male, were required to
participate in small groups focused on "helpdng one another achieve
personal change goals" as a course requirement. Kolb and Boyatzis
measured various need structures of the participants using the
Thematic Apperception Test to assess power, achievement, and affilia-
tion needs in helpers who were perceived to be effective, ineffective,
and non-helpers. They conclude that effective helpers were those who
were moderately high in all three motives and in self-image. Feed-
back from effective helpers tended to be more positive and less
control-oriented than did that of the ineffective helpers. These
findings are suggestive of the characteristics of effective helpers,

but their study in non-organizational settings with forced participants

and dealing only with interpersonal issues, limits potential

3O

applicability to OD. Nor will this review elaborate the methodo-
logical problems with this study (use Of projectives to measure need
structures, absence of needed data to assess certain statistical
procedures, etc.). Finally, although the authors reported that they
planned a follow-up study about the characteristics of effective
receivers of help, it has apparently never appeared in the literature.

Thus, overall, the number Of data-based 0D studies is very small
and of uneven quality. Most researchers (e.g., Kahn, 1974; Franklin,
1976) call for greater empiricism in OD research, particularly into
the specific role of the consultant in the process of change, but
such studies have simply not yet been conducted in any significant or
meaningful way.

A Clinical-Organizational Psychological (COP) Model of
Organizational Effectiveness and Change

Frost (Frost, Wakeley, and Ruh, 1974) has presented two conditions
as essential for organizational (and individual) effectiveness:

1.) Accurate awareness Of situational reality

2.) Appropriate response to that reality

In other (unpublished) work, he has elaborated, from the per-
spective of over thirty years of the practice of OD consultation, the
implications Of these two conditions for the practice of organizational
consultation. Specifically, he suggests a Clinical-Organizational
Psychological (COP) model of changing that is composed of four
primary elements:

1.) Awareness and Identification

2.) Confrontation

31

3.) Consolidation

4.) Integration

In this section, definitions and illustrations of these concepts
will be provided to explain the model, after which a more specific
operationalization of the model will be developed.

Awareness and Identification

Definition: Making the perceived problem explicit and establish-

ing the identity of its constituent parts.

The COP Model assumes that problems must first be accessible to
awareness before they can be resolved. Examples from clinical psych-
ology illustrate that underlying problems are Often obscured because
they are not available to conscious awareness. Before the patient
can confront core issues, access must first be available to the
client's present perceptions. Creation of an atmosphere (i.e., a
relationship) where the client feels it is safe to share his inner
world, is one of the first tasks of many approaches to psychotherapy,
a necessary but not sufficient condition for change. This process
enables the client to begin to bring into awareness and to share with
another person his present perceptions of reality, however distorted.
For example, a client who is very angry in body posture, gestures,
and voice tone, but whose words deny the emotion, may be helped by
a therapist to acknowledge that he is feeling something, and perhaps
later, that his feeling is anger. As long as this awareness and
identity are absent, understanding of the sources generating the emotion
and their relevance to the broader behavioral patterns that are

causing the difficulty is unlikely. Similarly, in the organizational

32

setting, a group of management and union workers might be meeting to
discuss a management-directed change in Operating procedures. Although
the employees may have strong reservations about the proposed prOgram
(though they may not even be aware of the concerns), they may not
discuss their views. Instead, they may give surface acquiescence,

but later assure the failure of the program. The "real" issues are not
accessible to the group; whatever legitimate need for change there
might be is obscured.

Becoming aware of a problem, i.e., admitting its existence, is
frequently a difficult first step. Elaborate systems or infra-
structures may be constructed in an individual, a family, a group,
or even an organization, to protect themselves, sometimes quite appro-
priately, from painful realities. Some families, for example, have
tacit agreements that no one mentions the father's drinking problem;
individuals may employ extensive denial systems to keep out of
awareness their overwhelming sense Of inferiority. In an organizational
example, a chief executive became visibly upset and soon changed the
subject when confronted by a key staff member's perception that the
human relations climate in the company was conspicuously low. On
his next encounter with the consultant, the chief executive presented
a highly rationalized explanation to the effect that there was no
problem, that management, in fact, was quite generous to its
employees (a perception not supported by the company's low wage rate
for its community, high turnover and absenteeism, and persistent

failure to earn productivity bonuses). In this instance, awareness

of the situational reality was too difficult to confront and no problem

33

awareness or identification was possible.

In the COP Model of consultation, awareness must be followed by
identification of the nature of the perceived problem. A psycho—
therapy client with difficulties making and keeping friends may define
his problem as his being too inferior for anyone to possibly like.

With this identification, the therapist can begin helping the client

to search for the underlying concerns (behavioral, Oedipal, or whatever)
that account for the Perceived inferiority. Or, an organization may
become aware of an increasingly dissatisfied workforce. It identifies
the cause of the problem as being the company's inability to earn
productivity bonuses. Helping the client discover how bonuses are
earned may focus its attention on the difficulties underlying their
absence in this particular organization.

In several respects, the "awareness and identification" phase is
similar to Argyris' first stage of generating valid and useful infor-
mation, as elaborated above. It also has some of the elements of
Lippitt et. al.'s phase 3A, identification and clarification of the
change problem. The major difference lies in the further specification
of the ends to which these tasks are directed, i.e., accurate aware-
ness of situational reality, and the conceptualization of the
specific dimensions associated with these tasks (elaborated below).
Confrontation

Definition: Comparison of system behavior and system actions;

assessment of the meaning of "awareness" for the system's behavior.

The second portion of the COP Model calls for perceptions gener-

ated in the first step to be confronted against the system's actual

34

behavior and against a larger reality focus (i.e., non—system issues).
This is viewed as the continuation of the process of establishing
accurate awareness of situational reality. A criticism of many
approaches to organizational intervention is that they are often

very solution oriented and technique-bound, suggesting procedures

for intervening without convincing rationales of their relevance to
this particular set of problems. The COP Model views confrontation
of the system's problems (frequently a heavily emotional process)
against relevant reality issues from both within and outside the
system to be a crucial precursor to effective problem solution.

As the client system's awareness and identification of problems
becomes accessible, it should be possible to confront and work through
the various discrepancies in its perceptions. Such discrepancies
may be among different views of the same problem, or between percep-
tions and relevant issues from outside the system. In the clinical
setting, family therapists Often help client systems confront their
differing views of the same situation to obtain a common agreement
about the nature of the problem. In the organizational arena,
clients may need to confront the implications of their new aware-
nesses with the need for changes within the system.

Like awareness and identification, the confrontation stage is
not seen as being an end in itself. Its value lies in sorting through
the various perceptions to come to a commonly agreed perception of
situational reality. and an assessment of the need for appropriate
response to these insights. The process of confrontation may require

reconciliation of conflicting views, and thus requires a superordinate

35

background against which discrepancies can be resolved. If, for
example, a couple experiencing profound marital unhappiness becomes
aware of different, incompatible needs and also discovers that its
former relationship has been irretrievably lost, reconciliation is
not likely. Or, an individual who insists that he is Moses or a
deity must ultimately come face to face with his distortion of
reality and the divergence of his self—perceptions from the rest of
the world's. Or, in organizations, a conflict between the sales and
manufacturing vice-presidents may be settled only against the broader
issues of the needs of the company.

The idea of confrontation as an integral element of changing has
received relatively little attention in the literature. The psycho-
therapy literature has discussed confrontation somewhat (e.g., Adler
and Myerson, 1973), but primarily from the standpoint of technical
issues related to the practice of therapy. In the organizational
setting, Davis (1967) cites confrontation as the "missing element" in
the behavioral science literature. As he puts it, "There is no
real growth.[or] development in the organization or in the individuals
within it, if they do not confront and deal directly with their
problems." (Davis, 1967, p. 4) Finally, Argyris (e.g., 1970, 1971)
confronts clients' behavior toward him (as a consultant) as an
analogue of their more general behavioral patterns.

Consolidation

Definition: Coordination of the need for appropriate action

with responsibility for its execution.

In this phase of problem resolution, the various aspects of the

36

problem are begun to be put together. A psychotherapy client, for
example, may successfully uncover many distorted perceptions of

reality. He may come to realize that part Of his undesired overt
behavior is explained by conflicts which were not formerly in

conscious awareness. As he begins to connect the underlying conflicts
with the presenting symptomatology, the process of consolidation begins.
0r again, a client may discover that he is behaving in ways that will
not allow his needs to be met (e.g., Glasser, 1965). Experiential
recognition and acknowledgment of the needs may be followed by

action more appropriate to their satisfaction.

The essence Of consolidation, as conceived here, is the accept-
ance of responsibility for problems and for action that will solve the
difficulties. This is not necessarily a neatly sequential process
in actual consultation or therapy. There appears to be a sense of
disjointed flow to the process of change, whether in therapy or in
0D consultation. Seldom is the process Of change a straight, linear
function. In psychotherapy, this process is often called "working
through" of conflicts. It is common, for instance, for a steady
period of emotional trauma associated with confrontation of painful
issues to be followed by a plateau in which little of much apparent
depth is occurring. Such a period seems necessary for the client
to begin to adjust to and internalize his new awarenesses.

In this phase of the COP Model, the irrational perceptions are
replaced by increasingly accurate awareness of situational reality,
and increasing recognition of the need for action compatible with

these awarenesses and commitment to changed behavioral patterns.

37

In this conceptualization, consolidation as a phase of intervention
has not been directly addressed in the 0D literature. More frequently,
approaches to problem resolution in organizations articulate the

need for problem identification and for solution generation and
implementation with little specification of how to move from one

step to the next. Sofer (1961) does provide descriptive data of

the vicissitudes of organizational change, the movement from euphoria
to depression to a more realistic assessment of problems and their
solutions, but he does not integrate this description with the

need for appropriate action on the part of the consultant. Most

of the rest of the OD literature does not even go this far.
Integration

Definition: Combining the various elements into a coordinated

whole.

The final step in problem resolution as conceived in this model
is the integration of the preceding steps with behavioral actions.

In psychotherapy, the concept of integration refers to the organ-
ization of various traits or tendencies into one harmonious person-
ality. The various parts are put together into a complete whole. An
obsessive—compulsive neurotic, for example, who uncovers the etiological
factors in the disturbance and connects the pervasive underlying
loneliness with its dynamic origins moves to integration when his
behavior begins to be affected by these realizations. If therapy has
been successful, the symptomatology (e.g., avoiding stepping on
sidewalk cracks, continual preoccupation with certain numbers or

thoughts, etc.) may begin to disappear as the elaborate defensive

38

structure is no longer needed for protection from unwanted awareness.
Similarly, an organization that has persistently avoided certain
perceptions about its declining position in the marketplace and its
decreasing value to the best in the labor market, may come to terms
with its realities through a lengthy, perhaps painful process. It
may recognize that, despite its (defended) perceptions, environmental
challenges threaten its existence. It may discover, for example,
that its own worst "enemy" is not its militant union, but an increas-
ingly tough group of competitors and decreasingly available raw
materials. It may then reOOgnize that unless it can work collectively
with its labor force, problem resolution is not possible. Unless it
can educate its employees to the nature and demands of external
reality, its resources will be diverted to other, Often conflicting,
aims. By such steps, it has begun a tedious process Of integration.
This too is a change phase that has received little attention
in the literature. Bennis wrote in 1965 that we know least about
the process of implementation, with its attendant phases of under-
standing and commitment; today, his critique remains more true than
not. French and Bell (1973) discuss OD's emphasis on goal setting,
and stress the need for clients to learn goal setting and planning
abilities. In contrast, the COP Model does not value goal setting
by itself, but only when integrated with the two primary consultation
tasks; helping the client develop accurate perceptions of situational
reality,and responding to these perceptions with appropriate action.
Obviously, the setting of targets or goals with a client might be

an appropriate focus at some stage of the consultation process,

39

but it is not, in this approach, a primary consultation activity.
Finally, Frohman (1970) operationalizes Lewin's third change phase,
freezing, as "integration," but he seems to use the concept to refer
primarily to movement to participative management, which is more
narrow in focus than the present conceptualization.

Having reviewed the general theory of the COP Model, the next

task is its specific Operationalization for research purposes.

Operationalization of the Prgposed Model

 

The four-phase COP Model of intervention presented above puts
forth a hypothesized theory Of consultation directed toward increased
organizational effectiveness, a model that is based on clinical and
organizational psychological perspectives. Here, the basic model
is operationalized with specific dimensions of the consultant's tasks.
Four primary consultation tasks, and a number of more specific varia-
bles are identified as the operationalized version of the COP Model.
These are summarized in Table l, and elaborated in the following
discussion, along with a review of literature relevant to each Of

these consultative tasks.

Problem Awareness and Identification (Awareness and Identification)
Definition: Assessment, with the client system's participation,
of the nature of the system's problem, including cognitive,
value, and affective components.

The awareness and identification phases of the COP Model are here
conceptualized as problem awareness and identification. By this is

meant awareness that something is dysfunctional or unsatisfactory in

1.

40

Table 1

Summary of Operationalized COP Model Consultation Dimensions

Problem Awareness and Identification (Awareness and Identification)
1.1 Facilitation of Information Flow

A. Creation of an atmosphere of psychological safety
(PSYSAFTY)

B. Elicitation of clients' perceived situational reality
(PRCVRLTY)

0. Process facilitation (PROCSFAC)

D. Linkage agent (LINKAGEN)

1.2 Problem Conceptualization and Re—definition (PROBRDF)

 

A. Identification of discrepancies in perceptions and
assumptions (DISCREP)
B. Focusing on the needs of the organization (ORGNEED)
C. Focusing on external reality issues (EXRLTY)
D. Theory input (THRYINPT)
Assessment of the Need and Capacity for Change (Confrontation)
2.1 Goal Identification and Evaluation (GOALFOCS)
2.2 Identification of Reasons for Changing (WHYDIFF)
2.3 Identification of the Implications of Changp (WHATDIFF)

Problem Ownership (Consolidation) (PROBOWN)

 

Joint Authorship of the Solution (Integration)

4.1 Assurance that ApprOpriate Resources are Included in the
Generation of Solutions (SLTNRESRCS)

4.2 Identification, Expression, and Testing of Alternative Solutions

A. Assurance that the rOposed solution is relevant to
the actual problem ESLTNRELV)

B. Assurance that the proposed solution is consistent with
the demands of external reality (SLTNLINK)

_ 4.3 implementation and Evaluation of the Solution(§) (SLTNIMEV)

41

the client system, a concept similar to Lippitt et. al.'s "felt need,"
and the identification of the nature and parameters of the problem.

Problem awareness and identification are implicit in many approach-
es to CD. Sofer (1961) speaks of the assembly and analysis of facts
about problems as a major intervention task of the consultant.
Levinson (1972) has developed an extensive guide for organizational
diagnosis. Adopting a clinical psychOIOgical perspective, Levinson
suggests that extensive assessment of the organization from genetic,
descriptive, and interpretive views is essential to effective consul-
tation. He goes on to isolate the relevant organizational variables
which he feels serve as a data—gathering guide for practitioners.

The organizational consultant, in Levinson's view, must serve as a
detective in much the same manner as an expert internist or clinical
psychologist, a view shared, in part, by Steele (1969).

Indeed, most approaches to OD view diagnosis as an essential
preliminary to changing organizations (e.g., R. Lippitt, 1959; Bennis,
1966; Beckhard, 1969; Burke and Schmidt, 1970; Kolb and Frohman, 1970;
French and Bell, 1973; Beer, 1976b). But diagnosis is really just

one aspect of "problem identification,‘ since any diagnostic
formulation assumes an underlying model of health and dysfunction.
Organizational consultation differs from medicine in its diagnostic
approach in that, at this stage, there is no generally accepted
understanding of what is functional and what is unhealthy in organi-
zations, i.e., no taxonomy of health and disease. As Walton and
Warwick (1973) observe, however, "Kpap gets defined as a problem

is...one Of the most crucial and politically salient aspects of

0D." (p. 684)

42

In some schema of 0D, the particular diagnostic model of the
consultant is of less relevance than that the clients be helped to
generate their own diagnostic information. Argyris (e.g., 1970) sees
this as one of the interventionist's most important tasks in consulta-
tion. The implication of this approach is that the primary value of
the consultant may be in assisting the process of information gener-
ation by the client system, rather than supplying a conceptual scheme
or model to the clients. Strauss (1976) has recently corroborated
this view by his statement that OD's main contribution is process-
related: "Effective 0D occurs when organizational members learn pp!
to diagnose what changes are required and p93 to make these with
minimal pain and lasting effectiveness--and hpfl to learn from the
learning process..."(Strauss, 1976, p. 674; emphases his). Similarly,
Sofer (1961) notes that more important than the facts identified in
organizational consultation is the manner in which problems are defined
and reconceptualized. This process forces assumptions to be made
explicit, the assessment of contradictions, and a depiction of the
social and psycholOgical forces at work in the situation.

While few theorists or practitioners would disagree that "problem
identification" is a basic consultation task, not all share the view
presented here that cognitive, emotional, and value components must
all be considered and worked through. Quite possibly, it is the
failure to deal with all three aspects of problems, and the often-
encountered inability of the consultant to work with each level, that
limits the scope and potentiality of many approaches to planned change.

An example may be helpful. A consultant working with a top management

43

group may be presented the dilemma of continuing production and
efficiency losses despite highly paid wages and relatively benign
personnel practices. In discussing the issue with the group, the
consultant discovers that management is acting on the assumption that
employees are aware of the relevance of the issue to certain competi-
tive realities which demand increased performance. In fact they are
not. The consultant then learns that management has insisted,

through the company's union, that employees' performance must improve.
The rationale for this need has not been presented. Management does
not understand why the workers need to know the details of the
problems and is reluctant to share the information. (Cognitive
dimensions) Further discussion reveals that management has a funda-
mental attitude Of superiority toward its employees. (Value dimension)
Partly, this stems from an emotionally charged union organizing
history that has left the management leaders bitter and uncooperative.
(Affective dimension) The "real" problem here, then, is inextrica—
bly bound with several components, and the failure to confront all
aspects or levels of the problem may interfere with change. This
example also illustrates the fact that in consultation, the issues
are seldom neatly arranged so that they can be approached one at a
time in a step-wise or sequential fashion.

Because behavior is multiply determined and Often overdetermined
(e.g., Argyris, 1976c), problem behavior, and its identification, must
be addressed on a variety of levels. It may also be that change
efforts must be approached on a variety of levels, and thus also

be over-determined. This perspective is adopted by Bennie (1965)

who argues that for effective implementation of planned change
efforts, the program must include emotional and value elements,

in addition to cognitive ones. He has also argued that Of all the
possible power bases for the consultant, probably the most effective
one is value-based.

The technology of problem awareness and identification, i.e.,
means for implementing these goals, varies considerably among differ—
ent approaches. In this model, two tasks are viewed as fundamental
to problem awareness and identification: the facilitation of infor-
mation flow (1.1), and problem conceptualization and re-definition
(1.2). Like Argyris (1970), it is accepted as axiomatic that without
information about the problem at hand, no solution is possible. The
focus here, however, is first on perceived information, rather than

Argyris valid and relevant information." It is suggested that
before objective information can be generated, clients must be able
first to surface information and views that are presently valid for
them. Often, this information will be distorted, cluttered, and
irrelevant to organizational needs and external realities. However,
this distorted information must be dealt with before constructive
change can begin. This conceptualization is similar to Katz and
Kahn's (1978) "value expression" as being one of the sources of
internalized motivation. The consultant's first task, then, in

the COP Model, is to help create an atmosphere in which sharing such

perceptions is possible.

Four consultant roles are identified in the area of facilitation

45

Of information flow (1.1):
A. Creation of an atmosphere of psychological safety (PSYSAFTY)
B. Elicitation of clients' perceived situational reality
(PRCVRLTY)
c. Process facilitation (PROCSFAC)
D. Linkage agent (LINKAGEN)

1.1A PSYSAFTY

 

An implication of the hypothesis that problem identification
requires the consultant to begin with the client system's present
perceptions of reality is that the consultant must be able to elicit
current views regarding the cognitive, value, and affective aspects
of system problems. Central to this task is the creation of an
atmosphere of psychological safety, i.e., an atmosphere in which the
sharing of perceptions is safe and negative consequences do not follow.

For the consultant's part, two roles are seen as essential to
this end: the change agent must be seen as genuinely wanting to
help the organization, i.e., his motivation must be seen as benevolent,
and must also be seen as in fact competent to help. These issues
may be conceptualized as "process" variables Of consultation that
overlap with the PSYSAFTY variable, which is viewed as a "content"
aspect of the consultant's tasks. The position taken by this
research is that PSYSAFTY is the necessary variable from which clients
may confront their perceptions and presumably move toward more
accurate perceptual awarenesses.
1.1B PRCVRLTY

The consultant may begin work with a client by efforts to

find out the systemfls present level Of functioning. For example,

46

the question "Why are you here?" or "Why did you call me in?" might

be used as early focal points to elicit information. In the process
Of facilitating information flow by such questions, the intervention-
ist begins the diagnostic process. Clients may respond to ambiguous
questions such as these in a variety of ways. The manner chosen by
the client may reveal a great deal about the intellectual level of

the system, its values, and its means Of dealing with emotional
issues. In one group, for example, all communication may flow through
the supervisor; in another, an atmosphere of fear and cover—up

may be apparent; more rarely, norms of Openness, trust, and helpfulness
may predominate. Sometimes, the non-verbal cues are as important as
the verbal ones. If, for example, the group's overt statements are
clear and consistent, but they are staring at one another in a hostile
manner as they speak, this inconsistency says something important
about the group.

Consultant activities in this area will depend on a number Of
variables, including the interventionist's perceptiveness and diag-
nostic abilities. The client, however, should experience the sessions
as drawing them out, helping to make their assumptions explicit,
providing an opportunity to "level" with one another, etc.

1.1C PROCSFAC

 

One means of facilitating information flow is the analysis of
group process variables. Schein (1969b) has provided an extensive
review of the elements of process consultation. He defines process

conSultation as ...a set of activities on the part of the consultant

which help the client to perceive, understand, and act upon process

47

events which occur in the client's environment." (1969b, p. 9)
"Process" elements of human interactions address their manner of
occurrence, the "how" rather than the "what" of communications.
Because the elements of processing have been detailed elsewhere

(by Schein and others), they will not be elaborated here. Process
approaches are integral to many intervention schemes. Sashkin,
Morris, and Horst (1973) note, for example, that Argyris' theory of
intervention assumes that needed data are within the client system
and the problem Of the consultant is to free communication blockages
or develop new communication links.

The present approach views process intervention as a sometimes
useful tool in organizational consultation, but not as necessary or
sufficient. This research takes the position that learning how to
think conceptually, or to know an organization's needs, or to respond
to an increasingly demanding external environment do not come solely
from free communication. Process interventions may help facilitate
perception exchange, but are unlikely to resolve complicated organi-
zational problems.

1.1D LINKAGEN

 

The consultant can also facilitate information flow by serving
as a communications link between various parts of the organization.
There are various ways that the interventionist can do this. In the
least controversial way, conflicting parties might be brought together
(e.g., Beckhard, 1967). Others approach this function by directly
communicating views from one Person or group to another within the

organization. The latter approach is rejected by Argyris (1970)

48

who advocates strict confidentiality of information shared with the
consultant. However implemented, the LINKAGEN concept puts the
consultant in an active communications exchange role.

The OD literature has given some attention to the concept of
linkage. Lippitt et a1. (1958) view one important helping role as
being "mediating and stimulating new connections within the client
system." They point to different ways that this can be accomplished,
e.g., collecting information and reporting it back to the representa—
tives of the group as a whole. Buchanan (1967) extracted "linkage"
as a key factor in distinguishing successful and unsuccessful organ-
izational change efforts. Ferguson (1968) noted the consultant's
role as "linking agent" and "communications conveyor." The former
refers to the role of noting instances where the work relationship
requires a collaboration among groups, in situations where that collab-
oration does not presently exist, and structurally intervening to
bring the groups together and confront their differences. In Ferguson's
latter case, the consultant himself serves as purveyor of information.
The concept is also discussed by Sashkin, Morris, and Horst (1973).

1.2 Problem Conceptualization and Re—definition

 

Once information or attitudes have been elicited, the next
step in the model calls for the conceptualization and definition or
re-definition of the nature of the problem(s) affecting the system.
Sofer (1961) has remarked that the assembly or elicitation of infor-
mation is Of less importance than problem conceptualization and re-
definition. As he notes, the extricated facts and Opinions may be

new only to the consultant. Seashore and Van Egmond (1959) also cite

49

the importance Of this task, referring to it as "re—defining the
situation." Just as in psychotherapy, where reframing and inter-
pretation are often regarded as being among the therapist's major tools
(e.g., Paul, 1973), so problem conceptualization and re-definition
are viewed as critical OD consultation tools.

Three consultant activities are Operationalized dimensions
associated with this phase:

A. Identification of discrepancies in perceptions and assump-

tions (DISCREP)

B. Focusing on the needs of the organization (ORGNEED)

C. Focusing on external reality issues (EXRLTY)

D. Theory input (THRYINPT)

1.2A DISCREP

 

The fact has already been discussed that much initial material
in consultation may be irrelevant, irrational, and confused. In this
phase, the consultant helps make the contradictions explicit and
helps clients to think through erroneous assumptions they may be
making. An example of the latter is the case of two groups in an
organization, each of which assumes the other is responsible for a
certain organizational task; helping them see the incompatability
of their assumptions might be a task of the consultant.

Challenging clients' fundamental assumptions can also be a potent
aid in helping clients to conceptualize their problems. For example,
a supervisor may complain that he feels his employees are not "loyal"

enough. Others may lament the low quality of young help they are

getting. In both these cases, the consultant might intervene to

50

question the client's assumptions: e.g., what is loyalty? where
does it come from? what supervisory policies are most likely to
result in loyalty?; or, in the second example, is the organization
getting the best help available in the local labor market? what would
be required to attract and keep better talent? how can the firm
best nurture and develop the people resources it presently has?
These are questions that force the client to make explicit fundamental
assumptions, and also to re—conceptualize what is viewed as being
a problem. (Because of the ambiguity Of this task, it is often
experienced as being frustrating by the client.)

This approach has not been adequately addressed in the literature
to date. Sofer (1961) speaks of the need to "redefine assumptions

and expectations,’ although only as an attempt to clarify contradictory
expectations among different individuals and groups. Similarly,
Ferguson (1968) writes of the consultant's helping to "clarify in
the formulation of issues," but he apparently views this primarily
as a general exercise in understanding issues more clearly, rather
than reconceptualizing them. Although he has not specifically
addressed consultation in his writings to date, management theorist
and consultant Peter Drucker seems to rely heavily on this approach
in his writings (e.g., 1974).

This consultation role can be very difficult for a consultant.
It is true, as Lippitt et. a1. (1958) note, that some consultants
intervene from an assumption of problem generality and may approach

all organizations as if the problem area which is their specialty

isthe source of the client's difficulty. It is hard, however,

51

to see how such an approach could be more than accidentally success-
ful, i.e., the presenting or underlying problem happens to match the
consultant's specialty. That such "packaged" approaches to organiza-
tional change flourish is as much a function of the recipients of
help as its purveyors. As Strauss puts it: "American companies are
suckers for gimmicks...Companies try one attractive package after
another, just as long as each promises a painless solution to their
problems." (Strauss, 1976, p. 669) What is being put forth by the
COP Model is an approach that begins with the client's perceptions of
reality, and moves beyond these perceptions to core, underlying
issues.
1.2B ORGNEED

The (external) consultant has the unique Opportunity of viewing
the entire organization as a system. It is precisely the external
focus that makes the contribution so important, and the possibilities
of relatively undistorted perception by the consultant more likely.
In a real sense, the consultant's client is the organization, not
just the individuals within it (Beckhard, 1969; Huse, 1975). Thus,
in terms of helping the organization identify and understand its
needs, the consultant can provide organization-wide perspectives
that can serve as a superordinate background for the resolution of
problems.

The consultant's tasks in this dimension is to help the individuals
focus on the broader picture, the overall organization and its needs.
Particularly in less healthy organizations, the territorial imperative

is the standard, and within-company competition, rather than

52

cooperation toward common goals, is the rule. (Steele, 1975, has
introduced the concept Of Orgol--organization overlearning--to refer
to the use of strategies suitable to one area in a different, inappro-
priate one, e.g., in—house competition rather than OOOperation.)

By shifting foci to the relevant needs of the organization, the
interventionist can shortcut unproductive, no-win interactions.
Discovery of commonality of fate and mutual interdependence can be a
powerful motivator, a perception supported by the social psychological
literature (e.g., Sherif and Sherif, 1953; Latane, Eckman, and Joy,
1966).

1.2C EXRLTY

 

Another factor which can assist in problem re-definition is
helping the client become more aware of the relevant external realities
that affect a problem, i.e., factors from outside the organization
itself. The consultant may challenge the system regarding the extent
to which it has adequately confronted the demands of its external
environment. In one manufacturing company, for example, the chief
of sales consistently issued production orders to manufacturing
departments based primarily on reactive data generated from recent
market sales. The consultant's persistent inquiries regarding what
systematic customer data was available pointed to a severe information
gap with external factors that affect the organization.

The external consultant is always marginal to the organization.
(Argyris, 1970) He or she exists only temporarily in the organizational
environment and is always to some extent an "outsider." While this

position can be threatening to clients (and consultants), it can also

53

provide a valuable perspective for them. Helping the client learn
more about the demands of the organization-environment interface is
an important service the consultant can provide.

Lawrence and Lorsch (e.g., 1967, 1969) have identified the
importance of this boundary interface to organizational development
efforts. They stress the need for organizations to match the demands
imposed by the environment. Similarly, one of the few characteristics
distinguishing Buchanan's (1967) successful and unsuccessful OD
change projects was the presence in the former cases of a linkage
function to the external system in which the organization operated.
Lippitt et. a1. (1958) discuss the concept of distorted perceptions
by the client of external reality, but do not feel that such distortions
are commonly experienced by organizational entities. They do agree,
however, that the consultant can provide valuable external perspectives.
1.2D THRYINPT

Part of the consultant's contribution in the area of problem
conceptualization or redefinition may be theoretical. This may
be explicit in the sense of didactically presenting theories of
human behavior or organizational effectiveness, or indirect, presenting
theoretical views by carefully selected questions, or other such '
approaches. For example, an organization struggling with problems
of apathetic workers may be presented a theoretical explanation or
model of apathy, or may be asked to think conceptually about the
causes of apathy.

' Almost all approaches to OD assume some theoretical base about

how organizations and individuals function. Some (e.g., Argyris)

54

are more carefully developed than others. Here the focus is less upon
the content of specific theories, than on the process of theory input.
The COP Model does not advocate a specific theory about human behavior
since it is intended as a general approach to organizational inter-
vention that would be applicable in many different settings and with
many different problems.

In this regard, Buchanan (1967), in a review of case studies of
organization development, concluded that a distinguishing difference
between the successful and unsuccessful cases was the introduction of
an appropriate model by the change agent. The successful studies
introduced a new model which the client could use as a basis for
developing improvement goals, and which was seen by the clients as
relevant to the problem—solving processes of the organization. This
suggests that the specific content of the change agent's model may
be of less relevance to effective consultation than that the organi-
zation be able to relate it to its own problems, and its means Of
coping with them. A possible implication of this is that the inter-
ventionist might best begin interventions where the organization is,
rather than superimposing a diagnostic model which, even if "correct,"
is rejected by the client as irrelevant or incomprehensible. In
other words, the consultant may need to juxtapose theoretical contri-
butions in a way that is compatible with how members of an organization
see their working environment.

2.0 Assessment of the Need and Capacity for Changg (Confrontation)
' Definition: Helping the client system determine if it is able

and willing to change, and if so, in what directions.

55

The role of confrontation in the consultation process is
operationalized here as the assessment of the need and capacity
for change. If the client system has succeeded in identifying that
it has a problem and in isolating the general parameters and nature
of the problem, the next step is deciding what, if anything, it is
going to do to change. In general, the OD literature conceptualizes
this as "motivation to change" (e.g., Lippitt et. al., 1958). A common
contention is that if the client is to successfully carry out change
efforts, there must be a fairly compelling dissatisfaction with
the present way of Operating. For example, a potential psychotherapy
client's felt hurt on intake is commonly used as one index of the
likelihood of being able to benefit from therapy.

But unlike psychotherapy, where early determination of change
motivation is Often essential, confrontation of the OD client's
willingness and capacity to change may take some time. Many organiza-
tions, for example, approach installation of an OD system like the
Scanlon Plan as if it were a simple incentive program. In working
with a prospective client, the consultant may continually inquire "why
a Scanlon P1an?", thus leading to problem identification. Subse-
quently, the client system must be confronted with its ability to
tackle the problems at hand and whether the Scanlon Plan is an
appropriate vehicle for doing so. Focus on the Scanlon Plan is
secondary; confrontation of the system's willingness to address its
problems is primary. (Frost, Wakeley, and Ruh, 1974)

'Motivation and ability to change, then, are explicitly confronted

with the client after problem awareness and identification have

56

been attempted. Beckhard (1959) observed that the consultant needs
to continuously assess the readiness and capacity of the client system
to change, but he did not specify how the client should be involved
in his process. Ferguson (1968), despite developing a long list Of
consultation roles, does not include the need to confront the client
with its own behavior and what is going to be done about it. Gibb
(1959) stresses the importance of examining the client's motives for
change as part of the diagnostic phase. R. Lippitt (1959) also
emphasizes the importance of assessing change motivation, though he
does not adequately concretize how and when this needs to be done.
The vicissitudes of client system motivation, the fact that expecta—
tions will fluctuate from expecting "instant cures" to feeling that
the problems are insoluble, the movement from euphoria to paranoia,
are highlighted by Sofer (1961).

When the problem dimensions have become explicit, the client
system is faced with a critical choice--is there a reason to do things
differently, given the attendant pain that can accompany change?

If this choice is not clearly confronted, passivity, or a decision

to ignore problems may result. This may be an apprOpriate decision
in many settings, but should be explicitly made. Confrontation, then,
is viewed as a transitional stage between accurate perception of
situational reality, and the appropriate response to it.

Three interventionist tasks are identified as relevant to this
phase:

2.1 Goal identification and evaluation (GOALFOCS)

2.2 Identification of reasons for changing (WHYDIFF)

2.3 Identification of the implications of change (WHATDIFF)

57

2.1 GOALFOCS

 

Often, client systems profess to be committed to certain rather
noble-sounding goals, but their behavior is clearly discrepant with
them. For example, organizations often claim to be interested in being
"fair" to employees, or will assert that they want subordinates to
bring problems to them, no matter how difficult. Despite these overt
statements, the consultant may Observe that "fairness" is translated
as patronization, or that employees do not surface problems with their
supervisors because the bosses are perceived as being too threatening
or there are penalties for those who app honest and Open. Confron—
tation of an organization with such discrepancies can be a focal point
for consultant intervention: is the organization serious about its
professed goals, is it genuinely committed to their accomplishment?
Thus, the consultant may help the organization identify differences
between professed goals and actual achievements, or its stated beliefs
and its actual behavior. (Blake and Mouton, 1964; Ferguson, 1968).
Argyris (1976a, c; Argyris and Schon, 1974) recently identified the
pervasiveness of the discrepancy between claimed and actual values
and behavior (and the enormous difficulties of change).

A second focus appropriate to this intervention is the goals
themselves: what goals are presently held-~implicitly as well as
explicitly? Is the need for these goals clear and convincing? Are
the goals capable of achievement, given the client system's resources?
Are they challenging enough? At a conference in which a corporate
mandate had been presented, for example, one department's representa-

tives were asked whether the articulated profit goal was achievable.

58

Actually, they reported, it was not too discrepant with what was
presently being earned. Was it, then, the consultant inquired,
challenging enough? The group replied that meeting the figure would
actually be something Of a challenge since it had never before actually
achieved that level. Why the figure was necessary for the organiza-
tion, and whether it could actually be achieved, were unclear to

the group.

This second focus of goal confrontation might center on the
creation by the consultant of "images of potentiality," a term apparent-
ly coined by Ronald Lippitt (1959). In essence, these are ideals
about what the client system might become; the discrepancy between
this image and present reality might potentially be a motivator. This
technique might be applied by presenting examples of other organizations
who have struggled with similar problems and have reached some degree
of success in their resolution. Whether the client system is challen-
ged by such vignettes or dismisses them as irrelevant can be one
diagnostic indicator of its willingness and readiness for change.

2.2 WHYDIFF

 

Frost maintains that if organizational reality has been adequately
identified, the need for appropriate action will Often follow fairly
easily. "The identification of the current situation usually helps
define the need for change in the organization." (Frost, Wakeley, and
Ruh, 1974, p. 57) This is consistent with Maier's (1963) concept of
"problem mindedness," which suggests that concentration on problem
specification instead of on solutions, as an effective vehicle to prob-

lem resolution.

59

Ultimately, the system must come face to face (i.e., be confront-
ed) with its motivation for change. If there are no convincing reasons
for behaving differently, permanent change is unlikely. As noted in
the earlier discussion, this is usually conceptualized in the liter-
ature as "motivation for change." The present focus departs from
this literature by suggesting that this is not the first consultant
task, but must flow out of identification of current reality. It
also suggests that the "need for change" should be explicitly confronted
with the client.

2. 3 WHATDIFF

 

The conceptualization behind the WHATDIFF variable is that the
client system must not only have compelling reasons to initiate change,
but that it must also delineate the steps that it would have to take
if it were to change. This requires the system to spell out the
implications of changing, the specific things that would have to
happen to implement the change(s). In one organization, for example,
data was persistently uncovered indicating that a central production
function was not being managed adequately to meet the system's demands.
Although the chief executive was aware of the problem, he continually
refused to resolve it because the head of the function was a family
relative. The executive saw the need for the change (as did the
rest of the organization), but was unwilling to take the action that
would be required, even though there was a clear reason for change.

In some instances, the system may need assistance in learning
that it does indeed have the capacity for change. In the clinical
setting, having worked through a client's problems of isolation and

differentness, the therapist may then help the client discover that,

60

despite her awkwardness and shyness, she does have the ability to
relate to peOple. Organizationally, a company may experience extreme
reluctance to take a needed risk to inaugurate a costly marketing
program; the consultant might help the organization discover that

it has the ability to move in new directions.

There is no literature that approaches this concept of WHATDIFF
in the way presented here. The clinical psychotherapy literature
comes closest in its discussion Of technique, but there is apparently
no theoretical approach consistent with this concept as it has here

been integrated with the foregoing variables.

3.0 Problem Ownership (PROBOWN) (Consolidation)

 

Definition: Acceptance of respOnsibility for the problem and

its solution by appropriate organizational resources.

Few theoretical models of organization development have directly
addressed the issue of problem ownership. Here it is hypothesized as
an essential component of effective problem resolution.

The essence of PROBOWN is the acceptance of responsibility. Glass-
er (1965) has identified responsibility as a crucial element of his
"reality therapy." In other psychotherapeutic approaches, ownership
deals with such issues as the patient's coming to grips with his
ultimate metaphysical aloneness, his self—responsibility for what he
does with his life now, no matter how traumatic the past (e.g.,
Branden, 1975). In organizational consultation, problem ownership
refers to the acceptance of responsibility for what happens in and
to the organization by the appropriate resources. It is not just

that the locus of responsibility and choice belongs in the client

61

system, as Argyris (1970) has detailed; but further, that the owner-
ship is accepted by the appropriate parties. One attribute of the
Scanlon Plan, for example, is its system-wide focus on the needs and
problems of the organization through its committee structure (Frost,
Wakeley, and Ruh, 1974). Implicitly, this system reinforces problem
ownership by providing a company-sanctioned means for this to occur.
The Scanlon structure says to employees, in essence, that the system
has problems, it needs to know about them, it needs everyone's
assistance in their resolution.

Without problem ownership, solutions that get generated and
accepted will tend to be viewed as external, and the chance of their
being suitable to the underlying problem, minimal. If problems are
not Owned by the appropriate source, the developed solutions will
tend to be someone else's problems. Argyris (1971), for example,
has cited the willingness of top management groups to mandate team
development or T-group activities for lower echelons of management,
without having had the experiences themselves. Similarly, Strauss
(1976) has cited the U.S. Army example in which OD activities that
were not adequately understood or adopted by commanders at the top
organizational levels were Ordered to be implemented at lower levels.
Finally, it is not surprising that clients who enter therapy in lieu
of jail sentences (e.g., sex offenders) or at someone else's insistence
do not tend to be successful clients in psychotherapy.

The CD literature does not overtly address the issue of problem
ownership in much detail. Argyris (1970) postulates "internal

commitment" as his third consultation task, but this seems to be

62

directed to the solution generated rather than to the problems to

which the solutions are addressed. Presumably, however, his second
step, free choice, and his first, the generation of valid and useful
information, create conditions where problem ownership is necessary.

In other writings, Argyris (1971) has suggested that personal causation
and responsibility are moderators of whether externally or internally
induced change is appropriate. In still another work (e.g., 1968), he
has suggested that organizational members will take increasing respon-
sibility when certain conditions are present (directly verifiable
information, minimally evaluative feedback, effective group functioning,
and psychological success). Finally, Argyris and Schon (1975) have
developed a theory of education that includes as one focal point the
necessity for "personally caused experience," which requires individuals
to themselves produce the behavior from which they can learn, implicitly,
a self-responsibility-based approach.

Implicit recognition of the need for problem ownership is also
provided by Schein, who notes that the process consultant's fundamental
task in the diagnostic phases of his work is "...to get the group
to pparg in diagnosis and to help the group lgapp to diagnose its
own process" (Schein, 1969, p. 44; emphases his). Huse (1975) notes
that the resistance to change literature documents individuals' ten-
dency to oppose changes which they have not been part of creating.
Similarly, Menninger (1958) suggests that, in psychotherapy inter-
pretation, it is the patient and not the therapist who should be able
to take the "credit" for attained insights. Chein et a1. (1948)

Observe' that diagnosis does not always lead to action, often because

63

Of insufficient involvement by those who must make the change.

In short, PROBOWN is viewed as a precursor of effective problem
solution. Effective problem solving requires the relevant organization-
al resources to accept their own role in generating the difficulties,

and their own responsibility for change.

4.0 Joint Authorship of the Solution (Integration)

Definition: Development of solutions to the problem and its

underlyinngimensions, with internal commitment by the client

system.

"Integration" as the terminal change goal is Operationalized as
the "solution" aspects of consultation. Effective change results not
just in solving the presenting problem, but in resolving it in such
a way that it does not recur (Argyris, 1970). This may Often involve
solution of the underlying problem beneath the surface issues.

Many OD theorists and practitioners have focused on problem solu-
tion aspects of the consultant's role. G. Lippitt (1959) and Lippitt
et. a1. (1958) view this as a primary role for the consultant.
Specifically, they point to the consultant's need to assist in the
examination of alternative solutions and goals, and the transfer of
these intentions to actual change efforts. These tasks are also noted
by Gordon Lippitt (1959). Often, this portion of consultation is
conceptualized as the "action" stage (e.g., Kolb and Frohman, 1970).
The distinction drawn here is the essentiality of integrated problem
solution, i.e., integrated with the basic reasons for system difficulty

in the first place. Sofer (1961) posits that working with one problem

without recognition Of the larger problem of which it is a part, will
almost certainly not succeed. Kolb and Frohman (1970) note, for exam-
ple, that problems in action implementation are usually traceable
to unresolved issues in the earlier phases of consultation.

Analagously, the problem of integration in insight psychotherapy
requires the resolution of unresolved conflicts underlying present
behavior, and their consolidation with the problematic behavior. In
integration, these foci are merged with the client's increasing
awareness of his needs, and his own responsibility for meeting them.
Organizations present a larger, social context, however, and problem
solution requires participation by all relevant parties to an issue.
Clearly, who is relevant will vary with the nature of the difficulty.
There is considerable research evidence, however, to support the notion
that participation is a powerful psychological variable that helps
overcome resistance and helps to develop internal commitment to
solutions generated (e.g., Coch and French, 1948; Lewin, 1951; Lowin,
1968; Katz and Kahn, 1978).

Returning to the OD literature, Greiner (1967) isolates six
phases of successful change efforts. His fourth phase is the invention
of new solutions and development of commitment to new courses of
action. Consistent with his central organizing concept of power redis-
tribution, Greiner contends that in the successful cases he analyzed,
the searches for creative solutions were widespread, and the solutions
generated through shared power. Lewin (1951) considered "freezing"
to be an essential and last step in the change process. However,

neither he nor subsequent researchers who adopted his model have

65

effectively operationalized the freezing process. An exception to
this generalization is Frohman's doctoral dissertation research (1970),
which operationalized "freezing" as "integration." However, Frohman
defines integration primarily as movement to more participative
management, a more solution—specific conceptualization than the
present one. Bartlett and Kayser (1973) speculate that the third
phase of Lewin's model is the most critical: they conclude that many
change efforts fail, despite successful unfreezing and movement,
because of inadequate attention to refreezing. Schein (1961) also
uses Lewin's model, and considers the central task in the "refreezing"
stage as integration of the changes made with the environment.

In short, many varying approaches to consultation view problem
solution and stabilization as important elements of organizational
change efforts. Most do not offer a rationale for how the consultant
assists in this process of why integration with the previous consul-
tation tasks is essential. Furthermore, the conceptual base of
most theories of change is fairly primitive regarding this phase, as
with the others. Most rely, implicitly or explicitly, on Lewin's
three stage model of change, but without adequate integration with the
model itself or of subsequent research. There are also typically
few theoretical or pragmatic specifications of who should be included
in the generation Of solutions, to what extent, and why. Finally,
the exact role of the consultant in the process is typically unspecified.

Four specific consultation variables are suggested in this

phase of the model:

66

4.1 Assurance that apprOpriate resources are included in the
generation of solutions (SLTNRESRCS)

4.2A Assurance that the proposed solution is relevant to the
actual problem (SLTNRELV)

4.2B Assurance that the pr0posed solution is consistent with
the demands of external reality (SLTNLINK)

4.3 Implementation and evaluation of the solution(s) (SLTNIMEV)

413mmmm§

 

As noted above, participation is an extremely important variable
in developing internal commitment to solutions generated. If peOple
are to effectively commit to solutions, it helps that they have
been a part of the process that led to what they are being asked to
commit. The persistently convincing evidence (e.g., Katz and Kahn,
1978) of the pragmatic effects of participation will not be reviewed
here. The point is that in problem solution, the consultant can
help to assure that relevant organizational resources are participating
in solution development. Obviously, all organizational members cannot
be included in all decisions made in the organization. The consultant
can help assure the relevance of those who are included in solution
develOpment, and help the client system assess who else might be
appropriate. As already noted, the Scanlon Plan institutionalizes
participation through a committee structure, and thus might be a
valuable aid to a consultant.

A second aspect to the consultant's role in SLTNRESRCS is the

evaluation of organizational personnel as to their relevance for

67

various organizational assignments. Clients may ask the consultant,
as a professional behavioral scientist, to assess the organization's
specific personnel needs, or the suitability of certain candidates
or individuals to the tasks at hand. Such a position of evaluation
by the interventionist may create role conflicts for the consultant
using the COP Model, since the task of evaluation may jeopardize
PSYSAFTY dimensions. There has been little in the theoretical or
applied OD or clinical literature about how consultants might best
evaluate personnel and still maintain a position of helpfulness

to the individuals within an organization.

4 . 2 SLTNLINK ; SLTNRELV
There are occasions when the external frame of reference of the
consultant can be helpful in demonstrating that what has been pr0posed
to solve a validly identified problem is not relevant to the "real"
problem. An organization, for example, that correctly identifies
a motivational problem in the firm stemming from employees' feeling
under-utilized may, in resorting to Old solutions, try to short
circuit the underlying problem by pr0posing salary increases for
these workers. If, however, these are employees with higher order
need strength (Hackman and Lawler, 1971), whose jobs are intrinsically
uninteresting, increased pay will not solve the basic problem. The
consultant can help re—direct the focus to the actual problem at hand.
The examination of alternative solutions is mentioned by R.
Lippitt (1959). as a phase in consultation. Argyris (1970) incorporates

this phase into his "free choice" consultant dimension, apparently

68

believing that the onus Of responsibility for ascertaining the
relationship between the proposed solution and the sought goals must
rest with the client. Huse (1975) also comments on the relevance of
this consultation activity.

Secondly, the consultant can provide an external reality focus
to help assure that the proposed solutions are appropriate to the
environment in which the organization must function (SLTNLINK). If
an organization identifies a need to increase its wage structure to
be competitive in the labor market, but does not tie this need into
the equally pressing need to lower prices to remain competitive, the
"solutions" generated by the firm may be disastrous. The intervention-
ist can assist in this phase by focusing organizational members'
attention on the relevant demands of the environment. This focus is

particularly well addressed by Lawrence and Lorsch (e.g., 1969).

4.3 SLTNIMEV

 

The final phase of the Operationalized COP Model is the actual
implementation of the solutions generated. The relevant consultative
roles in this phase will vary depending on the issues in question.

For example, in the process of educating an organization about the
need and relevance of a participatory management scheme, a presenta-
tion to workers might be made by management with the assistance of the
consultant. In another case, the consultant might be helpful in
finding resources appropriate to the needs of the organization.

Consistent with the philosophy Of consultation implicit in this
discussion, the consultant's role in implementation of solutions is

seldom direct. If the previous stages have been successfully implemented,

69

ownership will be where it belongs (i.e., in the client system) and
the consultant will primarily help in an advisory, sometimes suppor-
tive, role. Just as the therapist cannot actually assist a client in
beginning to look for new, more intimate interpersonal relationships,
so the organizational consultant cannot do the job for the client
system without undermining its own sense of self responsibility.

Lippitt et. a1. (1958) discuss the consultant's role in initiating
and sustaining change primarily as giving support during the process
of change itself. They also document the role the consultant can
play in stabilizing and generalizing the change process, a dimension
also noted by Beckhard (1959). The distinction Of the present model
is its integration of the solution implementation aspects of consult-
ation with the problem definition and ownership phases preceding it.
In the COP Model, the consultant does not impose or argue for solutions
as a first step, but only deals with "solutions" as they relate to
specific client problems and needs.

Finally, the interventionist can assure that the changes made
are properly evaluated. Sometimes this might be approached as simply
assuring that the system makes evaluations of what has been undertaken;
in other cases, technical assistance might be provided in designing
evaluation research.

These, then, are the major features of the operationalized COP

Model of consultation.

CHAPTER III
METHODOLOGY

This study explores several major research questions:

1) The construct validity of the proposed Clinical-Organizational
Psychological (COP) Model;

2) What dimensions, other than those developed in Chapter II, are
also important to clients;

3) The relationship between the COP Model dimensions and a criterion
of consultant effectiveness;

4) Differences in perceptions between experienced and inexperienced
clients; and

5) The relevance of certain process variables to consultant effect-
iveness.

It has already been noted that almost no literature exists to
date that empirically investigates clients' perceptions of the
consultative relationship. Contrary to more typical research in this
area, this study looks in depth at one intervention approach, attempt-
ing to identify relevant consultative variables for further, more
controlled research.

More specifically, this study is an in-depth research investi-
gation of the consultative relationship between a variety of clients
and one senior Organization Development (OD) consultant, Dr. Carl
Frost. The elaboration of Frost's model presented in Chapter II is
tested against actual client experience. While it is clear that there

are obvious limitations to generalizability from studying only one

70

71

consultant, even from the perspective of many different clients, there
is also value to a careful identification and understanding of what a
senior, nationally recognized, professional practitioner does in the
field, and how these activities are affected by characteristics of

the client. It is the author's contention that the field of CD must
first identify the relevant consultation variables before more defin-
itive and complex statements Of interrelationships can be ascertained.
So, this research is intentionally exploratory.

The process of my elaboration Of Frost's fundamental consulta-
tive model suggested that the four basic consultative tasks are
multidimensional, each dimension representing a separate but related
group of consultative duties. This resulted in the four operational-
ized variables and 16 more specific variables related to each of
the broader groupings.

To test whether these dimensions show empirical validity, two
sources of data were sought from Dr. Frost's clients. First, a
critical incidents methodology (Flanagan, 1954) was employed whereby
clients were asked to give examples of consultative behavior that was
especially helpful or especially non-helpful to the client. The
format for these questions is presented in Appendix A. This data
was elicited prior to more specific questions in the instrument to
prevent prejudicing the "set" of the respondents, or the examples that
they reported. A second source of evidence regarding consultative
variables was a series of items developed to specifically test the COP

Model's content dimensions and certain processing variables, as well

72

as a measure of (perceived) consultant effectiveness.

The procedure used in the development of this instrument was
as follows. The consultative dimensions were identified and defined
in the elaboration of the COP Model (Chapter II). Questions were then
written by the author, attempting to measure the various aspects of
each dimension. Then, a retranslation technique (Smith and Kendall,
1963) was used in which Frost, two other psychology professors, and
four advanced graduate students in psychology categorized the proposed
items into the defined categories. Only those items showing inter-
rater agreement in classification greater than or equal to 70% were
included in the final questionnaire. The 17 content scales and the
six processing scales are those presented and defined in Chapter II.
The final items were then randomized, with about one third of them
phrased negatively as an attempt to reduce response bias. A Likert
response format was used for each item, with the following response

choices:

DEFINITELY FALSE
MOSTLY FALSE
SLIGHTLY FALSE
UNCERTAIN
SLIGHTLY TRUE
MOSTLY TRUE
DEFINITELY TRUE

VOU-PWNH
I

Additionally, several demographic-type items were included in
the questionnaire, as well as a criterion measure. The former
attempted to assess individual and organizational variables that
might be expected to moderate the client-consultant relationship.

These included: 1) the employee's organizational level; 2) his or

73

her organizational position; 3) the length of time employed with

the organization; 4) the number of consulting sessions experienced in
the last year; 5) the total number of consulting sessions experienced;
and 6) the pp; of consultation, i.e., whether the client had worked
with Frost primarily individually or in group settings.

It was also recognized that a criterion Of overall effectiveness
would be desirable to distinguish between those §s finding the
consultant helpful and who were able to benefit from his services and
those who did not. While a behavioral measure would have been preferred,
this was not possible due to the financial and time limitations of
the study. Instead a perceived measure of evaluation was employed.
The clients were asked to rate the consultant on his overall effect-
iveness to their organization and to them personally, and on a number
of "pseudo-behavioral" measures, e.g., whether the client would recommend
working with the consultant to a friend within the company. The
specific items included in the criterion measure are presented and
discussed in Chapter IV.

Finally, an attempt was made to assess the perceived helpfulness
of the consultant on the various consultative dimensions. Initially
it had been planned to have the clients respond twice to each item,
once rating whether the consultant performed the specific dimension,
and secondly, if he did, the perceived extent of his helpfulness when
doing so. However, the already long questionnaire length made this
unfeasible, as well as the fact that it would create a number of

complicated data analysis difficulties, and so respondents were simply

74

asked to rate, in a single question for each dimension, the consultant's
overall helpfulness to them in performing the dimension.

Finally, the entire instrument was pre-tested using five clients
at different organizations and at different organizational levels,
each of whom knew the consultant well. Minor changes in the typography
and wordings were made, and the color-coded Dimensions of Consulting

(DOC) questionnaire was then assembled in its final form.

Data Sources

Because both experienced and inexperienced clients (EXPCLIs and
INEXPCLIs) were being studied, two groups of clients and client
organizations were sought for the study: those with long and short
term relationships with the consultant. Frost provided the names of
virtually all clients with whom he had had long—term relationships.
These were all organizational personnel from companies with whom he
had worked for extensive time periods. The operationalized definition
of a short-term consultative relationship was somewhat arbitrarily
set by Frost and the author to be fewer then five sessions with
the consultant. New companies, with whom the consultant had just
begun a consulting relationship were utilized for the latter group
of subjects. Respondents who had had.fewer than five sessions with
Dr. Frost, but who worked in a company with whom he had had a long-
term consulting relationship, and (more rarely) those who had had
greater than five sessions but worked for a company which had just
established a consulting relationship with Frost, were not included

in the study. This is because these groups were a) small in number

75

and b) represented a potential source of response bias, possibly
representing different perspectives than their organizational colleagues,
or possibly biased in their perceptions by the opinions and experi-
ences of these personnel.

The companies used in this study were all manufacturing organi-
zations with small to moderate sized work forces, primarily located
in Michigan, Illinois, North Carolina, and Tennessee. Companies with
established consulting relationships all either had a Scanlon Plan
or were in the final stages of implementation. The "new" companies
generally had contacted the consultant about installing a Scanlon Plan,
but were all in the exploratory stages of investigation and did not

have a Scanlon Plan in Operation.

Questionnaire Adpinistration

Questionnaires were delivered in person to individuals in
organizations in the local area. For other companies, the instrument
was mailed. In both cases, a cover letter was enclosed explaining
the nature of the project. This letter was on Scanlon Plan Associates'
stationery and was signed by the President of this group. In companies-
with longer-term consultative relationships, a highly-placed company
manager, usually the company's president or chief operations officer,
co-signed the letter to lend local support and validation to the
project. A sample cover letter is included as Appendix B. Personnel
not responding after about three weeks were sent a follow-up letter;
if there was still no response after about three additional weeks, a

final follow-up letter was mailed, asking for return of the questionnaire

76

even if the respondent could not complete it. Copies of sample

follow-up letters are also included in Appendix B.

Data Analysis

 

The data from returned questionnaires was keypunched with the
assistance of a research assistant. The critical incidents data
was coded by the author, an advanced undergraduate psychology student,
and a graduate student in psychology. Each of these individuals was
trained in the definitions and conceptualizations of the COP Model.
Each critical incident (CI) was divided into units representing a
discrete instance of consultant behavior to enable each coder to
respond to the same behavioral dimension. After individual codings,
consultation was held between the three coders in cases of disagreement
on the coding. If the discrepancy could not be resolved so that at
least two of the three raters agreed on the coding, the CI was not
used in the data analysis.

The questions that had been developed to specifically measure
consulting dimensions defined in this study were analyzed using cluster
analysis. This was mandated by sample size limitations and by the
interpretative difficulties that arise with orthogonal factors. Using
oblique cluster analysis procedures, three criteria were employed
for scale development: 1) content validity of the items; 2) internal
consistency; and 3) analysis of interscale relationships.

Items were initially written and categorized using the retrans-
lation technique already discussed. Coefficient alphas were then

computed from the empirical data as a measure of internal consistency,

77

using the a priori categorizations. Items that lowered the scale's
internal consistency were discarded. Finally, the pattern of relation—
ships among the various scales was analyzed, and scales re-assigned

or dropped that did not demonstrate interpretable inter-scale relation-
ships. Because Of their greater knowledge of the consultant's behavior,
EXPCLIs were used as the sample for scale development. A separate
assessment of the statistical properties of the INEXPCLIs' data was

also performed.

Research Hypotheses

Specific, operationalized hypotheses that are tested in this
study are as follows:

Hypothesis 1: Dimensions specified in the COP consultative

model will be validated by client perceptions of the
consultant's behavior.

This hypothesis is basically a test of the COP model's empirical
validity. The first step in the assessment Of consultation variables
must be the development of instrumentation that is capable of measuring
hypothesized consultative dimensions. The validity of the model
rests first on the instrument's psychometric properties, i.e., does
it consistently measure what it purports to, is its content appro-
priate to the variables under consideration, and does it have a
predictable pattern of relationships among its variables.

Two data sources are used to test the model. First, the critical
incident data is coded to see if the examples of helpful and nonhelp—

ful behavior conform to the dimensions specified in the model.

78

Secondly, the DOC questionnaire items that were deve10ped to assess
specific variables under consideration, is assessed for its psycho-
metric properties, whether these items emerge as content appropriate.
internally consistent scales with predictable interscale correlations
(moderate to high within group correlations with lower correlations
among the scales associated with different major dimensions of the
model). Ideally, the instrument would then be compared with other
measures of consultative variables to further assess its construct
validity. However, the virtual non-existence of other consultation
assessment instruments to deal with similar variables makes this
impossible.

Hypothesis 1A: Dimensions not associated with the COP Model

 

will also be present in the critical incidents data.

This study is exploratory in nature, and it is recognized that
the rather crude level of psychological hypothesizing and instrument
construction will neglect variables that might also be relevant to
clients. This is an added.advantage of the critical incidents method
of gathering data: it permits the generation of whatever data is
relevant to the client, regardless of what has been thought to be
important by the researcher or the consultant. It has already been
noted that Yalom (1975) and others have demonstrated that there are
often dimensions that are important to clients that the therapist
had not predicted to be relevant or important. This hypothesis attempts
to capture similar phenomena with consultative clients.

The method used to test this hypothesis will be the assessment

79

of the residual cluster of critical incidents that are not codable
using the COP Model dimensions. These CIs will be reviewed and
classified according to whatever groupings appear to best summarize
the data.

Hypothesis 2: Clients with limited experience with the consultant
(INEXPCLIs) will perceive a different pattern of behavioral
dimensions performed by the consultant than those with
longer-term consultative experience.

Hypothesis 2A: Specifically, consultation dimensions associa-

 

ted with Problem Awareness and Identification (1.0) and
Assessment of the Need and Capacity for Change (2.0) will
predominate in the INEXPCLIs group; dimensions associated
with Problem Ownership (3.0) and Joint Authorship of the
Solution (4.0) will additionally be represented in the
EXPCLIs group.

These two hypotheses suggest that the two groups of clients, the
EXPCLIs and the INEXPCLIs, will differ in the consultative dimensions
performed by the consultant. Specifically, as a test of the COP
model, it is suggested that all of the model's dimensions should be
represented by the consultant in his activities with longer-term
clients, but that only those dimensions associated with the first half
of the model will be used by the consultant in the earlier stages of
the consultative relationship.

This hypothesis is an attempt to measure a more difficult-to-
test assertion, viz., that the consultant-client relationship changes

over time, and in a predictable fashion. The ideal manner of testing

80

this assertion would be a longitudinal study in which the same clients
repeatedly give measures of their perceptions of the consultant after
each session. Such a methodology was not feasible in this study due to
expense and time limitations, and so a cross—sectional approach was
utilized.

Multivariate discriminant analysis will be used to assess the
variables that maximally discriminate between the two groups. A step-
wise analysis procedure using Rao's V as an entry criterion (Klecka,
1975) permits entry of variables into a discriminant analysis based
on their ability to separate two groups. Multivariate and univariate
analyses will also be used conjunctively to understand the nature of
the group differences.

Hypothesis_3: The pattern of perceived consultation variables

 

will differ among those rating the consultant highly
and those rating him less highly on a criterion measure.
This is an attempt to assess group differences between those
rating the consultant highly on the criterion measure and those
rating him less highly. The purpose is to assess which variables are
most helpful to clients and which less so; in other words, which are
associated with favorable and unfavorable consultant relationships.
Because the two groups of EXPCLIs and INEXPCLIs are conceptually
distinct groups (Hypothesis 2), each will be analyzed separately.
The methodology used for the analyses will be a multiple regression
procedure, using the criterion as the dependent variable, and the

COP Model dimensions, as measured by the DOC structured items, as

the independent variables or predictors. The Beta weights associated

81

with each analysis will be analyzed, as well as the order in which
each variable is entered into the regression equation in a step—wise
entry procedure.

The question then arises as to whether the relationships that
emerge from these analyses will be moderated by any of the demographic
or type of consultation variables on which data was collected.
Specifically,

Hypothesis 3A: The relationship among the predictors and
criterion that are found in Hypothesis 3 will not be
moderated by the type of consultation (group or individual),
the frequency or recency of the number of sessions, the
incumbent's organizational level or position, or his
number of years with the company.

This hypothesis will be tested by entering these variables into

a multiple regression analysis first, assessing their overall relation-
ship with the criterion, and if significant, the specific nature of
their contributions, and the differences in the Beta weights associated
with the predictors with and without the potential "moderators"
included in the equation.

Hypothesis 4: Process variables associated with the COP Model

 

will be validated by client perceptions of the consultant's
behavior.
This hypothesis is simply a test of the content and construct
validities of the processing variables to be measured in this
study, i.e., the manner in which the content variables are executed

as opposed to the specific behaviors themselves. These processing

82

variables include:

COG: The consultant's ability to work with the cognitive

aspects of problems;

VAL: The consultant's ability to work with the value dimensions

of problems, i.e., issues of "right and wrong";

EMO: The consultant's perceived ability to work with the

affective, or emotional, aspects of problems;

BEG: Whether the consultant is perceived as beginning with

the client's perspectives, issues, and agenda, or his own;
BEN: The perceived interest of the consultant in the client;
i.e., whether he is perceived as wanting to be of help
to the client; and
COMP: The perceived ability of the consultant to be of help
to the client.

Questions were composed to assess these dimensions in the same
manner (and at the same time) as the original COP Model content
dimension items. Psychometric analysis of their validity will utilize
the same procedure as the major COP Model dimensions.

Hypothesispj: A different pattern of consultative processing

 

variables will be associated with EXPCLIs than with INEXPCLIs.

Hypothesis 6: A different pattern of consultative processing

 

dimensions will be associated with those clients rating
the consultant high and those rating him less high on a
criterion measure.

These two hypotheses are simply the analogues of Hypotheses

2 and 3, only regarding the processing variables rather than the

83

content variables. The methodologies used to test these two hypo—
theses will be identical to those discussed above. As with their
analogues, these two hypotheses are exploratory in nature and attempt
to search for group differences between EXPCLIs' and INEXPCLIs'

perceptions of the consultant on these dimensions.

CHAPTER IV

RESULTS AND DISCUSSION

Characteristics of the Samples

 

A total of 101 questionnaires were distributed to EXPCLIs and
155 to INEXPCLIs (Tables 2 and 3). Among the EXPCLIs group, the
overall return rate of "useable" questionnaires was 74.3%; adding
those who replied but could not complete the questionnaire, or who
returned it with so much missing data that it could not be used,
this figure increases to 91.1% (Table 2). A summary of the reasons
for questionnaires returned by the EXPCLI group that were unuseable
is provided in Table 4. Clearly, the chief reason was the perception
of too-limited contact (or too limited recent contact) with the
consultant. Overall, these results for the EXPCLI group indicate
a level of return sufficiently large to have reasonable assurance
of the adequacy and representativeness of the sample. It should also
be noted that two companies (C and I) were excluded from the data
analyses. The former was excluded because the consulting relationship
with this company was terminated shortly before this project began,
and because of the author's own consultation with the company, raising
the strong possibility of systematic bias. Company I was excluded
because it had only one respondent, providing too small a sample
to assure representativeness.

Also excluded from subsequent analyses was a group of seven
respondents who stated that they had had fewer than five sessions

with the consultant. As discussed in Chapter 111, this would have

84

85

Table 2

Distribution and Return of Questionnaires to EXPCLIs

 

Total No. No. Returned

NO . Returned ,

 

 

Company Distributed Useablei Unuseable NO Response
.2 .111122. 2.1.2. 2.11%

31 24 / 77.4 3 / 9.7 4 / 12.9

19 16 / 84.2 3 / 15.8 o/ 0.0

c3 13 11 / 84.6 2 / 15.4 0 / 0.0
D 9 6 / 66.7 2 / 22.2 1 / 11.1
E 9 4 / 44.4 4 / 44.4 1 / 11.1
F 8 4/ 50.0 1/12.5 3/37.5
G 3 3 / 100.0 0 / 0.0 0 / 0.0
H 8 6 / 75.0 2 / 25.0 0 / 0.0
I3 1 1 /100.0 0 / 0.0 0 / 0.0
101 75 / 74.3 17 / 16.8 9 / 8.9

 

Includes 7 questionnaires returned from clients with fewer than
five sessions, which were not used in the data analyses.

2 For an explanation of the reasons for being unuseable, see Table 4.

3 Questionnaires from these companies were not used in the statistical
analyses for reasons explained in the text.

86

Table 3

Distribution and Return of Questionnaires to INEXPCLIs

 

 

 

 

 

Company Eiiiiiifited N°tsiitiiied' N°' 3:322:31; N. Response
2. 1141381 P_ZJE 111111

J3 1 1 /100.0 0 / 0.0 o / 0.0
K 4 2 / 50.0 0 / 0.0 2 / 50.0
L3 1 1 /100.0 0 / 0.0 0 / 0.0
M 6 3 / 50.0 1 / 16.7 2 / 33.3
N3 48 8 / 16.7 5 / 10.4 35 / 72.9
0 4 4 /100.0 0 / 0.0 0 / 0.0
P 3 2 / 66.7 1 / 33.3 0 / 0.0
Q 9 7 / 77.8 0 / 0.0 2 / 22.2
R 3 0 / 0.0 3 /100.0 0 / 0.0
s 8 3 / 37.5 0 / 0.0 5 / 62.5
T 68 35 / 51.5 9 / 13.2 24 / 35.2

155 66 / 42.6 19 / 12.3 70 / 45.2

 

1 Includes 2 questionnaires returned from clients with greater

than five sessions; these were not used in the data analyses.
For an explanation of the reasons for being unuseable, see Table 4.

3 Questionnaires from these companies were not used in the statistical
analyses for reasons explained in the text.

87

constituted a third group (INEXPCLIs in "experienced" companies),
but the numbers were not large enough to conduct separate analyses
that would be statistically meaningful. Overall, then, a final
sample of 56 was used for the EXPCLIs group.

The group of INEXPCLIs proved, as expected, to be much more
problematic. Because this group represented those who did not know the
consultant well (having had fewer than five sessions with him), there
was considerably more difficulty in completing the instrument, so much
so that the original closing date for completion of this study had to
be extended in order to obtain an adequately large sample. Even by
distributing questionnaires to several additional companies, and modify-
ing the cover letter to attempt to deal with some of the resistance
issues (Appendix B), the return rates remained lower than desirable.
Although the final sample size for this group was at an acceptable level
for statistical comparison with the EXPCLIs group, the small percentage
completed at each individual company (Table 3) raises questions about
the representativeness Of the sample, and the issue of possible
systematic bias. The final follow-up letter attempted to assess the
degree of such bias by soliciting reasons for non-response by the
participants, but few in this group responded. Most of those who did
provide a reason for non-completion (Table 4) listed limited contact
with the consultant as the reason why they did not complete it. In
a few cases, "clients" indicated that they had never met the consultant.

A second, perhaps equally plausible, possible explanation that
could account for the low return rate of this group is the fact that
the early experience of an intensive consultative relationship is

itself intrusive and unsettling, and provides, psychologically, limited

88

Table 4

Reasons for "Returned, Unuseable" Questionnaires

 

Haaapp EXPCLIs INEXPCLIs
Extensive missing data 2 4
Limited experience with the consultant 11 12

H9 experience with the consultant 3
Questionnaire too long 2

Leaving/left company 2

17 19

 

89

Table 5

Selected Characteristics of the Samples

 

Years Employed with the Company

 

l - 5
6 - 10
11 - 15
16 - 20
Over 20

Organizational Level
Supervisory
Middle Management
Top Management

Organizational Position

 

Foreman
Professional/Technical
Department Level Management

Corporate Level Management

Total Number of Sessions with the Consultant

 

1 - 2
3 - 4
5 - 6
7 - 8
9 - 10
Over 10

EXPCLIs

E

11

7
9
10

19

56

27
20

56

19
22

56

t'kn so C) c>

4o
56

Z,

19.6
12.5
16.1
17.9
33-9

100.0

16.1
48.2

35-7

100.0

14.3
12.5
33-9
39.3

100.0

0.0
0.0
12.5
8.9

7.1
71.4

100.0

INEXPCLIS

E.

%.

27.8
38.9
14.8

5.6
13.0

v 100.0

24.1
46.3
29.6

100.0

18.5
13.0
38-9
29.6

100.0

44.4
65.6
0.0
0.0
0.0
0.0

100 I0

Table 5 (cont'd)

Total Number of Sessions with the Consultant
in the last year

Type

O

1 - 2

3 — 4
5 - 6
7 - 8
9 - 10
Over 10

of Consultation

Individual

Small Groups

Large Groups

Evenly Distributed Among These

2 3.6
12 21.4
19 33-9

6 10.7

6 10.7

3 5.4
8 14.3

—*

56 100.0

21 37-5
4 7.1
9 16.1

22 39-3

56 100.0

3 5-6
31 57-4
20 37.0

0 0.0

0 0.0

0 0.0

0 0.0
54 100.0

1 1.9
14 25.9
35 64.8

4 7.4

54 100.0

 

91

possibility for introspection about the process itself. Limited exper-
iential data from new clients provide some support for this hypothesis,
but the question was not assessed in any systematic manner. Whatever
the case, there are grounds for questioning the representativeness of
the INEXPCLI sample data, and the results from this group must therefore
be interpreted cautiously, and be regarded at best as being suggestive,
not conclusive.

Finally, the characteristics of the samples on several dimensions
are assessed. Table 5 presents data describing the samples on six
variables: the respondent's number of years employed by the company,
organizational level and position, the total number of sessions the
client has had with the consultant, the number of sessions experienced
in the last year, and the type of consultation (whether conducted
primarily on an individual basis, in small groups (five or fewer peOple),
or in large groups. Review of this data suggests essential sample
comparability regarding organizational level and position of the
respondents. The EXPCLI group, however, has had somewhat more exper-
ience with their organizations, and is more likely to have had
individual or small group consultation, or some combination, than the
INEXPCLI group, which is most likely to have worked with the consultant
in larger groups. This information may be useful in helping to
understand intergroup differences. This finding must be interpreted in
the light of another statistic, however: the number of sessions
experienced in the last year. Over half of the EXPCLI sample had had
fewer than five sessions with Dr. Frost in the year preceding completion
of the questionnaire, suggesting that their perceptions are frequently

retrospective.

92

Hypothesis 1

Hypothesis 1 is an assessment of the construct validity of the
proposed COP consultative mode. Two sources of evidence are avail—
able for its test: the critical incidents (CIs) supplied by the
clients, and the questionnaire data from items written for the DOC
questionnaire.

If the model that has been developed in this research is
verified by client experience, the CIs generated by §s should be under—
standable and categorizable according to the model's dimensions,
though it is recognized (Hypothesis 1A) that other dimensions will
also be of importance to clients.

The CIs presented qualitative data that was often very rich in
clinical detail and which, in many instances, reflected finely tuned
perceptual skills on the part Of the clients. An example both of
this point and of the scoring process outlined in Chapter III is
provided by the following CI, generated by an EXPCLI:

This is the first interview I ever had with Dr. Frost...It
was shortly after [a colleague] and I had heard his speech to
a manufacturers' group. On the way home from that meeting we
decided together that Dr. Frost had been talking about something
we were in need of. It was the first meeting that led to other
meetings. They resulted in this long connection [with the
consultant].

At this meeting the experience I remember was /his arm
length handling of the interview. He was interested, he was
courteous, but didn't appear to be at all anxious to commit
himself./ A I think he was trying to find out how genuine
we were about better relations with our employees./B

At a later meeting he was very adamant in his prOpOsal that
if he began work with us I would have to be part of the
Scanlon Plan./C
The divisions in this CI represent the units that were coded

by the raters. Each unit was coded using the numbering system

93

from Table 1 to represent the consultative dimensions of the model;
in cases where the model's dimensions did not fit the CI, a new

category was created. For the above C1, the raters agreed on the

coding:
Rater 1 Rater 2 Rater 3
A 1 .1A 1 .1A 1 .1A (PSYSAFTY)
B 2.2 2.2 2.2 (WHYDIFF)
c 4.1 4.1 4.1 (SLTNRESRCS)

Thus, in coding the units supplied by the respondent, the first CI
was viewed as being an instance of the PSYSAFTY variable (1.1A),
the second as WHYDIFF (2.2), and the third as SLTNRESRCS (4.1).
A second example, generated by an INEXPCLI, is also illustrative:
Dr. Frost persisted in the question, why change?. The
concept of the need for a compellipg reason for change was
given new dimension and importance. I believe this experience

has caused some critical reevaluations and a climate of
greater honesty with our group and stated goals.

 

This CI was judged to have one major dimension, and was coded
as 2.2, WHYDIFF. In cases where at least two of the raters could not
agree on a coding, the CI was not included in subsequent analyses.
Each unit of a specific example of helpful or non-helpful behavior
was considered to be a separate CI for the purpose Of these analyses.
Tables 6 and 7 present the summarizations of the coded CI data
using the full model presented in Table 1 (including the abbreviations
identified there). In Table 6, the coded dimensions are summarized
by client grouping, i.e., EXPCLI and INEXPCLI, using the detailed
specification Of the COP Model. In a few cases, the CIs could not

be categorized into a specific consultation task operationalized in

Sit

o.OOH mom o.OOH rNH o.OOH mmN

o.HH or N.oH mH n.HH AN
m.HN re o.s mu s.sN mm
h.m mm o.H N N.s HH
m. H 0.0 o a. H
r. m m. H m. N
N.N rN m.N m o.oH nN
H.m mm m.N m r.NH on
0.0 o 0.0 o 6.0 o
o.m HH .m.N m s.m r
ommm mm N.NH NN m.r HN
N.N r mum: m: H.N w:
r.m HN m.s 6H s.: HH
o.m sH o.s a H.N m
0.0 o 0.6 o 0.0 o
m.Hm me r.os Na N.NN so
w.rH or m.mN mm W.HH NM
s.H o H.m s a. N
N.N rN m.m s m.r HN
r. m o.H N s. H
N.m 0H s.: c m.n mH
o.ON ms Nwmm mm c.mH mm
mum mm c.H N mnm HN
m.H s m.N m N.H a
s.s AN m.oH sH c.n mH
H.s mH m.c r o.m s
c.o N r. H s.o H
m.o H r. H 6.0 o
w m w .m w m
Hnsos nHHoaxMzH nHHoasm

A<BOB Qz<mu
monmzmzHQ mmmeo

0.: asses
smszSm m .s

ngHzeHm mN.s
5%sz «N5
msoHHsHom Ho mcHomoe can coHHmoHaHpscoH N.s

mommmmzeqm H.:
soapsaom one mo magmuonus< pcwow 0.:
szomorav chnwosso aoHcoha o.m

o.N Haeoe
aaHsesr: m.N

thmea N.N

mooaq<oo «.N
omcsco Hos mpaommso a cooz one mo pcosmmomm< o.N

N.“ AdBOB

Bmszsz QN
Nequm UN
amm20mo mm
<N

.H
.H
.H
ammomHa .H

soHsHsHHmoor can soHHssHHcssaoocoo soHpowa N.H

H.H A<BOE
zmo<szA QH.H
o<hmoomm UH.H
weqm>omm mH.H
>9h<mwmm <«.fi

roam coapoenomcH mo coaampaaaoma H.a
:owpdoamapcooH a mmocoudz< 50H90hm o.H

 

soanoaHm moo

 

anoeHosH HooHuso space so swagger

0 manna

95

 

 

A<BOB Qz<mu
mZOHmZHZHD mambo
0.: .H<eoe
>mszeHm m.s
szAZBAm mN.:
>AmmZBAm <N.:
mcoavsflom mo mcfi
upmoa a soapNOHMHvsovH N.:
mommmmZBAm H.d

o.chm one mo mwcwhosv5< acaow o.d
¢.m m Azzomommv macmwoczo EOHQOAm o.m

 

N.oN mm r.ms NcN s.rH 0N m.HN s0H N.0m HN r.mo scH
m.0 mN N.s NH N.: 0 .nw.n N N.N NH .m.s 0H
m.r mm 0.NH as 0.H m: n.n mu m.NH mm 0.sH or
a mu 0 N s N:
0 H 0 0 0 H
m 0 H 0 N 0
r 0N H N s NH
0H NH 0 m 0H 0H
0 0 0 0
m. H r.N 0H N. H 0.H N 0.0 0
s.H m: n.0H mm r. .wu s.rH HN s.H .Lmu N.s NH
0 r 0 m: 0 w:
0 NH H 0 m m
H 0H 0 a H a
0 0 0 0 0 0
e.m mH s.0N 00H m.m s N.mm mm o.N m1 N.sN rm
ms Hm w: NN s mm
N s H m H H
N 0N H o H 0N
0 m 0 N 0 H
0 0H 0 0 0 mH
0.0 NN s.sH mm n.m .ml N.NH mm 0.0 mH N.mH Hm
mu sH 0 N mu mm
m s N H H m
a HN N NH s 0
0 NH N c H r
0 N 0 H 0 H
H 0 H 0 0 0
w .m w. _m m. m .w .m w m. .w .m
HsNaHossoz HsNaHor HsHmHossoz mmwmwmm HsaaHoesoz Hssawwm
asses mHHcaxmzH mHHoaxs

o.~ .A<Hoe
aaH0earr m.N
aaHmwm: N.N
mooaq<oo H.N
omcsno mom hvaomnmo
a 0002 one mo pcosmmowm< o.N
N.H .Hseoe
emszmma 9N.H
xequm 0N.H
ammzumo mN.H
mmmomHm <N.H
coapacwmovom a compass
andwmoocoo soaponm N.H
H.H .Hssoe
zmu<szA QH.H
u<amoomm UH.H
weqm>umm mH.H
~9a<mwmm <N.H
roam soapsshomsH
mo coapmvfiaaoch «.«

sonnoHNchosH
a mmocoumz< soapoym o.H

 

:OHmeEHQ moo

 

nsoanoaH0 HsNaHossoz 0cm HsaaHor an eonHmano .wssoeHosH HnoHpro 00000 No sumassm

m OHQOB

96

the model, but could be coded as a more general dimension, e.g., 1.0
was used as a coding rather than the more specific, 1.1A. Summariza-
tions are provided for each major grouping of variables, as well as
overall. Table 7 is simply a more detailed elaboration of Table 6,
breaking down the frequency analyses using the Helpful/Nonhelpful
categorizations.

Before more specific analysis of each of the major model
dimensions, it is useful to discuss the general findings. First, the
fit of the data to the a priori defined dimensions is generally quite
good. The CIs are spread relatively evenly among the model's dimensions,
particularly for the EXPCLIs group. There were close to twice as
many CIs generated by this group than by the INEXPCLIs, however, and
this may account for the larger representation of many dimensions
in the EXPCLI sample. Overall, a rather large number of 01s were
generated (N = 363), which provides reasonable assurance that a wide
sample of client-perceived consultant behavior had been generated.

There is relatively less support from this data for the following
variables that were defined in this study: PROCSFAC, DISCREP, EXRLTY,
WHATDIFF, PROBOWN, SLTNRELV, and SLTNLINK. The number of cases that
had to be coded as a more general dimension was few, with the majority
in the Problem Conceptualization and Redefinition (1.2), and the
Identification and Testing of Solutions (4.2) categories. Clients
were generally less likely to give negative critical incidents,
suggesting a possible favorable response set in completing the instrument.
The dimensions for which negative CIs were given by the EXPCLI sample

were primarily PRCVRLTY, LINKAGEN, THRYINPT, SLTNRESRCS, Identification

and Testing of Solutions, and SLTNIMEV. There are twice as many nega-
tive CIs given for the Solution variables than for any other category.
For the INEXPCLI group, there are few negative CIs generated, but
those that were are primarily coded in the first two groupings of

the model (1.1 and 1.2). This is probably due to the fact that the
majority of this group's CIs fall into these codings, as the model
would predict. In general, then, the model receives support from this
data, but several of the variables are of apparently less importance
to clients. The data is limited by an apparently positive response
set and reluctance to identify negative instances of consultant I
behavior. Finally, there are definite differences between the EXPCLI
and the INEXPCLI groups' data.

Earlier it was suggested that much of the clinical richness of
the CI data was lost in the process of translation into summary data.
At this point it is useful to study this data in more detail. This
will provide both a greater qualitative understanding of the variables
that receive support from the study, and a more thorough understanding
of the summary data already presented. The specific consultation
dimensions will be discussed in the order in which they appear in
the model.

PSYSAFTY

Creation of an Atmosphere of Psychological Safety, 1.1A, appears
to reflect a number of more specific consultation tasks. First,
clients find helpful that the consultant is generally able to relate
to them on levels apprOpriate to them. A critical component of this

communication skill appears to be the consultant's awareness of and

98

empathy with the client's world, generally through technical knowledge
of the individual's work requirements, an ability to communicate,

in effect, "I know something about your work world, and am therefore
competent to talk with you." In the following illustrative CI, this
is described as "building the ego" of the client, but it is perhaps
more accurately termed, establishing the competence of the consultant
to the client, as well as his interest in the client.

He came to my company in 1975 while I was...manager of
[a department]. He toured the plant. During this tour he
talked with some of the employees of the department. His
ability to communicate with the workers at their level and
with some knowledge of the complexit of the equipment
helped build their ego. [Employee X will never forget his
first exposure to Dr. Frost. This built the credibility
necessary to have a good response to the[ Employee Partici-
pation Plan].

There appear to be additional elements to PSYSAFTY as well. It
seems not just to be the communication skill already stated, but
the ability to develop a sense of trust, i.e., that the consultant
is genuinely interested in the client's well being. This idea is
clearly implicit in many of the CIs generated, and is analagous
to Strupp's (1976) research on the necessary but not sufficient
component of "human" qualities needed for effectiveness as a thera-
pist. Again,this is not always Obvious or directly coded in the
CIs (only 4.1% of all of them were coded in this category). For
example:

...our facility recently installed a version Of the

Scanlon Plan. In...l977...just prior to all employees

voting on acceptance of the plan, Dr. Frost assisted in

meetings (36 hours of meetings in two days) to answer

any questions and review the final proposal. This type

of assistance and participation was typical of Dr. Frost's
interest in our company.

99

Thus, for this respondent, the "trust" dimension is implicit in
the amount of time the consultant was willing to devote to his organ-
ization. Others are much more explicit on this dimension:

He has become a part of|:our company's] family. He is

loved by all as a co-worker and he is more a company employee

than a consultant in the minds of many peOple.

I do have the strong feeling he's a man I can trust and

that he can be helpful to me and our company whether we

go scanlon or not.

It is also apparent that another aspect of the consultant's
ability to establish an atmosphere of psychological safety is the
perception that he has something to offer to the client. This is
illustrated by the following CI, which includes several behavioral
units coded in different parts of the model, but is here cited in
its entirety to understand its overall relationship to this dimension:

I was beginning my role as president of the Scanlon [special
interest] group. Not having much experience or prior direction
as to the role of president and goals of the group, I decided
to ask Dr. Frost for some assistance. His patient listening
and ideas for programs, plus possible resource peOple helped

me to contribute effectively to the group and its goals

during the year.

Like one of the Previously cited examples, the PSYSAFTY aspects of
this one are implicit, but they relate to the client's perceptions
that the consultant has something specifically to offer him.

In short, these CIs point to three specific aspects of the task
of establishing psychOIOgical safety: ability to communicate
on the client's level, benevolent interest in the client, and having
something to offer that is valued.

There were relatively few examples of negative C1s for this

variable, and those that were generated were often implicit

100

in 01s more appropriately coded in other dimensions. However, PSYSAFTY
appears to be jeopardized when the consultant's role is not clear,
when he is not perceived as benevolent in intentions, and when he is
not viewed as having anything of value to the client. For example:
I feel that his relationship to the general manager of

our facility needs "up front" definition for one is never

sure whether he is assessing the personal competency and

readiness fcr Scanlon on behalf of the general manager or

if he is there as a truely neutral but interested party

trying to help.
This suggests that the "client" here is not the respondent, but one
of his superiors, and that the respondent does not feel that he can

make full use of the consultant. This problem will be discussed

in more detail below.

1.1B PRCVRLTY

 

The idea behind this dimension is that Of helping the client
to discuss his views as they now exist, even if they are not rational
or fully defensible. Two CIs illustrate this concept well:

...Dr. Frost did a very fine job in leading the discussion

of what really makes the Participative Management work. He
cleverly drew out personal Opinions, so that they could
evaluate as to their effect on the Participative Management
Plan. The discussion was lively. Dr. Frost showed a fine
ability to have people express what was really on their minds.

In my first meeting with Dr. Frost...he asked why I felt
that I was promoted to my present position. I found Dr.
Frost very easy to talk to on explaining to him my background
with the company and my reasons why I felt that I was
selected for this position and at no time did I question what
Dr. Frost wanted to hear. I was sure that what he wanted from
me was only my true feelings about the situation.

The exact manner in which Frost helps draw out others' Opinions

and get them to share their perceptions is not well detailed by these

101

015. It appears to be both the ability to listen, and the ability to
elicit Opinions that are Of most relevance here. It is hypothesized
that this dimension is strongly related to PSYSAFTY, since it is

- unlikely that people will share their perceptions if they perceive
negative consequences to follow. Support for this hypothesis is
provided by the statistical analyses of the DOC questionnaire items,
presented later in this chapter.

As for the negative aspects of this dimension, they are again
related to the concept Of misuse of elicited information. Clients
reporting negative examples typically mentioned things that they had
told the consultant that were later misused, or a perception that the
elicitation Of information is not for benevolent purposes, e.g.,

Two months ago Dr. Frost visited my office. The conver—

sation seemed to be centered around trivia and I felt, as I

Often do, that he was just "pumping" me for information that

he could pass on to higher management.

It is important to recall that this study deals with perceived
data, and so we do not have an independent source Of understanding
whether the perception reported here was valid or not. What is clear
is that for this client the consultant was unlikely to be able to
Obtain information from him. Nor is the present study designed to
assess interactions between consultation style and client psychological
variables. More sophisticated psychological research would be
appropriate to understand the differences in individual perceptions

between those who perceive the same psychological stimulus (the

consultant) as being supportive and those perceiving him as threatening.

102

1.1C PROCSFAC

 

Very few (1.9% Of all CIs) examples were coded into this category,
Process Facilitation. Essentially, this consultant is not Often
perceived by these samples as utilizing process facilitation, at
least as the term is conventionally used to refer to having people
explicitly examine the manner or process of their communication or
managerial style (e.g., Argyris, 1970; Schein, 1969). A few clients
did perceive helpful incidents of this sort, e.g.,

The President of the company was dominating meetings with
his staff and Dr. Frost. This was dampening the creative
participation of the staff. Dr. Frost observed this and
commented to the President about his dominance and even
verbal dueling with one key staff member.

Few clients reported incidents (either helpful or nonhelpful) of
direct feedback regarding their own behavior; usually clients reported
feedback to someone else as being facilitative Of group process.

There is a second dimension also reported, which involves
processing in the sense of facilitation of the conduct of meetings,
Often by posing stimulating questions.

...Dr. Frost supplied baiting questions to sustain the

discussion among the group, resulting in beneficial replies

and results.

Whether this type of consultant activity is best conceptualized as

being process facilitation or another dimension of consulting is

somewhat moot, given the small number of CIs coded here.

1.1D LINKAGEN

 

In developing the COP Model, LINKAGEN was conceptualized as

the consultant's serving as a bridge, or communication link, among

103

various parts of the organization. The consultant does this either
by eliciting and passing on information from one part of the organi-
zation to another, or by bringing together various sections of the
organization to address some problem or need. In the present samples,
this dimension was perceived more frequently by EXPCLIs than by
INEXPCLIs (8.9% vs. 1.6%), but, particularly for the former group, was
the source of almost as many nonhelpful as helpful examples. Two
CIs illustrate the positive aspects of the activity:
I believe that when Dr. Frost meets or talks with someone
in our Company, peOple look more at him as the one who holds
the top and the bottom together. I believe this has been
very helpful to our company.
I have a problem with the C.E.O. ['s] direct interference

in my function as a manager. Dr. Frost called a meeting of
all managers and the C.E.O., and voiced the problem to the

group...

The implicit message of several Of these Cls, however, is a
tendency, especially at lower organizational levels, to look upon
the consultant as a "champion," perhaps as a substitute for taking
action on their own. Both of the following, cited by clients as
examples of nonhelpful behavior, are illustrative.

Not going to top management sooner on the problem of

them not being open candid and using participative manage-

ment. They talked about it, but middle management had to

use it.

In not being forceful enough with [the] Office of the
President on certain organizational rationale and changes.

A second difficulty with this activity is that the sources of
the information passed along may not be reliable or valid, in which
case, not contributing to effective problem awareness or identification.

Or, the consultant may rely on the same data sources too persistently,

104

thus not getting a clear picture. Finally, the consultant may
jeopardize his trust by inapproPriately communicating information,

and may also lose trust with those to whom he communicates the
information. These difficulties are well illustrated by the following
negative CIs.

He talks to the same more senior employees and usually
takes their input as gospel. He should talk with some of the
new employees so to have a better sample of employee feeling.

A meeting was held between Dr. Frost and myself, he began
talking about problems we as an organization had and that we
were not facing or taking problem ownership. It was very
general and appeared his information was coming from one or
two sources that were part of the problem. It was not specific,
not really identifiable, and very, very frustrating. Especially
when it appeared he was only interested in his information and
not what else was available. Generalities, coming from [the
consultant] are vague and at times half-truths. I feel at times
he gets "suckered into" situations based on contacts established
over the years. At times in this situation I feel he should
do more research and identification.

He would criticize performance but not be specific as
to where it came from or a proper perspective with other
important operational matters. His intent was to help,

I'm sure, but it left me with a feeling of frustration.

Thus, when acting as a "linkage agent" the consultant obtains
information on organizational issues, serves at times as an advocate,
and "creates settings" (Sarason, 1972) in which employees can
interface with one another. In accomplishing these tasks, however,
he is viewed by some as a threat, whose role in gathering this data
and transmitting it is unclear or threatening, or as an information

source whose data is biased or outdated.

1.2 Problem Conceptualization and Re-Definition
The second set of dimensions are different in kind from the

ones just discussed. They have to do with the understanding and

105

analysis of system problems in such a manner that the "core" or
fundamental issues are uncovered. Collectively, these dimensions
account for 31.9% of all CIs generated, and thus represent the
largest single group. There is a somewhat higher percentage Of
these examples among the INEXPCLI group than among the EXPCLI
group (40.6% v. 27.2%).

Two of the four dimensions, ORGNEED (1.2B) and THRYINPT (1.2D)
account for the vast majority of the CIs generated in this grouping
of variables. The remaining two (DISCREP, 1.2A; and EXRLTY, 1.20)
did not receive much support from the data (.8% and 1.7% respectively
of all of the 01s submitted). A relatively large number of C1s for
both samples were coded in the more general Problem Conceptualization
and Redefinition category (1.2), and it may therefore be the case
that these represent more specific variables that were not identified
by the COP Model, and that should be substituted for DISCREP and
EXRLTY. Those variables that were supported will be discussed in

more detail.

1.28 ORGNEED-

 

A division emerged in this grouping among those 013 that
focused on the organization's needs and those that focused on the
needs of specific personnel (Often the respondent) within the
organization. The consultant appears to provide both, focusing both
on the organization's needs and the fit between these and the needs
of personnel within the company. The following two CIs are

illustrative.

106

In highlighting the need and helping us to process
changes in our top management team. By working with us as
individuals and collectively, in making us look at our
strength and weakness and what was needed for the future...

I was struggling with a competence issue with a couple
of my subordinates. By having a discussion Of looking at
what
1; the company's requirements of me
2 what I require the subordinate to do to allow me to

complete tasks and carry out responsibilities
3) a stack up of skills and abilities required vs. those
the subordinate hada...

These CIs (which represent additional dimensions than just the
one here under consideration) focus on the needs both of the organiza-
tion, but also Of the individual employees, suggesting that the
client receptivity to the consultant may be enhanced by a simultaneous
concern for both. It is also clear that when the consultant discusses
each of these issues, it is not from the perspective of general
theory, but in the concrete terms of the client and client organization.
For example:

I had consulted with [another consultant] on our directions—
goals-needs and organization planning.
Dr. Frost helped me evaluate [the other consultant's] advice.

We then reviewed the peOple in our organization--their talents,

interests, and potential.

Dr. Frost helped all our tOp management people to see

how they fit into the needs of the company....

This approach is consistent with the earlier-discussed communi—
cation skill (1.1A) of being able to communicate with all levels of
employees in language appropriate to each; it presumably lends
credibility and authenticity that the consultant is aware of and able

to contribute to their specific concrete problems, not just to provide

general explanations.

107

1.2B THRYINPT

A fairly large percentage of the CIs gave examples of specific
theoretical input that they regarded as helpful (16.5% of all CIs;
11. 5% and 25.8% Of the EXPCLIs' and INEXPCLIs' CIs respectively).
The fact that this proportion was over twice as high in the INEXPCLI
group might be expected since in the longer term consultative
relationships, the consultant's theoretical perspective and values
could be predicted to become well known over time.

In general, the examples of effective consultant behavior in
this area are specific about various aspects of Frost's views regarding
organizational effectiveness. Since this has been well detailed
elsewhere (e.g., Frost, Wakeley, and Ruh, 1974) it will not be
reviewed here. One example of such a CI is the following:

This happened about two years ago. While it is a unique
occurrence slightly outside the normal consulting function
it is an example of how he has been helpful over a long time.
The specifics-—he wrote a book which clearly spells out:
identity, participation, equity. This sharpens the focus of
one using the Scanlon Plan and establishes priorities....it
illustrates how Dr. Frost helps one understand the fundamental
revolution that occurs when one embarks on the Scanlon
experience.

The negative examples of THRYINPT have less to do with the
content of the theory and more to do with the manner in which the
theory is presented, with the general complaint of being overly
theoretical, too abstract, or vague, e.g.,

Dr. Frost seems to want to dwell mostly on theory rather
than practical application and implementation. I am not saying
that he does not feel that the latter is important; but he seems
much more "at home" on an abstract or theoretical level than
he does on the level where ideas must be put into use.

His concepts of what should be considered and accomplished
are thorough and logical; and he seems to spend most of his
time in this area. As to pr_one makes it happen, he seems
to fall a little short.

108

A private meeting held in my office about 2% years ago

was conducted to evaluate the possibility of [a certain program].

Frost was so hypothetical, evasive, academic, and theoretical

that I was turned off.

When discussing application of Scanlon Plan to core group

of management, all of whom had read his book, one member of

the group asked a question and we received a 1% hour lecture

which most, including myself, felt a waste of time....

These CIs suggest a real sensitivity on the part of these
clients, who are all managerial personnel in private sectOr manu-
facturing organizations, to the pragmatic versus the theoretical.

This may well reflect their sense either of being intellectually
inferior, or their belief that the pragmatic (the essence of their
daily activities) is of most importance. This data suggest that
psychological consultants who work with industrial clients might be
best Off minimizing their direct theory inputs, or at least maximizing
the integration of theory with pragmatic data.

It was already mentioned that several of the C15 in this grouping
had to be classified as a general example of problem conceptualization
_Or redefinition, rather than as an instance of the more specific
operationalized variables. These CIs provide a useful perspective
on Frost's approach to this area. In contrast to a more programmed
style of consultation, which presents as a fait accompli a particular
program or package, Frost's style (and what is seen as being helpful
by clients) is Socratic. For example:

I think his ability is helpful to talk to you about a problem
you bring to him and after you discuss it you find that you

are coming up with your own answer, which in the discussing

he has helped you see...

The situation I am thinking about occurred almost 10 years
ago. The timing was only a year or two into my full time

109

professional career. I met with the consultant...for almost
half a day. It was a consulting session which I had requested
and had been advocated to have by my boss. As I recall, I was
particularly frustrated with some of the shortcomings of my
boss and other key People around me. The consultant was very
helpful because in his version of consultation he did not tell
me anything. Instead, he effectively helped me reflect on

the situation and get it back into perspective. I remember
at the time that I was frustrated with the fact that he
wouldn't draw his own conclusions. Later I was able to see
the wisdom in his approach.

He has always offerred questions without taking a position
which requires me to think and plan more thoroughly;

What this suggests is an indirect approach to problem identification
and reconceptualization, in which the consultant does not give advice
or direct answers, but rather helps the client think through the
problem himself. Such an approach is seen both as being of value
(note the impact of the example from ten years previously), but also
as being frustrating. Not all clients are able to benefit from this
approach, or find it helpful.

Dr. Frost was not particularly helpful in an example where
we asked his opinion on what he thought would be right or wrong
for us to do in proceeding in a specific instance.

...He refused to give any kind of an indication as to what
he thought was right or wrong, and indicated that it was the
company's problem to solve, which it was indeed, but what he
was asked for was his opinion or guidance...

[The consultant] is subtle to the point of exhaustion...we...
can't always afford to be philosophical and dikadantish about our
questing for truth, knowledge, and meaning...A strong and direct
Opinion from a respected authority would be a breath of fresh
air. One might disagree but at least the lines for a dialogue
could be more exactly drawn....There are times and circumstances
in which I believe a consultant with true concern must hit
forcefully with the courage of his convictions....the polite-
ness of academia's rhetoric is not appropriate to certain games
and fields of battle. The need for change must be forcefully
sold as the pursuit of profit.

Thus, for some clients, and perhaps for some organizations, the

question-asking approach, which does not take a stand or provide

110

"answers,' is perceived as more frustrating than helpful. This may

suggest that an individual differences approach to organizational
consultation is appropriate, and that modification of the basic
model is necessary if one is to have any impact. It may be speculated
that the psychological variable of tolerance for ambiguity is a
moderator of the receptivity to this approach, but a systematic test
awaits future research.

In any case, in the area of problem conceptualization and
redefinition, the consultant's behavior appears to be an attempt
to get the client to generate his own answers to problems through
the vehicle of raising critical questions which seem to address core
issues and which presumably serve a purpose analagous to interpreta—
tion in psychotherapy: bringing into conscious awareness aspects
of problems of which one was previously unaware. The essence——and
the limitations--Of this approach are captured by this detailed,
but significant, CI:

Dr. Frost was visiting our company one day...At the time, we
were having a severe conflict in a certain area and wanted his
assistance....True to form (yet contrary to my previous example)
Dr. Frost was very broad, and seemed unwilling to commit himself
to any specific advice. The conversation centered around
comments such as "you should seek some help in this" and "what
do yap think should be done"--very Frostian. The problem was,
however, specific...rather than a conceptual or large scale
exercise, and I guess I was looking for specific assistance.

(There is somewhat of a conclusion here. While Frost is a
genius at refusing to take over other's problems--he is not a
hand-holder or a "you do this and everything will be fine" type
of man--and this talent is essential in being effective in
getting people...in the Scanlon Plan, or any other endeavor—-
to stand on their own-~that same genius limits his expressions
of specific answers or advice...to specific questions...It's
not like "Dr. Frost, what time is it?"--"Well, what time would
yop like it to be?", but that's the idea.)

111

2.1 GOALFOCS; 2.2 WHYDIFE152.3H WHATDIFF

 

Frost (1978) contends that effective change must be preceded
by "compelling reasons" directing the change. He views part of
his consultation task as the assessment and perhaps stimulation of
such reasons. Overall 11.9% of the CIs were coded as representing
the three specific variables operationalized here, prOportionately
more in the INEXPCLI than the EXPCLI groups. Examples from GOALFOCS
and WHATDIFF were given relatively less frequently than the WHYDIFF
categories, though both GOALFOCS and WHYDIFF were well represented
among the INEXPCLI sample. Because of the limited number of WHATDIFF
examples, and because most CIs tended to include mixtures of more
than one of these dimensions, they are discussed collectively in
this section.

A chief component of the consultant's interventions in these
areas is the concept of organizational and individual identity, a
review of the historical origins of the present staff and organizational
functioning. Such assessment apparently helps clients focus on
their present goals and the (often implicit) directions in which they
are going. For example:

First consulting visit of about 3 hours-—when we assessed
where we were--our backgrounds-—and started us thinking on
where and what we want to be.

Our first personal meeting resulted in a challenge by Dr.
Frost to search and define the reason for our existence and the

reason we should continue. Due to his prodding, we have openly
defined and accepted common objectives throughout the executive

group.
The WHYDIFF variable, i.e., the need for the client system to

develop convincing reasons to change, is explicitly mentioned as

112

helpful by several examples, particularly by new clients.

Dr. Frost persisted in the question, why change? The
concept of a need for a compellipg_reason for change was
given new dimension and importance. I believe this experi-
ence has caused some critical reevaluations and a climate
of greater honesty with our group and stated goals.

 

Dr. Frost questioned each individual as to why did we

want to change existing circumstances. To each answer that

was given, another "why" was asked. This particular approach

caused you to think of every aspect possible of why you wanted

to change, why you needed to change and helped to emphasize

the critical needs and eliminate general concepts.

This approach, particularly with business clients, is a para-
doxical one (Watzlawick, Weakland, and Fisch, 1974). The client,
especially an industrial client, generally expects the consultant
to approach him with a program or a "package" and to be given a
"hard sell," thus approaching the client in a manner with which he
is familiar (e.g., Strauss, 1976). Instead, Frost insists that the
client's reasons to change are not convincing, and that they are
therefore not ready to do so. Often this creates the paradox that
the clients are more desirous of the change because they have been
told that they are not yet ready for it, as might also be predicted
by reactance theory.

For the purposes of this research, it is also of note that this
process is more Often reported by INEXPCLIs than EXPCLIs as exemplary
of helpful consultant behavior. Thus, contrary to the prediction
of a sequential process in which the Operationalized "confrontation"
variables follow earlier model stages, it appears that the consultant

is more likely to use confrontation, at least as Operationalized

here, in earlier consultation stages than had been expected.

113

Another dynamic that is noteworthy is the fact that clients
perceive the confrontation variables as being directed to them as
individuals as well as to them as organizational members.' This is
illustrated by the following CIs:

During his initial visit at our company...Dr. Frost queried
each person (six in all) to the effect, 'What are you doing
here? Is this the best job you can get?‘

The effect and ongoing effects of challenging each person
to have a reason to be here have been dramatic and helpful...
...By working with us as individuals and collectively, in
making us look at our strength and weakness and what was
needed for the future. Also, to help those individuals
affected...lay the ground work for the transition....

The effect of this dual focus, a persistent theme in this data, is
presumably to cause the client to feel that the consultant is concerned
about and wanting to help both the individual, the organization, and

to facilitate the fit between the two.

It must finally be noted that the dimensions represented here
are Often complex and overlapping with other dimensions. In particular,
several of these examples have implicit elements of ORGNEED (1.2B),
in that the clients are asked to focus on both the needs of the
organization and of themselves in determining whether there is a
convincing reason to change. Thus, it may be expected that these
variables will be correlated with earlier model dimensions when

measured by "objective" means.

3,0 PROBOWN

 

In this study, the basic model's "consolidation" phase was
operationalized as "problem ownership," the consultant's working

with the client system to accept responsibility for the problem(s)

114

presumably identified and understood by the earlier—discussed
consultation variables.

CIs coded as representing this category were a small percentage
of the totals (3.0% of the total sample; 3.4% of EXPCLIs' examples
and 2.3% of the INEXPCLIs'). Almost all were examples of helpful
rather than nonhelpful consultant behavior. There were some examples
of PROBOWN that were implicit in behavioral incidents that were
coded in other dimensions, especially THRYINPT, e.g.,

...Dr. Frost was very enthusiastic about the Scanlon Plan but

made it perfectly clear it was up to each of us to make ppr_pyp

decision concerning the Scanlon Plan.
In contrast, the incidents representing this category dealt with
specific behavior on the part of the consultant helping the client

to accept problem ownership of organizational issues, e.g.,

He made me pinpoint my responsibilities to myself, to
my peOple, and to the company.

...After we left our meeting we were committed to do all we
could to get things straight.

The vehicle the consultant uses to establish such ownership
is not always clear. It appears to be direct suggestion in many
cases, but there is another means that is only hinted at by the more
perceptive Observers. Perhaps the best example of this is the CI
discussed earlier in this chapter to illustrate the coding procedure:
...the experience I remember was his arm length's handling
of the interview. He was interested, he was courteous, but
didn't appear to be at all anxious to commit himself. I

think he was trying to find out how genuine we were about
better relations with our employees.

This approach to PROBOWN is perhaps best conceptualized as a clinical

skill, the ability to be concerned but not to accept the problem as

115

one's own, and implicitly communicates the message to the client that
the consultant will not accept ownership of the client's problems,

or responsibility for their solution.

4.0 Joint Authorship of the Solution

Overall, CIs associated with this cluster of variables accounted
for 21.5% of the total sample. They were much more dominant in the
EXPCLI sample than the INEXPCLI sample, as would be expected from the
hypothesizing about the sequential nature of the model. The CIs in
this area accounted for 29.4% of the EXPCLIs' examples, and only 7%
of the INEXPCLIs'. These CIs represent consultant behavior associated
with the evaluation or recommendation of organizational personnel to
be associated with problem solution (SLTNRESRCS, 4.1), the actual
generation of specific solutions themselves (4.2), including the
assurance that the solution is relevant to the underlying problem
(SLTNRELV, 4.2A), and consistent with external demands on the
organization (SLTNLINK, 4.2B). Lastly, the consultant is sometimes
involved in the actual implementation and evaluation of specific
solutions to problems (SLTNIMEV, 4.3).

Although the data for the INEXPCLIs is too small to allow
substantive analysis or conclusions, the data provided by EXPCLIs
shows a majority of the CIs in this grouping to be in the "helpful"
category, but a sizeable minority to be in the nonhelpful area
(Helpful: 17% of the group's CIs; Nonhelpful: 12.3%). This suggests
that consultant involvement in solution deve10pment is a potential

source of dissatisfaction with his activities, that perhaps his

116

more helpful contribution is in helping to define and understand

problems, rather than in solving them.

4 . 1 SLTNRESRCH

 

In examples coded in this category, the consultant was viewed
as suggesting resources for the solution of organizational problems
or as evaluating organizational personnel regarding their suitability
to the demands of the organization. At times, he is seen as helping
to integrate a staff member into the organization. Examples include:

The company's pricing policy and strategy and the decision-
making process for it was outmoded and inconsistent and inade-
quate for the new competitive markets the company had entered.
Dr. Frost developed an acceptance by the executives of a new
person (myself) into the process with a totally different
outlook and expertise. This was essential to enable me to make
a contribution through an Opportunity to display my solutions
to these problems.

I was about to make a new key appointment. I went through
my analysis of he need, the candidates, and my decision.
[The consultant asked me, 'Did you ever think of...?'

This caused a review--I selected the person he raised the
question about--that person is now one of our key executives...

But it is also true that the consultant is viewed as being
nonhelpful when he evaluates personnel or suggests peOple for certain
positions, e.g.,

[ The consultant] persuaded Management to promote me to a certain
position.
I accepted the job--got the De tment back in the black,
but resigned...because[ a managerfiaiefused to spend any money
for equipment, or process changes costing money.
I got to a point close to a basket case--
Later [the consultant] said it was an impossible situation
from the start...

A supervisor had all his data to fire a very poor performing
employee. Three supervisors before him had tried to do the
same thing and Management stopped it.

[ Dr. Frost] talked the supervisor out of it. The man
continued to be a major problem & three years after this

117

situation the supervisor was still upset with [the consultant].
Recently he made surface judgments concerning the technical
qualifications of one of our departments without knowledge
of the department's work or personnel. He spoke of the lack
of competence which was hearsay, not his own personal knowledge...
Thus, the consultant's involvement in decisions Of personnel
evaluation, and his specific suggestions of staff members for particular
positions, appears to be perceived as negative as well as positive.
It may also be that the consultant loses his leverage as a disinter-
ested, benevolent outside party, and comes to be feared as a threat
to clients' jobs. This is graphically illustrated by the following:
...There is a definite fear in some of the newer peOple that
he is a hatchet man for tOp management. Some of the 10 to 14
year peOple feel the same. This feeling gets in the way of
progress. In my 20 years at[ my company] I have never heard
anyone explain[ his] role.
It may also be that it is the departure from the COP Model of client-
generated solutions to problems which the consultant has helped define

that represents the threat to helpfulness.

4.2 Identification and Testing of Solutions

Most of the CIs coded in this grouping of variables fell in
the general category, 4.2, rather than the more specific SLTNRELV
(4.2A) and SLTNLINK (4.28). This may reflect an artifact that the
C13 in this group were Often very technical and specific, and did not
supply enough information to determine if they supported the model
presented in this research. For example:

I was in the midst of developing a service group within the
organization. We had defined manpower measurements, program
and procedures. Dr. Frost was helpful in determining how this

group would have to approach certain managers of key areas...
to establish a positive impact of the program and also to

118

point out areas where we had to improve to maintain a working
relationship with other parts of the organization over time.

It is clear from this example that the consultant was involved

in specific solution deve10pment. It is not clear whether these

"solutions" were tied into any underlying problem, or that they

assured the relevance to external reality issues.

In some of these 018, there is an advice giving, almost "preachy

quality to the involvement, e.g.,

...Dr. Frost listened very attentively, and then was quite

firm with me. He pointed out that while he reOOgnized what

I was trying to do, he also recognized that I was Obviously

very personally involved in this problem and furthermore was
trying to fulfill a task that essentially was out of my league...
Rather, I could best fulfill a necessary function by being the
ever present question, challenging top management, but not
trying to fill their role.

In other cases, the specificity of the advice compromises

the basic model, even if the advice may be sound, e.g.,

4-3

Dr. Frost suggested that I move from the office[ at one
location] to the [present] plant site...to deal more directly
with plant management problems and low productivity. It
resulted in my seeing the problems in a new perspective...

SLTNIMEV

There were relatively few examples from this category (3.6% of

all C15), and they do not differ in content or process in substantive

ways from the other variables related to the 4.0 category. The

major distinction in the CIs coded here was that the consultant's

role in the "solution development" involved the actually putting

specific solutions into place, e.g.,

...he helped me to organize the administration...so it could
function competently.

119

Also coded here were CIs having to do with the process of implemen-
tation of "solutions," which were Often examples of nonhelpful behavior,
e.g.,

Dr. Frost mentioned the possibility of myself spending

time with another company. This was in response to a request

from a person from this plant to give my views and experiences

in a Scanlon operation...

That was the end of communication on the subject as I heard

no more...

At one time the Co. decided that a Survey was necessary.

Without letting the peOple know what and why, they called

meetings the same day and loaded us all up because of the

way this was handled. I believe it was a waste. I also

believe that Dr. Frost was the head man on this.

It would appear then from an overall assessment of the consultant's
involvements in the "solution authorship" area, that these activities
are more likely to be performed with EXPCLIs than with INEXPCLIs, as
the model would predict. However, the consultant is almost as likely
to be perceived as being nonhelpful as being helpful when Operating
in this area. Probably it is the case that if his suggested resources
or solutions work out, he is favorably evaluated; if they do not, he
is unfavorably evaluated. The more important conclusion is that
these forays into solution authorship have more potential for
departing from the COP Model, and that they are associated (perhaps
because of this) with negative as well as positive consultant activity.
It is presumably the compromise of the basic tenets of the model
that account for this conclusion.

Summary of the Critical Incidents Data

Summarizing, the CI data suggest a consultant who:

1.) Can establish trust with personnel at different organizational

levels and with those with whom he has had limited contact;

120

2.) Who establishes such trust by being able to communicate on a
level appropriate to the client, by a benevolent interest in the
client, and by projecting an ability to be Of help;

3.) Who is skilled at facilitating clients' expression of their own
views;

4.) Who functions as a "linkage agent" among various parts of the
organization by interacting on a variety of levels, often by
communicating perceptions held by personnel in one part Of the organi—
zation to those in other parts, often at higher levels;

5.) Who, in the capacity of linkage agent, sometimes relies on the
same informational sources too heavily, or jeOpardizes trust establish-
ed with individual clients;

6.) Who helps identify and conceptualize problems by a non-directive
approach, often Socratic, in which he pushes peOple to do their own
thinking rather than to rely on answers provided by him;

7.) Who,in being indirect,frustrates certain clients, especially
those looking for "expert advice";

8.) Who confronts clients, especially new clients, with the need

to understand their history and present identity, and to elaborate
"convincing" reasons to change before embarking on change projects;
and in so doing, often functions paradoxically;

9.) Who generally insists that clients accept ownership of their
organizational and individual problems;

10,) Who at times functions more directively by suggesting resources,
evaluating organizational personnel, suggesting specific solutions

to problems, and helping to implement them;

121

11.) Who, in becoming involved in "solutiOns" is viewed frequently
as being nonhelpful as well as helpful;

12.) Who focuses both on individuals within organizations and upon
organizations themselves; who facilitates trust through the former

and gains leverage to intervene through the latter;

13.) Who is generally viewed as being more helpful than nonhelpful
by these samples of clients;

14.) Who is viewed differently by EXPCLIs and INEXPCLIs; and

15.) Whose actions, as reported by these samples, are generally

consistent with the COP consultative model elaborated in this research.

Analysis of the Questionnaire Data

A second source of evidence for Hypothesis 1 is provided by
the DOC questionnaire data that used an "Objective" format. The
same instrument was administered to both EXPCLIs and INEXPCLIs,
with the former group used as the basis for scale development (for
reasons explained in Chapter III).
Psychometric Analysis

A priori item assignment to defined scales, using the previously
discussed procedure, is presented in Appendix C. Items enclosed
in brackets indicate ones that were dropped from the initial
scale assignment to improve the scales' statistical characteristics.
The final scale items used for the present analyses are the non-
bracketed items in Appendix C.

Table 8 presents the measures of internal consistency for

the original and revised scales, and the number of items before

122

Table 8

Reliabilities1 of Original and Revised COP Scales, EXPCLIs

 

 

 

 

 

 

Scale Name n of Items ReliabilityCi)
A Priori /’Revised A Priori 7 Revised
1.1A PSYSAFTY 7 / 7 .89 / .89
1.1B PRCVRLTY 5 / 5 .85 / .85
1.10 PROCSFAC 5 / 4 .70 / .79
1.1D LINKAGEN 5 / 4 .73 / .77
1.2 PROBRDF 8 / 8 .91 / .91
1.2A DISCREP 5 / 4 .79 / .83
1.2B ORGNEED 5 / 5 .74 / .74
1.20 EXRLTY 5 / 5 .69 / .69
1.2D THRYINPT 5 / 4 .54 / .60
2.1 GOALFOCS 5 / 4 .29 / .36
WHYDIFF 6 / 5 .49 / .60
WHATDIFF 6 / 5 .82 / .85
3.0 PROBOWN 5 / 5 .73 / .73
4.1 SLTNRESRCS 5 / 5 .58 / .58
4.2A SLTNRELV 6 / 5 .72 / .79
4.28 SLTNLINK 5 / 5 .78 / .78
4.3 SLTNIMEV 7 / 7 .84 / .84
Process Variables:
P1 COG 5 / 5 .39 / .39
P 2 VAL 5 / 4 .74 / .77
P 3 EMO 6 / 5 .87 / .87
P 4 BEG 5 / 4 .60 / .61
P 5 BEN 8 / 8 .87 / .87
P 6 COMPT 5 / 5 .91 / .91

 

1 Coefficient alphas

123

and after the revisions. This analysis indicates that the scales
have generally high measures of internal consistency. Only three
scales (GOALFOCS, 2.1; SLTNRESRCS, 4.1; and 00c) show coefficient
alphas of less than .60 after removal of weak items, and one of these
(SLTNRESRCS) had an alpha of .58. In the revision process, only

one item in each of 11 scales needed to be removed to improve

scale reliabilities. In no case were items reassigned to other
scales. Generally the increase in scale reliability after removal

of the deficient items was slight (Range: .002 - .09; ii= .502). All
of the revised scales were included in subsequent analyses (despite
the relatively low alphas of two of the scales).

Scale Intercorrelations

 

Table 9 presents the intercorrelations among all of the COP Model
variables for the EXPCLIs (the processing variables are analyzed
separately). A separate analysis is also conducted later in this
chapter for the INEXPCLIs. All of the inter—scale correlations are
corrected for attenuation using the reliability estimates in the
diagonal. Scales are grouped in the order of the original COP Model.
This matrix shows that strong intercorrelations exist among most of
the scale variables, suggesting the presence of a general "halo"
effect in the clients' evaluation of the consultant, i.e., the ten-
dency to evaluate him similarly across dimensions.

The presence of a "general factor" serves to obscure the
relationships of the defined scale variables. In order to obtain a
better estimate of the actual scale score relationships, an attempt

was made to remove this "general factor.’ Specifically, a new scale

124

.on n z camemm «accommfid one

cw mopmefipwo mpflafinmflaoh one mafia: :Ofipmzcoppm How dopoohhoo

 

 

 

 

 

 

 

 

 

 

H
Aewv mm mm so am mm no em me as an no on we we or on >mszeqm m.s
Arsv mm mm om mm mm am so mm mm on mm mm an mm as szgzeqm mm.:

Amsv mm as we ed as Ho as mm mm or we on mm mm >qmmzeqm <N.:

Ammv do as on om me- on am we so a: we as no mommmmzsqm H.:

Amsv or as so do- or so am am so me as do zzomomm o.m

nwv em or mm as do me no so mm mm mm saHaeam: m.m

Aoov ma oo mm mm mm Hm mm mm or me aaHmsmz m.m

Acmv me- mm mm mm mm mm or as or moosg<oo H.m

Aocv cm as do mo- r em as no. mo emszmre QN.H

Aacv so mm mm or we or on segmxm om.H

Aesv or mm on mm mm mm amazomo mm.fi

Amrv or mm or on ma ammomHm <N.a

Afimv mm ms or do hammers N.H

Assv on me on zmu<szq mH.H

Adda so an mammoomm oa.H

Amrv we seqm>omm mH.H

Aomv seaamwmm «H.H

 

 

 

m.: mN.: <N.: H.: o.m New N.N H.N QN.H UN.H mN.H <N.H N.H QH.H UH.H mH.H <H.H

H

onEdm qumxm .wqowdeoHHoo oadomnhopaH

m magma.

125

was defined as being the sum of all scales and this factor was
then statistically partialled from the matrix. This partialled
matrix, corrected for attenuation using the new reliability estimates,
is presented in Table 10.
The partialled matrix, generally speaking, shows strong residual
variance left after the removal of the scale sum. Twelve of the
17 scales have coefficient alphas after partialling greater than .50.
A careful review of the partialled matrix suggests the follow—
ing conclusions:
1.) GOALFOCS does not hold up as a tenable cluster (a = .02). While
it shows strong correlations with other variables, especially those
associated with the first half of the model (as would be predicted),
all of these correlations are highly elevated by the correction for
attenuation. This scale is therefore dropped from subsequent
analyses.
2.) There is generally decreasing internal consistency as one moves
from PSYSAFTY to SLTNIMEV, suggesting less variance independent of
the general factor for the scales associated with the latter portion
of the model;
3.) The scales PROCSFAC, LINKAGEN, ORGNEED, THRYINPT, and SLTNLINK
do not fit in well with their a priori assigned groups, and are
reassigned or omitted, as discussed below;
4.) The scale PROBOWN appears to be related to other dimensions,
especially those in the PROBRDF categories, contrary to predictions;
however, the scale's low measure of internal consistency (a = .27)

elevates these apparent relationships considerably. It does appear

126

.mm H z mamsdm nadcowmwv on» :H mopdeflpmo hpdaflpwfiaoh can mafia: coavmszovvw Mom uopoohhoo

 

 

 

 

 

 

 

 

 

 

 

 

 

 

H
Ascv mH- ma mm as. a: mo mo- mH so no- as rm- mH on nH- or- smszegm m.e
Ammv NH- no. or- mH- mm mm no- am so- on- 0H- Hm- on- no am equzeqm mm.o
Asmv cw mo- Hm- mo or- mm- mm. we so- no mH- oH Hm Ho- >qmmzeHm <m.:
Ach do Ho em- Ho we. so- me so me me: cm mm. mm. mommmmzegm H.:
Asmv cH- 00 or- :3. HH- Hm mm m: mH- or- mH- mH zzomomm o.m
AHQV mm or. no mo- mo- no- or- om mo mm- mm- aaHaeams m.m
AomV mH Hm- as. Hm am. «0. mm- as- no mo- aaHmsmz m.m
Amov mm- on am- am me He Hm mm m: moosqaou H.N
Aswv 1Nm 11mm- mo- so- mm HH Hm- mo- emszmme mN.H
Aamv as- an Hm or so or- en. semem om.H
Ammv 3H 2H NH mm- mo mm mmmzumo mm.H
Asmv em on. no no mm mmmomHm <N.H
Asmv can mH mH oH- hammers N.H
Amnv mo- mH- AH- zmueszq QH.H
Amnv oH mm- oaamoomm oH.H
Area on seam>omm mH.H
Amsv sesamsmm «H.H
m.: mm.o <N.h H.e o.m m.m N.N H.N QN.H om.H mN.H «N.H m.H QH.H oH.H mH.H <H.H

 

oHasem HHomxm .xHeeoz coHecHoesoo oHoom-scecH HcHHcHeeem

H
OH memB

127

to be more strongly related to scales in the first rather than second

half of the model;

5.) The scale ORGNEED shows more consistent relationships to variables

in the 1.1 (Facilitation of information flow) category, than the

originally assigned 1.2 (Problem conceptualization and redefinition)

grouping; it is therefore reassigned to this group;

6.) The scales LINKAGEN and THRYINPT show a relationship to one

another, but no consistent relationship to other scales. Recalling

the pattern of CIs associated with these two groupings, i.e., strongly

negative as well as strongly positive examples, the pattern of associ—

ation with each other and generally negative correlations with other

variables may be interpreted as reflecting an underlying dimension

of ambivalence regarding the consultant's activities in this area;

7.) The scale PROCSFAC appears to be more closely associated with

the "solution authorship" grouping (4.0), contrary to predictions. It

will also be recalled that very few CIs were given that were coded as

being examples of this dimension. Perhaps clients view what little

consultant activity that does occur in this dimension as "solution

oriented" in that it is implicitly associated with a judgment that

the present way of processing needs to be changed; this is speculative

reasoning, however;

8.) SLTNLINK, which does not fit in well with its theoretical assign-

ment, does not show an interpretable pattern of correlations with

any other scales and is therefore dropped from future analyses.
Because of these findings, a revised "best fit" of the model and

data was made (Table 11). This shows a much neater, more consistent

128

.om u z mamsdm ”accowcfid map ca mopcefivmo hpfiaflpmaaon on» mnams mums coon
.mpdu .mHAvmxm

o>c£ cowvmdcoppc Mom mGOHvooHHoo use .doaadfiphdm ohm mGOHPmHophoo oacomuyopca one
one one Hoooz moo one no eeHe ence: one mcHeconowaow .oH oHpee mo soHecwamHesooow c nH chce ane

 

 

 

 

 

 

 

 

 

 

 

 

H
.ldmmv mm HH mH mm- we. we Hm- es. an no- mm so- Hm- mo- emszmme om.H
Aonv no- rH mH- me. om mm- mH- or on- or. NH mH- AH- zmu<szq QH.H
Ammv on! cH cm No a:. co. so we mH mm- OH or. oaamoomm oH.H
Asov m: mm as No as- we do mm- mo- mH- um >mszeHm m.:
Asmv rm Hm- mo mo- me. so- no N: Hm Ho- >Hmmzeqm <N.:
AcHV Ho am. am so- so we me an- em. mommmmzeqm H.r
AHcv mm cH- no. mo- es- mo- mm- mm- saHseams m.m
Aomv co so. am. No- Hm so mo- saHowmz m.m
Asmv HH- mm m: Hr mH- mH zzomomm o.m
Army em Hm as. or- an. semem om.H
Asmv an :H no mm ammomHa «N.H
Asmv :H aH oH- hammers N.H
Ammv mo mm mamzomo mN.H
Area on segm>omm mH.H
Area ses<mwmm «H.H
om.H QH.H UH.H mam .«qu .mwm .mqm mew awn om.H em.H N.H mN.H mH.H «H.H

 

oHasem HHomxm .Hoeoz moo eonH>om

Ha oHQwB

129

representation of the data that does not compromise the original
conceptualizations in significant ways. Analysis of this Revised
COP Model suggests:

1.) A fairly tight first cluster constructed with the addition of
ORGNEED to PSYSAFTY and PRCVRLTY, and the deletion of LINKAGEN and
PROCSFAC. The content of this group remains the same, i.e., expli—
cation of clients' views of reality, with ORGNEED interpreted to be
a means by which the consultant elicits information, and by which
he presumably communicates a benevolent interest in and concern with
the organization (although the measure of internal consistency for
ORGNEED after partialling the general factor is lower than desirable,
-23):

2.) The second cluster, Problem conceptualization and redefinition,
forms a tight grouping, although the addition of PROBOWN is at best
a marginal fit. This is because two of its inter-scale correlations
are higher than its low coefficient alpha; another of these
correlations is in the negative direction. The content of the
overall grouping is consistent with the original interpretation made
in the delineation of the model;

3.) The removal of GOALFOCS from the next grouping (due to its low
measure of internal consistency) leaves only the WHYDIFF and
WHATDIFF clusters for the "confrontation" variables. Although this
grouping is especially good for the WHYDIFF variable, the WHATDIFF
scale shows higher intercorrelations with two scales, SLTNIMEV
(r = .49) and THRYINPT (rxy = .65). While these correlations may

xy
be chance associations, they threaten the validation of this clustering,

130

which at best is tentative. It will be recalled that few examples
of consultant behavior from the CI data were classified as instances
of WHATDIFF;
4.) The "Solution authorship" grouping forms a relatively tight
cluster, but the SLTNRESRCS variable fits in less well, primarily
due to its low coefficient alpha (.16). PROCSFAC, though grouping
well with this set of scales on a statistical basis, has little
theoretical (content) rationale for inclusion here;
5.) A small residual cluster of two scales, LINKAGEN and THRYINPT
also emerges, as previously discussed. Data supporting the speculated
"ambivalence" hypothesis is suggested by the fact that these two scales
had the lowest mean scores of all the scales for the EXPCLIs' group.
In any case, these scales are eliminated from further analyses
due to their deviation from the COP Model and the lack of conclusive
data for alternative hypotheses to explain the observed findings;
6.) The two remaining scales, GOALFOCS and SLTNLINK are omitted
from further analyses due to their inconsistencies with the COP Model
that have already been discussed.
Scale Analysis of the INEXPCLIs' Data

Because subsequent analyses compare data obtained from the EXPCLI
sample with that of the INEXPCLI sample, it is important to note the
psychometric properties of the latter group's questionnaire data.
To permit comparisons with the EXPCLI sample, the same items were
used for the scale components for both groups, even though these were
not in all cases the best items for the INEXPCLI scales. Table 12

presents the inter—scale correlations for the INEXPCLIs, corrected

.:m u z onsdm “HocomoHo map :H mopmermo thHHQoHHoM one mnHm: sOHposcoppm Mom oopoohhoo

H

 

 

131

 

 

 

 

 

 

AHov oo so os oo so no mm on no ss ms oo os so no mo
Amsv mm as oo no so Ho os Ho oo so ms on so so Hs
Ammo ms Ho so so oH cs ss as no so oo oo mo so

Hoov so so ms No so so so Ho so ms so Ho Ho

Aoov om om om NH no so as so oo oo ms Ho

Aosv mm so on so oo as so oo oo ms ss

Ammo on mm on os oo so No No No so

AHov om so os so ms os os so oH

Aoov so Hs on on no- ms oH oo

Aoov so Ho os os mo os oo

Aooo so ms os os so Hs

Amov mm on oo oo so

Homo no no so ss

1. Aoov ms oH on

Aoov Hs mo

Aeov oo

Hoov

 

 

 

 

film om.s «N.s H.s o.m, new N.N H.o oo.H om.H oo.H sm.H N.H oH.H oH.H oH.H <H.H

>mszeam
szQZBAm
>Ammzeam

mummmmZBAm

zzomomm
mmHQH<m3
hhHowmz
moomq<ou
BQZHMmmB
wequm
ommzumo
mmmomHQ
mammOmm
sz<MzHH
U<hmoomm
MBAm>omm
MBh<mwmm

m.s
oo.s
em.s

H.s

o.m

m.N

N.N

H.N
sN.H
oN.H
oN.H
aN.H

N.H
oH.H
oH.H
oH.H
«H.H

 

onsom qumxmzH .chHpmHoHHoo oHoomihoch

H
NH oHQdB

132

for attenuation using the reliability estimates in the diagonal. As
evident, the measures of internal consistency are again high, but the
matrix is plagued with the same "general factor" problem encountered
with the EXPCLI sample. Accordingly, this matrix was also partialled,
again using as the general factor the sum of the individual scale
scores. This partialled matrix, with revised reliability estimates,
is presented in Table 13.

Again, the variance remaining in each scale after partialling
is substantial, with the exception of SLTNRESRCS, which is omitted
from further analyses due to its low estimate of internal consistency.
Other scales show quite high measures of internal consistency, with the
exception of EXRLTY (a = .22). The grouping of scales in Table 13 is
in the order hypothesized by the a priori COP Model. Analysis of
this configuration suggests that the model does not fit exactly as
predicted. Nor does the Revised COP Model developed for the EXPCLI
sample constitute the best fit with the INEXPCLIs' data. Rather, a
slightly different arrangement appears to be most consistent with
the data. This Revised COP Model for the INEXPCLIs is presented in
Table 14.

This re—configuration retains the essential features of the
COP Model; however, several of the scales are reassigned. As with
the EXPCLIs, ORGNEED fits better with the variables associated
with 1.1 (Facilitation of information flow) than with its original
placement (1.2, Problem conceptualization and redefinition). PROCSFAC

drOps out of this category also, just as it did with the EXPCLI

29H: moHoom 39H: opHHoqsssoo one How oo.H mopSPHpmpsm AHumH .conoo duo Hopcsmv vow: summonm Hopszoo one

.monmdo pcoHOHmmooo o>Hpomoc

.:m n z oHQSdm «Honomoos who :H mopwavao ovHHopoHHoH may moon: GOHpmzsoppm How covooshoo

N

 

 

133

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

H
Asov oN- No NHn-osO- os HH- Oo- HO oH oO- Os- oH- NN oO- so- oN- soszsqo o.s
Aoov mo oN oO oN- Hs sH so oN- sO ss- oo- oN Ns- ss- oO- ozngsHo oN.s
AOov NH- Hs OH so ON- sH HO- oO oo- Hs- so- oN- so- NO- >ooozsgo «N.s
NAoH-v HO- oo- oH- HH NH- Ho oO NH- sH HO Ho oO NO oOoooozsoo H.s
AHsV.~oO oo oo- Ho- OH- oo- NO- OH HH- NN oN- Ho stoOoa O.o
Aoov Ho ON- HH- NH oO- oO- oN- oH- oH- Hs- HH ssHOeoos o.N
Hosv sO NO Ho- No- sO oO- oN- oN- oO- oO- asHooo: N.N
Hosv OH os oO- Ho HO sO- No- sN Oo- oOOoHooo H.N
Asov ms oO oO NO sH- NH so- os- sazHHoos ON.H
ANNV so oO os- OO oN oO- so- osqogo ON.H
ANsv sN- oo- oH oH- Hs NH OoonoO oN.H
Asov no OH so No sH- mooooHo <N.H
Aoov sH- ss ss HH- sooooom N.H
Aoov sH so- oO- zoosngq oH.H
AHsv oN- oN- Osgoooom OH.H
AHoV os spoo>Oom oH.H
Assv seasosom «H.H
o.s oN.s «N.s H.s awn o.N N.N H.N ON.H ON.H oN.H «N.H N.H oH.H OH.H oH.H «H.H

 

oHaeco qumxozH .wHweoz soHeoHowsoO oHeoo-woesH OoHHsHewoa

H

mH oHDoB

134

.:m n z onsdm “Honoond can so moposopmo thHHDoHHoH map wchs come soon
o>o£ :OHpcscoepm sow m:0HpooHHoo new .OoHHmeHdm who mGOHPoHohHoo oHoom-MopCH one

.33 .oHonozH
one soc Hoeoz mOO one oo :eHo soon. on» osHesooowoow .NH oHpos so coHeowaoHssooow o oH oHpos ans

 

 

 

 

 

 

 

 

 

H
Asov oN- No sO- os HH- Oo- oO- HO oH Os- oH- oO- so- oN- >oszsqo o.s
Aoov so sO sN- Ns sH Ns- so oN- ss- oo- sO ss oO- oquzsoo oN.s
AOov Hs OH so ON- oN- sH HO- os- Hs- oO so- NO- >qoozsoo «N.s
AHsv oO oo .oo- NN Ho- so- NO- OH oo- oN- Ho stoOom O.o
soov Ho ON- oH- HH- NH oO- oN- oO- Hs- HH asHOssoz o.N
Aosv sO oN- NO Hs- sO sO- No- oO- oO- ssHosoz N.N
Aosv No- OH os Ho HO sO- sN Oo- oOOoqsoo H.N
AHsV NH oN so ss oH- sN- sN- OosoOOom OH.H
Asov os oO NO oO so- ss- emszoms oN.H
ANNV sO os- so sO- so- osqoxm ON.H
Asov no sN- No sH- mooooHo oN.H
Aoov oo- ss HH- aOooOom N.H
ANsV Hs NH OoozooO oN.H
AHov os osqo>Oom oH.H
Assv ossoosom «H.H
ON.s oN.s «N.s O.o o.N N.N H.N OH.H oN.H ON.H «N.H N.H oN.H oH.H <H.H

 

HoHseoo qumxmzH .Hoeoz mOO OonH>oo
sH oHpos

135

group. The a priori grouping of scales associated with 1.2 (Problem
conceptualization and redefinition) is more consistent with the
INEXPCLIs' data than was the case for the EXPCLIs. The addition

of the PROCSFAC scale (1.10) is a minor threat to interpretability,
but the fit to the other scales is far less than perfect. As with
the EXPCLIs' data, PROBOWN does not emerge as unrelated to other
scale groupings. It appears to be most closely associated with

2.0 (Confrontation). GOALFOCS, which had been hypothesized to be
grouped with the Confrontation scales, shows a more consistent
grouping with the Problem conceptualization and redefinition scales.
As with the more experienced clients, LINKAGEN shows no consistent
relationship to other scales, and so is dropped.

A comparison of the Revised COP Models for both EXPCLIs and
INEXPCLIs is presented in Table 15. The basic structure of the model
emerges essentially intact; however, the scales do not perfectly
fit the model for either sample, even after reassignment, though
the "fit" of the revised analyses to the model is generally quite
good.

For both EXPCLIs and INEXPCLIs, it is of interest to know
in more concrete detail what it was that was partialled from the
matrices of scale inter-correlations. An analysis was therefore
made of those items with high part-whole correlations with the
general factor. As presented in Table 16, using only those scales
that showed validity for both EXPCLIs and INEXPCLIs, it is apparent
that for both groups,.over half of the high loading items (rxy> 60)

fall into the model's first two dimensions, i.e., 1.1, Facilitation

136

Table 15

Comparison of Revised COP Model, EXPCLIs and INEXPCLIs1

 

EXPCLIs INEXPCLIS

1.1 Facilitation of Information Flow

 

1.1A PSYSAFTY 1.1A PSYSAFTY
1.1B PRCVRLTY 1.1B PRCVRLTY
1.2B ORGNEED 1.2B ORGNEED

1.2 Problem Conceptualization and Re—definition

1.2 PROBRDF 1.2 PROBRDF
1.2A DISCREP 1.2A DISCREP
1.2C EXRLTY 1.2C EXRLTY
3.0 PROBOWN 1.2D THRYINPT

1.1C PROCSFAC
2.1 GOALFOCS

2.0 Assessment of the Need and Capacity for Change

 

2.2 WHYDIFF 2.2 WHYDIFF
2.3 WHATDIFF 2.3 WHATDIFF
3.0 PROBOWN

4.0 Joint Authorship of the Solution

 

4.1 SLTNRESRCS 4.2B SLTNLINK
4.2A SLTNRELV 4.2A SLTNRELV
4.3 SLTNIMEV ' 4.3 SLTNIMEV

1.1C PROCSFAC q

 

1 A listing of the scales demonstrating construct validity for each

sample. (See text for a detailed explanation.)

137

Table 16

Breakdown of Items with High Loadings1 on Each Group's General Factor

 

EXPCLIs INEXPCLIs
a 2 .r_1 2

1.1 Facilitation of Information
Flow

1.1A PSYSAFTY 3 2
1.1B PRCVRLTY 3 4
1.2B ORGNEED _3_ _J;_
9 25.7 8 24.2
1.2 Problem Conceptualization
and Redefinition
1.2 PROBRDF 6 7
1.2A DISCREP 2 2
1.20 EXRLTY _2_, __q_
10 28.6 9 27.3
2.0 Need & Capacity for Change
2.2 WHYDIFF 1 2
2.3 WHATDIFF _g_ _g_
3 8.6 4 12.1
4.0 Solution Authorship
4.2A SLTNRELV 2 1
9.3 SLTNIMEV _;__ _g_
3 8.6 3 9.1
Other 10 28.6 9 27.2
Total 35 100 33 100

 

r
KY

 

138

of information flow and 1.2, Problem conceptualization and redef-
inition. This suggests that for both groups of Clients, even though
there is representation in the "general factor" of items from almost
all scales, the two primary contributions come from consultation
elements associated with helping them define and conceptualize
problems. If the latter is viewed as a facilitative role and the
former as a structuring one, a strong hypothesis is suggested, consis—
tent with a great deal of literature in the psychotherapy and leader-
ship areas, that the general perception clients have of this
consultant is his "benevolence" and his "structuring" capacities.
These two variables represent, in Strupp's terms, personal qualities
and technical skills (Strupp, 1976), or, in leadership research,
consideration and initiating structure (e.g., Korman, 1966). In
short, an interpretation of the "general factor" found in this
research is that it represents two psychological variables similar to
those found in the psychotherapy and leadership research. A
systematic test of the centrality of these dimensions to consultation

more generally awaits further research.

fiypothesis 1A
This hypothesis is an attempt to find other aspects of the
consultative relationship that were helpful or nonhelpful that
were not identified or included in the original COP model. The
data for this research question are provided by those CIs that
were not codable according to the model presented in this research.
Review of Table 6 shows that only 11% of all Cls could not

be coded to the model (11.5% of the EXPCLIs' CIs and 10.2% of the

139

INEXPCLIs' CIs). These behavioral examples were cited for both
helpful and nonhelpful consultant activity, with nonhelpful
dimensions predominating in the EXPCLI group. Table 17 summarizes
the new categorizations for both client groups.
Other Helpful Dimensions

Two major categories of helpful consultant behavior not classi-
fiable according to the model emerged. One group of Cls suggested
that "moral support" or reinforcement of the client's viewpoint
was valuable. This is exemplified by the following:

...Dr. Frost's appraisal of the situation...was very helpful

in that it confirmed and supported my estimate of the situation.

...his support stren thened my advice and helped in getting

a company executiv to adopt the advice and change his

approach.

The second grouping that did not fit well with any existing
categories of the COP Model was the generation of enthusiasm among
organizational personnel, e.g.,

...There were many positive comments frOm this endeavor

but most of all, the personnel in my specific area were

turned on and started input for improvement and increased

productivity...

The miscellaneous examples included isolated instances of

consultant behavior that was viewed as helpful, e.g., testimony
before a labor relations board, or the consultant's persistence
in keeping after an issue.
Other Nonhelpful Dimensions

The consultation examples of nonhelpful behavior that were
not codable using the COP Model were primarily given by EXPCLIs

and primarily dealt with variables having to do with the process

140

Table 17
Analysis of Critical Incidents Not Classifiable

According to the COP Model

 

Helpful Dimensions

 

Direct reinforcement or support
Enthusiasm/enhancing motivation

Miscellaneous

Nonhelpful Dimensions
Too indirect or vague
Nonavailability

No followthrough
Negativism

Not supportive

Miscellaneous

EXPCLIS

INEXPCLIS

H-C’N

Total

H
«a \n \n so

wwwxo

N

 

141

of consultation, rather than the content of the interventions. A
relatively large number specifically relate to the consultant‘s
being too indirect, abstract, or vague, so that the client is not
clear what point(s) the consultant is attempting to make. For example:
...[The consultant] is more direct now...prior to that I
had many situations where the questions were so obscure
that I couldn't "make a connection."...
Other dimensions include non-supportiveness (the inverse of
the "supportiveness" discussed above), negativism, and variables
dealing with the process rather than content of consultation, viz.,
not being available, not following through, or not keeping appointments.
Thus, essentially three variables emerge from the non-codable
CIs: reinforcement or support, inducement of motivation, and
variables related to the way the consultant implements his consult-
ation. The fact that so few new categories were needed and that so

few CIs could not be coded using the COP model's original dimensions,

lends indirect evidence of validity to its component variables.

Hypotheses 2; 2A

The question next arises as to the ways, other than the scale
structure differences already discussed, in which INEXPCLIs differ
from EXPCLIs in their use of the consultant and their perceptions
of his behavior. These hypotheses specifically suggest a) that there
will be differences among the two groups, and b) that the differences
will follow a specific pattern, a generally hierarchical structure
with early model dimensions dominant among those with limited

consultative experience and all model dimensions associated with the

142

EXPCLIs group. For this analysis, only those scales showing psycho-
metric stability for pgpp EXPCLIs and INEXPCLIs were used (see Table
15), as a conservative measure of group differences.

Table 18 presents the mean scores and standard deviations for
those scales that supported the COP Model in the previous analyses.
From this listing, it can be seen that in every case the clients from
the EXPCLI sample rated the consultant more highly than those in
the INEXPCLI sample. It is also noteworthy that in all but two cases
(WHATDIFF and SLTNIMEV) the standard deviations were smaller for the
experienced sample, suggesting more consistency in this group's
perceptions of the consultant.

To assess the significance of the differences between the two
samples, a multivariate analysis of variance (MANOVA) was performed
(Table 19), which was strongly statistically significant. From
analysis of the univariate ANOVAs, it can be seen that 7 of the
ten scales show statistically significant mean differences (p < .05
or better). However, the variables showing these differences
are not always the ones predicted to vary by Hypothesis 3A.

This hypothesis (2A) suggests that dimensions later in the
COP Model will differ more strongly than dimensions associated with
the earlier model components. To test it, a step down discriminant
analysis procedure was employed. If the hypothesis is supported,
the model's latter dimensions should enter the discriminant function
first, since they should account for a larger proportion of inter-
group differences. From Table 20 it can be seen that this hypothesis

receives mixed results. It is true that the most discriminating

143

Table 18

Means and Standard Deviations of DOC Scales1

 

 

§9§l§ ggpgpgsz lNEXPCLIs3
_i_ Std. Dev. _i_ Std. Dev.

PSYSAFTY 43.80 6.66 39.65 7.02
PRCVRLTY 31.68 3.80 27.74 5.22
ORGNEED 30.63 3.96 28.26 4.16
PROBRDF 47.29 6.55 42.85 7.88
DISCREP 22.96 3.65 21.52 4.03
WHYDIFF 28.68 4.67 27.91 4.87
WHATDIFF 28.00 4.90 27.28 4.56
SLTNRELV 30.50 3.98 26.54 4.24
SLTNIMEV 38.36 7.78 34.35 6.28
EXRLTY 26.30 4.74 26.22 4.12
1

EXPCLI and INEXPCLI samples.

2 N = 56

351:5“,

Includes only those scales showing construct validity for both

 

144

Table 19

Univariate and Multivariate ANOVAs to Test for Differences

Between EXPCLI and INEXPCLI Groups

 

DOC Scale

PSYSAFTY

PRCVRLTY

ORGNEED

PROBRDF

DISCREP

WHYDIFF

WHATDIFF

SLTNRELV

SLTNIMEV

Mean
Square

474.700
426.288
153.859
540.447
57.463
16.349
14.339
431.747
441.019

Univariate
F1

10.150
20.566
9.348
10-333
3.900
.719
.639

Significance
(p less than)_

.002
.000
.003
.002
.051
.398
.426
.000

.004

F Ratio for Multivariate Test2 of Equality of Mean Vectors = 9.7116

p < .00001

1 Degrees of Freedom

2 Degrees of Freedom

1, 108

9- 972

 

145

Table 20

Discriminant Function Analysis to Test for Differences Between
EXPCLI and INEXPCLI Groups

 

 

Standardized 1
Order of Entry Discriminant Function Change in Significance

Coefficients Rao's V (p less than)
SLTNRELV —.9077 25.57 .000
EXRLTY .6626 7.28 .007
WHATDIFF .5833 5.97 .015
PRCVRLTY -.8424 12.496 .000
WHYDIFF .3468 4.001 .045
DISCREP .3025 1.882 .170
SLTNIMEV -.1521 .486 .486
PROBRDF -.0813 .077 .782
PSYSAFTY —.0352 .029 .865

ORGNEED is not in the equation.

1Chi-square of change in Rao's V.

 

146

scale, SLTNRELV, is from the last part of the model, as expected.
It is also true that there is generally a reverse hierarchy, with
scales from early stages of the model accounting for very little of
the discriminant variability, and scales from the latter part of
the model accounting for relatively more. However, there are
exceptions. For example, SLTNIMEV does not enter the equation until
near the end and shows fairly limited discriminatory power. EXRLTY
and PRCVRLTY also show comparatively high standardized discriminant
function coefficients, despite their being associated with the
earliest model dimensions. In short, data provide mixed evidence
for the hypotheses tested. It suggests that clients do vary in
the ways in which they utilize the consultant, and that generally they
do so as predicted, but with several exceptions.

A possible explanation for this finding is that Hypothesis 2A
is really not well tested here because the INEXPCLIs group have
not reached even the earliest stages of the model. In other words,
the pattern of group differences does not more neatly fit the
hypothesis because the INEXPCLIs have not yet attained an adequate
level of even the most basic consultation dimensions presented in
this study. Although the clients from the INEXPCLI sample did rate
the consultant much more highly than would be expected if this is
true, it must be recalled that the large "halo" factor was present
even for the INEXPCLIs. Recalling the lower than desirable return
rates of questionnaires for this group, it can be conjectured that a

more adequate sample of INEXPCLIs would rate the consultant even

lower on these dimensions. If this is the case, it may well be

147

that the "critical mass" at which the prOposed theoretical differences

emerge is not in fact five sessions, but perhaps considerably more.

Hypothesisg3_

This hypothesis, also exploratory, suggests that there will be
differences among those rating the consultant favorably and those
rating him unfavorably on a criterion measure, i.e., that there will
be intra— as well as inter—group differences in the capacity to
effectively make use of the consultant's interventions. Separate
analyses are conducted for both groups, given the hypothesized and
actual differences between them.

The Criterion

As discussed in Chapter III, the criterion used in this study
is a perceived rather than behavioral measure. It consists of the
combination of items listed in Table 21 into a single criterion
score. Coefficient alphas for each group are also reported in this
table, and show fairly strong internal consistency to the criterion
scale.

EXPCLIs

The results of multiple regression analysis for the EXPCLI
group is presented in Table 22. To permit inter—group comparisons,
only those variables included in the Revised COP Model for pgph
the EXPCLI and the INEXPCLI groups are included in the analysis.

This procedure shows that the majority (63.2%) of the variance
is accounted for by two consultation dimensions: PROBRDF and

PSYSAFTY. Interpretation of this finding is complicated by the

148

Table 21

Criterion Items and Reliability Estimates

 

 

 

Item Number Item Content
27. I would strongly recommend working with him to
a close friend in the organization.
62. I spend very little time thinking about issues
we discuss when I'm not with him.
124. He has a great deal to offer me.
132. I cannot imagine why this company pays for

his services.

140. How helpful, OVERALL, have your experiences
with Dr. Frost been to ygu personally?

141. How helpful, OVERALL, do you feel Dr. Frost has
been to your company?

Reliability Estimates

 

 

Group Coefficient Alpha
EXPCLIsl .87
INEXPCLIsZ .84

1 N = 56

2

N=54

 

Multiple Regression Analyses of Revised COP Model Scales on the

149

Table 22

1

Criterion of Client Ratings of Consultant Effectiveness, EXPCLIs

 

 

 

 

. Beta 3 Zero Order . . . 2 2
Variable Weight Correlations E Significance R_ R Change
PROBRDF .202 .740 65.213 .000 .547 .547
PSYSAFTY .179 .678 12.215 .001 .632 .085
SLTNRELV .215 .671 2.446 .124 .648 .016
DISCREP .162 .675 .901 .347 .655 .007
PRCVRLTY .084 .683 .253 .617 .657 .002
EXRLTY -.096 .503 .749 .391 .657 .000
SLTNIMEV -.038 .416 .098 .755 .657 .000
WHATDIFF -.045 .450 .087 .769 .658 .001
ORGNEED -.052 .612 .116 .735 .658 .000
1N=56
2 Includes only those variables from both the EXPCLI and INEXPCLI

Revised COP Models;

the equation.

WHYDIFF does not meet criteria for entry in

3 Beta weights are regression coefficients of variables entered on
the last step of the analyses.

150

problem of multicollinearity, i.e., the rather high intercorrelations
among the scales in the original (unpartialled) matrix. Because

of this difficulty, the reliability of the regression coefficients

is lessened, and they must be cautiously interpreted.

Two data sources were also examined to attempt to obtain a
less contaminated estimate of the relationship between the COP
"predictors" and the criterion. First, scale scores were combined,
using the Revised COP Model as the basis for the combination.
Although the scale intercorrelations (Table 23) are decreased some-
what by this procedure, there are still relatively strong inter-
scale correlations. The multiple regression procedure performed
on this data gives essentially the same solution as the first one:
a strong "problem definition" component and a relatively less strong
"affective" component.

A second, more helpful, data source is the zero order correla-
tion coefficients between the scale scores and the criterion.

Table 24 presents both the unpartialled and partialled correlations
between the COP variables and the criterion. In the latter case,
the "general factor" has been partialled from the correlation.

In both cases, a correction for attenuation has been made in the
criterion only. Again, the unpartialled correlations are generally
quite strong. The partialled correlations, the best estimate of
the true client perceptions, indicate the strongest correlations
with the criterion are with the PSYSAFTY and PROBRDF variables.
Lower or negative correlations are found with the WHYDIFF, WHATDIFF,

SLTNRELV, and SLTNIMEV variables.

151

Table 23
Combined Scales Analyses, EXPCLIs
Reliability Estimates of Combined Scales
Scale Revised<1
Scale 1

PSYSAFTY .92
PRCVRLTY
ORGNEED

Scale 2

PROBRDF .93
DISCREP
EXRLTY

Scale 3

WHYDIFF .81
WHATDIFF

Scale 4

SLTNRELV .88
SLTNIMEV

Intercorrelation Matrix2

 

sole.
1 (92)
2 85 (93)
3 7o 7o <81)
___9 ....... 93 _____ éQ-__-§i-___£§§2 ____________________________________
Beta Regression Analysis
§2§l§. ngghp F_ Significance RE_ RE_§h§gg§
2 .185 74.15 .000 -579 .579
1 .130 8.89 .004 .639 .061
4 .025 .34 .801 .642 .002
N = 56

2Coefficient alphas in the diagonal; correlations are corrected for
attenuation.

3Scale 3 does not meet the entry requirement for the regression
equation.

152

Table 24

Correlations Between Revised COP Model Variables
and the Criterion, EXPCLIsl’ 2

 

 

 

 

 

V . Zero Order ny ny After Removal of3the
ariable General Factor
PSYSAFTY ~73 ~35
PRCVRLTY .73 .20
ORGNEED .65 .02
PROBRDF .79 .26
DISCREP .72 .27
EXRLTY .54 -.18
WHYDIFF .55 —.01
WHATDIFF .48 -.26
SLTNRELV .72 .10
SLTNIMEV .45 -.28

1 N = 56

2 Corrected for attenuation in the criterion only.

3 Correlations between the criterion and listed variables after
partialling of the "general factor." A correction for attenua-
tion has been made only in the criterion, using the revised
reliability estimates after partialling.

153

Collectively, these findings add strength to the previous
conclusions regarding the primacy of two major consultative dimen-
sions: ability to establish a relationship with the client and

ability to help the client identify and re—define problems.

INEXPCLIs

The same analyses were performed separately for the INEXPCLIs
using the same variables as for the EXPCLIs. Tables 25, 26, and
27 present the results of these analyses. Contrary to the
previous finding that the EXPCLIs and INEXPCLIs differed in
significant ways from one another, the present data indicates
the same general pattern of predictors associated with favorable
consultant evaluation: a central factor of problem identification
and re—definition, and a second factor of perceived benevolence.

The similarities in the patterns between the INEXPCLIs and
EXPCLIs may in part be accounted for by the exclusion of variables
that showed support for the EXPCLI or INEXPCLI groups independently.
However, separate analyses performed for each group using all
variables from each group's Revised COP Model indicates no change
in the general findings already reported. What this suggests
is that INEXPCLIs may differ from EXPCLIs not in the quality of
their relationship, but in the quantity of it, i.e., INEXPCLIs
see the consultant as performing essentially the same variables
as do EXPCLIs, but to a generally lesser extent. This also suggests
that for both groups, the consultant will be perceived as being

more helpful to the extent he is viewed as helping clients to

154

Table 25

Multiple Regression Analyses of Revised COP Model Scales on the 1
Criterion of Client Ratings of Consultant Effectiveness, INEXPCLIs

 

 

 

. 2 Beta 3 Zero Order . . . R Square
Variable Weight Correlation E. Significance R quare Change
PROBRDF .310 .718 55.48 .000 .516 .516
ORGNEED .425 .665 12.85 .001 .614 .097
PSYSAFTY .147 .622 2.13 .150 .629 .016
WHYDIFF -.153 .289 1.13 .293 .638 .008
SLTNRELV .065 .414 .153 .698 .639 .001
DISCREP .042 .596 .040 .842 .639 .000
WHATDIFF .029 .597 .021 .885 .639 .000

 

2 Includes only those variables from both the EXPCLI and INEXPCLI

Revised COP Models; PRCRLTY, EXRLTY, and SLTNIMEV do not meet
criteria for entry into the equation.

3 Beta weights are the regression coefficients of variables entered
on the last step of the analyses.

155

Table 26

Analysis of Combined Scales, INEXPCLIs
Reliability Estimates of Combined Scales

Scale Revised

Scale 1

PSYSAFTY .90
PRCVRLTY
ORGNEED

Scale 2

PROBRDF .94
DISCREP
EXRLTY

Scale 3

WHYDIFF .78
WHATDIFF

Scale 4

SLTNRELV .81
SLTNIMEV

‘—-—--‘--‘_—--~n—--‘-~-~----—-’“-----_—---———”_*‘--------’---—---_---‘

 

§Eals
1 (9o)
2 84 (94)
3 75 78 (78)
4 67 61 83 (81)
Beta Regression Analysis 2 2
§pglg Weight E. Significange_ B_. B__Qfléflfis
1 .18 62.58 .000 .546 .546
2 .19 5.75 .020 .592 .046
4 .10 .86 .360 .599 .007
3 -.10 1.21 .277 .609 .010
1 N = 54

Coefficient alphas in the diagonal; correlations are corrected
for attenuation.

156

Table 27

Correlations Between Revised COP Model Variables
and the Criterion, INEXPCLIs

 

 

 

Zero Order Correlation After Removal
Variable Correlation of the "General Factor"
PSYSAFTY .68 .16
PRCVRLTY .71 .17
ORGNEED .79 .26
PROBRDF . 79 .1 9
DISCREP .65 .04
EXRLTY .67 .16
WHYDIFF‘ .32 --35
WHATDIFF .65 -.08
SLTNRELV .45 -.O9
SLTNIMEV .53 -.ll

 

2 Corrected for attenuation in the criterion only.

3 Correlations between the criterion and listed variables after
partialling of the "general factor." A correction for attenuation
has been made only in the criterion variables, not in the predictor,
using the revised reliability estimates after partialling.

 

157

redefine problems and as creating a safe atmosphere to discuss
perceptions. When he tries to confront clients, or tries to deal
more specifically with solutions to problems, he is less likely
to be seen as helpful. This conclusion must be interpreted in
light of the recollection that this study deals with perceived
measures of consultant effectiveness, and that clients may actually
be helped by confrontation or other variables, even though they
do not experience or evaluate such efforts positively.
Hypothesis 35

Finally, the regression equations are re-computed using as
covariates the consultation and demographic variables discussed
earlier in this chapter. Tables 28 and 29 present these results
for the EXPCLIs and INEXPCLIs respectively. In neither case is
the overall F for the addition of these variables as a group signi—
ficant. The potential effect is noted, however, for the number
of sessions held in the last year (for EXPCLIs), suggesting that
the frequency of recent contact may affect perceived consultant
effectiveness. Variables showing similar effects for the INEXPCLI
sample include the number of sessions held in the last year, the
organizational level, and the organizational position of the
respondent. These findings, cautiously interpreted, may suggest
that for INEXPCLIs, people higher in the organization and who have
had more recent consultative experience may do better with the

consultant.

158

Table 28

Multiple Regression Analysis Using Selected Variables
as Covariates, EXPCLIs

1

 

 

 

2

 

 

2 Beta F to 2 R Overall
Covariates Wp§L Enter Signif. R__ Changg, F Signif.
Years Employed .060 1.139 .291 .035 .035 1.501 .198
Type Consultation -.332 .168 .684 .036 .001
Sessions/Last Yr. .402 4.469 .040 .147 .111
Orgt'l. Level .149 .382 .540 .151 .004
Total # Sessions -.480 .036 .851 .152 .001
Orgt'l. Position -.046 .201 .656 .155 .003
Variables3
PROBRDF .259 48.544 .000 .580 .425 9.470 .000
PSYSAFTY .193 11.261 .002 .661 .081 11.465 .000
SLTNRELV .226 2.092 .155 .676 .015 10.661 .000
EXRLTY -.106 .732 .397 .681 .005 9.612 .000
DISCREP .111 .245 .623 .683 .002 8.614 .000
ORGNEED -.113 .185 .670 .684 .001 7.765 .000
WHATDIFF .080 .303 .585 .687 .003 7.074 .000
SLTNIMEV -.041 .139 .711 .688 .001 6.444 .000
PRCVRLTY .652 .091 .765 .688 .000 5.888 .000
1N=56
2

See text for explanation of covariates and response choices for each.

3 WHYDIFF does not meet minimal entry requirements for inclusion in the
equation.

159

Table 29

Multiple Regression Analysis Using Selected Variables

as Covariates, INEXPCLIs

1

 

 

 

 

 

3 DISCREP and EXRLTY do not meet minimal entry requirements for
inclusion in the equation.

2 Beta F to 2 R2 Overall
Covariates Wig; Enter Signif. R_ Change, F Signif.
Years Employed -.566 .733 .396 .002 .002 1.834 .113
Orgt'l. Position —2.922 4.583 .038 .007 .005
Type Consultation .175 .176 .677 .016 .009
Sessions/Last Yr. 1.261 3.394 .072 .104 .088
Total # Sessions -.480 .014 .908 .104 .000
Orgt'l. Level 3.150 4.950 .031 .190 .086
Variables
PROBRDF .360 48.674 .000 .606 .416 10.121 .000
ORGNEED .458 12.262 .001 .691 .085 12.556 .000
PSYSAFTY .119 1.777 .189 .703 .012 11.551 .000
SLTNIMEV —.103 1.294 .261 .711 .008 10.595 .000
PRCVRLTY .111 .490 .488 .715 .004 9.562 .000
WHYDIFF -.063 .167 .685 .716 .001 8.605 .000
SLTNRELV .042 .080 .778 .716 .000 7.771 .000
WHATDIFF .016 .901 .716 .716 .000 7.040 .000
1 N = 54
2 See text for explanation of covariates and response choices for each.

160

Hypgtheses 4, 5, and 6

 

The next three hypotheses examine the "process" variables
explored in this study. Table 8 has already listed the number
of scale items for each process variable and the a priori and
revised estimates of reliability (coefficient alphas). Generally,
these are adequate, except for COG whose coefficient alpha is
disappointingly low (.39). Inter-scale correlations are presented
in Table 30 for the EXPCLI sample. After correction for attenuation,
the inter-scale correlations are moderate to high (.48 - .94),
suggesting the Presence again of "halo" in the clients' perceptions.
The same partialling procedure was attempted with the process
variables as with the COP Model variables. This is also presented
in Table 30. After partialling a general factor composed of the
sum of all of the process items, two of the scales show negative
alphas indicating that they do not hold up after removing the
general factor. The remaining scales show moderate to high
measures of internal consistency.

Although no specific hypotheses were generated about the
nature of the relationships among the process variables, the
obtained data for the EXPCLI sample are essentially uninterpret-
able, since most of the scales show negative correlations with
one another. The one interesting finding is the correlation of
.85 between VAL and COMPT after partialling, suggesting that the
consultant's ability to work with the value aspects of problems

is positively associated with clients' perceptions of his being

161

Table 30

Reliabilities and Interscale Correlations
on Process Variables, EXPCLIs1

 

Before PartiallingE

12212212
000 (39)
VAL 71 (77)
EMO 72 94 (87)
BEG 70 71 75 (61)
BEN 64 78 85 58 (87)

COMPT 45 86 76 48 71 (91)

After Partiallingg

Variable
00c (-07)3
VAL -08 (-12)3
EMO -49 00 (38)
BEG -11 -28 -44 (86)
BEN —39 ~44 —27 —59 (59)

COMPT -44 85 -17 -49 -15 (73)

 

1 N = 56

2 Correlations are corrected for attenuation using the reliability
estimates (coefficient alphas) in the diagonal.

3 Corrections for attenuation have been made using the reliability
estimates in the diagonal. In cases of negative alphas, the
number one was substituted for the negative value.

162

capable of helping them.

Similar analyses were next performed for the INEXPCLI sample.
These results are reported in Table 31. Essentially, these data
reveal the same patterns as the first sample: high measures of
internal consistency, but also high interscale correlations, in
many cases, higher than the coefficient alphas. Although COG, VAL,
and EMO are more strongly associated with one another (in both
the partialled and unpartialled matrices), especially for the
INEXPCLIs, these correlations are often less than the measure of
internal consistency. Similarly, BEN and COMPT, thought to have
been two aspects of trust, do not show consistent relationships
with one another as compared to their relationships with other
scales. In short, the construct validity for the process variables
receives little support from these data, contrary to the results
from the COP Model.

It will still be useful to examine the remaining two hypotheses
but primarily as suggestive information, given that the validity
of the scales is open to question.

Hypothesis 6 suggested that there would be differences between
EXPCLIs and INEXPCLIs in their perceptions of the consultant
regarding the process variables. Table 32 presents the mean scores
and standard deviations of the scales for the EXPCLI and INEXPCLI
samples. A MANOVA performed on this data was statistically
significant. In general, we see the same pattern with this data

as the COP Model data: EXPCLI rate the consultant more highly on

163

Table 31

Reliabilities and Interscale Correlations
on Process Variables, INEXPCLIs1

 

Before Partialling?

 

Variable
COG (68)
VAL 83 (79)
END 89 93 (84)
BEG 53 70 62 (69)
BEN 67 82 85 73 (88)
COMPT 65 71 67 58 91 (88)
After Partialling3
COG (-10)3
VAL 00 (46)
EMO 19 23 (49)
BEG -42 -31 —70 (45)
BEN -71 —88 -55 -36 (49)

COMPT -40 -70 -96 -64 48 (43)

 

2 Correlations are corrected for attenuation using the reliability
estimates (coefficient alphas) in the diagonal.

3 Corrections for attenuation have been made using the revised
reliability estimates in the diagonal. In the case of the
negative alpha, the number one was substituted for the negative
value.

164

Table 32

Comparison of EXPCLI and INEXPCLI Samples on the Process Variables1

 

 

 

 

Variable EXPCLIs INEXPCLIs
_X_ Sti.Dev. _X; Sti.Dev.
COG 30.29 3.13 28.39 4.18
VAL 25.24 2.84 21.52 4.45
EMO 30.64 4.66 27.39 5.09
BEG 21.78 4.34 19.15 4.53
BEN 52.53 5.86 46.72 7.45
COMPT 32.69 3.71 30.69 5.14
Univariate and Multivariate ANOVAs2

Hypothesis Significance
Variable Mean Square Univariate F (p less than)
COG 98.574 7.263 .008
VAL 376.628 27.130 .000
EMO 287.357 12.098 .000
BEG 188.996 9.592 .002
BEN 918.210 20.475 .000
COMPT 109.616 5.478 .021

F Ratio for Multivariate Test of Equality of Mean Vectors =

4.28 (p less than .00013)

165

Table 32 (cont'd.)

Discriminant Function Analysis

 

 

Standardized Discrim. Change in Significance
Order of Entry Function Coefficients Rao's V (p less than)3
VAL 1.124 27.130 .000
COMPT -.631 5.486 .019
ENG -.417 1.529 .216
BEN .508 .968 .325
BEG .062 .093 .761
COG -.033 .026 .872

 

1 EXPCLI N = 56; INEXPCLI N = 54

2 Univariate Degrees of Freedom = (1, 107)

Multivariate Degrees of Freedom = (6, 100)

3 Chi-square of change in Rao's V.

166

the various scales than do INEXPCLIs, and show less diversity in
their perceptions. A Discriminant Function Analysis was performed
on the data following the MANOVA using the already-discussed Rao's
V criterion in an attempt to maximize the group differences.

Using the step-wise procedure, it is VAL which shows the most
discriminating power. Discrimination coefficients are largest,
however, for BEN and BEG. All of these analyses are complicated
by the inter-scale correlations and lack of construct validation
for the process scales. Perhaps the most that can be meaningfully
concluded is that EXPCLIs and INEXPCLIs do not perceive the
consultant in the same manner in terms of his utilization of process
variables.

Lastly (Hypothesis 7), intra-group differences on the criterion
measure are explored for the process variables. Although regression
analyses were performed for both EXPCLI and INEXPCLI groups, the
nature of the already discussed statistical complications makes
such data of limited value. The multicollinearity problem and the
absence of a meaningful basis, either theoretically or empirically,
on which to combine scales, suggests that the most interpretable
data are the simple correlations between scale variables and
the criterion measure.

Table 33 presents both the zero order correlation coefficients
between the criterion and the process variables for the two
sample groups, and the correlations after partialling of the

so—called general factor for the process variables. The strong

167

Table 33

Correlations Between Process Variables and Criterion1

 

 

 

 

EXPCLIs INEXPCLIs
gero Order Afier gero Order Afier
Variable xy partialling xy partiallipg
COG .64 —.16 .69 -.09
VAL .87 .22 .81 .09
END .77 -.11 .72 -.28
BEG .59 -.25 .53 ' -.46
BEN .69 —.23 .82 .07
COMPT .87 .63 .91 .81

 

1

EXPCLI Sample N = 56
INEXPCLI Sample N = 54

2 Correlation between predictor and criterion after partialling

of the "general process factor"; see text for details.

168

correlations between COMPT and the criterion are expected because
the two ask similar questions (see Appendix C). However, the
strong association between VAL and the criterion for both groups,
and its positive association after partialling, suggests that

these clients positively evaluate the consultant's ability to

deal with issues regarding values as they relate to their
particular concerns. These associations are stronger and apparently
more persistent than dealing with cognitive or affective aspects

of problems, beginning with where the client is, or perceived
benevolence of the consultant. Future research would benefit

from the deve10pment of better measures and more sophisticated
theorizing regarding the role of values in organizational consulta-

tion.

CHAPTER V
SUMMARY AND CONCLUSIONS

Of what significance has been the study of a single organi-
zational interventionist? What generalities can be drawn from
this investigation? In what directions should future research
regarding the practice of organizational consultation move?

The primary contributions of this research are its exposition
and operationalization of an integrated model of consultation, its
attempt to validate the model against client experience, and its
demonstration of the feasibility of a research methodology
for future studies of organizational intervention.

Specifically, this study has shown:

1.) The Clinical—Organizational Psychological (COP) Model of
organizational intervention, as revised by clients' empirically
measured perceptions, shows promising construct validity as an
integrated approach to consultation;

2.) The model's a priori dimensions generally receive vali-
dation from the study, although some of the variables do not.
Specifically, a Revised COP Model based on analyses of both
experienced clients (EXPCLIs) and inexperienced clients (INEXPCLIs)
suggested the construct validity of the following model dimensions:
PSYSAFTY, PRCVRLTY, ORGNEED, PROBRDF, DISCREP, EXRLTY, WHYDIFF,
WHATDIFF, SLTNRELV, and SLTNIMEV. Other variables in the original
model were supported for one client group or the other, but not
for both.

3.) The variables receiving and not receiving support from

169

170

the critical incident (CI) data were generally consistent with

the DOC findings summarized in #2. However, there were several
differences. Specifically, DISCREP, EXRLTY, WHATDIFF, and SLTNRELV
received less support from the CI data.

4.) Nonhelpful CIs in the EXPCLI sample were predominantly
given for the following variables: PRCVRLTY, LINKAGEN, THRYINPT,
SLTNRESRCS, SLTNIMEV, and the general 4.2 category (Identification,
expression, and testing of alternative solutions). Four of these
variables did not receive validation support from the DOC question-
naire scale measures of the variables. This finding leads to the
idea that these are dimensions of consultation that are qualitatively
different from the others. In other words, these variables were
sources of unhelpfulness as well as helpfulness, suggesting that
these variables must be cautiously, and selectively, employed by
the interventionist. The specific danger appears to be that these
variables can either be seen as jeopardizing the consultant's role
as a benevolent outsider (in the case of PRCVRLTY, LINKAGEN, and
SLTNRESRCS), or, in the case of THRYINPT, as alienating clients
by presenting overly abstract, or theoretical views, to which the
client cannot relate. (The latter finding may be specific to
the business samples used in this study and not generalizable
to other organizational consultation, e.g., with hospitals or
educational institutions, whose members might be more receptive
to theoretical input.)

5.) As methodological tools, the C13 and the structured (DOC)

171

questionnaire items, provide generally consistent, but qualitatively
different, types of data. The DOC scales allowed quantifiable percep-
tions in a form useful for further statistical procedures. In contrast
the CIs gave rich and detailed clinical data, which clarified and
extended the DOC results. As such, the two data sources complemented
each other well, and provided a more interpretable picture of the
client—consultant relationship than either would have individually.

6.) From the CI data, a number of subdimensions are suggested

for further research. These include:

PSYSAFTY:

1. Ability to communicate at the level appropriate to the
client;

2. Benevolence of interest in helping the client;

3. Projection of competence, i.e., ability to help the client.

PRCVRLTY:

1. Ability to listen effectively to the client;

2. Elicitation of opinions held by clients.

PROBRDF:

1. Use of a "Socratic" method which helps clients discover
new insights about problems, and for which they are able
to accept responsibility for discovering;

2. Ability to use the client's views of reality as a base
point for understanding, conceptualizing, and redefining
the problem(s).

GOALFOCS:

An ability to help clients focus on their history and

current identities as a means of identifying present

172

organizational and/or individual needs and future directions.

PROBOWN:

An often indirectly communicated refusal to accept responsi-

bility for client problems.

7.) INEXPCLIs view the consultant differently from EXPCLIs.
In this study, there was an across-the-board tendency for INEXPCLIs
to rate the consultant lower (on average) than EXPCLIs, and a
general tendency to exhibit greater variability in their ratings.
There is limited support for the idea that the COP Model operates
sequentially, beginning with earlier model phases and progressing
consecutively. However, it is also plausible that INEXPCLIs require
more than five sessions with the consultant to reach even the
earliest stages of the model.

8.) Consultation, as practiced by this interventionist, while
multifaceted and complex, appears to be reducible (in client percep-
tions) to two primary consultation variables: a facilitative,
information-eliciting role, and a problem conceptualization and
redefinition role. The other variables appear to be of less impor-
tance, or perhaps variations and elaborations of these basic tasks,
at least in terms of general factor structure, and in predicting a
criterion measure of consultant effectiveness.

9.) Use of perceived measures of consultation has advantages
and disadvantages. When given an open-ended format, clients can
be remarkably detailed in their reporting of anecdotal information.

On the other hand, the tendency to rate certain variables or

173

consultation tasks negatively (e.g., confrontation variables) may
simply reflect the clients' dissatisfaction with unpleasant inter—
ventions. Colloquially stated, "good medicine may taste bad." If
so, negative client evaluations may need to be evaluated in reference
to more objective, and perhaps longer term, criteria.

10.) Consultation, as practiced by this consultant, addresses
both individual and organizational issues, and the interface between
the two. Clients perceive the consultant as being interested both
in them as individuals and in the organizations of which they are
a part. Such a dual focus reflects the fact that the COP Model
is indeed clinical gpd organizational.

11.) Variables not operationalized by the study, but surfaced
by CIs from clients, included: direct provision of support to
clients, engendering of enthusiasm and "motivation" among organiza-
tional members, and persistence in.advocating ideas or client actions.

12.) Generally, content variables (i.e., the COP Model dimensions)
were viewed by clients as being sources of helpful consultant behavior;
unhelpful behavior was generally related to "process" issues (i.e.,
the manner in which the content variables were implemented).

13.) The "objective" measures of the process variables defined
in this study did not appear to adequately measure the variables as
originally conceived. Preliminary evidence, however, suggests the
importance of the consultant's ability to handle the yglpg dimensions
of problems, perhaps even more than the ability to work on cognitive

or affective levels.

174

These, then, are the major findings from this research. Even
though this has been a rather lengthy study and report, more striking
than what has been included in the research is what it has gypluded.
Because the study has dealt with perceived measures of consultation,
it has only parenthetically addressed the issue of the consultant's
actual tactics and strategies and other subtleties not necessarily
visible to even the most discerning clients. The complex issues of
ppygp (e.g., Alderfer, 1976), which are clearly relevant to how the
consultant intervenes, and the specific organizational personnel
he works with, are underaddressed by this research. It is precisely
because of the subtlety with which this consultant uses leverage
that it was not adequately depicted by the data gathered and reported
here. No critical incident captured (or perhaps could capture) the
choice of personnel with whom Frost intervenes within an organization,
the process by which he chooses to interact with one member of the
company rather than another, or the differential impact of varying
frequencies of contact with different organizations or personnel.

Nor did this data address the implicit communications inherent
in Frost's approach. Examples of these omissions include: the
projections of organizational personnel to a senior, experienced
consultant who is typically accompanied by protage'graduate students,
the effect of knowing that the consultant's fee was given by him
to the university where he is employed, or the impact of limited
availability, meaning that clients often have to wait for long periods
of time to obtain his services. From the psychotherapy and social

psychological literature, it can be conjectured that these are

175

important and potentially influential variables affecting the
outcome of consultation, but for whom and to what degree was

not studied here. Overall, then, the surprising feature of studying
only one consultant is not the inclusiveness of the venture, but the
complexity and omissions inherent in the task.

Nevertheless, future research can benefit from the tactics and
findings of this study. First, a methodology has been developed,
both through critical incidents and the DOC questionnaire, as revised,
to efficiently gather information about clients' perceptions of
consultants in organizational settings. Secondly, there is now some
empirical basis for further research regarding concrete variables
of consultation. Those variables showing support in the Revised
COP Model need now to be tested in different settings and with
different consultants to discover whether they show generality and
are discriminators between effective and ineffective consultation.
(It may also be true that effective consultation makes use of these
COP variables, even when practitioners conceptualize their activities
differently.)

It is also clear from this study that there is a strong need
for longitudinal research to investigate how the consultant-client
relationship changes over time. The present exploration demonstrated
definite differences between EXPCLIs and INEXPCLIs, but did not
adequately document the way these differences change over the long
run. Such a longitudinal approach might take numerous forms, but
one promising idea is the completion of an abbreviated form of
instrumentation by clients after each session with the consultant

(or after every few sessions). The process of change could then

176

be mapped with a greater degree of precision to determine how the
ability to use outside help evolves, and what critical variables
account for such change. Among those who are never able to make
effective use of the consultant, there may be certain aspects of
the relationship between client and consultant that appear early
on that may be diagnostic, or which might be attended to on a
preventive basis.

Another research area that is suggested by this study is the
development of more definitive designs to investigate the role of
individual difference variables as moderators of the relationship
between consultant and client. Some of the variables uncovered
in this research could be studied in a laboratory setting (e.g.,
PRCVRLTY and PROBRDF), though some are fundamentally organizational
in nature, and would therefore be more difficult to study from an
individual differences perspective.

It would also be desirable to know more about consultation
activities in the highly-rated "problem conceptualization and
redefinition" area. Clearly, the clients studied here viewed Frost's
work in this dimension as being helpful. The CIs that were coded
in this grouping do give a flavor of the specific activities performed
by the consultant in this area, but do not fully capture their
essence. To say that the approach is Socratic, or that it asks
critical questions pointing to answers which the client can then
discover himself, does not fully explain the nature of the process.
The variable appears to be situation-specific, in that it is

directed to the client's particular problems, but simultaneously

177

theoretical, in that it implicitly advocates certain perspectives

by the questions asked. It is also impactful, in that it is perceived
as being very potent by clients, but at the same time keeps its
distance, enabling clients to take credit for insights obtained.

Such an approach is quite different from a programmed or "packaged"
intervention, which admittedly would be much easier to describe

and evaluate. Nevertheless, the value of this consultation dimension
suggests the need to study its nature in more specific detail,

perhaps using tape recordings of actual client—consultant inter-
actions.

There are also implications from this research for the profession-
al training of OD conSultants. The training of future consultants
would appear to be a more difficult task even than the analysis of
the consultative behavior of practitioners, particularly since the
types of things that clients found helpful were varied and complex.
Two tasks would appear to be fundamental, abstracting from the present
research. First, is the training of students in the ability to
undertake and develOp a relationship with clients in organizational
settings, that is at once both psychologically safe, and eliciting
of clients' views. In this regard, perhaps the best work yet done
about the nature of training of OD professionals, is the little-known
work of Sofer's (1961), in which he outlines in explicit detail
a training prOgram for fledgling consultants that would include
training in psychotherapy and the acceptance of responsibility for

establishing a "relationship" with clients that is inherent in

178

therapeutic work.Ifit is difficult for neophyte clinicians to become
proficient in the therapeutic enterprise (and it is), how much

more difficult it is for a consultant to learn to take responsibility
for the therapy of an organization, with its component individual,
group, and system-wide phenomena. Considering that many OD
organizations of professional consultants often consider acceptable
training the completion of an advanced degree in the "behavioral
sciences" or even in business administration, the present research
would suggest that such prerequisites are much too limited.

The second task that was viewed as especially helpful by
clients was the group of variables in the PROBRDF categories. This
is in many respects more of a cOgnitive skill, but at the same time
a clinical one. It is apparently not enough in dealing with business
clients to simply present new models or "solutions." Rather, the
client may be more significantly helped by a consultant who can
integrate theories with the practical realities of the clients'
existence, who can identify and help clients discover significant
organizational problems, and re-think them to discover underlying
issues. Indeed, one of the persistent implicit themes of this
research is the value of being able to enter another's reality and
to frame things in the client's own terms. Again, such skills are
not easily imparted to novice practitioners, but the need for an
intensive, rigorous training prOgram that incorporates these per-
spectives has apparently not yet been addressed by graduate training
institutions.

Overall, then, the future research and training needs in this

179

area of organizational consultation are large and underaddressed.
The present study has hopefully pointed the way to some directions
that will be incorporated in future studies. But it is clear that
clients in organizational settings, usually businesspeOple, can
benefit from an integrated perspective that is both clinical and
organizational in nature. Such interventions will also lead to
better insights about the nature of organizational psychology, of
how change takes place, and of the individual—organizational inter-
face. The present research may be viewed as a small step in a very

large amount of work left to be done.

APPEND IX A

1.

180

APPENDIX A
Critical Incident Question Format

Think of a particular experience you personally have had with
Dr. Frost that was ESPECIALLY HELPFUL to you. Keep the specific
example firmly in mind.

A. About how long ago did this situation occur?

B. Please describe the experience: WHAT happened, and what
was it Dr. Frost did that was especially helpful to you?

Now, think of a particular experience you personally had with
Dr. Frost that was NOT particularly helpful or useful. Keep
the specific example firmly in mind.

A. About how long ago did this situation occur?

B. Describe the experience: WHAT happened, and what was it
Dr. Frost did that was not especially helpful to you?

Please think of a specific situation in which you feel Dr.
Frost was ESPECIALLY HELPFUL to your COMPANY (regardless of
whether or not you personally were involved). Keep the
particular example firmly in mind.

A. About how long ago did this situation occur?

B. Describe the experience: What happened, and what was it
Dr. Frost did that was especially helpful to your company?

Please think of a specific situation in which you feel Dr.
Frost was NOT especially helpful to your COMPANY (regardless
of whether or not you were personally involved). Fix the
particular example firmly in mind.

A. About how long ago did this situation occur?
B. Describe the experience: What happened, and what was it

Dr. Frost did that was not especially helpful to your
company? - -

APPENDIX B

181
APPENDIX B

Cover and Followup Letters

July 11, 1977

Dear

The Scanlon Plan Associates and Michigan State University are
conducting a research study investigating the role of the external
consultant in Scanlon Plan organizations. This research is
interested in finding out what services clients find useful in

a consultant, and how different peOple make use of a consultant.
Specifically, the study is focusing on the relationship of various
Scanlon company employees to one consultant, Dr. Carl Frost,

who has worked with your own organization for some time.

People like yourself who have worked with.Dr. Frost (for different
periods of time) are being asked to complete the attached question-
naire. It deals with the types of activities you see the consultant
performing, how helpful these activities are to you, and some of
your perceptions of your own relationship with the consultant.

The Scanlon Plan Associates is presently exploring several
possibilities in the "new member outreach" and "internal member
development" programs for which the present study can provide
useful information. The question predictably arises: is the
consultant really necessary for installing and maintaining a
Scanlon Plan?

We encourage your participation in this project. Only a limited
number of peOple are being asked to complete this form. It is
therefore especially important that each person included in the
study return his or her completed questionnaire. It takes about
one hour to complete. Your replies are CONFIDENTIAL. Although
Dr. Frost has approved this project, your replies will not be
seen by him or by anypperson inpyour organization. After all the
data has been received and analyzed, you will be provided feed-
back of the results of this study. Any such feedback, however,
will combine the responses of many different people, and your own
responses will in no way be individually identified.

182

July 11, 1977
Page —2-

This questionnaire is being distributed to you by Michigan State
University (MSU) researchers. The principal investigator is Mr.

Rod Lowman. Any questions you have about the form can be directed
to him at MSU, telephone (517) 353-6422. Rod, or another university
associate, will pick up your completed form on .

The Scanlon Plan Associates, and Michigan State University sincerely
appreciate your participation in this study, part of the continuing
interest in learning how Scanlon Plan organizations can operate

more effectively. We thank you in advance for your contribution.

Sincerely,

Richard H . Ruch
President, Scanlon Plan Associates

Enc.

183

N1 ICHIGAN STATE UNIVERSITY

 

DEPAR'I‘MFNT OF PSYCHOLOGY EAST LANSING ' MICHIGAN ° 48824
(“.05 HAIL

August 1, 1977

Dear

Last month, we mailed you a questionnaire for a research study
involving the role of the external consultant in Scanlon Plan
organizations. You were asked to complete a form regarding your
own experiences with Dr. Carl Frost, who has had a long-term
consulting relationship with your company.

To date, we have not yet received your completed questionnaire,
which is very much needed to complete the study. We recognize

that you are a busy person and have many commitments. However,

we are dealing with a very small sample in this study and the
return of each participant's questionnaire is of great impor-
tance. We are also working under some very tight time restrictions
for the completion of this project.

Would you be kind enough to please send in your completed form
to us as soon as possible? (An addressed, stamped envelope was
included in the packet originally sent to you.) It may be,
because we did not speak with you personally at the time the
form was distributed, that you have questions about the project.
If so, please call me at (517) 353-6422 and we can discuss them.

Thank you for your involvement with this Scanlon Plan Associates-
sponsored study.

Sincerely,

Rodney L. Lowman
Research Assistant

P.S. If you have just sent the form in, please forgive this "reminder".

184

MICHIGAN STATE UNIVERSITY

 

I)l:l’.»‘\R'I'MI‘N’I~ OF PSYCHOLOGY EAST LANSING ° MICHIGAN ' 48824
()1 US HALI.

August 1, 1977

Dear

Some time ago we sent you a questionnaire as part of a research
study investigating your perceptions of your relationship with
Dr. Frost, who has done some consulting work with you and your
company. A followup letter was mailed to you several weeks after
the initial distribution of the form.

At this point, we are unclear about your intentions. We can
certainly still use your completed questionnaire, particularly
since this study is using a very small sample, and the return
of each respondent's completed form is very important. We

are asking, therefore, this last time, for you to kindly return
the questionnaire in the stamped, addressed envelope you were
originally sent.

If, however, you have decided not to participate in this study,
we would appreciate your returning the questionnaire in the
envelope provided. We would also appreciate, in confidence, a
statement from you indicating why you chose not to complete

the form. This will enable us to determine if there is any
systematic bias in the study among those few, who, like yourself,
have not yet returned the completed questionnaire.

We eagerly await your reply.

Cordially,

Rodney L . Lowman
Research Assistant

185

Dear

Michigan State University and the Scanlon Plan Associates are
conducting a research study investigating the role of the external
consultant in various organizations, both Scanlon Plan and non-
Scanlon companies. This research is interested in finding out
what services clients find useful in a consultant, and how
different peOple make use of a consultant. Specifically, the
study is focusing on the relationship of various organizational
personnel to one consultant, Dr. Carl Frost, who has done some
work with your own company.

We know that your own involvement with Dr. Frost has been limited,
perhaps only having seen him once or twice. Nevertheless, we

need to know how peOple use a consultant at the beginning of a
consulting relationship (we are asking others who have had more
experience to complete the form as well), and so we are asking

you to complete the attached questionnaire. It deals with the
types of activities you see the consultant performing, how

helpful these activities are to you, and some of your perceptions
of your own relationship with the consultant. If there are some
items you feel do not describe your relationship with the consult-
ant, note that you may use the middle or "uncertain" category, and
thus do not have to leave items blank.

The Scanlon Plan Associates is presently exploring several
possibilities in its "new member outreach" and "internal member
development" programs for which the present study can provide
useful information. The question predictably arises: is the
consultant really necessary for installing and maintaining a
Scanlon Plan?

We encourage your participation in this project. Only a limited
number of people are being asked to complete this form. It is
therefore especially important that each person included in the
study return his or her completed questionnaire. It takes about
one hour to complete. Your replies are CONFIDENTIAL. Although

186

Page -2—

Dr. Frost has approved this project, your replies will not be
seen by him or by any person in your organization. After all

the data has been received and analyzed, you will be provided
feedback of the results of this study. Any such feedback,
however, will combine the responses of many different people,

and your own responses will in no way be individually identified.

This questionnaire is being distributed to you by Michigan

State University (MSU) researchers. The principal investigator
is Mr. Rod Lowman. Any questions you have about the form can

be directed to him at MSU, telephone (517) 353—6422. When you
have completed the questionnaire, please return it in the stamped,
addressed envelope enclosed. Your prompt return of the form

will be most helpful.

The Scanlon Plan Associates and Michigan State University sincerely
appreciate your participation in this study, part of the continuing
interest in learning how industrial organizations can operate more
effectively. We thank you in advance for your contribution.

Sincerely,

Richard H. Ruch
President, Scanlon Plan Associates

Enc.

APPEND IX C

Appendix C

Dimensions of Consulting Scales1

1.1A PSYSAFTY

 

30.

37.
820
84.

*

103.

*
111.

119.

I can safely express disagreements with aspects of company
policy to him.

He will not misuse information I share with him.
Our relationship seems safe and comfortable.
It's safe to say almost anything to him.

I sometimes find it necessary to withold information from
him because I'm not sure how it will be used.

I feel ill at ease when I'm with him.

He is a person I can really trust.

1 .1B PRCVBL'ly

 

2.

4.

12.

66.
115.

He usually wants to know what I think about current organi-
zational issues.

When I work with him, he tries to get me to describe things
the way I now see them.

He helps me express my present thoughts, ideas, and feelings
to him.

He "draws me out" to say how I see things now.

He encourages me to express my opinions.

1.1C PROCSFAC

 

[22.

26.
46.

How someth ng gets said seems as important to him as what
gets said.

He gives me feedback about how I come across to others.

We look at how decisions are made, not just at the decisions
themselves.

 

1 Asterisked items were reverse scores. Items in brackets were ulti-
mately deleted from the scales on the basis of internal consistency
analyses.

187

188

49. He encourages me to observe how I come across to others.

29. He is often concerned about the impact my managerial style
has on other people in the organization.

1.1D LINKAGEN

 

1. He shares information with me from other departments in the
company.

24. Through him, I get new perspectives about things going on in
other parts of the company.

[:28. He brings together peOple from different parts of the
company to discuss problems.

85. He passes along the thoughts or feelings of other members of
the company to me.

*
130. I seldom learn much about what's going on in other parts of
the company when I talk with him.

1.2 PROBRDF

 

36. He helps me get to the central core of problems.

 

50. He sometimes shows me that the "problem" I thought was the
difficulty has larger dimensions.

53. He helps me see through the irrational and irrelevant parts
of problems and get at the central issues.

56. Working with him, I discover the broader dimensions of
problems.

78. Sometimes I learn from him that there is a larger problem
underlying the surface ones.

122. He helps me re—define problems and get at their basic roots.
134. He helps me put problems in a conceptual framework.

137. He often provides a new frame of reference from which to
view issues.

1.2A DISCREP

 

[8. He helps me see the inconsistencies in my ideas.]

20. He challenges me on things I might take for granted or assume
as "given."

920
97-
121 o

189

He makes me question my basic assumptions about things.
He helps me discover contradictions in my thinking.

He helps me see when I am being dishonest with myself.

1.23 ORGNEED

 

 

74. He reminds me of the needs of our organization.
*
81. Our focus is seldom on the needs of the organization as a
whole.
86. He seems to have definite views about what our company needs.
*-
90. In our discussions, he seldom encourages me to deal with
the needs of the company.
101. He helps me think about the needs of our company as a whole.
1.2C EXRLTY
33. Through him, I learn more about the competitive realities
with which our company must deal.
58. He shares information about how other organizations handle
similar problems.
61. He brings in material from outside the organization that
provides a new frame of reference.
69. He often helps me come to grips with difficult issues impinging
from outside the company.
73. He reminds me of the problems we must deal with in our

business environment.

1.2D THRYINPT

 

[15.
54.

59.

96.
113.

He often puts things into a theoretical context.]

Part of what he provides is information about theories of
why people behave as they do.

He has a strong theoretical bent that influences the issues
with which we deal.

He has definite views about why people behave as they do.

Sometimes in our sessions, he discusses theories about why
people act the way they do.

190

2.1 GOALFOCS

 

 

*
E 7. He seldom challenges our goals.)
39. He asks us to re-examine the goals to which we are committed.
75. He sometimes questions whether we will be able to reach our
present goals.
83. Sometimes it seems as if he thinks the goals we set for our
organization are not challenging enough.
98. Working with him, I sometimes learn that we are not acting
so as to reach our stated goals.
2.2 WHYDIFF
*-
5. He seldom directly deals with the need for change.
[10. When I suggest specific changes to him, he seems more conce ned
about the need for the changes than the changes themselves.
*-
72. Change is the most important issue to him, no matter how we
bring it about.
104. He insists that we have compelling reasons if we are going
to do things differently.
110. He asks me to consider why change is needed.
125. Having convincing reasons to change does not seem particularly

important to him.

2.3 WHATDIFF

89.

102.

He is concerned about what would have to happen if change were
to occur.

He insists that we spell out the details of changing before
we actually go through with anything new.]

He helps me determine what would have to happen to really
change.

Before implementing any change program, he helps me think
through what will be required for the change to take place
successfully.

He asks me to identify the specific steps needed to be taken
before a change program is implemented.

191

 

106. We sometimes look at what would have to happen if things
were to be done in a new way.
3:0 PROBOWN
42. I am challenged by him to own up to problems that are my
responsibility.
44. He helps people discover their own role in a problem.
91. He helps define whose problem is whose.
129. He insists that each person accept his/her share of responsi—
bility for a problem.
* .
135. He rarely seems concerned about establishing problem owner—

Ship 0

4.1 SLTNRESRCS

 

45.

47.

760

*114.

When it comes to generating solutions to problems, he helps
assure thatall relevant parties to a problem are included in
its solution.

If I am included in developing the solution to an organizational
problem, he is concerned about why I am part of the solution.

Who works on the solution to an organizational problem seems
to be an important issue to him.

He helps me see who ought to be included in developing
solutions to problems.

It seems less important to him that the right resources are
included in problem solutions than that the problem gets
solved.

4.2A SLTNRELV

 

11.

21.
*

51.

[99.

Just any solution to a given problem is not good enough for
him; it must be shown to be relevant to the problem.

He is concerned about solving the underlying, core problems,
not just in coming up with superficial solutions.

We have not worked much on the solutions to problems, only
on the problems themselves.

At times he seems to think our solutions to problems are too
shallow and don't get at the heart of the problem.]

117.

123.

192

To him, the proposed solution to a problem must be one that
deals directly with the problem rather than cover it up.

Coming up with the "right solution" to a problem is often
emphasized by him.

4.2B SLTNLINK

 

23.

60.

80.

108.

138.

To him, solutions to problems must do more than just solve
our own internal problems; they must also keep our various
"publics" satisfied.

He may know what makes the organization run, but he has little
understanding for the competitive realities which our solutions
to problems must satisfy.

When we solve problems, it seems important to him that our
solutions satisfy the competitive demands placed on the
organization.

He believes it is important to satisfy all the demands placed
on the organization-—outside as well as within the company.

He reminds me of the importance of satisfying all of our
company's investors (stockholders, vendors, customers, etc.)

4.3, SLTNIMEV

 

13.
18.

19.
40.
105.

112.

128.

COG

38.

He helps evaluate the solutions we put into practice.

Evaluating whether or not our solutions work is emphasized
by him-

He helps us put our solutions into actual practice.
He helps develop and implement solutions to problems.

He stresses followup evaluation and analysis of any solutions
we might develop to problems.

He helps develop and implement solutions to problems.

He strongly emphasizes the need to evaluate how effectively
our programs work.

He stresses intellectual understanding.

He can really handle ideas well.

65-

87c
*
133.

14.
[*70.

109.
120.
126.

pg
52.
67.
[*88.
107.
118.

193

He seems comfortable discussing things from an idea, or
abstract, point of view.

He can be helpful discussing thoughts and ideas.

He finds it difficult to deal with the theoretical aspects
of problems.

He really doesn't care what values people hold.

It's tough to know where he stands on questions involving moral
issues. ]

He is able to help me with problems concerning values or
standards.

If I have a problem that involves issues of what's "right
and wrong" I'd feel free to talk with him about it.

He can be helpful discussing issues about what's right and
wrong.

It's OK to express emotions when you're with him.

Often, he does not seem to understand my feelings.

He understands my words, but does not know how I fggl.]

He is not afraid to confront and deal with emotional issues.

If I have a problem that involves my feelings, I would feel
comfortable discussing it with him.

He seems to get a little frightened when the emotional issues
come up 0

We typically start with his agenda, even though I may have
things I want to discuss.

It seems aj if he tries to get me to accept his ideas and
standards.

In our sessions together, we usually begin with where I_
am, rather than with the things he wants to cover.

77-

*136.

BEN

*16.
55-

'X’
71.
93-
95-

127 o
139.
COMPT

*
310

57-
100.

116.

194

The issues with which I am concerned take precedence over
things he might want to discuss.

He usually has his own agenda in our meetings.

I feel he is genuinely interested in helping me.
He is indifferent to me.

I never have the slightest doubt about his interest in help-
ing me.

When he sees me, he seems to be "just doing a job."
He cares about me.

He is curious about our company, but does not really care
about what happens to it.

I feel he is genuinely interested in helping my company.

He cares about our organization.

I question his ability to help our company.

I trust his ability to work with our organization.
I think he's pretty competent as a consultant.

I have confidence in his consulting skills.

He is certainly able to help our organization.

APPENDIX D

APPENDIX D

Glossary of Abbreviations Used in this Study

COP: Clinical—Organizational Psychological

OD: Organization ' Development

DOC: Dimensions of Consulting

EXPCLI: Experienced client

INEXPCLI: Inexperienced client

PSYSAFTY: Psychological safety

PRCVRLTY: Perceived situational reality

PROCSFAC: Process facilitation

LINKAGEN: Linkage agent

DISCREP: Discrepancies in perceptions and assumptions
ORGNEED: Organizational needs

EXRLTY: External reality

THRYINPT: Theory input

GOALFOCS: Goal identification and evaluation

WHYDIFF: Identification of reasons for changing

WHATDIFF: Identification of the implications of change
PROBOWN: Problem ownership

SLTNRESRCS: Resources for solutions

SLTNRELV: Relevance of the solution for the actual problem
SLTNLINK: Relevance of the solution to external reality demands

SLTNIMEV: Implementation and evaluation of solutions

195

LIST OF REFERENCES

LIST OF REFERENCES

Adler, G. and Myerson, P. G. Confrontation lp_p§ychotherapy. New York:
Science House, 1973.

 

 

Albee, G. The protestant ethic, sex, and psychotherapy. American
Psychologist, 1977, 3g (2), l50-6l.

Alderfer, C. P. Change process in organizations. In M. D. Dunnette
(Ed.), Handbook pf industrial and organizatioqgl psychology.
Chicago: Rand McNally, 1976.

Alderfer, C. P. and Brown, L. D. Learning from changing. Beverly
Hills: Sage Publications, 1975.

Argyris, C. Explorations in consultant/client relationship. Human
Organization, 20, 1961, 121-33.

Argyris, C. Interpersonal competence and organizational effective—
ness. Homewood, Illinois: Irwin, 1962.

Argyris, C. Explorations in inter-personal competence. Journal pf
Applied Behavioral Science, 1965, l, 58—83.

Argyris, C. Conditions for competence acquisition and therapy. Journal
p§_Applied Behavioral Science, 1968, 4, 147-77.

Argyris, C. Intervention theory and method. Reading, Massachusetts:
Addison—Wesley, 1970.

Argyris, C. Management.§£é Organizational developmg_t. The pgph_fpom
§A_pp_YB, New York: McGraw-Hill, 1971.

Argyris, C. Increaslng leadership effectiveness. New York: John
Wiley & Son, 1976(a).

Argyris, C. Problems and new directions for industrial psychology.
In M. D. Dunnette (Ed.), Handbook pf_industrial and organizational
(psychology. Chicago: Rand McNally, 1976(b).

Argyris, C. Theories of action that inhibit individual learning.
American Psychologist, 1976(c), 3l, 638-654.

Argyris, C. & Schon, D. A. Theory lg practice: Increasing professional
effectiveness. San Francisco: JosseyéBass, 1974.

 

Baritz, L. The servants pf power. Middletown, Conn.: Wesleyan Univ.,

1960.

 

196

197

Bartlett, A. C., and Kayser, T. A. Toward a theory of changing
behavior: An elaboration on the role of influence and coercion.
In A. C. Bartlett and T. A. Kayser (Eds.), Changing organiza-
tional behavior. Englewood Cliffs: Prentice—Hall, 1973.

 

Beckhard, R. The confrontation meeting. Harvard Business Review,

1967. 32. 149-55-

 

Beckhard, R. Organization deve10pment: Strategies and models.
Reading, Mass.: Addison-Wesley, 1969.

Beer, M. On gaining influence and power for OD. Journal pf Applied
Behavioral Science, l976(a), l2, 44—51.

Beer, M. The technology of organization development. In M. D. Dunnette,
Handbook pf industrial and organizational psychology. Chicago:
Rand McNally, l976(b).

 

Benne, K. D. and Birnbaum, M. Principles of changing. In W. G.
Bennis, K. D. Benne, and R. Chin, The planning pf change (2nd
ed.). New York: Holt, Rinehart, & Winston, 1969.

 

Bennis, W. G. Theory and method in applying behavioral science to
planned organizational change. Journal pf Applied Behavioral

Science, 1965, l (4), 337-360.

 

Bennis, W. G. Changing organizations. New York: McGraw—Hill, 1966.

 

Bennis, W. G. Organization development: Its nature, origin, and
prospects. Reading, Mass.: Addison-Wesley, 1969.

 

Bergin, A. E. and Strupp, H. H. Changing frontiers lg the science
pf_p§ychotherapy. Chicago: Aldine, 1972.

 

 

Blake, R. R. and Mouton, J. S. The manggerial grid. Houston: Gulf,
1964.

 

Blake, R. R., Mouton, J. S., Barnes, L. B., & Greiner, L. E. Break-
through in organization development. Harvard Business Review,

512. (6). 1964. 133-155.

 

Bowers, D. G. OD techniques and their results in 23 organizations:
The Michigan ICL study. Journal pf_Applied Behavioral Science,

1973. 9 (1). 21—43.

Bowers, D. G., Franklin, J. L., Pecorella, P. A. Matching problems,
precursors, and interventions in OD: A systematic approach.
Journal Qf_Applied Behavioral Science, 1975, ll_(4), 391-409.

 

198

Branden, N. An informal discussion of biocentric therapy. In. B. N.
Ard Jr., Counseling and psychotherapy: Classics lg theories
and issues. Palo Alto: Science and Behavior Books, 1975.

 

Buchanan, P. C. Crucial issues in organizational deve10pment. In
G. Watson (Ed.), Change lg school systems. Washington, D. 0.:
National Training Labs, 1967.

Burke, W. W. Organization development in transition. Journal pf
Applied Behavioral Science, 1976, l2, 22—43o

Burke, W. W. and Schmidt, W. H. Primary target for change: The
manager or the organization? In H. A. Hornstein et al., Social
intervention: A_behavioral science approach. New York: Free
Press, 1971.

 

Cartwright, D. Achieving change in people: Some applications of
group dynamics theory. Human Relations, 1951, A, 381-393.

 

Chein, 1., Cook, S. W., & Harding, J. The field of action research.
American Psychologist, 1948, 3, 43-50.

Coch, L. and French, J. Overcoming resistance to change. Human
Relations, 1948, l, 512—32.

Cummings, L. L. Address, Department of Psychology, Michigan State
University, April 15, 1976.

Davis, S. A. An organic problem-solving method of organizational
change. Journal pf Applied Behavioral Science, 1967, 3 (1),
3'21 0

 

Drucker, P. F. Management Tasks reaponsibilities practices. New
York: Harper & Row, 1974.

 

Ferguson, C. K. Concerning the nature of human systems and the
consultant's role. Journal pf Applied Behavioral Science,

1968. it (2). 179-193.

 

Ferry, S. On labels, holism, and OD research. Journal pf Applied
Behavioral Science, 1973, 9 (1), 122-23.

 

Fiedler, F. E. A comparison of therapeutic relationships in psycho—
analytic, non-directive and Adlerian therapy. Journal pf
Consulting Psychology, 1950, l&, 436—445.

Fiedler, F. E. Factor analyses of psychoanalytic, non-directive, and
Adlerian therapeutic relationships. Journal pf Consulting
Psychology, 1951, l5, 32-38.

 

199

Flanagan, J. C. The critical incident technique. Psycholpgical
Bulletin, 1954, 5l, 327-355.

Franklin, J. L. Characteristics of successful and unsuccessful organ—
ization development. Journal pf Applied Behavioral Science,

1976. 12 (4). 471-492.

French, W. L. and Bell, C. H. Organization development. Englewood
Cliffs: Prentice—Hall, 1973.

 

Friedlander, F. A comparative study of consulting processes and
group development. Journal pf Applied Bahavioral Science,

1968: E: 377'399-

 

Friedlander, F. OD reaches adolescence: An exploration of its
underlying values. Journal pf_Applied Behavioral Science,
1976: 12. 7‘21-

Friedlander, F. & Brown, L. D. Organization development. In Annual
review pf_psychology, 25. Palo Alto, CA: Annual Reviews, 1974.

Frohman, M. A. AA empirical study pf a model and strategies for
planned organizational chang_. Unpublished Ph.D. dissertation,
University of Michigan, 1970.

 

 

Frost, C. F. The Scanlon Plan: Anyone for free enterprise?. MSU
Business Topics, 1978, 26, 25-33.

 

Frost, C. F., Wakeley, J. H., and Ruh, R. A. The Scanlon Plan for
organization development: Identity, participation, and equipy.
East Lansing: Michigan State University Press, 1974.

 

 

Gardner, J. W. Self renewal The individual and the innovative society.
New York: Harper & Row, 1964.

 

Gibb, J. R. The role of the consultant. Journal 9: Social Issues,

1959. 15 (2). 1—4.

 

Glasser, W. Reality therapy A new approach pp psychiatpy. New York:
Harper & Row, 1965.

 

Golembiewski, R. T., Billingsley, K., and Yeager, S. Measuring change
and persistence in human affairs: Types of change generated by
00 designs. Journal pf_Applied Behavioral Science, 1976, 12.(2).

133-157.

Greenson, R. R. The technique and ppactice 9f_p§ychoanaly§is.(Vol. 1)
New York: International Universities Press, 1967.

 

 

Greiner, L. E. Patterns of organization change. Harvard Business

Review, 1967, 32 (3), ll9-l30.

 

200

Hackman, J. R. and Lawler, E. E. Employee reactions to job character-
istics. Journal pf Applied Psychology Monograph, 1971, 55, 259—
286.

 

Hamner, E. F. Interpretive technique: A primer. In E. F. Hamner,
Use pf_ipteppretation ip treatment Technique and art. New
York: 1968, 31-42.

 

 

 

Hornstein, H. A., Bunker, B. B., Burke, W. W., Gindes, M. and Lewicki,
R. T. Social intervention: A behavioral science approach. New
York: Free Press, 1971.

 

 

Hornstein, H. A., Bunker, B. B., Hornstein, M. G. Some conceptual
issues in individual and group oriented strategies of intervention
into organizations. Journal pi_§pplied Behavioral Science, 2,

1971. 557-567.

 

Hunter, J. E., & Cohen, S. H. Package. Mimeograph, Computer Institute
for Social Science Research, 1971, Michigan State University.

Huse, E. F. Organizatipp development and change. St. Paul, MN: West
Publishing Co., 1975.

Kahn, R. L. Organizational development: Some problems and proposals.
Journal pi Applied Behavioral Science, 1974, _l_O_ (4), 485-502.

 

Karon, B. The psychoanalytic treatment of schizophrenia. In P. Magaro
(Ed.), The construction pi madness: Emergipg conceptions and
intervention§_into the psychotic ppocess. New York: Pergamon,

1976.

 

Katz, D., & Kahn, R. L. The social psychology 2: organizations (2nd
Ed.). New York: John Wiley, 1978.

 

 

Klecka, W. R. In N. H. Nie et al., Statistical package for the
social sciences (2nd Ed.). New York: McGraw Hill, 1975,

u3u—uo7.

 

Kolb, D. A. and Boyatzis, R. E. On the dynamics of the helping relation—
ship. In D. A. Kolb, D. M. Rubin, and J. M. McIntyre, Organiza-
tional psychology 5 book pi; readings (2nd Ed.). Englewood
Cliffs: Prentice Hall, 1971.

Kolb, D. and Frohman, A. An organizational development approach to
consulting. Sloan Management Review, 1970, la, 51-65.

 

 

Korman, A. K. "Consideration," "initiating structure," and organiza—
tional criteria—-A review. Personnel Psychology, 1966, l9,
349-361-'

Latane, E., Eckman, J. & Joy, V. Shared stress and interpersonal attract-
ion. Journal 9: Personalipy and Social Psychology, 1966, 1, 80-94.

 

201

Lawrence, P. R., and Lorsch, J. W. Organization and environment.
Boston: Division of Research, Harvard Business School, 1967.

 

Lawrence, P. R., and Lorsch, J. W. Develppipg organizations: Diagnosis
and action. Reading, MA: Addison—Wesley, 1969.

 

 

Levinson, H. Organizational diagnosis. Cambridge: Harvard Univ.
Press, 1972.

Lewin, K. (Ed.), Field theory_ip_social scienga. New York: Harper &
Brothers, 1951.

 

 

Lippitt, G. Consulting with a national organization: A case study.
Journal 9: Social Issues, 1959, li.(2). 20-27.

 

Lippitt, G. Organization renewal. Englewood Cliffs: Prentice Hall,
1969.

 

Lippitt, R. Dimensions of the consultant's job. Journal p: Social
Issues, 1959, l5 (2), 5—12.

Lippitt, R., Watson, J., & Westley, B. The dynamics p£_planned chang_.
New York: Harcourt, Brace & World, 1958.

 

 

Lowin, A. Participative decision making: A model, literature critique,
and prescriptions for research. Organizational Behavior and Human
Performance, 1968, 3, 68—106.

 

Lowman, R. L. Collaboration in work settings: Significant omissions.
Journal pf_§pplied Behavioral Science, 1978, l3, 127—128.

 

Maier, N. R. F. Problem solving discussions and conferences. New York:
McGraw Hill, 1963.

Mann, F. C. Studying and creating change: A means to understanding
social organization. In Research ip industrial human relations.
Industrial Relations Research Assoc., 1957, Pub. No. 17, 146—167.

 

Maslow, A. H. A philosophy of psychology: The need for a mature
science of human nature. Main Currents ip Modern Thought, 1957,

13, 127—32.

 

McGregor, D. The human side pf enterprise. New York: McGraw Hill,

1960.

 

 

Meehl, P. Psychodiagnosi§_Selected papers. Minneapolis: Univ. of
Minnesota Press, 1973.

Menninger, I. Theory p£_psychoanalytic techniqpe. New York: Basic
Books, 1958.

 

202

Miller, G. A. PsycholOgy as a means of promoting human welfare.
American Psychologist, 1969, g3, 1063-1075.

Paul, G., and Bernstein, D. Anxiety and clinical problems: Systematic
desensitization and related techniques. New York: General Learn-
ing Press, 1973.

 

Paul, I. H. Letters pp Simon pp the conduct 9: psychotherapy. New
York: International Univ. Press, 1973.

 

 

Polansky, N., and Kounin, J. Clients reactions to initial interviews:
A field study. Human Relati0pa, 1956, 9, 237-265.

 

Polanyi, M. Personal knowledge: Towards a_post—critical philosophy.
Chicago: University of Chicago Press, 1958.

 

Revised Ethical Standards of Psychologists. In APA Monitor, 1977,
§ (3). 22-23. -

Rogers, C. The necessary and sufficient conditions of therapeutic
personality change. Journal p£_Consulting Psychology, g1: 1957,
95—103.

 

Rogers, C. 9p becoming a person. New York: Houghton Mifflin, 1961.

Sarason, S. The creation pi_setting§ and the future societies. San
Francisco: Jossey-Bass, 1972.

 

Sarason, S. The paychological sense pf community: Prospacts for a
community psychology. San Francisco: Jossey-Bass, 1974.

 

Sarason, S. Work, a 'n , and social changa: Professionals and the
one life-one career imperative. New York: Free Press, 1977.

 

Sarason, S., Sarason, E. K., and Cowden, P. Aging and the nature of
work. American Paycholpgist, 1975, 39, 584-592.

 

Sashkin, M., Morris, W. C., and Horst, L. A comparison of social and
organizational change models. Information flow and data use
processes. Paychology Review, 1973, pg (6), 510-526.

Schein, E. H. Management development as a process of influence.
Industrial Management Review, 1961, a (2).

 

Schein, E. H. The mechanism of change. In W. G. Bennis, K. D.
Benne, and R. Chin, The plannipg of change (2nd Ed.), New York:
Holt, Rinehart and Winston, 1969(37.

Schein, E. H.‘ Process consultation: Its role ip organization
development. Reading, MA: Addison-Wesley, 1969(b).

 

 

 

203

Shakow, D. What is clinical psychology. American PsychOIOgast, 1976,
11.-9 553-560 -

 

Sherif, M. and Sherif, C. W. Groups ip harmony and tension. New York:
Harper & Row, 1953.

 

Singer, E. Key concepts ip psychotherapy (2nd Ed.). New York: Basic
Books, 1970.

 

Smith, P. C. and Kendall, L. M. Retranslation of expectations: An
approach to the construction of unambiguous anchors for rating
scales. Journal pf_§pplied Psychology, 1963, 32, 149-155.

Sofer, C. The organization from within. Chicago: Quadrangle Books,
1961.

 

Sommer, R. No, not research. I said evaluation! APA Monitor, §_(4),

1977. 1-

Steele, F. Consultants and detectives. Journal p: Applied Behavioral
Science, 1969, 5_(2), 187-202.

Steele, F. Consultipg_for organizational chang_. Amherst: Univ. of
Massachusetts Press, 1975.

Strauss, G. Organization development. In R. Dubin, Handbook pf
work organization and socie_y. Chicago: Rand McNally, 1976,
617'6850

Strupp, H. H. Clinical psychology, irrationalism, and the erosion
of excellence. American Paychologist, 1976, 31 (8), 561-571(a).

 

Strupp, H. H. Themes in psychotherapy research. In J. L. Claghorn
(Ed.), Successful psyphotherapy. New York: Brunner—Mazel,

l976(b).

Tannenbaum, R. Organizational change has to come through individual
change. Innovations, 1971, pg, 36-43.

Tichy, N. M. Agents of planned social change: Congruence of values,
cognitions and actions. Administrative Science Quarterly, 1974,
19 (2), 164-182.

Torbert, W. Some questions on Bowers' study of different OD tech-
niques. Journal pf Applied Behavioral Science, 1973, 9, 668—671.

 

Walton, R. E. Two strategies of social change and their dilemmas.
Journal pf Applied Behavioral Science, 1965, 1, 167—179.

Walton, R. E., and Warwick, D. P. The ethics of organization develop—
ment. Journal pf Applied Behavioral Science, 1973, 9 (6), 681-698.

 

204

Watzlawick, P., Weakland, T. H., and Fisch, R. Change: Principles
9: ppoblem formation and problam resolution. New York: Norton,

1974.

 

Wedge, B. A psychiatric model of intercession in intergroup conflict.
Journal p: Applied Behavioral Science, 1971, 2, 733-761.

 

Yalom, I. D. The theory and ppactice pf gpoup psychotherapy (2nd Ed.).
New York: Basic Books, 1975.

 

Zubin, J. and Spring, B. Vulnerability-—A new view of schizophrenia.
Journal g Abnormal Psychology, 1977, pg (2), 103-126.