INFLUENCE, EMPHASIS, REFERENCE
GROUP ORIENTATION AND
COMMUNICATION WITHIN THE
AUTONOMOUS UNIVERSITY DEPARTMENT

Thesis for the Degree of Ph. D.
MICHIGAN STATE UNIVERSITY
DONALD JAMES REICHARD
1970

w. .-

J LIBRARY
Michigan 5mm

University

     
     

Tm WWWW

3 1293 10463 2256

This is to certify that the

thesis entitled

INFLUENCE, EMPHASIS, REFERENCE GROUP
ORIENTATION AND COMMUNICATION
WITHIN THE AUTONOMOUS
UNIVERSITY DEPARTMENT

presented by

DONALD JAMES REICHARD

has been accepted towards fulfillment
of the requirements for

__P_Ll_,_fl._ degree in Jighgniducat ion

6241M

Major professor

Date

 

0-169

 

. " BINDING av I

 
 
 

  

IIIMB 8 WW
,BIJI'II’ BRIBERY IIIB.

I IAIIY HINDI“

 

 

 

 

 

 

 

 

 

 

 

 

 

 

a- . *W-vu" ‘

 

ABSTRACT

INFLUENCE, EMPHASIS, REFERENCE GROUP
ORIENTATION AND COMMUNICATION
WITHIN THE AUTONOMOUS
UNIVERSITY DEPARTMENT

BY
Donald James Reichard

Problem. The purpose of this study was to investigate the
manner and conditions under which departments and individ-
uals within departments characterized by high, medium, and
low degrees of perceived autonomy differed in regard to:
(1) characteristics such as Cartter report rating, the
percentage of faculty members in a department holding
joint appointments, and the percentage of faculty members
in a given department noting one or more problems classi-
fied by source as internal to or external to the depart-
ment; (2) the influence in departmental affairs of such
principle actors within the department as the department
chairman, the departmental faculty as a whole, and indi-
vidual faculty members; (3) the emphasis actually placed
upon such primary departmental goals as undergraduate in-

struction, graduate instruction, and basic research; (4)

Donald James Reichard

the reference group orientation of departmental faculty to
the university, the department, or the discipline; and (5)
the degree to which discussion of departmental affairs
took place between individual members of a department and
the department chairman, chairmen of special departmental
committees, departmental faculty and graduate students,
the dean, university administrators, and faculty or chair-
men of other departments or institutes.

Analysis. Departmental autonomy was defined as being in-
versely related to the combined perceived influence of the
dean and university administrators in departmental affairs.
Data were collected from 1,319 respondents from a total of
ninety-seven departments of English, history, chemistry,
mathematics, psychology, electrical engineering, and man-
agement in fifteen universities of diverse backgrounds
selected from a national sample. Departmental autonomy
was regarded as an independent variable. Employing the
process of elaboration, twenty-four test factors were
introduced as control variables in the chi square analysis
of the relationships between departmental autonomy and the
individual dependent variables. One-way and repeated
measures analyses of variance as well as Spearman rank
order correlations were also employed in the analysis of
data.

Findings. Four of the hypotheses were supported including

those which hypothesized (1) an inverse relationship

Donald James Reichard

between the degree of departmental autonomy and the influ-
ence of the department chairman as well as direct rela-
tionships between the degree of departmental autonomy and
(2) the influence of the departmental faculty as a whole,
(3) emphasis upon graduate instruction, and (4) emphasis
upon basic research. Although the remaining twelve hy-
pothesized overall relationships between the degree of
departmental autonomy and individual dependent variables
were not supported by the data, a number of conditions
were specified in which each of the hypothesized relation-
ships did appear to hold true.

In addition, the perception of the influence of
individual faculty members was found to be significantly
lower in departments of low rather than medium or high
autonomy. The combined amount of internal discussion of
departmental affairs with the departmental faculty as a
whole, the department chairman, chairmen of special de-
partmental committees, and graduate students was also
significantly lower in departments characterized by a low
rather than a medium or high degree of autonomy. All of
the significant relationships were modified to some extent,
however, when the twenty-four control variables were in-
troduced into the analysis.

An attempt was made to describe a number of char-
acteristics associated with a medium degree of departmental

autonomy which might be regarded as desirable outcomes for

Donald James Reichard

the university. Further research into the nature of prob-
lems associated with excessive degrees of departmental
autonomy was recommended which would incorporate financial
characteristics of the department into the definition of

departmental autonomy.

INFLUENCE, EMPHASIS, REFERENCE GROUP
ORIENTATION AND COMMUNICATION
WITHIN THE AUTONOMOUS

UNIVERSITY DEPARTMENT

BY

Donald James Reichard

A THESIS

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

DOCTOR OF PHILOSOPHY

Department of Administration and Higher Education

1970

ACKNOWLEDGMENTS

The writer is deeply indebted to a number of indi-
viduals who aided in the development of his thesis.
Special thanks go to his thesis director, Dr. Paul L.
Dressel, who has provided assistance and encouragement not
only at the dissertation stage but throughout the entire
doctoral program. Conversations with and suggestions of-
fered by Dr. Philip M. Marcus have been of great value
throughout the writer's association with the Esso Depart-
mental Study Project. The writer also gratefully acknowl—
edges the aid extended by Dr. Max Raines as a member of
the doctoral committee.

The many hours of assistance so willingly and ably
lent by Mrs. Marion M. Jennette in all phases of the study
are deeply appreciated. David Wright has been most helpful
in applying his expertise in the use of the computer to
the problems of data analysis. Last, but certainly not
least, the writer expresses his gratitude to his wife
Rosalind whose help, understanding, and encouragement in

so many ways have meant so very much.

ii

TABLE OF CONTENTS

ACKNOWLEDGMENTS O O O O O O O O O O O O O O 0

LIST OF TABLES O O O O O O O O O O O O O O 0

Chapter

I.

II.

III.

IV.

INTRODUCTION AND STATEMENT OF THE PROBLEM

Introduction . . . . . . . . . . . .
Need for Study . . . . . . . . . . .
Problem and Purpose . . . . . . . .

REVIEW OF RELATED LITERATURE . . . . .

Introduction . . . . . . . . . . . .
Summary . . . . . . . . . . . . . .

STATEMENT OF HYPOTHESES AND DEFINITION
TERMS AND PROCEDURES . . . . . . . . .

Statement of Hypotheses . . . . . .
The Sample . . . . . . . . . . . . .
Construction of Variables . . . . .
Analysis of Data . . . . . . . . . .

ANALYS I S OF DATA 0 O O O O C O O O O I

Introduction . . . . . . . . . . . .
Analysis . . . . . . . . . . . . .
Departmental Characteristics . . .
Influence of Various Actors . .
Emphasis Upon Departmental Goals .
Reference Group Orientations . . .
Communication . . . . . . . . . .
Summary . . . . . . . . . . . . . .

iii

Page

ii

[.4

CDU‘IH

l3

13
40

44

44
46
48
57

66

66
68
69
79
87
94
99
117

Chapter . Page

V. SUMMARY, DISCUSSION, IMPLICATIONS AND
RECOMMENDATIONS FOR FURTHER RESEARCH . . . . . 120

Summary of Findings . . . . . . . . . . . . 120
Discussion . . . . . . . . . . . . . . . . . 125
Implications . . . . . . . . . . . . . . . . 136
Recommendations for Further Research . . . . 138

APPENDIX C O O O O O O O O O O O O O O O C O O O O O 141

BIBLIOGMPHY O O I O O O O I O O O .0 O C O O O O O O 1 5 3

iv

Table

1.

LIST OF TABLES

Spearman rank order correlations between
departmental autonomy and various test
factors and dependent variables for
total sample and various subgroups . . .

Relationship between Cartter report ratings
and faculty perception of departmental
autonomy . . . . . . . . . . . . . . . .

Relationship between percentage of faculty
members holding joint appointments in
departments and faculty perception of
departmental autonomy . . . . . . . . . .

Relationship between percentage of faculty
members in departments noting one or more
problems classified by source as internal
to the department and faculty perception
of departmental autonomy . . . . . . . .

Relationship between percentage of faculty
members in departments noting one or more
problems classified by source as external
to the department and faculty perception
of departmental autonomy . . . . . . . .

Relationship between faculty perception of
the influence of the department chairman
and faculty perception of departmental
autonomy . . . . . . . . . . . . . . . .

Relationship between faculty perception of
the influence of the departmental faculty
as a whole and faculty perception of
departmental autonomy . . . . . . . . . .

Relationship between faculty perception of
personal influence and faculty perception
of departmental autonomy . . . . . . . .

Page

70

72

73

75

77

79

82

84

Table Page

9. Relationship between faculty perception of
emphasis placed upon undergraduate instruc-
tion and faculty perception of departmental
autonomy . . . . . . . . . . . . . . . . . . 88

10. Relationship between faculty perception of
emphasis placed upon graduate instruction
and faculty perception of departmental
autonomy . . . . . . . . . . . . . . . . . . 90

11. Relationship between faculty perception of
emphasis placed upon basic research and
faculty perception of departmental autonomy . 92

12. Relationship between reference group orien-
tations of faculty and faculty perception
of departmental autonomy . . . . . . . . . . 95

13. Relationship between individual and group
communication means and faculty perception
of departmental autonomy . . . . . . . . . . 101

14. Relationship between the amount of discussion
with the faculty in one's department and
faculty perception of departmental autonomy . 103

15. Relationship between the amount of discussion
with the department chairman and faculty
perception of departmental autonomy . . . . . 104

16. Relationship between the amount of discussion
with chairmen of special departmental com-
mittees and faculty perception of depart-
mental autonomy . . . . . . . . . . . . . . . 106

17. Relationship between the amount of discussion
with graduate students and faculty perception
of departmental autonomy . . . . . . . . . . 108

18. Relationship between the amount of discussion
with faculty outside of department and
faculty perception of departmental autonomy . 111

19. Relationship between the amount of discussion

with the dean and faculty perception of
departmental autonomy . . . . . . . . . . . . 113

vi

Table

20. Relationship between the amount of discussion
with chairmen of other departments or
institutes and faculty perception of
departmental autonomy . . . . . . . . . . .

21. Relationship between the amount of discussion
with university administrators and faculty
perception of departmental autonomy . . . .

Appendix

A-l. Relationship between percentage of faculty
members holding joint appointments in
departments and faculty perception of
departmental autonomy as modified by the
introduction of twenty-four test factors .

A-2. Relationship between faculty perception of
the influence of the department chairman
and faculty perception of departmental
autonomy as modified by the introduction
of twenty-four test factors . . . . . . . .

A-3. Relationship between faculty perception of
the influence of the departmental faculty
as a whole and faculty perception of
departmental autonomy as modified by the
introduction of twenty-four test factors .

A-4. Relationship between faculty perception of
personal influence and faculty perception
of departmental autonomy as modified by
the introduction of twenty-four test
factors . . . . . . . . . . . . . . . . . .

A-S. Relationship between faculty perception of
emphasis placed upon undergraduate in-
struction and faculty perception of
departmental autonomy as modified by the
introduction of twenty-four test factors .

A-6. Relationship between faculty perception of
emphasis placed upon graduate instruction
and faculty perception of departmental
autonomy as modified by the introduction of
twenty-four test factors . . . . . . . . .

vii

Page

115

116

141

142

143

144

145

146

Table Page

A-7. Relationship between faculty perception of
emphasis placed upon basic research and
faculty perception of departmental au-
tonomy as modified by the introduction of
twenty-four test factors . . . . . . . . . . 147

A-8. Relationship between reference.group orien-
tations of faculty and faculty perception
of departmental autonomy as modified by the
introduction of twenty-four test factors . . 148

A-9. Relationship between the amount of discussion
with the department chairman and faculty
perception of departmental autonomy as
modified by the introduction of twenty-
four test factors . . . . . . . . . . . . . . 149

A-lO. Relationship between the amount of discussion
with chairmen of special departmental com-
mittees and faculty perception of depart-
mental autonomy as modified by the
introduction of twenty-four test factors . . 150

A—ll. Relationship between the amount of discussion
outside of department and faculty perception
of departmental autonomy as modified by the
introduction of twenty-four test factors . . 151

A-12. Significant Spearman rank order correlation

coefficients between the degree of depart-
mental autonomy and additional variables . . 152

viii

CHAPTER I

INTRODUCTION AND STATEMENT OF THE PROBLEM

Introduction

 

The literature of higher education is replete with
widely varied organizational conceptions of the university.
Litchfield and Kerr have described the awesome size and
lack of unity inherent in the multiversity composed of a
federation of interests.1 On the other hand, Millett sees
the actions within the academic community as more conver-
gent than divergent.2 In a more humorous vein, Monson3
has likened the university to a dispensing machine, a zoo,
or Mammoth Cave while Moran has summarized some of the

existing conceptions of a university and urged that

 

1Edward H. Litchfield, "Organization in Large

American Universities: The Faculties," Journal of Hi her
Education, XXX (October, 1959), 353-364; Also Edward H.
LitchieId, "Organization in Large American Universities:.
The Administration," Journal of Higher Education, XXX
(December, 1959), 491-503; andICIark Kerr, THe Uses of the
University (Cambridge, Massachusetts: Harvard University
Press, 1964).

2John D. Millett, The Academic Community (New York:
McGraw-Hill Book Company, Inc., 19627.

 

 

 

 

3Charles H. Monson, Jr., "Metaphors for the Uni-
versity," Educational Record, LII (Winter, 1967), 22-29.

comprehensive investigation of the university be made in

order to determine "what kind of an organization the

modern university actually is."4
Despite the exhortations to study the university

as an organization, March's Handbook on Organizations.omits

 

completely a discussion of research findings focusing upon
the university.$ It is thus not surprising to find a re-
view of empirical research on the university department to
be both brief and widely scattered in focus.6 The univer-
sity department rather than the college or university as a
whole has become the primary organizational unit in which
the individual faculty member functions. The president of
a university who could at one time set educational policy,
obtain outside funds, and hire faculty by himself has been
joined by professors who can do likewise. Thus, partic-
ularly since World War II when federal funds for research
became more readily available to the university, university

expansion has been in the hands of individual entrepreneurs

 

4William E. Moran, "The Study of University Organ—
izations," Journal of Higher Education, XXXIX (March,
1968), 144-151.

5James G. March, Handbook of Organizations (Chi-
cago: Rand McNally and Company, 1965).

6Paul L. Dressel, F. Craig Johnson, and Philip M.
Marcus, The Confidence Crisis (San Francisco: Jossey-
Bass, 1970), pp. 241-48.

 

with little or no responsibility for coordinating their
scholarly endeavors with those of the university or, at
times, even their own department.

The social organization of the campus, according
to Clark, has moved from a unitary to a federal structure,
from single to multiple value systems, from non-profes—
sional to professional work, and from consensual to bu-
reaucratic coordination. Even though the elements of
bureaucracy in higher education are present, their influ-
ence is often overestimated in the light of many faculty
members' conditional loyalty to the institution as well as
their resistance to rules, standards and/or supervision.7

Universities tend to adapt their organizational
structure to the needs of professional staff members rather
than requiring their faculty to adapt to its structure.
The all-powerful influence of the "invisible hand" attri-
buted to the administrator is often much more apparent to
the faculty than to the administrator himself. Limited in
his ability to anticipate the autonomous actions of the
faculty in various departments in obtaining funds for new
centers, prOgrams, or institutes, the administrator finds

himself hOping and praying that somewhere there is an

 

7Burton R. Clark, "Organizational Adaptation to
Professionals," in Professionalization edited by Howard
M. Vollmer and Donald L. MiIIs (EngIewood Cliffs, New
Jersey: Prentice-Hall, Inc., 1966), p. 284.

 

"invisible hand" that can bring order to a complicated
system.

The problem encountered by the university is one
faced by any organization employing professional staff.

It is one of recognizing the professional autonomy of the
individual faculty member while, at the same time, attempt-
ing to integrate and coordinate his activities within the
context of the organizational environment. When profes-
sional orientation gives way to organizational demands, as
Kornhauser has noted with reference to industrial research
Operations, creativity suffers.8 On the other hand when
the organizational orientation gives in to professional
demands, professional expertise is not utilized in a man-
ner which takes into account the goals or resources of the
employing institution.

The ideal situation is probably one in which the
organization does not wholly absorb professionals, nor do
professionals wholly absorb the organization. The striving
for such an equilibrium should have a favorable outcome
for the university in that the tension between autonomy
and the integration of professional groups, as Kornhauser

suggests, may well tend to bring about a more effective

 

8William Kornhauser, Scientists in Industry
(Berkeley: University of California Press, 1963), p. 157.

 

structure than is attained when autonomy and integration
are isolated from one another or when one absorbs the

other.9

Need for Study

 

So little research has been done on how colleges and
universities are organized and administered that it is
fair to say, in fact, that the field has not been
touched.10 . . . T. R. McConnell
McConnell notes further than a conceptual frame-
work is 1acking which aids one in thinking systematically
about college and university organization and in formulat-
ing a set of hypotheses for investigation. He thus indi-
cates a need (due to the undeveloped state of the art) for
studies which are descriptive and analytical rather than
evaluative or experimental.
In the same publication, Anderson attempted to
relate the problems of higher education to the sociological
literature in regard to complex organizations. In so doing

he described the concepts of bureaucracy, collegiality,

and community and how these concepts might be applicable

 

9Ibid., p. 197.

10T. R. McConnell, "Needed Research in College and
University Organization and Administration," in The Study_
of Academic Administration, ed. by Terry F. Lunsford
(Boulder, Colorado: Western Interstate Commission for
Higher Education, 1963), p. 113.

 

to American colleges and universities.11 Anderson did not
state specific hypotheses to be investigated as did
McConnell. McConnell's hypotheses were, however, of much
too broad a scope to be applicable to this study.

Bolman was more specific in his recommendations
for needed research in the administration of higher educa-
tion.12 Among the areas he recommended for investigation
were patterns of organization and educational aims, in-
flexibilities in departmental and collegiate organization,
power perceptions and preferences, delineation of communi—
cation problems, and the phenomenon of dichotomy between
administration and faculty.

A fourth observer of the higher education scene
who stated some general areas for further research was
Moran.13 Moran's article is notable for its description
of various conceptions of the university offered by such
as Millett, Litchfield, and Kerr. It, too, however, fails
to state specific hypotheses to be investigated in regard

to departmental organization or autonomy.

 

11G. Lester Anderson, "The Organizational Character
of American Colleges and Universities," in The Stud of
Academic Administration, ed. by Terry F. Lunsford (BouIder,
CoIOrado: Western Interstate Commission for Higher Educa-
tion, 1963), pp. 1-19.

12Frederick deW. Bolman, "Needed Research in Ad-
ministration of Higher Education," Educational Record,
XLVI (Spring, 1965). pp. 166-76.

13

 

 

Moran, op. cit.

 

Only when we turn to the broad-scale comparative
study of complex organizations as recommended, for example,
by Blau and Scott, do we begin to acquire a conceptual
framework which enables one to study the processes common
to all organizations.14 A first step is developing a con-
ception of institutional types and the problems encountered
by such organizations. Thus we find the university clas-
sified, in respect to its teaching function, primarily as
a service institution. As such, it is faced with the
conflict between professional service to clients and ad-.
ministrative procedures. Service suffers if the profes-
sional staff members become primarily concerned with
furthering their own careers. In this type of organiza—
tion, the individual being served is generally regarded as
not qualified to judge what is best for him and thus be-
comes vulnerable to exploitation.

Similarly, classified on the basis of cui bono, or

 

who benefits, as recommended by Blau and Scott, the uni-
versity is classified as a commonweal organization in
regard to its research function. .Supposedly, the public
at large benefits in this type of organization. A prin-
ciple problem is that of assuring external democratic

control so that the public may have the means to control

 

14Peter M. Blau and W. Richard Scott, Formal Or-
ganizations: A Comparative Approach (San Francisco:
Chandler Publishing Company, 1962).

 

 

the ends served by the organization. Also of great concern
to this type of organization is encouraging the creativity
of its employees.

The fact that the university's functions are so
broad that it cannot be classified solely by type as fal-
ling into any one particular category makes it all the
more vital that the university department, which incorp-
orates both the research function of the commonweal organ-
ization and the teaching function found in service organ-s
izations, should be studied with reference to the influence,
emphasis, reference group and communication processes
common to all organizations. In this regard a recent
volume based on correlational data, and entitled The Con-

fidence Crisis: An Analysis of University Departments, by

 

Dressel, Johnson, and Marcus, has provided a valuable

description and analysis of departmental operations.15

Problem and Purpose

 

This study represents an extension of the earlier
study by Dressel, Johnson, and Marcus, using as a data
base the same data collected in conjunction with the E330
Departmental Study Project which led to the publication of

The Confidence Crisis. A prominent thesis of this volume

 

was that university departments are out of control for a

 

15Dressel, Johnson, and Marcus, op. cit.

number of reasons including excessive concern for attract-
ing outside research support, the extra institutional
orientation of both faculty members and departmental
goals, as well as outright resistance to the coordination
of departmental activities by elements external to the
department.

Despite the generally recognized existence of au-
tonomous university departments, however, there have been
no investigations to determine how departments and indi-
viduals within departments with varying degrees of per-
ceived autonomy differ in regard to patterns of influence,
emphasis, communication, reference group orientation, or
such characteristics as size, source or amount of support
and Cartter report rating of the department.16

Perhaps just as important as the lack of substan-
tive information in regard to the problem just noted is
the lack of methodological sophistication so common to
much of the survey research conducted in education. In
this regard, Trow notes that "The bulk of survey research
in education has been little more.than a matter of asking
some people some questions and reporting the distribution

17

of their responses." If survey research is to move from

 

16Allan M. Cartter, An Assessment of_gua1it in
Graduate Education (Washington, D.C.: AmeriCan CounciI on
Education, 1966).

17Martin Trow, "Survey Research and Education," in
Survey Research in the Social Sciences, ed. by Charles Y.
Glock’TNew York: Russell Sage Foundation, 1967), p. 319.

 

 

10

a description of relationships to an explanation of rela-
tionships, it is necessary to specify the conditions which
either maximize or minimize a relationship. Only then may
survey research move toward a probable explanation of
causal relationships.

Therefore, in light of the lack of research and/or
methodological deficiencies in current research, it is the
purpose of this study to investigate the manner and condi-
Eigng under which departments and individuals within de-
partments with varying degrees of perceived autonomy differ
in regard to:

1. Characteristics such as (a) Cartter report rating,

 

(b) the percentage of faculty members in a department
holding joint appointments, the percentage of faculty mem-
bers in a given department noting one or more problem
classified by source as (c) internal to (controllable by)
or (d) external to (not controllable) by, the department.

2. The influence in departmental affairs of suCh

 

principle actors within the department as (a) the depart-
ment chairman, (b) the departmental faculty as a whole,
and (c) the individual faculty member.

3. The emphasis actually placed upon such primary

 

departmental goals as (a) undergraduate instruction, (b)

graduate instruction, and (c) basic research.

11

4. The reference_group orientation of departmental

 

faculty to (a) the university, (b) the department, or (c)
the discipline.

5. The degree to which communication takes place be-

 

tween individual members of a department and (a) the entire
university, (b) persons external to the department both
above (deans or university administrators) and outside of
the department (chairman or faculty in other departments),
and (c) persons internal to the department (department
chairman, chairmen of special departmental committees,
departmental faculty and graduate students).

As previously noted, the data for this study were
collected in conjunction with the E350 Departmental Study
Project in which data were gathered from 1,319 faculty
members in departments of English, history, chemistry,
mathematics, electrical engineering, and management in
fifteen universities. The findings from the Departmental

Study Project reported in The Confidence Crisis were.based

 

upon the analysis of ninety-three variables constructed
from a faculty questionnaire and information supplied from.
the departmental records of ninety-seven departments in-
cluded in this study.

This research seeks to determine the correlates of
departmental autonomy and the conditions under which sig-
nificant relationships exist between the degree of depart-

mental autonomy, newly created variables, and selected

12

variables found to be important in departmental operations.
Seventeen additional variables were created. In the pro-
cess, departmental autonomy was redefined as being inver-
sely related to the combined influence of the dean and
university administration in departmental affairs. This
study focuses upon thirty-one variables. Nine of these
variables, including departmental autonomy, were newly.
created. Findings relative to the eight individual commu-.

nication variables were reported in The Confidence Crisis.

 

The remaining fourteen variables were included
because analysis in previously reported research showed
they were associated with differential response patterns
in regard to the nature, source, and resolution of depart-
mental problems and the perception of departmental
autonomy.18

The review of related literature which follows has
as a primary aim the reporting of research findings which

may serve as probable sources for the hypotheses to be

stated at the beginning of Chapter III.

 

18Dressel, Johnson, and Marcus, op. cit., pp. 98,

 

110.

CHAPTER II

REVIEW OF RELATED LITERATURE

Introduction

 

Citation of the results of a survey of holdings of

a major university library system (Cornell) in the general
area of higher education serves to point out the difficul-
ties this writer had in reviewing literature directly ap-
plicable to the functions of autonomous university depart-
ments. At the conclusion of her review, Olive noted:

In the thorough bibliographic search undertaken here,

it finally became apparent that failure to find pub-

lished research was not due to difficulties in locat—

ing and identifying the material, but to the fact that

few research studies exist.

The above conclusion, coupled with the previously

noted fact that March's Handbook of Organizations omits

 

completely any reference to research focusing upon the
university as an organization makes the "universe" of lit-
erature rather limited. The situation is aided somewhat
by the Encyclopedia of Educational Research: Fourth Edi-

tion which contains pertinent summary articles authored

 

lBetsy Ann Olive, "The Administration of Higher
Education: A Bibliographical Survey," Administrative
Scienceguarterly, XI (March, 1967), 671-77.

 

 

13

14

by Millett on "College and University Administration,"
Lorimer and Dressel on "Faculty Characteristics--College
and University," and Fattu on "Research Organizations."2

When, however, a search for literature focusing
specifically upon the university department is made the
available stock is lessened considerably. Further attempts
to limit the search to literature dealing with the manner
in which autonomous university departments operate would
virtually eliminate the need for this chapter.

On the other hand, if one were to attempt to review
the literature with reference to the concepts of bureau-
cratization, socialization, professionalization or the
reference group orientations of various actors within a
complex organization, all of which are related to this
study, the volume of literature would be prohibitive.
Therefore, an attempt has been made to cite only selected
literature which:

1. bears directly on the consequences or problems
encountered in the operation of autonomous university
departments,

2. helps to define the concept of autonomy, or

 

2John D. Millett, "College and University Adminis-
tration," pp. 161-69; Margaret F. Lorimer and Paul L.
Dressel, "Faculty Characteristics--College and University,"
pp. 488-494; and Nicholas A. Fattu, "Research Organiza-
tions," pp. 1152-1158 in Encyclopedia of Educational Re-
search: Fourth Edition, ed. by RBbert L. Ebel (The Mac
Millan Company, 1969).

 

 

15

3. reports research findings related to the influence,
emphasis, reference group orientations or communication
patterns of professors or scientists within a university
or laboratory setting which may be examined in terms of

their probable relationship to departmental autonomy.

Departmental Autonomy

 

A number of viewpoints have been put forth as to
the desirability of departmentalization. Regardless of
its utility or effectiveness, however, Demareth, Stephens,
and Taylor, have indicated that the academic department is
to the university, what the cell is to the body.3 Some
doubt may then be cast on the health of the organism, as
the authors state that the department is often an obstacle
to innovation wherein faculty of similar interests cling
together in order to resist intrusions from outsiders.

This tendency toward stagnation has been documented
by Hefferlin who examined the manner in which changes
came about in the educational programs of colleges and
universities. Approximately forty-one per cent of his
respondents indicated that faculty or departments repre-
sented the greatest obstacle to academic change. Eighteen

per cent indicated general inertia or conservatism as the

 

3Nicholas J. Demareth, Robert W. Stephens, and R.
Robb Taylor, Power, Presidents and Professors (New York:
Basic Books, 1967f, pp. 181-88.

 

16

main obstacle to change while lack of funds or resources
and administration were cited by sixteen and nine per cent
of the respondents, respectively.4

On the other hand, Andersen takes a more charitable
view toward the department and has argued that the univer-
sity department has been and is likely to remain a viable
force in American education.5 This contention is ques-
tioned by Dressel and Reichard who traced the history of
departmental organization and point to some of the dys-
functions or needs not fulfilled by such a structure.6

Clark sees the department as the center of entre-
preneurial activity in seeking funds and states that
American universities are more innovative than European
universities, due in part to their openness to external
fund sources. He notes that this openness to external
fund sources also encourages competition for funds in lieu
of a guaranteed annual budget and results in decentralized

decision making, primarily at the departmental level.7

 

4JB Lon Hefferlin, The Dynamics of Academic Re-
form (San Francisco: Jossey-Bass Inc., 1969), p. 100.

5Kay J. Andersen, "The Ambivalent Department,"
Educational Record, IL (Spring, 1968), 206-12.

6Paul L. Dressel and Donald J. Reichard, "The Uni-
versity Department: Retrospect and Prospect," Journal of
Higher Education, XLI(May, 1970), 387-402.

7Terry Clark, "Institutionalization of Innovations
in Higher Education," Administrative Science Quarterly,
XIII (June, 1968), 16-21.

 

 

17

Dressel, Johnson, and Marcus have described this
pattern of competition among departments for university
funds as a "confidence game" in which individual depart-
ments attempt to acquire resources and the freedom to use
them without external interference.8 The same authors

suggested in a volume entitled, The Confidence Crisis: An

 

Analysis of University Departments, that a possible effect
of departmental autonomy might be that it leads to a grad-
ual commitment of university resources rather than the
development of a rational plan for the university as a
whole. Open-ended responses to this statement were almost
evenly split between those who agreed and disagreed that
this situation existed at their own institutions. Approx-
imately forty-six per cent of the respondents felt that
something could or should be done to correct autonomy while
forty-two per cent maintained that autonomy should not be
corrected.

Among those who felt that something should be done
to correct autonomy, there was much disagreement as to how
change was to be brought about. Approximately fifty per
cent of the responses recommended better planning or better
administrative leadership. However, recommendations for

more faculty involvement in decision making, better

 

8Paul L. Dressel, F. Craig Johnson.and Philip M.
Marcus, "Departmental Operations: The Confidence Game,"
Educational Record, L (Summer, 1969), 274-78.

 

18

c00peration among departments in planning, or flexible

adherence to a plan, constituted thirty-seven per cent of

the responses and seemingly indicated a distrust of admin-

istrative action.

Among faculty who urged that nothing be done to

correct departmental autonomy, responses included state-

ments to the effect that autonomy generally worked out

well (48.8%), departments have a better knowledge of needs

(21.3%), the administration's solution would be worse

(15.3%), and rational planning was impossible (13.9%).

All responses appeared to reflect administrative mistrust.

The effects of the "confidence game" are such as

to lead Dressel, Johnson and Marcus to conclude that:

The universities and the departments within them are
out of control. Administrators and faculties too
readily interpret their own aspirations as meeting or
transcending the educational needs of the clientele
which they serve. In seeking support to fulfill these
aspirations, they engage in half-truths and reluctant-
ly acquiesce to requests for data which are so select-
ed, manipulated, and presented as to support their
case. In the defense of administrators, it must be
said that they are caught between the insatiable de-
mands of departments and faculties more concerned with
self-advancement than with service and a supporting
clientele which often does not understand the univer-
sity and therefore is capable, wittingly or unwitting-
ly, of using the truth to impede or destroy its
effectiveness. In most cases, too, presidents and
vice-presidents do not know just what is going on in
the colleges and departments, for these units also
engage in half-truths and misrepresentations shrewdly
calculated to attain their own ends.

 

9Paul L. Dressel, F. Craig Johnson, and Philip M.

Marcus, The Confidence Crisis (San Francisco: Jossey-Bass
1110. I 197“, pp. 108-090

 

9

19

Departments and other units within the university
must be brought under control so that their resources
are allocated and used in accord with priorities set
for the university by the university in cooperation
with those who support it.

Faculty Autonomy

Just as departmental autonomy is a group charac-
teristic resulting in part from the complex organizational
nature of the university, faculty autonomy is the result
in part of the long-standing struggle of individuals for
professional autonomy. This struggle for academic freedom
has been well summarized by Beach who described the long-
standing antagonisms between professional administrators
and faculty members which led to the founding of the
American Association of University Professors in 1915.11

Indeed, administrators who complain that the fac-
ulty has no interest in the university as a whole might
well blame such "captains of erudition" as Nicholas Murray
Butler and William Rainey Harper who maintained tight per-
sonal control of their universities causing John Dewey,
the first AAUP president, to argue "that professors being
only human, lose interest in enterprises when they have no

"12

voice in the decisions which affect them. The outcome,

 

lOIbid., p. 232.

11Mark Beach, "Professional Versus Professorial
Control of Higher Education," Educational Record, IL
(Summer, 1968), 263-73.

12

 

Ibid., p. 272.

20

Beach concluded, was that:

Each party to the construction of the university sought
to mold the organization to its own cherished ends.

As a result, the republic of learning provided by the
new education threatened to become instead an anarchy
of scholarship.

If it did not produce anarchy, it did produce an Academic

Revolution, or a rise to power of professional scholars
14

 

and scientists described by Jencks and Riesman.
Indeed, the faculty, individually and collectively
have become such an influential power group that Kristol
states "the beginning of wisdom in thinking about our
universities, is to assume that the professors are a class
with a vested interest in, and an implicit ideological
commitment to the status quo, broadly defined, and that
reform will have to be imposed on them as upon everyone
else."15
On the other hand, in English universities, the
faculty were always dominant. According to Ashby, "For
seven hundred years we have believed that a university is

not a corporation with employees, it is a society with

members." The result is an academic life which Ashby

 

13Ibid., p. 273.

14ChristOpher Jencks and David Riesman, The Aca-
demic Revolution (New York: Doubleday and Company, Inc.,
1968).

15Irwin Kristol, "A Different Way to Restructure
the University," The New York Times Magazine, December 8,
1968, p. 50.

 

21

characterizes as being "anarchic, nomadic and schizo-
phrenic" and in which, as in America, scholars "ask not
what Professor X has done for Manchester, but what has
Professor X done for Chemistry?"16
According to Clark, with reference to American
institutions:
The role of faculty authority is shifting from protect-
ing the right of the entire guild, the right of the
collective faculty, to protecting the autonomy of the
separate disciplines, and the autonomy of the individ-
pal faculty member. The university and the large col;
age are fractured by expertness, not unified by It.
When professionalism is combined with bureaucrat-
ization, a situation arises in which the individual faculty
member must adapt to the impact of bureaucracy in terms of
his own behavior. In this regard Page, in his article
entitled, "Bureaucracy in Higher Education," borrowing
heavily from Merton, described four ideal-type categories
of adaptation which were employed by the academician.
These four types of individuals were termed the ritualist,
academic neurotic, robber baron, and academic rebel. Of

these types the robber barron portrays many of the traits

of the individuals who, it may be assumed, would

 

16Eric Ashby, "Anatomy of the Academic Life," Edu-
cational Record, IIL (Winter, 1967), 45-50.

l7Burton R. Clark, "Organizational Adaptation to
Professionals," in Professionalization, ed. by Howard M.
Vollmer and Donald L. Mills (Englewood Cliffs, New Jersey:
Prentice Hall, Inc., 1966), pp. 288, 290-91.

 

22

participate in what Dressel, Johnson, and Marcus have
termed the "confidence game":

The robber baron, like the ritualist, is well adapted
to his bureaucratic surroundings. But his adaptation
takes a completely different form; for, rather than
glorifying the routines of collegiate life, rather
than making ends of means, he will, when it serves his
own ends best, ignore bureaucratic propriety alto-
gether. The ability to by-pass prescribed methods, to
avoid consultations or procedures called for by the
institution's formal code, to cut red tape, demands
enormous skills, the courage of one's convictions, and
an intense desire to further one's own ends. The
robber baron manifests precisely these qualities, in
much the same way that his prototype in another realm
--say, a Carnegie or a Rockefeller--reveals these
traits. If the academic robber baron is to escape
disgrace and is to remain within the collegiate enter-
prise, he must possess a highly realistic knowledge of
the academic world, must recognize the functionally
strategic relationships in and between the formal and
informal structures, and must display manipulative
ability.

Robber barons need not be the villains of history;
sometimes they are heroes. And this is the case of
the academic robber baron as well. If their ends cor-
respond with, or are congenial to, the professed goals
of the institution itself, the robber baron may, in
fact, become a collegiate saint. Just as we forgive
and "canonize" our Carnegies and Russel Sages, once
they have made good on the grand scale (i.e., have
conformed to and confirmed the culturally approved
value of material success) and once they have endowed
posterity with socially useful edifices, so do we
exalt our academic robber barons when they have con-
formed to and confirmed the highest values of educa-
tion.

In addition to Page, Hall and Tittle examined the

academic department in terms of its bureaucratic structure.

 

18Charles H. Page, "Bureaucracy and Higher Educa-
tion," in Sociology of Education, ed. by Ronald M. Pavalko
(Itasca, Illinois: F. E. Peacock Publishers, Inc., 1968),
p. 463.

 

23

On an overall bureaucracy scale, using Guttman scaling,
the authors ranked the academic department tenth in bu-
reaucratization out of twenty-five varied types of organ-
izations. The academic department was characterized as
possessing such bureaucratic elements as impersonality of
operations, division of labor, specificity of procedures
and complexity of rules but was lacking in the elements of
hierarchy of authority and technical competency as a cri-

terion for hiring and promotion.19

The number of types of
organizations included, the manner in which a single aca-
demic department may be generalized to represent all de-
partments, and the admitted need on the part of the authors
for conceptual clarification in regard to the elements of

bureaucracy, however, all serve to urge restraint in the

interpretation of these findings.

Autonomy

 

Autonomy is a term used in many varied contexts.
It is often spoken of in regard to public institutions,
usually demanding freedom from legislative control. Sim-
ilarly it is often used in regard to individuals, generally
professionals, whose eXpertise in a particular area war-

rants personal autonomy or freedom in conducting one's

 

19Richard H. Hall and Charles R. Tittle, "A Note
on Bureaucracy and its Correlates," The American Journal
of Sociology, LXXII (November, 1966), 268-69.

 

 

24

business. In any case, autonomy may be defined in terms
of behavior which is not controlled by an external agency.
In conjunction with this study, autonomy is a group char-
acteristic defined as being inversely related to the com-
bined influence of the dean and university administration
in departmental affairs which is perceived by a group or
collectivity of individuals in a given department.

Whereas complex organizations are often viewed as
all-powerful bureaucracies with rules and regulations gov-
erning all aSpects of patterned behavior, Katz, with ref-
erence to the informal organization inherent in all organ-
izations, maintains that "autonomy patterns are distinctly

20 In a later

structured within complex organizations."
volume, Katz notes that "the underlying theme of these
studies is that a degree of independence exists within
most formal arrangements," and that "One is inclined to
think of autonomy as a separation from a social context,
or as activity that isolates an individual from other per-
sons. In contrast autonomy is here regarded as a force

21

that binds peOple together." Katz indicates that

 

20Fred E. Katz, "The School as a Complex Social
Organization," in Sociology of Education, ed. by Ronald

M. Pavalko (Itasca, Illinois: F. E. Peacock Publishers,
Inc., 1968): P. 427.

21Fred E. Katz, Autonomy and Organization: The
Limits of Social Control (New York: Random House, 1968),
p. 4.

 

25

autonomy is specified for positions by internal and ex-
ternal roles, each of which are enacted within and outside
of the organization.22
Engel, in an article entitled, "Professional Au-
tonomy and Bureaucratic Organization," found that bureauc-
racy was not necessarily inimical to the degree of
autonomy perceived by professional physicians. Thus
physicians working in privately owned organizations char-
acterized by a moderate degree of bureaucracy perceived
greater personal autonomy with respect to both clinical
practice and research than did physicians working in gov-
ernmentally associated organizations or as solo practi-
tioners in settings which were characterized as having
high and low degrees of bureaucracy, respectively. Engel
concluded that a moderate degree of bureaucratization can
be of great aid to professionals by supplying funds, equip-
ment, technical personnel, and other physical facilities
which enhance both his performance and his perception of
professional autonomy.23

Research reported by Kornhauser indicates that of

the three primary settings for research--the university,

 

22Katz, "The School as a Complex Social Organiza-
tion," p. 428.

23Gloria V. Engel, "Professional Autonomy and
Bureaucratic Organization," Administrative Science Quart-
erly, XV (March, 1970), 12-21.

 

26

industry, and government, the university provides the

greatest amount of freedom to pursue one's own research.

To document this point, Kornhauser cited a study bereltzer

of seventy-five per cent of the physiologists in the United

States. The study indicated that ninety per cent of the

physiologists employed by universities, seventy-one per

cent employed by government, and fifty per cent employed

by industry reported "complete" or a "great deal" of free-

dom in choosing research problems.24
Perhaps the most comprehensive examination of or-

ganizational features in relation to research productivity

is Pelz and Andrews' volume entitled, Scientists in Organ-

 

izations: Productive Climates for Research and Develop:

ment.25 The book was based upon the responses and perfor-

 

mance evaluations of 611 professionals in five independent
laboratories, 144 professors in seven departments within a
university, and 526 scientists and engineers from five
government laboratories. Of particular interest is the
fact that the study dealt with the concept of autonomy as
both a group (amount of coordination) and individual char-

acteristic (freedom to pursue one's own choice of goals or

 

24William Kornhauser, Scientists in Industry
(Berkeley: University of California Press, 1963 , p. 132.

25Donald C. Pelz and Frank M. Andrews, Scientists
in Organizations: Productive Climates for Research and
Development (New York: John Wiley and Sons, 1966).

 

 

 

 

27

objectives). The authors, in regard to the amount of
coordination most frequently associated with optimal per-
formance, note that "in loosely coordinated settings, the

most autonomous individuals did poorly--perhaps because

26

they were isolated from stimulation." As summarized by

Fattu:

The more loosely coordinated a situation, the more
essential it was for the individual to remain strongly
motivated if he were to achieve. A fairly high level
of autonomy was effective mainly in the middle range
of coordination. In loose coordination, where members
had considerable freedom, the most autonomous scient-
ists were below average in performance. It was sug-
gested that in loose settings the most autonomous
scientists tended to withdraw from outer stimulation,
thus weakening opportunity to improve their perfor-
mance. In rigid situations, autonomous persons were
inhibited. In middle-range situations high autonomy
was accompanied by several strong motivations and
stimulations, and the setting appeared to improve per-
formance. In the loosest settings full autonomy seemed
to encourage complacency rather than motivation and
narrow specialization rather than breadth. Isolation
was not a good climate for achievement.27

In regard to the autonomy or individual freedom of
the individual faculty member to pursue his own research
goals or objectives, Pelz and Andrews asked if coordination
was compatible with personal freedom. They concluded that

28

"best performance occurred when both were present." .It

appeared in development-oriented-laboratories that as a

 

26Ibid., p. 214.

27Fattu, 9p. cit., p. 1158.

 

28Pelz and Andrews, 9p. cit., p. 8.

 

28

number of such people as the individual Ph.D. level re-
searcher's colleagues, immediate supervisor, higher level
technical supervisors, subordinates, and others at various
levels were influential in important decisions, performance
increased in terms of the individual scientist's scientific
contributions, usefulness to the organization, and the

29 Thus the rela-

production of both papers and reports.
tionship between individual autonomy and productivity was
curvilinear with productivity highest when autonomy within

the organization and the autonomy of the individual were

moderate.

Influence of Various Actors within
the Department

A review of empirical literature on the department
notes several studies which serve to describe the functions
of various actors.30 However, most of the literature
fails to indicate the relative power of a number of indi-
viduals influential in the operations of a given
department.

Hill and French examined the perceptions of the
power of department chairmen held by professors in sixty-
five departments within five state supported four-year

colleges, and related them to the perceived productivity

 

29Ibid., pp. 11-12.

30Dressel, Johnson and Marcus, The Confidence

Crisis, pp. 241-48.

 

29

and satisfaction of the departmental faculty. The authors
infer that "it is the power of the chairman to speak ef-
fectively on behalf of the faculty that explains the
positive association between the chairman's power and the

satisfaction of professors."31

The power of the department
chairman was correlated negatively with professional output
in terms of publications but was positively related to
perceived faculty productivity in terms of the college's
own goals which for the most part, placed high emphasis
upon teaching. More important, in regard to the interest
of this study, however, was the finding that in terms of
the relative control exercised by each level over all
levels of authority,department chairmen were accorded less
influence than professors and a lesser amount of influence
than every other succeeding level of authority ranging
from middle to higher administration and state governing
boards.32
Gross and Grambsch substantiated the findings of
Hill and French in regard to the relative power of the
faculty vis-a-vis the department chairman. Drawing upon

the responses of some 7,224 faculty members and adminis-

trators in sixty-eight nondenominational universities,

 

31Winston W. Hill and.Wendell L. French, "Percep-
tions of the Power of Departmental Chairmen by Professors,‘
Administrative Science Quarterly, XI (March, 1967), 573.

32

Ibid., pp. 557-58.

30

Gross and Grambsch queried, "How much say do you believe
persons in (various) positions have in affecting the major
goals of the university?" Faculty members were found to
have a slightly greater influence in the overall power
structure of American universities than were department
chairmen. Although the influence of department chairmen
was less than that of all other actors within the univer-
sity, it was distinctly more powerful than that of actors
external to the university except for that of the
regents.33

Pelz and Andrews also investigated the relationship
between the influence of the individual scientist and the.
degree of his autonomy. They concluded that scientists in
organizations could "have high influence even though they
might lack full autonomy or freedom to go their own way."34
The overall conclusion was that "some scientists can be
creative when completely self-determining; but in our

35 They noted further.

sample they formed a small minority."
that "A few high-status Ph.D.s in research, possessing

high influence were effective when no one else helped

 

33Edward Gross and Paul V. Grambsch, University
Goals and Academic Power (Washington, D.C.: American
Council on EducaEion, 1968), pp. 76, 143.

34

 

 

Pelz and Andrews, op. cit., p. 17.

35Ibid., p. 25.

31

decide their goals, but those with lower status or influ-

ence worked best when several others were included.36

Emphasis

The sociological literature in regard to complex
organizations contains a large body of material dealing
with organizational goals. Much more limited,.however,
are the attempts to identify the variety of goals espoused.
by various elements within educational institutions..-Gross
and Grambsch's Universitnyoals and Academic Power provide

the most comprehensive examination of this subject.37

 

.The
authors compared the degree of emphasis which individual
faculty members and administrators perceive to be placed
upon forty-seven output and support goals with the amount
of emphasis faculty and administrators feel should be
placed upon such goals. In the analysis of their data,
Gross and Grambsch take into account the relationship be-
tween the goals which are or should be emphasized and such
global characteristics as size, location, type of control,
productivity, Cartter rating and graduate emphasis.
Although Gross and Grambsch's research focuses
upon the responses of individual faculty members rather

than the collective re3ponses of individuals within a

 

361bid., p. 19.

37Gross and Grambsch, 9p. cit.

 

32

given department, a look at some of their findings is
helpful in that such goals as carrying on pure research,
encouraging graduate work and emphasizing undergraduate
instruction are either the same or very closely related
to the goals of emphasizing basic research, graduate in-
struction, and undergraduate instruction which are of con-
cern to this study. In addition such goals as protecting
academic freedom, rewarding faculty for.contributions to
their profession or institutions, developing pride in or
faculty loyalty to the university, or letting the will of
the faculty prevail, are closely related to the question
of departmental autonomy or the individual autonomy of
faculty members.

The ranks of related.perceived and preferred goals

of interest to this study were as follows:38

Perceived Preferred

 

 

Goal Rank Rank
Protect academic freedom 1 1
Carry on pure research 7 16
Encourage graduate work 18 27
Give faculty maximum opportunity
to pursue careers 22 25
Reward for contribution to profession 26 21
Reward for contribution to institution 32 13
Develop pride in university- 33 r 23
Let will of faculty prevail 36 24
Develop faculty loyalty to institution 42 29
Emphasize undergraduate instruction 44 44
Preserve institutional character 46 47

 

381bid., pp. 28-29.

33

In general, student related goals,.whether they
pertained to intellectual or personal development, ranked
extremely low as did goals which evidenced a desire for
preservation of or orientation to the university as an
organic whole. On the other hand, goals which had to do
with professional advancement, faculty prerogative, and
particularly, the preservation of academic freedom, re-
ceived very high emphasis. Developing loyalty on the part
of the faculty and staff to the university rather than.
only to their job or professions and making sure that
salaries, teaching assignments, perquisites and privileges
always reflected the contribution that the person involved
is making to the functioning of the university all received
less perceived emphasis than was preferred. Conversely,
the goal of encouraging students to go into graduate work
received much more emphasis than was felt should be given.39

Large staff size was associated with-a greater
emphasis upon pure research and less emphasis upon devel-
Oping pride in the university. However, the remaining
global characteristics bore no significant relationship
to the goals of special interest to this study.40 In con-

trasting the preferred goals of the faculty and adminis-

trators, Gross and Grambsch found a surprisingly high

 

391bid., pp. 34-35.

40Ibid., p. 44.

34

amount of agreement between the two groups. Administrators
generally, however, placed more emphasis than faculty mem-
bers upon student development goals, and also placed
greater emphasis upon basing rewards on the contribution
that the person makes to the institution, developing fac-
ulty loyalty to the university, and developing pride in

the university. Conversely, faculty tended to place a
higher value than administrators on making sure that on

all important issues (not only curriculum), the will of

the faculty prevails and upon making the university a
place in which faculty have a maximum opportunity to pur-
sue their careers in a manner satisfactory to them by their

. . 41
own criteria.

Reference Group Orientation.

 

Reference group orientation may be inferred, in
part, from communication with an extra-institutional group
of colleagues owing allegiance to the same discipline.
Therefore, it is difficult to view certain findings in
regard to communication and reference group orientation as
being mutually exclusive. Similarly, influence and refer-
ence group orientation are not mutually exclusive. Thus,
Gouldner found that influence in college policy decisions

at "Coop College" (Antioch) increased steadily as one

 

411bid., p. 103.

35

moved from a cosmopolitan to a local orientation, except
in the case of extreme locals who exerted very little
influence.42

Kornhauser found that those whose career orienta-
tions were directed exclusively neither to the profession
nor the organization, but were mixed, enjoyed the greatest
influence.' Thus the percentage of scientists who were
regarded as possessing high influence was approximately
equal for those with professional (55%) and organizational
(53%) career orientations, while a high amount of influence...
was attributed to sixty-six per cent of the scientists whose
orientations were mixed.43

Glaser, in developing the concept of "The Local-
Cosmopolitan Scientist" suggests that cosmopolitanism and
localism may be seen as two dimensions of an individual's
orientation which are activated at the opportune time and
place depending on the organizational structure within
which the scientist works.44 When institutional goals
were in accord with those of individual scientists who

were highly motivated to advance knowledge, Glaser

 

42Alvin W. Gouldner, "Cosmopolitans and Locals:
Toward an Analysis of Latent Social Roles - I," Adminis-
trative Science Quarterly, II (December, 1957), 299.

43

 

 

Kornhauser, 9p. cit., p. 123.

 

44Barney G. Glaser, "The Local-CosmOpolitan Scien-
tist," American Journal of Sociology, LXIX (November,'
1963), 257.

36

suggested that the organization would benefit in terms of
the individual scientist's desire to fulfill the organiza-
tion's scientific goals as well as its nonscientific goals.
Thus, for example, a scientist would be willing to take
time from research in order to aid in such nonscientific
activities as staff recruitment because maintaining or
improving the research staff would preserve the prestige
and/or sources of support of the organization. Maintenance
or improvement of the organization's personnel and or fi-
nancial resources would, in turn, aid the scientist in
fulfilling his scientific goals.45 Caplow and McGee also
acknowledged:
In a handful of great universities, where many of the
departments believed to be the best in their fields
are found, a merger of orientations is possible.
There a man may simultaneously serve an institution
and a discipline and identify with both. But tensions
exist between the two orientations everywhere. It is
worthy of note that the publication requirements in
the highest ranking departments are the most rigid, so
that the men they select have already met the require-
ments imposed by the discipline.46
Several researchers have examined extra-depart-
mental communication networks which have generally been

regarded as indicators of a disciplinary orientation.

Hagstrom, using Yule's Q, found a very high correlation

 

451bid., p. 256.

46Theodore Caplow and Reece J. McGee, The Academic

Market lace, Basic Books (New York: John Wiley and Sons,
Inc., I958), p. 85.

37

(.85) between extradepartmental communication with those
in the same discipline and research productivity of aca-
demic scientists while the correlation between extrade-
partmental and intradepartmental communication (.54) was
much lower.47

Crane looked at the diffusion of agricultural in-
novations in order to examine the types of social ties
existing among scientists who had published in a particular
problem area. Analysis of sociometric connectedness re-
vealed that a tie with one or more highly productive
scientists brought other scientists of less productivity
into a large network of influence and communication. The
groups were not tightly knit enough or sufficiently closed
to external influence however to constitute an elite group
of mutually interacting and productive scientists within a
research area which Price has referred to as an "invisible
college."48

Also related to a disciplinary orientation is the
question of the loyalty of faculty members to their insti—
tutions. In this regard, Lewis noted that the problem of

faculty loyalty is significant because of the tension be-

tween loyalty and expertise which is a central need of

 

47Warren 0. Hagstrom, The Scientific Community
(New York: Basic Books, 1965), p.750.

48Diana Crane, "Social Structure in A Group of
Scientists: A Test of the 'Invisible College' Hypothesis,"
American Sociological Review, XXXIV (June, 1969), 348.

 

38

institutions of higher learning. Professional prestige
was found to be more powerful than university prestige in
determining faculty loyalty. In summation, Lewis stated:
It is found that those with high institutional prestige
and those with low professional prestige are more
loyal than their counterparts,.and that those with a
combination of high institutional prestige and low
professional prestige are the most loyal while those
with low institutional prestige and higher professional
prestige are the least loyal. 9
All of these findings suggest that.reference group
orientations, whether they are to the university, depart-

ment, or discipline, are quite complex.

Communication

 

Communications within the university are rarely
the subject of investigation. Hagstrom did, however, in-
terview seventy-nine university scientists, and as pre-
viously noted, in applying Yule's Q, reported a correlation
of .54 between intradepartmental and extradepartmental
communication. Also, of interest were moderately high
correlations between intradepartmental communication and
participation in professional societies (.55) and the cor-
relation between intradepartmental.communication and the

production of articles for publication (.42).50

 

49Lionel S. Lewis, "On Prestige and Loyalty of
University Faculty," Administrative Scienceguarterly, XI
(March, 1967), 629.

50

 

Hagstrom, op. cit., p. 50.

39

The general conclusion reached by Pelz and Andrews
after an extensive examination of communication patterns,
primarily in independent or government research laboratories
rather than a university setting, was that "effective
scientists both sought and received more contact with col-
leagues."51 Scientists who were most productive either
spent a large total amount of time communicating or spent
brief periods of time talking with a number of individuals.52

In an earlier publication, Pelz found that "highest
performance is shown by those who, on the average, have
daily contact with five colleagues possessing values dis-

similar from their own."53

Choosing colleagues with whom
one had daily contact and who were dissimilar in terms of
values and experiences was associated with high performance
while low performance was common when contact was on a
weekly basis. Conversely, weekly as opposed to daily con-
tact with similar colleagues was associated with high
performance. The Opposite was true in regard to contact

with the single most important colleague rather than a

group of colleagues. Here, performance went up when one

 

51Pelz and Andrews, op. cit., p. 35.

52Ibid., p. 49.

53Donald C. Pelz, Motivation of the Engineering
and Research Specialist, General Management Series, Number
186 (New York: AmeriCan Management Association Inc.,
1957), p. 37.

 

 

40

had daily contact with a single most important colleague
having the same values while performance decreased when
there was daily contact with one's colleagues who held

opposing values.54

Summary

This review of literature has been wide ranging,
and has served to affirm Olive's conclusion that although
much lip service is paid to the need for research on the
administration of higher education, few research studies

exist. Indeed, The Confidence Crisis by Dressel, Johnson,.

 

and Marcus, is the only comprehensive study of the univer-
sity department which examines the issue of departmental
autonomy and the influence, emphasis, reference group, and
communication patterns which take place within it.
Discussions of the concept of autonomy as related
to the manner in which professionals function in complex
organizations have value in theory but research findings
in this regard are reported primarily with reference to
non-university settings. Pelz and Andrews as well as
Engel have indicated that a medium degree of coordination
or bureaucratization, respectively, may have optimum out-
comes in terms of the performance of professionals in

organizations. If a medium degree of autonomy may be

 

54Ibido ’ pp. 39-42.

41

roughly equated with a medium degree of coordination or
bureaucratization, and if the Cartter report may be inter-
preted as a rough criterion of performance, these findings
may lead one to hypothesize that a medium degree of de-
partmental autonomy may be most frequently associated with
a favorable Cartter rating.

Hypotheses that the internal, external, and total
amounts of communication would be highest in departments
of medium autonomy would also stem in part from the high
association Pelz and Andrews found between performance and
the amount of communication with various elements within
the organization. Thus, due to the perhaps unwarranted
assumptions necessary in generalizing concepts, and the
differences in settings, the findings of Engel, and Pelz
and Andrews may or may not be either applicable or repli-
cable within a university setting.

Much of the literature cited in regard to the
influence, emphasis, reference group orientation, and
communication patterns, even if obtained from university
settings, generally has a slightly different focus from
this study. Hill and French as well as Gross and Grambsch
find the influence of the department chairman, in terms of
determining university-wide goals, to be less influential
than that of the collective faculty in a department.
Whether this finding would hold in regard to departmental

rather than university matters is questionable.

42

The department chairman is often referred to as

the man in the middle between

faculty. This would lead one

the dean and the departmental

to believe that when the

dean and university administration were relatively influ-

ential, and therefore, the autonomy of the department was

low, the chairman's influence
Similarly, when the influence
administration are relatively
tonomy is high, the influence

expected to be relatively low

and collectively should enjoy:

condition.

would be relatively high.

of the dean and university
low, and departmental au-

of the chairman would be

as the faculty, individually,

greater influence under this

In regard to goals, Gross and Grambsch's examina-

tion of university-wide

serves to point out the

rather than departmental goals

seeming contradiction between em-

phasis upon concern for or loyalty to the institution and

the desire that rewards
into account the individual's

tion.

for faculty members should take'

contribution to the institu-

Caplow and McGee as well as Glaser suggest, however,.

that there need be no conflict between institutional loy-

alty and a disciplinary reference orientation if the in-

stitution can provide the resources which allow the faculty

member to pursue his own goals.

Crane and Hagstrom both document the extra-insti-

tutional communication patterns of academicians.which

would seemingly indicate a disciplinary reference group

43

orientation. Thus Crane's examination of the "invisible
college" hypothesis and Hagstrom's examination of communi-
cation patterns with one's colleagues in the discipline
lead one to hypothesize a direct relationship between a
disciplinary orientation and the degree of departmental
autonomy. Conversely, a low university orientation would
appear to be logically associated with a low degree of
departmental autonomy while a departmental orientation.may
represent a middle ground between the.university and the
discipline most commonly found in departments with a medium.
degree of autonomy. In any event, reference group orien-
tations appear to be so complex that few simple relation-
ships may be expected to exist.

Finally, the attitudes conveyed toward departmental
autonomy and the lack of agreement in regard to the desir-
ability of departmental autonomy as reported by Dressel,
Johnson and Marcus suggest that if one were to wish to
obtain a clearer picture as to the implications of depart-
mental autonomy, he would do well to investigate further
the relationship between departmental autonomy and the
influence, emphasis, reference group orientation, and com-
munication patterns within the university department.
Toward this end, we now turn to the statement of hypotheses.

and description of the procedures employed in this study.

CHAPTER III

STATEMENT OF HYPOTHESES AND DEFINITION

OF TERMS AND PROCEDURES

This chapter is composed of four main sections.
Section one states the hypotheses to be investigated in
the study. Section two describes the sample from which
the data were drawn and the response rates. The third
portion of the chapter notes the manner in which the
variables were constructed. The final section of the
chapter indicates the manner in which the data were
analyzed through the use of the process of elaboration,
Spearman rank order correlation coefficients, and analysis

of variance.

Statement of Hypotheses

 

On the basis of the preceding review of literature
and prior work in the area, the following sixteen rela-
tionships are hypothesized:

1. Characteristics

 

A. The relationship between departmental autonomy v
and a favorable rating in the Cartter report
will be curvilinear. Departments with a med-
ium degree of autonomy will be rated in the.

44

45

Cartter report more frequently than departments
with a high or low degree of autonomy.

The percentage of faculty members within a
department holding joint appointments will be
inversely related to the degree of departmental
autonomy.

The percentage of faculty members in a given
department indicating one or more problems
classified by source as internal or under the
control of the department, will be inversely
related to the degree of departmental autonomy.

The percentage of faculty members in a given
department indicating one or more problems
classified by source as external or beyond the
control of the department will be directly
related to the degree of departmental autonomy.

Influence

 

A.

The influence of the department chairman in
departmental affairs will be inversely related
to the degree of departmental autonomy.

The influence of the departmental faculty as a
whole in departmental affairs will be directly
related to the degree of departmental autonomy.

The influence of the.individual faculty member
in departmental affairs will be directly re-
lated to the degree of departmental autonomy.

Emphasis

 

A.

Emphasis upon undergraduate instruction as a
departmental goal will be inversely related to
the degree of departmental autonomy.

Emphasis upon graduate instruction as a de-
partmental goal will be directly related to
the degree of departmental autonomy.

Emphasis upon baSic research as a departmental
goal will be directly related to the degree of
departmental autonomy.

Reference Group Orientation

 

A.

Orientation to the university as a reference
group will be inversely related to the degree
of departmental autonomy.

46

B. The relationship between departmental autonomy
and orientation to the department as a refer-
ence group will be curvilinear. A higher per-
centage of individuals within departments with
a medium degree of autonomy than in departments
of high or low autonomy will be oriented to
the department as a reference group.

C. Orientation to the discipline as a reference
group will be directly related to the degree
of departmental autonomy.

5. Communication

 

A. The relationship between departmental autonomy.
and the total amount of communication by a
department will be curvilinear. Individuals
within departments with a medium degree of au-
tonomy will communicate more frequently with
persons internal or external to the departments
than will faculty in departments with a high
or low degree of autonomy.

B. The relationship between departmental autonomy
and the amount of communication with persons
external to the department will be curvilinear.
Individuals within departments with a medium
degree of autonomy will communicate more fre-
quently with persons outside of their own
department than will persons in departments
with a high or low degree of autonomy.

C. The relationship between departmental autonomy
and the amount of internal communication within
departments will be curvilinear. Individuals
within departments with a medium degree of an-
tonomy will communicate more frequently with
persons within their own department than will
persons in departments with a high or low
degree of autonomy.

The Sample

 

The sample was drawn from faculty in departments
of English, history, chemistry, mathematics, psychology,

electrical engineering, and management at each of fifteen

47

universities. The universities and the departments within
them varied greatly in regard to such factors as region,
source of support, size, the ability to attract research
funds from external sources, and Cartter ratings. The

following universities were included in the sample:

University of Arizona Northwestern University
Boston College University of Notre Dame

University of Cincinnati Pennsylvania State University.

University of Denver University of Southern California
University of Florida Syracuse University
University of Georgia Temple University

Louisiana State University University of Tennessee

University of North Carolina at Chapel Hill

Because each of the universities did not have all
seven departments and because the data from three or four
departmentsvere either incomplete or unusable, the number
of departments included in the study was reduced from one
hundred and five to ninety-seven.

A total of 1,210 usable responses were obtained
from A faculty (faculty included in ranks of assistant
professor and above), representing a response rate of
fifty-one per cent. An additional one hundred and nine
responses were obtained from B faculty (faculty included in
ranks of instructor and below) yielding a total of 1,319

usable responses. The response rates of A faculty ranged,

48

according to discipline, from a low of thirty-five per
cent for mathematics to a high.of sixty-eight per cent
among faculty in departments of management. Among univer-
sities, the response rate for A faculty ranged from a low
of thirty-five-per cent to a high of ninety per cent.

An intensive examination of the correlates of high
response rates revealed that several factors were asso-
ciated with them. For example, response rates were higher
in departments ESE rated in the Cartter report, in private
institutions, in departments indicating a strong orienta-
tion to the university as a reference group as well as.in
departments where the dean or chairmen of special depart-
mental committees were particularly influential. Gener-
ally, response rates were higher in departments placing
greater emphasis upon graduate and undergraduate instruc-
tion while the rate of response was much lower in depart-

ments emphasizing basic research.

Construction of Variables

 

Data for this study were collected in conjunction
with the Esso Departmental Study Project. A number of
findings of the Esso Project have.been published previously

in a volume entitled The Confidence Crisis. Appendix C of

 

this volume contains all of the schedules and forms used in

collecting data for this study as well as the E330 Project.l

 

lDressel, Johnson, Marcus, 9p. cit., pp. 249-62.

 

49

A total of thirty-one variables are examined in
this study. Twenty-seven of these variables were derived
from the responses of faculty members to a "Departmental
Study Inventory," henceforth referred to as the faculty
questionnaire. Three variables were constructed from in-.
formation supplied by each of the departments, henceforth
referred to as the "Basic Data." The remaining variable
was the department's rating in the Cartter report.2 Here,
departments listed as extremely attractive, attractive, or
acceptable plus, in regard to effectiveness of graduate
departments 9£_distinguished, strong, good, or adequate
plus in regard to the effectiveness of graduate faculty
were termed "Cartter-rated departments." Departments.not
listed in regard to either the effectiveness of graduate
departments or of graduate faculty were termed "Cartter
non-rated departments." In order to indicate the precise
manner in which these variables were constructed, the
exact wording of the question from the faculty question-
naire or basic data is given beginning with the items

constructed from the faculty questionnaire.

Faculty Questionnaire Variables

Of the twenty—seven variables constructed from the

faculty questionnaire, twenty variables were constructed

 

2Allan M. Cartter, An Assessment of Quality in
Graduate Education (Washington, D.C.: American Council on
Educatién, I966).

 

50

from individual questionnaire items while seven were group
variables constructed by adding together two or more indi-

vidual variables. The following definitions are offered:

Departmental Autonomy

Faculty members were asked to indicate in general
on a five-point scale ranging from "Very great influence,"
to "Of no influence at all," how much.say or influence the
university administrators had in determining "what goes on.
in your department." Responses of individuals within each
department were summed and a departmental mean was computed
for the influence of the dean and for the influence of the
university administration. These two means were averaged
in order to derive the measure of departmental autonomy.
The means for departmental autonomy were ranked from low
to high for each of the ninety-seven departments in the
study.‘ The lowest mean score received a rank of one and

indicated the greatest amount of departmental autonomy.

Influence

 

The influence of the department faculty as a whole,
the department head or chairman, and individual faculty
members were derived from the responses to this question:
"In general, how much say or influence do each of the fol-
lowing have over what goes on in your department? (a) the
department faculty as a whole, (b) the department head or

chairman, (c) you, personally." This was the same question

51

and the options were the same as those used in assessing
the influence of the dean or the influence of the univer-
sity administration in order to derive a definition of

departmental autonomy. Options were:

(a) Very great influence. (five points)
(b) Great influence (four points)
(c) Some influence (three points)
(d) Slight influence (two points)

(e) Of no influence at all (one point)

Responses of individuals within each department were summed
and a departmental mean was computed for each of the three

individual influence variables.

Emphasis

Variables indicating the degree of emphasis placed.
upon such departmental goals as undergraduate instruction,..
graduate instruction and basic research were constructed.
from responses to the following question: "Within your
department, how much emphasis is placed on each of the
following? (a) undergraduate instruction, (b) instruction
of graduate students, (c) basic research." The options

were

(a) A very great amount (five points)
(b) A great amount (four points)
(c) Some (three points)
(d) A slight amount (two points)
(e) None at all (one point)

Responses of individuals within each department were summed

52

and a departmental mean was computed for each Of the indi-

vidual emphasis variables.

Reference Group Orientation

 

The orientation of individual faculty members to a
university, departmental, or disciplinary reference group
was indicated by the responses to the following
question: "In general, do you usually think Of yourself
primarily as a member Of your: (a) university, (b) de-
partment, or (c) discipline?" The percentage of faculty
members in a given department choosing each of the three
options represented the definition Of each Of these three

reference group orientation variables.

Communication

Faculty members were asked: "To what extent do
you discuss your Opinions and ideas about the department
with each Of the following? (a) the chairman Of your
department, (b) faculty in your department, (c) chairmen
of special department committees, (d) the dean Of your
school or college, (e) graduate students, (f) chairmen Of
other departments or institutes, (9) faculty in other de-
partments or institutes, (h) university administrators
(e.g., President or Vice-President, etc.)." The Options
on a five-point scale were:

(a) TO a very great extent (five points)

(b) To a great extent (four points)
(c) TO some extent (three points)

53

(d) TO a slight extent (two points)

(e) To no extent at all (one point)
Responses Of individuals within each department were summed
and a departmental mean was computed from each of the eight
discussion or communication variables.

In addition, five group communication variables
were constructed:

(1) The amount of total communication was derived by aver-

 

aging the eight individual discussion means noted directly
above.

(2) The amount Of internal communication was derived by

 

averaging the mean amounts Of discussion with (a) the
chairman, (b) the faculty, (c) chairmen of special depart-.
mental committees, and (e) graduate students within the
department.

(3) The amount of external communication was derived by

 

averaging the mean amounts of discussion with (d) the dean
Of the school or college, (f) chairmen of other departments
or institutes, (g) faculty in other departments or insti-
tutes, and (h) university administrators.

(4) The amount Of upward communication was derived by

 

averaging the mean amounts Of discussion with (d) the dean
Of the school or college, and (h).university administrators.

(5) The amount Of outward communication was derived by

 

averaging the mean amounts Of discussion with (f) chairmen

54

Of other departments or institutes, and (9) faculty in

other departments or institutes.

Characteristics

 

1. Joint appointments. Faculty members were
asked "DO you hold a joint appointment?" Their Options
were: (a) Yes, or (b) NO. The variable was derived by
computing the percentage of faculty members in each de-
partment indicating they held joint appointments.

2. Democracy. Faculty members were asked "How
are decisions reached in your department in each Of the
following? (a) recruitment and selection Of new faculty,
(b) promotions and tenure, (c) salary increases, (d) leaves
of absence, (e) travel authorization, (f) budget items for
supplies and equipment, (9) teaching assignments, (h) Of-
fice and research space, (i) award Of assistantships, fel-
lowships, and scholarships, (j) requirements for majors
and graduate students." The options for each Of the ten

instances, on a four-point scale, were as follows:

(a) vote Of all members Of

departmental staff (four points)
(b) vote of all tenured
faculty (three points)

(c) department chairman in

consultation with an

advisory group (two points)
(d) department chairman

acting within estab-

lished policy (one point)

55

Responses Of individuals within each department were summed
and a departmental mean was computed for each of the ten
items. The means Of each of these items were averaged in
order to derive the democracy Of decision making variable.

3. Source Of problem internal. On an Open-ended

 

question, faculty members were asked "What are two or three
major problems or issues which you see in the Operation Of
your department? Are these generated by internal or ex-
ternal factors? How might these be resolved?" Problems
noted were classified in regard tO.(a) nature, (b) source,
and (c) resolution. This particular variable was derived
for each department by noting the percentage Of faculty
members within a given department who noted one or more
problems which were classified by source as internal to or
controllable by the department.

4. Source Of problem external. This variable was

 

constructed in the same manner as (3.) directly above from
the percentage Of faculty members with each department who
noted one or more problems classified by source as exter-

nal to, or not controllable by the department.

Basic Data Variables

Size of A Faculty

 

Departments were asked to answer the following

question on the Basic Data collection forms: "In Fall

56

1967, what was the total number Of persons at each rank
affiliated with this department?" The Size of A Faculty
variable was derived by summing the entries for the ranks
Of professor, associate professor, and assistant professor.
Eighty-nine Of the ninety-seven departments were able to

supply this information.

Ratio Of Research to General
Fund’Dollars

 

 

Departments were asked to answer the following
question on the Basic Data collection forms for the year
1966-67 with reference to the expenditure of (1) general
funds and (2)grants and contract research funds: "What
was the total amount Of money expended for each Of the
following:

(1) Personnel

a. Academic staff

b. Clerical and technical staff

c. Fellowships, graduate assistants

d. Number of federal traineeships
(2) Operating Funds

a. Supplies, services, materials

b. New equipment

(3) Total?"

The ratio Of research to general funds variable was then
derived by dividing the total amount Of dollars expended
from grants and contract research funds in 1966-67 by the
total amount Of dollars expended from general fund dollars.
Complete data for this variable were available for only

seventy Of the ninety-seven departments. Thus approx-

imately one-third of the responses Obtained in the faculty

57

questionnaire were not usable when $ Res/S GF was used as

a control variable or test factor.

Publication Rate

Departments, in providing Basic Data, were asked
to respond to the following question: "How many different
members Of this department have published how many schol-

arly works in the past three years?"

a. textbooks were published by members

b. monographs were published by
members
c. advanced research treatises or research

articles were published by members.

The total number of publications Of all types was summed
and divided by the number Of A Faculty within the depart-
ment in order to form a very rough publication rate or
index. This information was supplied by only seventy-one.
Of the ninety-seven departments with a proportional reduc-
tion Of usable responses when rate of publications was

used as a control variable or test factor.

Analysis of Data

 

Data were analyzed through three primary means
including the use Of (l) the process Of elaboration, (2)
Spearman rank order correlation coefficients and (3) anal-

ysis of variance.

58

Elaboration

 

As explicated by Rosenberg in his volume, The Logic

 

Of Survey Analysis, the process of elaboration involves

 

the introduction Of a third variable into the analysis Of
a two variable relationship in order tO "explain" or
"specify" the relationship thus making it more exact.3
Elaboration is intended to aid in answering the questions
of "why" and "under what circumstances" a particular rela-
tionship holds true.

Thus, elaboration is employed in an attempt to
determine the existence or lack Of causal relationships.
It must be noted, however, that there is no way with
survey data that the analyst can guarantee that a rela-
tionship is causal. The best he can do, as Glock notes,
is to introduce all the third variables or test factors
which might explain the relationship and examine the re-
sults.4 If the relationship is replicated once a test
factor or third variable is introduced, the probability is
increased that the relationship is causal. If, however,

the relationship vanishes or is explained away when a

third test factor is introduced, the relationship would

 

3Morris Rosenberg, The Loglc of Survey Analysis
(New York: Basic Books, Inc., 1968), p. 20.

4Charles Y. Glock, "Survey Design and Analysis in
Sociology," in Surveleesearch in the Social Sciences, ed.
by Charles Y. Glock (New YOfk: RusseIl Sage Foundation,
1967), p. 21.

59

have to be regarded as spurious. Glock notes:
It is always possible, however, that a test factor
exists which would explain away the relation, had the.
analyst had the knowledge and foresight to collect
data on it and to introduce it into his analysis.
Thus, the survey analyst, as analysts Of any other
form Of data, can never conclude absolutely that two
variables are indeed causally related.5
Data were analyzed in this study making use Of the
elaboration process. Departmental autonomy was regarded
as an independent variable with the various influence,
emphasis, reference group, communication and some depart-.
mental characteristic variables regarded as dependent
variables. The investigator's concern was focused upon .
the consequences which followed from the distribution.on
an independent variable (departmental autonomy) rather
than the effects a number of independent variables have on
a particular dependent variable. Glock referred to this
type of study as an implications study.6 Here, as with
all studies incorporating the process of elaboration, the
probability that a relationship is causal is increased
when the relationship remains rather than vanishes once
theoretically important test factors have been introduced.
In order to determine whether a significant rela-

tionship existed between the independent variable (depart-

mental autonomy) and the dependent variables, data were

 

51bid., p. 21.

6Ibid., p. 37.

6O

analyzed through use of the chi square technique for de-
termining statistical significance. All Of the individual
dependent variables in regard to.influence, emphasis, ref-
erence group orientation, and communication had five pos-
sible response categories. The number Of response cate-
gories was collapsed to two, three, and in some instances
four categories as long as each Of the collapsed categories
contained a minimum of approximately ten.per cent of the
total number Of responses to an item.

The overall two variable relationships, noting.the
chi square value Obtained for the relationship between
departmental autonomy and individual dependent variables,
are noted in the tables appearing in Chapter IV. If the
overall relationship between departmental autonomy and the
dependent variable was significant, twenty-four test fac-
tors were introduced into the analysis in order to indicate.
the conditions under which the significant relationship
either held or vanished or the conditions under which
another significant sub-pattern of variation appeared.. If.
the overall relationship between departmental autonomy and
a dependent variable was not significant, the various.test
factors were introduced in order to indicate the conditions
under which significant relationships emerged. The twenty-
four test factors were:

Cartter Rating Reference University

Support Reference Department

61

Size A Faculty Reference Discipline

Democracy Discussion Department Head

$ Res./$ General Fund Discussion Department Faculty

Publication Rate Discussion Dean

Influence Of Department Discussion University
Faculty Administration

Influence Of Department Total Communication
Head

Influence Of Individual External Communication
Faculty

Emphasis on Undergraduate Upward Communication
Instruction

Emphasis on Graduate Outward Communication
Instruction

Emphasis on Basic Research Internal Communication

Tables A-l to A-ll in the Appendix, which are re-
ferred to in Chapter IV, indicate whether an overall rela-
tionship between departmental autonomy and a particular
dependent variable emerged, was replicated, explained away,
or altered for each category Of the twenty-four test fac-
tors. Data were analyzed through the use Of the NUCROS
program developed by the Computer Institute for Social
Science Research at Michigan State University. The Program
produces cross-classifications, also called cross-tabula-
tions, analysis Of crossbreaks or contingency tables in
two, three, or four dimensions.

Departments were ranked high, medium or low in
regard to departmental autonomy which was regarded as the
independent variable. The individual influence, emphasis

reference group, communication, and departmental

62

characteristics were regarded as dependent variables. The
NUCROS program produced separate chi square tables for
each category Of the test factor. Thus, for example, the
relationship between departmental autonomy and the influ-
ence Of the department chairman was indicated in two
separate chi square tables--for departments rated in the
Cartter report and for departments not rated in the Cartter
report. Separate chi square tables were also produced for
departments in public supported institutions and for de-
partments in private supported institutions. Three sepa-
rate chi square tables indicated the relationship between
the degree Of departmental autonomy and the influence of
the department chairman for each of the three (high, med-
ium and low) categories for each Of the remaining twenty-
two test factors. Thus total number of contingent rela-
tionships and separate chi square values produced was
seventy (3 x 22 + 2 x 2).

If, for example, an inverse relationship was hy-
pothesized between the degree of departmental autonomy
(high, medium or low) and the influence Of the department
chairman (very great; great; some,sdight or none), each Of
the seventy conditional relationships was examined to see
if the percentage Of faculty members attributing a very
great amount Of influence to the department chairman (1)
exhibited significant variation and (2) decreased as au-

tonomy increased. If the above conditions were met or if

63

the percentage of faculty members attributing some, slight
or no influence increased as department autonomy increased,
the overall relationship was viewed as being supported.
Tables A-l - A-ll indicate only the instances, out of sev-
enty possible conditional relationships, when there was a
significant relationship between the degree of departmental
autonomy and the individual dependent variable as indicated

by a significant chi square value.

Spearman Rank Order Correlation
Coefficients

 

 

Whereas in the process Of elaboration a significant
chi square indicated that there was significant variation
in the particular sub-sample considered, it did not indi-
cate the magnitude of any existing linear relationships.
Spearman rank order correlation coefficients were more
appropriate for this purpose.

By averaging the responses for all members Of a
department to most Of the individual department variables,
by computing the percentage Of faculty members within a
department holding a particular reference group orienta-
tion, or by computing the percentage Of faculty members in
each department indicating one or more problems classified
by source as either internal or external to the department,
each Of the departments could be ranked on each variable

in regard to each Of the other departments.

64

Because it was possible to Obtain an ordinal rank-
ing Of the departments for each variable, data were anal-
yzed through the use Of the Spearman rank order correlation
coefficient. When data were missing for particular vari-
ables, the number Of pairs correlated was reduced. When

ranks were tied, they were broken randomly.

Analysis of Variance

 

A simple one-way analysis Of variance was employed
in order to Obtain mean values, on a five-point scale, for
each Of the individual influence, emphasis, reference
group, and discussion variables as well as for the five
group communication variables. This was necessary because,
although the chi square analysis was valuable in noting
the distribution at either end Of the five-point scale,
variation in the middle categories could be great enough
to convey a misleading picture if only differences in the
extreme categories were reported.

Through the use of Scheffe's method Of multiple
comparisons individual means were compared to determine
whether or not they were significantly different from each
other. It was thus possible, for example, to compare the
mean amounts Of discussion of departmental affairs with
the department chairman in departments Of high, medium,

and low autonomy.

65

Data were also analyzed through the use of a re-
peated measures analysis of variance. This procedure
allows one to compare the means of a number of items
answered by the same individuals, thus enabling one, for
example, to note in departments of high, medium or low
autonomy whether discussion Of departmental matters was
significantly greater with the department chairman, de—
partmental faculty as a whole, or the chairmen of special
departmental committees. Contrasts Of individual means
were, again, made through the use of Scheffe's method of

multiple comparisons.

CHAPTER IV

ANALYSIS OF DATA

Introduction

 

In attempting to construct a portrait of university
departments with varying degrees of autonomy, departmental
autonomy was regarded as an independent variable. As pre-
viously noted, several areas of interest were examined.
These areas and the dependent variables considered for
each in regard to departmental autonomy were as follows:

1. Departmental Characteristics including (a) Cartter
report rating, (b) the percentage of faculty in a given
department holding joint appointments in other departments,
(c) the percentage of faculty members in a given department
indicating one or more problems classified by source as
internal, or within the control Of the department, (d) the
percentage of faculty members in a given department indi-
cating one or more problems classified by source as exter-
nal or beyond the control of the department.

2. The influence of various actors within the depart-
mgpp§_including (a) the department chairman, (b) the de-
partmental faculty as a whole, and (c) the individual

faculty member.

66

67

3. The emphasis placed upon various departmental goals

 

including (a) undergraduate instruction, (b) graduate in-
struction, and (c) basic research.
4. The reference group orientation of individual fac-

ulty members to (a) the university, (b) the department,

 

and (c) the discipline.
5. The communication system including communication

with such individuals and elements as the (a) dean, (b) de-

partment chairman, (c) faculty within one's own department,
(d) chairmen of special departmental committees, (e) univer-
sity administration, (f) graduate students, (g) faculty in
other departments, (h) chairmen Of other departments or
institutes. In addition, several group communication
variables were constructed through the combination Of
individual communication variables. These group variables
were: (a) total communication representing the mean for
all eight individual variables; (b) external communication
representing the mean for four variables including discus-
sion with the dean, university administration, faculty in
other departments, and chairmen of other departments or
institutes; (c) upward communication representing the mean
for discussion with university administration and discus-
sion with the dean; (d) outward communication representing
the mean for discussion with faculty in other departments
and discussion with chairmen Of other departments or in-

stitutes; (e) internal communication representing the mean

68

for four variables including discussion with the department
chairman, faculty members within the department, chairmen
Of special departmental committees, and graduate students

within the department.

In investigating the relationship between depart-
mental autonomy (regarded as the independent variable) and
the individual dependent variables, the following proced-
ures were followed:

1. An hypothesis was stated.

2. Data tending to support or reject the hypothesized
relationship from a chi square analysis and analysis Of
variance were noted. The Spearman rank order correlation
between departmental autonomy and a particular variable
was noted for the total group Of ninety-seven departments.
Spearman rank order correlations were also noted for var-
ious subgroups of the total sample. These correlations
were noted for departments characterized as having (a)
high, (b) medium, or (c) low autonomy and departments
which were (d) rated or (e) not rated in the Cartter
report.

3. Test factors, twenty-four in all, were introduced
in order to examine the conditional relationships which

emerged between the independent variable (departmental

69

autonomy) and individual dependent variables. Data were
then analyzed in the manner described in Chapter III.

4. The range of responses, giving the high and low
percentages Of faculty within a given department indicating
a particular response, was given.

Table 1 presents a summary Of the Spearman rank
order correlations between the degree of departmental au-
tonomy and (1) the test factors noted above and (2) all
dependent variables. These correlations are noted for
faculty in all departments, faculty in departments charac-
terized by high, medium, or low autonomy, and faculty in

departments rated or not rated in the Cartter report.

Departmental Characteristics

 

Cartter Ratipg

 

Hypothesis.--The relationship between departmental
autonomy and a favorable rating in the Cartter report will
be curvilinear. Departments with a medium degree Of au-
tonomy will be rated in the Cartter report more frequently
than departments with a high or low degree of autonomy.

Table 2 indicates that the hypothesized relation-
ship was not supported by the data. Although the findings
of Pelz and Andrews may be broadly construed as indicat-
ing that productivity, largely in research laboratories,
was highest in settings with moderate coordination, medium

autonomy was not associated with high Cartter ratings.

7O

 

¢N.+ mo.| smm.t va.l «mm.| «sh~.| .pmoa Hoeuo muasomm cowmmsomfin
50.: HN.+ mH.I om.| mo.+ mo.| mucmpsum Omuw cOflmmsomHQ
o~.u -.+ H~.I mo.+ oo.n mo.+ .ssoo .nmmo pmmm aoemmsomso
m~.u sa.+ tom.u sa.+ mo.- mo.+ puma .uamo cosmmsomeo
mo.u ma.+ mo.- mo.- so.n mo.- seasons .pmma aoemmsomna
Ho.l vo.+ emm.| OH.I «mm.| semm.t coaumOHcsEEoo pum3uso
mo.+ mo.+ vo.| mo.+ hH.I v0.1 coameflcsEEOU Onmzma
HH.+ no.+ vm.| mo.| NN.I OH.I coauMOHcsEEOU Hmcuouxm
mm.| mN.+ hN.I Ho.| mo.+ mo.+ cOHDOOHCSEEOU HmcHoucH
OH.I mH.+ som.| HH.I wo.l mo.l coaumowcsfieoo Hmuoe
mm.| mH.+ smm.+ oa.+ HH.+ eH.+ ocflamflomwo mom
4cm.+ H~.I 4mm.u 4H.I no.u ma.u pamspnmmma mom
ma.l HH.+ NH.+ mo.+ mo.: mo.| wuflmuo>aca mom
ma.| mo.+ mo.| no.1 hH.+ «mH.+ noumomom OHmmm mammnmam
oa.u «H.+ HH.+ 44mm.+ no.+ «mm.+ .umaH mumspmuo memmnmsm
no.u mo.u mo.+ Ho.u oa.u ma.u zoeuosuumaH pmuma memmnmsm
Hm.| mH.+ mm.| smm.+ OH.| «H.+ oocmuscH Hospfl>flch
wo.+ «mm.+ Ho.| «hm.+ seeme.+ «semm.+ huasomm .umoo mo mocosHmcH
«om.l ¢H.I «ssmm.l «esmv.l ««#m©.l sssmm.l Ummm .flmmD MO OOGODHMCH
oo.| mo.+ mm.| mo.+ hH.I Ho.+ Hmcuouxm aneoum mo oousom
mo.+ «mm.| mH.+ vH.I no.+ mH.| accumucH anooum mo oonsom
mo.| om.+ mo.| mo.+ mo.+ NH.+ oumm GOHOMOHHQOm
mm.n eo.: mm.+ Hm.u ma.+ mo.+ page Hmnmcmo m\soumomom m
OH.I mH.+ mo.+ hH.+ «sov.+ ssemm.+ momHOOEOO
NH.I oa.+ ~o.u mH.I mm.u oa.u auasomm 4 mo muem
«Hm.l no.1 mo.+ om.| mo.| ma.t mucoEucflommd “neon
maocousfi hfiocousa wfiocous< .mumoo .mpmoo mucoauummon manmflum>
30a Edwpoz swam poummlcoz poumm Hafi

Houuumu Hmuunmo

 

 

.mmsoumnsm msowum> can mamfimm Hmuou Mom moanmflnm> ucmpcomop new muouomm ummu
msowum> 0cm meocousm Hmucoauummop coozuon mcowumHonHOo HOOHO xcmn coaumommII.H OHQOB

 

7l

 

Hoo. W m«««
Ho. W mes
mo W as
em.n so.+ Ho.+ mo.u HH.I Ha.u .sae muemum>ncp aoemmsomeo
ma.+ mo.+ eo.| ma.+ mH.I Ho.+ coma coammsomfla
mo.n ~H.+ -.I ma.n 40.: mH.I .ummo nonuo pawn aoemmsomfia
meocousm meocousm mEOcousfl .mumoo .memoo mucofiuummoa manmflum>

sou sweep: seem pmpmmuaoz pmpmm Ham
kuuumo Hmueumu

 

 

.OODGHDGOUII.H manna

72

Instead, a direct relationship between departmental au-
tonomy and a favorable Cartter rating was suggested al-
though the relationship was not statistically significant.

Table 2.--Re1ationship between Cartter report ratings and
faculty perception of departmental autonomy.

 

 

Departmental Autonomy

 

 

 

Cartter Ratings High Medium Low Total N
Rated 51.5 37.5 28.1 39.1 38
Not Rated 48.5 62.5 71.9 60.8 59

100.0 100.0 100.0 160.0 97
N 33 32 32

x2 = 3.78

df = 2

p > .05

JointAppOintments

 

Hypothesis.--The percentage of faculty members

 

within a department holding joint appointments will be
inversely related to the degree of departmental autonomy.
Data presented in Table 3 tend to be in the direc-
tion Of the stated hypothesis although the differences were
not statistically significant. Table A-1 and all others
also located in Appendix A are designed to indicate how a
relationship between departmental autonomy, the independent
variable, and a particular dependent variable, in this case,
joint appointments, is modified when the various test fac-

tors are introduced as control variables. The overall

73

relationship between the degree Of departmental autonomy

and the frequency Of joint appointments, as noted in Table

3 was not significant.

Table 3.--Relationship between percentage Of faculty members

holding joint appointments in departments and
faculty perception Of departmental autonomy.

 

 

 

 

 

 

Joint Departmental Autonomy
Appointments High Medium Low Total N
present 9.6 12.2 13.3 '11.7 127
Not Present 90.4 87.8 86.7 88.3 957
100.0 100.0 100.0 100.0 I084
N 363 353 368
x2 = 2.49
df = 2
p > .05

Column one of Table A-1, however, indicates the
conditions under which a significant relationship, which
was not present in the overall relationship, emerged.
Column two indicates the conditions under which there was
a significant amount of variation in the relationship be—
tween departmental autonomy and the percentage of faculty
members holding joint appointments, with joint appointments
being more frequent in departments characterized by medium
rather than high or low autonomy. Column three indicates
the conditions in which there was a significant amount Of
variation in the relationship between departmental autonomy

and the percentage Of faculty members holding joint

74

appointments, with joint appointments being least frequent
in departments characterized by a low rather than a high
or medium degree of autonomy.

Thus, for example, Table A-1 indicates that in
private supported institutions, there was an inverse rela-
tionship between the degree of departmental autonomy and
the percentage of joint appointments which was significant
at the .01 level Of significance. However, no such sig-
nificant relationship emerged between departmental autono-
my and the percentage Of faculty with joint appointments
within public supported institutions.

The most common pattern noted in Table A-1 was for
the frequency Of joint appointments to be greatest in de-
partments with medium autonomy. However, in departments
where faculty influence was low, in departments which were
large or placed strong emphasis upon basic research, and
were also characterized by medium autonomy, a smaller per-
centage of faculty with joint appointments existed than in
departments with high or low autonomy. Thus joint ap-
pointments were much more frequent in departments with
medium.autonomy and high total, external and outward com-
mmnication particularly with the deans and university
administration. Joint appointments were also more common
in departments characterized by a low degree of democracy
and influence Of the departmental faculty as a whole in

combination with a medium degree of autonomy. The

75

initiative for making joint appointments in departments
with medium autonomy thus appears as if it may lie outside
Of the department. The highest incidence of faculty with
joint appointments in a department (30.8%) was in a de-
partment where a medium degree Of autonomy and a high
degree of influence attributed to the university adminis-
tration were combined. No joint appointments were report-
ed in departments where high autonomy was also associated
with a low emphasis upon basic research, low influence
attributed to the departmental faculty as a whole, or low

democracy in departmental decision making.

Internal Problems

 

Hypothesis.--The percentage Of faculty members in

 

a given department indicating one or more problems clas-

sified by source as internal or under the control Of the

department, will be inversely related to the degree of

departmental autonomy.

Table 4.--Re1ationship between percentage Of faculty mem-
bers in departments noting one or more problems
classified by source as internal to the depart-

ment and faculty perception Of departmental
autonomy.

 

 

Source Of

 

 

 

Problem Departmental Autonomy
Internal High Medium Low Total N
Indicated 60.5 59.2 66.9 62.3 693
Not Indicated 39.5 40.8 33.1 37.7 419
100.0 100.0 100.0 100.0 1112
N 367 355 390
x2 = 5.56
df = 2

p > .05

76

An examination Of the data in Table 4 reveals
that the hypothesis was not confirmed. There was, how-
ever, a tendency for low autonomy to be associated with a
high percentage Of faculty members within given depart-
ments indicating the existence Of problems classified by
source as internal, a trend which approached but did not
reach statistical significance. Few notable relationships
emerged when various test factors were introduced into
the analysis. Generally, the lowest percentages Of fac-
ulty indicating internal problems were in departments
with medium autonomy while faculty in departments with
low autonomy were most likely tO indicate problems clas-
sified by source as internal. This was the case in de-
partments also characterized by private support, low
publication rates, low external communication and in
departments attributing high influence to the faculty,
individually and collectively, high reference group
orientation to the university and high amounts of discus-
sion with either the dean or departmental faculty. Med-
ium autonomy thus appears to be consistently related to
fewer problems which could be controlled by departmental
action.

The percentage Of faculty members noting one or
more problems classified by source as internal ranged
from a high Of 77.7 per cent in departments characterized

by a low degree Of autonomy and high discussion of

77

departmental affairs among the departmental faculty as a
whole to a low of 43.1 per cent in departments character-
ized by highly influential faculty and a medium degree Of

departmental autonomy.

External Problems

 

Hypothesis.--The percentage of faculty members in
a given department indicating one or more problems clas-
sified by source as external to or beyond the control Of
the department will be directly related to the degree of
departmental autonomy.
Table 5.--Relationship between percentage Of faculty mem-
bers in departments noting one or more problems
classified by source as external tO the depart-

ment and faculty perception of departmental
autonomy.

 

 

Source Of

 

 

 

Problem Departmental Autonomy
External High Medium Low Total N
Indicated 64.0 66.5 61.8 64.0 712
Not Indicated 36.0 33.5 38.2 36.0 400
_1oo.o 100.0 —“0'100. —T100. III?
N 367 355 390
x2 = 1.77
df = 2
p > .05

As indicated in Table 5, there was only a very
slight overall relationship between the percentage Of

faculty members noting one or more problems classified by

78

source as external, or beyond the control of the depart-
ment, and the degree Of departmental autonomy. Only nine
significant relationships emerged when the twenty-four
test factors were introduced into the analysis and none Of
the significant relationships supported the hypothesized
relationship. Instead, heightened awareness Of problems
with an external source was most evident in departments
characterized by a medium degree of autonomy while the
fewest external problems were in departments with low
autonomy.

High perception of problems caused by sources
beyond the control Of the department was particularly
evident in departments which were rated in the Cartter
report or departments which placed low emphasis upon un-
dergraduate instruction. Cartter-rated departments are
more likely than non-rated departments to seek and receive
research funds or money for improved facilities or addi-
tional faculty, all Of which are not directly controllable
by the department. Thus it is not surprising that such
departments would have a larger percentage of problems
classified by source as external.

Perception Of externally caused problems was also
associated with departments characterized by high total
and external communication, particularly with the dean or
university administration. Thus these consultative pat-

terns, presumably, are viewed by the departmental faculty

79

as constituting interference with departmental prerogatives
rather than providing administrative support. The per-
centage of faculty members noting one or more problems
classified by source as external ranged from a high of

76.7 per cent in departments characterized by high total
communication and medium autonomy to a low of 50.0 per

cent in departments with a medium orientation to the de-
partment as a reference group and a high degree of

autonomy.

Influence of Various Actors

 

Influence of Department Chairman

 

Hypothesis.--The influence of the department

 

chairman in departmental affairs will be inversely related
to the degree of departmental autonomy.
Table 6.--Re1ationship between faculty perception Of the

influence of the department chairman and faculty
perception Of departmental autonomy.

 

 

Influence of

 

 

 

Department Departmental Autonomy
Chairman High Medium Low Total N
Very Great 33.4 52.5 66.3 50.9 624
Great 43.6 40.3 28.4 37.3 457
Some, Slight or
None 23.0 7.2 5.3 12.8 144
100.0 100.0 100.0 150.0 I225
N 404 402 419
2
X = 123.93***
df = 4
p é .001

80

As noted in Table 6, a strong inverse relationship
was found to exist between the perceived influence Of the
department chairman and the degree of departmental autono-
my, thus lending strong support tO the hypothesis. The
Spearman rank order correlation coefficients between the
influence of the department chairman and the degree of
departmental autonomy for all departments and for all sub-
groups, except for departments of medium autonomy were,
for the most part, highly significant. The strongest
relationship was evident among departments rated favorably
in the Cartter report (rS = -.68). Furthermore, the mean
amounts Of influence, on a five point scale, decreased as
one moved from conditions Of low (4.60) to medium (4.45)
and from medium to high (4.07) autonomy, with all differ-
ences being highly significant.

Table A-2 illustrates how the overall inverse sig-
nificant relationship between departmental autonomy and
the dependent variable noted in Table 6 was altered when
the twenty-four test factors were introduced into the
analysis. The original inverse relationship held, as
noted in column one, in sixty-five of the seventy condi-
tional relationships examined.

Column two indicates, among other things, that the
overall significant inverse relationship between the
degree Of departmental autonomy and the influence Of the

department chairman vanished in departments placing a

81

medium degree Of emphasis upon graduate instruction. The
inverse relationship held, however, at the .001 level Of
significance in departments placing either a high or low
emphasis upon graduate instruction. Additional modifica-
tions Of the overall relationship were evident only in
departments ranked high or medium in regard to the ratio
Of research to general fund dollars, in departments where
the total amount Of communication was high and in depart-
ments where the influence of the department chairman was
ranked medium. In these instances the significant rela-
tionship between the degree Of departmental autonomy and
the influence Of the department chairman vanished. There
were no instances, as revealed by column three Of Table
A-2, in which there was significant variation between the
degree Of departmental autonomy and the influence Of the
department chairman which were not in keeping with the
overall inverse relationship between these two variables.
The percentage Of faculty members attributing a
very great amount of influence to the department chairman
ranged from a high Of 77.4 per cent in departments with
low upward communication and low autonomy tO a low Of 18.6
per cent in departments with a high ratio of research dol-

lars to general fund dollars and high autonomy.

82

Influence Of Departmental Faculty
as a Whole

Hypothesis.--The influence of the departmental

 

faculty as a whole in departmental affairs will be directly
related tO the degree of departmental autonomy.
Table 7.--Re1ationship between faculty perception Of the

influence of the departmental faculty as a whole
and faculty perception of departmental autonomy.

 

 

Influence Of

 

 

 

 

Departmental Departmental Autonomy
Faculty High Medium Low Total N
Very Great 33.9 25.9 13.8 24.4 301
Great 35.9 32.1 30.0 32.6 402
Some 22.4 31.1 33.8 29.1 359
Slight or None 7.9 10.9 22.4 13.9 171
100.0 100.0 I00.0 100.0 1233

N 408 405 402

2
X = 81.34***
df = 6
p .5: .001

Table 7 indicates a strong, direct, highly signif-
icant relationship between the degree of departmental au-
tonomy and the influence of the departmental faculty as a
whole. The relationship was strongest'(rS = +.49) in de-
partments which were rated favorably in the Cartter report.
As one moved from conditions Of low (3.29) to medium
(3.70) and from medium to high (3.95) autonomy the mean
amounts Of influence attributed to the departmental faculty

on a five point scale, increased significantly. Under the

83

condition Of high autonomy the relative influence of the
faculty approached, but did not equal that of the depart-
ment chairman (4.07) in departmental affairs.

The relationship between the perceived influence
Of the departmental faculty as a whole and the perceived
degree of departmental autonomy, although strong, was not,
however, as strong as the relationship between autonomy
and the influence Of the department chairman. Thus Table
A-3 reveals that the hypothesized relationship held in
fifty-one Of the seventy as compared to sixty-five of
seventy Of the conditional relationships noted. The degree
Of democracy in departmental decision making and the in-
fluence Of the university administration were particularly
important test factors. Thus the significant relationship
between the influence Of the departmental faculty as a
whole and the degree of departmental autonomy vanished in
departments characterized as either high or low in regard
to these test factors. Under the conditions Of high pub-
lication rates, high orientation to the university as a
reference group, and low discussion with university admin-
istrators the direct relationship between departmental
autonomy and the influence of the departmental faculty as
a whole also vanished.

The third column of Table A-3 indicates seven con-
ditions under which the one significant sub-pattern of

variation emerged: the tendency for faculty in departments

84

with a medium degree of autonomy to regard themselves as
more influential than faculty in departments with high or
low autonomy. Here, under such conditions as a weak de-
partment chairman, highly influential individual faculty
members, and high outward communication, true collegial
power is likely to exist. The percentage of faculty mem-
bers attributing a very great amount of influence to the
departmental faculty as a whole ranged from a high Of 49.6
per cent when discussion with departmental faculty and
departmental autonomy were high to a low Of 8.3 per cent
when discussion with departmental faculty and departmental

autonomy were low.

Influence Of Individual Faculty
Members

 

Hypothesis.--The influence of the individual fac-
ulty member in departmental affairs will be directly re-

lated tO the degree Of departmental autonomy.

Table 8.--Re1ationship between faculty perception Of per-
sonal influence and faculty perception Of depart-
mental autonomy.

 

 

Personal Influence Departmental Autonomy
Of Faculty Member High Medium Low Total N

 

Very Great, Great

 

 

 

or Some 58.9 57.5 48.9 55.0 671
Slight 34.7 33.4 31.2 33.1 403
None 6.4 9.0 19.9 11.9 145

100.0 100.0 IEUTU' IUETJ III?

N 404 398 417
x2 = 40.49***
df = 4

<
p = .001

85

While there was a suggestion Of a direct relation-
ship between the personal influence Of faculty members and
the degree of departmental autonomy, the main factor con-
tributing to the large chi square value noted in Table 8
was the relatively low percentage Of faculty members (48.9%)
in departments with low autonomy who felt that they had at
least some influence in departmental affairs. Thus the
mean amount of influence attributed to individual faculty
members, on a five point scale, was significantly lower in
departments Of low autonomy (2.39), than that Of faculty in
departments Of medium (2.60), or high (2.64) autonomy, where
the relative influence Of individual faculty members was
essentially the same. The Spearman rank order correlation
coefficients, except for those departments which were not
rated in the Cartter report (rS = +.26), were not statis-
tically significant.

A number Of modifications Of the overall direct
relationship between personal influence and departmental
autonomy occurred when the twenty-four test factors noted
in Table A-4 were introduced into the analysis. The hypo-
thesized relationship held in exactly 50 per cent (35) of
the cases while it vanished in twenty-nine cases and ex-
hibited other significant patterns Of variation in nine
other instances.

The overall direct relationship between personal

influence and departmental autonomy vanished, most notably

86

in departments with high rankings on all of the group com-
munications variables. The relationship also vanished in
Cartter-rated departments and in departments characterized
by high or medium amounts of democracy in decision making
and high or medium percentages Of research dollars from
outside sources. Taken together, these findings suggest a
highly competent community of scholars where differences in
the perception Of one's personal influence are slight be-
cause the degree Of personal influence exerted is uniformly
high.

A consistent sub-pattern Of variation which occurred.
in each Of nine instances when there was significant varia-
tion from the overall pattern was for one's perception Of
his own personal influence to be highest in departments
with a medium degree Of autonomy and lowest in departments
with low autonomy. Thus, there was not a single instance
among the forty-seven significant relationships noted when
low departmental autonomy was not associated with a lower
perception Of personal influence than was the case in de-
partments characterized by a medium or high degree Of
departmental autonomy.

The percentage Of faculty members attributing a
very great, great, or some amount of influence to themselves
as individuals ranged from a high Of 73.4 per cent when the
influence Of the departmental faculty as a whole was high

and was combined with a medium degree Of autonomy to a low

87

Of 36.2 per cent when the total amount of communication and

departmental autonomy were low.

Emphasis Upon Departmental Goals

 

We now turn to the examination of the relationship
between the degree of departmental autonomy and the degree
Of emphasis placed upon three primary departmental goals:
(1) undergraduate instruction, (2) graduate instruction,
and (3) basic research. Here, in general, as indicated by
the mean amount of emphasis attributed to the various de-
partmental goals, on a five point scale, the faculty in our
sample placed a significantly higher emphasis upon graduate
instruction (4.14) as Opposed to basic research (3.74) or
undergraduate instruction (3.71). Under conditions Of high
and medium autonomy the relative emphasis placed upon the
three departmental goals remained the same as stated above.
Under the condition of low departmental autonomy, emphasis
upon graduate instruction (4.01) was also clearly strongest.
However, emphasis upon undergraduate instruction (3.74) was
somewhat higher than the emphasis placed upon basic research
in departments with low autonomy (3.56), although the dif-

ference was not statistically significant.

Emphasis Uppn Undergraduate Instruction

 

Hypothesis.--Emphasis upon undergraduate instruction
as a departmental goal will be inversely related to the

degree of departmental autonomy.

88

Table 9.—-Relationship between faculty perception Of empha-
sis placed upon undergraduate instruction and
faculty perception of departmental autonomy.

 

 

Emphasis Upon

 

 

Undergraduate Departmental Autonomy
Instruction High Medium Low Total N

Very Great 14.3 21.0 20.9 18.8 238

Great 45.6 45.8 43.4 44.9 570

Some, Slight or None 40.1 33.3 35.6 36.3 461
105.5 100.5 155.5 I50.5 I262

N 419 415 435

x2 = 9.66*

df = 4

p ; .05

Although there was significant variation in the
overall relationship between the degree of emphasis placed
upon undergraduate instruction and the degree Of depart-
mental autonomy, the relationship as evidenced by the lack
of significant Spearman rank order correlation coeffi-
cients, was not linear. Nor were the mean emphases placed
upon undergraduate instruction under conditions Of high
(3.63), medium (3.76) and low autonomy (3.74) significantly
different from each other. The hypothesis was, therefore,
not confirmed. The most significant tendency indicated in
Table 9 was for the faculty in departments with high au-
tonomy to place a relatively low emphasis upon undergrad-
uate instruction.

As the overall relationship between the degree of

emphasis placed upon undergraduate instruction and the

89

degree of departmental autonomy was one of borderline sig-
nificance, it was not surprising that significant rela-
tionships were evident in only twenty-nine of the seventy
conditional relationships noted in Table A-5. The trend
for faculty in departments characterized by high autonomy
to indicate less emphasis upon undergraduate instruction
than faculty in departments with medium or low autonomy,
however, held for twenty-seven of the twenty-nine signif-
icant relationships.

The percentage of faculty members indicating that a
very great amount of emphasis was placed upon undergraduate
instruction ranged from a high of 46.4 per cent in depart-
ments where democracy was high and autonomy was low to a
low Of 5.1 per cent in departments when a low amount Of
discussion was coupled with a high degree of departmental
autonomy. Because the overall relationship between
departmental autonomy and the degree Of emphasis placed on
undergraduate instruction was tenuous at best, further
attempts to interpret the data might only lead to a mis-

representation Of the actual situation.

Emphasis Upon Graduate Instruction

 

Hypothesis.--Emphasis upon graduate instruction as

 

a departmental goal will be directly related to the degree

of departmental autonomy.

90

Table lO.--Relationship between faculty perception of
emphasis placed upon graduate instruction and
faculty perception Of departmental autonomy.

 

 

 

 

 

 

Emphasis Upon Departmental Autonomy

Graduate Instruction High Medium Low Total N

Very Great 36.5 35.9 26.3 32.8 412

Great 51.2 50.5 54.8 52.2 655

Some, Slight or None 12.3 13.6 18.9 15.0 188
100.0 100.0 100.5 100.0 1255

N 414 412 429

x2 = 16.08**

df = 4

p § .01

The data in Table 10 tend to support the hypothes-
ized relationship between the degree of emphasis placed
upon graduate instruction and the degree Of departmental
autonomy. Most notable, however, was the tendency for
faculty in departments with a low degree of autonomy to
attribute a low emphasis to graduate instruction as a de-
partmental goal. Thus the mean amount Of emphasis placed
upon graduate instruction under the condition Of low de-
partmental autonomy (4.01), was significantly lower than
the means for departments characterized by medium (4.19),
or high (4.23) autonomy.

The Spearman rank order correlation coefficient,
between the degree Of emphasis upon graduate instruction
and the degree Of departmental autonomy, as noted in Table

1,were significant for the entire sample of ninety-seven

91

departments (rS = +.23) and particularly for departments
not rated in the Cartter report (rS = +.33). However, the
direct significant relationship held, as noted in Table
A-6, for only seventeen of the seventy conditional rela-
tionships examined. Generally, departments placing high
emphasis upon graduate instruction were characterized by a
high total amount Of communication, particularly within
the department. Such departments were ppp_oriented to the
discipline and were more likely to be non-rated in the
Cartter report. In each of the six instances in which
there was significant variation not in accordance with the
overall relationship, emphasis placed upon graduate in-
struction was lowest in departments with low autonomy and
highest in departments with medium autonomy. Thus in all
twenty-five Of the significant relationships examined, low
autonomy was associated with a low emphasis upon graduate
instruction.

The percentage of faculty members indicating that
a very great amount of emphasis was placed upon graduate
instruction ranged from a high of 48.9 per cent when a
medium orientation to the department as a reference group
was combined with a high degree of autonomy, to a low Of
18.7 per cent when a low degree Of autonomy was combined
with a low ratio Of research to general fund dollars. As
was the case in examining the relationship between autonomy

and emphasis upon undergraduate instruction, the strength

92

Of the overall relationship between emphasis upon graduate
instruction and departmental autonomy was too unstable to

permit further attempts at interpretation.

Emphasis Upon Basic Research

Hypothesis.--Emphasis upon basic research as a

 

departmental goal will be directly related to the degree
Of departmental autonomy.
Table ll.--Relationship between faculty perception Of

emphasis placed upon basic research and faculty
perception of departmental autonomy.

 

 

 

 

 

Emphasis Upon Departmental Autonomy
Basic Research High Medium Low Total N
Very Great 29.8 26.3 19.8 25.2 316
Great 43.1 39.4 33.4 38.6 483
Some, Slight or None 27.1 34.3 46.8 36.2 453
100.0 100.0 1_oo'—.0' Io—o'.6 I257
N 413 414 425
x2 = 36.90***
df = 4
é .001

Data in Table 11 lend support to the stated hypo-
thesis. There was considerable variation produced due tO
the relatively small percentage Of faculty in departments
with low autonomy (19.8%) who attributed a very great
amount Of emphasis tO basic research as a departmental
goal. However, the Spearman rank order correlation coef-

ficients were of marginal significance at best (rS = +.18)

93

for all departments. This is counterbalanced by the fact
that the mean emphasis placed upon basic research increased,
on a five point scale, from 3.56 to 3.78 to 3.91 under con-
ditions of low, medium, and high autonomy, respectively.

The difference in means for departments of high and low
autonomy and for departments Of medium and low autonomy
were significant at the .001 and .01 levels of signif-
icance, respectively.

Thus, as noted in Table A-7, when the twenty-four
test factors were introduced, the overall relationship
held in forty-two of seventy instances. Twelve significant
relationships emerged which were at variance with the
overall pattern. In forty-six Of the fifty—four signif-
icant relationships examined, faculty in departments with
low autonomy were less likely than faculty in departments
Of high or medium autonomy tO indicate strong emphasis
upon basic research.

In general, where emphasis upon basic research was
weak, as when the size Of the department was small, democ-
racy in decision making was low, funds from external
sources were slight, the chairman Of the department was
highly influential, discussion with departmental faculty
was low, and discussion with the dean or with faculty and
chairmen Of other departments was high, the overall sig-
nificant relationship between the degree of departmental

autonomy and the amount of emphasis placed upon basic

94

research vanished. On the other hand, when individual
faculty members were perceived as having high or moderate
influence, the relationship also vanished, probably be-
cause under these conditions, emphasis upon basic research
was very high.

Emphasis upon basic research was highest in de-
partments where a high publication rate was combined with
a medium degree Of autonomy. In this instance, 62.7 per
cent of the faculty indicated a very great amount of in-
fluence was placed upon basic research. This finding tends
to corroborate Pelz and Andrews' finding that productivity
(a high publication rate) is enhanced under conditions Of
moderate coordination or autonomy. At the other extreme,
only 5.9 per cent Of the faculty in departments character-
ized by a low degree of democracy in departmental decision
making and high departmental autonomy indicated that a

very great emphasis was placed upon basic research.

Reference Group Orientations

Discussion of findings in regard to the reference
group orientation Of individual faculty members requires a
slight change in format. Table 12 presents the data rele-
vant to all three stated hypotheses. As noted by the ex-
tremely small chi square value, virtually no variation in
reference group orientation was evident among faculty

working in settings of varying departmental autonomy.

95

Table 12.--Relationship between reference group orienta-
tions Of faculty and faculty perception of
departmental autonomy.

 

 

 

 

 

 

Reference Departmental Autonomy
Group High Medium Low Total N
University 15.8 17.0 16.9 16.6 203
Department 38.8 41.9 39.6 40.1 491
Discipline 45.4 41.1 43.4 43.3 531
100.0 100.0 105.0 100.0 1225
N 405 401 419
x2 = 1.60
df = 4
p 3 .05

Table A-8 indicates the nine conditions under
which significant variation was evident relevant to the
stated hypotheses. The three test factors dealing with
departmental emphasis upon undergraduate instruction,
graduate instruction and basic research appeared to be
prominently associated with the presence of significant
relationships between an individual's reference group and

the degree Of departmental autonomy.

University Reference Group
Orientation

 

 

Hypothesis.—-Orientation to the university as a
reference group will be inversely related to the degree of
departmental autonomy.

The Spearman rank order correlation coefficients

between the degree Of departmental autonomy and the degree

96

Of orientation to the university as a reference group, as
noted in Table 1, yielded no significant relationships.
Orientation tO the university as a reference group ranged
from a low Of 9.2 per cent of faculty in departments char-
acterized by low influence of individual faculty members
and a medium degree Of autonomy to a high of 31.8 per cent
of the faculty in departments placing a low emphasis upon
basic research and having a high degree Of departmental
autonomy.

As noted in Table A-8, the hypothesized inverse
relationship between departmental autonomy and orientation
to the university as a reference group emerged in only one
of the nine relationships in which significant variation
existed. The converse, a direct relationship between uni-
versity orientation and the degree Of departmental autonomy
was evident in two of the nine conditions: when emphasis
upon basic research and, somewhat unexpectedly, when dis-

cussion with university administration were low.

Departmental Reference Group
Orientation

 

 

Hypothesis.--The relationship between departmental

 

autonomy and orientation to the department as a reference
group will be curvilinear. A higher percentage of indi-
viduals within departments with a medium degree Of autonomy
than in departments with a high or low degree of autonomy

will be oriented to the department as a reference group.

97

The Spearman rank order correlation coefficients
between departmental reference group orientation and de-
partmental autonomy revealed a negative orientation to the
department (rS = -.32) among faculty in highly autonomous
departments and a positive orientation to the department
(rS = +.30) on the part Of faculty in departments with low
autonomy.

Orientation to the department as a reference group
ranged from a high Of 52.8 per cent among faculty in de-
partments with a medium degree Of autonomy and a low amount
Of discussion with university administration to a low Of
19.8 per cent Of the faculty in departments where the col-
legial power Of the departmental faculty as a whole and
departmental autonomy were high. The hypothesis that
orientation to the department as a reference group would
be highest in departments characterized by a medium degree
Of departmental autonomy was supported in seven of the
nine conditions under which significant relationships
emerged. Thus, high orientation to the department appeared
to occur when medium autonomy was associated with low
emphasis upon graduate instruction or basic research.
Departmental orientations were also prominent when an ex-
ternal power pattern was evident whereby faculty, individ-
ually, and collectively, have low influence although the
total amount Of communication, particularly with the

university administration and department chairman are low.

98

Reference Group Discipline

 

Hypothesis.--Orientation to the discipline as a

 

reference group will be directly related to the degree Of
departmental autonomy.

The Spearman rank order correlation coefficients
between the degree Of disciplinary reference group orien-
tation and the degree of departmental orientation, as noted
in Table l, were generally inconsistent. The Opposite
direction Of the correlations for departments with high
(rS = +.29) and low autonomy (rS = -.22) indicate how
important the degree Of autonomy, itself, may be in deter-
mining reference group orientation. It is not surprising,
then, that the direct hypothesized relationship between
the degree of disciplinary orientation and the degree of
departmental autonomy held in only two Of nine instances.

In seven of the nine conditional relationships
noted in Table A-8, however, a disciplinary orientation
was strongest in departments where high autonomy was com-
bined with low emphasis upon undergraduate or graduate
instruction, low individual or collective influence Of
departmental faculty and low total communication, partic-
ularly with the department chairman and with university
administration. Thus, except for the varying degrees Of
autonomy and differing emphases upon undergraduate in-
struction and basic research, orientation to the depart-
ment and tO the discipline appear to be associated with

many of the same factors.

99

Orientation to the discipline as a reference group
ranged from highs Of 59.0 per cent in departments with
high autonomy and low emphasis upon graduate instruction
and 59.8 per cent in departments with high autonomy and
low discussion with the department head to a low of 32.4
per cent among faculty in departments with a medium degree
Of autonomy and low discussion with university administra-

tion.

Communication

 

Group Communication Means

 

Hypotheses:

 

A. The relationship between departmental autonomy
and the total amount of communication by a department will
be curvilinear. Individuals within departments with a
medium degree of autonomy will communicate more frequently
with persons internal or external to the department than
will persons in departments with high or low degrees of
autonomy.

B. The relationship between departmental autonomy
and the amount Of communication with persons external to
the department will be curvilinear. Individuals within
departments with a medium degree Of autonomy will communi-
cate more frequently with persons outside Of their own
department than will persons in departments with high or

low degrees Of autonomy.

100

C. The relationship between departmental autonomy
and the amount Of internal communication within departments
will be curvilinear. Individuals within departments with
a medium degree of autonomy will communicate more frequent-
ly with persons within their own department than will per-
sons in departments with high or low degrees of autonomy.

Table 13 presents the individual and group commu-
nication means for departments of high, medium, and low
autonomy and notes the contrasts of means which, when com-
pared, were significantly different from one another. The
only group communication means in which significant varia-
tion existed were in regard to internal communication. In
this instance, the mean for internal communication was
significantly lower in departments Of low autonomy (3.01)
than the means in departments Of high (3.17) or medium
(3.18) autonomy. The group internal communication mean
(3.14) was much higher than the mean for external communi-

cation (1.61).

Internal Communication

As might be expected, and as noted by the individ-
ual communication means in Table 13 which were all signif-
icantly different from each other, the greatest amount of
internal communication took place among faculty within the
department (3.88), while the department chairman (3.52),

chairmen of special departmental committees (2.96), and

101

 

 

I I mv.a ov.a me.a H¢.H cowumnu
ImAQHEOd muflmuo>flco Suez
I I mn.a ve.a mm.a mn.a cmoo cues
I I mm.a om.a nm.a mm.a couumoeasssoo cheese
I I mw.a mv.H mv.H mv.a mousewumcH HO
.mumoo Hosuo GOEHHOSU :DH3
I mo. Hm.a mm.H Hm.H me.H .mnama umauo as seasons anus
I I em.a mm.a ve.a mm.a coeumoecsaeou oumsuso
I I HO.H ve.a om.a mm.a :oHumoecsEEoo amcuouxm
I I mH.m no.m mH.m nH.m mucopsum mumspmuo cuflz
Ho. mo. om.m om.m mo.m oo.m meannessoo Hmuamspummmo
Hmwoomm mo cofiuflmsu nee:
Ho. Hoo. ~m.m mm.m oo.m mo.m amsuemno pamspummmo sues
I I mm.m om.m mm.m mm.m suaaomm unassummma anus
Ho. Ho. wH.m Ho.m ma.m na.m coauOOHcSEEOU accumch
I I mm.m mm.~ oe.~ nm.~ couscouaassoo Hmuoe
mIm MIH NIH Hmuoe 30a sweep: roam
memo: mo mummuucou m N a
HGOOHMHcmflm mEocous¢ Hmucosunmmoa

 

 

.>Eocousm kucofiuummop mo coaemoouom
muHsomm can mcmoa OOHDOOAOOEEOO msoum pcm Hospfl>flpcfl coo3uon mflnchHumHomII.mH magma

102

graduate students (2.23) were consulted significantly less
Often. Findings in regard to each of these individual
communication variables are reported in the order of the
frequency Of consultation. Hypotheses are not stated
relating the individual communication variables to the
degree of departmental autonomy because implicit in the
hypotheses that the group communication means will be
highest in departments Of medium autonomy is the assump-
tion that individual communication means will also be

higher under this condition.

Discussion with Departmental

FacuIty

Although communication was most frequent with one's
colleagues within his own department the data in Tables 13
and 14 indicate virtually no difference in regard to the
frequency with which faculty in departments with varying
degreesof autonomy consult with each other. Communication
with departmental colleagues was, thus, uniformly high.

When the twenty-four test factors were introduced
into the analysis of the relationship between the degree
Of discussion with one's departmental colleagues and the
degree Of departmental autonomy, significant variation was
noted in only one Of the seventy conditional relationships.
As more than one significant relationship could have been
expected to occur by chance, further reporting Of findings

is not warranted.

103

(Cable l4.--Relationship between the amount Of discussion
with the faculty in one's department and faculty
perception of departmental autonomy.

 

 

 

 

 

 

Discussion with Departmental Autonomy
Departmental Faculty High Medium Low Total N
Very Great 24.1 25.4 25.0 24.8 306
Great 47.2 43.7 41.7 44.2 544
Some, Slight or None 28.7 30.9 33.3 31.0 382

100.0 100.0 100.0 100.0 I237

N 407 405 420
x2 = 3.01
df = 4
p > .05

Discussion with Department Chairman

 

If one examines the percentage Of faculty in de-
partments Of high, medium and low autonomy who indicated
either a slight degree Of communication or no communication
at all with the department chairman, as noted in Table 15,
a direct relationship is evident. At the other extreme, a
very great amount Of communication with the department
chairman was most common in departments characterized by a
medium degree of autonomy. Further indication Of a direct
relationship is given, however, when one notes that the
mean amounts Of discussion with the department chairman
were significantly higher in departments with high (3.63)
or medium (3.60) autonomy than in departments with a low
(3.32) degree Of autonomy. The correlation between de-
partmental autonomy and discussion with the department

chairman for all departments (rS = +.09) was insignificant

104

with negative correlations of borderline significance in
Idepartments Of high (rS = -.30) and low (rS = -.29)
autonomy.

Table 15.--Relationship between the amount Of discussion

with the department chairman and faculty per-
ception of departmental autonomy.

 

 

 

 

 

 

Discussion with Departmental Autonomy
Department Chairman High Medium Low Total N
Very Great 24.9 28.0 21.3 24.7 304
Great 30.3 27.0 22.2 26.5 326
Some 31.3 27.8 30.9 30.0 369
Slight or None 13.5 17.2 25.6 18.8 232
, 100.0 100.0 100.0 100.0 123I
N 406 407 418
x2 = 26.63***
df = 6
§ .001

Table A-9 shows that, with the introduction Of
twenty-four test factors, a significant direct relationship
between departmental autonomy and discussion with the
department chairman was present in nineteen Of the seventy
conditional relationships examined. A significant sub-
pattern Of variation in fourteen instances was for commu-
nication with the department chairman to be highest in
departments with a medium degree of autonomy. Departments
exhibiting this sub-pattern also exhibited a great deal of

communication with all other elements Of the university, a

105

high university reference orientation, high emphasis upon
undergraduate instruction and high influence of the de-
partmental faculty as a whole.

The percentage of faculty members who indicated
that they discussed departmental affairs to a very great
extent with the department chairman ranged from a high of
41.5 per cent in departments within privately supported
institutions with a medium degree of autonomy to a low Of
13.9 per cent in departments characterized by a medium
amount Of discussion Of departmental matters with depart-

mental faculty and a low degree of autonomy.

Discussion with Chairmen of Special
Departmental Committees

 

 

NO consistent linear relationships between the
degree Of communication with the chairmen of special de-
partmental committees and the degree of departmental au-
tonomy, as noted by the Spearman rank order correlation
coefficients in Table l, were evident. The mean amounts
Of communication with the chairmen of special departmental
committees, as indicated in Table 13, were, however, sig-
nificantly higher in departments of high (3.00) and medium
autonomy (3.09) than in departments with low autonomy
(2.80). This trend was evident also in the display Of
data in Table 16 and contributed, undoubtedly, to the

significance of the chi square value which was obtained.

106

Table l6.--Relationship between the amount Of discussion
with chairmen of special departmental committees
and faculty perception Of departmental autonomy.

 

 

Discussion with
Chairmen Of Special

 

 

 

Departmental Departmental Autonomy
Committees High Medium Low Total N

Very Great or Great 33.7 35.3 27.6 32.1 392

Slight 36.6 37.8 35.0 36.5 445

None 29.7 26.8 37.4 31.4 383
100.0 100.0 100.0 100.0 1720'

N 404 399 417

x2 = 12.55*

df = 4

p é .05

Seeking to Obtain further evidence as to the
strength Of the relationship between departmental autonomy
and communication with the chairmen Of special departmental
committees and the conditions under which this relationship
was present, twenty-four test factors were introduced into
the analysis. The results are noted in Table A-10. Sig-
nificant variation was noted in only nineteen of the
seventy conditional relationships examined. In thirteen
of the nineteen relationships, the.overall trend, which
was for the most discussion with the chairmen Of special
departmental committees to take place in departments with
medium autonomy and for the least amount Of communication
to take place in departments with low autonomy, was re-

peated. The fact that this trend was particularly evident

107

in larger departments and in departments in public sup-
ported institutions which tended to be larger, suggests
that these committees may be Operational in nature, re-
flecting a division of labor found in most complex organ-
izations. Discussion with the chairmen Of special
departmental committees appeared to be associated with a
relatively high total amount of communication, particularly
with one's colleagues and with the university administra-
tion.

The percentage of faculty indicating a very great
or great amount Of discussion with the chairmen of special
departmental committees ranged from a high Of 41.4 per
cent for faculty in departments with medium autonomy and more
than thirty faculty members to a low Of 17.5 per cent in
departments characterized by high autonomy as well as a

low emphasis upon basic research.

Discussion with Graduate Students

 

Although the communication means for discussion
with graduate students noted in Table 13 tended to increase
as the degree of autonomy increased and a very slight
direct relationship might be detected between the two
variables from the data in Table 17, the insignificance
of the chi square and the Spearman rank order correlation

coefficients fail to indicate a clear relationship.

108

Table l7.--Re1ationship between the amount of discussion
with graduate students and faculty perception
Of departmental autonomy.

 

 

Discussion with Departmental Autonomy
Graduate Students High Medium Low Total N

 

Very Great, Great

 

 

or Some 34.2 33.0 29.6 32.3 397
Slight 38.9 40.2 39.9 39.6 488
None 26.9 26.8 30.5 28.1 346

100.0 —_100 0 I"_0’00. 100T. 1731'

N 409 403 419
x2 = 2.93
df = 4
p > .05

Similarly, when the twenty-four test factors were
introduced into the analysis, only ten Of the seventy
conditional relationships examined indicated a significant
amount Of variation. Among the ten significant relation-
ships, discussion with graduate students showed a direct
relationship to the degree of departmental autonomy in six
instances and an inverse relationship in two instances.
Discussion with graduate students was most frequent under
conditions Of medium autonomy in four instances. Discus-
sion with graduate students was more common when high au-
tonomy was coupled with public support, a high amount of
upward communication, particularly with the university
administration, or a low degree Of support from outside

research funds.

109

These findings suggest that communication with
graduate students may be greater when graduate student
support comes from the university rather than outside
funds. Thus, for example, a teaching assistantship may
provide an opportunity for a greater amount Of communica-
tion with a larger number of faculty than would a research
assistantship. The greater degree Of upward communication
found when discussion with graduate students was high
might, in turn, be due tO the politicking involved with
securing support for a department's graduate students.

The percentage Of faculty indicating at least some discus-
sion with graduate students about departmental affairs
ranged from a high of 42.9 per cent in departments charac-
terized by a medium degree of autonomy and a low orienta-
tion tO the department as a reference group to a low of
19.4 per cent when low autonomy was combined with a high

ratio Of research to general fund dollars.

External Communication

As indicated by the group communication means in
Table 13, external communication (1.61) was much less fre-
quent than internal communication (3.14). Within the
category Of external communication the mean amount of out-
ward communication to chairmen of other departments or
institutes and faculty of other departments (1.61), was
approximately as frequent as upward communication to the

dean or other university administrators (1.59).

110

Generally, as shown by the communication means in
Table 13, when faculty members wanted to discuss depart-
mental matters with someone outside Of their immediate
department, they turned to faculty members outside of
their department (1.81), before consulting with the dean
(1.72), chairmen Of other departments or institutes (1.48)

or university administration (1.45).

Discussion with Faculty in Other
Departments

 

The data in Table 18 would lend some support to an
hypothesis that an inverse relationship exists between the
degree Of departmental autonomy and the amount Of discus-
sion with faculty outside of the department. Communication
with faculty outside of the department, as noted by the
communication means in Table 13, decreased as autonomy
increased with faculty in departments with high autonomy
(1.73) communicating with faculty outside Of the department
significantly less Often than faculty in departments with
a low degree Of autonomy (1.88).

The analysis of data through the use Of the chi
square technique for determining statistical significance,
Spearman rank order correlation coefficients, and analysis
of variance all indicated a significant inverse relation-
ship between departmental autonomy and communication with
faculty outside of the department. However, when the

twenty-four test factors were introduced into the analysis

111

the significant inverse relationship held in only ten Of
the seventy conditions examined while in five additional
instances communication with external faculty was highest
within departments characterized by medium autonomy.
Table 18.--Relationship between the amount Of discussion

with faculty outside Of department and faculty
perception Of departmental autonomy.

 

 

Discussion with
Faculty Outside Of Departmental Autonomy
Department High Medium Low Total N

 

Very Great, Great,

 

 

 

Some or Slight 52.2 61.4 61.7 58.5 712
None 47.8 38.6 38.3 41.5 505
100.0 100.0 I0070’ IUUTU' IIIT
N 405 394 418
x2 = 9.49**
df = 2
é .01

When combined with a high degree Of departmental
autonomy such factors as public support, a high amount Of
democracy in departmental decision making, a low degree of
outside research support, and low orientation to the uni-
versity as a reference group were associated with a rela-
tively high amount of communication with faculty outside
the department. Similarly, when a medium degree of de-
partmental autonomy was combined with high personal in-
fluence Of individual professors or small departmental
size (5-19 members), discussion with faculty outside of

one's department was more frequent.

112

Again, the data point tO the importance of depart-
mental size as related tO communication with various ele-
ments internal and external tO the department. Thus sig-
nificant Spearman rank order correlation coefficients for
all departments between the number of faculty in the ranks
Of assistant professor and above and communication with

various persons were as follows:

rS with Size

Variables of A Faculty
Discussion with Dean -.49***
Discussion with Department Chairman -.40***

Discussion with Chairmen of Special
Departmental Committees +.32**

Discussion with Chairmen Of Other
Departments or Institutes -.32**

Discussion with Faculty in Other Departments -.25*

The percentage of faculty members indicating at
least a slight degree of communication with faculty outside
Of one's department ranged from a high Of 76.7 per cent in
departments with less than twenty faculty and a medium
degree Of autonomy to a low Of 45.3 per cent Of the faculty
in departments combining high autonomy with a low reference

group orientation to the university.

Discussion with the Dean

 

None of the Spearman rank order correlation coef-

ficients in Table l and none of the contrasts of

113

communication means in Table 13 were significant. At
least a slight degree of communication with the dean some-
what unexpectedly, was associated more Often with high
autonomy while communication was least frequent when a
medium degree Of autonomy was present. When the twenty-
four test factors were introduced into the analysis, sig-
nificant variation existed in eleven Of the seventy
conditional relationships.

Table l9.--Relationship between the amount of discussion

with the dean and faculty perception Of depart-
mental autonomy.

 

 

Amount Of Discussion Departmental Autonomy
With the Dean High Medium Low Total N

 

Very Great, Great,

 

 

 

Some or Slight 47.9 40.8 45.4 44.7 551
None 52.1 59.2 54.6 55.3 681
100.0 100.0 100.0 100.0 12 2
N 407 404 421
x2 = 4.21
df = 2
p > .05

In ten out Of eleven instances, in accordance
with the general trend, communication with the dean was
least frequent when a medium degree of departmental au-
tonomy was perceived. Thus, when a medium degree Of de-
partmental autonomy was also associated with such factors

as large departmental size or high orientation to either

114

the department or the discipline as a reference group,
communication with the dean was relatively less common
than in departments with a high or low degree of autonomy.
The percentage of faculty indicating at least some
discussion of departmental matters with the dean ranged
from a high Of 65.7 per cent in departments where high au-
tonomy was combined with a highly influential dean to a
low Of 27.8 per cent in large departments with more than
thirty faculty members in the rank of assistant professor

and above and a medium degree of perceived autonomy.

Discussion with Chairmen Of Other
Departments or Institutes

 

NO consistent relationships between the degree of
departmental autonomy and the amount Of communication with
the chairmen of other departments or institutes were ap-
parent through the chi square analysis noted in Table 20
or through the examination of Spearman rank order correla-
tion coefficients and discussion means noted in Table 13.

When the twenty-four test factors were introduced
into the analysis, only seven of the seventy conditional
relationships which emerged exhibited any significant var—
iation. Here, the combination of a low degree of autonomy
with small departmental size, a weak department chairman,
or a low orientation to the university as a reference
group were most prominently associated with the discussion
of departmental affairs with the chairman of other depart-

ments or institutes.

115

Table 20.--Relationship between the amount of discussion
with chairmen of other departments or institutes
and faculty perception Of departmental autonomy.

 

 

Discussion with
Chairmen Of Other Departmental Autonomy
Departments High Medium Low Total N

 

Very Great, Great,

 

Some or Slight 35.5 36.9 36.0 36.1 444
None 64.5 63.1 64.0 63.9 785
I00.5 150.5 155.5 155.5 I229
N 406 404 419
x2 = 0.18
df = 2
p > .05

A medium degree of departmental autonomy combined
with a low orientation to the department was also asso-
ciated with relatively greater communication with the
chairmen of other departments or institutes. The percent-
age Of faculty members in a department who discussed de-
partmental matters with the chairmen Of other departments
or institutes ranged from a high Of 59.3 per cent in de-
partments with weak chairmen and low autonomy to a low of
20.9 per cent in departments characterized by high autonomy
and a medium degree of orientation to the department as a

reference group.

Discussion with University Administrators
The chi square value noted in Table 21, the Spear-

man rank order correlation coefficients between the degree

116

Of departmental autonomy and the percentage Of faculty who
indicate that they have some communication with university
administrators, and the communication means in Table 13
point to no consistent relationship between the two
variables.

Table 21.--Re1ationship between the amount Of discussion

with university administrators and faculty
perception of departmental autonomy.

 

 

Discussion with
University Departmental Autonomy
Administrators High Medium Low Total N

 

Very Great, Great,

 

 

Some or Slight 30.2 33.6 30.5 31.4 385
None 69.8 66.4 69.5 65.6 840
100.0 T—00.0 PTO .0' I002'0' I275
N 404 402 419
x2 = 1.30
df = 2
p > .05

Of the six significant conditional relationships
that emerged when the twenty-four test factors were intro-
duced, five indicated that discussion with university ad-
ministrators was more frequent under conditions of medium
rather than high or low departmental autonomy. Thus the
combination Of a medium degree of autonomy with a high
orientation to the university as a reference group, high
discussion with the dean, small departmental size, low

democracy in departmental decision making or a weak

117

department chairman were all associated with a relatively
high percentage Of faculty who discussed departmental af-
fairs with university administrators. The percentage Of
faculty members who discussed departmental affairs with
university administrators ranged from a low Of 24.3 per
cent in departments where low autonomy was combined with
high discussion with one's faculty colleagues in his own
department to a high of 51.0 per cent of the faculty in
departments characterized by a medium degree Of autonomy
as well as a high orientation to the university as a

reference group.

Summary

Attempts to summarize the findings Of this study
are difficult due both to the volume Of data and to the
lack Of any clear decision-rules as to when the data point
either to a confirmation or rejection of a given hypothe-
sis. In general, a hypothesis was regarded as being sup-
ported if the chi square analysis of data, Spearman rank
order correlation coefficients, and the means for approp-
riate variables Obtained from an analysis Of variance were
consistent with each other.

Applying the criterion of consistency, only four
Of the sixteen stated hypotheses were supported including

those which hypothesized:

118

1. An inverse relationship between the degree Of
departmental autonomy and the influence Of the
department chairman,

2. A direct relationship between the degree Of de-
partmental autonomy and the influence Of the de-
partmental faculty as a whole,

3. A direct relationship between the degree Of de-
partmental autonomy and emphasis upon graduate
instruction,

4. A direct relationship between the degree Of de-
partmental autonomy and emphasis upon basic
research.

In addition, for three Of the sixteen stated hy-
potheses, a significant relationship between the degree Of
departmental autonomy and a dependent variable was evident.
However, the data did not support these hypothesized rela-
tionships between the degree of departmental autonomy and:

1. The individual faculty member's personal influence,

2. The amount of emphasis placed upon undergraduate
instruction,

3. Internal communication.
A more complete explanation Of the manner in which these
findings varied from those which were expected is found in
the summary section at the beginning Of Chapter V.

The remaining nine hypotheses were not supported
by the data, nor was significant variation between the
degree Of departmental autonomy and the particular depend-
ent variable evident. However, in each Of these nine
instances, when data were analyzed through the process Of

elaboration, except for the influence Of the departmental

119

faculty as a whole, at least six of the seventy conditional
relationships examined showed significant variation between
the degree Of departmental autonomy and the particular
dependent variable. Also, in three Of the nine relation-
ships where no significant relationship was evident, data
tended to be in the direction of the stated hypothesis.
This was the case in regard to the hypothesized relation-
ship between departmental autonomy and:

1. Favorable ratings in the Cartter report,

2. The percentage of faculty members within a given
department holding joint appointments,

3. Orientation to the department as a reference group.
In addition, although hypotheses were not stated
in regard to the eight individual discussion variables,
significant variation was found to exist between the degree
Of departmental autonomy and:
1. Discussion with the department chairman,

2. Discussion with the chairmen Of special depart-
mental committees,

3. Discussion with faculty outside Of the department.
A more complete explanation of these findings is also found

in the summary section at the beginning of Chapter V.

CHAPTER V

SUMMARY, DISCUSSION, IMPLICATIONS AND

RECOMMENDATIONS FOR FURTHER RESEARCH

The first section of this chapter provides a sum-
mary Of findings. The second section is devoted to a dis-
cussion Of these findings while part three explores some
Of the implications Of the study and the last section

indicates some recommendations for further research.

Summagy Of Findings

 

The purpose Of this study was to examine the manner
and conditions under which departments and individuals
within departments characterized by high, medium, or low
degrees of departmental autonomy differed in regard to
departmental characteristics, influence of various actors
within the department, emphasis upon primary departmental
goals, reference group orientation, and communication pat-

terns. The findings are summarized below.

Departmental Characteristics

 

1. NO significant relationship was found to exist
between the degree Of departmental autonomy and a
favorable rating in the Cartter report (Table 2).

120

121

NO significant relationship was found to exist
between the degree Of departmental autonomy and
the percentage Of faculty members within a depart-
ment who held joint appointments (Table 3).

NO significant relationship was found to exist
between the degree Of departmental autonomy and
the percentage of faculty members in departments
who noted one or more problems designated by
source as internal. However, there was a strong
tendency for low departmental autonomy to be asso-
ciated with a high percentage Of faculty members
who noted one or more problems classified by
source as internal tO the department (Table 4).

NO significant relationship was found to exist
between the degree Of departmental autonomy and
the percentage of faculty members in a given de-
partment who noted one or more problems classified
by source as external to the department (Table 5).

Influence

 

1.

An extremely strong inverse relationship was found
to exist between the degree of departmental au-
tonomy and the perceived influence of the depart-
ment chairman. This relationship held for virtually
all departments except for those which had exten-
sive support from fund sources external tO the
university (Table 6).

A strong direct relationship between the degree of
departmental autonomy and the influence Of the
departmental faculty as a whole was found to exist
under most conditions. A prominent sub-pattern of
variation showed the influence Of the departmental
faculty as a whole to be particularly high in de-
partments with a medium degree Of autonomy (Table
7).

Significant non-linear variation was found tO
exist between the degree Of departmental autonomy
and the individual faculty member's perception Of
his own influence. Individuals in departments
where the degree of autonomy was perceived as low
were particularly apt to express low perceptions
Of their own personal influence (Table 8).

122

Emphasis

l.

2.

Significant non-linear variation was found to
exist between the degree of departmental autonomy
and the amount of emphasis placed upon undergrad-
uate instruction as a departmental goal. There
was a tendency for faculty in departments with a
high degree of autonomy to place a low emphasis
upon undergraduate instruction although the mean
amounts Of emphasis upon undergraduate instruction
under conditions Of high, medium, and low autonomy
were not significantly different from each other
(Table 9).

A significant direct relationship was found to
exist between the degree of departmental autonomy
and the amount of emphasis placed upon graduate
instruction as a departmental goal due largely to
the fact that faculty members in departments with
a low degree of autonomy placed a low emphasis
upon graduate instruction (Table 10).

A significant direct relationship was found to ,
exist between the degree Of departmental autonomy
and the amount Of emphasis placed upon basic re-
search as a departmental goal. Emphasis upon
basic research was particularly low when the
degree of departmental autonomy was also low
(Table 11).

Reference Group Orientation

 

1.

NO significant relationship was found to exist
between the degree of departmental autonomy and

the percentage Of faculty members who were oriented
tO the university as a reference group (Table 12).

NO overall significant linear relationship was
found to exist between the degree of departmental
autonomy and the percentage Of faculty members who
were oriented to the department as a reference
group. However, in departments with high and low
degrees Of autonomy, there were significant inverse
and significant direct relationships, respectively,
between the degree Of departmental autonomy and
the percentage Of faculty members within a given
department who declared a departmental reference
group orientation (Table 12).

123

NO overall significant relationship was found to
exist between the degree Of departmental autonomy
and the percentage of faculty members who were
oriented to the discipline as a reference group
(Table 12).

Communication

 

1.

NO significant relationship was found to exist
between the degree Of departmental autonomy and

the average amounts Of total or external discussion
Of departmental affairs with various elements
within the university (Table 13).

Significant non-linear variation was found to
exist between the degree Of departmental autonomy
and the amount of internal communication in regard
to departmental affairs. The average amount Of
internal communication in a given department with
the departmental faculty as a whole, the department
chairman, the chairmen of special departmental
committees and graduate students was significantly
lower in departments characterized by a low rather
than a medium or high degree Of departmental au-
tonomy (Table 13).

Discussion of departmental affairs with one's col-
leagues in his own department was significantly
greater than communication with the department
chairman or any other element within the univer-
sity. However, no apparent relationship was found
to exist between the degree of departmental autono-
my and the amount Of discussion Of departmental
affairs with other members Of the departmental
faculty (Table 14).

Significant inverse linear variation was found tO
exist between the degree Of departmental autonomy
and the degree to which departmental affairs were
discussed with the department chairman. Discus-
sion Of departmental affairs with the department
chairman was significantly less frequent in de-
partments Of low autonomy than in departments Of
medium or high autonomy (Table 15).

Significant non-linear variation was found to
exist between the degree Of departmental autonomy
and the amount Of discussion of departmental con-
cerns with the chairmen Of special departmental
committees. Discussion Of departmental affairs

10.

11.

124

was significantly greater in departments Of high
and medium autonomy than in departments with a low
degree Of autonomy (Table 16).

There was no significant relationship between the
degree of departmental autonomy and the amount of
discussion Of departmental affairs with graduate
students (Table 17).

A significant inverse relationship was found tO
exist between the degree Of departmental autonomy
and the amount Of discussion of departmental af-
fairs with faculty outside Of the department, par-
ticularly in Cartter-rated departments, or in
departments characterized by high autonomy. Dis-
cussion Of departmental affairs with faculty mem-
bers outside of the department was significantly
lower in departments with high autonomy than in
departments characterized by a low degree Of
autonomy (Table 18).

NO significant relationship was found to exist
between the degree Of departmental autonomy and
the degree to which departmental concerns were
discussed with the dean (Table 19).

NO significant relationship was found to exist
between the degree of departmental autonomy and
the amount of discussion of departmental affairs
with chairmen Of other departments or institutes
(Table 20).

NO significant relationship was found tO exist
between the degree Of departmental autonomy and
the amount of discussion Of departmental affairs
with university administrators (Table 21).

Size of the department, as indicated by the number
Of faculty members within each department in the
rank of assistant professor and above, was found
to be directly related to the amount Of discussion
with the chairmen of special departmental commit-
tees and inversely related to the amount Of dis-
cussion with the dean, department chairman and the
faculty or chairmen of other departments or
institutes.

125

Discussion

 

In this section we turn our attention first to an
examination of some Of the reasons why it is probable that
a number Of the study's hypotheses were not confirmed.
Here, an Opportunity is presented tO compare the findings
Of this study with those previously cited in the review Of
literature.

Next, a composite summary of findings is reviewed
noting the probable reaction Of administrators and faculty
to departmental autonomy. The review serves tO point out
the probable lack of agreement between the two groups as
tO the desirability Of a high degree Of departmental
autonomy.

Finally, on the basis Of the study's findings and
on the basis of a suggestion by Kornhauser, an attempt is
made to make a case for the desirability Of characteristics
associated with departments possessing a medium degree Of
autonomy.

The hypothesis that departments with a medium
degree Of autonomy would be most likely to be favorably
rated in the Cartter report may have been erroneous on
several counts. The hypothesis was based primarily on the
research conducted by Pelz and Andrews who found that, for
scientists in research organizations, a medium degree Of
coordination provided a setting which appeared to improve

performance.

126

One of the most obvious difficulties in attempting
to generalize this finding to this particular sample was
that most of Pelz and Andrews' respondents were scientists
in private or governmental research laboratories, not uni-
versity faculty members in a number of disparate disci-
plines. As such, their performance was judged in terms Of
the individual's contributions and usefulness to the or-
ganization as well as his output Of reports, papers, and/or
patents. On the other hand, it is uncertain just what a
favorable rating in the Cartter report means.

Basically, the Cartter report appears to represent
the verdict Of a jury Of one's peers in his discipline.
Thus, such criteria as prestige and academic notability
rather than performance in terms of institutional goals
are emphasized. Dressel, Johnson and Marcus found that
Cartter-rated departments are less involved in local in-
stitutional matters, and presumably enjoy greater freedom
from university restrictions. Data noted in Table 2 would
tend to support this finding as the percentage of depart-
ments rated favorably in the Cartter report tended to in-
crease as the degree Of departmental autonomy increased,
although the trend was not statistically significant. In
retrospect, perhaps a Cartter report rating is more Of an
indicator Of extra-organizational outlook than a multi-

faceted measure Of performance.

127

The inability Of this study to replicate the find-
ings Of previous studies may be due to any number of fac-
tors including different sampling populations. However,
it also points further to the paucity of research in regard
to important processes within the university. Gross and
Grambsch as well as Hill and French found in their studies
that faculty enjoyed greater power in regard to college or
university goals or policy than did the department chair-
man. However, in this study, most likely because the
referrent was the amount of influence various actors ex-
erted in regard to departmental affairs--not matters Of
university-wide policy, the power Of the department chair-
man was found tO be significantly greater than the influ-
ence of the departmental faculty as a whole, except under
the condition of high autonomy. Here, the influence Of
the department chairman was still greater than that of the
faculty but the difference was not significant.

In regard tO the relative emphasis which faculty
perceive to be placed upon three primary departmental
goals: those Of undergraduate instruction, graduate in-
struction and basic research, the findings Of this study
were also quite different from those reported by Gross and
Grambsch. Although Gross and Grambsch's sample was drawn
from some sixty-eight universities as Opposed to the fif-

teen universities included in this study, both studies

128

included only those faculty in departments which were Of-
fering graduate degree programs. Thus the differences are
somewhat surprising.

Gross and Grambsch ranked the university-wide goals
of carrying on pure research, encouraging graduate work,
and emphasizing undergraduate instruction 7th, 18th, and
44th, respectively, in terms of their perceived importance.
This study, with reference to departmental, rather than
university goals found the perceived emphasis upon graduate
instruction to be significantly higher than emphasis upon
either basic research or undergraduate instruction.

Perhaps, again, the difference between the univer-
sity and departmental reference points Of the two studies
in regard tO goals is crucial with the higher emphasis
which Gross and Grambsch's respondents perceived in regard
to pure research reflecting the traditional image of the
university as the discoverer Of new knowledge. However,
in reality, the lower emphasis accorded to basic research
in this study may reflect the failure to allocate adequate
resources at the departmental level. This may in turn
cause the goal Of basic research to be displaced by that
Of graduate instruction. In any case, undergraduate in-
struction appears to be truly the least of the brethren as
both studies indicate that this goal receives the least
emphasis among the three primary goals. Indeed, the rela-

tively constant amount of emphasis placed upon

129

undergraduate instruction under conditions Of high, medium
and low autonomy may suggest that it, like the kid brother,
is an ever present tag-along which must be tolerated.

Under the condition of low departmental autonomy,
however, undergraduate instruction was found tO receive a
slightly greater emphasis than basic research. Returning
to the analogy of the kid brother, this might suggest that
as Opportunities (funds) are unavailable to play with more
desirable companions (basic research), as was generally
found to be the case when autonomy was low, the little
brother (undergraduate instruction) becomes more and more
attractive as a playmate.

This study found virtually nO linear relationship
between the degree of departmental autonomy and the per-
centage of faculty in a department with university, de-
partmental, or disciplinary reference group orientations.
The few instances in which significant variation did exist
were Often in regard to the varying degrees Of emphasis
placed upon undergraduate and graduate instruction and
basic research. Perhaps, if a large enough sample were
available, the relative effects Of departmental autonomy
and departmental emphasis or goals could be distinguished
from each other.

In regard to communication among various elements
internal or external to the department but within the

university, the findings Of this study are not directly

130

comparable to those of Crane or Hagstrom who looked at
disciplinary communication networks. Similarly, the find-
ings Of this study are not directly comparable to those of
Pelz and Andrews whose findings in regard to communication
were related to performance measures. The perhaps mistaken
assumption to equate a medium degree of autonomy with a
medium degree of coordination which Pelz and Andrews found
to be associated with highest performance led the writer
tO hypothesize that communication among various elements
internal and external to the department would be highest
in departments characterized by a medium degree Of
autonomy.

In actuality, however, although a medium degree Of
autonomy was associated with the greatest amount Of commu-
nication with the chairmen of Special departmental commit-
tees, most differences were small. Indeed, perhaps the
most significant finding which emerged from the analysis
Of the communication variables was in regard to the amount
Of discussion Of departmental affairs with departmental
faculty in departments with varying degrees of autonomy.
Here, although discussion with departmental faculty was
more frequent than discussion with any other element in-
ternal or external to the department, there was virtually
no difference in the amount of discussion which tOOk place,
even when all twenty-four test factors were introduced

into the analysis.

131

Somewhat unexpectedly, low autonomy was associated
with the least amount Of communication with the department
chairman or with the chairmen Of special departmental com-
mittees. These findings do, however, help account for the
low amount Of internal communication in regard tO depart-
mental affairs found in departments with low autonomy.

The administrator with concern for the university
as a whole and the individual faculty member who lOOk at
the departmental characteristics which are correlated with
departmental autonomy as noted in Table 1, may have mixed
reactions. Depending upon whether the administrator pre-
fers a strong or weak department chairman, he will look
with disfavor or favor upon the relatively weak influence.
exerted by department chairman in departments with high or
medium autonomy. He may also view the greater democracy
Of departmental decision making which characterizes de-
partments Of relatively high autonomy as threatening.
Similarly, if the administrator views the vesting Of col-
legial power in the hands Of the departmental faculty as
harmful tO university interest, he will be unhappy about
the collective power Of faculty members in autonomous
departments.

In regard to the perceived emphasis placed upon
particular departmental goals, the administrator may view
the positive correlation between departmental autonomy and

emphasis upon graduate instruction or basic research as

132

contributing further to the compartmentalization Of the
university. On the other hand, faculty are likely to
regard increased departmental autonomy as a precondition
necessary for achievement of the valued goals of advancing
knowledge and training scholars.

Findings in regard tO.the relationship between
departmental and disciplinary reference group orientations
Of faculty and the degree Of departmental autonomy are
likely to disturb the administrator. Here, particularly
in departments characterized by high autonomy, orientation
to the department was low. Similarly, the administrator
who was intent upon improving communication among various
elements within the university is likely to attempt to
curb autonomy when he finds that in departments with high
autonomy, there is a negative correlation between the
amount of departmental autonomy and the total amount Of
communication with other elements, both internal and ex-
ternal to the department. The low total and outward com-
munication, particularly with faculty in other departments
and with the department chairman noted in high autonomy
departments, on the other hand, are probably valued by the
faculty member in that external interference is minimal.

The administrator and the faculty member, again,
depending upon their individual perspectives when they
look at some Of the additional correlates of departmental

autonomy indicated in Table A-12, are likely to regard

133

departmental autonomy as desirable or undesirable in that
it is associated with a low preference for emphasizing
undergraduate instruction, low influence Of university
committees, high concern for prestige, and increased em-
phasis upon professional activities outside of the univer-
sity. The administrator, feeling that many other depart-
ments may be just as deserving, may also fail tO appreciate
the fact that highly autonomous departments, and partic-
ularly Cartter-rated departments, enjoy a number Of per-
quisites which come with the relatively high amount Of
general fund dollars expended per faculty member, i.e.,
greater access to funds for supplies and equipment and a
higher percentage Of faculty members with private Offices..
The administrator, as well as the faculty member,.
may on the other hand, take pride in the ability Of au-
tonomous, and particularly Cartter-rated departments, to
attract research dollars although this pride may be tem-
pered by the remembrance of the headaches encountered in
securing these funds. Administrators and faculty may also
tend to look with favor upon the positive correlation
between departmental autonomy and the production Of Ph.D.
degrees. Particularly disturbing to the administrator,
however, would be the tendency, as noted in Table A-12,
for faculty members in departments with high autonomy to

state that departmental autonomy should not be controlled.

134

The preceding section failed to indicate correlates
Of either a high or low degree Of departmental autonomy
which both faculty and administrators could regard as
being desirable. Much Of the difficulty was, no doubt,
due to the different perspectiveswhich the two groups may
generally be expected to adopt.in regard to institutional
matters. Thus, in this section, findings relative to de-
partments characterized by a medium degree of autonomy are
noted in an effort to cite outcomes which might generally
be regarded as favorable tO the university.

Admittedly, there may be much argument as to what
characteristics are desirable from an institutional view-
point. However, the analysis Of data through the process
of elaboration indicated that a medium degree of autonomy
was positively associated with several communication and
departmental characteristic variables. Although there was
no overall significant relationship between the percentage
Of faculty members holding joint appointments and the
degree Of departmental autonomy, a medium degree of au-
tonomy was most frequently associated with the incidence
of joint appointments.

It appeared that a smaller percentage of faculty
members in departments with medium autonomy indicated the
existence Of departmental problems of an internal nature
which could be resolved by the department. On the other

hand, a medium degree Of departmental autonomy was

135

associated with a relatively high percentage of external
problems, most notably in Cartter-rated departments and in
departments characterized by.a high amount Of total commu-
nication and external communication, particularly with the
dean. Taken together these findings suggest that a medium
degree of departmental autonomy may be associated with
rigorous pursuit of outside funds as evidenced by a greater
number Of external problems, relatively little internal
conflict, and a fair amount Of communication with various
elements within the university.

Although there were few conditions under which
significant relationships emerged between the degree Of
departmental autonomy and a particular reference group
orientation, a strong departmental reference orientation
was most often associated with a medium degree of depart-
mental autonomy. A medium degree Of departmental autonomy
was also associated with increased communication. Thus,
the amount of discussion Of departmental affairs with
various elements such as the department chairman, chairmen
Of special departmental committees and university adminis-
trators was highest in departments Of medium autonomy in
several instances. To the degree that the above findings
associated with a medium degree of departmental autonomy
appear to represent a lessening of internal conflict, a
probable rigorous pursuit Of outside funding possibilities,

an intra-institutional reference group orientation (to the

136

department), and increased communication with other persons
within the university, a medium degree of departmental
autonomy may represent, as Kornhauser has suggested, an

ideal toward which universities should strive.

Implications

 

The favorable outcomes for the university found tO
be associated with a medium degree Of departmental autono-
my may Offer a ray of hope to the beleaguered administrator
who, amid all the tumult and shouting Of the disparate
disciplines, is inclined toward much hand-wringing when he
contemplates the fragmented nature of the multiversity.
They serve to indicate that a medium degree Of departmental
autonomy can be associated with a number of characteristics
which seem to indicate a degree of "departmental vitality."
Thus, they tend to support the theses Of such as Engel,
Glaser, Caplow and McGee, and Pelz and Andrews that insti-
tutions and individuals need not be constantly at Odds.

To be sure, a number Of serious problems are.asso-
ciated with excessive degrees Of departmental autonomy.

But these problems are not unique to universities. The
entrepreneurialism Of the highly autonomous departments
illustrates the dilemma between managerial planning and
initiative which Blau and Scott note is common to all or-
ganizations. On the other hand, departments with low
autonomy and relatively parochial concerns may exhibit

tendencies toward stagnation which Hefferlin has noted.

137

The dilemma between hierarchical coordination and
the free flow Of communication, also noted by Blau and
Scott, would seem to be lessened in the university which
exhibits, in comparison to other formal organizations, a
rather flat power structure. Direct contact with other
units rather than hierarchical.coordination could be en-
hanced through a greater number Of joint appointments and
the establishment of interdisciplinary teaching and re-
search units. Holders Of joint appointments can make.
valuable contributions to departments in that they repre-
sent "marginal men" who, if they are "loving critics" can
Offer fresh perspectives on a wide range Of problems.

The most serious dilemma Of formal organization
facing the university which was noted by Blau and Scott is
the conflict between bureaucratic discipline and profes-
sional expertness. The problem is heightened by both the.
multiplicity or lack of goals which are quantifiable and
the extra-institutional orientation Of many faculty members
which facilitate the perennial clash between faculty and
administration.

The clash is indicative, in part, Of the problem
which all organizations face in determining reward systems.
It is complicated both by the fact that the university
reward system, unlike in other organizations, is primarily
based on values which are external to the university and
do not take into account rewards for service to the

institution.

138

Indeed, if additional subunits were added to the
university which were designed to reward a particular
competency, the benefits which would accrue from the fac-
ulty member's being able to pick the "league" in which he
wishes to compete may outweigh the disadvantages Of further
compartmentalization.

Better attempts should be made to acquaint faculty..
members with institutional goals, rewards, and the limita-
tions Of institutional resources. A corps Of professor-
administrants who return to their faculty positions after
a specified period could be helpful in this regard. At
the same time, however, faculty members must recognize the
necessity of accommodating their own personal interests to
these goals and resources if the university is to continue

as a rational entity.

Recommendations for Further Research

This research is significant from the standpoint
that the question Of autonomy or coordination is basic to
the literature Of complex organizations. The question Of
excessive departmental autonomy as oftentimes manifested
by the department's disregard for institutional concerns
interests administrators of higher education in that such.
organizational behavior is antithetical to the supposedly

rational nature of the university.‘

139

In view of the limited volume and scope of research
literature in higher education, this study represents a
first attempt to construct a profile of the autonomous
university department. Methodologically, this study is
significant in that it, unlike most survey research in.
education, as noted by Trow, has not only described rela-
tionships but has also specified additional conditions
under which the relationship is either changed or repli-
cated.

The study was not without limitations. With some
Of these difficulties in mind, the following recommenda-
tions are made for further research in regard to depart-
mental autonomy:

1. Additional research should draw upon a large.
enough number Of institutions which are able to supply
basic input-output data so that financial characteristics
of a department such as the ratio Of research to general
fund dollars may be included in the definition of depart-
mental autonomy without having to disregard the responses
Of those departments unable to supply such data.

2. Additional research should employ a large enough
sample so that differences among disciplines may be.anal-
yzed. A larger sample would also permit in-depth analyses
Of two-variable relationships by allowing for the simul-

taneous introduction of two or more test factors.

140

3. Due to the lack of adequate or readily available
performance measures in higher education, additional re-
search along the lines of that undertaken by the Educa-
tional Testing Service should seek to identify desirable
characteristics which indicate "institutional vitality."1
These characteristics should then be analyzed in regard to
the degree of departmental autonomy in order to determine
the relationships between these variables.

The abuses caused by inordinate departmental self-
interest may be due in part tO the basic lack Of informa-
tion in regard to university Operations and processes. If
this is true, the ability to provide the information nec-
essary to carry out the above recommendations for research

may, in itself, go a long way toward limiting excessive

departmental autonomy.

 

1R. E. Peterson and D. E. Loye, eds., Conversations

 

Toward a Definition Of Institutional Vitality TPrinceton:
Educational Testing Service, 1968).

 

APPEND IX

141

 

 

 

 

 

9 h N v 9 b N N
I I I I I I I I I 309 .p0z £09: .8500 90£H09£H
I I I I I 9o. I I I 309 .U0S £09m .EEOU U9039so
I 9o. I I I Hoe. I I I 369 .602 smHm .esoo pants:
I I I I I 9oo. I I I 309 .p0z £09m ..EEOU 90:909xm
I I I I I I I I 9o. 369 .pm: :69: a69umo9aassoo 96969
I I I I I mo. I mo. I 369 .pm: a59m .spa .59aa .msomfla
I mo. I I I 9oo. I I I 309 .00: £093 £009 £09mmsom9o
I mo. I 9oo. I I I I I 309 .p0z £m9m .omm .9m0o .050099
I 9o. I I I I I I 90. 309 .602 £m9m @009 .9909 .msom9o
I I I I I I I I I 309 .pm: smHm 0a9999om9o .mmm
I mo. I Hoo. I I I I I 309 .O02 £m9m 9c0fi990m0a .m0m
I I I I I mo. I 90. I 309 .U0z £O9m m99mu0>9cb .mmm
I I 900. I 9o. . I 90. I I 309 .602 £093 £0900m0m 09009 .mEm
I I I I I I I mo. I 309 .O02 £09m .9mcH .pmuw .@89
I I I I I I I I I 309 .p0z £m9m .9mcH .Omumo .mam
I 99. I I I 9o. I I I 309 .Umz £09m .omm .>9p£H .mcH
I I I I I I I I I 309 .pm: £599 660m .9309 .mp9
mo. I I 900. I I I I I 309 .p0z £09m .omm .9m0o .mcH
I I I I I I I I I 309 .pm: sm9m 096m ao9umo99nsm
I I I I I I I I I 309 .pm: tm9m .9 .a00 m\.m0m w
I 9o. I 90. I I I I I 309 .002 £m9m >0090080o
I I mo. I I I 9oo. I I 309 .U02 £m9m M99500m 9 0N9m
«z I I «z I I «z 90. I «z 696>9um 699nsm anomasm
92 I I «z I I 92 I I «z pmumm coups
9oz m£990m 9099900
m N 9 m N 9 m N 9 m N 9
0099000900 0099000900 m09uom09mo 0099000900
.c09sd 309 £993 .£O9s¢ .U0S £993 0009059 90900m 9009

.09909 £9 90309
.m9m .muma< .9n

.m9m0o £9 H0£m9m m9£m£099090m
.m9m .m9mmm .9b 0mu0>£9 .m9m

90900m 900B

 

 

.0909009 9009 HDOMI>9O039 mo £099OSOOH9£9 0£9
m£ ©0999OOE mm >EOCO950 909c0fi9umm0p mo £099m0ou0m m99somm can 09:05990m0p £9
m9£089£90mm0 9:9On m£9p90£ 0909505 w995009 mo 0mm9£0090m £00390£ m9£m£099090mII.9I¢ 09£09

142

 

 

 

 

 

 

 

 

 

6 6 66
6 6 6 6 m 6 mm 96 99
I I I I I I 966. 966. 966. 369 .662 2692 .2266 96296929
I I I I I I 966. 966. 966. 369 .662 2692 .2266 6963966
I I I I I I 966. 966. 966. 369 .662 2692 .2266 696366
I I I I I I 966. 966. 966. 369 .662 2692 .2266 96296926
I I I I I x 9oo. 90o. I 309 .602 £m9m 2099609222800 96909
I I I I I I 966. 966. 966. 369 .662 2692 .262 .3926 .626696
I I I I I I 966. 966. 966. 369 .662 2692 2666 2696626696
I I I I I I 966. 966. 966. 369 .662 2692 .666 .9666 .666696
I I I I I I 9oo. 90o. 9oo. 309 .602 £092 660m .9209 .620092
I I I I I I 66. 966. 966. 369 .662 2692 6299696696 .962
I I I I I I 966. 966. 966. 369 .662 2692 9262996666 .962
I I I I I I 966. 966. 966. 369 .662 2692 6996963926 .962
I I I I I I 966. 906. 966. 309 .602 £m9m £0960m0m 09269 .289
I I I I x I 966. I 966. 369 .662 2692 .9629 .6696 .626
I I I I I I 966. 966. 96. 369 .662 2692 .9629 .66966 .626
I I I I I I 966. 66. 66. 369 .662 2692 .666 .39629 .929
I I I I x I 966. I mo. 309 ..602 £692 660m .9202 .929
I I I I I I 966. 966. 966. 369 .662 2692 .666 .9666 .929
I I I I I I 66. 966. 966. 369 .662 2692 6962 26996699266
I I I I 2 2 966. I I 369 .662 2692 .6 .266 6\.662 6
I I I I I I 966. 966. 966. 369 .662 2692 666966266
I I I I I I 66. 966. 966. 369 .662 2692 6996666 2 6696
22 I I 22 I I 22 966. 966. 22 6963996 699226 9966666
22 I I 22 I I 22 966. 966. 22 66962 66wmm 629962 9699966
6 m 9 6 m 9 6 m 9 m m 9
60990m096u 6099000960 60990m096o 6099060960
wamwhm mmfimflcwxw MUHOS HOHOMh UmOB
m9£m2099690m m9£m2099690m m9£m2099690m 909066 9009
.m9m 90£9O 92609992m9m 0m90>29 .m9m

 

 

.0909069 9609 9209I>92039 mo 209902609929
0£9 29 60999608 66 28020926 969202996206 90 209920090@ 2992069 626 26E996£0
9202996206 0£9 90 002029929 0£9 90 2099200902 2992069 200390£ m9£m2099690mII.NI2 0996B

143

.28020926 829608 2993 69208996206 29 9602292 663 09023 6 66 2992069 9208996206 029

09 002029929 96096092 029 22992299996 69208996206 29 6902808 2992069 90 0269200902 0292

 

 

 

 

 

2 99 96 I
6 9 6 6 6 6 69 66 69
I I I I I I 96. 96. 966. 369 .662 2692 .2266 96296929
I I 6966. I I I 96. 966. I 369 .662 2692 .2266 6963966
I I I I I I 966. 966. 96. 369 .662 2692 .2266 696366
I I 696. I I I 966. 966. I 369 .662 2692 .2266 96296922
I 29o. I I I I 90o. I 9oo. 309 .602 2292 2099609228800 96909
I I I 2 I I I 966. 966. 369 .662 2692 .262 .3926 .666696
I I I I I I 966. 966. 9c. 309 .602 2292 2600 2096620692
I I I I I I 96. 966. 966. 369 .662 2692 .666 .9666 .666696
I I I I x I 9oo. I 90o. 309 .602 2292 6602 .9202 .620692
I I I I I I 966. 66. 96. 369 .662 2692 6299696696 .962
290. I I I I I I mo. 9oo. 309 .602 2292 9208996202 .902
I I I I I 2 966. 966. I 369 .662 2692 6996963926 .962
29cc. I 90o I I I I 900. I 309 .602 2292 20960602 09662 .282
I I I I x I 9oo. I 96o. 309 .602 2292 .9629 .6690 .282
I I I I I I 966. 966. 966. 369 .662 2692 .9629 66966 .622
I I 296 I I I 66. 966. I 369 .662 2692 .666 .39629 .929
666. I I I I I I 966. 96. 369 .662 2692 6662 .9666 .929
I I I 2 I 2 I 66. I 369 .662 2692 .666 .9666 .929
I I I I I 2 966. 966. I 369 .662 2692 6962 26996699266
I I I I 2 I 966. I 966. 369 .662 2692 .6 .266 6\.662 6
I I I x I x I 90o. I 309 .602 2292 206900802
I I I 2 I I I 96. 966. 369 .662 2692 2996666 2 6696
22 I I 22 I I 22 966. 966. 22 6963996 699266 9966666
22 I I 22 I I 22 966. 966. 22 66962 66mm” 629962 9699966
6 6 9 6 m 9 6 6 9 6 6 9
6099020960 6099020960 6099020960 6099020960
6960224 6026926> 66902 909062 9609
292620996902 292620996902 292620996902 909062 9609
.696 96296 92669992696 966996 .696

 

 

.6909069 9609 9209I292039 90 209902609929 029 22 6099
I9608 66 28020926 969208996206 90 2099200902 2992069 626 09023 6 66 2992069
969208996206 029 90 002029929 029 90 2099200902 2992069 2003902 292620996902II.6I2 09269

144

.28020926 309 2993 69208996206 29 960309 626 28020926 90 009206 829608 6

2993 69202996206 29 9602292 663 002029929 96206902 6.902808 2992069 6 90 2099200902 0292

 

 

 

 

 

 

6 mm mm
6 m m 6 99 99 69 99 69
I I I I x x 900. I I 309 .602 2292 .8800 96290929
I I I I I 2 mo. 900. I 309 .602 2292 .2200 6963920
I I I I I x 9oo. mo. I 309 .602 2292 .2200 696322
29cc. I I I I x I 90o. I 309 .602 2292 .2200 96290922
I 290. I I I x 960. I I 309 .602 2292 2099609222800 96909
I I I I I I 66. 966. 96. 369 .662 2692 .262 .3926 .666696
I I I I x I 906. I 96. 309 .602 2292 2602 2096620692
I I I x I I I 9oo. 90. 309 .602 2292 .062 .9202 .620692
I 666. I I I 2 966. I I 369 .662 2692 6662 .9666 .666696
I I 266 I 2 I 966. I I 369 .662 2692 6299696696 .962
2mo. I I I I I I mo. 90o. 309 .602 2292 9202996202 .902
29o. I 29o I I I I .9oo. I 309 .602 2292 299690>92D .902
I I I x I x I 9o. I 309 .602 2292 20960602 09662 .282
I I I I I 2 9o. me. I 309 .602 2292 .9629 .6690 .222
I I I I I I mo. 9oo. 90. 309 .602 2292 .9629 66929 .282
I I I I 2 2 66. I I 369 .662 2692 .666 .39629 .929
2966. I I I I I I 96. 66. 369 .662 2692 6662 .9666 .929
I I I 2 2 I I I 96. 369 .662 2692 .666 .9666 .929
I I 666 I 2 I 96. I I 369 .662 2692 6962 26996699266
I I I I 2 2 966. I I 369 .662 2692 .6 .266 6\.662 6
I I I I x x 9oo. I I 309 .602 2292 206900202
I I I x x I I I mo. 309 .602 2292 2992062 2 029m
22 I I 22 2 I 22 I 966. 22 6963996 699266 9966666
22 I I 22 2 I 22 I 966. 22 66962 66mm“ 629962 9699966
6 N 9 m N 9 m m 9 m m 9
6099020960 6099020960 6099020960 6099020960
6960224 6026926> 66902 909062 9609
292620996902 292620996902 292620996902 909062 9609
.696 96296 92669992696 .966996 .696

 

 

.6909069 9609 9209I292039
90 209902609929 029 22 60999602 66 28020926 969202996206 90 2099200902
2992069 626 002029929 96206902 90 2099200902 2992069 2003902 292620996902II.6I2 09269

145

.28020926 229602 2993 69208996206 29 663 20990299629 0962669290622 2022 69662280 96096090

+

.28020926 2292 2993 69208996206 29 62209 69662280 96609
“28020926 829608 2993 69208996206 29 663 20990299629 0962669290622 2022 69662220 960960906

 

 

 

 

 

M9 .96 29
m m m 69 m9 6 m 0 m
I I I I 2 2 9o. I I 309 .602 2292 .2800 96290929
I I 9o. 2 I I I 9o. I 309 6602 2292 .8800 6963920
I 699096. + I 2 I 2 I I I 302 .602 2292 .2800 696322
I I I 2 2 2 I I I 302 .602 2292 .2800 96290922
I I I 2 2 2 I I I 309 .602 2292 2099609228800 96909
90. I mo. I 2 I I I I 309 .602 2292 .262 .3922 .620692
+ I «90. + I 2 I 2 I I I 309 .602 2292 2602 2096620692
I I I 2 I 2 I 90. I 309 .602 2292 .062 .9202 .620692
I I I I 2 I mo. I mo. 309 .602 2292 6602 .9202 .620692
996. I I I I I I 66. 66. 362 .66: 2692 6299696696 .962
I I 96. 2 I I I 96. I 362 .662 2692 9262996666 .962
I 9oo. + I I I 2 90o. I I 309 .602 2292 2996903922 .902
I I 9o. 2 2 I I I I 309 .602 2292 20960602 09662 .282
I 6mo. I I I 2 9o. I I 309 .602 2292 .9629 .6690 .282
I I I 2 2 2 I I I 309 .602 2292 .9629 .6692: .222
I I I I 2 2 96. I I 362 .662 2692 .666 .39629 .929
I I I 2 2 2 I I I 362 .662 2692 6662 .9666 .929
I I 6966. 2 2 I I I I 362 .662 2692 .666 .9666 .929
I I I 2 I I I 66. 66. 362 .662 2692 6962 26996699266
I I I 2 I 2 I 90. I 309 .602 2292 .2 .200 6\.602 6
I I I 2 2 I I I 9oo. 309 .602 2292 . 206900802
I «mo. I 2 I 2 I I I 309 .602 2292 2992062 4 0296
22 6966. I 22 I I 22 I 966. 22 6963996 699266 9966666
22 I I 22 2 2 22 I I 22 66962 66wmm 629962 9699960
m m 9 m N 9 m m 9 m N 9
6099020960 6099020960 6099020960 6099020960
6960222 60269263 66902 909062 9609
292620996902 292620996902 292620996902 909062 9609
.296 90290 9260929229m 0690>2m.29m

 

 

.6909069 9609 9209I292039 90 209902609929
029 22 60999602 66 28020926 969208996206 90 2099200902 2992069 626 20990299629
0962669290622 2022 600692 69662280 90 2099200902 2992069 2003902 292620996902II.mI< 09269

.wfiocousm achoE nqu mucmfiuummmc cH ummsmHn
can aaocousm 30H nuH3 mummEpHmmmo GH umm3oH mm3 coHuosuumcH oumacmum coma mHmmnmEma

 

 

 

 

 

m Irv 5H
m N m mH mH NH v m oH
I I I x x I I I mo. 30H .wmz 55H: .EEOU HmcumucH
I I I x I x I Ho. I 30H .Umz amHm .EEOU wum3uso
I I I I x x Ho. I I 30H .60: 50H: .EEOU GHM3QD
I I I x x x I I I 30H .vmz anm .8500 Hmcumuxm
I I I x x I I I Ho. 30H .cmz anm :oHumoHcsafioo Hmuoa
I I I x x I I I mo. 30H .vmz anm .Eufl .>H:D .msomHo
I I «mo x x I I I I 30H .Umz anm amma QOHmmaomHa
I «Ho. I x I I I I mo. 30A .602 anm .omm .ummn .mnomHo
I I I x x I I I mo. 309 .cm2 smHm ammm .umma .msomHQ
I Imo. I I I x Hoo. I I 3oq .cmz smfim mcHHmHomHa .mmm
*Ho. I I I I I I Hoo. mo. 30g .cmz swam unmsunmmma .mmm
I I «mo x x I I I I 30H .00: anm muHmHm>HcD .mmm
I I I x x x I I I 30H .602 anm noummmmm UHmmm .mEm
% I. I I x x x I I I 30H .002 gmHm .umcH .cmuw .mem
1. I I I x I I I Hoo. mo. 30H .Umz anm .umcH .UmumD .mfim
«mo. I I I x x I I I 3oq .cmz smflm .omm .>HwaH .mcH
I I I x x I I I mo. 30H .602 anm Ummm .ummo .mcH
I I I x x x I I I 30H .cmz smHm .omm .umma .mcH
I I I x x x I I I 30: .omz anm mumm aoHumoHHnsm
I I I I x x Hoo. I I son .wmz smflm .m .cmw w\.mmm m
I I I I x x mo. I I 30A .002 anm mommoofimo
I I I x x x I I I 30H .Umz anm muHDomm fl mNHm
«z I I 42 x I «z I Hoo. «z mum>Hum UHHnsm uuommsm
«z I I «z x I «z I ace. «2 cmumm cmwmm magnum umuuumu
m m H m m H m N H m m H
ammuommumu mmHuommumu mmHuommumo mmwuommumu
mummmm¢ mmsmHam> mvHom uoaomm umma
mHamcoHumHmm mHnmcoHHMHmm mHamGOHuMHmm Houomm umme
.mHm Hmnuo HGMOHmmcmmm uomumo .mHm

 

 

.muouomm ummu “DOMImucw3u mo GOHuozcouucH
mnu ha UwHMHvoE mm aeocousm HmucmEpHmmmw mo coHummoumm muHsomm can coHuosHum
IcH mumavmum coma cmomHm mHmmnmEm mo GOHumoouwm muHsomm cmmzumn QHSmQOHumemII.mI¢ mHnme

147

.meocousm EdHqu suHs mucmEuHmme cH ummsmHn mmz noummmmu

UHmmn coma mHmmsmEm*

 

 

 

 

 

.»anounm 30H :qu mucmEpHmmmc GH umm3oH mm3 noummmmu OHmmn coma mHmmcmEm+
HH wH «H
m m m m o m HH mH mH
I I I I x I Hoo. I Hoo. 30H .cmz anm .saou HmcumuaH
I I I I x x Hoo. I I 30H .nmz anm .5500 cumzuso
I I I I I I Hoo. Ho. Hoo. 30H .nmz smHm .eeoo nnmsmo
I I I I I I Hoo. mo. Ho. 30H .wmz anm .5500 Hmcnmuxm
I Imo. I I I I Hoo. I Hoo. 30H .woz smHm coHumOHnsesou Hmuoe
Imo. I I I I I I Hoo. Hoo. 30H .mm: anm .awm .>HcoI.msomHa
I I I I I x Hoo. Hoo. I 30H .cmz gmHm came :onmsomHa
I I I x I I I Hoo. Hoo. 30H .cmz anm .omm .pmma .msomHa
Hoo. I IHoo. I I I I I I 30H .cmz anm wmmm .umma .msomHo
I I I I I I Hoo. mo. Hoo. 30H .cmz smHm mcHHmHomHa .mmm
I+Ho. I I I I I I Hoo. mo. 30H .omz anm unmaunmmmo .mmm
I I «Ho. I I I Hoo. Ho. I 30H .cmz smHm muHmum>Has .mmm
I I I x x x I I I 30H .602 amHm nonmwmmm UHmmm .mfim
I I I I x I Hoo. I Ho. 30H .cmz anm .umcH .cmuw .mem
«mo. I I I I I I Hoo. Hoo. 30H .wmz smHm .umcH .omums .msm
mo. I I I x x I I I 30H .mm: anm .omm .>chH .mcH
I I I I I x Hoo. Hoo. I 30H .omz anm cmmm .ummo .mcH
I+mo. I I I x I I I Hoo. 30H .mm: smHm .omm .umwa .qu
I I I+Hoo. I I I Hoo. mo. I 30H .omz anm mumm coHumoHHnsm
I I I x I I I Ho. Hoo. 30H .umz gmHm .m .nmw w\.mmm m
I I I x I I I mo. Ho. 30H .Umz smHm MUMHOOEmQ
I I I x I I I Ho. mo. 30H .cmz anm auHsomm H mNHm
«z I I «z I I «2 mo. Hoo. «z mum>Hum oHHnsm uuommsm
Hz «+Hoo. I «z I I «z I Hoo. «z cmumm wowmm mcHumm nmuuumu
m m H; m m H m m H m m H
mmHuommumo mmHuommumo mmHuomwumu mmmuommumu
mumwmmd mwSchm> mUHom Houomm umma
mHanOHNMHmm mHnmcoHumHmm mHanOHumHmm Houomm puma
.mHm “mayo unmoHMHamHm uomuHo .mHm

 

 

.muouomm ummp HsOMImucmau mo
:oHuoswouucH man an UmHMHCOE mm hfiocousm kucmfiuummmc mo COHummoumm hanomm can
noummmmu UHmmn com: UmUmHm mHmmnmfim mo COHumwoumm wuHSOMM :mm3uwn mHnmcoHumHmMII.hI« mHnme

148

mmummw ans m nuHs mucweuummmc GH ummann

.hEocounm mo

mm3 GOHumuamHuo mzonm mocmummmn mumcHHmHomHQ«

 

 

 

 

 

 

 

 

oz h mm» N oz N mm» b oz w mm» H
I I I I I I I I I 30H .Umz anm .8500 HmcnmunH
I I I I I I I I I 30H .602 :mHm .EEOU cum3uao
I I I I I I I I I 30H .Umz anm .3800 Unmzms
I I I I I I I I I 30H .vmz anm .8500 Hmcumgxm
*oz I I mm» I I oz I I 30H .Uwz anm coHuMOHGSEEOU Hmuoa
*oz I I mm» I I oz I I 30H .vmz anm .Evd .>H:D .msomHo
I I I I I I I I I 30H .Umz anm ammo conmdomHo
I I I I I I I I I 30H .Umz anm .omm .ummo .msomHo
*oz I oz mm» I oz oz I 02 30H .Umz anm cmmm .ummo .msomHQ
I I I I I I I I I 30H .cwz smHm mcHHmHomHa .mmm
I I I I I I I I I 30H .wwz anm ucmEuHmmmo .mmm
I I I I I I I I I 30H .cmz anm muHmHm>HcD .mmm
02 I I mm» I I oz I I 3oq .Umz :mHm soumwmmm OHmmm .mEm
«oz I I mm» I I oz I I BOA .cmz anm .umcH .Umno .mfim
«mmw I I 02 I I mm» I I 30H .vmz anm .umcH .cmnmo .mfim
Immw I I mm» I I oz I I 30H .Umz smHm .omm .>HUGH .qu
I I I I I I I I I 30H .cmz anm cmmm .ummo .qu
«oz I I mmw I I 02 I I 30H .Umz smHm .omm .ummo .qu
I I I I I I I I I 30H .mm: anm mumm aoHumoHHnsm
I I I I I I I I I 30H .Umz zmHm m .cmo m\.mmm m
I I I I I I I I I 30H .Umz :mHm mongoosmo
I I I I I I I I I 30H .00: smHm auHsomm d mNHm
«z I I <2 I I <2 I I <2 mum>HHm UHHQDm uuommam
«z I I «z I I Hz I I «z wmumm cmwmm maHumm umuuumu
m N H m N H m N H m N H
mmHHommumo mmHuommqu mmHHommumo mmHuommumo
masocouDN on N uw¢.Mdmmmz N>Eocou54 on Houomm umma
UmHMHmm mHuoman HWM umwsmmm mHnmcoHHMHmm uouomm ummB
wcHHmHomHo .mwm .ummo .mmm mmuw>cH D .mmm

 

 

.muouomm ummu unom
Imucmzu mo :oHuoawouunH man an omHMH©OE mm mfiocoudm Hmucwfiuummmo mo COHummo
Iumm muHsomm cam muHsomm mo mcoHumucmHuo msoum wocmnmumu awm3umn mHzmcoHuMHmmII.mI« mHnme

149

 

 

 

 

 

wH mm @H
N H HH v MH m m CH v
I I mo. x x I I I I 30H .vmz anm .5500 HmcumucH
I I Hoo. I x I I I I 30H .umz smHm .asoo unususo
I I Hoo. x x I I I I 30H .Umz anm .5500 wum3ms
I I mo. x I I I Ho. I 30H .mm: anm .2200 Hanumpxm
I I me. I x I mo. I I 30H .nmz anm coHumoHcsasou Hmuoa
I I Hoo. x I I I Ho. I 30H .w«: gmHm .aw¢ .>st .msomHa
I I Hoo. x I I I mo. I 30H .omz anm cmma GOHmmsomHo
I I Ho. x I I I mo. I 30H .Uwz anm .omm .ummo .msomHQ
I I I x x x I I I 30H .co: anm Ummm .ummo .msomHa
I I I I I x Hoo. Ho. I 30H .uwz anm quHmHomHo .mmm
mo. I I I x I I I Hoo 30H .cmz anm ucmeunmmmo .mmm
I I Hoo. x I I I Hoo. I 30H .omz :mHm auHmum>Hco .mmm
mo. I I I I x I mo. I 30H .cmz smHm noummmmm onmm .mem
I I I x x x I I I 30H .Umz anm .HmGH .cmuw .mfim
I I Hoo. x I I I Hoo. I 30H .cmz smHm .umcH .Umnmo .mEm
I I I x x x I I I 30H .cmz anm .omm .>H©:H .mcH
I I I x I x I Ho. I 30H .mm: smHm wmmm .ummo .mcH
I I Hoo. x x I I I I 30H .mm: amHm .omm .ummo .mcH
I I I I x x Ho. I I 30H .cwz anm mumm coHumoHHnsm
I I I I x x Hoo. I I 30H .Umz anm .m .cmu m\.mmm m
I I I I x x Ho. I I 30H .602 anm homuoofimo
I I I x x I I I Ho. 30H .Umz anm thsomm ¢ mNHm
«z Hoo. I «z I I «z I Hoo. «z mum>Hum oHHnsm uuommsm
«z I I «z I I «z Hoe. Hoo. Hz venom cmwmm mcHumm nmuuumo
m N H m N H m N H m N H
mmHHommumo mmHHommumU mmHHommumo mmHHommumu
.und .cmz .mumma mmancm> mcHom uouomm umma
CH Umwm .ummn mHnmcoHuMme QHSmCOHHMHmm Houomm umma

cqu .5800 umoz unmoHMHcmHm uomuHo .mHm

 

 

.muouUMM ummu HDOMImucm3u mo coHuosc
IouucH man an @mHMHvoE mm >Eocoudm Hmucmsuummmc mo QOHummonwm muHsomm Gum
:mEHHmno ucwfiuummwu map nuHs conmzomHv mo unsofim map cmmzpmn QHSmcoHuMHmmII.mI¢ anme

150

.ucwvam mm3 haocoudm Hmucmfiuummow mo mmnmmo map can mmmuuHEEoo Hmuamepnmmmu
HMHommm mo :wEuHmco may nuH3 GOHmmsomHv mo uGSOEm may :ww3umn mHanOHHMHmH uomuHc ¢«

 

 

 

 

 

m , Hm mH
o N v «H mH mH m m N
I I I x x x I I I 30H .602 anm .5800 HmcumucH
I I I x x x I I I 30A .602 smHm .EEOU unmzudo
I I I I x x Hoo. I I 30H .cmz smHm .EEOU wumzms
I I I I x x Ho. I I 30H .Umz :mHm .EEOU Hmcumuxm
I I mo x I I I mo. I 30H .Umz anm quuMOHGSEEOU Hmuoa
I I Imo x I I I mo. I 30H .cmz anm .eca .>Hco .msomHa
I I I I x x Hoo. I I 30H .Umz anm cmma GOHmmsomHo
I I «mo x I I I Ho. I 30H .Umz smHm .omm .pmma .msomHQ
I I I I x x Ho. I I 30H .omz amHm comm .ummn .muomHo
I Imo. Imo x I I I I I 30H .cmz anm mcHHmHomHo .mmm
I I I x x x I I I 30H .60: zmHm uamfipnmmmo .mmm
I I I I x x mo. I I 30H .cmz anm huHmHm>HCD .mmm
I I I I x x Ho. I I 30H .cmz anm nonmmmmm OHmmm .mfim
I I I x x x I I I 30H .00: anm .umcH .wmnu .mfim
I «Hoe. I x I x I I I 30H .cmz smHm .umaH .omumo .mem
I I I I x x Ho. I I 30H .cmz smHm .umm .>HuaH .mcH
I I I I x x mo. I I 30H .cmz anm cmmm .pmma .mcH
I I I x x x I I I 30H .Umz anm .omm .umma .wcH
I I I x x x I I I 30H .cmz anm mumm coHumoHHnsm
I I I x x x I I I 30H .Umz smHm .m .cmw w\.mmm m
I I I x x x I I I 30H .cmz anm womuuofimo
I I I x x I I I Hoo. 30H .cmz anm muHsomm fl mNHw
«z I I «z x I «z I Ho. «2 mum>Hum oHHnum uuommsm
«z I I «z x x Hz I I Hz cmumm wwwwm mchmm umuuumo
m N H m N H m N H m N H
mmHHommumo meHommumo mmHHommumo mmHnommumo
ucmmmum mmancm> 30H cwmmmmmwn%us¢ Hepomm umma
mHsmcoHumHmm mHsmcoHHMHmm . . . mm uouomm umma
.mHm umnuo ucmonHamHm co: suH3 my a

CH

.OmHQ umoz

 

.muouomm ummu H:OMImucmzp mo coHuosmouucH as» an cmHMHUoE
mm aaocousm Hmucmapummmw mo coHummoumm huHsomm can mmmuuHEEoo Hmucmauumm
Iwu HMHommm mo :mEuHmno nUH3 aonmsome mo unscam may cmmzumn mHnmnoHuMHmmII.oHI< mHnma

151

 

 

 

 

 

H H m 5H mH mH m w m
I I I x x x I I I 30H .602 anm .EEOU HmaumucH
I I mo. x x I I I I 30H .umz anm .esoo wumzuso
I I I x x x I I I 30H .602 anm .8500 chasm:
I I I x x x I I I 30H .cmz anm ..EEOU Hmcumuxm
I I mo. I x I mo. I I 30H .mm: anm quumOHasaaoo Hmuoa
I I I x I x I Ho. I 30H .vmz anm .864 .>HCD .msomHn
I Ho. I x I x I I I 309 .cm2 anm cmma conmaomHn
I I I x x x I I I 30H .wmz zmHm .omm .ummo .mnomHa
I I I x x x I I I 30H .cmz anm cmmm .umma .msomHo
I I I x x x I I I 30H .omz anm mcHHmHomHn .mmm
I I I x I x I mo. I 3oq .Umz anm ucmsunmmmn .mmm
I I I I x x Ho. I I 309 .Umz anm wuHmHm>HcD .mwm
I I I x I x I Ho. I 30H .Umz anm nonmmmmm UHmmm .mfim
I I I x x x I I I 30H .wmz anm .umaH .cmuw .mem
I I I x x x I I I 309 .nm: gmHm .umcH .cmuma .mem
I I mo. I x x I I I 30H .602 anm .omm .>H©cH .mcH
I I I x x x I I I 30H .omz smHm cmmm .umma .maH
I I I x x x I I I 30H .602 30H: .omm .ummo .ucH
I I I x x x I I I 30H .Uwz anm mumm GOHumoHHnsm
I I I I x x mo. I I 30H .wmz 30H: .m .cmo w\.mwm m
I I I x x I I I mo. 30H .wwz :mHm homuoofima
Ho. I I I x I I I mo. 30H .wmz anm aanomm « muHm
I I I «z x I «z I mo. «2 mum>Hum oHHnsm uuommsm
I I I «z I x «2 mo. I «z mmumm nmwmm mcHumm umuuumo
m N H m N H m N H m N H

mmHHommumo mmHuommumo mmHuommumo mmHHommHMU

INEocousd mmschm> mcHom Houomm umme
.Umz suHB .mumma mHanOHumHmm mHnmaoHUMHmm uouomm umma

CH .8500 umoz pamonHamHm mmnm>cH .mHm

 

 

.muouomm umwu HSOMImucw3u mo onuonUouucH
man an UmHMHGQE mm >80:0p9m Hmuameunmmww mo coHummonm MuHaomm can name
qummmo mo mUHmuso muHsomm nuH3 GOHmmsoch mo “macaw may :mm3uwn mHanOHumHmmII.HHI¢ mHQma

152

 

 

Hoo. W m««« Ho. W QHH mo. M Q«

. bmmH
«ON + mmwumma .4.m ou mumswmuw OHumm
ImH.+ hmmH auHsomm H\.H.z
Im~.+ Im~.+ nmmH muHsomm H\.n.nm
Imm.I IHm.+ «Hm.+ INN.+ hmmH muHaomm H\.o.nm
*NH.+ «mN.+ mumm mmcommmm
«Ham.+ mmHummum huHsomm How cnmoaou
«mm.+ .>H:D mnHmuso .ummo mchoEoum m
IHm.I nmmHIhmmH oHumm H HH mmcmnu w
«HN.I nmmH muHsomm Hmuoe .mummH 302 w
«Hmv.+ *mN.+ mmOHmmo mum>HHm nqu muHsomm d w
ImvnI nmaH muHsomm H w
HNm.+ nmmH muommmmonm HHsm w
«mm.+ bmmH Aunommsmv mwcsm maHumnwmo
«Nm.I hmmH Aunommamv mocam mcHumummo
«Hmv.+ «hN.+ Hmnfimz muHsomm ¢\wcsm Hmumcmw m
«om.4 . .Hmnamz wuHaomm H\£onmmmmm m
Ivm.+ I-.+ amasmz muHsomm H\Hmuoa w
«««mm.I ««*ov.I huHme>HGD w mmmHHoo mo hfiocousd
IINH.I I*om.I *Hm.I IIImm.I mmmHHoo mo msoaousH
«mm.+ ummcmso ma uoz stosm meocousH
IHov.I cmHHouucoo an oHsosm maocousH

*«Hmm.I H0N.+ UHmmm comb mammnmamsmwmmwmm
. . mmum

. . coHposuumcH mumnvmnm
«mN I *«mN I Inmwco comb mHmmcmfim wmuummmnm
HhH.+ mucmwsum mumsvmuw mocmnHmaH
.«Hm.I «HHH.I *HHm.I «gHH.I H*«mm.I mmmuuHEEou muHmum>HaD mucmsHmcH
. «Hov.+ «««mm.+ monmz :onHoma GH mommoofimo

mEOGOpafi maocousd mfiocouSH .mummo .mumma .mummo
wmummIcoz wmumm m QHHHM>
30H schmz anm umuuumo umuuumo HHH H

 

 

.mmHanum> HMGOHunww can afiocousm Hmucmfiuummwc mo
mmumwv may cmm3umn mucmHonmmoo QOHuMHmuuoo Hmwuo xcmu unfiummmm HGMOHMHcmHmII.NHI4 mHQMB

BIBLIOGRAPHY

BIBLIOGRAPHY
29.9132

Anderson, G. Lester. "The Organizational Character of
American Colleges and Universities." The Study-of
Academicgdministration. Edited by Terry F.
Lunsford. Boulder, Colorado: Western Interstate
Commission for Higher Education, 1963.

 

Blau, Peter M., and Richard Scott. Egrmal Organizations:
A Comparative Approach. San Francisco: CHandIer
Publishing Company, 962.

Caplow, Theodore, and Reece J. McGee. The Academic Market-
place. New York: Basic Books,_1958.

 

 

Cartter, Allan M. An Assessment of Quality in Graduate
Education. Washington, D.C.: American Council
on Education, 1966.

Demareth, Nicholas J.; Robert W. Stephens; and R. Robb
Taylor. Power, Presidents, and Professors. New
York: Basic Books, 1967.

Dressel, Paul L.; F. Craig Johnson; and Philip M. Marcus.
The Confidence Crisis. San Francisco: Jossey-
Bass Inc., 1970.

Fattu, Nicholas A. "Research Organizations." Encyclopedia
of Educational Research: Fourth Edition. Edited
by Robert L. EbeIZ The MacMillan Company, 1969.

 

 

Glock, Charles Y. "Survey Design and Analysis in Sociol-
ogy." Survey Research in the Social Sciences.
Edited by Charles Y. Glock. New York: Russell
Sage Foundation, 1967.

Gross, Edward, and Paul V. Grambsch. Universitinoals and
Academic Power. Washington, D.C.: American
CounciI on Education, 1968.

 

 

Hagstrom, Warren 0. The Scientific Community. New York:
Basic Books, 19 5.

 

153

154

Hefferlin, JB Lon. The Dynamics of Academic Reform. San
Francisco: Jossey-Bass Inc., 1969.

Jencks, ChristOpher, and David Riesman. The Academic
Revolution. New York: Doubleday and Company,
Inc., 1968.

 

 

Katz, Fred E. Autonomy and Organization: The Limits of
Social Control. New York: Random House, 1968.

 

 

Katz, Fred E. "The School as a Complex Social Organiza-
tion." Sociolo of Education. Edited by Ronald
M. Pavalko. Itasca, Ill1n01s: F. E. Peacock
Publishers, Inc., 1968.

Kerr, Clark. The Uses of the University. Cambridge,
Massachusetts: Harvard University Press, 1964.

 

Kornhauser, William. Scientists in Industry. Berkeley:
University of CaliforniaT Press, 1963.

 

Lorimer, Margaret F., and Paul L. Dressel. "Faculty Char-
acteristics--College and University." Encyclopedia
of Educational Research: Fourth Edition. E 1te

By Robert L. EEeI. The MacMillan Company, 1969.

 

McConnell, T. R. "Needed Research in College and Univer-
sity Organization and Administration." The Study
of Academic Administration. Edited by Terry F.
Lunsford._Boulder, Eblorado: Western Interstate
Commission for Higher Education, 1963.

 

March, James G. Handbook_of Organizations. Chicago:
Rand McNally and Company, 1965.

 

Millett, John D. The Academic Community. New York:
McGraw-Hili Book Company, Inc., 1962.

 

Millett, John D. "College and University Administration."
Enc clo edia of Educational Research: Fourth Edi-
t1on. Edited by Robert L. EbeI. The MacMilIan
Company, 1969.

 

 

Page, Charles H. "Bureaucracy and Higher Education."
Sociology of Education. Edited by Ronald M.
PavaIko. Itasca, Ill1nois: F. E. Peacock Pub-
lishers, Inc., 1968.

Pelz, Donald C. Motivation of the Engineering and Research.
Specialist. General Management Series, Number 186.
New York: American Management Association Inc.,
1957.

 

 

155

Pelz, Donald C., and Frank M. Andrews. Scientists in Or-
ganizations: Productive Climates for Research and
Development. New York: John Wiley and Sons, Inc.,
1966.

 

 

Peterson, R. E., and D. E. Loye, eds. Conversations Toward
a Definition of Institutional Vitality. Princeton:
EducationaIiTesting Service, 1966}

 

 

 

Rosenberg, Morris. The L0 ic of Survey Analysis. New
York: Basic Books Inc., 1968.

Trow, Martin. "Education and Survey Research." Survey
Resgarch in the Social_Scienqg§. Edited by Charles
Y. Glock. New York: Russell sage Foundation,
1967. I

 

Articles and Periodicals

 

Andersen, Kay J. "The Ambivalent Department." Educational

 

Record, IL (Spring, 1968), 206-13.

Ashby, Eric. "Anatomy of the Academic Life." Educational
Record, IIL (Winter, 1967), 45-50.

 

Beach, Mark. "Professional Versus Professorial Control of
Higher Education." Educational Record, IL (summer,

 

Bolman, Frederick deW. "Needed Research in Administration
of Higher Education." Educational Record, XLVI
(Spring, 1965), 166-76.

 

Clark, Terry. "Institutionalization of Innovations in
Higher Education." Administrative Science Quart-
erly, XIII (June, 19687, 1-25.

 

Crane, Diana. "Social Structure in A.Group of Scientists:
A Test of the 'Invisible College' Hypothesis."
American Sociological Review, XXXIV (June, 1969),

 

Dressel, Paul L.; F. Craig Johnson: and Philip M. Marcus.
"Departmental Operations: The Confidence Game."
Educational Record, L (Summer, 1969), 274-78.

 

Dressel, Paul L., and Donald J..Reichard. "The University
Department: Retrospect and Prospect." Journal of
Higher Education, XLI (May, 1970), 387-402.

 

 

156

Engel, Gloria V. "Professional Autonomy and Bureaucratic
Organization." Administrative Science Quarterly,

 

Glaser, Barney G. "The Local-Cosmopolitan Scientist."
American Journal of Sociology, LXIX (November,
1963), 249-59.

Gouldner, Alvin W. "Cosmopolitans and Locals: Toward an
Analysis of Latent Social Roles--I." Administra-
tive Science Quarterly, II (December, 19575, 581-
306.

Hall, Richard H., and Charles R. Tittle. "A Note on Bu-
reaucracy and Its Correlates." The American
Journal of Sociology, LXXII (November, 1966),
267-72.

 

Hill, Winston W., and Wendell L. French. \"Perceptions of
the Power of Departmental Chairmen by Professors."
Administrative Science Quarterly, XI (March, 1967),
548-74. ii

Kristol, Irwin. "A Different Way to Restructure the Uni-
versity." The New York Times Magazine, December
8, 1968, p. 56.

Lewis, Lionel S. "On Prestige and Loyalty of University
Faculty." Administrative Science Quarterly, XI

Litchfield, Edward H. "Organization in Large American
Universities: The Faculties." Journal of Higher
Education, XXX (October, 1959), 353:364.

 

Litchfield, Edward H. "Organization in Large American
Universities: The Administration." Journal of
Higher Education, XXX (December, 19595, 391—553.

Monson, Charles H;,Jr. "Metaphors for the University."
Educational Record, IIL (Winter, 1967), 22-29.

Moran, William E. "The Study of University Organizations."
Journal of Higher Education, XXXIX (March, 1968),
1444151.

Olive, Betsy Ann. "The Administration of Higher Education:
A Bibliographical Survey." Administrative Science
Quarterly, XI (March, 1967), 67I;77T*

157

General References

Clark, Burton R. "Organizational Adaptation to Profes-
sionals." Professionalization. Edited by Howard
M. Vollmer and Donald L. Milli. Englewood Cliffs,
New Jersey: Prentice Hall, Inc., 1966. .

Corson, John J. Governance of Coll_ges and Universities.
New York: McGraw-Hill Book Company, Inc., 1960.

Ebel, Robert L., ed. Encyclgpedia of Educational Research:
Fourth Edition. The MacMillan Company, 1969..

 

 

 

Glock, Charles Y., ed. Surve Research_in the Social
Sciences. New York: Russell Sage Foundation,

Gouldner, Alvin W. "Cosmopolitans and Locals: Toward an
Analysis of Latent Social Roles--II." Administra-
tive Scienceguarterly, II (March, 1958), 144-80.

 

 

Gross, Edward. "Universities as Organizations: A Research
Approach." American Sociological Review, XXXIII
(August, 1968), 518-41:

 

Haas, Eugene, and Linda Collen. "Administrative Practices
in University Departments." Administrative Science
Quarterly, VIII (June, 1963), 44-60.

 

Hall, Richard H. "Professionalization and Bureaucratiza-
tion." American Sociological Review, XXXIII (Oc—

 

Hungate, Thad L. Management in Higher Education. New
York: Bureau of Publications, Teachers College,
Columbia University, 1964.

 

Knowles, Asa 8., ed. Handbook of College and Universit
Administration: Academic. New York: McGraw-H111
Book Company, 1970.

 

 

Knowles, Asa 8., ed. Handbook of College and Universit *
Administration: General. New York: McGraw-Hill
Book Company, 1970.

 

 

Lunsford, Terry F., ed. The Study of Academic Administra-
tion. Boulder, Colorado: Western Interstate Com-
mission for Higher Education, 1963.

 

 

Pavalko, Ronald M., ed. Sociolo of Education. Itasca,
Illinois: F. E. Peacock Publishers, Inc., 1968.