Group Title: comparative and programmatic approach to organizational control
Title: A comparative and programmatic approach to organizational control
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00098304/00001
 Material Information
Title: A comparative and programmatic approach to organizational control a case-study of two hospitals
Alternate Title: Organizational control, A comparative and programmatic approach to
Physical Description: xxi, 320 leaves : diagrs. ; 28cm.
Language: English
Creator: Corum, B. H., 1933-
Publication Date: 1975
Copyright Date: 1975
 Subjects
Subject: Hospitals -- Administration   ( lcsh )
Organization   ( lcsh )
Management thesis Ph. D   ( lcsh )
Dissertations, Academic -- Management -- UF   ( lcsh )
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Thesis: Thesis--University of Florida.
Bibliography: Bibliography: leaves 310-318.
Statement of Responsibility: by B. H. Corum.
General Note: Typescript.
General Note: Vita.
 Record Information
Bibliographic ID: UF00098304
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000167388
oclc - 02855115
notis - AAT3778

Downloads

This item has the following downloads:

comparativeprogr00coru ( PDF )


Full Text
















A COMPARATIVE AND PROGRAmmATIC
APPROACH TO ORGANIZATIONAL CONTROLs
A CASE-STUDY OF TWO HOSPITALS






By



B.H. CORUM.-


A DISSERTATION PRESENTED TO THE GRADUATE
COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL
FULFILLMENT OF THE REQUIREMENTS FCR THE DEGREE OF
DOCTOR OF PHILOSOPHY




I,


UNIVERSITY OF FLORIDA"
1975






























Dedicated
to
My Family













ACKNOWLEOGEMENTS

After more than twenty academic years of exposure to

various educational environments the author finds himself

without the ability to recall all of those who have con-

tributed to his learning experiences. Therefore any attempt

to enumerate the contributors would obviously omit some

who have unintentionally been forgotten. Recent events

however, dictate the expression of sincere appreciation to

the members of the supervisory committee for their guidance

and patience in directing this particular research project.

Acknowledgement is also gratefully given to the admin-

istrators and staffs of the hospitals which participated

in this study, and to Dr. Harry M. Hughes and the staff of

the Biometrics Division of the USAF School of Aerospace

medicine, Brooks Air Force Base, Texas for their guidance

and computer support during the interpretation and analyses

of the research data which had been collected.

Lastly, for the author's wife Carol and daughter

Renee words are not expressive enough to describe the

feeling held for their unending contributions of patience,

understanding and support from the beginning to the end

of this undertaking.













TABLE OF CONTENTS

Page

ACKNOWLEDGEMENTS .................. ................ iii

LIST OF TABLES .... ....................... ............ ix

LIST OF FIGURES..................................... xvi

ABSTRACT................................... ........ xviii

Chapter

I. INTRODUCTION... .............. ................ 1

The Scope of the Problem..................... 1

Overview....... .. ... ...... ...... ... 1

The Problem Area......................... 6

Purpose of the Study........................ 8

Significance of the Study.................... 9

methodology................. ............. 11

Assumptions..... ... ................ ......... 16

Organization of the Study.................... 17

Notes ................. ........... .. .. e.. .... 19

II. THE NATURE OF CONTROL......................... 22

Overview.... .. ........ ........... .. ... ..... 22

What is Control....................... .. ... 23

What are its Functions....................... 30

What are the Bases of Control......... ...... 31

Summary ...................... ..*..* ...** *... 38
Notes..................a... ..... ...... ...... 39







Chapter Page

III.. THE CONTROL GRAPH AND ITS APPLICATION TO THE
STUDY OF CONTROL IN ORGANIZATIONS.......... 43

The Control Graph......................... 43

Applications............................... 51

Prototypes.................. ............ 51

Distribution of Control................ 51

Total Amount of Control................ 52

Perceived and Desired Control.......... 55

Control and Uniformity.................. 57

Organizational Type Comparisons........ 59

Bases of Control or Power............... 60

Summary................ ..... .......... ...... 62

Notes..................... ............ ..... 63

IV. RESEARCH DESIGN, METHOD, AND PROCEDURE ...... 67

Overview........................**......... 67

Selection of Hospitals................... 68

Selection of the Individual Groups and
Respondents.... ."..... .................. 77

Research Instruments and Data Collection... 79

Data Collection....................... 88

Response Rates.. ...................... 93

Areas of Investigation, Hypotheses Tested
and methods of measurement........ ....... 94

Prototypes......... .......... .......... 97

Perceived control.................. 97

Desired control.................... 98







Chapter Page

Distribution of Control................ 99

Influence of the Medical Staff......... 101

Bases of Control or Power and Satis-
faction.........** ................ 101

methodological Limitations............... 104

Summary .,......... ..... ....... ....... 105

Notes........ .......... ..... ... ......... 108

V. THE RESEARCH FINDINGS ............. ........ 112

Introduction........ ... .......... ......... 112

Prototypes .......... .., ,,......... .. 116

Perceived Control................... 116

Desired Control........ ............ 119

Perceived vs. Desired Control.......... 124

Statistical Analyses.................. 128

Church affiliated hospital
(Hospital A)........ ........... 130

Nonchurch affiliated hospital
.(Hospital B) ..................... 134

Comparison between hospitals/ques-
tions (perceived and desired).... 134

Distribution of Control.................. 139

Statistical Analyses................... 146

Church affiliated hospital
(Hospital A)..................... 146

Nonchurch affiliated hospital
(Hospital B).................... 147

Comparison between hospitals/ques-
tions (7-12)..................... 147

vi







Chapter Page

Active vs. passive control......... 156

Note on active vs. passive control. 157

Bases of Control and Satisfaction.......... 158

Bases of Control........... ....... .. 158

Satisfaction........................... 162

Influence of the medical Staff............. 166

Perceived Control............ .......... 166

Desired Control ................ ... 167

Active vs. Passive Control............. 169

Summary........................ ... . 170

Notes............ .. ...... ... .... ... 172

VI. SUMMARY, CONCLUSIONS AND IMPLICATIONS FOR FUR-
THER RESEARCH............................... 173

Summary..... ................ ......... ... 173

Conclusions................................ 177

On Patterns of Control................ 177

On the Bases of Supervisory Control and
Satisfaction . ......................... 178

On the Influence of the Medical Staff.. 179

Implications for Further Research.......... 180

Patterns of Control................... 180

Bases of Control....................... 181

Influence of the medical Staff......... 181

Summary.....o.. .. ...o.... ........ ... 182







Chapter Page

APPENDICES ............ ............ .............. ..... 183

Appendix A.... .................................. 184

Appendix B ........... .... ... ............ .... ... 186

Appendix C......... ........................... 203

Appendix 0............ ...... ............... ... 290

BIBLIOGRAPHY.. ............ ...... ..... ... ......... 310

BIOGRAPHICAL SKETCH................................. 319


viii













LIST OF TABLES


Table Page

1. Attained Gross and Net Response Rates for all
Respondent Groups Combined, but Separately for
Each Participating Hospital...............* ...... 95

2. Final Number of Respondents from Each Hierar-
chical Level and Each Participating Hospital..... 96

3. Number of Participants in Each Response Group
at Each Hospital................... .............. 113

4. Research Question Titles.............. ............ 114

5. Repeated measurements for Analysis of Variance
(ANOVA)- Church Affiliated Hospital.............. 131

6. Mean for Each Hierarchical Level and Question
for the Church Affiliated Hospital............... 133

7. Repeated measurements for Analysis of Variance
(ANOVA)- Nonchurch Affiliated Hospital.......... 135

8. Mean for Each Hierarchical Level and Question
for the Nonchurch Affiliated Hospital............ 137

9. Sources of Variation for Disproportionate
Analysis of Variance on Data for Both Hospitals/
Question........................ ............ 138

10. Response Group by Question by Hierarchical Level
Means Both Hospitals ........................... 140

11. Response Group by Question by Hierarchical Level
means Both Hospitals ......................... 141

12. Repeated measurements for Analysis of Variance
(ANOVA)- Church Affiliated Hospital.............. 148

13. Question by Hierarchical Level means for the
Church Affiliated Hospital....................... 150







Table Page

14. Repeated measurements for Analysis of Variance
(ANOVA)- Nonchurch Affiliated Hospital........... 151

15. Question by Hierarchical Level means for the
Nonchurch Affiliated Hospital.................... 153

16. Active vs. Passive Control Means, Plus Asso-
ciated Differences, at the Church Affiliated
Hospital............ .... .................... ... 159

17. Active vs. Passive Control Means, Plus Asso-
ciated Differences, at the Nonchurch Affiliated
Hospital........... ........... .................. 160

18. Ranking of the Bases of Control (Power)......... 163

19. Mean Ratings of Bases of Control (Power)........ 164

20. Correlations with Satisfaction measures.......... 165

21. Responses as to Amount of Perceived Control by
Hierarchical Level -Church Affiliated Hospital... 204

22. Responses as to Amount of Perceived Control by
Hierarchical Level Nonchurch Affiliated Hos-
pital.............. ......... ......... ............ 208

23. Responses as to Amount of Desired Control by
Hierarchical Level Church Affiliated Hospital.. 212

24. Responses as to Amount of Desired Control by
Hierarchical Level Nonchurch Affiliated Hos-
pital .......... ........ ...... .............. ..... 216

25. Response Group Means Church Affiliated Hos-
pital ....... ............ .... .. .. .... ......... 220

26. Mean for Each Question by Response Group Church
Affiliated Hospital .............................. 221

27. Response Group and Hierarchical Level Means -
Church Affiliated Hospital...................... 222

26. Question and Hierarchical Level Means Church
Affiliated Hospital.............................. 223

29. Response Group and Question and Hierarchical
Level Means Church Affiliated Hospital......... 224









30. Response Group means Nonchurch Affiliated
Hospital...... *...... .. .......... ....... .... ... 225

31. Mean for Each Question by Response Group Non-
church Affiliated Hospital...................... 226

32. Response Group and Hierarchical Level Means Non-
church Affiliated Hospital ............ .......... 227

33. Question and Hierarchical Level means Nonchurch
Affiliated Hospital ............................. 228

34. Response Group and Question and Hierarchical
Level Means Nonchurch Affiliated Hospital...... 229

35. Combined Responses as to the Amount of Active
Control by Hierarchical Level Church Affili-
ated Hospital....... ........................... 230

36. Combined Responses as to the Amount of Active
Control by Hierarchical Level -, Church Affili-
ated Hospital.................................. 231

37. Combined Responses as to the Amount of Active
Control by Hierarchical Level Church Affili-
ated Hospital................ ................... 232

38. Combined Responses as to the Amount of Active
Control by Hierarchical Level Church Affili-
ated Hospital ............................ ...... 233

39. Combined Responses as to the Amount of Active
Control by Hierarchical Level Church Affili-
ated Hospital................ ................. 234

40. Combined Responses as to the Amount of Active
Control by Hierarchical Level Church Affiliated
Hospital............ ................... ....... 235

41. Combined Responses as to the Amount of Passive
Control by Hierarchical Level Church Affili-
ated Hospital................. .................... 236

42. Combined Responses as to the Amount of Active
Control by Hierarchical Level Nonchurch
Affiliated Hospital............... .............. 237


Table


Page








Table Page

43. Combined Responses as to the Amount of Active
Control by Hierarchical Level Nonchurch
Affiliated Hospital.. ....................... ..... 238

44. Combined Responses as to the Amount of Active
Control by Hierarchical Level Nonchurch Affil-
iated Hospital........................ ............ 239

45. Combined Responses as to the Amount of Active
Control by Hierarchical Level Nonchurch Affil-
iated Hospital.. ................................ 240

46. Combined Responses as to the Amount of Active
Control by Hierarchical Level Nonchurch Affil-
iated Hospital. ................... ..... ... .. 241

47. Combined Responses as to the Amount of Active
Control by Hierarchical Level Nonchurch Affil-
iated Hospital................................... 242

48. Combined Responses as to the Amount of Passive
Control by Hierarchical Level Nonchurch Affil-
iated Hospital............................ ...... 243

49. Response Group means Church Affiliated Hos-
pital ............ .... ........ ........ ... ....... 244

50. Mean for Each Question by Response Group Church
Affiliated Hospital.............................. 245

51. Response Group and Hierarchical Level means -
Church Affiliated Hospital....................... 246

52. Question and Hierarchical Level Means Church
Affiliated Hospital........................... 247

53. Response Group and Question and Hierarchical
Level Means Church Affiliated Hospital.......... 248

54. Response Group Means Nonchurch Affiliated Hos-
pital........... ........ .... ......... .... .. 250

55. mean for Each Question by Response Group Non-
church Affiliated Hospital.......... ............ 251

56. Response Group and Hierarchical Level means -
Nonchurch Affiliated Hospital................... 252










57. Question and Hierarchical Level means Non-
church Affiliated Hospital................. ..... 253

58. Response Group and Question and Hierarchical
Level means Nonchurch Affiliated Hospital...... 254

59. Two-way Disproportionate with Repeated measure-
ments Analysis of Variance Test of Main Effects
Assuming Zero Interactions Question 5.......... 256

60. Two-way Disproportionate with Repeated measure-
ments Analysis of Variance Test of lain Effects
Assuming Zero Interactions Question 6......... 258

61. Two-way Disproportionate with Repeated measure-
ments Analysis of Variance Test of main Effects
Assuming Zero Interactions Question 7.......... 260

62. Two-way Disproportionate with Repeated Measure-
ments Analysis of Variance Test of Main Effects
Assuming Zero Interactions Question 8........ 262

63. Two-way Disproportionate with Repeated measure-
ments Analysis of Variance Test of main Effects
Assuming Zero Interactions Question 9.......... 264

64. Two-way Disproportionate with Repeated measure-
ments Analysis of Variance Test of Main Effects
Assuming Zero Interactions Question 10..... ... 266

65. Two-way Disproportionate with Repeated measure-
ments Analysis of Variance Test of main Effects
Assuming Zero Interactions Question 11........ 268

66. Two-way Disproportionate with Repeated measure-
ments Analysis of Variance Test of Main Effects
Assuming Zero Interactions Question 12......... 270

67. Number of Responses by Priority for Each of the
Five Bases of Control Church Affiliated Hos-
pital ........ .. . ................ .. .. ....... 272

68. Number of Responses by Priority for Each of the
Five Bases of Control Nonchurch Affiliated
Hospital.**.....*. .... .............,,. ............ 274


xiii


Table


Page







Table Page

69. Number of Responses as to Satisfaction With the
Way Immediate Supervisors Were Doing Their Jobs
by Response Group Church Affiliated Hospital... 276

70. Number of Responses as to Satisfaction With the
Way Immediate Supervisors Were Doing Their Jobs
by Response Group Nonchurch Affiliated Hos-
pital......................... ....... ...... ... 277

71. Correlation Coefficients......................... 278

72. Correlation Coefficients........................ 279

73. Correlation Coefficients......................... 280

74. Correlation Coefficients........................ 281

75. Correlation Coefficients ....................... .. 282

76. Responses as to the Amount of Control the medical
Staff has on How the Hospital Functions Church
Affiliated Hospital.............................. 283

77. Responses as to the Amount of Control the fledical
Staff Should Have on How the Hospital Functions -
Church Affiliated Hospital..................... 284

78. Responses as to the Amount of Control the medical
Staff has on How the Hospital Functions Non-
church Affiliated Hospital............. .......... 285

79. Responses as to the Amount of Control the medical
Staff Should Have on How the Hospital Functions -
Nonchurch Affiliated Hospital.................... 286

80. Combined Responses as to the Amount of Active
Control Exercised by the medical Staff Church
Affiliated Hospital.................... ......... 287

81. Combined Responses as to the Amount of Active
Control Exercised by the medical Staff Non-
church Affiliated Hospital....................... 288

82. Combined Responses by Organization as to the
Amount of Passive Control Exerted Upon the Med-
ical Staff ........ ............. ................ 289







Table Page

83. Contingency Table 1 (Church Affiliated Hospital). 294

84. Contingency Table 2 (Nonchurch Affiliated Hos-
pital).. ......... ........................ 295

85. Contingency Table 3 (Church Affiliated Hospital). 296

86. Contingency Table 4 (Nonchurch Affiliated Hos-
pital) ............. .......... ........... .. .... 297

87. Contingency Table 5 (Church Affiliated Hospital). 298

88. Contingency Table 6 (Nonchurch Affiliated Hos-
pital) .............. ............. ... .......... 299

89. Contingency Table 7 (Church Affiliated Hospital). 300

90. Contingency Table 8 (Nonchurch Affiliated Hos-
pital) ............... ........................... 301

91. Contingency Table 9 (Church Affiliated Hospital). 302

92. Contingency Table 10 (Nonchurch Affiliated Hos-
pital)....... ...................... .. ...... .. ... 303

93. Contingency Table 11 (Church Affiliated Hos-
pital) ....... ..... ............. .... .......... .. 304

94. Contingency Table 12 (Nonchurch Affiliated Hos-
pital)....... .... .......... ..................... 305

95. Contingency Table 13 (Church Affiliated Hos-
pital)....................................... .... 306

96. Contingency Table 14 (Nonchurch Affiliated Hos-
pital) ................... .................. .. 308













LIST OF FIGURES


Figure Page

1. The Control Process ........... ...... ........ .. 28

2. The Control Graph.. ................ ... ... ...... 44

3. Autocratic model .... ..................... ...... 45

4. Democratic Model................................. 45

5. Dictatoral Model .... ....... .......................... 46

6. Anarchic and Polyarchic Models................... 46

7. Increase in Total Amount of Control............. 55

8. Perceived Control Curves for the Church Affiliated
Hospital as Reported by the Different Response
Groups ............................... ............ 117

9. Perceived Control Curve for the Church Affiliated
Hospital Based Upon the Combined Responses of all
the Participants From That Organization.......... 118

10. Perceived Control Curves for the Nonchurch Affil-
iated Hospital as Reported by the Different
Response Groups............... ................. 120

11. Perceived Control Curve for the Nonchurch Affil-
iated Hospital Based Upon the Combined Re-
sponses of all the Participants From That
Organization .................................... 121

12. Desired Control Curves for the Church Affiliated
Hospital as Reported by the Different Response
Groups............ .... ... ...... ..... ..... ........ 122

13. Desired Control Curve for the Church Affiliated
Hospital Based Upon the Combined Responses of
all the Participants From That Organization...... 123








Figure Page

14. Desired Control Curves for the Nonchurch Affil- ,
iated Hospital as Reported by the Different
Response Groups...............*. *................ 125

15. Desired Control Curve for the Nonchurch Affil-
iated Hospital Based Upon the Combined Responses
of all the Participants From That Organization... 126

16. Perceived vs. Desired Control Curves for the
Church Affiliated Hospital Combined Responses.. 127

17. Perceived vs. Desired Control Curves for the Non-
church Affiliated Hospital Combined Responses.. 129

18. Active vs. Passive Controls Church Affiliated
Hospital ......................... ................ 142

19. Active vs. Passive Controls Nonchurch Affil-
iated Hospital .................................. 143

20. Question by Hierarchical Level means for the
Church Affiliated Hospital...................... 154

21. Question by Hierarchical Level means for the
Nonchurch Affiliated Hospital................... 155


xvii













Abstract of Dissertation Presented to the Graduate
Council of the University of Florida in Partial Fulfillment
of the.Requirements for the Degree of Doctor of Philosophy



A COMPARATIVE AND PROGRAMMATIC APPROACH TO
ORGANIZATIONAL CONTROLi A CASE-STUDY OF TWO
HOSPITALS



By

B.H. Corum

June 1975



Chairman: Or. William V. Wilmot, Jr.
major Oepartmenti Management



Introduction

Despite the importance of the hospital in today's

society relatively little research has been performed on

the distribution and balance of influence in the hospital

as an organization. On the other hand the process by which

members determine or influence how things get done in an

organization has been the subject of extensive research

in such organizations as unions, voluntary associations,

colleges, business and industrial organizations. Studies

similar to the ones which have been performed in these

other organizations simply do not exist for hospitals.

xviii







Purpose of the Study

Therefore the purpose of this case-study was twofold.

First, was to compare the patterns of control by hierarch-

ical level in two nongovernmental, not-for-profit hospitals

(one church operated, the other nonchurch operated) to

determine what, if any, significant differences existed

between the two. This was done by assessing (a) perceived

vs. desired control and (b) active vs. passive control.

Second, was to determine the relationship between the

perceptions of individuals at various hierarchical levels

(using the French and Raven fivefold typology) as to the

bases of supervisory control in each institution and then

to correlate those with their perceptions as to satisfac-

tion with the way immediate supervisors were doing their

jobs.


Research methodology

The methodology was essentially the same as that

used in several other studies of organizational control

in which the control graph approach and the French and

Raven fivefold typology were utilized. Data were gathered,

from members of various groups and hierarchical levels in

the two institutions, by the use of an influence question-

naire. These data were then subjected to both descriptive

and statistical analysis.







Conclusions



On Patterns of Control

Based upon the research findings, concerning the

patterns of control which were addressed by this case-study

it was concluded that there were no significant differences

in the patterns of control in the two hospitals which

participated in the study. This conclusion was supported

by the following findings.

1. Both hospitals were perceived of as being&

a. characterized by oligarchicc" control

structures; and

b. in need of a more equalitarian distribution

of control than that which existed in them.

2. The receipt of control in both organizations was

perceived of as being a more general principle than the

exercise of control.


On the Bases of Supervisory Control and Satisfaction

Why did the respondents comply with the requests of

their organizational superiors? Of the five bases of

supervisory control measured in these two institutions,

it was concluded that the single most important reason was

because subordinates respected the competence and good

judgment of their superiors (expert control). The least

likely reason for compliance was because the supervisors







could apply pressure or penalize those who did not cooperate

(coercive control).

Additionally it was concluded that the strongest and

most consistently positive correlations existed between

expert control and satisfaction with the way immediate

supervisors were doing their jobs. Conversely the most

negative correlations existed between coercive control and

satisfaction with the way immediate supervisors were doing

their jobs.


On the Influence of the Medical Staff

The medical staffs in both hospitals were perceived

of as being very influential groups which exercised more

control within each of the organizations than was generally

believed they should. Additionally, while they were

perceived of as exercising a "great deal" of control, it

was the opinion of the majority of the respondents from

each institution that they were subject to only "some"

control within the organizations, a condition which is

considered to be a potential source of tension and/or

misunderstanding.













CHAPTER ,I


INTRODUCTION



The Scope of the Problem



Overview

Historians have tried over the years to give an ade-

quate definition of a hospital. All definitions have nec-

essarily varied with the purposes of the hospital. Pur-

poses have changed with the needs of the people and with

the state of medical knowledge of the times when the defi-

nition was made.

One thing is certain, hospitals have always existed

in some form or other.1 "There has always been some place

where the tired, the sick, the injured, the poor, the aged,

the destitute and the disabled could go to rest, to repair

their sick and broken bodies, to relieve their pain and dis-

comfort, to receive solace and comfort, or to pass their

last few moments on earth."2

Regardless of size or purpose, a hospital must neces-

sarily follow a basic pattern of organization just as any

other business or industry if it is to achieve its purposes.

There are many types of systems and organizations in







the hospital field which exercise control over and assume

responsibility for the functioning of a hospital, but ac-

cording to Or. Charles U. Letourneau;

All systems and organizations conform basically
to the four major functions of government of any kind
of organization. These areas
1. legislation
2. execution
3. administration
4. evaluation

These four functions of government are most commonly
associated with the constitutional law applicable to
government organizations but the principles apply
equally well to industry, to business and to educational,
religious and hospital institutions. A hospital may be
a part of a vast institutional system or may be a small
local institution serving a relatively isolated commu-
nity.3

At the end of the The Community General Hospital, its

authors, Georgopoulos and mann, states

The community general hospital could easily claim
the dubious honor of being one of the least researched
modern large-scale organizations. In spite of its
crucial function of aiding the integration and stability
of society, through the maintenance of a level of health
that permits other social institutions to accomplish
their objectives, and in spite of its far-reaching im-
pact upon nearly every facet of everyday life particu-
larly our economy, standards of living and community
welfare the community general hospital has not received
more than a fraction of the scientific attention that
its importance as an organization would warrant. As yet
our understanding of its functioning, problems and
characteristics is extremely limited... and the same is
true, only more so, regarding comparative studies of
hospitals.4

The significance of this comment cannot be allowed to

escape attention, since research in the field of health

itself the diagnosis, treatment and prevention of disease -

must rank among the oldest of guests for security, not only

among the sciences as they are established today, but in








the previous epochs of alchemy, magic and superstition

through which they all, including medicine, have passed.

Why, when the history of medicine itself is so richly docu-

mented and, of later years, so suggestive for the other

sciences, has so little regard been paid to the develop-

ment of hospital organization itself?

Dr. Reginald W. Revans has suggested that to some ex-

tent the answer to this question is found in the fact "that

studies of hospitals were not needed.'5 "In the past the

hospital created no serious problems for those whose hands

were on the levers of social control, for only the poor

were driven to seek its shelter. The rich were nursed at

home and died in their own beds."6 Along these same lines

Dr. Sam A. Edwards has suggested that:

Studies of these early hospitals were not needed or
desired for the following reasons:
1. Paucity of hospitals.
2. Simplicity of the operation of hospitals.
3. Limited social objectives of hospitals.
4. Hospital social objectives were not recognized as
of any value to the community, with the possible excep-
tion of giving comfort to the poor... a purely local
activity.
5. The method of financing hospital care dictated
control by a small group who associated their interests
with the immediate community.
6. Demand for hospital care was not of sufficient
importance to be significant.
7. The financing of hospitals was not an economic
problem, as development in the following areas was
limited
a. Technology.
b. Medicine.
c. Social objectives that hospitals could attain to.
d. Public acceptance of hospitals as institutions
useful to the entire community and to each of its members.
e. The need for inter-hospital cooperation was non-
existent or not recognized. 7
8. Lack of governmental interest.








Someone in this day and age would have difficulty in under-

standing the expansive and benevolent amateurism with which

these early hospitals were established. The suggestion

that research might be needed either to identify their

problems or to point out their solutions would have been so

remote from their patronizing self-confidence as to lack

all meaning. In commenting upon this situation Or. Revans

states

The Encyclopedia Britannica for 1911, for example, in
its article on nursing, remarks that whatever other prob-
lems the profession may face in the future, it will never
be short of recruits. Who, in their senses, mould spend
years examining the role and status of nurses, given this
happy state of affairs? Bullock, in his 1954 report on
the profession's needs for self-realization would, to the
first daughters of Florence Nightingale, have sounded not
only pretentious but also indecent.8

Additionally these early hospitals had none of the problems

with which they are now perplexed. So long as they were run

by amateurs only for the deserving poor they made no demands

upon official consciences and so called for no official

examination. Even if he had existed the research worker

had no channel of entry either from the university or the

government department.

One does not create overnight a tradition of hospital

research. The fact that a better understanding is needed

of present day problems no longer needs emphasis. But re-

sources to carry out the search for understanding are prac-

tically non-existent. Few studies get to the heart of the








problem of "what makes for a good hospital," on which alone

a tradition of hospital research is built. Nor does one

have to look far back into American history to find equally

telling illustrations of how little progress has been made

in eliminating the ignorance of these vital problems in the

best use of precious resources.

The Commonwealth Fund, some 30 years ago, produced a

report on the small community hospital and, in its chapter

on organization and administration, said of the doctors:

men already accustomed to having their own way in
a proprietary situation promptly identify the hospital
with themselves, thinking of the operating room as their
operating room, the nurses as their nurses, the superin-
tendent as their agent.,. these attitudes may lead to
irritability in the staff and early friction with the
superintendent, if not with the business men who stand
behind him. There is a good deal of jockeying for posi-
tion and much depends upon the tact and firmness with
which the hospital is run during this period.... Per-
haps a local disaster a tornado or a fire throws
into relief the advantage of the hospital to the commu-
nity and shows the doctors their own capacity for a
quick and smooth cooperation....9

In reviewing these comments Dr. Revans says

The author of these lines has nothing to learn about
one of the most obstinate of all hospital problems -
the cult of individualism among the medical staff -
but his drastic remedy, teamwork in disaster, is one
on which the resources of social science should be
able to improve. Nor is this all. A full program
of research into the organization and management of
hospitals would help to understand this distressing
condition among the medical staff, not merely to
ameliorate or even cure it, but to prevent it by
adequate methods of emotional inoculation in early
life. But, judged against what is needed in the way
of social therapy, present knowledge is pitiably
slender.10

There is an advantage to this lack of knowledge and

that is in this field no misconceived theory need be








disproven before contact can be made with what may prove

useful. Also, a very wide variety of ideas can be adapted

from other fields, from mechanical engineering to political

science, and from nonparametric statistics to social psy-

chology. But, according to Dr. Revans:

The main research problem remains that of defining,
analyzing and modifying the attitudes of those who,
in the hospitals, command the heights of power. All
three must be achieved definition, analysis and
modification for in the real world of suffering and
anxiety, academic or scholarly studies that do not
lend themselves to improving the human condition
have but a secondary place. They may be brilliant
excursions into the fields of statistics, anthropology,
economics or social theory, but if they do not help
to resolve the problems of hospital effectiveness
they are not research into management or organization.
These management problems can be understood only if
and when those who are actually managing personally
join or take over the research needed to resolve the
problems. This demands a radical change of view, to
recognize that, while help is available, salvation
will not be by outside experts.11



The Problem Area

These introductory comments have identified the broad

general area in need of research as that of "research into

the organization and management of hospitals." This case-

study in particular is concerned with what Rensis Likert

has referred to as a fundamental aspect of organization:

the process by which members determine or influence how

things get done in an organization (the process of con-

trol).12In the hospital, as in any complex organization some

groups and individuals have more influence on the operations

of the organization than others. And the part each group or





7

person plays in the organization depends, among other things,

on the relative amount of influence that it has. Moreover

a particular group may be perceived by its members, and/or

others in the organization, as wielding more, or less, in-

fluence than it should insofar as organizational functioning

is concerned. The prevailing distribution of influence in

the organization may or may not coincide with the distribu-

tion that is preferred by those concerned. Imbalances of

influence may be present in the organization. Such imbal-

ances, or discrepancies between prevailing and desired pat-

terns of influence in the system,.when large enough and un-

mitigated, can at least in the opinion of Georgopoulos and

Mann based upon the results of their study, "result in power

conflicts, intraorganizational strains, and dissatisfactions

among organizational members, ultimately affecting the per-

formance of the organization adversely."13 This position is

supported by other studies such as the one by March and

Simon14 who propose that disagreements between participants

regarding organizational facts and ideals (including those

related to control) contribute to intergroup conflict.

Additionally there is the study by Blake and mouton15 who

have underscored the importance of mutual understanding and

agreement in attaining organizational effectiveness. While

more recently mcMahon and Perritt16 have reported, from a

study of two manufacturing plants, that they found a high

degree of concordance was directly related to measures of

effectiveness. Condordance they define as "the degree of





8

agreement among hierarchical echelons' perceptions of the

organizational control structure."l

It is, therefore, important to know something about

the distribution and balance of influence in the hospital,

especially about the influence of key groups in the organ-

ization. Yet, despite the importance of this subject rela-

tively little research has been performed in this area.

While the number of hospitals has increased by approximately

250 since 1960, only one major research study is available

which makes any reference to the influence of key groups in

the hospital.18 On the other hand the process by which

members determine or influence how things get done in an

organization has been the subject of extensive research in

such organizations as unions,1920 voluntary associations,21'22

colleges,25 business, 16,2425'26 and industrial organiza-

tions.27,28 Studies similar to these do not exist for

hospitals.


Purpose of the Study

With this background information in mind, the purpose

of this case-study becomes relatively clear. It is twofold

First, to compare the patterns of control by hierarchical

level in two hospitals (of different types) to determine

what, if any, significant differences exist between the two.

This is done by assessing (a) perceived vs. desired control

and (b) active vs. passive control. (These terms are defined

in the methodology section of this chapter.) Second, to







determine the relationship between the perceptions of the

individuals at various levels (using the French and Raven

fivefold typology) as to the bases of supervisory power

in each institution and then to correlate these with their

perceptions as to satisfaction with the way their immediate

supervisors were doing their jobs.


Significance of the Study

Recent research has indicated that there is a direct

relationship between the amounts of control exercised by

members at all organizational echelons, higher performance

and increased satisfaction. 1721'25'2629 This same research

has also pointed out that there are certain relationships

between control structure and member consensus. While it is

not the purpose to examine the relationship between control

and performance in these hospitals, it is within the purpose

to determine the prevailing distribution of influence which

exist within them. This distribution (which for the two

hospitals involved is discussed and developed in detail in

subsequent chapters) does not always coincide with the dis-

tribution that is preferred by those concerned and previous

studies have demonstrated that such imbalances, or discrep-

ancies between prevailing and desired patterns of influence,

when large enough and unmitigated, can result in power con-

flicts, intraorganizational strains, and dissatisfactions

among members of the organization which ultimately affect

the performance of the organization adversely. The fact







that this study demonstrates that such imbalances or dis-

crepancies do exist in these institutions should serve as

a warning to the managerial personnel of these organizations

as to the existence of a real or potentially imminent

problem area that can, if uncorrected, adversely affect the

performance of their own and other similar organizations.

Additionally, it should serve as an indicator of a problem

area in hospital administration which is deserving of and

in need of further research.

Likert30 and Tannenbaum31 have suggested that the

processes underlying a system of high control and its

effects derive essentially from the satisfaction of the ego

motives of the individual, such as the desire for status,

achievement, and acceptance. If their interpretation is

correct, then one would expect reward, referent, and expert

power to be the more important bases underlying control and

its implications. In contrast, if the more traditional

Weberian view is indeed correct, then the more important

bases of control and its effects would be legitimate author-

ity and the manipulation of rewards and sanctions. Closely

related to this point is the fact that there are empirical

studies which have shown a direct relationship between con-

trol, bases of control, performance and satisfaction with

supervisory personnel.23'26,32

Use of the French and Raven fivefold typology in this

case-study permitted a categorization, within these insti-

tutions, of the more important bases of supervisory control








as perceived by members of the organizations. Results of

this categorization tend to support the earlier findings

of Likert and Tannenbaum as opposed to the more traditional

Weberian view. The bases of control obtained by use of the

above mentioned typology were then correlated to perceptions

of satisfaction with the way immediate supervisors were doing

their jobs to demonstrate that a definite relationship did

in fact exist in these institutions between these two vari-

ables. Certainly these results point out that this is an-

other area which is in need of additional investigation and

study.

Finally significance can be found in the fact that this

is one of the first, if not the first, attempts to apply the

control graph approach to the study of control patterns in

hospitals. The success of this research should demonstrate

that there are many additional areas in need of study to

which this approach can be applied.


Methodology

There is a serious problem that exists in any study

of control and that is the one of measurement. In general

researchers have obtained data about control either from

available records describing the legal or structural char-

acteristics, of organizations or from informants who respond

to questions concerning how or where in the organization

decisions are made or how influence is exercised.

In 1963 Evan reviewed a number of indices that






12

illustrate the measurement of control in industrial organi-
33
zations.3 These included, span of control the number of

levels of hierarchy the ratio of administrative to pro-

duction personnel "time-span of discretion," which is de-

fined as "the maximum length of time an employee is author-

ized to make decisions on his own initiative which commit

a given amount of the resources of the organizationl"4

the hierarchical level at which given classes of decisions

are made; and the formal limitations that apply to the

decision-making authority of management.

More recently, Whisler, leyer, Baum and Sorensen have

conducted an empirical as well as analytical study in which

they focused upon the three general measures of control that

have been suggested in organizational literature (1) indi-

vidual compensation, (2) perceptions of interpersonal

influence recorded on a questionnaire, and (3) the span of
35
control in the formal organization.3 These three measures

are recognized as being based upon different concepts of

the process of control in organizations. For the concept of

control identified as "control over system output" (system

control), the compensation paid the individual by the organi-

zation is used as the measure of control. Where control is

defined as "perceived interpersonal control," scaled per-

ceptions of individual influence is the measure of control.

And finally for the concept of control which is "formally

defined (or intended) interpersonal control," the measure
is the span of control.36 Each of these measures can be







appropriate depending upon the concept of control to be

measured. In determining which measure to use the researcher

should take into consideration several factors. First to be

considered is the relevance of the control construct (concept)

to the other variables studied. Where the other variables

relate to the organization as a whole, for example changes

in technology, changes in size or dispersion and differences.

in the environment (cultural or demographic) then the system

control concept with its "individual compensation" measure

is likely to be most relevant. When the other variables are

internal in nature, such as technology and task complexity

then the span of control measure is probably the most relevant.

Perceived control is apt to be most relevant when psycholog-

ical variables are studied.37 The second factor to be taken

into consideration is ease of use. In this regard, quali-

tative differences in the various measures can be seen.

Questionnaire data are costly and difficult to gather. Com-

pensation data are often confidential, especially in private

businesses. Formal organization structures on the other

hand are normally available, provided one has a reliable

and knowledgeable informant or provided that the organization

maintains and preserves organizational charts. The avail-

ability of the different kinds of data desired is often

related to the organization or research site within which

the research is to be conducted. The final factor to be

discussed, which should be taken into consideration, is that
of the research design. If the research is designed to








compare the present with the past, then the influence ques-

tionnaire measure is unfeasible (unless the questionnaire

had for some reason been administered previously). In this

type of project it would be necessary to use either the span

of control or compensation measures. The influence question-

naire or some variation of it; however, may be the only

feasible measure to be used when the researcher encounters

informally organized groups, or organizations that do not

use monetary compensation.39

Taking into consideration the concept of control (Chapter

II) and the research design (Chapter IV) the work described

in the remainder of this paper relies for measures of con-

trol largely on the averaged judgments by organization

members in response to questionnaire items dealing with the

amount of influence or control exercised by various groups

in their organization. This approach to measurement of con-

trol has limitations; yet it seems to be more suitable than

the available alternatives for the measurement of the partic-

ular concepts with which this research is concerned.

The specific methodology which is used in this study

is essentially the same as that which has been used in other

studies of organizational control in which the control graph

approach and the French and Raven fivefold typology have

been utilized. An influence questionnaire, which was

developed by using research questions from previous studies

conducted by Dr. Arnold 5. Tannenbaum and others at the
Survey Research Center, Institute for Social Research,








Department of Psychology, University of michigan, was dis-

tributed in two very carefully chosen hospitals for the

purpose of gathering certain data from members of various

groups and hierarchical levels in the two institutions.

These data which have been collected are subjected to both

descriptive and statistical analysis in the following

chapters of this paper.

The control graph technique is especially utilized in

this study tos

1. measure the perceptions of various members from

several different groups and hierarchical levels, in each

of the hospitals, as to the type of control structure which

characterizes their institution (perceived control). In his

recent work Tannenbaum uses the words "actual control" in

his discussions of this concept.29

2. Measure the perceptions of this same group of indi-

viduals as to how control should be distributed in their

institution (desired control). Once again Tannenbaum uses

the words "ideal control" to describe this concept in his

book.29

3. Compare the perceived control with the desired control

for each institution based upon the perceptions of this

group of individuals.

4. measure the perceptions of this same group as to

the "active" and "passive" control curves which characterize

their institution.
"Active Control" in this context means the extent






16

to which the actor (either an individual or a group)

exercises control in the organization.

"Passive Control" on the other hand means the extent

to which the actor is controlled within the organization.

5. Measure the perceptions of these same respondents

as to the amount of perceived and desired control (influence)

that the medical staff has or should have on how their

institution functions on how it is run and how it operates.

The French and Raven fivefold typology is used in the

study to measure the perceptions of these same individuals

as to what constitutes the bases of supervisory control in

each of these institutions.



Assumptions

In adopting the control graph approach to the measure-

ment of control, the assumption is made that organization

members as a group are able to provide reasonably valid

and reliable data. There are those who would call this

assumption into question saying that it is apparent that

organization members differ in their judgment about control.

In response to this, it is important to bear in mind that

the reliability of the measures, which are intended as

organizational indices, is a function of the number of

respondents chosen from each of the organizations studied.

Thus, although the reliability of scores based on an indi-

vidual's responses may be low, averaged responses may be
quite stable. The fact that individual respondents may be





17

unsure of their answers and that they may be in error does

not in itself vitiate the method, provided that respondents

give better than chance answers, that the errors are random,

and that a sufficient number of respondents are available.

It is assumed in this study that these conditions do in

fact prevail. Additionally it is assumed, based on the

results of previous research, that the control graph approach

to the study of control in organizations, was and is an

empirically reliable and valid approach to the study of this

concept.


Organization of the Study

The remainder of this paper is devoted to developing

and testing these concepts. Chapter II elaborates on the

nature of control, answering such questions ass (1) What

is control? (2) What are its functions? and (3) What are

the bases for control? Chapter III discusses at some length

the "control graph" and its application to the study of

control in organizations. Chapter IV is devoted to the

research design, method and procedure used in the study.

Also included in this chapter are such items as the (1) se-

lection of the participating hospitals; (2) selection of

the respondents; and (3) methodological limitations of the

study. Finally Chapter V presents the findings of this

investigation, while the sixth chapter summarizes the disser-

tation, drawing those conclusions which seem reasonable in

light of the results, and indicates the implications this





18

inquiry holds for further research.

Data not an integral portion of the body of the paper

but supplemental in nature, are included in appendices,

and are cross-referenced in the appropriate places in the

text.













NOTES

1. Charles U. Letourneau, The Hospital Administrator
(Chicagos Starling Publications, 1969J, p.i.

2. Ibid.

3. Ibid., p.8.

4. Basil S. Georgopoulos and Floyd C. Mann, The Community
General Hospital (New Yorks The macmillan Company,
Y9b2), p.b88.

5. Reginald W. Revans, "Research Into Hospital management
and Organization," The milbank memorial Fund Quar-
terl, Volume 44, Number 3, Part 2 (July, 196b),
p.207.

6. Ibid.

7. Sam Allen Edwards, A Comparison of Controls as Found In
Hospitals with those In Industries Classified as
Public Utilities, lDoctoral dissertation, State
University of Iowa, 1960), pp.1-2.

B. Revans, p.209.

9. Henry J. Southmayd and Geddes Smith, Small Community
Hospitals (New Yorks The Commonwealth Fund, 1944).

10. Revans, p.210.

11. Ibid.

12. Rensis Likert, "Foreword," in Control in Organizations.
ed. Arnold 5. Tannenbaum (New Yorkt icGraw-Hill, Inc.,
1968), pp.vii-viii.

13. Georgopoulos and mann, p.566.

14. James G. March and Herbert A. Simon, Organizations
(New Yorks John Wiley and Sons, 1963).

15. Robert R. Blake and Jane S. mouton, The managerial Grid
(Houston: Gulf Publishing, 1964).







16. J. Timothy Mclahon and G.U. Perritt, "The Control Struc-
ture of Organizationss An Empirical Examination,"
Academy of Management Journal, Volume 14, Number 3
(September, 1971), pp.327-340.

17. and "Toward a Contingency Theory of
Organizational Control," Academy of management
Journal, Volume 16, Number 4 (December, 1973), p.625.

18. Georgopoulos and Mann, pp.566-575.

19. Arnold S. Tannenbaum and Robert L. Kahn, "Organizational
Control Structures A General Descriptive Technique
as Applied to Four Local Unions," Human Relations,
Volume 10, Number 2 (May, 1957), pp.127-14U.

20. "Control Structure and Union Functions,"
American Journal of Sociology, Volume 61, Number 6
(may, 1956), pp.53b-b4b.

21. "Control and Effectiveness in a Voluntary
Organization," American Journal in Sociology,
Volume 67, Number 1 LJuly, 1961), pp.33-46.

22. Clagett G. Smith and Arnold S. Tannenbaum, "Organiza-
tional Control Structures A Comparative Analysis,"
Human Relations, Volume 16, Number 4 (November,
1963), pp.299-326.

23. Jerald G. Bachman, David G. Bowers and Philip M. Marcus,
"Bases of Supervisory Powert A Comparative Study in
Five Organizational Settings," in Control in Organi-
zations. ed. Arnold 5. Tannenbaum (New Yorki McSraw-
Hill, nc., 1968), pp.229-238.

24. David G. Bowers, "Organizational Control in an Insurance
Company," Sociometry, Volume 27, Number 2 (June,
1964), pp.230-244.

25. Clagett G. Smith and Oguz N. Ari, "Organizational Con-
trol Structure and member Consensus," American Journal
of Sociology, Volume 69, Number 6 (May, 1964)
pp.623-638.

26. John M. Ivancevich, "An Analysis of Control, Bases of
Control, and Satisfaction in an Organizational Set-
ting," Academy of management Journal, Volume 13,
Number 4-(December, 191J), pp.427-436.







27. Arnold S. Tannenbaum and Basil S. Georgopoulos, "The
Distribution of Control in Formal Organizations,"
Social Forces, Volume 36, Number 1 (October, 1957),
pp.44-b0.

28. Josip Zupanov and Arnold S. Tannenbaum, "The Distribu-
tion of Control in Some Yugoslav Industrial Organiza-
tions as Perceived by Members," in Control in Organi-
zations. ed. Arnold S. Tannenbaum (New YorkT MicGraw-
Hill, Inc., 1968), pp.91-109.

29. Arnold 5. Tannenbaum, Control in Organizations (New York;
mcGraw-Hill, Inc., 1958).

30. Rensis Likert, New Patterns of management (New Yorki
mcGraw-Hill, Inc., 1961).

31. Arnold 5. Tannenbaum, "Control in Organizationsi Indi-
vidual Adjustment and Organizational Performance,"
Administrative Science Quarterly, Volume 7, Number 2
kJune, 192)J, pp.236-257.

32. Jerald G. Bachman, Clagett G. Smith, and Jonathan A.
Slesinger, "Control, Performance, and Satisfactioni
An Analysis of Structural and Individual Effects,"
Journal of Personality and Social Psychology, Volume 4,
Number 2 LAugust, 19bbJ, pp.127-136.

33. William M. Evan, "Indices of Hierarchical Structure of
Industrial Organizations management Science, Volume
9, Number 3 (April, 1963), pp.468-477.

34. Ibid., p.472.

35. Thomas L. Whisler, Harald Meyer, Bernard H. Baum, and
Peter F. Sorensen Jr., "Centralization of Organiza-
tional Controls An Empirical Study of Its meaning and
measurement," Journal of Business, Volume 40, Number 1
(January, 1967), pp.10-26.

36. Ibid., p.10.

37. Ibid., p.22.

38. Ibid., p.23.

39. Ibid., p.24.













CHAPTER II


THE NATURE OF CONTROL



Overview

What exactly does control mean? When this question

was asked of a number of managers, in both government and

industry, the answers showed a general lack of agreement.

It is important that managers have a clear understand-

ing of this concept. A manager who does not understand

control cannot be expected to exercise it in the most

efficient and effective manner. Nor can staff men whose

duty it is to design systems and procedures for their

organizations design efficient systems unless they possess

a clear understanding of control. And certainly anyone who

is subject to control by others has to understand clearly

what that means if he is to be contented in their relation-

ship.

When control is not understood, good management is a

very improbable result. This is especially true when -

as frequently it is control is identified with management,

or is confused with certain devices of management, such as





23

objectives, plans, organization charts, policy statements,

delegation of authority, procedures, and the like. Sherwin

has stated that "the manager who believes managing and con-

trolling are the same thing has wasted one word and needs

a second to be invented."2 He goes on to add that "one

who believes he has provided for control when he has estab-

lished objectives, plans, policies, organization charts, and

so forth, has made himself vulnerable to really serious

consequences. A clear understanding of control is there-

fore indispensable in an effective manager.

Understanding control really means understanding at

least three principal things about its What it is; what its

functions are; and finally the bases for its existence. By

addressing these three things I have framed a concept of

control that serves as a basis for the research which was

conducted as a part of this dissertation.


What is Control

The word control is difficult to define as it appears

to have different meanings in different contexts when used

by different authors. For example Clark states
"Control" means, primarily, coercion orders backed
by irresistible power. In a sense, no coercion is truly
irresistible, or almost none. One can always break the
law if one will take the consequences... and sometimes
the penalty is less than the profits of the offense.
But the earmark of coercive control is penalties, imposed
by a power which can, if it will, make them heavier than
anyone but the most desperate would deliberately incur.4

He readily admits there are other means of exercising control.







Newman, on the other hand, defines control ass

...seeing that operating results conform as nearly as
possible to the plans. This involves the establishment
of standards, motivation of people to achieve these
standards, comparison of actual results against the
standard, and necessary corrective action when performance
deviates from the plan.5

Dubin, in writing of control within organizations but

nevertheless applicable to society in general, states

Control within an organization has two major dimensions
We can conceive of control as the process of developing
systems of standards for the guidance of organization
behavior; we can view control as a system for enforcing
standards of organization behavior.6

Roucek defines control as "a collective term for those

processes, planned or unplanned, by which individuals are

taught, persuaded, or compelled to conform to the usages and

life-values of groups." He expands his definition by stating

furthers

Control occurs when one group determines the behavior
of another group, when the group controls the conduct of
its own members, or when individuals influence the re-
sponses of others.... Control, consequently, operates
on three levels group over group, the group over its
members, and individuals over their fellows. In other
words, control takes place when a person is induced or
forced to act according to the wishes of others whether
or not in accordance with his own individual interests.?

Hill and Egan have defined controls

As the selection of guidelines for the decisions
of lower participants as well as the establishment of
rules to enforce conformity to the standards of per-
formance which are set by superiors.8

more recently Tannenbaum has defined control ass

Any process in which a person or group of persons
or organization of persons determines, that is, inten-
tionally affects, the behavior of another person, group,
or organization.9





25

It becomes apparent from these quotes that "the word

control has the serious shortcoming of having different mean-

ing in different contexts"10 This attribute has been noted

by such authors as Drucker,1 Kast and Rosenzweig,12
13 14
Litterer,3 and Luneski.4 Each points out that management

control may be viewed in two parts. One relates to the

achievement of effective control over subordinates through

the direction of their activities. The second relates to

the evaluation of the desired outcome of an activity and the

making of corrections when necessary. This dichotomy has

been summarized well by Reeves and Woodwards

In the literature relating to organizational behavior
there is ambiguity in the use of the word control. The
confusion arises largely because to control can also mean
to direct. Precisely defined control refers solely to
the task of ensuring that activities are producing the
desired results. Control in this sense is limited to
monitoring the outcome of activities, reviewing feed back
information about this outcome, and if necessary taking
corrective action.15

Partially because of this confusion, control is considered

to be "one of the thorniest problems of management today."15

While it has been widely discussed, there are still some

writers who content that it lacks a common area of under-

standing. It has "scarcely any generally accepted principles,

and everyone in the field, therefore, works by intuition and

folklore."17 Rowe has noted

Although management control is widely discussed, little
has been done to formulate a body of principles for use
in business system design.18


Furthermore, Jerome has pointed outs





26

Principles and procedures and substantive content
simply have not been rigorously developed in the area
of executive control.19

More recently, lockler has written

In spite of the fact that management control is one
of the basic management functions, there is no comprehen-
sive body of management control theory and principles
to which executives can turn for guidance in performing
their management control functions.20

Recognizing this ambiguity regarding the use of the term

control and the alleged lack of control theory it is necessary

to state that the definition of control used in this paper

is the one provided by Tannenbaum which has been referenced

above. In effect, this eliminates from consideration the

works of those authors who use the term control to refer

solely to the traditional "constant cyclic-type activity of

plan-do-compare-correct" with its "continuous, concomitant

system of communication or flow of information." For a

rather comprehensive review of the work done in this area

the reader is referred to a recent article written by Giglioni

and Bedeian, who concluded by stating

Even though control theory has not achieved the level
of sophistication of some other management functions, it
has developed to a point that affords the executive ample
opportunity to maintain the operations of his firm under
check. Unquestionably however, continued interest and
research in this area are necessary to bring control
theory to new levels of sophistication and, above all,
pragmatism.21

Although the definition of control provided by Tannenbaum

and utilized in this study conforms essentially to what many
22
authors mean by control, power, or influence,2 there cer-

tainly are differences of opinion regarding the definition







of these terms.23,24 For example, some writers prefer to

think of power as an exclusively coercive form of control.

Weber was the first of the classic authors on organization

to reject this limited notion of power, and many contemporary

social scientists, including Tannenbaum, are inclined to

think of power as having bases in addition to, although by

no means excluding, coercive ones. Some authors like to

think of power in terms of differentials or ratios that

describe the relative "strengths" of persons in a system.

In this view power is essentially the effect that one person

has on a second compared with that which the second has on

the first. This is an important index of power relations,

but it is conceptually a derivative of the more general

definition proposed by Tannenbaum and the one used in this

paper. A number of authors prefer to distinguish power

from control by defining power essentially as the ability

or capacity to exercise control, that is, as "potential

control." For example, Goldhamer and Shils state that "a

person may be said to have power to the extent that he

influences the behavior of others in accordance with his own

intentions."25 While Etzioni says that "power is an actor's

ability to induce or influence another actor to carry out

his directives or any other norms he supports."26 Both of

these definitions are consistent in essential respects with

the one used in this paper, although Etzioni's statement

implies what is called "potential control." Many authors
use the term "authority" to refer to the formal right to





28

exercise control, and I do likewise in this paper.
The meaning of control, then as defined in this paper,
can be -seen in a simple graph which represents control as
a cycle beginning with an intent on the part of one person,
followed by an influence attempt addressed to another person,
who then acts in some way that fulfills the intent of the
first. Figure 1 presents the control process in its
simplest form.





Intent of
person A


tha leads
fulfills to



(\
behavior of influence
person B att mpt





resulting in

Figure 1. The Control Process





29


This graph was originated by Tannenbaum who states that

There are, of course, many elements in addition
to those indicated that are important in understanding
this process. These include the assumptions and values
of the actors, the 'basis of power" that help explain
B's response, and the great variety of means by which
A attempts to influence B. Such means may be direct or
indirect; they may include orders or requests, threats
or promises, and so forth. The behavior of B may in-
volve relationships with other persons or it may involve
actions in relation to technological elements, such as
tools, computers, or production lines. Thus technology
may enter into the cycle at various points, creating
what has been called a "sociotechnical" system. For
example, computers may provide A with information that
leads him to request B to do one thing rather than
another. Or A may simply use the computer to tell B.
A may also speed up a production line, which illustrates
another form of influence attempt on B.

The intentions of A may be initiated by him, or
they may be the intentions of others that are acquired
by A. These intentions may imply quite specific actions
for B, as when a supervisor gives detailed instructions
to a subordinate; or they may be very general, although
no less real, as in the formulation of organizational
policy. The behavior of B, which is the object of A's
intentions, may, in our definition, be covert as well
as overt. A, for example, may have intentions regarding
the intentions of B, and vice versa.27

It can be stated then that Figure 1, although sim-

plified, represents the essence of the control process, as

it has been defined, Such a cycle, as that depicted in the

graph, includes essentially what Etzioni refers to as "com-

pliance." The control cycle is a basic unit of organization

structure; organizations are composed of large numbers of

such cycles in interrelationship. If a cycle breaks down

at any point, for whatever reason, control cannot be said

to exist.28 Chronic breakdowns of such cycles imply a

breakdown in the organization itself.







What are its Functions

Just as control has been variously defined, so has

the concepts as to the functions of control varied. Con-

trol has been conceptualized in both a narrow and broad

sense, as evident in the following quotes from different

writers.

In the broad social sense Ross has stated that

The function of control is to-preserve that indis-
pensable condition of common life, social order. When
this order becomes harder to maintain, there is a demand
for more and better control. When this order becomes
easier to maintain, the ever-present demand for indivi-
dual freedom and for toleration makes itself felt. The
supply of social control is evoked, as it were by the
demand for it, and is adjusted to that demand.29

Young condenses this definition further by stating that the

function of control is "to bring about comformity, soli-

darity, and continuity of a particular group of society."30

Roucek, in commenting on Young's definition, states

These purposes may possibly guide far-seeing states-
men of social scientists, but most individuals who en-
deavor to control their fellow men show little perspec-
tive in their efforts. Cften they merely struggle to
increase the acceptance of the modes of conduct that
they themselves prefer.31

While writing of control within the organization, as opposed

to society in general, Hill and Egan have stated that

Control is a critical element in the administrative
process. Its purpose is to insure that work activities
are directed toward the accomplishments of stated ob-
jectives. Therefore, it compliments both the determina-
tion of goals and the subsequent structuring of work
flows which are intended to accomplish these aims. Since
both of these elements are continuing processes, regu-
latory mechanisms also must extend over time. As a re-
sult they bear a vital relationship to adaptation and
innovation.32







In another place these same authors state:

If control is to approach its ideal use, it must
perform three functions (1) evaluation, (2) enforcement,
and (3) motivation. Too often, however, emphasis is
concentrated on the second function, managers fail to
pay sufficient attention to the monitoring processes
which must be used to gather, sift, and evaluate infor-
mation pertinent to the activities to be governed.
Frequently, even less attention is devoted to con-
structing a system which is designed to motivate de-
sirable forms of behavior. Failure to give sufficient
weights to each of these functions can lead to the
creation of dysfunctional control procedures.33


Tannenbaum, on the other hand, writing about the functions

of control, in an organization, states that:

A social organization is an ordered arrangement of
individual human interactions. Control processes help
circumscribe idiosyncratic behaviors and keep them con-
formant to the rational plan of the organization.
Organizations require a certain amount of conformity as
well as the integration of diverse activities. It is
the function of control to bring about conformance to
organizational requirements and achievement of the
ultimate purposes of the organization. The coordi-
nation and order created out of the diverse interests
and potentially diffuse behaviors of members is largely
a function of control.34

Since I have previously elected to accept Tannenbaum's

definition of control, for the sake of continuity I also

recognize and accept his concepts as to the functions of

control.



What are the Bases for Control

Theories on the bases of control have changed through

the years as the concepts of control have changed, and

although the theoretical analysis of control in social

systems has been utilized for a long time, empirical re-

search has only recently been initiated in organizations.





32

The "human-relations" approach that inspired a great

deal of research in organizations avoided explicit reference

to social power or control, partly because these terms

carried connotations that were inconsistent with the ideal

of the harmonious, conflict-free organization. This re-

search was concerned implicitly with enhancing the control

exercised by management through devising more effective

techniques of supervision and through reducing "resistances"

on the part of workers to managerial policies. Therefore,

some advocates of human relations were committed, implicitly

at least, to enhancing control within organizations while

denying its importance a contradiction that, according

to Crozier, may have contributed to the charge that human

relations was manipulative.35

Traditionally, the concept of power has been associated

with forms of tyranny, elitism or authoritarianism, or with

conflict and struggle. Almost all the literature on the

power of leadership, according to Bell, stems from the works

of Aristotle and machiavelli and is committed to the image

of the mindless masses and the image of the strong-willed

leader."3 Bendix maintains that historically, ideologies

of management have grown up specifically to justify the

employers' exercise of authority, which was associated in

one way or another with the subordination or expliotation

of workers.37

iWhile many of the classical conceptions of control,

including those of Weber in bureaucracies and Michels in








political organizations, have been valuable in analyses of

contemporary organizations, the changing character of soci-

eties and organizations over the years is making apparent

some of the limitations of the older concepts. The emphasis

in contemporary social science on quantitative research has

also contributed to changes in interpretations of the con-

trol process because of the need to develop concepts that

are operational as well as theoretically meaningful. At

the same time, research findings themselves have led to

reinterpretations of some of the older conceptions.38

The changes, of course, with which this section is

concerned is the changes that have taken places in analyses

of the bases of power. Coercion played a prominent role

in the traditional analyses, consistent with the presumed

conflict between leaders and followers. Leaders were to be

obeyed out of fear of punishment or hope for reward. Weber,

however, has argued that the stability of social systems

depends on acceptance by followers of the right of leaders

to exercise control. This implies legitimate authority,

and Weber defines three types (1) "Charismatic" authority,

according to which leaders are thought to be endowed with

extraordinary, sometimes magical powers. Charisma on the

part of a leader elicits obedience out of awe. It is illus-

trated in its pure form by "the prophet, the warrior hero,

the great demagogue." (2) "Traditional" authority, which

appertains to those who have the right to rule by virtue

of birth or class, The traditional leader is obeyed because








it is the thing to do and because he or other members

of his family have always been followed. Examples are

certain patriarchs, monarchs, or feudal lords. (3) "Legal"

authority, which applies to those who hold leadership posi-

tions because of demonstrated technical competence. The

legal leader is obeyed out of a sense of duty to the law.

In the ideal bureaucracy, leadership is based almost exclu-

sively on legal authority.3

The nature of authority visualized within this frame-

work is consistent with many of the traditional analyses

Weber's authority figures are prophets, warriors, dema-

gogues, partriarchs, lords and bureaucrats. However, more

recent analyses have outlined additional bases of power.

Simon, for example, points to the importance of social

acceptance and approval.40 Approval and disapproval repre-

sent forms of reward and punishment, but they deserve special

consideration because they are frequently dispensed, not

only by the designated leader, but also by others. There-

fore, a subordinate may obey a supervisor, not so much

because of the rewards and punishments meted out by the

supervisor, as because of the approval and disapproval by

the subordinate's own peers. Confidence may be another

basis for acceptance of a supervisor's authority. A sub-

ordinate may, for example, trust the judgment and therefore

accept the authority of a superior in areas where the leader

has expertise. French and Raven make a further distinction

between the influence of a leader based on confidence by








subordinates in the leaders' expert knowledge and "infor-

mational influence" based on acceptance by subordinates of
41
the logic of the arguments that the leader offers.4 An

expert leader, then, may exercise control, not simply

because he is an acknowledged authority, but because his

decisions, being based on expertise, are logical, appro-

priate, and convincing. Subordinates are persuaded that

the decisions are correct.

Some of these concepts represent radical departures

from the traditional ones where coercion played a prominent

role in the analyses of the bases of power.

French and Raven have developed a fivefold typology

which suggests a number of different categories of bases

of control.42 This framework offers a clear distinction

between the various categories of control. They list five

specific bases of power which are as follows

1. Reward Powers This is based on a subordinate's

perception that a superior has the ability to mediate

rewards for him.

2. Coercive Powers This is based on a subordinate's

perception that a superior has the ability to mediate

punishments for him.

3. Legitimate Powers This is based on internalized

values which dictate that there is a legitimate right to

influence and an obligation to accept this influence. The

organizational position of the superior is a major factor
of the legitimate power base.








4. Referent Power: This is based on the desire of

a subordinate to identify with a superior. The identifi-

cation of the subordinate can be maintained if he behaves,

believes, or perceives as the superior does.

5. Expert Power: This is based on a subordinate's

perception that a leader has some special knowledge or

expertise in a given area. Experience, training, reputation,

and demonstrated ability are among the many reasons why a

subordinate attributes expertness to a superior.

Recently there has been another base of control sug-

gested by Katz and Kahn. They call their new power variable,

"incremental influence," which they say is stated as follows:

...we consider the essence of organizational leader-
ship to be the influential increment over and above the
mechanical compliance with routine directives of the
organization.43

The five bases proposed by French and Raven and the

incremental concept offered by Katz and Kahn afford re-

searchers an important conceptual distinction. Reward

power, coercive power, and legitimate power are exercised

primarily in areas which are largely specified by the organ-

ization. For example, a person's position in the organiza-

tional structure largely dictates his degree of power in

these three areas. However, the exercise of referent and

expert power are idiosyncratic in character: i.e., these

power bases are uniquely determined by the behavior of the

superior and his ability to interact with subordinates.

Therefore, the superior's ability to influence his subordi-

nates, based on the referent and expert power forms,







constitutes an increment which is beyond that dictated by

his position in the organizational structure. Consequently,

referent power and expert power are the bases of incremental

influence and are operationalized as the combination of
44
referent power and expert power.4

Two other changes which have occurred in the conception

of control are worthy of mention before this section is

completed. First there is the change which relates to the

assumptions concerning the mutuality unilaterality of

control. A view common to traditional analyses argues

that the control process is unilateral; one either leads

or is led, is strong or weak, controls or is controlled.

Simmel, in spite of his general adherence to the traditional

conflict view of power, noted a more subtle interaction

underlying the appearance of "pure superiority" on the part

of one person and the "purely passive being led" of another

"All leaders are also led; in innumerable cases the master

is the slave of his slaves."45 Several social scientists

emphasize the fact that contemporary analyses are more

likely than the earlier ones to consider relationships of

mutual as well as unilateral power, of followers influencing

leaders, as well as vice versa. Finally there is the change

which has taken place in the assumptions as to the total

amount of control in an organization. Traditional analyses

of social power assume that the total amount of power in a

social system is a fixed quantity and that leaders and

followers are engaged in a "zero sum games"i increasing the





38

power of one party must be accompained by a corresponding

decrease in the power of the other. Now social scientists
46 4748 49
such as Lammers,46 Likert,47 Parsons, Tannenbaum and Kahn4

are inclined to question the generality of this assumption.

They believe that the total amount of power in a social

system can grow and that the leaders and followers can

therefore enhance their power jointly. The converse is

also true in their opinion.


Summary

In summarizing the Nature of Control it can be stated:

(1) that there is a certain amount of ambiguity associated

with the term "control" but that as it is used in this

paper it refers to any process in which a person or group

of persons or organization of persons determine, that is,

intentionally affects, the behavior of another person,

group or organization; (2) that its functions are diverse

but the recognized one in this thesis is the one contributed

to Tannenbaum, namely that its function is to bring about

conformance to organizational requirements and achievement

of the ultimate purposes of the organization; and (3) that

there are at least six bases of control which have been sub-

jected to empirical research reward, coercive, legitimate,

referent, expert and incremental.

Organizational control has been studied from a number of

different approaches, one of which is the control graph

approach the subject of Chapter III.












NOTES

1. Douglas S. Sherwin, "The Meaning of Control," Dun's
Review and modern Industry, Volume 67, Number-i
(January, 1956b, pp.45-46, 83-84.

2. Ibid., p.45.

3. Ibid.

4. John M. Clark, Social Control Business (New Yorks McGram-
Hill, Inc., 1939), p.5.

5. William H. Newman, Administrative Action (New Yorks
Prentice-Hall, Inc., 1951), p.4.

6. Robert Dubin, Human Relations in Administration (Engle-
wood Cliffs: Prentice-Hall, Inc., 1955), p.281.

7. Joseph S. Roucek, Social Control (2d ed.; Princeton,
N.J.I Van Nostrand Company, Inc., 1956), p.3.

8. Walter A. Hill and Douglas M. Egan, Readings in Organi-
zation Theory: A Behavioral Approach kBoston: Allyn
and Bacon, Inc., 196J), p.bUb.

9. Arnold 5. Tannenbaum, Control in Organizations (New Yorks
fcGraw-Hill, Inc., 1968J, p.b.

10. William T. Jerome III, Executive Control The Catalyst
(New Yorks John Wiley and Sons, Inc., 1961J, p.42.

11. Peter F. Orucker, The Practice of management (New Yorks
Harper and Row, 1954J, p.150.

12. Fremont E. Kast and James E. Rosenzweig, Organization
and Management (New Yorks McGraw-Hill, Inc., 197OJ,
p.467.

13. Joseph A. Litterer, The Analysis of Organizations (New
York John Wiley and Sons, Inc., 1965), p.233.

14. Chris Luneski, "Some Aspects of the meaning of Control,"
Accounting Review, Volume 39 (July 1964), p.593.







15. Tom K. Reeves and Joan Woodward, "The Study of managerial
Control," in Industrial Organizations Behaviour and
Control. ed. Joan Woodmard (Londons Oxford University
Press, 1970), p.38.

16. Alex W. Rathe, "management Controls in Business," in
management Control Systems. eds. Donald G. malcolm
and Alan J. Rowe (New Yorks John Uiley and Sons,
Inc., 1960), p.30.

17. Robert N. Anthony, Planning and Control Systemsi Frame-
work for Analysis (Bostons Division of Research,
Graduate School of Business Administration, Harvard
University, 1965). p.vii.

18. Alan J. Rome, "A Research Approach in management Controls,"
in management Control Systems. eds. Donald G. Malcolm
and Alan J. Home (New Yorks John Wiley and Sons, Inc.,
1960), p.274.

19. Jerome, p.26.

20. Robert J. Mockler, "Developing the Science of management
Control," Financial Executive, Volume 35 (December,
1967), p.B0.

21. Giovanni B. Giglioni and Arthur G. Bedeian, "A Con-
spectus of management Control Theorys 1900-1972,"
Academy of management Journal, Volume 17, Number 2
(June, 1974), pp.292-30b.

22. Robert A. Dahl, "The Concept of Power," Behavioral Science,
Volume 2, Number 3 (July, 1957), pp.201-21b.

23. Dorwin Cartwright, "Influence, Leadership, Control,"
Handbook of Organizations. ed. James march (Chicagos
Rand tcNally, 1965).

24. John Schopler, "Social Power," in Advances in Experi-
mental Social Psychology. ed. Leonard Berkowitz
(Volume 2; New Yorks Academic Press, 1965).

25. Herbert Goldhamer and Edward A. Shils, "Types of Power
and Status," American Journal of Sociology, Volume
45, Number 2 TSeptember, 1939), p.1 1.

26. Amitai Etzioni, A Comparative Analysis of Complex Organ-
izations (New Yorki Free Press, 1961), p.4.

27. Tannenbaum, pp.5-6.


28. Ibid., p.7.








29. Edward A. Ross, Social Control and,the Foundations of
Sociology (Boston: Beacon Press, 1959), p.114.

30. Kimball Young, Sociology (Cincinnati: American Book
Company, 1942), p.898.

31. Roucek, p.7.

32. Hill and Egan, p.506.

33. Ibid.

34. Tannenbaum, p.3.

35. Michel Crozier, The Bureaucratic Phenomenon (Chicagos
The University of Chicago Press, 1964), pp.145-150.

36. Daniel Bell, "Notes on Authoritarian and Democratic
Leadership," in Studies in Leadership. ed. Alvin W.
Gouldner (New Yorks Harper and Row, 1950).

37. Reinhard Bendix, Work and Authority in Industry (New
York John Wiley and Sons, 19bb).

38. Tannenbaum, p.9.

39. max Weber, "The Three Types of Legitimate Rule," in
Complex Organizations: A Sociological Reader. ed.
Amitai Ltzioni (New Yorks Holt, Rinehart, and
Winston, Inc., 1961).

40. Herbert A. Simon, "Authority," in Research in Industrial
Relations. ed. C.M. Arnesberg et al. INew Yorks
Harper and Row, 1957).

41. John R.P. French, Jr. and Burtram Raven, "The Bases of
Social Power," in Group Dynamics: Research and
Theory. ed. Dorwin Cartwright and Alvin Zander (New
York Harper and Row, 1960), pp.607-623.

42. Ibid.

43. Daniel Katz and Robert L. Kahn, The Social Psychology
of Organizations (New York John Wiley and 5ons,
19bb), p.301.

44. Ibid. p.302.

45. Kurt H. Wolff, The Sociology of Georg Simmel (Glencoe,
Ill Free Press, 1950), pp.18b-lBS.

46. C.J. Lammers, "Power and Participation in Decision-
Making in Formal Organizations," American Journal
of Sociology, Volume 73, Number 2 (September 1967),
pp.201-216.





42

47. Rensis Likert, New Patterns of Management (New Yorks
McGram-Hill, Inc., 1961).

48. Talcott Parsons, "On the Concept of Political Power,"
Proceedings American Philosophy Society, Volume 107,
Number 53 19b3i) pp.232-2b2.

49. Arnold S. Tannenbaum and Robert L. Kahn, "Organizational
Control Structures A General Descriptive Technique as
Applied to Four Local Unions," Human Relations, Volume
10, Number 2 (May, 1957), pp.12(-140.













CHAPTER III


THE CONTROL GRAPH AND ITS APPLICATION TO THE
STUDY OF CONTROL IN ORGANIZATIONS



The Control Graoh

The control structure of an organization can be repre-

sented in terms of a general schema which has been called

the "control graph." Such a graph was first applied by

Tannenbaum and Kahn in 1957 to a study of four trade-union

locals. This schema characterizes the control structure

of an organization in terms of two axes. The horizontal

axis of this graph represents a scale of hierarchical levels

in an organization. It may run from rank-and-file members

at the low end through various levels to the president

(top management) at the high end. The vertical axis of the

control graph represents the amount of control over the

organization's policies and actions that is exercised by

each of the hierarchical levels. This may vary according

to Tannenbaum and Kahn's approach from "none" to "a very

great deal of control." Thus,"having a great deal of control"

means that persons at the hierarchical level under consider-

ation determine in large degree the specific actions and

policies of the organization. "Having no control" on

this dimension means that all persons at a given level have

43









no "say" or influence in determining the policies and ac-

tions of the organization. A curve can be created by plot-

ting and connecting the points that show the amount of con-

trol characteristic of each hierarchical level. Figure 2,

graphically illustrates this concept.

It is clear from this graph that an infinite number of

Amount of
control
exercised


A very great
deal


A great deal



Quite a bit



Some



Little or
none A
Top middle Bottom
Level Level Level
Hierarchical Echelons
Figure 2. The Control Graph


curves of varying shapes are possible. For example, a curve

may have a negative slope, indicating that the amount of

control increases as one goes up the hierarchy (Figure 3).

It is also conceivable for the curve to have a positive

slope (Figure 4). This shape of curve applies to an








organization where individuals at the lower level as a group

have more control than the individuals at the uppermost

level, even though these may be active and effective leaders.

In some organizations, there may be very little increase

in the degree to which the various levels institute control

until the top of the organization is reached; there a great

increase takes place. This type of organization is con-

trolled by only a few individuals (Figure 5). Other organi-

zations may be characterized by a relatively flat curve.

Such a curve may be low and flat, indicating a very low

degree of control throughout the organization (dotted line

in Figure 6). On the other hand, a flat curve might be

high on the vertical axis, indicating that people at all




High High



a 0
4-3 .1-
o 0











Top Bottom Top Bottom
Hierarchical Levels Hierarchical Levels

Figure 3. Autocratic model Figure 4. Democratic model
0 0
.3 4\3






Low Low ^/

Top Bottom Top Bottom
Hierarchical Levels Hierarchical Levels

Figure 3. Autocratic Model Figure 4. Democratic Model





46

levels in the organization have a great deal of influence

(solid line in Figure 6).2

It should also be readily apparent that these curves

may differ from one another, not only in their shape, but

also in their average height, suggesting, at least theoret-

ically, that organizations may differ in their total amount

of control, as well as in the relative amount of control

exercised by the respective hierarchical levels. If this

be true, then it becomes apparent that the control graph

illustrates a concept that is in opposition to what some

writers refer to as "the dominant tendency in the litera-

ture... that there is a fixed quantity of power in any
:3
relational system...."





High High


o 0


U U
o 0






Lou Low

Top Bottom Top Bottom
Hierarchical Levels Hierarchical Levels

Figure 5. Dictatoral model Figure 6. Anarchic and Poly-
archic Models





47

The assumption of a variable amount of control in

organizations represents to Tannenbaum, an assumption of

basic theoretical and practical importance. He states

Theoretically, this assumption opens up a number
of possibilities that would not otherwise be apparent.
Consequently it allows us to resolve what might other-
wise appear to be opposing and irreconcilable arguments
concerning the implications of control in organizations.
For example, one argument holds that the enhancement of
control by rank-and-file members is essential for in-
creasing organizational effectiveness, because involve-
ment in decision making by these persons, especially in
the context of a "democratic society," is necessary to
foster conditions of identification, motivation, and
loyalty. On the other hand, the conflicting argument
goes, a high degree of control by leaders is necessary
for the efficient direction and administration of organ-
izations. Our use of the control graph has led us to
question the "fixed-pie" assumption underlying this
controversy and has raised the question of why increased
control exercised by both leaders and members does not
create conditions for more effective organizational
performance.4

Control curves then give us, by implication, a good

deal of information about an organization. They tell us

something about how control is distributed in an organiza-

tion also something about the total amount of control that

is instituted in that organization. They can also tell us

whether the actor is doing the controlling (active control)

or is being controlled (passive control). They can be used

to describe the amount of control and the distribution of

control that is desired by the membership. This can then

be compared with the perceived control curves. If these two

curves, perceived and desired, were superimposed, the dis-

crepancies between them would be expected to have an important

bearing on membership participation, involvement, and loyalty





48

to the organization. In an organization where the perceived

control curve differs sharply from the desired curve one

would expect to find a corresponding degree of dissat-

isfaction, frustration, and disaffection among the members.

This can be contrasted with another hypothetical organiza-

tion which has the same perceived control curve but in which

the desired control curve corresponds more closely to the

perceived curve and in which one would expect and predict a

greater degree of membership satisfaction and loyalty.5

Several capabilities of the control graph approach to

the study of control in organizations have been suggested

in the literature. Some of them are as follows

1. It provides a convenient device for characterizing

and thinking about control in social systems.6

2. It provides a method of description which is both

quantitative and conceptually meaningful.7 In this sense it

has been stated that it can be used to conceptualize the

prescriptions of major organizational theories in terms of

the amount and distribution of control affecting organizational

functioning.B

3. It illustrates the importance of two distinct aspects

of control in organizations the distribution of control and

the total amount of control.9

4. It has been offered as one approach to the compara-

tive study of organizations with the advantages of being a

general, quantitative technique with conceptual as well as
operational potentialities.10







5. It opens up to the process of scientific testing a

number of hypotheses that have been discussed primarily in

speculative terms.11 For example, it can be used to test

hypotheses which emanate from bureaucratic theory, power

equalization and participative management models as well as

the theory of a possibly increase in the amount of total

control.12

This is an impressive list but it is not the writer's

intention to suggest that these are unique characteristics

of the control graph approach only, or to create the impres-

sion that the approach is without weaknesses, because it does

have some definite limitations. One of the most criticized

limitations pertains to the method of measuring control.

The control graph approach relies for measures of control

largely on the averaged judgments by organization members

to questionnaire items dealing with the amount of influence

or control exercised by various groups in their organization

and, of course, organization members differ in their judgments

about control. Proponents of this approach logically explain

away this limitation by maintaining that averaged responses

are quite stable. And by stating that the fact that in-

dividual respondents may be unsure of their answers and

that they may be in error does not in itself vitiate the

method, provided that respondents give better than chance

answers, that the errors are random, and that a sufficient

number of respondents are available.







Another limitation pertains to the fact that there are

aspects of control in organizations that appear to have

broad implications, but are not fully reflected in the con-

trol graphs as they are presently drawn. While the graphs

are designed to describe the amount of control that indi-

viduals at various hierarchical levels exercise, they do

not describe the means through which this control is

exercised.

Finally there is the limitation that these curves may

be better diagrammed in terms of specific areas of control

rather than in terms of general control in the organization.

Critics of the approach maintain that if specific areas are

used, a satisfactory way of integrating these to provide an

overall picture must be developed.

Despite these limitations the control curves have been

used successfully to test relationships between aspects of

control and other organizational factors. Many questions

and hypotheses relating control to other functions in organ-

izations have been proposed in the literature. For example,

how does control relate to conformity behavior, to partici-

pation, to leadership characteristics, to the ideology or

philosophy of an organization? What effects do such vari-

ables have on the control structure of an organization, and

in what ways are they affected by that structure? These

and other implications and questions relating to the use of


these curves are summarized below.





51

Applications



Prototypes

The control curves characterize in terms of two contin-

uous variables a number of organizational types that have

heretofore been treated as important but discrete. Such

concepts as democracy, autocracy, and laissez-faire as

distinct types or classes of control structure are brought

into a single schema. Thus, this approach provides a

unitary way of looking at these types, and at the same time

working with the many variations between these extremes.

Several articles have been written on this subject, but

one of the more informative ones is the article written

by Tannenbaum in which he discusses the various types of

organizational control, based on a study of four local
13
unions. Other supporting studies include the ones accom-

plished by Tannenbaum and Kahn, Likert;14 Mann and Hoffman;15

Williams, Hoffman and Mann;1 and McMahon and Perritt.1



Distribution of Control

The control curve presents a picture of the control

distribution in an organization. Organizational control,

however, is a more fluid and dynamic process than is

suggested by the control curves, which reflect the situation

at a given point in time. The question of alignments and

coalitions is one that should be considered.18 For example,
one group may be lower than another in the amount of control







it can institute in the organization, but it might add to

its effective control by joining forces with a third group.

Groups might do this on a temporary basis relative to a

specific issue, or they might form a more permanent faction

or clique.

The distribution of control in formal organizations

has been the subject of investigation by Tannenbaum and
19
Georgopoulos. In summarizing the results of their study

the authors make the following statements,

We have presented an analytical framework for the
study of the distribution of control in formal organi-
zations, illustrated with data from recent research,
and have indicated some of the issues involved as well
as some of the directions which further research may
follow. Beginning with the "control graph," we have
elaborated on four major concepts pertaining to the
distribution of controls (1) active control, (2) passive
control, (3) orientations of control, and (4) sources
of control. In each case, we have proposed some hy-
potheses which could be fruitfully investigated within
a distribution of control approach to the study of
formal organization.20

The distribution of control in foreign industrial

organizations has also been the subject of investigation.

with the idea in mind that organizations in all societies

share common characteristics.21 The universality and

centrality of control in organizations suggest it as an

important area for study and particularly as an area within

which comparative research may profitably be conducted.


Total Amount of Control

The issue of total amount of control in a system has
been of concern to social scientists more implicitly than







explicitly.22 most analyses of control have been concerned

with the relative control exercised by groups within

organizations rather than with total amount. The litera-

ture, therefore, provides little guidance concerning the

conditions under which the amount of control in a system

may expand.

As was pointed out in Chapter II, traditional analyses

of social power assumed that the total amount of power in

a social system was a fixed quantity and that leaders and

followers were engaged in a "zero sum game," where increas-

ing the power of one party must be accompanied by a corre-

sponding decrease in the power of the other.

Today there are several social scientists who are

inclined to question this assumption of a fixed quantity

of total amount of power. Among this group one finds such

names as Deutsch, Lammers, Likert, Parsons, Tannenbaum and

Kahn. In their opinion the total amount of power in a

social system can grow, and leaders and followers can

therefore enhance their power jointly. Conversely, the

total power may, in their opinions, also decline, with all

the groups in the system suffering corresponding decreases.

The control graph is of important theoretical interest

concerning this subject since it generates two organizational

or system measures! the total amount of control represented

by the height of the control curve and the distribution of

control represented by the slope. These measures have been

used to describe an organization's control structure and in






54

hypothesis testing by operationalizing prescriptions of

different organizational theories.3

In his book, Control in Organizations, Tannenbaum3

presents a collection of programmatic studies in which the

control graph is used to identify relationship between the

amount and distribution of control and measures of organi-

zational effectiveness defined largely in terms of members'

satisfaction and some production indicators. Specifically,

it is demonstrated that organizational effectiveness is

directly related to the amount of total control (the mean

height of the control graph)24 It is in this work that

Tannenbaum uses the control graph approach to attack the

controversy concerning the quantity of power in a social

system.

Our use of the control graph has led us to question
the "fixed-pie" assumption underlying this controversy
and has raised the question of why increased control
exercised by both leaders and members does not create
conditions for more effective organizational per-
formance.25

He demonstrates how, by using the control curves, an

increase in the total amount of control can be measured.

Curve X, in Figure 7, illustrates this concept. He says

"by comparison with curve A, curve X is both more 'dem-

ocratic,' in the sense of greater control by lower echelons,

and more oligarchicc,' in the sense of greater control by

upper echelons which in traditional terms, is a contra-

diction."26 For a more definitive explanation of how this,

in his opinion, can occur and for a discussion of some of the








approaches he sees as being possible to enhancing the total

amount of control the reader is directed to pages 14 through

23 of his previously referenced book.3





Amount of
control
exercised


A very great_
deal

B

A great deal- N



Quite a bit-
N.


Some-

%X

Little or-
none I A

Top middle Bottom
Level Level Level

Hierarchical Echelons

Figure 7. Increase in Total Amount of Control



Perceived and Desired Control

Control curves have been drawn on the basis of re-

sponses to questions asked of members regarding the amount

of control which various echelons or levels exercise. In





56

addition to providing a description of the situation as

members see it, the graph has also been used to characterize

the pattern of control which members desire. These concepts

have often been referred to in the literature as "actual"

and "ideal" control respectively. Comparisons have then

been made of perceived and desired control based on the

judgments of members of the organizations. Smith and

Tannenbaum used this approach to accomplish a comparative

analysis of several aspects of organizational control for

a number of organizations.27 Tannenbaum used this approach

in a study of control and effectiveness in the "The League

of Woman Voters."28

French, Israel and As, have presented experimental

data to show that discrepancies between the influence which

members perceive to exist in an organization and that which

they feel should exist are related to aspects of worker -

management relations and to member satisfaction.29

Finally, March and Simon have argued that disagree-

ments between members and leaders regarding organizational

facts and ideals, including facts and ideals about control,

are among the conditions contributing to intergroup conflict

within organizations, and that one might also expect thesp

discrepancies to have some bearing on member satisfaction

and productivity.3

From these and other studies231 it can be seen that the

control graph approach can be used to study relationships

between control, organizational effectiveness, and member

attitudes.








Control and Uniformity

The relationship between control and member uniformity

has been traditionally subsumed under the concept of social

norm. This concept can be defined simply as the continuous

uniformity in expectations, attitudes, or behavior within

a group regarding an activity developed and maintained by

processes of control. Central to this definition of norms

is the premise that they are a function of control. The

control graph approach has been used to study the relation-

ships of varying patterns of organizational control to

member uniformity. For example Smith and Ari used it in the

study of a nationwide service organization which had opera-

tions in several metropolitan areas of the United States.32

The findings of their research suggest that the pattern of

control which tends to be associated with member consensus

is that predicted by a high amount of control exercised by

members at all echelons, leaders as well as rank-and-file

members. High total control tends to be conducive to con-

sensus both within the work group and between the rank-and-

file and the supervisory levels. The findings further suggest

that high total control was efficacious in promoting member

consensus in the organizations under study because it was

associated with significant influence by the rank-and-file

members upon the operation of the organization. This says

the authors, "is substantiated, in part, by the significant

relationship between total control and the morale of the

members (r=.72) and by the significant relationships between





58

the influence of the rank and file upon the operation of

the station and the measures of work group and hierarchical

consensus.33 The high-producing station was found to be

characterized by high total control, high member consensus

and high member morale. The multiple correlation of total

control and general station consensus with member morale was

.72.

Recent research in several organizations has indicated

that the manner in which control is structured is related

to organizational effectiveness. These studies suggest

the importance in some organizations of high rank-and-file

control relative to leadership control and, more generally,

the importance of a high amount of control exercised by
34
members at all echelons in the organization. The inter-

pretations offered of these findings suggest that these

patterns of control may be conducive to high organizational

effectiveness, in part, through the uniformity with respect

to organizational standards and policies which they promote.

These interpretations seem to suggest one particularly

significant process explaining the efficacy of these patterns

of control in promoting high organizational performance,

namely, the coordination and regulation of member behavior

with respect to organizational norms. The resulting uniformity

derives its significance from the fact that it is basic to

the concerted member effort underlying effective organizational

performance. The importance for organizational functioning

of such variables as member consensus and reciprocal role








expectations has been suggested in a number of studies. Basil

Georgopoulos, e.g., found aspects of the "normative system

of the organization" such as "normative complementarity" and

"group consensus" to be significantly related to organizational

productivity

One thing is evident from these references and that is

that the control graph approach to the study of control can

be used to consider the relationships of patterns of control

to member uniformity and then to evaluate their implications

for organizational effectiveness.



Organizational Type Comparisons

The need for comparative approaches, to the study of

organizational control, is great but comparative studies

are beset with serious conceptual as well as methodological

problems. The control graph method has been offered as one

approach to the comparative study of organizations. It has

been claimed to have the advantages of being a general,

quantitative technique with conceptual as well as operational

potentialities.36

Blau,37 Blau and Scott, Etzioni,39 Gouldner,40 and

Likert41 are among the writers who have recently attempted

conceptual categorizations of organizations based partly on

differences in control which suggest the fruitfulness of

comparative analytic approaches in understanding this phe-

nomenon.
Smith and Tannenbaum recently utilized the control graph







method to accomplish a comparative analysis of organizational

control in approximately 200 geographically separate organi-

zational units from a number of larger organizations.42

They presented data which they suggested was illustrative

of the potential of the method and suggestive of a number of

hypotheses about organizational control which, in their

opinion, were amenable to empirical tests.

Therefore it appears as if there is sufficient evidence

in the literature to support the claim by Tannenbaum and Kahn,

in their 1957 article, that the control graph approach could

be effectively utilized in comparative analyses of organiza-

tions.



Bases of Control or Power

Control in any organization may be exerted through

several different channels. As was pointed out in Chapter

II, French and Raven have developed a fivefold typology which

suggests a number of different categories of bases of control.

They propose that there are five bases of powers reward power;

coercive power; legitimate power referent power; and expert

power. To this list Katz and Kahn have added incremental

influence.

Student has applied the French and Raven power typology

and the incremental concept to a study of supervisory in-

fluence and work group performance in a manufacturing firm.43
more specifically, his research is an analysis of some per-

formance correlates of the first-line supervisors, incremental-

influence.







Using the French and Raven typology, Bachman, Smith and

Slesinger investigated salesmen satisfaction and performance.44

Their research was concerned with the relationship between

organizational effectiveness and social control in organiza-

tions. In particular it was designed to explore two aspects

of control the distribution of control among organizational

levels (control graph), and the bases for this control.

Using the same approach that Tannenbaum and Smith had

used previously,45 Ivancevich analyzed the relationship

between control, bases of control, and three categories of

satisfaction.46 The findings of his research tend to support

portions of previous control-satisfaction investigations.

Bachman, Bowers and Marcus using the French and Raven

typology recently accomplished a comparative study, based

on five different organizational settings, of the bases of

supervisory power.47 Their research was concerned with two

interrelated problems Why do people comply with the requests

of organizational "superiors?" And how are these various

reasons related to the total amount of control and to organ-

izational effectiveness? Stated another way it was concerned

with the bases of supervisory power and its effects.

In summary then it can be stated that the control curves

emphasize the importance of control in organizations, and

provide a means of taking a more holistic view of it. Studies

of leadership, influence, power and the bases of power can

be made with such a framework, as evident by the above stated

projects.







Summary

The control graph is an analytical tool for the analysis

of control in organizations. This graph shows a line which

represents the amount of control (vertical axis) exercised

by each of the hierarchical levels within an organization

(horizontal axis) as perceived by a sample taken from all

levels. Specific points are located by plotting mean re-

sponses to the question, "How much influence does (top

management, middle management, lower management, rank-and-file

members) exert...?

This tool has been used to test relationships between

various aspects of control and other organizational factors

such as satisfaction, effectiveness, conflict and productivity.

In view of this comprehensiveness inherent in the "control

graph" formulation, it is clear that application of this

approach should be made to the study of control in hospitals,

a project not previously attempted but which serves as the

focal point of this investigation.













NOTES

1. Arnold S. Tannenbaum and Robert L. Kahn, "Organizational
Control Structures A General Discriptive Technique
as Applied to Four Local Unions," Human Relations,
Volume 10, Number 2 (May, 1957), pp.127-140.

2. Ibid.

3. Arnold S. Tannenbaum, Control in Organizations (New Yorks
McGraw-Hill, Inc., 1968), p.12.

4. Ibid., pp.12-13.

5. Tannenbaum and Kahn, pp.127-140.

6. Arnold S. Tannenbaum and Basil S. Georgopoulos, "The
Distribution of Control in Formal Organizations,"
Social Forces, Volume 36, Number 1 (October, 1957),
pp.44-50.

7. Arnold 5. Tannenbaum, "Control Structure and Union
Functions," American Journal of Sociology, Volume
61, Number 6 (may, 1956), pp.536-545.

8. J. Timothy McMahon, "management Control Structure and
Organizational Effectiveness," Academy of management
Proceedings, (August, 1972), p.162.

9. Tannenbaum, "Control Structure and Union Functions,"
pp.536-545.

10. Clagett G. Smith and Arnold S. Tannenbaum, "Organizational
Control Structures A Comparative Analysis," Human Re-
lations, Volume 16, Number 4 (November, 1963), pp.299-
32b.

11. Tannenbaum and Kahn, pp.127-140.

12. Mcmahon, p.162.

13. Tannenbaum, "Control Structure and Union Functions,"
pp.536-545.

14. Rensis Likert, "Influence and National Sovereignty,"
in Festschrift for Gardner Murphy. ed. John G.
Peatman and Eugene L. Hartley (New Yorks Harper
and Bros., 1960), pp.214-227.








15. Floyd C. Mann and L. Richard Hoffman, Automation and
the worker: A Study of Social Change in Power Plants
INew Yorks Henry Holt and Co., 19bJ).

16. Lawrence K. Williams, L. Richard Hoffman, and Floyd C.
Mann, "An Investigation of the Control Graphi In-
fluence in a Staff Organization," Social Forces,
Volume 37, Number 3 (march, 1959), pp.189-195.

17. J. Timothy McMahon and G.W. Perritt, "The Control Struc-
ture of Organizationst An Empirical Examination,"
Academy of management Journal, Volume 14, Number 3
(September, Ig9ti, pp.327-340.

18. Tannenbaum and Kahn, p.137.

19. Tannenbaum and Georgopoulos, pp.44-50.

20. Ibid., p.50.

21. Josip Zupanov and Arnold S. Tannenbaum, "The Distribu-
tion of Control in Some Yugoslav Industrial Organi-
zation as Perceived by members," in Control in
Organizations. ed. Arnold 5. Tannenbaum LNew Yorks
McGraw-Hill, Inc., 1968), pp.91-112.

22. Ralf Dahrendorf, Class and Class Conflict in Industrial
Society (Stanford Calif. IStanford University Press,
1959).

23. J. Timothy McMahon and G.W. Perritt, "Toward a Contingency
Theory of Organizational Control," Academy of Manage-
ment Journal, Volume 16, Number 4 (December, 197),
p.625.

24. Ibid.

25. Tannenbaum, Control in Organizations, p.13.

26. Ibid.

27. Smith and Tannenbaum, pp.299-326.

28. Arnold S. Tannenbaum, "Control and Effectiveness in a
Voluntary Organization," American Journal of Soci-
ology, Volume 67, Number 1 (July, 1961), pp.33-4b.

29. John R.P. French, Jr., Joachim Isreal, and Dagfinn As,
"An Experiment in Participation in a Norwegian
Factory," Human Relations, Volume 13, Number 1
(February, 1960), pp.3-19.








30. James G. March and Herbert A. Simon, Organizations
(New Yorks John Wiley and Sons, 1963).

31. David G. Bowers, "Organizational Control in an Insurance
Company," Sociometry, Volume 27, Number 2 (June, 1964),
pp.230-244.

32. Clagett G. Smith and Oguz N. Ari, "Organizational
Control Structure and member Consensus," American
Journal of Sociology, Volume 69, Number 6 (may,
19b4), pp.b23-b68.

33. Ibid., p.636.

34. Ibid., p.623.

35. Basil S. Georgopoulos, The Normative Structure of Social
Systems: A Study of Organizational Effectiveness
(Doctoral dissertation, University of michigan,
1957).

36. Tannenbaum and Kahn, pp.127-140.

37. Peter M. Blau, Bureaucracy in modern Society (New Yorks
Random House, 195b).

38. and W. Richard Scott, Formal Organizations:
A Comparative Approach (San Franciscos Chandler Pub-
lishing Co., 19b2).

39. Amitai Etzioni, A Comparative Analysis of Complex Organ-
izations (New York: Free Press, 19i1).

40. Alvin W. Boulder, Patterns of Industrial Bureaucracy
(New Yorks Free Press, 1954).

41. Rensis Likert, New Patterns of management (New Yorks
McGraw-Hill Inc., 1901J.

42. Smith and Tannenbaum, pp.299-326.

43. Kurt R. Student, "Supervisory Influence and Work-Group
Performance," Journal of Applied Psychology, Volume
52, Number 3 (June, 196B), pp.18B-194.

44. Jerald G. Bachman, Clagett G. Smith and Jonathan A.
Slesinger, "Control, Performance, and Satisfactions
An Analysis of Structural and Individual Effects,"
Journal of Personality and Social Psychology,
Volume 4, Number 2 (August, 19bb), pp.12Y-136.





66

45. Arnold S. Tannenbaum and Clagett G. Smith, "The Effects
of member Influence in An Organizationn Phenomenology
Versus Organization Structure," Journal of Abnormal
and Social Psychology, Volume 69. Number 4 kCctober,
1964), pp.401-410.

46. John m. Ivancevich, "An Analysis of Control, Bases of
Control, and Satisfaction in an Organizational
Setting," Academy of management Journal, Volume 13,,
Number 4 (December, 1970), pp.427-43b.

47. Jerald G. Bachman, David G. Bowers, and Philip M.
Marcus, "Bases of Supervisory Powers A Comparative
Study in Five Organizational Settings," in Control
in Organizations. ed. Arnold S. Tannenbaum (New Yorks
McGram-Hill, Inc., 1968), pp.229-238.













CHAPTER IV


RESEARCH DESIGN, METHOD, AND PROCEDURE



Overview

The main purpose of this chapter is to describe the

significant aspects of the research design, method, and

procedure.

Following this brief overview will be a discussion of

the procedure and rationale used to select the particular

hospitals for study. Then the how, and why, certain

groups of hospital personnel were specified to represent

each participating institution, as well as how individual.

group members were selected to take part in the study will

be reviewed. In the next section there will be a descrip-

tion of the main research instruments used to collect data,

the kinds of data collected, and the response rates attained.

Next there will be a discussion of the areas of investiga-

tion, the hypotheses tested and the methods of measurement.

The final section of this chapter will be devoted to a

discussion of the main methodological limitations of the

study. The division of this chapter into these sections

is not intended to insinuate that the different aspects







of the research design can be neatly separated and treated

independently of one another. There is a high degree of

interrelatedness among the separately treated features of

the design just as has been suggested exists in the "scien-

tific method of problem solving."


Selection of Hospitals

Of the approximately 7,125 hospitals in the United States

listed in the 1973 Guide Issue of Hospitals, about 6,300

were short-term institutions of different types. And of the

latter, at least 5,900 were nonfederal general hospitals,

i.e., institutions engaged in the care and treatment of

acute diseases and illnesses. Of all these nonfederal, short-

term general hospitals in the nation, nearly 2,575 were

either state and local governmental institutions or pro-

prietary hospitals. This left about 3,325 hospitals that

were nongovernmental, not-for-profit institutions. These

hospitals represented close to 47 percent of all hospitals

in the nation and they had a combined total of almost 630,000

beds and were accomplishing approximately 22,500,000 admiss-

ions annually. This is the type of hospital with which the

ordinary person is most likely to be familiar or have had

contact as a patient or visitor.

These hospitals were all alike in some ways. For example

they! (1) were all of the same general type (2) shared a

number of needs and experienced certain important organi-

zational problems in commons and (3) were all concerned with








the same general objective that of providing adequate

patient care effectively. Because of the extensive coverage

in academic, professional, and trade journals of the subject

of hospital objectives it becomes necessary at this point to

digress from the main subject under discussion and cover in

general terms some of the material published in these journals

concerning hospital objectives.

A review of the literature reveals that hospital objec-

tives have been discussed from a number of different perspec-

tives. From the "community needs" standpoint one finds such

quotes as the following.

The hospital is the professional monitor of the quality
and quantity of care rendered not only on its own premises
but throughout its community.2

The objective of the hospital is to provide the best
in patient care at a cost the patient and the community
can stand.3

Generally, individual hospitals wish to allocate the
limited monies they have available in a manner which will
result in the maximum obtainable benefits (for the com-
munity) from any given level of expenditures benefits
referring to both (1) those which can be readily measured
in dollar terms, and (2) those which result in a higher
level of health care for the community (in terms of lower
mortality, lower morbidity, improved health for target
populations, etc.) These at least are the publicly stated
goals.4

Health care institutions that have as an objective the
provision of adequate community health facilities must
choose from among alternative uses of resources in their
efforts to meet this goal.5

The goals of the hospital should be to meet the needs
of the community served by the hospital, whatever those
may be. The hospital must have as its goal the satisfac-
tion of the needs of the community that it serves even
though these may not accurately represent its demands.6








The notion that the voluntary hospital was established
by the community to serve its health care needs is un-
deniably a true statement.?

The hospital must provide a value to the community
which is superior to that obtainable in any other manner.8

The basic objective of the management of the Society
can be stated as follows utilizing appropriate, effec-
tive and efficient management techniques to assist each
hospital (Society) and its community in the creation of
a comprehensive medical care center which will fulfill
community needs in the most effective way in a voluntary
setting.9

In addition to the community needs standpoint there is a

volume of material which addresses hospital objectives in

general terms. The following statements are typical of

what is found in some of the literature.

The medical care industry has as its prime goal the
development and maintenance of optimum health levels.10

The primary or basic objective of any health care
institution is that of providing quality services at
optimum cost.11

Most of the hospital administrators and assistant
administrators interviewed indicated that hospital organ-
izations have two basic objectives quality patient care
and financial soundness (efficiency).12

Their objective is delivering to each patient the
services required to treat effectively his illness or
injury and then discharging him.13

The Samaritan Health Service went carefully through a
project study period and came up with five basic objec-
tivess (1) the containment of costs accomplished through
numerous efficiencies brought about by centralized pro-
grams, (2) the accessibility of health care made available
to people regardless of income or geographic remoteness,
(3) the systematic phasing out of duplication and frag-
mentation of both services and facilities, (4) the esca-
lation and broadening of the quality of patient care,
and (5) the innovative break with tradition keeping
people out of hospitals and treating them while still on
their feet rather than flat on their backs after they
have become catastrophic statistics.14







There is also evidence that in the minds of some of the

writers there is a vagueness about the objectives of hos-

pitals.

Objectives for the hospital are generally vague and 15
express broad general intentions to render public service.

The absence of definable and generally acceptable
hospital organization objectives makes a normative
approach to hospital performance infeasible.16

The absence of a clear-cut primary objective for non-
governmental, not-for-profit hospitals to fill the role
played by profit or wealth maximization in most economic
studies, has long been a source of discouragement for
economists interested in studying the health sector.17

Some of the writers even venture to speak of the future

objectives of hospitals.

The health care market is made up of many kinds of
illness for which various kinds of institutions and
various kinds of services are needed. The social thrust
of the future will be for total and comprehensive health
care services under single management. In the future,
hospital authorities will no longer operate individual,
autonomous hospitals devoted to one aspect of the total
need. They will operate hospital systems in which the
organizational structure will be different, and the
skills needed will be different, but all will be related
to the total needs of the community.18

The hospital will become both the primary operational
center for community health services and the primary center
for comprehensive health planning at the community level.19

Finally there is in the literature a discussion of hospital

objectives from an economic models perspective. One of the

leading articles in this area is the one written by Richard
20
W. Foster. In this article the author discusses the six

most popular economic models of hospital behavior. He

begins by stating that "the practicing administrator will

find none of these models realistic. He is likely to feel

that the forces identified by the models are real, but that





72

many other forces have been neglected.21

Considering all of the literature which discusses

hospital objectives from these five perspectives (1)

community needs, (2) general terms, (3) vagueness, (4)

future, and (5) economic models, the general objective for

the nongovernmental, not-for-profit hospitals previously

stated by the author appears to be quite in character.

Continuing with the discussion of the nongovernmental

not-for-profit hospitals which was discontinued on page 69

it is pointed out that even though these hospitals were

similar in a number of ways, they still, constituted a

very heterogeneous group of organizations. For one thing,

some of the hospitals were extremely small having 25 or so

beds, while a few others were extremely large, having more

than 750 beds each. The majority, of course, fell between

these two extremes, but even a range of 25 to 750 beds is

very wide. A 50-bed hospital, for example, is likely to

be very different from a 450-bed hospital, and the same

may be said for a 100-bed hospital in comparison to a

400-bed hospital, although the differences may be smaller

in the latter case.

The differences that may be associated with the size

of the hospitals virtually defy enumeration. There are

differences in the size and kind of staff and personnel

required by a small as compared to a large institution.

There are differences in the environment within which

larger and smaller institutions operate; for example, a








small rural community would be unlikely to be supporting

a 500-bed hospital. In general, with increased organiza-

tional size, there is more departmentalization, more

specialization, more heterogeneity, and more complexity
22
in organization and operations.2

The hospitals in question differed not only in size

and size-related characteristics, but also on a number of

other dimensions. Included among these were (1) profes-

sional accreditation by the Joint Commission on Accredita-

tion of Hospitals -while most of the hospitals were

accredited, some were not; and (2) regional-geographic

location, and the type of community within which the

hospitals operated some hospitals were located in New Eng-

land, some in the south, and others in the.east, west or

midwest, and similarly, some served huge metropolitan

areas, while others served small cities or towns. Of

course, there were other dimensions on which these hospitals

varied. But for the present purpose the differences cited

here are sufficient to illustrate the point that a great

many significant differences a good deal of heterogeneity

or variance characterizes this population of general

hospitals.23

From the standpoint of research design, the crucial

problem that differences of this kind pose may be stated

as follows

Unless the researcher takes cognizance of the
heterogeneity of the population with which he is
dealing, either by controlling many of the differences
involved through his initial study design or by making








sure to study the effects these differences may have
upon the phenomena he proposes to investigate, he will
end up with many spurious results or "impure" findings,
which he mill be unable to explain. And ideally, of
course, the researcher wants to be able to understand
and explain the phenomena he is studying....24

The problem of how to guard against spurious results
due to great heterogeneity in the population of hos-
pitals may be handled in either of two main ways.
First, the researcher may restrict his initial popula-
tion to a sub-population, thus reducing much of the
unwanted heterogeneity. He may impose specific re-
strictions and qualifications that the hospitals in
the study should meet (based on differences in the
population he considers important enough to avoid,
control for, or keep constant without actually measuring
them), and instead of dealing with all of the popu-
lation, deal with a much smaller sub-class of hospitals
which meet certain criteria. Alternately, rather than
restrict the population, he may have a large enough
number of hospitals participate in the study, so as
to capture much of the heterogeneity prevailing in
the whole population.... He may increase the size
of his sample to a number which permits him to rep-
resent reasonably well the entire population of
hospitals by his sample....25

Restricting the population has the advantage of
obtaining relatively pure rather than spurious results
with relatively few hospitals participating in the
study, and the limitation of not being able to generalize
the results as much as might be desired. Increasing the
size of the sample has the advantage of permitting
greater generalization of the findings, and the dis-
advantage of requiring greater costs, time, efforts,
and energies. Assuming limited funds, the researcher
is almost inevitably forced toward restricting the
population and away from a relatively large sample.26

In this case-study only two hospitals were used, this, of

course, has the disadvantage of not being able to "gener-

alize" the results beyond the hospitals involved, which is

one of the major limiting factors of the study.

Needless to say, the difficulties posed by the hetro-

geneity of the population, and by considerations of economy

and feasibility, were not the only factors used to determine








the selection of the particular hospitals which were asked

to participate in this study. Included among the other

determining factors were the nature of the study itself,

i.e., the particular aims of the study and the questions

that were to be answered, also the very character of the

phenomena in which the researcher was interested.

The very nature of the study was an important deter-

minant in the selection of the hospitals. Georgopoulos and

mann state that

If the research is of an exploratory kind, aiming
to yield some unavailable information, stimulate
insights, or aid the formulation of hypotheses in
anticipation of more systematic studies then it is
not crucial for the researcher to study a large or a
"representative sample" of organizations. The same
also applies if the research is of the "case-study"
type, not aiming at generalization of results or
rigorous hypothesis testing, or if the research happens
to be too circumscribed in its objectives as is the
case with many small experimental studies. If, on the
other hand, the research is of the explanatory kind
aiming to test rigorously hypotheses of wide generality,
or to yield results by studying a sample of the total
population and then generalize these results to the
whole population then the organizations actually studied
must constitute a probability sample that is"represent-
ative" of the whole population.27

This study belongs to the "case-study" category of research

and therefore according to the above reference, it was not

crucial for the researcher to study a large or a represent-

ative sample of hospitals.

Georgopoulos and mann further state that#

If the research is in part exploratory and de-
scriptive and in part explanatory and analytical,
designed both to develop hypotheses for further study
and to test hypotheses based on already available








research and theory, then a design that restricts the
population to which the results may potentially apply,
but includes an adequate number of organizations in
the research to permit the use of sound analytical
procedures, may yield the best possible solution.
Such a design provides a reasonably good alternative
in place of the ideal "representative sample" design,
or in place of a large sample design, which is not
feasible because of considerations of costs and
economy and/or because of lack of sufficient prior
knowledge about the phenomena with which the research
is concerned.28

This case-study was designed to be in part both exploratory

and explanatory. In terms of this design, generalization

of the findings beyond the hospitals studied can not be

attempted on statistical grounds, but logical inferences

can be made based on theoretical principles.

Up to this point reference has only been made to how

the problems of spuriousness and economy, and the nature

of the research affected the decision concerning the kind

and number of hospitals that were to be included in the

study. This decision is further clarified by a listing of

the specific criteria which mere employed to select the

hospitals. They were as follows

1. Type of service rendered short-stay general hos-

pitals.

2. Sizes approximately 400 beds.

3. Ownership, and institutional control and affilia-

tions nonprofit, nongovernmental institutions (Categories

21 and 23 in the Guide Issue). One from category 21 (church

affiliated and operated) and one from category 23 (other).

4. Administrations administrated by a chief executive








officer (administrator, executive director, director,

executive vice president, etc.), under a policy-making body

at the local level known as the governing body (board of

directors, board of trustees, council, etc.)

5. Status fully accredited by the Joint Commission

on Accreditation of Hospitals.

6. Region and geographic locations located in the same

geographic region.

7. Workload of comparable size and complexity.

8. Staffsi of comparable size and kind.

9. missions or purposes similar in nature.

10. Facilities comparable facilities available in each

institution.

A detailed listing of the specific criteria as they

apply to the final two hospitals which were chosen to par-

ticipate in the study, is found in Appendix A. The purpose

of these specific criteria was to have the hospitals as

homogeneous as possible, except for item number three (3)

where one was to be affiliated with a church organization

and the other was to be nonchurch affiliated. The reason

behind this was to determine what, if any, significant

differences existed in the patterns of control by hierarchical

level between the two hospitals.


Selection of the Individual
Groups and Respondents

This study was designed to explore several facets of








influence among various hierarchical levels in two hos-

pitals. The hierarchical levels considered were those of

the formal hospital organization as depicted in Dr. Charles

U. Letourneau's book The Hospital Administrator.29

They are as follows

1. Governing Body (board of directors, board of

trustees, council, etc.).

2. Chief Executive Officer (hospital administrator,

executive director, director, etc.).

3. Directors of Various Services (director of profes-

sional services, director of personnel, director of mate-

rials management, director of finance, etc.).

4. Department Heads (department of medicine, department

of surgery, department of nursing, department of radiology,

laboratory, department of physical medicine, hospital

pharmacy, dietary department, etc.).

5. Unit Managers and other first line supervisory

personnel (A unit manager is a coordinator of activities

within the department, a controller of materials and costs

and supervisor of administrative personnel. The unit

manager serves in a staff capacity to the department head.

He may be referred to by such titles as service manager,

administrative coordinator, administrative assistant, etc.).

6. Nonsupervisory employees.

From this enumeration of the hierarchical levels it

should be apparent that this study is primarily concerned

with the overall institutional authority line which origi-
nates with the governing body, however, recognition is also




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - Version 2.9.7 - mvs