University of Southern California Dissertations and Theses

Social-Psychological Factors In Organizational Attachment

(USC Thesis Other)

Social-Psychological Factors In Organizational Attachment

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content SOCIAL-PSYCHOLOGICAL FACTORS IN
ORGANIZATIONAL ATTACHMENT
by
Lincoln Jacob Fry
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirement for the Degree
DOCTOR OF PHILOSOPHY
(Sociology)
August 1972
INFORMATION TO USERS
This dissertation was produced from a microfilm copy of the original docum ent.
While the most advanced technological means to photograph and reproduce this
document have been used, the quality is heavily dependent upon the quality of
the original subm itted.
The following explanation of techniques is provided to help you understand
markings or patterns which may appear on this reproduction.
1. The sign or “ target" for pages apparently lacking from the docum ent
photographed is "Missing Page(s)". If it was possible to obtain the
missing page(s) or section, they are spliced into the film along with
adjacent pages. This may have necessitated cutting thru an image and
duplicating adjacent pages to insure you complete continuity.
2. When an image on the film is obliterated with a large round black
mark, it is an indication th a t the photographer suspected th a t th e
copy may have moved during exposure and thus cause a blurred
image. You will find a good image of the page in the adjacent frame.
3. When a map, drawing or chart, etc., was part of the material being
p h o to g rap h e d the photographer followed a definite m ethod in
"sectioning" th e material. It is customary to begin photoing at the
upper left hand corner of a large sheet and to continue photoing from
left to right in equal sections with a small overlap. If necessary,
sectioning is continued again — beginning below the first row and
continuing on until complete.
4. The majority of users indicate that the textual content is of greatest
value, however, a somewhat higher quality reproduction could be
made from “ photographs" if essential to the understanding o f the
dissertation. Silver prints of “ photographs" may be ordered at
additional charge by writing th e Order Department, giving the catalog
number, title, author and specific pages you wish reproduced.
University Microfilms
300 North Zeeb Road
Ann Arbor, Michigan 48106
A Xerox Education Com pany
; w l
Cl
SI
73-4912 1
FRY, Lincoln Jacob, 1937- I
SOCIAL-PSYCHOLOGICAL FACTORS IN ORGANIZATIONAL I
ATTACHMENT. \
!
University of Southern California, Ph.D., 1972 ’
Sociology, general i
&
University Microfilms, A X E R O X Company, Ann Arbor, Michigan
(^Copyright "by
LINCOLN JACOB FRY
1972
UNIVERSITY O F SOUTHERN CALIFORNIA
THE GRADUATE SCHOOL
UNIVERSITY PARK
LOS ANGELES, CALIFORNIA 9 0 0 0 7
This dissertation, written by
.......U.te.Qln. Jacob.. Fry................
under the direction of h.Xs... Dissertation Com
mittee, and approved by all its members, has
been presented to and accepted by The Graduate
School, in partial fulfillment of requirements of
the degree of
D O C T O R O F P H I L O S O P H Y
Dean
D ate...
DISSERTATION COMMITTEE
phatrman
PLEASE NOTE:
Some pages may have
indistinct print.
Filmed as received.
University Microfilms, A Xerox Education Company
ACKNOWLEDGMENT
I wish to express my appreciation to my committee,
Jon Miller, Solomon Kobrin, and Salvatore Merendino. I
wish to express my deepest gratitude to my chairman, Jon
Miller, for his inspiration and continuing help. Without
his guidance, this dissertation would not have been
possible.
ii
TABLE OP CONTENTS
Chapter Page
I. SOCIAL-PSYCHOLOGICAL FACTORS IN
ORGANIZATIONAL ATTACHMENT ................... 1
Review of the Literature ................... 8
Industrial settings ...................... 9
Professionals in organizations * ......... 16
The local-cosmopolitan construct ......... 18
Laboratory studies ........................ 27
Summary........................................31
The Present S t u d y .............................36
II. THEORETICAL CONSIDERATIONS AND DERIVATIONS
OP HYPOTHESES................................... 39
Reference Group Theory ...................... 4l
Occupations and professions as reference
groups..................................... 42
Exchange theory as a supplement to
reference group theory ................. 49
Derivation of Hypotheses ................... 52
Exploration of Other Variables ............. 56
Summary........................................57
III. THE SAMPLE AND METHOD OP ANALYSIS............. 59
The Sample..................................... 59
The Questionnaire............. 61
Measurement of attachment ............... 62
Reference group orientation ............. 64
Professional or formal training ......... 66
Organizational socialization ............. 66
Work satisfaction.......................... 66
Tension..................................... 68
Job freedom.................................69
iii
Chapter Page
Job clarity............................. 70
Interpersonal rewards .................... 72
Statistical Techniques ...................... 73
Measures of association .................. 73
Statistical program ...................... 73
IV. TEST OP THE HYPOTHESES AND DISCUSSION .... 75
Overview of the Findings...................75
Attachment and exploratory variables . . . 97
Summary of the test of the hypotheses . . 101
A Multiple Regression Analysis ............. 102
The statistical package for the
social sciences ........................ 104
Implications of the multiple
regression analysis .................... 109
Theoretical Implications of the Findings . . 110
Implications for Future Research ........... 112
Summary of the Findings..................114
i REFERENCES............................................116
i APPENDIX..............................................135
LIST OP TABLES
Table
1.
2 .
3.
4.
5.
Relationships Among Organizational
Attachment, Reference Group Orientation,
Education, Occupation, Employment
Length, Interpersonal Rewards, and Sex
Among 260 Members of Pour Professional
and Semi-professional Organizations . . . .
Relationships Among Attachment and
Education, Reference Group Orientation,
Length of Employment, and Interpersonal
Rewards Among 260 Members of Pour
Professional and Semi-professional
Organizations, with Statistical Controls. .
Relationships Among Attachment, Occupation,
Education, Reference Group Orientation,
Employment Length, Interpersonal Rewards,
and Sex in Each of Pour Professional and
Semiprofessional Organizations
'(Total = 260)...............................
Relationship Between Education and Reference
Group Orientation Among 260 Members of
Pour Professional and Semiprofessional
Organizations ............................
Relationship Between Reference Group
Orientation and Organizational Attachment
Among 260 Members of Pour Professional
and Semi-professional Organizations . . . .
Relationship Between Employment Length and
Organizational Attachment Among 260
Members of Pour Professional and Semi-
professional Organizations ...............
Page
77
78
80
83
86
91
v
Relationship Between Interpersonal Rewards
and Organizational Attachment Among 260
Members of Pour Professional and Semi-
professional Organizations .............
Relationships Between Various Interpersonal
Reward Measures and Organizational
Attachment in Pour Professional and
Semi-professional Organization1 .
(Total N = 260) ..........................
Relationships Between Organizational
Attachment and Job Tension, Work Satis
faction, Job Freedom, and Job Clarity,
with Controls, Among 260 Members of
Pour Professional and Semi-professional
Organizations . . ........................
Crosstabulations of Organizational Attach
ment and Job Tension, Work Satisfaction,
Job Freedom, and Job Clarity for Pour
Professional and Semi-professional
Organizations, with'Reference Group
Orientation Introduced as a Statistical
Control ...................................
Multiple /Pfegression Analysis Using Various
Independent Variables to Predict
Organizational Attachment Among 260
Members of Pour Professional and Semi-
professional Organizations .............
CHAPTER I
SOCIAL-PSYCHOLOGICAL FACTORS IN
ORGANIZATIONAL ATTACHMENT
This study is concerned with some of the
theoretical and empirical problems in the analysis of
organizational attachment. The approach taken is basically
social-psychological. A secondary question considered by
this study is whether or not the local-cosmopolitan
typology described by Gouldner (1957) is useful in
describing patterns of organizational attachment.
•"Organizational attachment" may be broadly defined
as the degree to which an individual is bound to an organi
zation. Attachment has been used synonymously with
relationship, bond, and tie (Gewirtz, 1969:16). For the
purpose of this study, organizational attachment will be
defined as the probability that an individual will maintain
his participation in his present organization. Socio
logically, organizational attachment is important because
modern society has been described as an organizational
society (Presthus, 1962). Organizations are ubiquitous in
modern society and a source of strain inherent in
1
2
organizations takes on general sociological significance.
The present research investigates organizational
attachment in four complex organizations. Hypotheses will
be derived and tested with data generated by questionnaire
items administered to all of the personnel of these four
organizations. The questions consist of sociometric and
structured items which deal with interpersonal relation
ships on the job, satisfaction and tension on the job, and
the individual’s perception of his or her relationship to
the organization.
A useful way of conceptualizing organizational
attachment is offered by Caplow (1964). According to
Caplow’s framework, all organizations, regardless of their
size, purpose, or cultural setting, may be visualized as
facing four basic problems: stability, integration,
voluntarism, and achievement.
While all four of these problems may be seen as
related to attachment, stability has attachment as its
central theme. Stability depends upon the tie the
individual has with the organization. Integration is also
strongly associated with attachment and implies that
members of organizations will come to prefer each other as
interaction partners as opposed to non-members. Voluntar
ism is usually defined as attractiveness and is composed of
sentiments such as pride, spirit, or morale. Organizations
meet achievement criteria by honoring style, inventiveness
and prowess, and cults will develop around persons who
embody those values to a high degree. Caplow's (1964:134-
135) identification of the four basic problems of
organizations is useful in summarizing the literature that
is relevant to organizational attachment.
The problem of organizational attachment has been
especially of interest to studies concerned with the
professions. Research in this area has focused on the
problem of integration and has differentiated between
occupational and organizational integration as polar types
(for example, see Gouldner, 1957:281-306). Industrial
studies, on the other hand, have been concerned with degree
of attachment as it relates to stability. The focus on
stability has led to a differentiation between formal and
informal social organization. This distinction is between
patterns of behavior resulting from rational authority and
planning as opposed to those patterns which emerge from a
particular work group (for example, see Faunce, 1967:
281-289).
Most of the industrial studies of informal
organization have concentrated upon industrial workers and
have shown little interest with work groups composed of
managerial personnel. The literature concerned with
professional personnel, on the other hand, has tradition
ally compared two occupational groups, usually engineers
and research scientists, and has ignored all other
organizational employees. Thus, the literature which may
be considered most relevant to organizational attachment
has concentrated primarily on two distinct occupational
categories, professional and manual workers.
Since an organization is a place where work is
performed, the importance of work to various occupational
groups is closely related to organizational attachment.
Are there more common factors which are related to
organizational attachment than cut across occupational
lines? One common answer to this question has been that
work is to be considered a central life interest for adults
in modern industrial society. This kind of statement is
seriously questioned by others. It is an important issue
for this research because of the way organizational
attachment has been defined; that is, the willingness of
an individual to stay in the organization.
Attachments can be positive or they can be
negative. Attachment can be defined as a positive,
affective tie which is voluntary in nature as opposed to
attachments which are negative and restrictive. This type
of attachment may be a function of restricted opportunity.
Some researchers talk about attachment and some talk about
the more common phenomenon, dependability (for instance,
see Chinoy, 1952: ^53-^59» and Dubin, 1956:131-1^2).
In an article entitled, "Necessary versus Voluntary
Behavior," Dubin (1968) notes that men can be fitted to
industrial tasks whose skill requirements are dictated by
technological considerations. Work technologies are
determined by criteria other than the social character
istics of man] cost, for example. The worker, then, may be
visualized as caught up in necessary behavior rather than
voluntary behavior. This behavior may be organizationally
relevant, but neutral in terms of the consequences for the
individual. Necessary behavior may be carried on by a
person who dislikes it; it may be that such behavior is
carried on with indifferance. Dubin considers the
indifferent category to be the most prevalent attitude
among industrial workers. He finds support for this posi
tion in past industrial studies. The point of Dubin’s
(1968:72-76) argument is that attachment may not be a
salient question when faced with a disinterested employee.
Dubin's view of the disinterested worker is in
competition with another version of the condition of the
industrial worker; that is, that he is alienated. One
version of the alienation argument sees the individual as
oppressed by the work organization. Oppression comes about
because of the incompatibility of the mature human
personality and the organization. Argyris (1957) presents
the view that modern western cultures stresses indepen
dence, versatility, flexibility, activism, and self-
control. Individuals expect to be able to express these
values and to be rewarded for behavior consistent with
them. However, organizations are inimical to self-
expression and adequate personality development. This is
because of organizational emphasis on utilitarian goals,
impersonality, specialization, coordination, and
supervision. These priorities act to block and punish
individual expressions of the ideal personality attributes
(Argyris, 1957:1-24). Argyris suggests that the more
mature a person is, the more strain he will experience in
organizations. It may also be suggested that persons who
possess certain occupational characteristics (independence,
for example) may suffer greater strain in some organiza
tions. As will be discussed later, it may be that
professions are occupations which socialize their members
into roles which are a source of strain in organizations.
Argyris1 view of organizational reality is in
clear contrast to the views already mentioned by Dubin
(1968:90-93) and Strauss (1968:93-103). Both of the
latter insist that there is little inherent conflict
between organizational demands and personal expectations.
Specifically, Dubin argues that individual conformity is
required by the nature of the organization and is
ordinarily guaranteed by coercive controls; however,
control is usually not stressful to workers because they
have not come to expect alternate arrangements (Dubin,
1968:73ff). The suggestion is that organizational
attachment cannot be polarized into clear patterns of
positive or negative evaluations. The worker is not led
by his expectations or his attitude towards work in the
organization to make such evaluations.
With these contrasting viewpoints in mind, this
research will examine the linkages between organizational
attachment and m^ny of the individual, occupational and
organizational variables which have been suggested as
important influences on attachment. Organizational
attachment is an important focus because of what it can
tell us about the forces which bind individuals to groups
in general. Stated another way, the ultimate objective of
research such as is undertaken here is to contribute to our
understanding of the social-psychological processes which
determine the relationship of an individual to a group.
For our purposes, these relationships will be conceptual
ized in terms of the following variables: 1) how an
individual defines a group and the degree to which he is
attached to the group; 2) the degree to which an
individual defines himself in terms of his membership in a
group and the degree of satisfaction he receives from group
membership; and 3) the interaction patterns that appear
within a group. All of these social-psychological
dimensions are related to organizational attachment.
8
Review of the Literature
The degree to which an individual is attached to a
group is a central issue in the social psychology of work
organizations. The central role of the work group in
determining organizational effectiveness has been
especially emphasized in the management literature (Likert,
1961:184-217). The literature related to attachment will
be reviewed in this chapter in three sections. The first
two sections may be defined as organizational and the
third section may be defined as laboratory or social-
psychological studies. The first section is concerned with
industrial studies which tend to see the work group and
management as competing for the employee’s attachment, or
to bind him in allegiance. The term "organizational
attachment" is not commonly used in industrial studies, but
other terms used in industrial studies are quite similar in
meaning. The second section, concerned with professionals
in organizations, visualizes the organization as competing
with outside professional groups for the employee's ties
or attachments (for instance, see Gouldner, 1957:281-306).
Organizational attachment is a major dimension used in
analyzing the relationship between professionals and their
employing organizations. Particular attention will be
given to those studies that deal with the distinction
between "locals" and "cosmopolitans." The third section
will be devoted to laboratory studies of attachment in
small groups. These studies permit a more selective
approach to specific group processes, greater ability to
manipulate certain variables, and the use of control groups
to a degree not possible in natural settings.
Industrial Settings
As previously noted, industrial studies tend to see
management and workers as opposing groups, whose interests-
are frequently in conflict. Management represents formal
organization and the worker's work groups represent
informal organization. Formal organization is concerned
with the division of labor, power, control, and rules and
regulations about how one'/s work is to be performed.
Informal organization generally refers to the social rela
tions that develop between vrorkers and is concerned with
work group pressures and relations-that develop between
the leaders and the followers of these work groups (see
Etzioni, 1964:32-49).
The Hawthorne studies were the pioneering research
which stimulated interest in the industrial work group.
The original interest of these studies was the effect
illumination and other physiological conditions had on
worker productivity, but it became evident after the
initial phases of these studies that worker productivity
could only be explained as social-psychological rather
10
than physiological phenomena (Homans, 1951:210-212). For
example, a major finding of the Hawthorne study was the
relationship between high productivity and small group
membership. The investigators noted that the development
of a cohesive group structure developed among the subjects
of the research. Cohesiveness resulted because the members
had participated as a working group. The study notes:
. . . workers began to help one another out
for the common good of the group. They had
become bound together by common sentiments
and feelings of loyalty. (Roethlisberger and
Dickson, 1964:86)
Group cohesiveness or the attachment of an individ-.
ual to a group became a major research interest following
the publication of the Hawthorne study. No general
agreement has been reached upon the use of terminology.
Bass (1960:60-80), for example, notes that cohesiveness,
attractiveness, solidarity, valence, potency, and
involvement are similar ways of describing the same thing.
Cohesiveness has often been operationalized as some form
of interpersonal choice without specification of the basis
on which choices were to be made; that is, the criterion
of choice is not specified (Hare, 1962:17-18, 126-166).
The relationship of group cohesiveness to work
output has also received-a great deal of attention. Coch
and French (1948:512-533) reported that workers in informal
groups lowered their productivity in order to resist
innovation. Roy (1952) reported on quota restriction and
11
wasted time in an industrial setting. The researcher was
a participant observer at the time of the study, and he
indicated that informal group processes are the determining
factor in the level of worker output. Roy made a
distinction between "goldbricking" and "quota restriction"
and indicated that the type of work assignment determined
which type of output restriction is likely to occur. Out
put was defined by the work group as appropriate to
management's incentive policies. These two studies indi
cated that group cohesiveness is detrimental to high
worker productivity.
Gross takes issue with the idea that informal
organization, specifically work group membership, naturally
opposes formal organization. He takes the position that
informal organization may supplement and complement formal
organization. With data collected in a small manufacturing
plant and a United States Air Force radar site, Gross
examined the ways in which the informal organization
helped to resolve problems created by extreme specializa
tion of jobs, formal channels of communication, and formal
personnel assignment procedures (Gross, 1953:368-373).
Seashore attempted to discover the factors which
produce group cohesiveness and the consequences of
cohesiveness for the work group. He found that the work
group is more likely to become cohesive if administrative
actions are designed with these ends: 1) to lend prestige
to group members; 2) to structure the organization so that
there is provision for groups of relatively small size;
and 3) to maintain a continuity in group membership over a
period of time (Seashore, 195^:97-102). Data for this
study were drawn from a population of 228 groups and
questionnaires were completed by 5,871 members of these
groups. Seashore demonstrated that workers in cohesive
groups were neither more nor less productive on the
average than workers in non-cohesive groups; however,
cohesive group workers were more uniform in their pro
ductivity— they all produced approximately the same amount
(Seashore, 195^:97-102).
Sociologists have shown an interest in the problem
of group solidarity and have studied the conditions under
which groups persist and the conditions under which they
fail to persist. Solidarity has been studied in terms of
defections from the group and these studies point out the
connection between the persistence of groups and inter
personal relations. Conformity to group norms has been
found to rest upon personal attachment to the individuals
regarded as representatives of the group (Shibutani, 1961:
581-588).
Closely related to the studies of work group
cohesiveness were the studies concerned with worker morale
or satisfaction. The Hawthorne study was interpreted as
implying that productivity was tied to morale. After
approximately thirty years of research, the results of
numerous studies are confusing. Some studies did show a
positive association between the morale of workers and
their level of productivity. The relationship found in
many studies was weak; there was no relationship at all in
a number of studies, and in some studies there was even a
negative relationship. There is no general agreement
about the meaning'of these studies, but there is general
agreement that the adjustment of individuals to organiza
tions is a complex matter and that other factors must be
explored which affect the relationship between worker
attitudes and productivity (Tannenbaum, 1966:33-38).
In a slightly different vein, Blauner reacted to
the generalization that the typical industrial worker was
alienated and estranged with skepticism. He reports there
is a remarkable consistency in the finding that the vast
majority of workers are moderately or highly satisfied,
rather than dissatisfied with their jobs; this finding
applies in virtually all occupations and industries
(Blauner, 1969:2^7)* The factor Blauner introduces to
understand differences in work satisfaction among workers
is occupational differences. The highest percentages of
satisfied workers, Blauner reports, are found in the
professional and managerial ranks, and the lowest among
unskilled laborers and workers on assembly lines. He lists
several factors which help to account for occupational
14
differences in satisfaction: 1) occupational prestige;
2) control over time, physical movement, pace of work,
technology, social environment, and freedom from direct
supervision; and 3) integrated work groups (Blauner, 1969:.
228-245).
Blauner's findings are extremely interesting to the
study of organizational attachment as visualized in terms
of Caplow's (1964) S.I.V.A. model of organizations. Morale
or satisfaction is seen by Caplow as synonymous with
voluntarism and voluntarism is seen as a negative function
of stability. An increase in voluntary cooperation (morale)
tends to diminish the authority that sustains the whole
system of cooperation. On the other hand, coercion
ordinarily decreases the willingness to be coerced (Caplow,
1964:139)* Blauner's study, then, may be interpreted as
indicating that the more prestige, the more control in
various areas of work, and the more integrated the work
group, the less stability or attachment an individual will
be likely to display.
A worker may be satisfied with his work, but work
may not be a very important area of his life. Dubin
investigated work as a central life interest, the role and
importance of primary social relations on the job, and
some sources of organizational attachments. He found that
only twenty-four per cent of the workers he studied could
be called "job-oriented." Moreover, only ten per cent of
15
these industrial workers perceived their important social
relationships as taking place at work. However, Dubin did
find that sixty-one per cent of the workers chose their
companies as the most meaningful context when their life
experiences in organizations were brought into focus. The
most significant formal organization in the life of a
worker is the employing organization (Dubin, 1956:131-142).
The formal structure of an organization may
contribute to organizational attachment through its rewards
system. Chinoy investigated the automobile worker, whom he
visualized as caught in a dilemma between tradition and
reality. The tradition is one of opportunity (or the
"American Dream") as opposed to reality, which consisted of
very limited opportunities. Of sixty-two men interviewed,
only eight men felt they had a promising future outside the
factory. Within the factory, five men had real hope of
advancement while the remaining forty-six men could see
little room for personal advancement, and therefore
restricted their ambitions. They still accepted the
opportunity tradition and accepted personal blame for their
failure. Self-blame was mitigated by possessing ambitious
hopes for their children. Although they constantly talked
about leaving the shop, they admitted when pressed by the
researcher, that they probably never could do so (Chinoy,
1952:453-459)* Organizational attachment in the case of
the automobile worker comes about by default; there just
16
isn’t anywhere else for him to go.
Again, it should be noted that the conceptualiza
tion of group attachment varies considerably from study to
study. A clear distinction can be found in the literature
between those who define attachment in terms of a positive
affective tie and those who see attachment to a given
situation as a function of restricted alternatives.
Professionals in Organizations
A sense of obligation to, and involvement with,
one’s work is held to be characteristic of a professional.
However, a professional’s relationship to his work is held
to run counter to a bureaucratic career. Perhaps the most
incompatible with the tenents of a bureaucratic organiza
tion is the professional’s occupational identification,
because it carries with it an emotional commitment to one’s
colleagues (Abrahamson, 1967:1-9)• Professionals bring
their own informal organization with them when they accept
employment in bureaucratic organizations. Commitment to
peers is seen as inherent in professional standing.
The problem of the professional in the organization
has most generally dealt with the conflict between pro
fessional and organizational identification. A number of
conceptual schemes have been presented which distinguish
between an organizational and professional orientation.
Reissman (19^9:305-310) identified four types of adaptions
17.
to the organization by civil servants who were specialists
in a state agency. These adaptions were:
1. Functional Bureaucrat— oriented towards a
professional group outside the organization.
2. Job Bureaucrat— oriented entirely toward
the organization.
3. Specialized Bureaucrat— oriented like the
functional bureaucrat but more aware of and more
identified with the bureaucracy.
4. Service Bureaucrat— oriented toward service
to a certain non-professional group and the
bureaucracy offered him the ability to achieve
his personal goals .
Similarly, Wilensky (1956:114, 129, 145, 153)
distinguished four types of staff specialists in trade
unions as follows:
1. The Professional Service type— oriented
toward the profession.
2. The Careerist type— oriented toward the
hierarchy of the union.
3- The Politico— oriented toward political
power and influence for self.
4. The Missionary— oriented toward some
abstract concept of the labor movement.
Kornhauser (1962:267) notes a tendency in past
studies of professionals in bureaucratic organizations to
argue for the differential significance of only one type
of orientation for a given social structure. This
assumption, he notes, ignores two obvious facts: 1) there
are multiple types of orientations in any complex social
system; and 2) there are multiple functions in any complex
social system. The division of labor in an organization
therefore requires a diversity of work orientations.
Kornhauser states that organizations and professions are
effective in achieving their goals only insofar as they
can induce and sustain satisfactory levels of participa
tion. They differ, however, in the kind of participation
and appropriate incentives depends upon the kind of work
to be done and on the nature of the standards to be upheld.
Conflict is generated because of the difference between
organizational and professional incentives. By means of
mutual adaptions, like time off for professional meetings
granted by the organization, and professional compliance
with organizational demands such as coming to work at the
specified time, accomodation is possible and strain may be
minimized (Kornhauser, 1962:274-276).
The Local-Cosmopolitan Construct
The professional's attachment pattern in the
complex organization has generally been described in terms
of a typology known as localism-cosmopolitanism. Using
terms borrowed from Merton, Gouldner (1957:290)
distinguished between locals and cosmopolitans as follows:
Locals: Those high on loyalty to the employing
organization, low on commitment to specialized
role skills, and likely to use an inner (Organ
izational) reference group orientation.
Cosmopolitans: Those low on loyalty to the
employing organization, high on commitment to
19
specialized role skills, and likely to use
an out (professional) reference group
orientation.
It would appear that localism-cosmopolitanism is a
well established distinction in sociology. This judgment
would seem well-founded based upon the frequency with which
the local-cosmopolitan distinction has appeared in the
literature. In fact, the frequent use of the local-
cosmopolitan construct is not related either to empirical
knowledge or to explanatory power about the organizational
processes these terms are supposed to represent.
A major problem inherent in past local-cosmopolitan
research has been the dimensionality of the construct.
According to Merton (1957:393)9 locals and cosmopolitans
were influential community leaders. Locals were defined
in terms of their interest in the local community with
little thought or energy devoted to the larger society.
Cosmopolitans, while finding it necessary to maintain some
interest and a minimum of relations within the community,
were oriented to the larger society and regarded themselves
as integral parts of the larger society (Merton, 1957:393).
G'ouldner (1958) contributed to the conceptual
cloudiness of the local-cosmopolitan construct when he
tested his earlier theoretical statement in a small,
private liberal arts college. He constructed Guttman type
scales to measure loyalty to the employing organization,
commitment to specialized professional skills, and
20
reference group orientation. His findings were that
commitment to professional skills and an orientation to an
outside reference group was associated with low loyalty to
the college. It is the generalization from this finding
that leads Gouldner (1958:466) to conclude that there seems
to be tension between an organization's bureaucratic need
for expertise and its need for loyalty.
Gouldner identified six distinct types of locals
and cosmopolitans in his study (1959 465-477). The first
four of these are locals and the remaining two are
cosmopolitans. Gouldner identified these attachments as
follows:
1. Dedicated— true believers who are identi
fied with and affirm the distinctive ideology of
their organization.
2. The True Bureaucrat— distinguished by
their orientation to the town or local community
in which their organization is located.
3. The Homeguard— characterized by the sub
group to which they belong, particularly females
and second rung administrators.
4. The Elders— oldest people in the group
and the longest employed.
5. The Outsiders— characterized by low
participation in the organization and low personal
influence by choice.
6. The Empire Builders— committed to the sub
groups to which they belong with strong emphasis
on group autonomy.
Glaser (1964) was concerned with the implications
of past research that characterized scientists as either
21
cosmopolitan or local. Scientists are presented as two
distinct types whose orientations and activities were, if
not directly opposed to each other, unrelated. Glaser's
findings were that both research and non-research activi
ties of the organization were seen as compatible to highly
motivated scientists. These scientists are seen by Glaser
as both cosmopolitan and local. They are cosmopolitan in
that they are oriented to achievement of the honorary
rewards in the scientific world and to their professional
colleagues. They are also local in that they are highly
motivated to do an important job in the organization and
have a sense of attachment to the organization. They are
also motivated toward higher level jobs in the organiza
tion; supervision, for instance. The conclusion of this
study is that there is a basic congruence between the
scientists' goals and the organizations' goals which, in
both cases, is to advance knowledge.
The literature reviewed to this point indicates
that distinct polar local and cosmopolitan types have not
been found by previous research. While Gouldner posits
pure types, the test of his theoretical formulation
presents findings that are not incompatible with Glaser's
study. Gouldner's Empire Builders were cosmopolitans whose
orientations were consistent with Glaser's local-
cosmopolitan scientists. Both visualized their departments
as vehicles to further their careers. A possible
22
explanation of this finding is presented by Miller (1967)
who found those differences to be determined by the
organizational unit to which an organizational participant
belongs.
The causes of localism-cosmopolitanism are con
sistently presented as related to formal or professional
training for cosmopolitanism and organizational socializa
tion for localism. Blau and Scott's (1962) study of case
workers in a public assistance agency found that graduate
training was the most significant factor in determining a
person's organizational attachment. They distinguished
four types of reference orientations (1962:64-74):
1. Professionals— workers having graduate
training and choosing outside reference groups.
2. Reference Group only— workers oriented to
outside reference groups but lacking graduate
training.
3. Training only— workers having graduate train
ing but not choosing outside reference groups.
4. Bureaucrats— workers without graduate
training and not oriented to outside reference
groups.
The findings of this study were as follows:
1. Caseworkers with graduate training were
more likely to choose an outside reference group.
2. Professionals were lowest in organizational
attachment.
3. Professionals exhibited professional
characteristics such as attendance at meetings,
as predicted by the study.
23
4. Professionals were more critical of
the agency.
Abrahamson (1967) studied five different industrial
laboratories. His findings support a view of the indus
trial laboratory as presenting a resocialization experience
for academically trained researchers. A cosmopolitan
orientation, a desire to do basic research, and advanced
academic training make organizational socialization more
difficult.
Certain occupational choices may predispose an
organizational participant to either localism or
cosmopolitanism. This may be true even if the choices are
in the professional or semi-professional occupational
categories. Engineers, for instance, have been described
as already high on organizational attachment, even while
still in school (Goldner and Ritti, 1967). Nurses, who
have usually been described as cosmopolitans, have also
been found to be highly attached at the same time to their
employing organization (Bennis, et al.., 1958). Etzioni
(1964:87-89) cites nurses, teachers, and.social workers
as the typical employees of semi-professional organiza
tions. While these- professions tend to be oriented like
full-fledged professions (that is, outside of their
employing organizations), they also tend to become
bureaucratized by organizations and tend to become more
administration conscious. According to Etzioni, this stems
24
from the fact that the administrators of these organiza
tions are most likely nurses, teachers, and social workers
that come up from the ranks.
Career concerns are held to be related to a local
or a cosmopolitan orientation. Glaser (1964:127-136)
suggests that scientists are "great men" oriented during
their professional training. With the kind of model, most
are doomed to failure and, in fact, they do become
"comparative failures." The scientist who recognizes
himself as in the comparative failure category does not
have to give up personal attachment to his chosen pro
fession. Instead, he has to adjust to the organization and
set his sights a little lower; his ambitions may be
"cooled out" and call for a change in reference orienta
tion.
Engineers, on the other hand, are seen as
management oriented. To alleviate feelings of failure on
their part when they fail to get into management, a
professional career ladder is set up for them. They are
professionalized to alleviate the results of their
organizational management orientation. It is instructive
that Goldner and Ritti (1967:489-502) entitle their article
"Professionalization as Career Immobility."
A final study on professional versus organizational
ties further confuses the causes or organizational
attachment. Goldberg, et al. (1965:704-710), reports that
25
organizational and. professional orientations are indepen
dent of each other. In fact, they found that personnel
did not choose between organizational and professional
rewards as has been previously reported in the literature,
but that personnel, in their study, varied in the extent
to which they sought personal gratification in general.
It did not matter if these personal gratifications came
from the organization or from the profession. The authors,
however, question the representativeness of their sample
(two per cent were Ph.D.s in a single case study) and
indicate the need for further empirical investigation.
The consequences of localism and cosmopolitanism
for organizational participants have generally not been
explicated by past research. The reported consequences
have been closely tied with the definitions of the two
types of orientations. The major consequence of
cosmopolitanism, for instance, has been lack of organiza
tional Integration or attachment (Abrahamson, 1967:102-109,
and Gouldner, 1958:476-477). Gouldner (1958:476-477) also
found isolation to be highly associated with the type of
cosmopolitanism he called "Outsider."
Divergent research findings, such as those reported
above, and conceptual problems have contributed to the
recommendation by one recent study (Grimes and Berger,
1970:407-416) that localism-cosmopolitanism should not be
used as an explanatory or descriptive variable. This
study proposes that research is needed on the construct
itself before it is utilized in further research. The
major conceptual'problem cited is the complex and confused
dimensionality of "the construct. Another recent study
(Thielbar, 1970:243-25*0 which proposed to assess the
prospects for a theory of localism-cosmopolitanism
concluded that localism-cosmopolitanism might still be
helpful in conceptualizing a theory of social participa
tion. But to date, no theory which satisfactorily
encompasses the local-cosmopolitan construct has been
forthcoming.
Despite the fact that the most extensive research
efforts in the broad area of organizational attachment have
been concerned with localism-cosmopolitanism, numerous
methodological problems have characterized these past
research efforts. Measurement and sampling problems are
two examples. Researchers have continued to operationalize
the construct in different ways, which has made the
comparability of results almost impossible to assess. Poor
sampling techniques have characterized local-cosmopolitan
research in general (Grimes and Berger, 1970:408-410).
Most research involves case studies (this includes Blau and
Scott, 1962; Kornhauser, 1962; Goldberg, et al., 1965; and
Abrahamson, 1967). Where they have not been confined to
single organizations, studies have been confined to
contrasting limited numbers of occupations. For example,
27
engineers and research scientists have often been compared
(see Marcson, 1961; Kornhauser, 1962; Glaser, 1964;
Abrahamson, 1967:102-109; and Miller, 1967:755-767).
Laboratory Studies
The advantages of the use of laboratory versus
natural material group settings in the study of small
groups has been the subject of a great deal of discussion.
Lewin and Lippitt provide a rationale for the use of
laboratory settings:
One has to create a set-up where group life
might be studied under rather free but well
defined conditions. Instead of utilizing the
groups in schools, clubs, and factories, one
should create groups experimentally because only
in this way the factors influencing group life
will not be left to chance but will be in the
hands of the experimenter. (Lewin and Lippitt,
1966:648)
While investigators in the laboratory tradition may
have created their own groups, their research interests and
the important variables in their studies are quite similar
to those found in the organizational literature. Cohesion,
for instance, has been a major research interest of
laboratory studies of small groups. The review of the
laboratory studies of small groups will be restricted to
studies of cohesion and especially cohesion as it is
related to the major independent variables suggested in the
review of the organizational literature.
The property of "groupness" has been a major
28
emphasis of laboratory studies, especially that property
defined as cohesiveness. Cohesion has been defined in some
studies in terms of interpersonal attraction and then
operationalized as the ratio of in-group to out-group
choices (Collins and Raven, 1968). A common sense approach
to organizational attachment (or the probability of one
staying or leaving) indicates that attachment should be
related to acceptance or rejection in a person’s inter
personal relationships at the group or organizational
level.
Back (1951) found greater communication and con
vergence in attitude to be a function of increased
cohesiveness. This was true whether cohesiveness was
based on: 1) personal attraction; 2) importance of the
task; or 3) prestige through membership. Differences were
reported in communication and social influence which were
accounted for by determining the basis for cohesiveness.
A widely reported variable in the study of group
cohesiveness is the tendency of individual group members
to influence and be influenced. Schacter, et al., (1951)>
informed subjects about the presumed congeniality of their
groups. Subjects were supposedly performing three
different tasks, and importance of quality and speed was
stressed. After the subjects were assigned to different
workrooms and given the impression that other members of
the group were working on other assignments, communication
between group members was by written notes delivered by
messengers. Notes were intercepted and a standard pre
written set was substituted to induce either high or low
production. The findings of this study were that cohesion
and influence interacted as determinants of production.
The data demonstrated a clear-cut and positive relationship
between experimentally induced cohesion and member ratings
of the personal attraction of their particular group
(Schachter, et al., 1951:229-238). Similar results are
reported by F.estinger, et al., (1952), and Berkowitz
(1954).
Organizational attachment may be a function of
group opinion, as was suggested earlier and the decision to
stay or to leave may be a function of one’s interpersonal
relations in an organization. Gerard (1954) reported that
subjects show greater resistance to changing an opinion if
it is an opinion expressed by a highly cohesive group of
which the subject was a member. Festinger (1950) has
suggested that group members direct their communication
toward a person holding deviant opinions. When a deviant
does not shift his opinion to the satisfaction of the
group, communication towards him falls off and he is
rejected by the group. Interpersonal influence in cohesive
groups is also found to be related to imitation. Grosser,
et al. (1951)» used an experimental group to discover that
a person was more likely to be imitated if he behaved in a
30
friendly fashion toward a coworker than when his behavior
was unfriendly. Collins and Raven (1969:171ff) define
these interpersonal influences due to group cohesiveness as
referent power and define such power as "based on persons
identification with another or a person's desire for such
identification."
Cohesiveness has also been found to be strongly
related to satisfaction or morale. Morale is a term (as is
cohesion) which has been used conceptually to combine
several classes of phenomenon. First, morale, as in
industrial studies, has been studied as it relates to group
productivity. Secondly, morale research has been centered
on the individual while evidence has suggested that the
important locus of morale has been in small social
groupings (Golembiewski, 1962:237-241). Elsewhere,
Golembiewski (1964) discusses the distinction between
morale and satisfaction. He distinguishes between output
and satisfaction because earlier studies tended to
associate these variables directly. Morale and satisfac
tion are distinguished with the recommendation that morale
be discarded because of lack of clarity as to what the
concept means (Golembiewski, 1964:103).
Hare (1962) indicates that morale and cohesion have
generally been used interchangeably. When a distinction
is made, cohesiveness is used to represent desire to
belong to a group whereas morale includes an emphasis on a
disposition to act toward a goal.
Communication patterns have been found to be highly
related to satisfaction. Leavitt (1951) suggested that the
relative independence of action of the person occupying a
control position, in constrast to the dependence of persons
at the periphery leads other members to perceive them as
the leaders; they are led to perceive themselves as leaders,
and this in turn explains why such persons have a higher
satisfaction score in experimental groups. Support for the
perception of one's self as leader and high satisfaction
scores have been given by studies conducted by Trow (1957)
and Shaw (1954).
Summary
The state of our knowledge about organizational
attachment is, at best, unclear. This is true whether we
are referring to what have been called the industrial, the
organizational, or the laboratory studies. Part of the
cloudiness that appears in these studies is a result of the
way terms are used interchangeably. Attachment, loyalty,
cohesion and solidarity are all terms used synonymously.
Some studies talk about morale, others talk about
productivity, still others talk about the probability of
staying or leaving, and yet others talk about the
probability of conflict. Perhaps the common thread that
runs through these studies is, in general, a reflection of
32
the individual's willingness to remain in the organization.
The problem is to determine how the dimensions of this
willingness to remain are related to each other. In
short, how can we conceptualize the suggestions of the
review of the literature?
Caplow's (1964) S.I.V.A. model is cited as a useful
way of conceptualizing the variables that relate to
organizational attachment. In fact, this model served as
an outline for the review of the literature. Caplow
presents the following chart which reveals the way the
variables cited are related to each other.
Figure I identifies why organizations may be seen
as "problem ridden""social systems. The major underlying
problems may be seen as follows:
1. Stability and integration are negative
functions of each other, while both are positive functions
of achievement.
2. Stability and voluntarism are positively
correlated with achievement but the relationship between
stability and voluntarism is unclear because few organiza
tions depend upon voluntarism for their continuity.
3. Integration and voluntarism are positively
related to each other.
The review of the literature indicates the
following relationships:
Figure I
Paired Relationships Among S.I.V.A. Variables
Resulting Tendency*
Direction of Change Stability Integration Voluntarism Achievement-
Stability plus
Stability minus
Integration plus
Integration minus
Voluntarism plus
Voluntarism minus
minus
plus
indeterm,
indeterm,
minus
plus
plus
minus
indeterm.
indeterm.
plus
minus
plus
minus
plus
minus
plus
minus
Achievement plus
Achievement minus
plus
minus
plus
minus
plus
minus
*In one variable at a time, holding the others constant. (Caplow, 1964:122)
U)
< jO
34
1. Organizational attachment is related to
orientation to peers versus orientation to management for
industrial workers, or orientation to colleagues versus
orientation to organization for professional employees.
2. Individual organizational attachment patterns
or individual attitudes are seen as related to group
factors such as the level of cohesion, solidarity,
satisfaction, interest, alienation, etcetera.
3. Organizational attachment is dependent upon a
number of social-psychological variables and such organi
zational phenomenon as morale or satisfaction are
interpretable in terms of these variables.
The review of the literature suggests a number of
social-psychological variables that seem to be related to
the dimensions comprising organizational attachment. It is
argued here that these variables are the key to understand
ing organizational attachment. While schemes for classi
fying variables such as presented by Caplow (1964) are
helpful, in the final analysis they do not assist in
clarifying the basic problem of individual adaption to an
organization. For instance, why might a professional
suddenly become concerned with stability as opposed to
integration? That is, why might a professional suddenly
become concerned with tenure versus professional colleague
relationships? This type of question is seen as
interpretable in terms of some of the social-psychlogocal
variables suggested by the review of the literature:
1) occupation— with special reference to training required
to enter an occupation (Blau and Scott, Abrahamson,
Glaser); 2) role— especially role requirements connected
with occupational categories (Gouldner, Glaser, Bennis,
et al., Goldner and Ritti); 3) expectancies— particularly
occupational expectancies (Coch and French, Roy,
Abrahamson, Goldner and Ritti, Glaser); 4.) attitudes— with
special attention to those attitudes that may be seen as
acquired through interpersonal relations, either from the
work group or colleague relationships (Homans, Roy,
Roethlingsberger and Dickson, Coch and French, Schachter,
Gerard, Abrahamson); and 5) rewards and inducements—
contrasting the sources of rewards and inducements as
between work groups or colleagues and organizational
sources (Gouldner,'Rbethllngsberger and Dickson,
Kornhauser, Goldner and Ritti).
The review of the literature also reveals a general
problem in all three areas of research, industrial,
organizational, and laboratory. This entire group of
studies suffers from a lack of theory. McGrath and
Altman's (1966) comment that smaTl group studies have
produced too much data and too little theory is applicable
to the review of the literature offered here. In fact,
definitions of terms and the absence of theory together
form the major problems with studies of organizational
36
attachment.
The Present Study
After summarizing the literature, the broad
question of how individuals in different occupations and
different organizational circumstances vary in terms of the
type and amount of organizational attachment they express
remains unanswered. It is not clear whether there are
systematic differences between those who are attached to a
professional reference group or an industrial work group,
and those who are not attached to any group. It is not
clear whether or not there is a conflict between organiza
tional attachment and attachment to a professional refer
ence group, or an organizational work group. All of these,
questions take their significance from the general question
of the determinants of individual attachments to groups and
organizations. The question is stated more broadly here
than has been done in past research, especially the
local-cosmopolitan research. The attempt is to determine
some of the factors that have more general organizational
significance in determining organizational attachment.
The present research reports on the questionnaire
responses of the members of four complex organizations.
The first of these organizations Is the research unit of
a large national food processing and marketing firm
(n = 92). This research component is directed toward
37
increasing the productivity and profitability of the parent
organization, despite the fact that a large number of
highly trained scientific personnel and technical
assistants are employed there. These technical people,
then, are not engaged in developing "pure" knowledge or in
providing services to the community. Personal research
interests and theoretical and other professional questions
are likely to be subordinated to the commercial interests
of the parent organization.
The second organization is a social research
organization which is attached to a major private uni
versity (n = 47). It is staffed primarily by professionals
trained in the social sciences.
The third organization is also a social research
organization which is administratively autonomous (n = 62).
This organization is seen as less concerned with purely
professional and scientific objectives when compared to
the other social research organization. Clearly, the
second and third organizations are highly professionalized
and approach the abstract model described by Etzioni (1964:
77ff).
The fourth organization may be defined as a semi-
professional, public-service organization. This organi
zation is the alcoholic rehabilitation program of a large
California county (n = 59). Pour clinics constitute this
organization, which is staffed by medical and psychiatric
practitioners, rehabilitation counselors, and social
welfare personnel.
With a sample size of two hundred and sixty
respondents from four complex organizations, hypotheses
about organizational attachment will be derived from
existing theoretical frameworks and from the review of the
literature. This will be the task of the following
chapter.
CHAPTER II
THEORETICAL CONSIDERATIONS AND DERIVATIONS
OP HYPOTHESES
The review of the literature presented in the first
chapter indicates that there are wide gaps in our
knowledge about the factors which contribute to organiza
tional attachment. These gaps are apparent whether the
discussion is about the causes or about the consequences
of organizational attachment. It is argued here that the
first step toward unraveling this conceptual confusion is
the consistent application of existing theoretical
framework. A clear indication of this need is presented
by the fact that theory has been largely absent in studies
which are relevant to organizational attachment. This
is reflected in the local-cosmopolitan studies as well as
in the laboratory and industrial studies. This chapter
explores the possibility of applying several existing
theoretical frameworks to the area of organizational
attachment. This process will be primarily inductive and
is suggested by the review of the literature.
Two theoretical frameworks which are suggested by
the review of the literature and which will be explored
39
here are reference group theory and exchange theory.
Taken together, these two approaches subsume the major
variables which the literature has suggested are important
for the analysis of group attachment. Reference group
theory is suggested by the local-cosmopolitan studies; in
fact, the reference group orientation is an integral part
of the definition of this distinction. Exchange theory,
especially as defined by Homans (1961), is suggested by the
alternate sources of reward available for organizational
participants— namely, rewards from work group associations
as opposed to rewards available from management or
interpersonal rewards bestowed by the organization. While
the major theoretical framework explored here is reference
group theory, exchange theory will also be utilized where
it may be integrated with reference group theory, or where
exchange theory supports alternative hypotheses or addi
tional variables which will further clarify the nature of
organizational attachment.
While this chapter is an attempt to formulate
hypotheses which link the independent variables identified
in the review of the literature to the dependent variable,
organizational attachment, another important consideration
is the relationship among these independent variables.
This research will also attempt to clarify some of these
relationships.
41
Reference Group Theory
l\&.ny of the influences on attachment which were
suggested by the review of the literature are important
variables for reference group theory. This framework is
particularly relevant for the analysis of peer group versus
organizational orientations. The idea that individuals
evaluate their actions from the perspective of a
particular reference group in which they are members is as
old as sociology. This notion is brought into somewhat
sharper focus by reference group theory. Individuals are
seen as taking on the attitudes, evaluations, and behavior
of groups to which they belong, or to which they aspire.
Kemper (1969:298) defines a reference group as:
A group, collectively, or person which the
actor takes into account in some manner in the
course of selecting a behavior from among a
set of alternatives, or in making a judgment
about a problematic issue. A reference group
helps to orient the actor in a certain course,
whether of action or attitude.
It is important to note that a reference group may
be an actual group, a person, a personification of an
abstraction, or a collectivity.
The concept, reference group, was first used by
Hyman (1942) and subsequently developed by Sherif (1948),
Newcomb (1952), Merton (1957), and Shibutani (1961).
While it has been suggested that reference group theory is
almost devoid of any "real theory," and that it is so
42
limited that it no longer is of much interest to
researchers (for instance, see Golembiewski, 1962:17)s the
review of the literature suggests that it might be well
adapted to the state of our knowledge of how organizational
attachment is related to the attitudes of the groups to
which an individual belongs or aspires to belong.
The question of most interest concerns the specific
functions that reference groups perform; that is, what do
they do for the individual? Kelley (1968) has described
two functions of reference groups: 1) the normative
function, and 2) the comparative function. On the one
hand, reference groups enforce or set standards of
behavior for the person, and on the other hand, reference
groups serve as comparison points against which a person
can evaluate himself and others. In other words, one can
learn appropriate behavior from reference group standards
and compare behavior from his reference group standards.
Occupations and Professions as Reference Groups
Occupational differences are very important in
understanding organizational attachment. The review of the
literature may be seen as contrasting two distinct
occupational categories, manual and professional workers.
Both of these occupational categories have been visualized
as having reference group orientations which may poten
tially undermine organizational attachment. For example,
manual workers are commonly pictured as indifferent or
alienated from their work and from their employing
organization. They may be seen as having reference groups
(their own work groups) which serve both of the reference
group functions. Workers are not seen as reacting to the
work situation as individuals, but as members of groups.
To the extent that this is true, individuals are not free
to set up production quotas by themselves; rather, these
are set and enforced by the groups which evolve in the work
place (see Etzioni, 1964:35). These groups may be seen as
normative reference groups because they provide the worker
with a guide to action by explicitly setting norms and
reinforcing common values. The worker is expected to
comply with these norms and values and deviation is
severely punished.
Professions may also be seen as normative
reference groups. Professions have been conceptualized as
occupations with distinctive core characteristics
including: 1) autonomy in work performance, 2) a service
orientation, and 3) command over a body of scientific
knowledge (Goode, 1961:902-914; Reiss, 1961:10-18; and
Hughes, 1963:655-658). The literature in what may be
called the ’ 'classic" professional tradition went so far as
to describe the professional person as an independent
entrepreneur who engaged in his profession because of a
"calling." He might have apprentices to assist him, or law
44
clerks, junior partners, or technicians, but as a profes
sional, he was his own master within the ethical and
associational framework of a society of colleagues (Vollmer
and Mills, 1966).
This picture of the professional has been
recognized as no longer accurate. However, the model is
still held to be an ideal type for the professional to
emulate, even though in reality, more and more profession
als are finding employment in complex organizations. In
this sense the notion of professionalism is a key element
in the normative and comparative functions which profes
sions serve for their members.
Blau and Scott (1967:174-177) may overstate the
case when they argue that bureaucratic discipline and
professional expertise represent a basic "dilemma” of
modern organization, but their description of organiza
tional reality clearly recognizes the reference group
nature of occupational commitments (also, see Parsons,
1947:58-60; and Gouldner, 1954:22-24). This can be seen
in their treatment of the sources of the dilemma which they
identify as: 1) difference in organizing principles:
organizational vested interests versus service and
professional ethics; 2) source of authority: formal
organizational authority versus technical expertise;
3) decisions of supervisors: compliance with supervisors
versus internalized professional standards; and 4) final
45
authority: prerogative of management versus the profes
sional colleague group.
Etzioni (1964:75-93) describes the professional-
organizational dilemma as basically a conflict between
administrative authority and professional authority. In
his terms, organizations range from fully bureaucratic to
fully professionalized; that is, from bureaucracies
administered by administrators and fully professional
organizations administered by professionals. The amount
of conflict between organizational and reference group
priorities therefore varies with the type of organization.
It is of interest to note that conflict is at a minimum
when professionals find themselves administered by other
professionals. Prom the reference group perspective,
conflict is minimized when there is a basic congruence
between colleague and administrative groups; that is, when
they are identical. This discussion is consistent with
Glaser's (1964) description of the local-cosmopolitan
scientist. In an industrial laboratory administered by
scientists, he found that a researcher's professional group
is largely coextensive with the administrative group.
The important part is not whether the contact
between occupations and organizations can be conceptualized
as a "dilemma" or a "conflict" in a given situation; rather,
the important question is whether the relationship can be
analyzed successfully in terms of the social-psychological
46
variables which were identified by the review of the
organizational studies. Whether professionals react to
these variables in a fundamentally different way from
non-professionals is also an open question, and one which
will receive some attention in this analysis.
One direction this exploration might take it
suggested by the localism-cosmopolitanism studies. They
suggest that one way to conceptualize an individual's
potential choice of reference groups is to concentrate on
the occupational socialization process. Socialization in
the most general sense is concerned with how individuals
learn to participate effectively in social interaction.
Brim (1966:3) defined socialization as:
The process by which individuals acquire
the knowledge, skills, and dispositions that,
enable them to participate as more or less
effective members of groups and society.
Adult socialization is a type of socialization that is
concerned with the learning of organizational processes,
career sequences, and occupational preparation. Career
sequences begin with formal education and training and
carry over into the work situation, usually an organiza
tional situation. Two types of adult socialization
processes are identified in this process: those connected
with formal or professional training, and those connected
with organizational socialization. Professional or formal
occupational training is more likely to take place in the
beginning of an occupational career and is intended to
socialize the individual into the particular ideology of
the occupation. Organizational socialization ordinarily
takes place later in the career, especially for profession
als, and is instrumental in changing the person’s
attachment from an outside orientation to an organizational
orientation.
Reference group theory, then, is concerned with how
membership in or aspiration to certain groups affects the
attitudes and behavior of the individual member or the
aspirant. Taking on the behavior or values of a certain
group may be seen as a form of socialization. Occupational
socialization is one mode of adult socialization which is
relevant to the professional— organizational dilemma.
Reference group theory is able to include a number of the
concerns indicated by the review of the literature. These
have to do with the rite of socialization and in-group
versus out-group orientation; the extent to which reference
group provides a comparison for self and for others; the
importance of group consensus as salient for the
individual's attitude and behavior; and finally, the
degree to which social-psycological variables affect
reference group orientation or selection.
One objective of this research is to explore the
possibility of formulating a model for future research in
the area of organizational attachment. As Wilier (1967:
15-22) points out, in the most simple form, a model
contains a rationale and nominally defined concepts which
are structured in such a way that they form a mechanism. A
rationale explains the nature of the included phenomena and
the definitions of concepts included in this rationale
determines the relationship of these concepts to each
other. This relationship of concepts form the structure
called the mechanism.
Using Wilier's terminology, reference group theory
provides a "rationale" which has an ambiguous "mechanism."
The conceptualization of behavior indicated by the
reference group orientation may be seen as a rationale,
but the mechanism suggested by this formulation is unclear.
This is consistent with Merton's (1957) major criticisms
of reference group theory. Reference group theory as it
now stands does not explain why certain reference groups
are chosen and how or why the individual comes to learn or
take on the orientation of a particular reference group.
It is argued here that reference group theory may benefit
by an attempt to integrate findings which are derived from
exchange theory, for exchange theory might provide the
mechanism necessary for an explanation of why certain
reference groups are chosen as opposed to others and how
or why individuals take on the orientations of these
reference groups.
49
Exchange Theory as a Supplement to Reference
Group Theory
Exchange theory assumes that individuals enter into
and continue interpersonal relationships because they are
somehow rewarding. Moreover, Homans (1961) assumes that
only people can reward people. A similar version of
exchange theory is presented by Thibaut and Kelley (1959)
based on face-to-face behavior in small groups. Blau, on
the other hand, formulates exchange theory more broadly
and allows for indirect transactions between members of a
society or exchanges between individuals and groups.
Attention is given to the structural aspects of exchange
and how exchange processes become institutionalized.
Homan's (1961:51-78) relies on concepts borrowed
from behavioral psychology and elementary economics. Prom
Skinner, the notion is borrowed that men engage in
behavior that is rewarded or "reinforced"; they become
conditioned to engage in that behavior at appropriate
times. Following economics, behavior may be visualized as
a function of its "payoff" in terms of the reward or
punishment it fetches. Behavior becomes an exchange
activity between two or more individuals that is more or
less rewarding or costly. Rewards are determined by units
of value and costs by some form of punishment, judged
against some standard of "distributive justice." Rewards
and investments are not always material, but may involve
50
psychic profits or sentiments such as approval. Homans
(1961:51-78) inductively built his theory from experimental
and field research, especially small group research. For
instance, cohesiveness is referred to as the values of the
different kinds of rewards available to members of a group.
The more valuable to the members of a group the activities
(or sentiments) they receive from each other or from the
environment, the more cohesive the group will be.
Exchange theory.is relevant for the analysis of
organizational attachment because organizations present a
number of possible rewards to participants. These may be
classified as material and non-material; non-material
rewards include authority, prestige, and friendship, among
others. Attachment, in this view, is a function of the
ties'which emerge in exchanges between interacting
individuals and between individuals and the larger
organization. In another light, individuals may be seen
as having investments in organizations over time, and
attachment may be a function of experiencing "distributive
justice" in the social exchanges of investments which take
place in the organization.
Among the more problematic features of exchange
theory, it should be noted that the same behavior,
conformity, may be rewarding and unrewarding. This is a
major criticism of exchange theory pointed out by
Abrahamson (1971)- Note also that the same behaviors
which produce attachment in some contexts may be seen as
leading to a lack of attachment in others. Blau (1964:
143-167) attempts to clarify these problems from an
expectations point of view. He notes the reference group
nature of expectations and provides an example of why
certain behavior may be rewarding in some instances and
unrewarding in others. Long range investments, like
professional education, are a "sunk cost." The only way
to safeguard investments such as education is by engaging
in competition. Competition requires mobility and at the
same time requires commitment or attachment. Organiza
tional attachment precludes the mobility necessary for
individuals to safeguard investments and to receive a fair
return for the major investments of their lives. Despite
such attempts at clarification, the analyst who wishes to
work with exchange theory must proceed cautiously.
Probably the best strategy is to begin by exploring some
of the simpler notions of exchange theory before more
complex theoretical analyses are undertaken.
Although both reference group theory and exchange
theory are subject to criticism, it is argued here that a
synthesis of the two will alleviate some of these short
comings. Specifically, whereas reference group theory
52
suffers from an inability to explain why certain reference
groups are chosen over others and how an individual takes
on the values or attitudes of reference groups to which he
belongs or aspires, exchange theory does provide at least
some useful clues as to why certain groups are chosen over
others in terms of rewards, investments, social approval,
and the like. Based on the ideas which link these two
theoretical approaches together, a series of meaningful
hypotheses can be derived.
Derivation of Hypotheses
The review of the literature has suggested a
number of variables that seem to be related to organiza
tional attachment. They include occupation, role,
expectancies, attitudes, and rewards and inducements. An
attempt is made here to link these variables to exchange
theory and reference group theory and to derive a series
of useful hypotheses.
The first hypothesis deals with professional or
formal training, especially the length of training. The
local-cosmopolitan studies, particularly, identified
training as an important determinant of organizational
attachment. Training involves the socialization of an
individual into an occupational community which may be
seen as a reference group. Members of an occupation or
profession may be seen as taking on the values and
53
attitudes of an occupation through training and therefore,
taking on an occupational reference group orientation.
(In exchange terminology, confirmation of this hypothesis
would indicate that training is an investment and the
willingness to continue lengthy training reflects the
belief that this will be rewarding in the sense that it
will maximize the interpersonal returns for the
individual.) This suggests the following hypothesis:
The greater the professional or formal
training, the greater the probability
an individual will assume an outside
(professional) reference group orien
tation.
An outside reference group orientation was
identified by the local-cosmopolitan studies, specifically
Gouldner's (1957;1958), as an orientation toward a group
which is outside the organization, and which, therefore,
detracts from organizational attachment. However, the
organizational literature has relied on an in-group/out-
group approach which has left the issue of organizational
attachment at best, unclear. The question of the effect
of group experiences in an organization remains unanswered.
What do group experiences have to do with attachment to
organizations in general? Reference group theory allows
predictions to be made about the nature of group
experiences as they are related to attachment. While an
outside reference group orientation may be associated with
low attachment, the distinction between "bureaucratic” and
"local-cosmopolitan" orientations is crucial to the utility
of reference group theory in predicting organizational
attachment. Finer distinctions are necessary and if they
cannot be made, it may be that occupational categories
alone may be as predictive of organizational atachment as
is reference group orientation. The effect of occupational
differences must be controlled if the utility of reference
group theory is to be tested, because occupational dif
ferences overlap with differences on other variables,
especially training. Theoretically, a straightforward
relationship is suggested, with individuals possessing a
strictly bureaucratic orientation more highly attached to
an organization than those who have been identified as
"local-cosmopolitan." A strictly outside reference group
orientation should identify an individual as possessing a
still lower degree of organizational attachment. This
discussion suggests the following hypothesis:
H2 With occupation constant, the greater
the probability than an individual
possesses a cosmopolitan (out-group)
orientation, the less the organizational
attachment he will display.
An orientation toward a professional reference
group is seen as partly a consequence of professional
training. Similarly, continuing employment in an
organization may be visualized as an extended socialization
process and in part as a re-socialization process into the
organization's bureaucratic ways. The person who has
recently completed his training should still possess a
stronger professional (outside) reference group orientation
than someone who has been employed in an organization for a
longer period of time. This is consistent with Homan's
(1961) idea of positive reinforcement and the extinction
process that takes place when one is removed from a source
of reward; that is, training. This suggests the following
hypothesis:
With formal or professional training
held constant, the greater the length
of organizational socialization
(employment), the greater the organi
zational attachment an individual will
display.
Exchange theory allows predictions about why
certain organizational orientations are taken by an
individual, based upon whether or not they are rewarding.
Most important, one can assume, based on exchange theory,
that the greater the reward a person receives in an
organization, the greater the organizational attachment
that person will display. But Glaser (1964) and Blau
(1964) also posit organizational attachment as a result of
56
comparative failure based on comparisons with groups
outside the organization. Which of these different ideas
is most realistic can be explored by examining the
following hypothesis:
Hjj With reference group orientation held
constant, the greater the interpersonal
rewards a person receives in an
organization, the greater the organi
zational attachment he will display.
If this hypothesis is confirmed, it will be a
strong indication that attachment is a positive tie to the
organization, and not the negative result of a restriction
of alternative outside associations.
Exploration of Other Variables
The review of the literature suggested a number of
other variables that appear to be related to organizational
attachment. This list of variables includes: 1) job
satisfaction; 2) morale; 3) job tension; and 4) job
freedom. Each of these variables can be related to one or
both of the theoretical approaches discussed here. How
ever, these variables are especially unclear conceptually
because there is a lack of understanding as to the kinds
of behavior to which they refer. It is also unclear how
they are related to each other and to organizational
attachment; in some cases inconsistent findings have been
57
reported. This is especially true for the relationship
between morale, or satisfaction, and attachment.
Because of this lack of clarity in the ways these
variables have been treated in the organizational
literature, they will be examined in a very exploratory
fashion here. The relationships among them will be
assessed and the relationship of each variable to organi
zational attachment will be explored. It is hoped that this
procedure will help clarify the processes of group
attachment by suggesting how these variables may be
treated in future research.
Summary
This chapter attempted to tie the major independent
variables suggested by the review of the literature to the
dependent variable, organizational attachment. The review
of the literature suggested several theoretical frameworks
as relevant to the problem of linking the independent
variables to organizational attachment. Two were cftosen as
most relevant; namely, reference group theory and exchange
theory. Reference group theory was seen as a framework
highly related to the problem of organizational attachment,
but the theory was noted as having some serious short
comings. The addition of exchange theory was posited as
one way of alleviating some of the inherent deficiencies
in reference group theory and hypotheses were formulated
58
which might have been derived from both frameworks. The
problem of the relationship of the independent variables
to organizational attachment, and to each other, was noted
and clarification of these relationships stated as a goal
of the analysis.
In the following chapter, the methods of analysis
will be presented. The sample, the subjects of the
research, the questionnaire, and the operational measures
of the variables will be presented.
CHAPTER III
THE SAMPLE AND METHOD OP ANALYSIS
The.data presented in this dissertation were
collected from the members of four organizations described
in Chapter I. Questionnaires were administered to all of
the members of these four organizations. The response rate
was between eighty-five and ninety-five per cent for the
four organizations, with the overall response rate slightly
above ninety per cent for the entire sample.
The Sample
The organizations described in Chapter I are highly
professionalized. Each employs primarily college trained
personnel and each is involved in the creation or
application of professional knowledge. The research unit
of the national food processing firm (n = 92) is staffed
by chemists, nutritionists, and related technical staff.
This organization differs from the other three organiza
tions whose personnel are primarily from social science or
clinical backgrounds. This organization may be considered
the least professional of the four organizations presented
in this research. This is because of the commercial
59
60
nature of the encompassing organization. However, the same
may not be true of the research unit personnel surveyed for
this study, who are better paid, enjoy greater prestige
than, for example, the members of the alcoholism program,
and probably experience less frustration than the
alcoholism personnel in the course of their employment.
The social science research organization, attached
to the private university (n = 47), is the most profes
sionalized of all the organizations included in this
sample. The social science professionals employed within
this administrative unit are engaged in a variety of
research projects and this organization most closely
resembles the full-fledged professional organization, as
described by Etzioni (1964:81-87).
The autonomous social' research organization
(n = 62) has a corporate board of directors and is a
private, non-profit organization. While many of the
professionals employed by this organization have the ties
to academic institutions in the surrounding city, the
organization is autonomous in that it does not have any
direct administrative linkages to any other organization.
This research organization is held to be slightly less
professionalized than the first research organization in
that the personnel employed are engaged in developing
practical and commercial uses for social science informa
tion. This is not to say that independent research is not
61
undertaken, but not to the extent as In the first research
organization.
The final organization in this sample, the
alcoholism rehabilitation program (n = 59) is a system of
clinics which share an administrative staff and admini
strative apparatus.. Personnel are divided into five
treatment teams which form a four clinic system. This
organization is basically autonomous, although included
under the jurisdiction of a large county department of
public health.
The sample consists of two hundred and sixty
employees of these four organizations. While there are
some differences between these organizations, they are not
consistent enough to rank one organization as more profes
sional than another with complete confidence. Consequently,
the ranking implied in the above discussion should be
accepted with considerable caution. (Demographic profiles
of the four organizations appear in the appendix.)
The Questionnaire
The questionnaire which was used included items
measuring attachment, reference group orientation, length
of employment, education, job tension, job satisfaction,
job freedom, and job clarity. A number of complex socio
metric questions were also included in this instrument.
62
Measurement of Attachment
The measure of the dependent variable is derived
from scalogram analysis, which may be described as a
procedure for evaluating sets of statements to determine
whether or not they meet the criteria necessary for them
to be considered a unidimensional scale. Specifically,
the Guttman scale is used (see Edwards, 1957:172-200).
Guttman scaling is based upon the assumption that a set of
items measure a single dimension. The degree to which a
group of items is both unidimensional and cumulative is
determined by the extent to which the most difficult item
to be passed is associated with scores on items which have
been ranked less difficult. The scalability of items may
be determined by statistical procedures. The Coefficient
of Reproducibility specifies the extent to which a
respondent's scale score is a predictor is his response
pattern. The Minimum Marginal Reproducibility is the
minimum reproducibility that could have occurred for the
scale and the proportion of respondents passing and failing
each item. The Coefficient of Scalability is obtained by
dividing the difference between the first two measures by
the difference between 1 and the Minimum Marginal
Reproducibility. Recommendations are that a Coefficient
of Reproducibility higher than .90, a Minimum Marginal
Reproducibility of at least .85, and a Coefficient of
63
Scalability of .50 are evidence for a valid scale (see
Edwards, 1957; and Nie, et al., 1970).
The scale of organizational attachment is derived
from responses of the following questions:
How do you like working for this company?
It's not a very good place to work.
It's all right, but there are many things
that should be changed.
It's a fairly good place, but quite a few
things should be changed.
It's a good place, but there are a few
things that should be changed.
It's a very good place— wouldn't change
anything.
Would you advise a friend to come and work for
this company?
I would not advise a friend to come and
work for this company.
I would advise a friend to come and work
for this company.
If you had a chance to do the same kind of work
for the same pay, but in another company, would you stay
here?
I would prefer to go to the other company.
I would stay in this company.
64
How do you feel about the progress you have made
in this company?
I have made little or no progress.
I have made some progress, but it should
have been much better.
I have made quite a lot of progress, but
it should have been better.
I have made a great deal of progress.
These questions produced a Guttman scale with the
following characteristics: Coefficient of Reproduci
bility = .96; Minimum Marginal Reproducility = .88;
Coefficient of Scalability = .71*
Reference Group Orientation
The following structured questions provide the
measure of reference group orientation:
1. Who in general is the best judge of your
abilities, competence, and achievements?
(Please check only one.)
Responses provided for this question were:
1. Members of your occupation or profession
2. Your immediate supervisor
3. Supervisor above your supervisor.
2. In general, where do you get most of your
intellectual stimulation?
Responses provided were :
1. Members of your occupation
2. Other people in the organization
3. Members of your occupation in the
organization or outside of the
organization
4. Periodicals, books, and/or other
publications.
An outside-professional reference group orientation
(called "cosmopolitan") will be interpreted as a choice of
profession as best judge of abilities and an outside-
occupational or literature reference as the source of
intellectual stimulation. A "bureaucratic" reference group
orientation will be derived from the choice of supervisor
as best judge of ability and other people in the organiza
tion as the source of intellectual stimulation. "Locals"
are an intermediate type derived from the other
combinations available, namely, profession as best judge of
abilities and members of occupation in the organization or
other people in the organization as the source of
intellectual stimulation. Also, supervisor as best judge
of abilities in combination with members of occupation in
the organization or members of occupation outside the
organization or a literature orientation was classified as
a "local" orientation.
66
Professional or Formal Training
Professional or formal training will be defined
operationally by the following questions:
1. Total number of years of schooling completed
at present time. (Please circle one.)
Grade School High School College Graduate School
12345678 1234 1234 12345
2. Have you had special training or education for
an occupation, whether or not you now held
this occupation (for example, college educa
tion, graduate school, trade school, apprentice
program, etcetera).
Yes_____ No_____
Organizational Socialization
Organizational socialization will be defined by
the following questions:
1. How long have you been employed where you
work?
Years Months
2. How many years have you worked full time,
at the occupation you have now?
Years
Work Satisfaction
A work satisfaction measure was derived from the
6 7
conditions or attitudes toward the job held by the
respondent at the time the questionnaire was completed.
These were:
1. Is there some other work, either here or
outside your company, which you would like
better than what you are now doing?
Responses provided were:
I would rather have some other job
I would rather have my present job.
2. Not counting all the other things that make
your particular job good or bad, how do
you like the kind of work you do?
Responses provided were:
I dislike it very much; would prefer
some other kind of work
I don’t like it very much; would prefer
some other kind of work
It's all right but there are other kinds
of work I like better
I like it very much, but there are other
kinds of work I like much better
It’s exactly the kind of work I like
best.
68
3. Taking all things into consideration
concerning your present job (work),
are you satisfied or dissatisfied with
it?
Responses provided were:
Very satisfied
______Satisfied
______Dissatisfied
Very dissatisfied
These three items produced a Guttman scale with
the following characteristics: Coefficient of
Reproducility = .94; Minimum Marginal Reproducibility =
.84; Coefficient of Scalability = .62.
Tension
High morale may be seen as inversely related to
the amount of tension a person perceives on the job.
Tension consists of a four item scale as follows:
1. Peeling that you have too heavy a load, one
that you can't possibly finish during an
ordinary workday.
2. Thinking that the amount of work you have
to do may interfere with how well it gets
done.
3. Peeling that you have too little authority
to carry out the responsibilities assigned
to you.
4. Peeling that you have too much responsibility
and authority delegated to you by your
supervisors.
Response categories provided for each
statement were:
1. Never
2. Rarely
3. Sometimes
4. Rather often
5. Nearly all the time.
These items produced a Guttman scale with the
following characteristics: Coefficient of
Reproducibility = .96; Minimum Marginal Reproducibility =
.85; Coefficient of Scalability = . 7^.
Job Freedom
One way of approaching the problem of satisfaction
is to ask how much freedom a person perceives on his job,
freedom to do his job, and freedom from interference.
The following items attempted to measure this dimension
of the work world:
1. Peeling that you have to do things on the
job that are against your better judgment.
; 70
2. Peeling unable to influence your immediate
supervisors decisions and actions that
affect you.
3. Thinking that you will not be able to
■ satisfy the conflicting demands of various
1 people over you.
Responses provided were:
; A. Never
1 B. Rarely
; C. Sometimes
|
D. Rather often
|
\ E. Nearly all the time.
i *
| 4. How much does your job give you a chance to
!
J do things you are best at?
| _____No chance at all
I
Very little chance
Some chance
Fairly good chance
| _____Very good chance.
The items produced a scale with the following
characteristics: Coefficient of Reproducibility = .87;
Minimum Marginal Reproducibility = .71; Coefficient of
Scalability = .54.
I
I
I Job Clarity
I
I A final measure which may be included under the
71
morale or satisfaction heading is defined as job clarity.
Job clarity refers generally to the knowledge of just what
the job entails and the knowledge of what is expected as
the tasks are carried out. The following items make up
this scale:
1. Not knowing what your supervisor thinks of
you, how he evaluates your performance.
2. The fact that you can’t get information
needed to carry out your job.
3. Not knowing just what people you work with
expect of you.
*1. Being unclear on just what the scope and
responsibilities of your job are.
Responses provided were:
A. Never
B. Rarely
C. Sometimes
D. Rather often
E. Nearly all the time.
. These items produced a scale with the following
characteristics: Coefficient of Reproducibility = .9^;
Minimum Marginal Reproducibility = .82; Coefficient of
Scalability = .65*
It is important to note the similarities among
the measures presented in this section, satisfaction,
72
tension, freedom, and job clarity. This similarity
indicates that caution must be exercised in interpreting
these measures as they relate to organizational attachment
or to each other.
Interpersonal Rewards
Sociometric questions were asked which requested
;the informant to name the five people who were his closest
friends, his work contacts, and the persons who he felt
were in control of his particular organization. These
I measures were utilized in several different ways.
First, a total interpersonal rewards score was
:derived for everyone in the sample. The total inter-
I
!personal reward measure is based upon the number of times
• a person was named as a friend, as a respected person, as a
I person seen as exercising control in the organization and
las a work contact. (Scores were collapsed into "low,"
I"medium" and "high" categories.) Individual scores were
j
;also derived which reported the amount of work contact
;individuals had with persons reported as friends, with
i
those co-workers they respected most, and those persons
whom they saw as in control in their respective
l
!organizations. Persons were again ranked in terms of low,
s
medium, and high as to their relative contact on each of
'these measures, by organization.
Statistical Techniques
‘Measures of Association
The measures presented in this chapter are all j
ordinal in nature. Reference group orientation, for
instance, allows the ordering of individuals in terms of
cosmopolitanism, but does not specify the differences
i i
i
'between ranks on this variable. Therefore, the measure !
, ' I
presented here is an ordinal measure of association, I
:Goodman and Kruskal’s (195*0 gamma. Gamma is based on the
i i
|predictability of order between two sets of ordinal I
i ;
I classifications. It expresses the probability of getting
i
j"like" rather than "unlike" orders in two classifications
I
I when individuals are chosen at random from a population.
|An individual who ranks higher or lower on two classifica-
: i
Itions as compared to another individual is said to be a
!like order. A rank higher on one classification but lower
|
!on the other is said to be an unlike order. While gamma
i
!expresses the magnitude of the degree of association
i
between two variables, the statistic has the disadvantage
of disregarding the cases in which individuals are tied on
| one or both variables. However, it presents a clear
|operational interpretation in terms of proportional
t
i
‘reduction in error.
i
i
i i
I Statistical Program j
| A computer program designed specifically for socialj
74
science data is presented in the Statistical Program for
the Social Sciences (Nie, et al., 1970). This program
was selected for this analysis because of the flexibility
it provides, including a comprehensive set of procedures
for data transformation. Specifically, the Guttman scaling
program and the crosstab procedures, with controls, were
utilized to test the hypotheses presented in Chapter II.
CHAPTER IV
TEST OP THE HYPOTHESES AND DISCUSSION j
i
I
Overview of the Findings
Of the hypotheses presented in Chapter II,
Hypothesis 1 is partially confirmed and Hypotheses 2, 3, ;
i
and 4 are disconfirmed. Hypothesis 1, which predicted that!
high educational attainment would be associated with the j
I
I
choice of an outside (cosmopolitan) reference group, is I
partially confirmed because high educational attainment is
associated with the choice of a cosmopolitan reference
group. However, this relationship is not without excep
tions, especially among those with the highest level of
education. Those persons most highly educated were more
likely to choose an inner (bureaucratic or local) reference
group orientation. Hypothesis 2 predicted that a high
level of attachment would be associated with a bureaucratic
reference group orientation; this is clearly disconfirmed.
Cosmopolitans were found to be most highly attached to
organizations. Hypothesis 3, predicting a strong relation
ship between level of attachment and length of employment,
is clearly disconfirmed, with highest attachment associated
with the lowest category on length of employment.
75
Hypothesis predicting a high level of interpersonal
rewards for those with high attachment, is also clearly
disconfirmed, with the highest level of rewards occurring I
among those with the lowest level of attachment. '
t
The hypotheses were tested by computing gamma ;
coefficients to describe the degree of association between !
the independent variables and the dependent variable,
|
organizational attachment. The independent variables were
also related to each other. These correlations are shown
in Table 1.
The intercorrelations between the variables which
^appear in the hypotheses were also subjected to statistical
control; these relationships appear in Table 2. i
I
A final statistical procedure involved examining
ithe intercorrelations among the major variables separately j
;by organization. These relationships appear in Table 3. j
I !
! j
:Hypothesis I: Partially Confirmed j
! The first hypothesis suggested that the length of
I formal or professional training would be associated with
!the choice of an outside (cosmopolitan) reference group
I
|orientation. This sample is highly educated, with 14 years
|
|of education and less ranked as "low," while "medium"
|education was from 15 to 17 years, and "high" educational
jattainment was 18 years and upward. Over half of the
|
jsample possessed graduate degrees.
TABLE 1
RELATIONSHIPS AMONG ORGANIZATIONAL ATTACHMENT, REFERENCE
GROUP ORIENTATION, EDUCATION, OCCUPATION, EMPLOYMENT
LENGTH, INTERPERSONAL REWARDS, AND SEX AMONG 260
MEMBERS OF FOUR PROFESSIONAL AND
SEMI-PROFESSIONAL ORGANIZATIONS
Attachment Reference Education Occupation Employ. Rewards Sex
Group Length
:Attachment .20 .00
C O
o
•
i
-.15
-.34 -.05
:Reference Group .36 .20 -.15 -.17 .25
I Education —
.75
.12
.49
.58
iOccupation —
• 29 .51 .69
|Employment Length —
.39
.24
i
|Rewards
— .40
j Sex
i
--
j aSee Chapter
|is gamma.
Ill for an explanation of measures. Measure of association
TABLE 2
RELATIONSHIPS' AMONG ATTACHMENT AND EDUCATION, REFERENCE GROUP
ORIENTATION, LENGTH OF EMPLOYMENT, AND INTERPERSONAL
REWARDS AMONG 260 MEMBERS OF FOUR PROFESSIONAL
AND SEMI-PROFESSIONAL ORGANIZATIONS,
WITH STATISTICAL CONTROLS
Zero-Order Relationship
(gamma)
i1. Reference Group-Education (.36)
i Controlling:
Occupation
Attachment
Employment Length
| Rewards
j Sex
I
|2. Reference Group-Attachment (.20)
Controlling:
Occupation
i Education
| Employment Length
Rewards
! Sex
Relationships Within Successive
Categories of Control Variables
(gamma)a
12 3 4
"Low"
.52 .22 .30
.ill .34 .33
.47
.26 .41
.28 .62
.59
.39 .29
Occupation .06 .16
.31
Education .13
.33
.20
Employment Length .16 .24 .16
Rewards .40
.07 .17
—a
co
TABLE 2— Continued
! 4
Relationships Within Successive
Categories of Control Variables
(gamma)a
(gamma) 1
"Low"
2
3
Attachment-Employment Length (-.15)
Controlling:
Occupation .06 -.19
-.24
Education -.08 -.23 -.15
Reference Group -.11 -.20 .16
Rewards
-.05 -.07
-.04
Sex -.08
-.19
Attachment-Rewards (-.34)
Controlling:
Occupation
-.59
-.30 -.32
Education -.54 -.39 -.35
Reference Group -.34 -.23
-.62
Employment Length -.25 -.45 -.25
Sex
-.43 -.30
-.19
-.42
Occupation has four categories, with clerical treated as low and managerial
| as high; reference group, attachment, rewards, length of employment, and education
are trichotomies. Reference group classifies bureaucrats as low and cosmopolitans as
high. The dichotomy sex classifies females as low
i
vo
TABLE 3
RELATIONSHIPS AMONG ATTACHMENT, OCCUPATION, EDUCATION, REFERENCE
GROUP ORIENTATION, EMPLOYMENT LENGTH, INTERPERSONAL REWARDS,
AND SEX IN EACH OF FOUR PROFESSIONAL AND SEMI-PROFESSIONAL
ORGANIZATIONS (TOTAL N = 260)
1. Semi-Autonomous Survey Research Organization (N = 47)
2 3 4
5
6
7
1. Education •
CT \
CO
•
—q
ro
.46
.31
.42 .82 !
2. Reference Group .22
• 47
.04 -.29 .68
3. Occupation .14
.29 .63 .94
4. Attachment -.09 -.31 .53
5• Employment Length .65
.21
6. Rewards
.25
7 • Sex
Autonomous Survey Research Organization (N = 62)
:
2 3 4
5
6
7
1. Education •50 .91 -.24 .28 .44
.87
2. Reference Group
• 39
-.11 .21 .00 .40
3. Occupation -.12 .42 .52 .80
4. Attachment .03 -.27 -•39
5. Employment Length .42
.25
6. Rewards .55
7. Sex
TABLE 3— Continued
Alcoholism Renabilitation Organization (N =
59)
2
3
4
5
6
7
1. Education ' .25 .86
-.03
-.04 .58
.34
2. Reference Group .15 .19
-.28
.07
-.18
3. Occupation
-.15 .31
.68
.43
4. Attachment -.20 -.27
-.16
5. Employment Length .21
.17
6. Rewards
.49
7. Sex
Commercial Research Organization (N = 92)
2
3
4
5
6
7
1. Education .15 .80
.07
.10
.51
.56
2. Reference Group .06 • 38 -.41 -.32 .30
3. Occupation .02 .36 .57
1.00
4. Attachment
-.35
-.41
-.09
5. Employment Length .41
.27
6. Rewards .39
7. Sex
a
gamma
The relationship between length of formal j
i
education and the choice of reference group is moderately I
positive (gamma = .36). Table 4 displays this relation- *
: ship. I
As Table 4 reveals, the strength of the relation- j
' I
ship between education and reference group results from the!
j
:large percentage of individuals with low educational j
:attainment who chose a "bureaucrat" orientation (60.7 j
. i
per cent of the low education category). While an outside
|(cosmopolitan) orientation is associated with high educa-
1
tional attainment (53.8 per cent of those identified as
jcosmopolitans were in the high educational category), the
;converse is not true. Note that nearly twice as many
ipersons high in educational attainment chose an inner
!(local) reference group (44.3 per cent to 23.9 per cent).
iIn fact, more persons high in educational attainment chose
!
;a bureaucratic orientation (31.8 per cent) than chose an
|outside reference group orientation. Still, it is correct
i
i
!to say that the probability of a cosmopolitan orientation
'increases with education.
I
| These findings are partially explained in terms of
I organizational differences. The magnitude of the gamma
i
|coefficients computed for each organization ranged from
i 1
! .15 to .68 (Table 3), with the commercial research
!
|organization showing the lowest relationship and the most
j
^professional research organization presenting the strongest
TABLE 4
RELATIONSHIP BETWEEN EDUCATION AND REFERENCE GROUP ORIENTATION
AMONG 260 MEMBERS OF FOUR PROFESSIONAL AND
SEMI-PROFESSIONAL ORGANIZATIONS
Reference Group Orientation
! Education
Bureaucratic
N (35)
Local
N. (35)
Cosmopolitan
N
Total
N (J5)
Low
51 (60.7) 27
(32.1) 6
(7.1)
84
(99-9)
Medium
39
(50.6) 26 (33.8) 12 (15.6)
77
(100.0)
High 28 (31.8)
39 (44.3)
21
(23.9)
88 (100.0)
Total 118 92
39 249
gamma = . 36.
Missing observations = 11,
; 84
relationship. Apparently, education is more relevant for j
reference group orientation in some organizations than
others. While it was noted that clear distinctions were j
not apparent between the degree of professionalization of J
: ]
; the organizations presented here, in this instance the !
; magnitude of the relationship between education and
i reference group orientation follows the crude rankings of i
professionalization presented in the description of the j
' i
| sample. j
While organizational differences tend to clarify
[
i the relationship between education and choice of reference
; i
j groupy it is important to note that in only two of the
i
i
! organizations does the number of cosmopolitans approach the
t
inumber of locals, and that in no single organization are
i
j there as many cosmopolitans as there are locals among those
!
| high in education. In fact, it is only in the organization
i . . . . .
I posited as most professional that the number of cosmopoli
tans is greater than the number of bureaucrats among those
i
! high in education. While important, education is quite
! clearly not related to reference group orientation in a
i
; simple way. Education in general is a reasonably good
j
predictor of bureaucratic-non-bureaucratic orientations,
but it is not adequate to distinguish between different
i
itypes of non-bureaucrats, except in the most profession-
! i
|alized surroundings. The conclusion which is suggested by
j
I this is a complex one. The data suggest that education has
85
, an important influence on reference group orientation, but
that the distinction between "locals” and "cosmopolitans"
is somewhat less salient than the literature on this
subject would suggest.1 (See Appendix.) The introduction
of control variables (Table 2) has some effect on the
relationship between education and reference group
orientation; however, overall, the relationship remains
basically unaltered.
; Hypothesis 2: Disconfirmed
The second hypothesis predicted that possession of
a cosmopolitan orientation would be associated with a low
: level of organizational attachment, while a bureaucratic
I reference group orientation would be associated with a high
;degree of organizational attachment. Confirmation of this
:hypothesis would have resulted in a negative relationship
J between reference group and attachment because a
jcosmopolitan orientation was treated as "high." Occupation
I
| was to be held constant to test this relationship. Table 5
i
idisplays this relationship without occupation held
i
'constant. The effect of occupation as a control variable
i
!
|in this relationship appears in Table 2, panel 2.
| The direction of the relationship between reference
;group orientation and organizational attachment without
!
! controls is relatively low and opposite to what was
|predicted (gamma = +.20). High attachment is associated
TABLE 5
RELATIONSHIP BETWEEN REFERENCE GROUP ORIENTATION AND ORGANIZATIONAL
ATTACHMENT AMONG 260 MEMBERS OF FOUR PROFESSIONAL
AND SEMI-PROFESSIONAL ORGANIZATIONS
Low
Reference Group N (%)
Bureaucratic 44
(37.3)
Local 36
(39.1)
Cosmopolitan
7 (17.9)
Total 87
Organizational Attachment
Medium High Total
N (%) N (*) N (*)
51
(43.2)
23 (19.5)
118 (100.0)
35 ‘
(38.0) 21 (22.8) 92 ( 99.9)
15 (38.5) 17
(43.6)
39
(100.0)
101 61 249
gamma = .20.
Missing observations = 11.
CO
CT\
; 87
with cosmopolitanism. In fact, nearly half (43.6 per cent)
of the cosmopolitans were highly attached. Locals are
slightly more attached than bureaucrats, while only 17*9
! per cent of the cosmopolitans were found to be low on
attachment. This compares to 39.1 per cent of the locals
and 37*3 per cent of the bureaucrats in the low attachment
category.
When occupation is controlled (Table 2), the
I relationship between reference group orientation and
| attachment is strongest for managerial personnel, declines
!slightly for professionals, becomes still smaller for
|technical and skilled, and almost disappears for clerical
i
iworkers. In other words, the strength of the positive
j
!relationship between reference group orientation and
i
jattachment steadily increases up the occupational ladder,
though it never becomes very high. Put another way, the
ihypothesized relationship between reference group
orientation is contradicted in every occupational category,
I but most firmly so in the highest occupational category.
i
|Bureaucrats in general are the least attached and
|especially if their occupational status is high. These
|findings could scarcely be farther from what was expected,
i
!
|although they are interpretable, as we shall presently see.
i When further controls are introduced into the
|relationship between reference group and attachment, it is
i
i
I seen that interpersonal rewards have the greatest impact
88
on the magnitude of this relationship, but in a direction
■ opposite to the effect of occupation. Where interpersonal
rewards are low, the relationship between cosmopolitanism
and attachment is strongest. Where rewards are not
lacking, cosmopolitanism shows no pronounced relationship
to attachment.
The conclusion which seems to fit these findings
best, though the relationships, are not strong enough to
;give it firm support, is that bureaucrats tend to be
i unattached, especially if their occupational status is
jhigh, and/or if the rewards they receive are low.
I
1 Conversely, this would suggest that cosmopolitans are more
I likely to be attached where their occupational status is
i
ihigh and/or their rewards are low. What these data suggest
!is that interpersonal rewards and occupational status mean
i
! very different things to those with bureaucratic
i
I orientations and those with a cosmopolitan outlook. These
i
i
!two categories are likely to be attached under very
I
|different circumstances. Among "bureaucrats," especially
i
I those of higher occupational status, organizational
I
i
;attachment may well be contingent upon receiving inter
personal rewards at work. Where those rewards are not
;forthcoming, attachment to the organization declines. In
icontrast, cosmopolitans apparently are more likely to
I "fall back" on the organization and express attachment to
!it when they cannot gain important interpersonal rewards
| from their co-workers, despite their high occupational
status. When they can count on interpersonal rewards from
their co-workers (whatever it is about them which produces
; this situation) they evidently have little need for the
:organization and consequently express less attachment to
: it.
In short, it would appear that reference group
orientation is relevant for group attachment, but in a very
! complex way. More than anything else, an individual’s
; reference group orientation would appear to affect
iattachment indirectly by conditioning his reliance upon
t
;interpersonal rewards at work. Among bureaucrats, rewards
I
I increase attachment; among cosmopolitans, rewards decrease
! attachment. What this may indicate is that an individual
j
|who is bureaucratically oriented requires rewards in order
!
! to remain attached to the organization. A cosmopolitan,
j
| on the other hand, is apparently liberated from the
j
|organization (in attitude, if not in fact) when he receives
ia high level of rewards from his co-workers. If he is
j
junable to command a high level of interpersonal rewards,
i
he is thrown back on the organization by default. In
either case, receiving rewards reinforces the individual's
I
sreference group orientation. Bureaucrats who are rewarded
i
iremain attached to the organization (remain bureaucrats);
|cosmopolitans who are rewarded retain their cosmopolitan
|outlook. Conversely, bureaucrats who are not rewarded
become alienated from the organization, while cosmopolitans 1
who are not rewarded lose their sense of independence from j
the organization.
Hypothesis 3: Disconfirmed
Hypothesis 3 predicted an association between the
length of employment and attachment. The prediction was
that the longer a person was employed, the greater the
attachment he would display. Education was held constant
!to test this relationship. Table 6 displays the relation-
i
jship between length of employment and attachment without
i
|controls, while the effect of education and other control
;variables are displayed in Table 2, panel 3.
| The overall relationship is negative, but not
i
I strong (gamma = -.15), and is affected in a curvilinear
imanner when education is controlled. Low educational
attainment reduces the relationship (gamma = -.08), medium
|education increases the magnitude (gamma = -.23), while
'high educational attainment reduces the relationship
j
I(gamma = -.15). Still, the conclusion must be that length
j
|of employment does not have a pronounced effect on
i
I
!attachment at any educational level.
j
j When length of employment and attachment is
i
|controlled by reference group, the relationship remains
'fairly consistent, for local and bureaucratic orientations,
i
jwith the overall level of the sample (gamma = -.20 and
TABLE 6
RELATIONSHIP BETWEEN EMPLOYMENT LENGTH AND ORGANIZATIONAL ATTACHMENT
AMONG 260 MEMBERS OF FOUR PROFESSIONAL AND
SEMI-PROFESSIONAL ORGANIZATIONS
Employment Length
Low
Medium
High
gamma = -. 15 •
Missing observations = 9.
Attachment
Low Medium High Total
N (%) N
27
(31.0) 32
29
(35.8)
33
32 (38.6) 36
Total 88 101
(%) N (%) N (%)
(36.8) 28 (32.2) 87 (100.0)
(40.7) 19 (23.5) 81
(100.0)
(43.4)
15
(18.1) 83 (100.0)
62
251
i gamma = -.11). However, in the cosmopolitan category, the !
i
■ relationship is reversed and becomes positive (gamma =
: .16). This is consistent with the finding that cosmo- 1
i
I politanism is related to high attachment. i
Introduction of interpersonal rewards as a control i
; j
i variable reduces the relationship between length of j
! employment and attachment in all categories of reward. I
: I
i i
I Occupational differences also affect the relationship, with!
| the strongest relationship again appearing in the j
Jmanagerial category, while it decreases down the occupa- j
I i
| tional ladder, almost disappearing for clerical personnel
j
| (Table 2). Organizational differences also affect this
I relationship to some extent. As displayed by Table 3, the
I magnitude of the relationship increases in the two least
|
jprofessionalized organizations (gamma = -.35 and gamma =
-.20), while it declines in the two most professionalized
organizations (gamma = .03 and gamma = -.09). The
implication of this finding appears to support, as do
I
reference group differences, the notion that profession
alism is related to attachment in this sample of
organizations. That is, while attachment decreases with
length of employment, cosmopolitanism and more highly
professionalized organizations tend to reverse or at least
1
minimize this negative relationship.
In short, length of employment is not systemati-
I
jcally related to attachment. The slight relationship which
93
; I
I appeared in the zero order correlation is not stable under i
I various control conditions, indicating that the effects of j
■ length of employment are largely an indirect reflection of |
| other variables. j
j I
Hypothesis Disconfirmed
i
Hypothesis 4 posited that high interpersonal
rewards would be associated with high organizational j
attachment. Reference group was held constant to test this
‘relationship. The overall relationship is presented in
! Table 7, without controlling on reference group,
i In this table, it is clear that rewards are
|negatively related to attachment (without controls,
|gamma = -.3*0 • When reference group orientation is con
trolled, the relationship displayed in Table 2, panel 4,
‘appears. The magnitude is.at the level of the sample as a
|
j whole for the bureaucratic reference group orientation
| (gamma = -.3*1); it decreases slightly for the local
I reference group orientation (gamma = -.23) and increases
|sharply for the cosmopolitan orientation (gamma = -.62).
!This is consistent with our earlier speculation that
|rewards are more likely to function as an alternate to
I attachment for cosmopolitans. On the other hand, the fact
I
ithat the relationship between attachment and rewards is
; still negative for bureaucrats indicates that our inter-
I
ipretation for Hypothesis 2 is not the whole story. There
TABLE 7
Rewards
Low
Medium
High
RELATIONSHIP BETWEEN INTERPERSONAL REWARDS AND ORGANIZATIONAL
ATTACHMENT AMONG 260 MEMBERS OF FOUR PROFESSIONAL
AND SEMI-PROFESSIONAL ORGANIZATIONS
Low
N (SO
Attachment
Medium
N (SO
High
N (JO
Total
N
10 (13-9) 36 (50.0) 26
41 (42.3) 33 (34.0) 23
37 (45-1) 32 (39.0) 13
(36.1) 72
(23.7) 97
(15.9) 82
(100.0)
(100.0)
(100. 0)
Total 88 101 62 251
gamma = -.34.
Missing observations = 9*
95
is still a lot of unexplained variation to be accounted
for.
The individual measures which were combined into
the composite measure of interpersonal reward were cross-
itabulated with organizational attachment. These include
the relative member of friendship choices, control choices,
work contacts, and prestige nominations received. They
were trichotomized by organization into low, medium, and
high categories. Additionally, contacts with friends,
'respected co-workers, and authority figures were calculated
|as a proportion of total work contacts, and these were
divided into low, medium, and high categories by organiza
tion. Finally, a measure was computed of the relative
i
i
!number of information contacts an individual received in
I
j
|the organization, again classified into low, medium, and
!high categories. The results of these crosstabulations
!appear in Table 8.
i
!
i Almost without exception, these relationships are
I
'negative for the sample as a whole, as well as for each
I
|organization taken separately. In conclusion, the
|relationship between interpersonal rewards and attachment
I is negative. This is a consistent finding, regardless of
!the measure of interpersonal reward utilized. Moreover,
jthis relationship is especially characteristic of
i cosmopolitans.
TABLE 8
RELATIONSHIPS3- BETWEEN VARIOUS INTERPERSONAL REWARD MEASURES AND
ORGANIZATIONAL ATTACHMENT IN FOUR PROFESSIONAL AND
SEMI-PROFESSIONAL ORGANIZATIONS
TOTAL N = 260)
Attachment Measures A B
Organ! zat ion1 3
C D E
Relative Number of Friends
OJ
-=r
•
1
-.37 -.29 -.38 -.37
Relative Number of Controls
-.35
-.06
00
0
•
1
-.31
-.26
Relative Prestige -.46
-.31 -.17 -.29
VO
00
•
1
Relative Work Contacts
-.25
-.24 -.10 -.21
VO
1 —1
•
1
Relative Information Contacts -.28
-.25 -.19
-.34 -.27
Respected Contacts
.09
-.16
-.13 -.35
-.11
Authority Contacts .08
-.39 -.25 .13 -.05
Friend Contacts -.14
O
•
1
H
CM
•
1
-.33 -.13
a
gamma
A refers to the commercial research organization (N = 42)
B refers to the semi-autonomous survey research organization (N = 47)
C refers to the autonomous survey research organization (N = 62)
D refers to the alcoholism rehabilitation organization (N = 59)
E refers to the entire sample (N = 260).
;Attachment and Exploratory Variables !
Organizational attachment was intercorrelated with :
the variables defined as exploratory in Chapter II. These j
include satisfaction, tension, freedom, and job clarity. i
:Table 9 displays these relationships, along with the
intercorrelations among the exploratory variables. j
I
| For the entire sample, only tension is positively j
related to attachment (gamma = .18). Among those for whom
|satisfaction, clarity, and freedom are low, the relation-
i
jship is more pronounced, though still not high. The
!
;introduction of reference group orientation as a control
|variable in the relationship between the exploratory
I variables and attachment is displayed in Table 10.
| While not pronounced, the introduction of reference
I
group orientation presents an instance where locals come
into sharper focus in this analysis. Locals are con-
j
sistently more affected than bureaucrats or cosmopolitans
I by the exploratory variables. Tension and satisfaction are
I more highly related to attachment for locals, and at the
jsame time, freedom and job clarity are more negatively
|related to attachment for locals than the other reference
|
|group orientations, except for job clarity. This finding
I is consistent with the common sense approach to reference
j
|group orientation which indicates that locals should have
ithe combined effects of both professional and organizational
TABLE 9
RELATIONSHIPS3- BETWEEN ORGANIZATIONAL ATTACHMENT AND JOB TENSION,
WORK SATISFACTION, JOB FREEDOM, AND JOB CLARITY, WITH CONTROLS,
AMONG 260 MEMBERS OF FOUR PROFESSIONAL AND
SEMI-PROFESSIONAL ORGANIZATIONS
Zero-Order Relationship
(gamma)
(1) Attachment-Tension (.18)
Controlling:
Satisfaction
Freedom
Job Clarity
(2) Attachment-Satisfaetion (.03)
Controlling:
Tension
Freedom
Job Clarity
(3) Attachment-Freedom (-.38)
Controlling:
Tension
Job Clarity
Satisfaction
Relationships Within Successive
Categories of Control
Variables
Low Medium High
.29 -.06 .11.
.25 -.20 .09
.31 .13 ' .02
.18 -.04 -.11
.00 .00 .08
-.29 .03 - .18
-.33 -.18 -.59
-.62 -.12 -.26
-.33 -.36 -.24
TABLE 9— Continued
Relationships Within Successive
Categories of Control
Variables
Zero-Order Relationship
(gamma)
I (4) Attachment-Job Clarity (-.40)
Controlling:
! Tension
I Freedom
! Satisfaction
Low Medium High
.25
.54
.47
-.38
-.19
-.32
-.52
-.14
-.18
gamma
V O
vo
TABLE 10
CROSSTABULATIONS OP ORGANIZATIONAL ATTACHMENT AND JOB TENSION, WORK
SATISFACTION, JOB FREEDOM, AND JOB CLARITY FOR FOUR PROFESSIONAL
AND SEMI-PROFESSIONAL ORGANIZATIONS, WITH REFERENCE GROUP
ORIENTATION INTRODUCED AS A STATISTICAL CONTROL
Zero--Order Relationship
Reference Group Orientation
(gamma) Bureaucratic Local Cosmopolitan
(1)
Attaehment-Tension (.18) .01
.39 .27
(2) Attachment-Satisfaction (.03) .02
.09
.01
(3)
Attachment-Freedom (-.38) -.32 -.48
-.39
(4) Attachment-Job Clarity (-.40)
-.45 -.36 -.21
a
gamma
101
pressures. Why these differences did not appear in the
analysis of the testing of the hypothesis remains an
unanswered question.
While the relationship between attachment and job
satisfaction is only negligibly negative for the entire
sample (gamma = -.03), the other two exploratory variables
which may be defined as "rewarding" (job clarity and job
freedom) are more clearly negatively related to attachment
(gamma = -.40 and gamma = -.38, respectively). These
relationships substantially hold up when controls are
applied. It is significant that the one variable in this
group which is not rewarding, i.e., tension, is positively
related to attachment. Again the results are consistent
with the findings discussed earlier. Those individuals to
whom rewards disproportionately accrue are noticeably less
attached to the organization than other personnel.
Summary of the Tests of the Hypotheses
The testing of the hypotheses resulted in the
partial confirmation of Hypothesis 1. High educational
attainment is moderately associated with the choice of an
outside reference group orientation. However, it was noted
that high educational attainment alone is not sufficient to
explain reference group choice.
Hypothesis 2 was not confirmed. A bureaucratic
reference group orientation is not associated with high
102
• i
; organizational attachment in this sample. The reverse is
true, with an outer (cosmopolitan) reference group
orientation associated with the highest level of organi-
, zational attachment.
Hypothesis 3 is also disconfirmed, with attachment '
decreasing as the length of employment increases; the
reverse of the prediction was found to be true. The I
; highest level of attachment was found in the category ;
1 employed the least amount of time. I
‘ \
Finally, Hypothesis 4 is disconfirmed, with
j interpersonal rewards clearly inversely related to
i
| attachment.
j Of all the exploratory variables, only tension is
i
! positively related to attachment. Satisfaction shows
i almost no relationship to attachment, while freedom and job
| clarity are moderately negatively related to attachment.
| A Multiple Regression Analysis
This analysis was described as inductive and
somewhat exploratory. The findings reported in the first
i
j section of this chapter are surprising, to say the least,
! and more importantly, they leave unanswered several
I
i
| questions which were stated as basic goals of this research.
i
I One question still unanswered concerns the major dimensions
| of organizational attachment. A related question concerns
! j
jthe relationship between the variables which are central toi
103 ;
this analysis. While organizational differences may appear!
to clarify some of these questions, great caution must be !
used in interpreting findings based upon organizational
l
variation. A legitimate basis for comparisons based on i
I
organizational differences is not apparent in this sample,
:since the critical differences in organizational context
cannot be specified. On the basis of the findings presented;
to this point, a multiple regression analysis was under- !
taken to further explore the relationship between i
; j
;organizational attachment and the independent variables j
i . i
;utilized in this research.
! Multiple regression is a statistical technique
;which attempts to predict a single dependent variable from
i
i
|any number of independent variables (Blalock, i960). It is
|typically used to indicate how much of the total variation
|in the dependent variable (organizational attachment in
jthis analysis) can be explained by all of the independent
!
i
|variables acting together. Until recently, the assumption
I
|was made that the data subjected to multiple regression
analysis should be at least interval in measurement.
Recently, the argument has been presented, especially in
isociology, that multiple regression is appropriate with
(ordinal data (for instance, see Hawkes, 1971)* The
jmeasurement techniques available to the sociologist have
;been the basis for this argument. Most of these techniques
Iproduce measurements that are no more than ordinal. The
104
choice becomes one of either using the less sensitive, less
powerful, nonrparametric techniques, and facing the
possibility of losing valuable information (see Labovitz,
1967), or proceeding as if the measurement were actually
interval. Several ways of proceeding are available,
including assigning numbers to rank order data (Labovitz,
1970a), or merely proceeding as if the measures were of
interval level and interpreting findings with caution.
This research proceeds on the latter choice.
The Statistical Package for the Social Sciences
Nie, et al., (1970) provides a multiple regression
program designed for ordinal data. The multiple regression
analysis proceeds on the basis of a correlation matrix
generated from the intercorrelations of all the variables
to be included in the analysis. The user has the option
of using either Kendall’s or Spearman’s rank order
correlation, with tau recommended for data containing a
large number of tied ranks. This matrix was generated
utilizing Kendall’s tau as the measure of association, and
is based upon the independent and exploratory variables.
(This matrix is displayed in the appendix.)
Multiple regression attempts to produce a linear
combination of independent variables which correlate as
highly as possible with the dependent variable. With a
large number of independent variables, an indication of
105
their relative importance can be found by relating the j
dependent variable to each independent variable in turn, !
always controlling for the remaining variables. In short, i
• i
| the best possible combination of variables, predicting the I
dependent variable, can be found among a large number of j
independent variables. Stepwise multiple regression i
, provides a means of choosing the best possible prediction j
; !
: with the fewest independent variables. The method con- |
; structs a prediction equation one independent variable at |
; !
| a time, with the first added the best predictor of the
I
| dependent variable. The second independent variable added
' provides the best prediction in conjunction with the first
i
I independent variable. Variables may be added until they
j
| no longer make a significant contribution to the
|prediction equation or the choice may be a certain number
!
!of variables to be included in the equation. Stepwise
|
regression was the method used in this analysis.
Multiple regression produces a statistic called
j
!the multiple correlation coefficient. The square of this
statistic may be interpreted as the variation explained in
|the dependent variable by the multiple regression equation.
i
j The stepwise regression procedure was used to
i
Jdetermine the best predictors of organizational attachment
| of all the independent variables in this analysis. The
i
! independent variables were also divided into two
I categories, one consisting of all the variables in the
106 ;
hypotheses and controls and the other consisting of the
exploratory variables plus reference group. The results !
produced by this procedure appear in Table 11. j
The first regression equation which included all of j
I
:the independent variables produced a multiple R = .41, |
which explains 17 per cent of the’ variance in organiza- I
!
: I
tional attachment. The five best predictors of reference
: I
;group, in order of their contribution to the regression j
equation, were job clarity, rewards, freedom, tension, and
:education. However, tension and education added little
jwhen included in the equation. Job clarity, rewards, and
;freedom, together, explained 16 per cent of the variation
i
I in attachment. Considering the three best predictors of
i
I
|attachment of the major variables, rewards alone explained
; 5 per cent of the variation in attachment. The addition
|of reference group and education contributed little to the
!regression equation and, together, the best three pre-
]
jdictors of attachment of the major variables in this
!
i •
|analysis explained only 6 per cent of the variation in
j
jattachment. The best three predictors of attachment of the
exploratory variables explained 13 per cent of the
i
!
|variation in attachment. Of these three, job clarity and
[freedom together explained 12 per cent of the variation,
jwhile the addition of tension added little.
i
! In short, three variables, rewards, clarity, and
I
i
!freedom appear to be most useful for predicting
TABLE 11
MULTIPLE REGRESSION ANALYSIS USING VARIOUS INDEPENDENT VARIABLES
TO PREDICT ORGANIZATIONAL ATTACHMENT AMONG 260 MEMBERS OF
FOUR PROFESSIONAL AND SEMI-PROFESSIONAL ORGANIZATIONS
Multiple R r R? r
All Independent Variables5
Job Clarity .30 .09
-.20
Rewards • 36 .13 -.23
Freedom .40 .16 -.16
Tension .41 .16 .14
Education .41
.17
.01
Major Variables*3
Rewards
Reference Group
Education
.22
.24
.25
.05
.06
.06
-.22
.12
.01
TABLE 11— Continued
Multiple R R2 r
| Exploratory Variables0
Job Clarity .30 .09
-.30
Freedom
.35
.12 -.30
Tension
• 35 .13
.14
I aReports the five best predictors of attachment among ten independent variables
| in the analysis.
v
Reports the three best predictors of attachment among the variables in the
I hypotheses with the addition of sex.
! o
| Reports the three best predictors of attachment among the exploratory
| variables with the addition of reference group.
!
O
CO
! 109
'organizational attachment. It is crucial to note that all ;
; I
:three of these variables are negatively related to
attachment. j
| Implications of the Multiple Regression Analysis j
It is important to note that the multiple j
. regression analysis produced multiples which are low and
| therefore should be interpreted with great caution. How- j
i ever, for heuristic purposes, the results of the regression!
; I
j analysis can be used to generate hypotheses for future ]
I :
I research on organizational attachment. I
! The multiple regression analysis supports the
; notion of attachment as a negative thing. 'Itewards,
| clarity, and freedom are the best predictors of low
organizational attachment, since they are negatively
: related to attachment. When combined with the results of
i
I
I the test of the hypotheses, the multiple regression
analysis suggests that a system of variables may be seen
j as interacting to produce differences in organizational
I attachment and in reference group orientation. The overall
[effect of this system is negative in terms of attachment.
I
| The positive influences on attachment are
apparently conditioned by reference group orientation
through tension and education. This system accounts for a J
! minute amount of variation in attachment; together they j
| i
i !
| only explained one per cent of the variation in
110
organizational attachment when added to the multiple
;regression equation.
; Reference group orientation has both direct and
;indirect negative effects on differences in attachment.
|There is a strong implication that in addition to their ;
!relation to each other, reference group and attachment are :
common correlates of clarity and rewards. However, the
i effect of clarity and rewards on attachment seems to best I
i i
;be explained in terms of their meaning to different I
I j
I reference group orientations. For example, the effect of j
i
j reference group on attachment seems to depend on the amount
i
!
|of rewards and the effect of rewards is different for
t
ibureaucrats and professionals. That is, the effect of j
| I
|rewards depends on the type of reference group
orientation.
I
i Theoretical Implications of the Findings
I
| A major criticism of the research that has been
done in the area of organizational attachment was described
as the absence of theory. More precisely, the criticism
was directed at the consistent lack of the application of
i
theoretical frameworks to research findings. Chapter II
proposed reference group and exchange theories as relevant
frameworks and suggested that a possible synthesis between
the two might overcome deficiencies in both, and at the
same time, help to clarify the dimensions of organizational
i
111
attachment. The theoretical implications of the findings
are crucial to this analysis if anything is to be
contributed to our knowledge of organizational attachment.
Reference group theory, as a framework capable of j
: encompassing attachment, was in competition with one of
the major variables presented in this study, occupations.
There was the possibility that occupation alone would be a j
better predictor of attachment, than reference group. |
1 Occupation, however, disappeared completely as a predictor j
|
; of attachment or as a correlate of reference group ;
■ i
!
: orientation in the multiple regression analysis. |
i The major theoretical implication of this research |
!
]
; deals with the relationship between reference group and I
| i
i exchange theories. The data show that the effect of i
! rewards depends on reference group orientation. Bureau-
i
: crats need rewards to be attached; cosmopolitans need
| rewards to be liberated from the organization. In other
j words, the presence or absence of rewards determines
1
i
j whether reference group is consistent with the degree of
’attachment. Reference group theory, in past research, has
! relied almost exclusively on social-psychological
j characteristics such a^s age, sex, and marital status. In
!
| this analysis, reference group theory, in combination with
| exchange theory, incorporated a wide range of variables
: within a single framework. The results of the testing of
| the hypotheses produced results which were interpretable in
112 i
terms of this framework, Reference group theory emerged
from this analysis as a framework which can encompass |
organizational attachment and is more valuable to the area j
of organizational analysis than seemed possible from the
|previous research findings! I
Implications for Future Research
i
j
Since this research was exploratory, the implica-
|
' tions for future research must be directed toward |
j i
clarification of the findings presented here. The j
. suggestion is that the major emphasis should be on
'reference groups, rewards, and organizational differences.
'A major unanswered question of'this research deals with the
! meaning of the "local" reference group. A bure.aucratic-
» non-bureaucratic distinction was apparent throughout the
i
I analysis, while clear local-cosmopolitan distinctions
|failed to emerge, except on the job tension items, and then
I the distinctions were never pronounced. It would certainly
i
|be essential to have a larger number of cosmopolitans in
i
|future research. In this study, there were many more
j
|locals than cosmopolitans, but the small number of
|cosmopolitans that did appear, contributed meaningful
jdistinctions. Finally, it is clear that more sophisticated
!measures of reference group are needed for future research.
j
| Clearer measures of reward are also required in
I future research. Organizational versus interpersonal j
113
distinctions are required to clarify the role of rewards in
determining organizational attachment. The implication of
this research is that rewards have a differential impact on
various reference group orientations, and that cosmopoli
tans are thrown back on the organization (attached) by
default. It would be necessary to have measures of
professional rewards, including publications or other
information of professional standing, to determine whether
or not the cosmopolitan is dependent upon an alternate
reward system.
Organizational differences seem crucial to future
research activity. This analysis suffered from an
inability to interpret divergent findings when they
appeared between organizations because organizational
contexts could not clearly be distinguished. As was
pointed out earlier, only a small number of cosmopolitans
were identified in this analysis. If even more profes
sionalized organizations than the present ones were
included in a future sample, these might be more profitably
compared with more bureaucratic organizations. By the same
reasoning, less professionalized organizations must also be
surveyed. This is just another way of saying that a
representative sample of organizations is needed for a
definitive test of the ideas put forth here.
Finally, a goal of future research would be to
determine why organizational attachment seems to be "by
11H
default." While the correlates of attachment in this
analysis were overshelmingly negative, the individual
|measures of attachment have no such connotation (see the
.description of these measures in Chapter III).
Summary of the Findings
Organizational attachment has been described here
ias a negative consequence of participation in highly
: professionalized organizations. The data have shown that
I reference group orientation is positively related to
j organizational attachment and, contrary to the implication
t
I of the review of the literature, a cosmopolitan reference
!group orientation is positively related to organizational
;attachment. On the other hand, bureaucratic and local
jreference group orientations were negatively related to
! organizational attachment. These differences are-partially
|
j explainable in terms of the effect of interpersonal rewards
on different reference group orientations. Bureaucrats
need rewards to be attached, while cosmopolitans need
|
|rewards to be liberated from the organization. Bureaucrats
I
jwho fail to receive rewards are not attached, while
J cosmopolitans who fail to receive rewards are attached by
j default. The most important finding to emerge in this
!
j study is that the presence or absence of rewards seems to
l
|determine whether reference group orientation remains
i
! consistent with the degree of attachment.
115
The correlates of attachment, revealed in the
course of this research, are overwhelmingly negative, with
length of employment, job freedom, job clarity, and
especially rewards all negatively related to attachment.
The suggestion was made in Chapter I that this might be the
case. However, this suggestion came mainly for research
which dealt with manual workers and support of this
position in a highly professionalized sample is indeed
surprising.
The findings support the proposed synthesis between
reference group and exchange theories and suggest that this
framework has some utility for future research in the area
of organizational attachment. The findings also support
the utility of the local-cosmopolitan framework in the area
of organizational attachment and contradict the notion that
it should be abandoned.
REFERENCES
116
117
REFERENCES
Abrahamson, Mark
1967 The Professional in the Organization.
Chicago: Rand McNally.
Altman, Irwin and Joseph E. McGrath
1966 Small Group Research’:: A Synthesis and
Critique of the Field. New York: Holt,
Rinehart and Winston.
Argyris, C.
1957 "The Individual and Organization: Some
1 Problems of Mutual Adjustment." Administrative
| Quarterly 1 (June): 1-24.
1957 Personality and Organization. New York :
Harper and Row.
I Back, K. W.
! 1951 "Influence Through Social Communication."
I Journal of Abnormal Social Psychology 46:
| 9-23.
|Bass, B. M.
' i960 Leadership, Psychology, and Organizational
Behavior. New York: Prentice-Hall.
! 118
Bennis, W. G., N. Berkowitz, M. Affinito, and M. Malone
1958 "Reference Groups and Loyalties in the
Out-Patient Department." Administrative
Science Quarterly 2: 481-500.
i Berkowitz, L.
1954 "Group Standards, Cohesiveness and Produc
tivity." Human Relations 7-* 509-519*
: Blalock, Hubert M.
■ i960 Social Statistics. New York: McGraw-Hill.
! Blau, Peter and Richard W. Scott.
; 1962 Formal Organizations. San Francisco:
i Chandler.
i
| Blauner, Robert
j
| 1969 "Work Satisfaction and Industrial Trends."
I in Amitai Etzioni (ed.) A Sociological Reader
I
| on Complex Organizations. New York: Holt,
1
! Rinehart and Winston.
i
| 1964 Alienation and Freedom: The Factory Worker
1
and his Industry. Chicago: University of
Chicago.
Brim, Orville G., Jr. and Stanton Wheeler
1966 Socialization after Childhood. New York:,
John Wiley. 18-33-
Caplow, Theodore
1964 Principles of Organization. New York:
Harcourt, Braice and World.
119
Cartwright, D and A. P. Zander (eds.)
1950 Group Dynamics: Research and Theory. 2nd ed.
Evanston, Illinois: Row Peterson.
Chinoy, Ely
1952 "The Tradition of Opportunity and the
Aspirations of Automobile Workers."
American Journal of Sociology 57 (March):
University of Chicago.
Coch, L. and J. R. P. French
1948 "Overcoming Resistance to Change." Human
Relations I: 512-532.
Collins, Barry E. and Bertram H. Raven
1968 "Group Structure: Attraction, Coalition,
Communication, and Power." in Lindsey and
Aronson (eds.). The Handbook of Social
Psychology 4. 2nd Ed.
Dubin, Robert
1968 "Necessary Versus Voluntary Behavior."
Human/Relations in Administration.
Englewood Cliffs: Prentice-Hall.
1968 "Personality and Organization." Human
Relations in Administrations. Englewood
Cliffs: Prentice-Hall.
Dubin, 'Robert
1956 "Industrial Workers Worlds: A Study of the
Central Life Interests of Industrial Workers
Social Problems 3 (January): 131-141.
Edwards, Allen L.
1957 Techniques of Attitude Scale Construction.
New York: Appleton-Century-Crofts.
Etzioni, Amitai
1964 Modern Organizations. New Jersey:
Prentice-Hall.
1958 "Democratic and Nondemocratic Supervision in
Industry." Journal of Human Relations 6, 4
(Summer): 47-51.
Parris, Robert E. L.
1964 Handbook of Modern Sociology. Chicago:
Rand McNally.
Paunce, William A.
1967 Problems of an Industrial Society. New York
McGraw-Hill.
1967 Readings in Industrial Sociology. New York:
Appleton-Century-Crofts.
Pestinger, L., H. Gerard, B. Hymovitch, H. Kelley, and
B. H. Raven.
1952 "The Influence Process in the Presence of
Extreme Deviants." Human Relations 5-*
121
Festinger, L. and J. Thibaut
1951 "Interpersonal Communication in Small Groups."
Journal of Abnormal Social Psychology 46:
92-99.
Gerard, H. B.
1954 "The Anchorage of Opinions in Pace-to-Pace
Groups." Human Relations 7: 313-325*
Gewirtz, Jacob L.
1969 "Mechanisms of Social Learning: Some Roles of
Stimulation and Behavior in Early Human
Development." Goslin’s Handbook of Social
Theory and Research. Chicago: Rand McNally.
Glaser, Barney (ed.)
1968 Organizational Careers: Sourcebook for Theory.
Chicago: Aldine.
1964 Organizational Scientists:. Their Professional
Careers. New York: Bobbs-Merrill.
1963 "Local-Cosmopolitan Scientists." American
Journal of Sociology 70 : 704-710.
Glaser, R., D. J. Klaus, and K. Egerman
1962 "Increasing Team Efficiency through Training."
The Acquisition and Extinction of a Team
Response 2. Pittsburgh: American Institute
for Research.
122
Goldberg, Louis C., Prank Baker, Albert H. Rubenstein
1965 "Local-Cosmopolitan: Unidimensional or
Multidimensional." Research Notes. American
Journal of Sociology JO: 704-710.
Goldhamer, Herbert and Edward A. Shils
1939 "Types of Power and Status." American
Journal of Sociology 45 (September): 171-182.
Goldner, Fred H. and R. R. Ritti !
1967 "Professionalization as Career Immobility." j
American Journal of Sociology (March): j
!
I
University of Chicago. |
Golembiewski, Robert T.
1964 "Small Groups and Large Organizations."
James G. March (ed.) Handbook of Organization.
Chicago: Rand McNally. 1
1962 The Small Group: An Analysis of Research
Concepts and Operations. Chicago: University
of Chicago.
Goode, William J.
1961 "Encroachment, Charlatanism, and the Emerging
Professions: Psychology, Sociology, and
Medicine." American Sociological Review 25
(December): 902-914.
Goslin, David A.
1
1969 Handbook of Socialization Theory and Research.
Chicago: Rand McNally.
123
Gouldner, Alvin W.
"Organizational Analysis." in Robert K.
Merton, et al., (eds.) Sociology Today. New j
York: Basic Books. j
"Cosmopolitans and Locals: Toward an Analysis ;
of Latent Social Roles -1." Administrative j
Science Quarterly 2 (March): 444-480.
I
"Cosmopolitans and Locals: Toward an Analysis ;
|
of Latent Social Roles -2." Administrative !
j
Science Quarterly (December): 281-306. !
Patterns of Industrial Bureaucracy. Glencoe,
Illinois: Free Press. i
|
"The Problem of Succession in Bureaucracy." |
Studies in Leadership: Leadership and
Democratic Action. New York: Harper.
Greer, Scott A.
1955 Social Organization. New York: Random House.
Grimes, Andrew J. and Philip K. Berger
1970 "Cosmopolitan-Local: Evaluation of the
Construct." Administrative Science Quarterly
15 (December): 407-416.
Gross, Edward
1953 "Some Functional Consequences of Primary
Controls in Formal Work Organizations."
j
American Sociological Review 19 (August):
368-373.
1959
1958
1957
1954
1950
124
Grosser, D., N. Polansky, and R. Lippitt
1951 "A Laboratory Study of Behavior Contagion."
Human Relations 4: 115-142.
Grusky, Oscar and George A. Miller
1970 The Sociology of Organizations: Basic Studies.
New York: Free Press.
Haire, Mason (ed.)
1959 Modern Organization Theory. New York: John
Wiley and Sons.
Hare, Paul A.
1966 Small Groups: Studies in Social Interaction.
New York: Alfred A. Knopf.
1962 Handbook of Small Group Research. New York:
Free Press.
Haug, Marie R. and M. B. Sussman
1969 "Professionalism and the Public."
Sociological Inquiry 39/1 (Winter): 57-63.
Hawkes, Roland K.
1971 "The Multivariate Analysis of Ordinal
Measures." American Journal of Sociology 76
(March): 908-926.
Homans, George Caspar
1964 "Bringing Men Back In." American Sociological
Review 29 (December): 809-818.
125
Homans, George Caspar
1961 Social Behavior: Its Elementary Forms.
R. K. Merton (ed.) New York: Harcourt, Brace
and World.
Horowitz, John J.
1970 Team Practice and the Specialist. Springfield,
Illinois: Charles G. Thomas.
Hughes, Everett C.
1963 "Professions." Daedalus (Fall): 655-668.
!Hyman, Herbert H.
1942 "The Psychology of ...Status. " Archives of
Sociology 269: 5-38, 80-86.
| Hyman, Herbert H. and Elanor Singer
1968 Readings in Reference Group Theory and
Research. New York: Free Press.
| Kahn, Robert L., Donald M. Wolfe, Robert P. Quinn, R. A.
1964 Rosenthal
j
! Organizational Stress: Studies in Role
j
| Conflict and Ambiguity. New York: John
| Wiley and Sons.
| Kemper, Theodore D.
! "1968 "Reference Groups, Socialization and
j
Achievement." American Sociological Review
I 33: 31-45.
! 1963 "The Relationship Between Self-Concept and the
Characteristics and Expectations of Significant
126
Others." Ph.D. Dissertation. New York:
New York University.
Kerlinger, Fred N.
1964 Foundations of Behavioral Research. New York
Holt, Rinehart and Winston.
Kornhauser, William
1962 Scientists in Industry: Conflict and
Accomodation. Berkeley: University of
California.
Kruskal, William and Leo Goodman
1959 "Measures of Association for Cross
Classifications, II." Journal of the
American Statistical Association 54 (March):
123-163.
Labovitz, Sanford
1970a "The Assignment of Numbers to Rank Order
Categories." American Sociological Review
35 (June): 515-524.
1970b "The Nonutility of Significance Tests;
The Significance of Tests of Significance
Reconsidered." Pacific Sociological Review
13 (Summer) : 141-148.
1967 "Some Observations on Measurement and
Statistics." Social Forces 46 (December):
151-160.
127
Lazarsfeld, Raul P. and Morris Rosenberg
1955 The Language of Social Research. New York:
Free Press. j
jLeavitt, Harold J.
j 1951 "Some Effects of Certain Communication
Patterns on Group Performance." Journal of i
; Abnormal Psychology 46: 38-50.
;Lewin, Kurt and Ronald Lippitt
1938 "An Experimental Approach to the Study of j
; Autocracy and Democracy: A Preliminary
i
' Note." Sociometry 1: 292-300.
f
|Liberman, Robert Paul
! 1971 "Behavioral Group Therapy: A Controlled
I
i Clinical Study." British Journal of
I
!
! Psychiatry 119. #552 (November): 535-544.
i
jLikert, Renses
| 1961 "An Emerging Theory of Organization, Leader
ship and Management." L. Petrullo and B. M.
Bass (eds.) Leadership and Interpersonal Behav-:'
i
! ior. New York: Holt, Rinehart and Winston.
i
| 290-309.
j
i 1961 New Patterns of Management. New York: McGraw.
i
jLindesmith, Alfred R. and Anselen L. Strauss
i
| 1969 Readings in Social Psychology. New York:
1
| Holt, Rinehart and Winston.
128
Mahoney, Thomas A. and William Weitzel
1969 "Managerial Models of Organizational
Effectiveness." American Sociological
Quarterly 14 (March): 357. j
Marcson, Simon
! ;
1961 "Organization and Authority in Industrial j
Research." Social Forces *10 (October): 72-80.1
i960 The Scientist in American Industry.- New j
York: Harper and Row. |
Marsh, James G. and Herbert A. Simon with Harold Guetzkow
1958 Organizations. New York: John Wiley and
Sons.
Martindale, Don
i960 The Nature and Tupes of Social Theory.
Boston: Houghton Mifflin.
McKelvey, W.
1969 "Exceptional Noncomplementarity and Style of
Interaction between Professional and
Organization." Administrative Science
Quarterly 14 (March): 21.
Merton, Robert K.
1957 Social Theory and Social Structure.
New York: Free Press.
Miller, George A.
1967 "Professionals in Bureaucracy: Alienation
Among Industrial Scientists and Engineers."
American Sociological Review 32 (October):
755-767.
Miller, Jon
1970 "Social-Psychological Implications of Weber's
Model of Bureaucracy: Relations Among
Expertise, Control, Authority, and
; Legitimacy." Social Forces (September):
91-101.
j
Miller, Jon and Sanford Labovitz
i
! 1971 "Individual Reaction to Organizational
Conflict and Change." (Forthcoming)
! »
I Mills, Theodore M.
I
I 1967 The Sociology of Small Groups. New Jersey:
I Prentice-Hall.
1
jMurray, V. V.
1
| 1966 . "Loyalty to Immediate Superiors at Alternate
1
I Hierarchical Levels in Bureaucracy."
i
I American Journal of Sociology 1 (July): 77-85
|Newcomb, Theodore M.
| 1951 Social Psychology. New York: Dryden Press.
1
1
|Nie, Norman H., Dale H. Bent, and C. Hadlai Hull
1
I 1970 Statistical Package for the Social Sciences.
! New York: McGraw-Hill.
130
Orzack, L. H.
1959 "Work as a Central Life Interest of
Professionals." Social Problems J: 125-132.
Parsons, Talcott and A. M. Henderson (eds.)
1947 Max Weber: The Social and Economic
Organization. New York: Free Press.
Pearlin, L. L.
1962 "Alienation from Work: A Study of Nursing
Personnel." American Sociological Review 27
(June): 314-326.
Pelz, Donald C. and Frank M. Andrews
1966 Scientists in Organizations: Productive
Climates for Research and Development.
New York: John Wiley.
Pepitone, Albert
1964 Attraction and Hostility. New York:
Atherton Press.
Presthus, Robert
1962 The Organizational Society. New York: Knopf
Proshansky, Harold and Bernard Seidenberg (eds.)
1966 Basic Studies in Social Psychology. New York
Holt, Rinehart and Winston.
Reiss, Albert J., Jr.
1961 Occupations and Social Status. Glencoe,
Illinois: Free Press.
131
;Reissman, Leonard
1949 "A Study of Role Conceptions in Bureaucracy
Social Forces 27: 305-310.
Rizzo, John, Sidney I. Lirtzman, and Robert J. House
1970 "Role Conflict and Ambiguity in Complex
Organizations." Administrative Science
Quarterly 15 (June): 150-163.
Roethlisberger, F. J. and W. J. Dickson
1964 Management and the Worker. (Chapter 5):'
73-74. Cambridge: Harvard University.
>Rose, Arnold Marshall (ed.)
j 1962 Human Behavior and Social Processes: An
Interactionist Approach. Boston: Houghton
I Mifflin.
|Roy, Donald
I 1952 "Quota Restriction and Goldbricking in a
i
Machine Shop." American Journal of
| Sociology 57 (March): 427-442.
!Schachter, S., N. Ellertson, D. McBride, and D. Gregory
i 1951 "An Experimental Study of Cohesiveness and
Productivity." Human Relations 4: 229-238
i
'Seashore, Stanley E.
j
| 1954 Group Cohesiveness in the Industrial Work
! Group. Ann Arbor Survey Research Center:
University of Michigan.
132
Secord, Paul P. and Carl W. Backman
1964 Social Psychology. New York: McGraw-Hill.
Shaw, M. E.
1954 "Group Structure and the Behavior of
Individuals in Small Groups." Journal
of Psychology 38: 139-149-
Sherif, Muzafer
1967 Social Interaction. Chicago: Aldine
Shibutani, Tametsu
1961 Society and Personality. Englewood Cliffs:
Prentice-Hall.
1955 "Reference Groups as Perspectives." American
Journal of Sociology 60: 562-569.
Steiner, Ivan D. and Martin Pishbein (eds.)
1965 Current Studies in Social Psychology. New
York: Holt, Rinehart and Winston.
Stinchecombe, Arthur L.
1968 Constructing Social Theories. Robert K.
Merton (ed.) New York: Harcourt, Brace
and World.
1959 "Bureaucratic and Craft Administration of
Production: A Comparative Study."
Administrative Science Quarterly 4 (September):
168-187.
133
Strauss, George
1968 "Personality Versus Organizational
Hypothesis." 93-104. Robert Dubin (ed.)
Human Relations in Administration. Englewood
Cliffs: Prentice-Hall.
Tannenbaum, A. S.
1968 Social Psychology of the Work Organization.
California: Wadsworth.
Thibaut, John W. and H. H. Kelley
1959 The Social Psychology of Groups. New York:
John Wiley.
Thielbar, Gerald
1970 "Localism-Cosmopolitanism: Prolegomenon to a
Theory of Social Participation."
Sociological Quarterly 11: 243-253.
Trow, D. B.
1957 "Autonomy and Job Satisfaction in Task-
Oriented Groups." Journal of Abnormal
Psychology 54: 204-209.
Udy, Stanley, Jr.
1959 "Bureaucracy and Rationality in Weber's
Organization Theory: An Empirical Study."
American Sociological Review 24 (December):
791-795.
134
Vollmer, H. M. and Donald L. Mills
1966 Professionalization. Englewood Cliffs:
Prentice-Hall.
Wallace, Walter L.
1969 Sociological Theory: An Introduction.
Chicago: Aldine.
Wilensky, Harold L. and Charles N. Lebeaux
1958 Industrial Society and Social Welfare. 48-49*
New York: Russell Sage Foundation.
1957 "Human Relations in the Workplace: An
Appraisal of Some Recent Research." 25-50.
Conrad M. Arensberg, et al., (eds.)
Research in Industrial Human Relations .
New York: Harper and Row.
Wilier, David
1967 Scientific Sociology: Theory and Method.
Englewood Cliffs: Prentice-Hall.
Zajonc, Robert Boleslaw
1966 Social Psychology: An Experimental Approach.
California: Brooks/Cole.
APPENDIX
Footnote
1. When the local-cosmopolitan distinction is collapsed
into a single category, non-bureaucrats, the relation
ship between education and reference group increases
(gamma = 44). This seems to indicate, at least in this
instance, that the local-cosmopolitan distinction adds
little in the way of understanding. The relationship
between education and the bureaucratic-non-bureaucratic
distinction is displayed below.
RELATIONSHIP BETWEEN EDUCATION AND REFERENCE GROUP
ORIENTATION AMONG 260 MEMBERS OF FOUR PROFESSIONAL
AND SEMI-PROFESSIONAL ORGANIZATIONS WITH REFERENCE
GROUP ORIENTATION TREATED AS BUREAUCRATS-NON-
BUREAUCRATS.
Education
Low
Medium
High
Reference Group
Bureaucratic Non-Bureaucratic Total
N (%) N (%) N (, o
51 (60.7) 33 (39.3) 84 (100.0)
39 (50.6) 38 (49.3) 77 ( 99-9)
28 (31.8) 60 (68.2) 88 (100.0)
Total 118 131 249
gamma =.44.
Missing observations = 11.
CORRELATION MATRIX
Kendall ' s Tau
2 3 4
5 6
7
8
9
10
11
Attachment .12 -.05 -.22 -.10 .01 -.02 .14 .04
-.29
o
on
•
l
Reference Group .14
-.09
-.08 .24
.13 .05 -
.04
CT\
O
«
1
-.13
Occupation
.37
.21 .56 .46 .18 -.02 .01 -.06
Rewards .28 .34 .23
.12 .02 .11 .05
Employment Length
.09
.16 .08 -.14 .02 .11
Education • 36 .12 -
.05
.00
i
•
o
-a
Sex
.13 .13
-.02 .01
Tension
.05 -.27 -.17
Satisfaction
-.07 -.07
Freedom .41
Job Clarity
H
U)
L _ .
DEMOGRAPHIC PROFILES OF FOUR PROFESSIONAL AND SEMI-PROFESSIONAL
ORGANIZATIONS (N = 260)
Semi-Autonomous
Survey
Research
Organization
(N = 47)
Autonomous
Survey
Research
Organization
(N = 62)
Alcoholism
Rehabilitation
Organization
(N = 59)
Commercial
Research
Organization
(N = 92)
Per Cent with '
Official Authority
19.5 14.5
18.6 25.6
Per Cent Male 60.0 49.2 49.2 72.7
Formal Training:
12 years 10.0 8.1
11.9
14.4
1-3 years college 20.0
35.5 27.1
35.6
4 years college
12.5
8.1 22.0
13.3 ■
years college
57.5
48.4 39.0 36.7
Per Cent with
Specialized Training
for Occupation 70.0 73.1 65.5 58.9
DEMOGRAPHIC PROFILES— Continued
Semi-Autonomous Autonomous Alcoholism Commercial
Survey Survey /'Rehabilitation Research
Research Research Organization Organization
Organization Organization
(N = 47) (N = 62) (N = 59) (N = 92)
Professional Rank:
1. Clerical
22.5 47.5
25.4 17.8
2. Service and
Technical
12.5 8.5
0.0 27.8
3.
Semi-
Professional 10.0
8.5
39.0
23.3
4. Managerial and
Professional 55-0 35.6 35.6
31.1
1. Custodial staff excluded from the survey.
2. Clerical staff, custodial staff, and students excluded.
3. Category includes lab and research assistants, technicians, etc.
4. Category includes nurses, social workers, counselors, etc.
5. Category includes those trained for supervision (most of whom are
professionally trained as well), and professional scientists (including
both M.D.'s and Ph.D.'s).
( —1
(jO
V O

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

An empirical test related to deviance theory through the use of dyadic interaction with the labelled retarded

PDF

The Deterrent Effect Of Criminal Justice Agencies On Felony Offenses

PDF

The Effects Of Generation, Religion, And Sex On The Relationship Of Family Vertical Solidarity And Mental Health In Lebanon

PDF

Bam: An Innovative Change Model--Barriers Encountered In The Implementation Of A Classical Research Design To Modify The Behavior And Attitude Of Staff And Inmates In A Correctional Institution

PDF

Structural Factors Affecting Fertility In Large United States Cities

PDF

Cultural Continuity And Discontinuity In Adolescent Socialization

PDF

Resource allocation among organizations within urban communities

PDF

Social Stratification In A Selected Community

PDF

The Effects Of Social Security Programs On Fertility Levels

PDF

Anomy And Verbal Behavior In Task-Oriented Small Groups: An Exploratory Study

PDF

Delinquency As A Function Of Intrafamily Relationships

PDF

Selected Social Psychological Factors Related To Viewers Of Television Programs

PDF

Factors In The Conviction Of Law Violators: The Drinking Driver

PDF

Studies In The Influence Of The 'Commedia Dell'Arte' On English Drama: 1605-1800

PDF

The Radical Movement Of The 1960'S: A Sectarian Type Of The Civil Religion

PDF

The Selfish Self: A Social Psychological Study Of Social Character

PDF

Southern Congressmen And Welfare Policy In The 1960'S: A Case Study Of Redistributive Politics

PDF

Some Social Factors Affecting The Power Structure And Status Of A Professional Association In Reference To Social Work

PDF

Passage Through The Expressive Student Subculture: The Initiation And Cessation Of Marijuana Use

PDF

In church with my ancestors: The changing shape of religious memory in the Republic of Armenia and the North American diaspora

Asset Metadata

Creator Fry, Lincoln Jacob (author)

Core Title Social-Psychological Factors In Organizational Attachment

Degree Doctor of Philosophy

Degree Program Sociology

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag OAI-PMH Harvest,Sociology, general

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Miller, Jon P. (committee chair), Kobrin, Solomon (committee member), Merendino, Salvatore (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-775838

Unique identifier UC11364363

Identifier 7304912.pdf (filename),usctheses-c18-775838 (legacy record id)

Legacy Identifier 7304912

Dmrecord 775838

Document Type Dissertation

Rights Fry, Lincoln Jacob

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA