Dimensions of Organizational Commitment

Dimensions of Organizational Commitment
Author(s): Helen P. Gouldner

Source: Administrative Science Quarterly, Vol. 4, No. 4 (Mar., 1960), pp. 468-490
Published by: Sage Publications, Inc. on behalf of the Johnson Graduate School of Management,
Cornell University
Stable URL: http://www.jstor.org/stable/2390769 .
Accessed: 14/06/2014 18:18
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
Sage Publications, Inc. and Johnson Graduate School of Management, Cornell University are collaborating
with JSTOR to digitize, preserve and extend access to Administrative Science Quarterly.
http://www.jstor.org
This content downloaded from 195.34.79.192 on Sat, 14 Jun 2014 18:19:00 PM

All use subject to JSTOR Terms and Conditions
HelenP. Gouldner
Dimensions of Organizational
Commitment
The author reports on a factor analytic study which distinguishes

different dimensions of organizational commitment among the mem-
bers of a voluntary association. Among the forms of organizational
commitment studied were: cosmopolitan integration-the degree to
which the individual is active in and feels himself a part of the varying
levels of a particular organization, and is active, as well, in other
organizations; and organizational introjection-the degree to which
the individual's "ideal" self-image includes a number of organizational-
ly approved qualities and values. In addition to the factors of organi-
zational commitment, two other factors were identified as reflecting
commitment to two specific values of the organization. The emergence
of these lends support to a basic hypothesis of the study, namely, that
commitment to the specific values of an organization is distinct from
commitment to the organization as a whole.1
Helen P. Gouldner is associate professor and chairman, Department
of Sociology, Lindenwood College.
IT has of course long been obvious that the members of an associ-

ation differ greatly in the degree of their commitment to the
'The author acknowledges with great pleasure the support of the Social Science
Research Council, many helpful suggestions from Leonard Broom, Richard T.
Morris, and Ralph H. Turner, interviewing assistance from Mary Bob Chesler,
assistance in computations from Kern Dickman, and aid in all phases of the research
from Alvin W. Gouldner.

ORGANIZATIONAL COMMITMENT 469
organization. At both poles variations in the degree of commit-
ment have been problems. Attention has been directed, on the
one hand, to the fanatical zeal of the political and religious secta-
rians and, on the other, to the apathy of the members of some
unions or fraternal and civic organizations.2 These extremes in a
democratic culture that regards them both with suspicion have
been of concern both to social scientists and citizens. Partly because
of this interest in the degree of organizational commitment, how-
ever, less attention has been paid to the fact that members differ
also in the form of their commitment. It is the main hypothesis of
this paper that organizational commitment is not a homogeneous
and unidimensional variable but is, instead, a multidimensional
phenomenon.
Along these lines, Herbert Simon was one of the first to suggest
that commitment to an organization as a whole is distinguishable
from a commitment to its specific values, policies, or goals.3Recent-
ly, James Coleman and others have made a similar distinction
between those who are "idea-oriented," forming attachments on
the basis of values and ideals, and those who are "group-oriented,"
forming attachments through group loyalties.4 This distinction
between over-all commitment and commitment to particular goals
is important for organizational analysis, because it suggests that
those who are committed only to the goals of an organization may
resign or withdraw should these be changed.
The objective of our research, then, was to identify some of the
components or dimensions of organizational commitments and to
begin to explore the relations among them. It seems obvious that
the problem of organizational commitment involves an issue of
2For studies of extreme differences in degree of commitment compare Philip
Selznick, The Organization Weapon: A Study of Bolshevik Strategy and Tactics
(New York, 1952), pp. 25-29, and Scott A. Greer, "The Participation of Ethnic
Minorities in the Labor Unions of Los Angeles County" (Ph.D. dissertation, Univer-
sity of California, Los Angeles, 1952). See also Arnold M. Rose, Union Solidarity
(Minneapolis, 1952).
"H. A. Simon, D. W. Smithburg, and V. A. Thompson, Public Administration
(New York, 1950), pp. 94-100.
4James S. Coleman, Multidimensional Scale Analysis, American Journal of Sociol-
ogy, 63 (1957), 255.
5As the term commitment is used here, it refers to those kinds of constraints which
are generated by the actor's own motivations, orientations, and behaviors.

470 ADMINISTRATIVE SCIENCE QUARTERLY
basic sociological interest, concerned as it is with the diverse ways

in which individuals attach themselves to groups. In one way or
another many investigations, including research in group cohesive-
ness, integration, influence, and morale either are implicitly
premised on, or explicitly postulate assumptions about organiza-
tional commitment and use this concept as a dependent, indepen-
dent, or intervening variable. It seems likely, therefore, that the
identification of distinct forms of organizational commitment will
increase our knowledge about some of the basic group processes.
For example, differences in the kind (as well as the amount) of
commitment an individual has to a group are likely to affect the
degree and kind of influence that the group can exert upon him.
Similarly, the identification of different forms of organizational
commitment should also contribute to our understanding of such
group attributes as cohesion. The studies of Festinger, Schachter,
Back, Emerson, and others" assume that group cohesion is a
unidimensional variable. However, if the individual's commitment
to the group is one force generating group cohesion, and if there
are several distinctive forms of commitment to a group, it is likely
that there are several distinguishable forms of group cohesion.
SETTING OF THE STUDY
The study reported here was carried out among members of the
Los Angeles League of Women Voters. The League is a large and
nationally organized women's voluntary association devoted to
''promoting political responsibility through informed and active
participation of citizens in government." The organization of the
League is modeled after the government, with national, state, and
local groups. League activities include providing nonpartisan,
factual information to voters on registration, voting procedures,
candidates, public officials, and issues. The members study and
"For a review of several studies in this area see L. Festinger, Informal Social Com-
munication, Psychological Review, 57 (1950), 271-282. See also Richard M. Emerson,
Deviation and Rejection: An Experimental Replication, American Sociological
Review, 19(1954), 688-693. For research and discussions on the problem of group
cohesiveness see K. W. Back, Influence through Social Communication, Journal of
Abnormal and Social Psychology, 46 (1951), 9-32; Dorwin Cartwright and Alvin
Zander, eds., Group Dynamics (Evanston, 1953), especially pp. 87-88; N. Gross and
W. E. Martin, On Group Cohesiveness, and Stanley Schachter's Comment, American
Journal of Sociology, 57 (1952), 546-562.

sometimes take action on various issues and problems in national,
state, and local government.
After several months of field work, which included observation
at meetings, interviews, and a study of League documents, a work-
ing sample was selected by taking every nth name from an alpha-
betical list of members provided by League leaders. From this
working sample, sixty interviews were completed and these con-
stitute the main body of data for the following analysis. The data
also include responses to mailed questionnaires, which had been
filled out and were collected at the time of the interview. Inter-
views with each member ranged from about two to four hours.
The sample is largely composed of highly educated, middle-to-
upper-class married women with two or three children of school
age.
THE HYPOTHESIZED DIMENSIONS
In constructing commitment measures the focus was not only
on items which would test the degree of commitment to the
organization but also on items which might reveal different forms
or dimensions of this commitment. A major distinction was made
between commitment to the organization as a whole and commit-
ment to specific values of the organization.7 Sample items on each
of these major dimensions are listed below.
Commitment to the Total Organization
One set of items sought to reveal "integration," that is, the
degree to which the individual feels a part of and has a sense of
belonging to the varying levels of the organization. These items
included the following:
Membersof the League differ in their sense of closeness to different
partsof the organization.Do you feel a real part of... (1) Your branch
group? (2) the city League as a whole? (3) the state League? (4) the
national League?
Categories for each of these items ranged on a five-point scale
from "not at all" to "very much."
A further item was developed to measure "introjection." This
item is limited in the present study to an indication of the degree
7Several other commitment variables, including involvement with organizational
subgroups have not been included in the present analysis. See Simon et al., op. cit.

to which the individual's ideal self-image includes a variety of

League-approved qualities and values. More specifically, it is a
reflection of the respondent's desire to mold her personality, char-
acter, or behavior in terms of general League ideals, values, and
norms. It is, in one part, a measure of the degree to which the
organization is an "identification" reference group.8 The item was
stated as follows:
Try for a minute to step outside yourself, to... look at yourself....
What kind of a League memberwould (respondent'sname) like to be?
What do you think you would say about yourself?
Responses were coded on a three-point scale ranging from "low
introjection," little or no indication of desire to mold personal
character to League values and no mention of specific desired
changes, to "high introjection," indication of such desire and
mention of two or more specific desired changes in self.
In addition to the items specifically designed to tap integration
and introjection, a number of other items related to organizational
commitment were developed. These included various measures of
activity and participation, highest office held, degree of enthusiasm
about the League, degree of confidence that the League would
achieve its over-all goals, preference for the League over other
organizations, and readiness to give up time spent in other organi-
zations or in social life outside the family to maintain present
level of activity in the League.
Commitment to Specific Organizational Values
On the hypothesis that commitment to a specific organizational
value is often independent of commitment to the organization as
a whole, and, further, that commitment to one organizational value
is sometimes independent of commitment to another, three values
specific to the League were singled out for special study. These
were: (1) cross-sectional membership-the desirability of getting
members from different income brackets, different political phi-
losophies, different levels of education, and so forth; (2) political
party responsibility-the desirability of including individuals con-
8Ralph H. Turner has isolated several types of reference groups including the
identification group which is defined as the source of the individual's major per-
spectives and values. See his Role-Taking, Role Standpoint, and Reference-Group
Behavior, American Journal of Sociology, 61 (1956), &27-328.

sistently supporting one political party; (3) soliciting funds from
the community-the desirability of going to the community for
financial aid to supplement dues and member contributions.9 The
major interest here is not these particular values (indeed, others
might have been chosen), but rather whether commitment to any
specific organizational value is independent of other forms of
organizational involvement.
These values were presented to the respondents in the form of
an idea (or principle or policy) that was preferable or desirable for
the organization, which differs from asking them for their personal
preference. On the whole, however, the data suggest a relationship
between organization preference and personal preference (i.e., if
a respondent preferred the value for the organization, she also
often preferred it for herself). Nevertheless, a distinction between
organizational and personal preference still proved useful. With-
out this distinction, for example, many of the respondents would
have indicated a preference for fund-raising events over soliciting
funds, had we asked them only for a personal choice, even though
in principle they did approve of the League soliciting funds as an
organization. As one woman commented, "It's all right as long as
I don't have to do it."
The items measuring value commitment were also designed to
test the degree of private commitment to the value concerned, as
opposed to getting at a more limited public commitment.10 (An
individual who is privately committed to a value will tend to main-
tain a consistent position in favor of this value in varying situations
and groups-even where the majority of a group is opposed to the
value-whereas an individual who is only publicly committed may
change her opinion according to the position of the group she
happens to be in.)"
The value commitment score was derived as follows: After some
discussion, the respondent was asked to sum up the position she
9Each of these are justified preferences. For example, soliciting funds from the
community is preferred over fund-raising events (bazaars, teas, theater parties)
"because statistics have shown that returns are greater per unit of time spent" and
so forth. This conforms to the definition of values currently in use.
:0A brief review of the literature on this distinction is found in Harold H. Kelley
and Edmund H. Volkart, The Resistance to Change of Group-Anchored Attitudes,
American Sociological Review, 17 (1952), 455-456.
:1Cf. Turner's discussion of the internalization of norms, op. cit., p. 3,26.

had taken on the value in actual League discussions (or would

take if the value were being discussed) on a six-point scale ranging
from "strongly opposed" to "very strongly in favor." A series of
questions probing for private doubts followed and the respondent
was then asked to place herself in a hypothetical League meeting
where the value was being discussed. In each case the group in
this hypothetical situation was opposed to the position the
respondent had initially taken on the value. Each respondent was
then asked to choose again from the same set of categories that
which would best sum up their private feelings in this situation.
The commitment score was computed by comparing the answers
in the first and second ratings, the resulting scale ranging from
"opposed and did not change positions" to "very strongly in favor
and did not change positions." This set of questions was repeated
for each of the three values, giving three separate value
commitment scores.12
Other items thought to be related to value commitment were
developed. These included questions for each value on personal
salience, private doubts, number of qualifications, discussion of
value with other League members, use of standard League argu-
ments in support of value position, and objectivity in discussing
the value. "Objectivity" (the weighing of pros and cons or some
posture of "getting the facts first") is a highly pervasive League
value, which affects the ability to become committed to any
specific value or policy.
THE FACTOR ANALYSIS
After the data were collected and coded, factor analysis was used
as a tool for refining and developing the hypothesized dimensions
and as a test of their independence.13 In all, forty-one items were
'"There is some evidence to suggest that, with all the obvious drawbacks of this
instrument, it did tap a greater degree of commitment than would have been the
case had we used a single question and had not forced the respondent to attend
to alternative reference groups. For one thing, many respondents did change their
position. Those with many doubts and qualifications were likely to score lower the
second time. There is evidence, too, that these portions of the interview were some-
what anxiety provoking, which might be expected in cases where there is disparity
between "private" and "public" opinion.
"3Factoranalysis is a method for identifying a multiplicity of items which cluster
together. It is also useful as a scanning device enabling the researcher to detect

intercorrelated using Pearson's product-moment correlation. Seven
factors were extracted by Thurstone's centroid method and were
rotated to both orthogonal and oblique simple structure, using
the "Quartimax" and "Oblimax" methods of rotation respective-
ly.14 (Rather than being graphic trial-and-errormethods, these are
analytical methods that have been adapted for use on electronic
computers.) In order to obtain the clearest solution possible, the
oblique analysis was carried several steps further than is ordinarily
done, to secure "primary"factors rather than stopping the analysis
with the "reference vectors."'15The results indicated five inter-
pretable orthogonal factors and six interpretable oblique factors
(see Appendix, Tables 1 and 2).
The factors are presented and interpreted below. (Only those
items with loadings above .30 are listed.) Inasmuch as many of the
factors include both negative and positive loadings, the response
category corresponding to the direction of the loading of any par-
ticular item is indicated with two asterisks to facilitate interpreta-
tion. Further, while it is often possible to interpret factors at either
end of a positive-negative continuum, interpretation is limited
here to only one "side" of the factor, usually the "positive" side.
Factor 1 reflects the degree of integration into the varying levels
and activities of the organization, particularly the middle organi-
zational levels. With a loading of .90, item 30 ("Do you feel a real
part of the city League as a whole?") is a good measure of this
factor. A sense of closeness, or feeling of belonging to other levels
dimensions other than those he had planned to include in a given domain or sample
of items. Furthermore, by indicating the items most highly loaded with the under-
lying dimension, it may provide a more parsimonious measure of that dimension,
allowing fewer items to be used as a measure of the dimension in later research.
'4Orthogonal dimensions are independent of and only randomly related to each
other; oblique dimensions, on the other hand, may or may not be correlated. In an
oblique rotation there is no attempt to maintain strict independence among the
factors.
The seven centroid factors represented 54 per cent of the common variance in
the total set of items. For a description of the "Quartimax" and "Oblimax" methods
of rotation see J. 0. Neuhaus and C. F. Wrigley, The Quartimax Method: An
Analytic Approach to Orthogonal Simple Structure, British Journal of Statistical
Psychology, 7 (1954), 81-91, and D. R. Saunders and C. Pinska, Analytic Rotation to
Simple Structure: Extension to an Oblique Solution, Research Bulletin, Princeton
Educational Testing Service, 1954.
'rSee Raymond B. Cattell, Factor Analysis (New York, 1952), pp. 219-2-23, 426.

Factor1 (Orthogonal):Integration
Load- Item
ings No.
.90 30 Feel a real part of city League: notat all. . .verymuch**

.82 43 Extent of activity: inactive.. .extremely active**
.77 31 Feel a real part of state League: notat all. . .verymuch**
.77 17 Attendance, League Day: never.. .everymeeting**
.74 29 Feel a real part of branch group: not at all. . .verymuch**
.72 32 Feel a real part of national League: notat all. . .verymuch**
.72 16 Attendance, branch meetings: never.. .everymeeting**
.67 41 Offices or committees: no officesor committee work. . .city board
member* *
.66 26 Enthusiasm about membership: a little. . .extremely**
.62 47 Introjection: low. . .high**
.57 52 Discussed or heard discussed cross-sectional membership: no,
yes* *
.54 20 Readiness to give up time in other organizations: wouldnot
give up. . .wouldreadilygive up**
.52 C30 Awareness of cross-sectional membership as League value:
no awareness.. .stronglyaware**
.46 27 Would rather belong to League than other women's organiza-
tions: no,yes **
.41 24 Preference for group discussions over outside speakers: no,
yes* *
.41 76 Commitment score on soliciting funds from the community:
opposed.. .verystronglyinfavor* *
.38 72 Discussed or heard discussed soliciting funds: no,yes**
.37 68 Arguments on soliciting funds: extra-Leaguearguments.. .
standardLeaguearguments**
.34 48 Arguments on cross-sectional membership: extra-League
arguments.. .standardLeaguearguments* *
.33 77 Personal importance, soliciting funds: none. . .a greatdeal**
of the organization-the state League, branch group, and national

League-have somewhat lower, but nonetheless high loadings.
The high loadings of items on attendance at meetings and general
level of activity further support the hypothesis that this dimension
is indicative of integration. This factor describes the "active minor-
ity," those members who often referred to their League activities
as a career or a profession from which they withdrew only, as one
woman put it, "to take a sabbatical leave."

The oblique solution has tended to divide this general ortho-
gonal factor into two dimensions. Oblique Factor A suggests what
has been termed "cosmopolitan integration." Oblique Factor B,
on the other hand, reflects "organizational introjection."
The items for Factor A are as follows:
FactorA (Oblique):Cosmopolitan Integration

Load- Item
ings No.
.91 32 Feel a real part of national League: notat all. . .verymuch**

.90 31 Feel a real part of state League: notat all. . .verymuch**
.84 30 Feel a real part of city League: notat all. . .verymuch**
.84 16 Attendance, branch meetings: never.. .everymeeting**
.81 43 Extentof activity:inactive.. .extremely active**
.73 17 Attendance, League Day: never.. .everymeeting**
.67 52 Discussed cross-sectionalmembership: no,yes**
.55 19 No. of other organizations in which respondent holds office:
none. . . threeor-more**
.53 29 Feel a real part of branch group: notat all. . .verymuch**
.52 24 Preference for group discussions over outside speakers: no,
yes* *
.45 27 Would rather belong to League than other women's organiza-
tions:no,yes* *
.46 18 No. of other organizations in which respondent is active:
none. . four ormore**
.44 20 Readiness to give up time in other organizations: would not
give up. . . wouldreadilygive up* *
.42 26 Enthusiasm about membership: a little. . extremely**
As with Factor 1 this oblique factor reflects integration. A

respondent with a high score feels herself to be an integral part of
the organization and is active in it. The factor contains both
cognitive and affective forms of commitment to the organization
as a whole.
In contrast to Factor 1, however, Factor A seems to indicate not
only a sense of integration into but a "cosmopolitan" orientation
toward the organization. While the factor consists of integration
or a sense of closeness to all levels of the organization, integration
with the higher, national and state levels is dominant. The factor
describes a woman who identifies with the League not merely in

terms of its local units but as a national organization or social

movement with society-wide goals."6Even though she prefers the
League, and is both loyal to and active in it, she is also active in
other organizations.
Factor B (Oblique): Introjection

Load- Item
ings No.
.82 47 Introjection: low. . .high**

.58 C30 Awareness of cross-sectional membership as League value:
no awareness. . .strongly aware* *
.56 41 Offices or committees: no officesor committeework. . .city board
member**
.45 26 Enthusiasm about membership: a little. . .extremely**
.45 29 Feel a real part of branch group: a little. . .verymuch**
.39 21 Readiness to give up time spent in social life outside the
family: would not give up. . .would readily give up* *
.38 67 Personal importance, political party responsibility: none. . .
a great deal* *
.33 69 Number of qualifications, soliciting funds from the com-
munity: none. . four or more* *
The highest ranking item on Factor B measures "organizational

introjection'.17 While Factor A, cosmopolitan integration, reflects
the degree to which respondents feel themselves to be a part of the
organization, this introjective dimension, Factor B, consists of the
degree to which the respondents feel the organization to be a part
of themselves. It is an affective and evaluative dimension. An
individual with a high score on this factor judges, guides, and
evaluates her behavior from the standpoint of a variety of League
values and norms.
Factor B also indicates that introjection is related to holding
"'A. W. Gouldner has analyzed the roles of "locals" and "cosmopolitans" in various
organizational settings. One of the specific dimensions he used to define "cosmo-
politans" is the use of an outer reference group orientation, which in this case may
be taken to mean orientation to the League as a national organization, rather than
exclusive orientation to local units, and to other organizations besides the League.
See Cosmopolitans and Locals: Toward an Analysis of Latent Social Roles-I,
Administrative Science Quarterly, 2 (1957), 281-306.
"7Thesixth orthogonal factor is a residual (see Table 1), but is similar to Factor B
in that the introjection item had one of the highest loadings (-.39).

office in the organization. This suggests that the more specific the
organizational expectations and the greater the pressure for con-
formity to these expectations (as is true for officeholders),the more
the expectations become introjected.18 Officers act out specific
organizationally approved patterns of behavior, which are rewarded
and become, in turn, part of the self-image. An alternative but not
contradictory hypothesis is that members with a high level of
organizational introjection will more often volunteer or be drafted
for offices and important committees. Here, again, the performance
of specific roles would tend to reinforce organizational introjection.
Let us now examine some of the other factors:
Commitment to a Specific Value: Cross-sectional Membership*

Factor2 FactorC Item
Orthogonal Oblique No.
Rank Load- Load- Rank
ings ings
1. .74 .81 2 56 Commitment score on cross-sectional
membership: opposed.. .strongly in fa-
vor* *
2. .72 .82 1 57 Personal importance, cross-sectional
membership: none. . .a great deal* *
3. -.64 -.76 3 49 No. of qualifications, cross-sectional
membership: none* .*. . four or more
4. -.53 -.69 4 55 Private doubts, cross-sectional member-
ship: not at all* . . .pretty definitely
5. .38 .15 8 47 Introjection: low. . .high**
6. .31 .20 7 48 Arguments on cross-sectional member-
ship: extra-Leaguearguments.. .standard
League arguments**
10. .23 .32 5 C34 Uses objectivity in discussing political
party responsibility: no,yes**
11. -.22 -.31 6 46 Emphasis on friendships in the League
vs. emphasis on educational aims:
friends**. . .educationalaims
*In the presentation of the remaining factors all factor loadings above .30 on either
the oblique or orthogonal solutions are presented for comparison.
:8Cf. Arnold M. Rose, The Adequacy of Women's Expectations for Adult Roles,
Social Forces, 30 (Oct. 1951), 69-77, and James G. March, Group Norms and the
Active Minority, American Sociological Review, 19 (1954), 738.

Orthogonal Factor 2 and Oblique Factor C involve a commit-

ment to the specific value or policy of cross-sectional membership
in the League. Respondents with a high score on this factor took a
positive position on the value and did not change their position
under pressure; they placed high personal salience on the value,
gave few qualifications and had few private doubts about their
position. The emergence of this as an orthogonal factor supports
the hypothesis that commitment to the values of an organization
may be independent of commitment to the organization as a whole.
This factor indicates commitment to another specific League
Commitment to a Specific Value: Political Party Responsibility

Factor3 FactorD Item
ings ings
1. -.66 -.74 1 65 Private doubts, political party responsi-
bility: not at all** . . .definitely
2. -.62 -.73 3 59 No. of qualifications, political party
responsibility: none*. . .four or more
3. .62 .63 4 67 Personal importance, political party
responsibility: none. . .a great deal**
4. .61 .73 2 66 Commitment score on political party
responsibility: opposed.. .stronglyin fa-
vor* *
5. .40 .40 5 C30 Awareness of cross-sectional member-
ship as League value: none. . strongly
aware* *
6. .37 .40 6 58 Arguments on political party responsi-
bility: extra-Leaguearguments.. .standard
League arguments* *
7. -.36 -.36 8 27 Would rather belong to the League than
other women's organizations: no,* * yes
8. -.35 -.40 7 28 Confidence in ability of the League to
achieve its goals: a little **. . .extremely
9. -.30 -.35 9 20 Readiness to give up time in other or-
ganizations: would not give up* *. .
wouldreadilygive up

value, responsibility to a political party. It loads heavily questions
on private doubts, number of qualifications, personal importance,
and the commitment score on political party responsibility.'9 The
emergence of this as an orthogonal factor again supports the
hypothesis that commitment to an organizational value is inde-
pendent of commitment to the total organization, as well as of
other organizational values.
Exposure to Organizational Culture

Factor4 FactorE Item
RankLoad- Load-Rank
ings ings
1. .63 .68 2 62 Discussed or heard discussedpolitical
party responsibility:no,yes**
2. .62 .70 1 C25 Uses objectivity in discussing cross-
sectionalmembership:no,yes**
3. .61 .61 3 58 Type of argumentson political party
responsibility:extra-Leaguearguments.
. .
standardLeaguearguments* *
4. .52 .47 6 68 Type of argumentson solicitingfunds
from the community:extra-League argu-
ments. .. standard Leaguearguments**
5. .51 .57 5 72 Discussedor heard discussedsoliciting
fundsfrom the community:no,yes**
6. .51 .57 4 48 Type of arguments on cross-sectional
membership:extra-League . .
arguments.
standardLeaguearguments* *
7. .38 .32 7 52 Discussedor heard discussedcross-sec-
tional membership:no,yes* *
This factor involves degree of exposure to League "culture,"

rather than a particular dimension of commitment. To the extent
that the term "socialization" refers to the use of skills and group
:'Other items on this factor suggest that commitment to this particular value-
political party responsibility-is a function of experiencesoutside the League; thus
for example, those who are committed to the value do not prefer the League over
other organizations.For those holding this value, it appears the League is not an
identification group.

values, as well as to the internalization of these skills and values,

the factor may also be thought to represent a dimension of
organizational socialization.
The isolation of this constellation of items identifies an impor-
tant dimension for further research. Knowledge and awareness of
group norms and values vary among group members and some
problems for further analysis are pointed up by Merton's question
"In which ways does group structure affect the distribution of
knowledge about the values and norms actually held by members
of the group?"20
Exclusiveness of Organizational Membership

Factor5 FactorF Item
ings ings
1. -.65 -.93 1 18 Number of other organizations in which
respondent is active: none"*. . four or
more
2. -.60 -.90 2 19 Number of other organizations in which
respondent holds office: none**. . .tthree
or more
3. .54 .47 4 76 Commitment score on soliciting funds
from the community: opposed.. .strongly
infavor* *
4. .49 .61 3 25 Preference for group discussion over
presented material: no,yes* *
5. .47 .33 5 24 Preference for group discussion over out-
side speakers: no,yes**
- .06 .32 6 47 Introjection: low. . .Ihigh**
This last factor loads heavily items on the number of organiza-

tions in which the respondent is active. The factor is mainly
indicative of a tendency toward exclusive membership in the
League. Items on this factor suggest that multiple organizational
membership is negatively related to commitment to the League
values of soliciting funds and group discussion. Both of these are
low-consensus values and both directly affect each individual mem-
ORobert K. Merton, Social Theory and Social Structure (Glencoe, Ill., 1957), p. 337.

ber. Two related hypotheses may account for the item loadings on
these values in this factor: (1) Those deeply involved in other
organizations may tend to view their League activities in limited
and utilitarian ways; they may prefer to get as much as they can
from the League without the additional time and energy required
by such activities as soliciting funds and group discussion. (9)
Other organizations may serve as alternative reference groups
which inhibit commitment to such values. Indeed, contrary prac-
tices in other organizations may be used as a justification for lack
of commitment, particularly where lack of commitment is a func-
tion of personal distaste in implementing the value. The majority
of women personally disliked to solicit funds, and many women
found group discussion methods frustrating and confessed they
sometimes preferred the direct presentation of factual material.
In such cases the individual can more legitimately justify her lack
of commitment by pointing out alternative practices in other
organizations than by expressions of personal dislike. By doing so
she is protected from the common rejoinder, "Well, if you were
really sold on the League, you wouldn't have any trouble selling
the League."
In any event, the qualitative data revealed a substantially large
group of women who seemed to typify what has been called the
"outsiders" and who belonged to a number of organizations.21
They are "in but not of} the organization. They approved of the
League and its activities but their primary loyalties were to other
organizations. These included the "political party actives," who
joined the League specifically to learn more about government;
community leaders concerned about schools or civil liberties, who
considered the League as an action organization that would help
them achieve rather specific desired ends; and women who "took"
the League rather than adult education or university extension
courses. Such cases involve a nonexclusive and utilitarian type of
membership where the organization is, in Turner's terms, more
frequently used as an "interaction" or "valuation" reference group
rather than as an "identification" group.22
21Alvin W. Gouldner, Cosmopolitans and Locals: Toward an Analysis of Latent
Social Roles-II, Administrative Science Quarterly, 2(1958), especially pp. 449-450.
22Groups whose members constitute merely conditions to an individual's actions
are defined as "interaction" reference groups; groups "which acquire value to the

INTERCORRELATIONS AMONG THE OBLIQUE

FACTORS
There can, of course, be no question of a correlation among
orthogonal factors since these are, by design, independent of each
other. Ordinarily however, there are correlations among oblique
factors, since these are not constructed so as to be independent of
each other, though they may turn out to be unrelated. In this anal-
ysis it has already been noted that there is a close congruence
between the oblique and orthogonal factors, and consequently
one should not expect that there will be high correlations among
the oblique factors. This is confirmed by Table 3 (see Appendix)
which indicates that correlations among the oblique factors are,
with certain exceptions, relatively low.23
One notable exception is the relation between Factor A, cos-
mopolitan integration, and a number of other dimensions. For
example, cosmopolitan integration is related to Factor B, introjec-
tion; that is, the degree to which individuals feel a part of varying
levels in the organization, particularly the higher levels, is positive-
ly related to the degree to which their self-image incorporates a
variety of organizationally approved qualities and values. This is
consistent with the earlier suggestion that both of these oblique
factors are derived from a splintering of the first very general inte-
gration factor yielded by the orthogonal solution.
The degree to which individuals feel a part of varying and
especially higher levels in the organization (cosmopolitan integra-
tion) is also related to the degree to which they are knowledgeable
about and aware of organizational policies (Factor E, organization-
al exposure) as well as to the degree to which they are committed
to the value of cross-sectionalmembership (Factor C). Finally, there
is a positive relation between Factor A (cosmopolitan integration)
and Factor F (exclusiveness of organizational membership)-if
individuals feel a part of varying and especially higher levels of the
organization, they are likely to belong to few or no other
organizations outside the League.
This last is a rather interesting and unexpected relationship,
individual because the standpoint of his identification groups designates them as
points of reference" might be called "valuation groups." See Turner, op. cit., p. 328.
23These correlations are not based upon correlations among the individuals' factor
scores, but derive directly from correlations among the factors themselves.

for in Factor A (cosmopolitan integration) items on multiple
organizational membership are positively loaded and are thus posi-
tively related to feeling a part of and being active in the League,
even though cosmopolitan integration is, as a whole, shown to
be related to few or no other memberships in organizations
outside the League when measured by Factor F (organizational
exclusiveness).
This suggests that active membership in a number of organiza-
tions outside of the League may have two quite different conse-
quences or implications. On the one hand, for some League women,
such outside activities may supplement and be integrated with
their activities in the League. These women may be those whose
organizational participation and whose leadership in one organi-
zation tend to pull them into a variety of organizations and an
expanding web of community functions. On the other hand, as
indicated in the earlier discussion of Factor F, membership in other
organizations may, for other women, imply a limited and utili-
tarian commitment to the League. The primary organizational
loyalties of these women may be elsewhere and their membership
in the League may be regarded by them as restricted to limited
educational or political purposes.
SUMMARY AND DISCUSSION
This paper reports on a factor analytic study of the dimensions
in organizational commitment among a group of women in the
League of Women Voters. Five orthogonal factors and six (pri-
mary) oblique factors were identified. The constellations of items
were similar in the orthogonal and oblique solutions.
Four of the oblique factors suggest various forms of organiza-
tional commitment.These include: cosmopolitan integration, the
degree to which the individual is active in and feels a part of the
varying levels of the particular organization, especially the higher
levels, and is active as well in other organizations; and organiza-
tional introjection, the degree to which the individual's ideal self-
image includes a variety of organizationally approved qualities
and values. These two oblique factors appear to be a breakdown
of a large, general integration factor obtained in the orthogonal
solution.
Two further commitment factors were identified as reflecting a

commitment to two specific organizational values, cross-sectional

membership and political party responsibility. The emergence of
these two factors lends support to the hypothesis that commitment
to the specific values of an organization is distinct from commit-
ment to the organization as a whole and further that commit-
ment to one organizational value is sometimes independent of
commitment to another.
The two remaining factors, or clusters of items, reflect (1) expo-
sure to organizational "culture"-knowledgeability and awareness
of organizational values and policies-and (2) exclusiveness of
organizational membership-organizational membership limited
to the particular organization.
Since various measures of participation have frequently been
used as indices of involvement or commitment to voluntary asso-
ciations, it is of interest here to indicate the relation of participa-
tion measures to other indices of commitment.24 In this study
questions on attendance at meetings and self-ratings on amount of
activity were heavily loaded on the cosmopolitan integration
factor. It is noteworthy, however, that holding office was loaded on
a different factor, the organizational introjection dimension. Thus
these two kinds of items, often used as measures of the same thing,
organizational commitment, actually seem to be measures reflect-
ing different kinds of organizational commitment.
This study has also raised some questions regarding Homan's
hypothesis that the higher an individual's rank in a group, the
more nearly his activities conform to the norms of that group.25
The elaboration of this hypothesis requires knowledge of the
kinds of norms and values to which an organizational elite conform
and those from which they may deviate.26The data here indicate
24See, for example, March, op. cit., especially pp. 735-736, and R. O. McWiiliams,
"A Study of the Relationship of Political Behavior to Social Group Membership"
(Ph.D. dissertation, University of Michigan, 1953).
25For a complete statement of this hypothesis and a review of the literature see
Henry W. Riecken and George C. Homans, "Psychological Aspects of Social Struc-
ture," in Gardner Lindzey, ed., Handbook of Social Psychology (Cambridge, Mass.,
1954), II, especially 789-794. See also Norman Hilmar, "Conflicting Social Norms in
a Formal Organization: A Study of Interpersonal Expectations" (Ph.D. dissertation,
Cornell University, 1955).
26See March, op. cit. Riecken and Homans themselves point out, "It has often
been observed that very high ranking members of groups (e.g., members of aristoc-

that commitment to some organizational values may be distinct
from participation, office holding, and integration-all of which
have been used, in addition to sociometric choice, as measures of
rank in testing Homan's hypothesis. The data suggest, further,
that it is among the organizational elite that some values, generally
taken for granted, are discussed and may become new issues.
The results of this research are, of course, highly tentative. The
sample is small and the context is limited. The study has, perhaps,
provided some first steps in identifying meaningful dimensions
of group commitment, a problem which is at the very core of
sociology.
racies, oligarchies, or other powerful classes), who have a secure social position, do
not conform as strictly to some group norms as do individuals of middle rank, nor
are they subjected to serious pressure to conform..." (op. cit., p. 793).

APPENDIX
Table 1. Orthogonal factor matrix.

Item 1 2 3 4 5 6 7*
16. Attendance, branch meetings .72 -.14 -.08 .14 -.02 .08 -.23
17. Attendance, League Day .77 .05 .08 .07 .06 .13 -.34
18. No. organizations, active -.03 -.07 .09 -.21 -.65 .31 .12
19. No. organizations, hold office .06 -.06 -.08 -.08 -.60 .32 .05
20. Give up time in other organizations .54 .13 -.30 -.19 .02 -.17 .04
21. Give up social life .21 .29 -.21 -.17 -.23 -.22 .42
24. Prefer group discussion over outside
speakers .41 .13 -.01 -.15 .47 .16 .14
25. Prefer group discussion over presented
material .25 -.11 .18 -.20 .49 -.15 .03
26. Enthusiasm for League .66 .14 .03 .02 -.03 -.20 .20
27. Rather belong to League than other or-
ganizations .46 .01 -.36 .20 .29 .01 .23
28. Confidence in League achieving goals .09 .21 -.35 .07 -.21 -.24 .17
29. Feel part of branch group .74 .05 -.01 .09 -.11 -.25 .15
30. Feel part of city League .90 .10 -.01 -.01 -.01 .02 .08
31. Feel part of state League .77 -.02 -.04 -.12 .10 .16 .15
32. Feel part of nat'l League .72 -.01 -.13 -.11 -.05 .15 .12
41. Highest office held .67 .26 .20 .07 -.10 -.20 -.23
43. Extent of activity .82 .01 .06 .11 .02 .08 -.12
46. Friendship vs. educational aims for
League -.06 -.22 .00 -.27 -.10 -.18 -.33
47. Introjection .62 .38 .22 -.01 .06 -.39 .03
48. Type of argument, csmt .34 .31 -.13 .51 -.02 -.18 -.10
49. No. qualifications, csm -.12 -.64 .08 .16 -.10 -.30 .08
52. Discussed csmwith League members .57 -.20 .16 .38 .09 .21 -.16
55. Private doubts, csm -.14 -.53 -.10 .05 .12 -.39 .26
56. Commitment, csm .29 .74 -.10 -.02 .05 .23 -.10
57. Personal importance, csm .20 .72 .02 .01 -.02 .26 .05
58. Type of argument, ppr t .26 -.06 .37 .61 .06 .10 .08
59. No. qualifications, ppr .03 .03 -.62 .09 .03 -.05 -.31
62. Discussed ppr with League members .14 -.05 -.02 .63 .11 -.12 .00
65. Private doubts, ppr .03 .13 -.66 .06 .04 -.12 .01
*The nature of this factor is obscure. It is certainly related to the value of soliciting
funds, and, possibly, some form of commitment to this value. However, we have been
able to advance no satisfactory hypothesis to account for the particular combination
of items in the factor. Since discussion of the factor would necessarily involve at this
time only a series of ad hoc explanations, the factor has not been included in the main
analysis. Part of the difficulty may be related to the fact that soliciting funds was a low-
consensus value, which at the time of the study was the object of much conflict and
discussion.
tSpecific League value, cross-sectional membership.
tSpecific League value, political party responsibility.

Table 1 (concluded).
Item 1 2 3 4 5 6 7*
66. Commitment, ppr .27 .07 .61 .25 .13 .33 .07
67. Personal importance, ppr .17 -.03 .62 .29 -.04 -.12 -.09
68. Type of argument, sfc? .37 -.02 .03 .52 .28 .16 -.14
69. No. qualifications, sfc .05 -.13 .24 .05 -.11 -.23 .57
72. Discussed sfc with League members .38 .00 .18 .51 .12 -.23 .15
75. Private doubts, sfc -.18 -.20 .05 .00 -.03 .00 .61
76. Commitment, sfc .41 .12 .02 .16 .54 .16 -.07
77. Personal importance, sfc .33 .23 -.09 .04 .02 -.06 -.54
C25. Objectivity, csm -.07 -.19 .01 .62 .02 -.18 .12
C30. Awareness csm as League value .52 .29 .40 .05 .24 -.16 .01
C34. Objectivity, ppr .00 .23 -.13 .26 .15 .19 .22
C43. Objectivity, sfc -.06 .02 -.02 -.01 .00 .14 .59
?Specific League value, soliciting funds from the community.
Table2. Primary factor pattern.
Item A B C D E F G*
16. Attendance, branch meetings .84 -.05 -.16 -.12 .10 -.20 -.26
17. Attendance, League Day .73 .08 .05 .06 -.01 -.09 -.35
18. No. organizations, active .46 -.19 .07 .20 -.18 -.93 .16
19. No. organizations, hold office .55 -.29 .07 .00 -.04 -.90 .07
20. Give up time in other organizations .44 .25 -.01 -.35 -.18 .08 -.05
21. Give up social life .05 .39 .15 -.20 -.08 -.12 .36
24. Prefers group discussion over outside
speakers .52 -.12 -.06 .03 -.26 .33 .13
25. Prefers group discussion over presented
material .07 .23 -.18 .16 -.29 .61 .02
26. Enthusiasm for League .42 .45 -.01 .02 .04 .02 .17
27. Rather belong to League than other or-
ganizations .45 -.11 -.02 -.36 .19 .20 .19
28. Confidence in League achieving goals -.07 .20 .07 -.40 .18 -.09 .11
29. Feel part of branch group .53 .45 -.13 -.04 .13 -.06 .12
30. Feel part of city League .84 .26 .04 .00 -.05 -.13 .05
31. Feel part of state League .90 .04 .00 .00 -.18 -.10 .13
32. Feel part of national League .91 .00 .00 -.09 -.15 -.26 .09
41. Highest office held .27 .56 .10 .15 .07 .01 -.25
43. Extent of activity .81 .11 -.03 .05 .06 -.13 -.13
46. Friendship vs. educational aims for
League .03 .07 -.31 -.08 -.28 .01 -.36
*See discussion of Factor 7, Table 1.

Table2 (concluded).
Item A B C D E F G*
47. Introjection -.01 .82 .15 .17 .00 .32 .00

48. Type of argument, csmt -.09 .24 .20 -.20 .57 .08 -.13
49. No. qualifications, csm .07 -.05 -.76 .01 .22 -.03 .11
52. Discussed csmwith League members .67 -.18 -.12 .17 .32 -.15 -.12
55. Private doubts, csm -.05 .02 -.69 -.18 .11 .28 .25
56. Commitment, csm -.05 .17 .81 -.05 -.06 .01 -.13
57. Personal importance, csm -.11 .18 .82 .11 -.01 -.08 .05
58. Type of argument, ppr .10 .01 .00 .40 .61 -.06 .17
59. No. qualifications, ppr .11 -.27 -.02 -.73 .12 .04 -.41
62. Discussed ppr with League members -.09 -.02 -.10 -.08 .68 .14 .02
65. Private doubts, ppr .03 -.14 .04 -.74 .13 .10 -.10
66. Commitment, ppr .17 .04 .24 .73 .16 -.07 .18
67. Personal importance, ppr -.15 .38 -.08 .63 .28 .03 .00
68. Type of argument, sfc? .27 -.19 .04 .02 .47 .14 -.11
69. No. qualifications, sfc -.04 .33 -.22 .29 .12 -.03 .61
72. Discussed sfcwith League members .04 .30 -.13 .10 .57 -.03 -.12
75. Private doubts, sfc -.01 -.06 -.16 .13 .04 -.09 .65
76. Commitment, sfc .23 -.05 .18 .03 .05 .47 -.08
77. Personal importance, sfc .10 .15 .15 -.18 .02 .09 -.59
C25. Objectivity, csm -.22 -.04 -.24 -.04 .70 .07 .15
C30. Awareness csmas League value .00 .58 .18 .40 .00 .38 .03
C34. Objectivity, ppr -.06 -.17 .32 -.07 .25 .05 .23
C43. Objectivity, sfc .09 -.09 .10 .10 .00 -.12 .62
tSpecific League value, cross-sectional membership.

tSpecific League value, political party responsibility.
?Specific League value, soliciting funds from the community.
Table3. Correlations among the primary factors.
A B C D E F G
A 1.00
B .35 1.00
C .38 .12 1.00
D .10 -.25 -.09 1.00
E .40 .13 .16 .16 1.00
F .53 -.15 .23 .19 .27 1.00
G -.10 -.04 -.03 -.27 -.14 .01 1.00


Dimensions of Organizational Commitment

Uploaded by

Copyright:

Available Formats

Dimensions of Organizational Commitment

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Dimensions of Organizational Commitment

Uploaded by

Copyright:

Available Formats

Dimensions of Organizational Commitment

Author(s): Helen P. Gouldner

This content downloaded from 195.34.79.192 on Sat, 14 Jun 2014 18:19:00 PM

The author reports on a factor analytic study which distinguishes

IT has of course long been obvious that the members of an associ-

This content downloaded from 195.34.79.192 on Sat, 14 Jun 2014 18:19:00 PM

This content downloaded from 195.34.79.192 on Sat, 14 Jun 2014 18:19:00 PM

basic sociological interest, concerned as it is with the diverse ways

This content downloaded from 195.34.79.192 on Sat, 14 Jun 2014 18:19:00 PM

This content downloaded from 195.34.79.192 on Sat, 14 Jun 2014 18:19:00 PM

to which the individual's ideal self-image includes a variety of

This content downloaded from 195.34.79.192 on Sat, 14 Jun 2014 18:19:00 PM

This content downloaded from 195.34.79.192 on Sat, 14 Jun 2014 18:19:00 PM

had taken on the value in actual League discussions (or would

This content downloaded from 195.34.79.192 on Sat, 14 Jun 2014 18:19:00 PM

This content downloaded from 195.34.79.192 on Sat, 14 Jun 2014 18:19:00 PM

.90 30 Feel a real part of city League: notat all. . .verymuch**

of the organization-the state League, branch group, and national

This content downloaded from 195.34.79.192 on Sat, 14 Jun 2014 18:19:00 PM

FactorA (Oblique):Cosmopolitan Integration

.91 32 Feel a real part of national League: notat all. . .verymuch**

As with Factor 1 this oblique factor reflects integration. A

This content downloaded from 195.34.79.192 on Sat, 14 Jun 2014 18:19:00 PM

terms of its local units but as a national organization or social

Factor B (Oblique): Introjection

.82 47 Introjection: low. . .high**

The highest ranking item on Factor B measures "organizational

This content downloaded from 195.34.79.192 on Sat, 14 Jun 2014 18:19:00 PM

Commitment to a Specific Value: Cross-sectional Membership*

This content downloaded from 195.34.79.192 on Sat, 14 Jun 2014 18:19:00 PM

Orthogonal Factor 2 and Oblique Factor C involve a commit-

Commitment to a Specific Value: Political Party Responsibility

This content downloaded from 195.34.79.192 on Sat, 14 Jun 2014 18:19:00 PM

Exposure to Organizational Culture

This factor involves degree of exposure to League "culture,"

This content downloaded from 195.34.79.192 on Sat, 14 Jun 2014 18:19:00 PM

values, as well as to the internalization of these skills and values,

Exclusiveness of Organizational Membership

This last factor loads heavily items on the number of organiza-

This content downloaded from 195.34.79.192 on Sat, 14 Jun 2014 18:19:00 PM

This content downloaded from 195.34.79.192 on Sat, 14 Jun 2014 18:19:00 PM

INTERCORRELATIONS AMONG THE OBLIQUE

This content downloaded from 195.34.79.192 on Sat, 14 Jun 2014 18:19:00 PM

This content downloaded from 195.34.79.192 on Sat, 14 Jun 2014 18:19:00 PM

commitment to two specific organizational values, cross-sectional

This content downloaded from 195.34.79.192 on Sat, 14 Jun 2014 18:19:00 PM

This content downloaded from 195.34.79.192 on Sat, 14 Jun 2014 18:19:00 PM

Table 1. Orthogonal factor matrix.

This content downloaded from 195.34.79.192 on Sat, 14 Jun 2014 18:19:00 PM

?Specific League value, soliciting funds from the community.

Table2. Primary factor pattern.

This content downloaded from 195.34.79.192 on Sat, 14 Jun 2014 18:19:00 PM

47. Introjection -.01 .82 .15 .17 .00 .32 .00

tSpecific League value, cross-sectional membership.

Table3. Correlations among the primary factors.

This content downloaded from 195.34.79.192 on Sat, 14 Jun 2014 18:19:00 PM

You might also like