Journal of Occupational Psychology (1990), 63, 49-61 Printed in Great Britain 49

© 1990 The British Psychological Society

Predicting employee commitment to

company and union: Divergent
models

Julian Barling* and Bill Wade

Department of Psychology, Queen's University, Kingston, Ontario KIL 3N6, Canada

Clive Fullagar
Kansas State University

The adequacy of Mowday, Porter & Steers' (1982) model of organizational commitment
for predicting both company and union commitment was assessed. Data were obtained
from 100 members of a white-collar union. Two rnonths before this study, they had
taken part in a three-week strike and were still involyed in a dispute with management.
Measures of company and union commitment were regressed separately on work
experiences (job satisfaction, job involvement and perceived organizational climate) and
personal characteristics (educational level, sex, company and union tenure). In neither of
the multiple regression analyses did the same predictor account for a significant portion
of the variance in both union and company commitment. Job satisfaction, organizational
climate and job involvement were significant predictors of company commitment,
accounting for 17.8 per cent of the variance. Union tenure was the only significant
predictor (5.3 per cent of the variance) of union commitment. The emergence of
divergent predictors of company and union commitment suggests the need for greater
specificity within the literature on 'organizational commitment'. Consistent with the
differing nature and function of unions and companies, it may be more appropriate to
construct separate models of company and union commitment.

The 1980s have been witness to an upswing in psychological research on labour

organizations, the formation of various cornmittees within the Arnerican Psychological
Association aimed at developing union research (e.g. the implementation in 1981 of the
Task Force on Union Negotiation), as well as the publication of special editions of
psychological journals which have specifically addressed the issue of psychology's
relationship with labour {American Psychologist, 1984, 39(4); International Review ofApplied
Psychology, 1981, lQ{2y, Journal of Occupational Psychology, 1986, 59(3)}. One area of
conceptual interest has been the generalization of theories of organizational commitment
(e.g. Porter & Smith, 1970; Steers, 1977) to the theoretical development and operation-
alization of union commitment (Fullagar, 1986; Fullagar & Barling, 1987, 1989^;
Gordon, Philpot, Burt, Thompson & Spiller, 1980; Gordon, Beauvais & Ladd, 1984;
Ladd, Gordon, Beauvais & Morgan, 1982).
* Requests for reprints.
50 Julian Barling, Bill Wade and Clive Fullagar
There is a considerable amount of evidence pointing to a positive relationship between
union and company commitment (Dean, 1954; Fukami & Larson, 1982; Gallagher,
1984; Purcell, I960). Such a relationship suggests that the two commitment constructs
may conceivably share certain common predictors. Indeed, Fukami & Larson (1984)
investigated whether union and company commitment could be predicted by the same
variables. They assessed whether Mowday et al.'s (1982) model of organizational
commitment isolated the unique and common predictors of both company and union
commitment. Fukami & Larson, however, provided little empirical evidence or justifi-
cation for the inclusion of work- and role-related variables as antecedents, especially of
union commitment. Nonetheless, based on Fukami & Larson's (1984) multiple regression
analysis rather than their zero order correlations, company commitment was predicted
significantly by personal characteristics (number of years with the company) and role-
related variables (job scope and job stress). On the other hand, union commitment was
predicted significantly by one kind of work experience, viz. social involvement. Yet it is
not clear why social involvement at work would predict union commitment: the four
items in their subscale may not be measuring work experiences at all (e.g. 'what percent of
the people you work with would you consider to be good friends of yours?' and 'what
percent of your friends work with the Courier?'). Fukami & Larson (1984) question
whether the work experiences and personal characteristics they operationalized were
appropriate for understanding union commitment, and suggest that future research still
focus on these two categories as possible predictors of union commitment.
Consequently, the present study assessed whether various work experiences and
personal characteristics found to be empirically associated with either union or company
commitment would predict company and union commitment. First, various facets of job
satisfaction (work, co-workers, supervision, pay and promotion satisfaction) were investi-
gated as predictors, as correlations in the range of .50 have been found between job
satisfaction and organizational commitment (Brooke, Russell & Price, 1988; Hall &
Schneider, 1972; Meyer & Allen, 1987; Mowday etal., 1982; Mowday, Steers & Porter,
1979). Dissatisfaction with extrinsic and intrinsic conditions of employment has also been
found to be a major reason why workers join unions and vote for a union (Bigoness, 1978;
DeCotiis & LeLouarn, 1981; Farber & Saks, 1980; German, Goldberg & Herman, 1976;
Hammer & Berman, 1981; Kochan, 1979; Maxey & Mohrman, 1980; Premack &
Hunter, 1988; Schriesheim, 1978; Zalesny, 1985). Regarding research on the
relationship between union commitment and job satisfaction, Gordon etal. (1980) found
either negative or non-significant associations between satisfaction of lower and higher
order needs and three dimensions of union conimitment (viz. feelings of responsibility to
the union, expressed willingness to work for the union and general belief in unionism).
Recently, extrinsic and intrinsic job dissatisfaction have been found to predict union
commitment amongst both skilled and unskilled workers (Fullagar & Barling, 1989c?).
Consequently, the present research hypothesized that although job satisfaction would be
negatively associated with union commitment and positively associated with organiz-
ational commitment, it would be a common predictor of both types of commitment.
Second, perceptions of organizational climate have been found to result from interper-
sonal and supervisor—subordinate relationships at work (Schneider, 1985). There is some
support for the hypothesis that the perceived quality of supervision predicts company
commitment (Mowday et al., 1982). Positive perceptions of the labour—management
 Predicting union and company commitment 51
relationship is the only variable to have been consistently related to dual commitment in
several studies (Angle & Perry, 1986; Gallagher, 1984; Larson & Fukami, 1985;Magenau
& Martin, 1985; Magenau, Martin & Peterson, 1988; Schriesheim & Tsui, 1980). The
present study, therefore, hypothesized that positive perceptions of organizational climate
would predict both commitment to the union and company.
Third, job involvement has also been found to be associated with both company
commitment (Mowday et al., 1982) and union loyalty (Fullagar & Barling, 1989^).
Regarding the former relationship, recent research (Brooke etal., 1988) has indicated that
job involvement, job satisfaction and organizational commitment are empirically distinct
rather than partially redundant concepts (Morrow, 1983), and job involvement is
positively related to union loyalty amongst skilled workers who have greater access to
decision-making structures within the employing organization (Fullagar & Barling,
1989^). Amongst semi-skilled and unskilled workers, on the other hand, job involvement
is negatively related to union loyalty. This is attributable to more alienated (i.e. less job
involved) workers having less opportunity for organizational involvement and the
satisfaction of intrinsic needs. With the present study's sample of white-collar workers, it
was hypothesized that job involvement would be positively associated with both
organizational and union commitment.
The current research also investigated personal characteristics as possible predictors of
company and union commitment. Much evidence suggests that organizational commit-
ment is positively related to age and tenure, and inversely related to education (Mowday et
al., 1982). Moreover, males exhibit higher levels of organizational commitment than
females (Angle & Perry, 1981). The little research that has been conducted on union
commitment has found no significant correlations with job grade, tenure, marital status,
and number of children (Gordon et al., 1980, 1984). Two personal variables which have
been associated with union commitment are member's gender (Gordon et al., 1980) and
race (Fullagar & Barling, 1989^). Women are more loyal to their union than men, even
though men are more likely to participate in union affairs (Gordon etal., 1980).* Race has
been found to moderate both the strength and nature of the relationship between certain
predictor variables and union commitment (Fullagar & Barling, 1989<«).t A side-bet
approach to commitment (Becker, I960) would suggest that where seniority clauses are
part of the collective agreement, union tenure would be related to union commitment.
Becker indicated that the longer the individual has spent with the organization, the more
he/she has invested, both behaviourally and financially. Consequently, through a process
of cognitive consonance, the individual becomes more attached and committed to the
organization. Fukami & Larson (1984) showed a significant relationship between com-
pany commitment and tenure, but no relationship between company tenure and union
commitment (union tenure was not reported). The present research, therefore, ascertained
whether demiographic characteristics, including union tenure, predicted company and
union commitment, and controlled for the possible effects of these sociodemographic
variables in the regression analysis by entering them first.

* This phenomenon is not due to gender per se, but rather diverse variables including the greater experience of sex-role
conflict amongst working women (Chusmir, 1982).
t Race was regarded as a marker variable denoting differences in occupational and social privilege, salary, history of trade
unionism, and political representation.
52 Julian Barling, Bill Wade and Clive Fullagar
Three other questions remaining from Fukami & Larson's (1984) research were also
addressed in this study. First, Fukami & Larson (1984) studied an all-male sample, even
though males and females have been found to differ on both company (Mowday et al.,
1982) and union (Gordon et al.., 1980) commitment. The present sample included both
male and female union members. Second, a subsequent description by Larson & Fukami
(1985) shows that there was an unusually high level of cooperation between the company
and union in their 1984 study. Since this may be atypical, the present research was"
conducted on members of a union involved in an ongoing labour dispute with manage-
ment. Finally, Fukami & Larson (1984) assessed the correlation between company and
union commitment. This phenomenon of dual loyalty or allegiance has been studied
previously (e.g. Dean, 1954; Purcell, 1954; Stagner, 1934), where it was hypothesized
that simultaneous or dual loyalty was unlikely because of the dissonance and role conflict it
engendered. Yet consistent with Fukami & Larson (1984), many studies have yielded
positive relationships between company and union commitment. The present study
afforded a test of the generality of previous findings as (a) the union and management in
this study were involved in an ongoing labour dispute, and (b) the methodological
problems common in previous research (cf. Schriesheim & Tsui, 1980) were avoided.

Method

Setting
All participants in the present study were employed on a full-time basis by the same community college in
south-western Ontario. Compulsory membership of the union was a condition of employment. Two months
prior to this study, the participants were involved in a three-week, province-wide strike which had been
preceded by 10 months, of union-management negotiations. Twenty-two community colleges and 7600
faculty were involved in this strike, thefirstby this union against the community college's Board of Regents.
The major issue involved in this prolonged labour dispute was essentially economic, namely the number of
hours of direct teaching contact required of faculty per week. The dispute had not been settled by the two
parties concerned after three weeks of striking, and the teachers and counsellors were legislated back to work.
Thus, the white-collar participants in this study were members of an active union.

Subjects
Questionnaires were mailed to the home addresses (obtained from the union) of all 305 of the 315 full-time
teaching staff and counsellors at this one college for whom home addresses were available. Fifteen were
returned by the post office as undeliverable. Of the final sample of 290, 100 questionnaires were duly
completed and returned, a response rate of 34.4 per cent. The mean age of the respondents was 44.1 years
(SD = 8.2), and their mean company and union tenure were 10.4 years (SD = 5.5) and 8.6 years
(SD = 4.4) respectively. Forty-five individuals had acquired university degrees or college diplomas, 47 a
graduate course or some graduate training, and the remaining eight had completed 13 years of formal
education. Although there is no way of guaranteeing the representative nature of the sample, the male ; female
ratio in the present sample (62 : 38) closely approximated that of the total population (66: 34) of this one
college. The only study variable on which males and females differed was educational level. Female
respondents had received significantly more formal schooling than males (^1, 97) = 2.30, p < .01).

Procedure
The questionnaires and a covering letter from the local union president indicating support for the research
were distributed to all respondents by mail, and returned in self-addressed, reply-paid envelopes. To ensure
 Predicting union and company commitment 53
confidentiality and anonymity, subjects were requested not to identify themselves in the questionnaires in any
way. Tbe union local received a report of the completed research, which was also available to any interested
union member.

Questionnaire materials
Mowday et at.'s (1979) nine-item short form of the Organizational Commitment Questionnaire assessed
company commitment. All nine items are positively worded and rated on a seven-point Likert-type scale, and
the short form correlates highly with the 15-item full scale. The Organizational Commitment Questionnaire
consistently yields satisfactory internal and temporal stability, discriminant, convergent and predictive
validity (Mowday et al., 1982). The 'union loyalty' factor derived from Gorden et al.'s (1980) Union
Commitment Questionnaire consists of the 10 items which consistently appear as the first factor, accounting
for most of the variance in union commitment across different factor analytic studies (Fullagar, 1986; Fullagar
& Barling, 1989^; Gordon etal., 1980; Ladd etal., 1982). All 10 items were rated on afive-pointLikert-type
scale in this study, and adequate internal reliability and construct validity have been documented in these
studies. (Unless specifically stated, reliability for all questionnaires for the present sample is adequate; see
Table 1.)
The 72-item Job Descriptive Index (Smith, Kendall & Hulin, 1969) assesses satisfaction with work, pay,
co-worker, promotion and supervision. The respondent rates whether each of the items presented is
descriptive of the specific dimension. Cook, Hepworth, Wall &Warr(1981) suggest that the Job Descriptive
Index remains the most psychometrically acceptable measure of job satisfaction. Taylor & Bowers' (1972)
13-item Organisational Climate Composite Indices measure technological readiness, human resource
primacy, communicationflow,motivational conditions and decision-making practices. Responding occurs on
afive-pointLikert-type scale. In the present study, however, the reliability of the Motivational Conditions
subscale was low (alpha = .45), and it was discarded from all analyses. Job involvement was measured with
Kanungo's (1982) 10-item (plus five filler items) Job Involvement Questionnaire. Responding occurs on a
six-point Likert-type scale and this questionnaire has been used in previous union commitment research
(Fullagar & Barling, 1989'*). This scale exhibits high internal and temporal stability, and its convergent and
discriminant validity are also satisfactory (Kanungo, 1982).

Results

The extent to which work experiences and demographic characteristics predict company
and union comirtitment were analysed in two separate multiple regression analyses.
However, although the linearity assumption was satisfied in all instances, the multi-
colinearity assumption was questionable. First, there were substantial correlations
between the demographic characteristics, viz. age, company and union tenure (see
Table 1). Thus, age was omitted from the multiple regression analyses, and union and
company tenure were assessed as separate predictors of union and company commitment
respectively. Second, all five work satisfaction facets were significantly intercorrelated,
there were significant correlations between all work satisfaction and all perceived
organizational climate indices, while the correlations amongst the perceived organiz-
ational climate indices were substantial (range: r = .63 to .82, all ^ < .001). To
minimize the possibility of a mukicolinearity problem, a principal components analysis
with varimax rotation was computed on the job satisfaction and organizational climate
variables. Two interpretable components emerged.* The first component was labelled
'perceived organizational climate' (alpha = .85; accounting for 53 per cent of the overall
variance). The items (with their loadings in parentheses) on this component were human

* DetailecJ results of the factor analysis are obtainable from the first author on request.
54 Julian Barling, Bill Wade and Clive Fullagar

.79

.39

.03
.02
.79
.03

-.04
.32
00

.63
.67
.06

.27
GS
q
1

-.10 -
.60

.07
.47

.32
1

o so
CN
O O
.28

.12

.22
.47
GS

-.04
1

so
.24

cr-
.40
GS

.27
.18
rCi (N 1—(

.22 -
00 CN 00

.01
CN
.06 -.08
.14 -.01 -.08
-.02

O O
-.18

-a r l' r \r^
!::i' 9
.07 -.03

o
-.07

.05 -.01
o CN
-.16

;^ o r rr
.22

ITS O o o cr-
.23

00 CS 00
-.07

r rl'
or
.20

.19
.11

.08
.00

.08

.08

-.06

O
l' I3
o O "U V
CN CN CN

1 >/N 00 CN
1 ^ so 00 00

5 1
c<N c^ ITN 00 tr\ OO 00
00 o c r-i <N 1-^
o
so cn t>
18.
10.

40.

00 ot)
ro § CN
tion

I a
Al
I
3
satisfaci :ion
satisfaci :ion

Co-worker

Supervi sor

c
^1 I I a-g
Compai

Pay sati

satisfaci

o
-I I
Work

3 ^
a M I a

I a 3Mat3Q5s
s ^ S 6
3 i^ ON «
tN
 , * • • • • : ; • .

Predicting union and company commitment 55

resource primacy (.88), decision-making practices (.87), communication flow (.85),
technological readiness (.80) and supervisor satisfaction (.63). The items in the second
component were work (.84), co-worker (.66), promotion (.44) and pay (.49) satisfaction.
This component was labelled 'job satisfaction' (11 per cent overall variance explained;
alpha = . 76). Factor scores were calculated for these two components using the regression
method.
Two stepwise multiple regression analyses were computed. As in Fukami & Larson's
(1984) study, the demographic variables, i.e. sex (coded as a dummy variable), level of
education, union tenure (in the case of union commitment), and company tenure (as a
possible predictor of company commitrnent) were entered first in both regression
equations to control for the effects of these demographic variables. The factor scores for job
satisfaction and then perceived organizational climate were then entered, followed by job
involvement. The order in which predictor variables are entered in a stepwise multiple
regression analysis influences the amount of variance each variable will explain (Pedhazur,
1982). Unlike Fukami & Larson, the order of entry of the work experience variables in the
hierarchical regression analysis in the present study was determined by their conceptual
and theoretical imLportance based on previous empirical research on union and company
commitment.*
The above six variables explained 19 per cent of the variance in company commitment.
However, only job satisfaction (F(l, 94) = 3.07, p < .05; beta = .29), perceived
organizational climate (F(l, 94) = 4.05, p < . 0 1 ; beta = .26) and job involvement
(F(l,94) = 3.54,/! < . 0 1 ; beta = .09) significantly predicted company commitment.
These three Variables respectively accounted for 7.9, 6.3and 1.2 per cent of the variance in

Table 2. Summary results of multiple regression analyses predicting commitment to

company and union

Company commitment Union commitment

Predictor variable beta B F R^ beta B p

Demographics
Company tenure .00 .09 .03 1.11 — — — — 2.24**
Union tenure — — —- — .05 .24 .47 5.46*
Education level .00 -.08 - .66 1.33 .06 .05 .39 2.86 0.76
Sex .03 -.19 -4.84 1.16 .06 .06 1.18 2.03 0.05
Work experiences
Work satisfaction .11 .29 4.16 3.07* .06 -.03 - .31 1.53 9.17**
Organiz. climate .18 .26 3.91 4.05** .07 -.11 -1.13 1.43 4.21**
Job involvement .19 .09 .28 3.54** .07 -.01 - .01 1.18 3.69**

*p < .05; **p < .01.

" Total variance explained at the end of each regression step.
In this instance, the regression parameter for company tenure was compared to that for union tenure.

* Given the significance of the order of variable entry in stepwise regression, alternative hierarchical orders were computed
for the entry of predictor variables in the regression equation. These alternative stepwise solutions yielded no significant
differences in the amount of variance explained by each variable.
56 Julian Barling, Bill Wade and Clive Fullagar
company commitment. On the contrary, only 7.1 per cent of the variance in union
commitment was accounted for by the six predictors. In this analysis, the only significant
predictor was union tenure ( f ( l , 94) = 5.46,^ < .05; beta = - . 2 4 ) , explaining 5.3
per cent ofthe variance in union commitment (see Table 2). As it was expected that union
and company commitment would share the same predictors, the difference between these
regression coefficients was tested for statistical significance (Cohen & Cohen, 1983). The
result of this test is reported as a / ratio (see Table 2). The unstandardized partial
regression coefficients differed significantly for tenure (company tenure vs. union tenure),
job satisfaction, perceived organizational climate and job involvement.
There was a significant, negative relationship between union and company commit-
ment (r (93) - - . 2 5 , ^^ > .01).

Discussion
Consistent with previous findings (Mowday et al., 1982), all individual facets (see
Table 1) and the global index (see Table 2) of job satisfaction predicted company
commitment. However, neither the specific job satisfaction facets correlated with union
commitment, nor did the global job satisfaction index predict commitment to the union
in the miultiple regression analysis (see Tables 1 and 2).
That job satisfaction, perceived organizational climate and job involvement were
correlated with company commitment replicates a large body of existing data (Brooke et
al., 1988; Mowday et al., 1982); other than pointing to the strength of the relationship
between organizational commitment and these work-related factors (15.5 per cent ofthe
variance explained), no further discussion is warranted. However, the fact that neither job
satisfaction nor job involvement predicted union commitment is not consistent with
previous findings (e.g. Fullagar & Barling, 1989^; Gordon «/^/., 1980, 1984). A number
of factors might account for this discrepancy. First, the nature ofthe union-management
relationship in the various studies differs spniewhat. The present study assessed the
correlation between job satisfaction and company and union commitment in an organiz-
ation where the union—management relationship was unfavourable. In such adversarial
situations, the role of job satisfaction in union commitment may be reduced. Specifically,
job dissatisfaction predicts increased initial union activity (Brett, 1980<«, b; DeCotiis &
LeLouarn, 1981). Second, one would have expected amongst a sample which had been on
strike for three weeks, and been legislated back to work, that union commitment would
have been relatively high and job satisfaction low. However, the dissatisfactions experi-
enced by the present sample may not have generalized to, or been measured by, the specific
job satisfaction dimensions included in the Job Descriptive Index. The specific grievance
expressed by the faculty concerned the weekly, direct teaching load, which was not
defined by any of the items in the JDI and which did not appear to spill over into
dissatisfaction with co-workers, pay, supervisors, promotional prospects or work. Fur-
thermore, the source of dissatisfaction was probably perceived as being the Board of
Regents and neither the work per se, nor direct supervision. If future research is to
concentrate on ascertaining the common work experience predictors of both union and
organizational commitment, specific grievances will have to be focused on and operationa-
lized, rather than relying on global indices of satisfaction. Finally, the breakdown in
 Predicting union and company commitment 57
labour negotiations may have been attributed to union incompetency and produced a
deterioration in membership commitment rather than ap increase.
This raises the issue of whether the perceived instrumeritality of the union in fulfilling
the needs and resolving the dissatisfaction of its members might be a more important
predictor of union corrimitment. Previous firidings have shown that job dissatisfaction is
highly related to pro-union voting and activity in certification elections (Brett, 1980a, b;
DeCotiis & LeLouarn, 1981), yet the relationship between job satisfaction and union
commitment amongst members in already certified unions is modest at best (Gordon et
al., 1980, 1984). However, perceived union instrumentality has been shown to moderate
the relationship between job dissatisfaction and union commitment (Fullagar & Barling,
1989'*). Workers who are dissatisfied with their jobs do not necessarily become commit-
ted to the union unless they perceive the union as instrumental in redressing their
grievances.
With regard to the lack of a relationship between job involvement and union
commitment, Fullagar & Barling (1989^?) found a significant, but weak, effect amongst a
sample of blue-collar workers. The discrepancy with the present results may be due to the
differing nature of the samples. Job involvement (or alienation) appears to be an important
predictor amongst workers of lower occupational status who enjoy less access to decision-
making processes. Furthermore, where the relationship between management and labour
is poor (as in the present instance), job characteristics may become less significant as
predictors of union commitment. Indeed, the significant negative correlation between
organizational and union commitment can probably be attributed to the adversarial
nature of the labour—management relationship (Angle & Perry, 1986; Gallagher, 1984;
Magenau & Martin, 1985).
The separate components of perceived organizational climate correlated significantly
with company but not union commitment (see Table 1). There is a growing amount of
evidence which would suggest that organizational climate is a moderator of the
relationship between union and company commitment, rather than a direct predictor of
both types of commitment (Angle & Perry, 1986; Gallagher, 1984; Magenau & Martin,
1985; Magenau e/^/., 1988; Schriesheim & Tsui, 1978). Where the union—management
relationship is favourable (e.g. Fukami & Larson, 1984), positive correlations emerge. In
situations where the union—management relationship is adversarial (e.g. the present
study), either negative correlations emerge or union and company commitment are
unrelated. Subjects in the present research were divided into those with favourable
perceptions of organizational climate and those with unfavourable perceptions on the basis
of a median split. Amongst the latter there was a significant negative correlation between
union and company commitment (r(46) — —.52, p < .01), whereas amongst those
subjects with positive perceptions of organizational climate, an insignificant correlation
between both types of commitment emerged (r(47) — —.06, p > .05). This would
confirm the mioderating role of organizational climate.
Unlike previous research, none of the demographic variables were associated with
company commitment (see Tables 1 and 2). On the other hand, union tenure was the only
significant predictor of union commitment. Given the seniority clause within the
collective agreement, this may be consistent with a side-bet approach to understanding
commitment. Meyer & Allen (1984) question whether a significant correlation between
loyalty (affective commitment) and tenure is a sufficient test of side-bet; they showed that
58 Julian Barling, Bill Wade and Clive Fullagar
side-bet theory can be assessed more adequately with measures of continuance
(membership) commitment. Nonetheless, where a seniority clause is operative, it is
possible that affective commitment follows continuance commitment to reduce any
possible role conflict or cognitive dissonance.
The findings of this study raised several issues. First, although job satisfaction may
precede company commitment (Meyer & Allen, 1987), it remains unclear if union tenure
is an antecedent, correlate or consequence of union commitment, and longitudinal data
are required. Second, a global measure of job satisfaction obscures possible contributions
of specific satisfaction facets to union commitment. Although the possibility of a
mukicolinearity problem precluded the use of unique measures, it remains unclear what
purpose intrinsic job dissatisfaction fulfils in the union commitment process: to date,
research has primarily investigated the role of extrinsic dissatisfaction in the unionization
process (Brett, 1980<?, b; Hamner & Smith, 1978). Third, it is important to replicate the
present findings amongst blue-collar unions, as the type of union studied may limit the
generalizability of any findings. However, the ongoing labour dispute and recent strike
experience of this white-collar union affords the results obtained in this study greater
external validity. Fourth, males and females have been sbown to enjoy different union
experiences (Izraeli, 1985) and exhibit differing levels of commitment (Gordon et al.,
1980). Future research should focus on gender as a possible moderator, rather than direct
predictor, of union commitment. Unfortunately, the small sample size in the present
study prevented an assessment of the moderating role of gender, although it was found
that no differences existed between male and female respondents on the union, work and
organization variables. That males and females did not differ with respect to either
organizational commitment or union commitment suggests that in situations where the
labour—management relationship is highly adversarial, gender differences in commitment
are minimal, and loyalty to both company and union become more uniform. Fifth, the
present sample consisted of members of a closed-shop union. The level, nature, and
determinants of union commitment may differ between companies that have two or more
competing labour organizations and companies governed by a single-union shop
agreement. For example, the low amounts of variance accounted for in the present study
may be attributable to the possible effects of a union shop, as freedom of association is an
important determinant of commitment (Salancik, 1977). The current results suggest that
variables which had been found to be significant correlates of union commitment in
situations where there is freedom to choose between membership of several unions, or even
whether one wants to join a union or not, may not generalize to the union shop situation.
Even though the present study investigated different kinds of work-experience
variables to Fukami & Larson (1984), the results still did not isolate a common predictor of
both company and union commitment. We suggest that the nature of companies and
unions as organizations is such that research shouid address 'divergent' rather than
'parallel' models of commitment, and focus explicitly on their differential predictors.
Although both unions and companies are social organizations, there are fundamental
differences between the two. Companies develop from the need to pool resources to
manufacture marketable goods; unions arise largely from dissatisfaction with employment
conditions (Brett, 1980^, b\ Tannenbaum, 1965). Besides other distinctions such as
(fl) whether the particular organization is incorporated for financial gain, and {h) the fact
that workers perceive the company as providing basic needs (Schein, 1980) and the union
 Predicting union and company commitment 59
as ensuring those basic needs are satisfied through the collective bargaining and grievance
processes, the motivation, involvement and commitment required by unions and
companies for their survival and growth differ.
It is argued, therefore, that divergent models be constructed and tested to understand
and predict union and company commitment. This is consistent with the suggestion that
organizational commitment is not a global, unitary construct. Rather, individuals
simultaneously experience a series of multiple commitments (Reichers, 1985). It follows
that the term 'organizational commitment' should be discarded in favour of the terminolo-
gical and conceptual precision afforded by the alternative use of 'company' and 'union'
commitment.

Acknowledgements
Portions of this research were supported by an Advisory Research Committee grant from Queen's University.
The financial assistance of the Social Sciences and Humanities Research Council of Canada (Grant No.:
410-85-1139) to the first author is also acknowledged. The authors express their appreciation to W. H .
Cooper, J. Meyer and G. Wilde for constructive comments regarding earlier versions of this manuscript.

References
Angle, H. J. & Perry, J . L. (1981). An empirical assessment of organizational commitment and organiza-
tional effectiveness. Administrative Science Quarterly, 26, 1-14.
Angle, H . J . &Perry,J. L. (1986). Dual commitment and labor—management relationship climates. AtarfaKy
of Management Journal, 29, 31—50.
Becker, H. S. (I960). Notes on the concept of commitment. American Journal of Sociology, 66, 32—42.
Bigoness, W . J. (1978). Correlates of faculty attitudes toward collective bargaining. Journal of Applied
Psychology, 63, 228-233.
Brett, J. M. (1980«). Why employees want unions. Organizational Dynamics, Spring, 47—59.
Brett, J. M. (1980^). Behavioral research on unions and union-management systems. In B. M. Staw (Ed.),
Research in Organizational Behavior, vol. 2. Greenwich, CT: JAI Press.
Brooke, P. B., Russell, D. W . &Price,J. L. (1988). Discriminant validation of measures of job satisfaction,
job involvement, and organizational coramitment. Journal of Applied Psychology, 73, 139-145.
Chixsmir, L. H. (1982). Job commitment and organizational woman. Academy of tAanagement Review, 7,
595-602.
Cohen, J. & Cohen, P. (\9Si). AppliedMultipleRegressionlCorrelation AnalysisfortheBehavioralSciences, 2nded.
Hillsdale, NJ: Lawrence Erlbaum.
Cook,J. D., Hepworth, S. J., Wall, T. D. & Warr, P. B. (1981). The Experience of Work: A Compendium and
Review of 249 Measures and their Use. London: Academic Press.
Dean, L. R. (1954). Union activity and dual loyalty. Industrial and Labor Relations Review, 1, 526-536.
DeCotiis, T. A. & lelouarn, J . (1981). A predictive study of voting behavior in a representative election
using union instrumentality and work perceptions. Organizational Behavior and Human Performance, 27,
103-118.
Farber, H. S. & Saks, D. H. (1980). Why workers want unions: The role of relative wages and job
chatactetistics. Journal of Political Economy, 88, 349-369.
Fukami, C. V. & Larson, E. W. (1982). The relationship between union commitment and organizational
commitment: Dual loyalty reexamined. Paper presented at the 42nd Annual Meeting of the Academy of
Management, New York.
Fukami, C. V. & Larson, E. W . (1984), Commitment to company and union: Parallel models. Journal of
Applied Psychology, 69, 367-371.
Fullagar, C. (1986). A factor analytic study of the validity of a union commitment scale. Journal of Applied
Psychology, 7 1 , 129-136.
60 Julian Barling, Bill Wade and Clive Fullagar
Fullagar, C. & Barling, J. (1987). Toward a model of union commitment. In D. Lipsky & D. Lewin (Eds),
Advances in Industrial and Labor Relations. Greenwich, CT: JAI Press.
Fullagar, C. & Barling, J. (1989«). A longitudinal test of a model of the antecedents and consequences of
unioa. comraitmtat. Journal of Applied Psychology, 14, 213-227.
Fullagar, C. & Barling, J. (1989^). The dimensionality of union commitment: A construct and discriminant
validity study. Manuscript submitted for publication.
Gallagher, D. G. (1984). The relationship between organizational and union commitment among federal
government employees. Proceedings of the Academy of Management, 44, 319-323.
Getman,J. G., Goldberg, S. B. &Herman, J. B. (1976). Union Representation Elections: Law and Reality. New
York: Russell Sage Foundation.
Gordon, M. E., Beauvais, L. L. & Ladd, R. T. (1984). The job satisfaction and union commitment of
unionized engineers. Industrial and Labor Relations Review. 37, 359—370.
Gordon, M. E. & Nurick, A. J. (1981). Psychological approaches to the study of unions and union-
management relations. Psychological Bulletin, 90, 293-306.
Gordon, M. E., Philpot, J. W . , Burt, R. E., Thompson, C. A. &Spiller, W. E. (1980). Commitment to the
union: Development of a measure and an examination of its covKhites. Journal of Applied Psychology, 65,
479-499.
Hall, D. T. & Schneider, B. (1972). Correlates of organizational identification as a function of career pattern
and organizational type. Administrative Science Quarterly, 17, 340—350.
Hammer, T. H . & Berman, M. (1981). The role of noneconomic factors in faculty union voting. Journal of
Applied Psychology, 66, 4 1 5 ^ 2 1 .
Hamner, W . C. & Smith, F. J. (1978). Work attitudes as predictors of unionization SiCtivity. Journal of
Applied Psychology, 63, 4 1 5 ^ 2 1 .
Izraeli, D. N . (1985). Sex differences in self-reported influence among union officers. Journal of Applied
Psychology, 70, 148-156.
Kanungo, R. N . (1982). Work Alienation: An Integrative Approach, New York: Praeger.
Kochan, T. A. (Z979). How American workers view labor unions. Monthly Labor Review, 101, 23—31.
Ladd, R. T., Gordon, M. E., Beauvais, L. L. & Morgan, R. L. (1982). Union commitment: Replication and
tvXir&ion. Journal of Applied Psychology, 61, 640-644.
Larson, E. W. & Fukami, C. V. (1985). Union commitment as a moderator of the relationship between
company commitment and worker behaviors. Paper presented at the Academy of Management Meeting,
Boston, MA.
Magenau, J. M. & Martin, J. E. (1985). Patterns of commitment among rank-and-file union members: A
canonical analysis. Proceedings of the 37th Annual Meeting of the IRRA.
Magenau, J . M., Martin, J. E. & Peterson, M. M. (1988). Dual and unilateral commitment among stewards
and rank-and-file union members. Academy of Management Journal, 31, 359-376.
Maxey, C. & Mohrman, S. A. (1980). Worker attitudes toward unions: A study integrating industrial
relations and organizational behavior perspectives. Proceedings of the 33rd Annual Convention of the Industrial
Relations Research Association, 33, 326-333.
Meyer, J. P. & Allen, N . J. (1984). Testing the 'side-bet theory' of organizational commitment: Some
methodological considerations. Journal of Applied Psychology, 69, i72-378.
Meyer, J . P. & Allen, N . J. (1987). A longitudinal analysis of the early development and consequences of
organizational commitment. Canadian Journal of Behavioral Science, 19(2), 199-215.
Morrow, P. C. (1983). Concept redundancy in organizational research: The case of work commitment.
Academy of Management Review, 8, 486-500.
Mowday, R. T., Porter, L. W . & Steers, R. M. (1982). Employee-Organization Linkages: The Psychology of
Commitment, Absenteeism and Turnover. New York: Academic Press.
Mowday, R. T., Steers, R. M. & Porter, L. W . (1979). The measurement of organizational commitment.
Journal of Vocational Behavior, 14, 224-247.
Pedhazur, E. J. (1982). Multiple Regression in Behavioral Research: Explanation and Prediction, 2nd ed. New
York: Holt, Rinehart & Winston.
Porter, L. W . & Smith, F. J. (1970). The etiology of organizational commitment. Unpublished paper.
University of California, Irvine.
Premack, S. L. & Hunter, J. E. (1988). Individual unionization decisions. Psychological Bulletin, 103,
223-234.
 Predicting union and company commitment 61
Purcell, T. V. (1954). Dual allegiance to company and union—packinghouse workers. Personnel Psychology, 7,
48-58.
Purcell, T. V. (I960). Blue Collar Man: Patterns of Dual Allegiance in Industry. Cambridge,, MA: Harvard
University Press.
Reichers, A. B. (1983). A review and reconceptualizarion of organizational commitment. Academy of
Management Review, 10, 465—^76.
Salancik, G. R. (1977). Commitment and the control of organizational behavior and belief. In B. M. Staw &
G. R. Salancik (Eds), Netit Directions in Organizational Behavior. Chicago: St Clair Press.
Schein, E. H. (1980). Organizational Psychology, 3rd ed. Englewood Cliffs, NJ: Prentice-Hall.
Schneider, B. (1985). Organizational behavior. Annual Review of Psychology, 36, 573—611.
Schriesheim, C. A. (1978). Job satisfaction, attitudes towards workers, and voting in a union representation
election. Journal of Applied Psychology, 63, 548—553.
Schriesheim, C. A. & Tsui, A. S. (1980). Dual commitment to company and union: Fact or fiction. Paper
presented at the Annual Convention of the American Psychological Association.
Smith, P. C , Kendal, L. M. & Hulin, C. L. (1969). The Measurement of Satisfaction in Work and Retirement.
Chicago: Rand McNally.
Stagner, R. (1954). Dual allegiance as a problem in modern society. Personnel Psychology, 1, 41—47.
Steers, R. M. (1977). Antecedents and outcomes of organizational commitment. Administrative Science
Quarterly, 27, 4 6 - 5 6 .
Tannenbaum, A. S. (1965). Unions. In J. G. March (Ed.), Handbook of Organizations. Chicago: Rand
McNally.
Taylor, J. C. & Bowers, D. G. (Eds) (1972). Survey of Organizations. Ann Arbor, MI: Institute for Social
Research, University of Michigan.
Zalesny, M. D. (1985). Comparison of economic and noneconomic factors in predicting faculty vote
preference in a union representation election. Journal of Applied Psychology, 70, 243-256.

Received 15 November 1988; revised version received 18 April 1989