A Meta Analysis of Organizational Identification

Journal of Vocational Behavior 66 (2005) 358–384
www.elsevier.com/locate/jvb
Organizational identification: A meta-analysisq

Michael Riketta*
Psychological Institute, University of Tübingen, Friedrichstr. 21, 72072 Tuebingen, Germany
Received 22 December 2003

Available online 19 August 2004
Abstract
The last two decades have witnessed a surge in interest in research on organizational iden-
tification (OI). This paper presents a comprehensive meta-analysis of this research (k = 96).
Results indicate that (a) OI is correlated with a wide range of work-related attitudes, behav-
iors, and context variables, (b) OI is empirically distinct from its closest conceptual neighbor,
attitudinal organizational commitment (AOC), and (c) the two most common OI measures
(the Mael scale and the Organizational Identification Questionnaire) produce very different
results. It is argued that OI scales, especially the Mael scale, may be preferable over AOC
scales for studies aimed at explaining, and partly also for studies aimed at predicting, work
behavior.
Ó 2004 Elsevier Inc. All rights reserved.
Keywords: Organizational identification; Organizational commitment; Work behavior; Meta-analysis
1. Introduction
Compared with other psychological variables assumed to be relevant to work

behavior, such as ability, job satisfaction, and work motivation, organizational
q
Michael Riketta, Psychological Institute, University of Tübingen, Germany. I am grateful to Rolf van
Dick for helpful comments on this research project and to Timo Förster for assistance with coding.
*
Fax: +49-7071-29-5899.
E-mail address: [email protected].
0001-8791/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.jvb.2004.05.005
M. Riketta / Journal of Vocational Behavior 66 (2005) 358–384 359
identification (OI) has received little attention as a unique research topic until re-
cently. Although the first detailed model of was OI proposed by March and Simon
(1958), only a few studies that explicitly dealt with OI were published in the follow-
ing 20 years (e.g., Brown, 1969; Lee, 1969; Patchen, 1970; Rotondi, 1975a, 1975b). In
the 1970s, Porter and his colleagues (e.g., Porter, Steers, Mowday, & Boulian, 1974;
Mowday, Steers, & Porter, 1979) included identification as a component of attitudi-
nal organizational commitment (AOC) in their seminal definition of this construct.
Since then, several researchers have treated the terms OI and AOC as synonyms
(e.g., Griffin & Bateman, 1986; Mathieu & Zajac, 1990).
Researchers in organizational behavior, social psychology, and communication
re-discovered OI as a unique construct in the late 1980s. After Ashforth and Mael
(1989) outlined the relevance of these social psychological theories to organizational
behavior research, the number of empirical and theoretical analyses focussing on OI
as a unique construct and employing these theories increased markedly in that dis-
cipline (e.g., Dutton, Dukerich, & Harquail, 1994; Elsbach, 1999; Mael & Ashforth,
1992; Pratt, 1998; Rousseau, 1998; Tyler, 1999; Wan-Huggins, Riordan, & Griffeth,
1998). At the same time, and seemingly independently of research in organizational
behavior, several social psychologists following the social identity tradition discov-
ered organizational settings as a new field of application of social identity theory
and self-categorization theory (e.g., Abrams, Ando, & Hinkle, 1998; Brown & Wil-
liams, 1984; Ellemers, de Gilder, & van den Heuvel, 1998; Haslam, 2001; van Knip-
penberg & van Schie, 2000). Finally, in communication research, Cheney, Tompkins,
and their colleagues studied the relation of OI to control, socialization, and commu-
nication in organizations (e.g., Barker & Tompkins, 1994; Bullis & Tompkins, 1989;
Cheney, 1983).
To date, about 80 journal articles dealing with OI have been published, about the
half of them since 1998 (PsycINFO, electronic database, update from October 2003).
Moreover, OI has been addressed in special issues of Academy of Management Re-
view (Albert, Ashforth, & Dutton, 2000) and Group Processes and Intergroup Rela-
tions (van Knippenberg & Hogg, 2002). Given this surge in interest in OI, work
that summarizes and organizes the literature on this construct appears desirable.
Although recent qualitative reviews on OI are available, they did not aim at a com-
prehensive overview of this research (e.g., Elsbach, 1999; Haslam, 2001; van Dick,
2004). The present article provides a more comprehensive review and meta-analysis
of empirical research on OI.
This review addresses three issues that are crucial to OI research. First, it gives a
comprehensive quantitative overview of the most often studied correlates of OI. Sec-
ond, it addresses the question of whether the operationalization of OI matters. Many
measures for OI are available. Yet, little is known about the consequences of using
one measure instead of the other. Third, this review presents evidence concerning the
empirical distinctiveness of OI and its closest conceptual neighbor, AOC. Many OI
researchers assert that this construct is distinct from AOC (e.g., Ashforth & Mael,
1989; Pratt, 1998; van Knippenberg & van Schie, 2000), while others challenge this
assumption on both empirical and theoretical grounds (e.g., Ouwerkerk, Ellemers, &
de Gilder, 1999; Sass & Canary, 1991; Stengel, 1987). The present review addresses
360 M. Riketta / Journal of Vocational Behavior 66 (2005) 358–384
this issue by comparing the results from the meta-analysis of OI research with results
from previous meta-analyses of AOC research.
To date, only one meta-analysis of OI research has been presented (Fontenot
& Scott, 2000; see also Fontenot & Scott, 2003). However, this prior work has
several limitations, which the present study overcomes. First, the Fontenot and
Scott meta-analysis focused on only four correlates (AOC, job satisfaction, intent
to leave, and organizational tenure). By contrast, in the present meta-analysis, all
correlates were considered for which a sufficient number of studies could be re-
trieved. Second, the prior meta-analysis included only studies conducted until
2000. Since then, many more OI studies have been published. Third, Fontenot
and Scott did not compare their results for OI with findings from AOC research.
Thus, they left open the question of whether the results were unique for OI re-
search. Finally, correlations were not disattenuated in the Fontenot and Scott
meta-analysis so that it probably underestimated population correlations (cf.
Hunter & Schmidt, 1990).
In the following, at first, I will present definitions of OI and discuss the relation
between OI and AOC. A meta-analysis of OI studies follows. Next, I will compare
the correlations of OI and AOC with other variables to explore the distinctiveness
of these two constructs.
2. Organizational identification and attitudinal organizational commitment
Many definitions of OI have been proposed. Most of them conceptualize OI as a

cognitive construct, in particular, as the congruence of individual and organizational
values (Hall, Schneider, & Nygren, 1970; Pratt, 1998; Stengel, 1987), as the ‘‘percep-
tion of oneness with or belongingness to’’ the organization (Ashforth & Mael, 1989,
p. 34), or as the process of incorporating the perception of oneself as a member of a
particular organization into oneÕs general self-definition (Dutton et al., 1994; Els-
bach, 1999; Rousseau, 1998). Having a different focus, OÕReilly and Chatman
(1986), following Kelman (1961), defined OI in affective-motivational terms, that
is, as based on attraction and the desire to maintain an emotionally satisfying self-
defining relationship with the identification object. Further, the definition derived
from social identity theory—which is the most pervasive theoretical framework in
contemporary OI research (Haslam, 2001; van Dick, 2001)—combines cognitive
and affective components. According to this theory, social identity is ‘‘that part of
an individualÕs self-concept which derives from his knowledge of his membership
of a social group (or groups) together with the value and emotional significance at-
tached to that membership’’ (Tajfel, 1978, p. 63; see Abrams et al., 1998; Benkhoff,
1997a; and Ouwerkerk et al., 1999; for an application of this definition to organiza-
tions). Finally, perhaps the most comprehensive definition of OI has been proposed
by Patchen (1970), who used the term OI for ‘‘a variety of separate, though related
phenomena . . . (1) feelings of solidarity with the organization; (2) [attitudinal and
behavioral] support for the organization; and (3) perception of shared characteristics
with other organizational members’’ (p. 155). Despite their heterogeneity, all these
definitions imply that the organizational member has linked his or her organizational
membership to his or her self-concept, either cognitively (e.g., feeling a part of the
organization; internalizing organizational values), emotionally (pride in member-
ship), or both. The link between self-concept and organization is used herein as
the working definition of OI.
Depending on its specific definition, OI is more or less similar to other concepts of
organizational behavior research such as involvement, satisfaction, and, above all,
AOC (Ashforth & Mael, 1989; Mael & Tetrick, 1992; Pratt, 1998). The distinction
between AOC and OI, because it is the fuzziest one, requires examination.
According to its seminal definition, AOC is ‘‘the relative strength of an indi-
vidualÕs identification with and involvement in a particular organization’’ (Mow-
day et al., 1979, p. 226). It has at least three related factors: (1) acceptance of the
organizationÕs goals and values; (2) willingness to work hard for the organization;
and (3) a strong desire to remain in the organization. Recently, the reconceptual-
ization of AOC by Allen and Meyer (1990) has received growing attention in re-
search. These authors distinguished three forms of commitment: affective,
continuance, and normative. The definition of affective organizational commit-
ment resembles Mowday et al.Õs definition and includes the following components:
‘‘employeeÕs emotional attachment to, identification with, and involvement in, the
organization’’ (Allen & Meyer, 1990, p. 1). In the present work, the term AOC
encompasses both the definition by Mowday et al. and the definition by Allen
and Meyer. If one looks at the cited definitions, there is a clear overlap between
AOC and OI. In particular, identification is explicitly included in these definitions
of AOC.
There is also some overlap at the operational level. The two most often used AOC
scales are the Organizational Commitment Questionnaire (OCQ; Mowday et al.,
1979) and the Affective Commitment Scale (ACS; Allen & Meyer, 1990). Items in
these scales refer to (a) emotional attachment to the organization (e.g., OCQ: ‘‘I
am proud to tell others that I am part of this organization’’; ACS: ‘‘I do not feel
emotionally attached to this organization’’), (b) involvement in organizational issues
(OCQ: ‘‘I really care about the fate of the organization’’; ACS: ‘‘I feel as if this orga-
nizationÕs problems are my own’’), (c) value congruence (OCQ: ‘‘I find that my val-
ues and the organizationÕs values are very similar’’; not represented in the ACS), and
(d) willingness to stay with the organization (OCQ: ‘‘There is not too much to be
gained by sticking with this organization indefinitely’’; ACS: ‘‘I would be very happy
to spend the rest of my career with this organization’’). Similar items are included in
many OI measures. For example, eight of the 25 items of the Organizational Identi-
fication Questionnaire (OIQ; Cheney, 1983), one of the most often used OI measures,
are virtually identical with ACS and OCQ items.
Several researchers have asserted that AOC is a broader or vaguer construct
than OI (e.g., Ashforth & Mael, 1989; Edwards, 2002; Pratt, 1998; van Dick,
2004). The most influential scholars distinguishing between OI and AOC are Ash-
forth and Mael (1989), who stated that OI is ‘‘a perceptual cognitive construct
that is not necessarily associated with any specific behaviors or affective states’’
(p. 21). Their view has found many supporters (e.g., Edwards, 2003; Elsbach,
1999; Iyer, Bamber, & Barefield, 1997; Mael & Tetrick, 1992; Pratt, 1998; van
Knippenberg & van Schie, 2000; Wan-Huggins et al., 1998; but see Harquail,
1998; for a critique of this cognitively narrowed view). One of the most common
OI scales, the Mael scale (Mael & Tetrick, 1992), is based on this definition. Its
items do not overlap with items from OCQ and ACS (e.g., ‘‘When I talk about
this organization, I usually say ÔweÕ rather than they’’; ‘‘When someone praises
this organization, it feels like a personal compliment’’; ‘‘This organizationÕs suc-
cesses are my successes’’).
Other researchers evidently considered both AOC and OI as closely related or
even interchangeable, using the terms as synonyms (e.g., Benkhoff, 1997a; Ouwer-
kerk et al., 1999; Wallace, 1993), citing studies on OI as examples of commitment
research (e.g., A. Cohen, 1992; Mathieu & Zajac, 1990) or vice versa (e.g., Elsbach,
1999), and/or incorporating items from OI scales in their commitment measures (e.g.,
Buchanan, 1974; Mottola, Bachman, Gaertner, & Dovidio, 1997) or vice versa (e.g.,
Cheney, 1983). Finally, some scholars did not mention AOC at all in their studies on
OI (partly because that concept was not yet introduced at that time of their studies)
so that it is unclear how they saw the relationship between OI and AOC (e.g., Padaki
& Gandhi, 1981; Patchen, 1970; Rotondi, 1975a, 1975b). All in all, there is no con-
sensus among OI researchers regarding the conceptual relation between OI and
AOC.
Empirical evidence on the OI–AOC relation is also inconsistent. The correlation
between OI and AOC varied across studies between values around 0 (Beerman, 1987;
van Dick, Wagner, & Gautam, 2002; van Dick, Wagner, Lemmer, & Britt Lima,
2002) and .8 or above (Potvin, 1991; Sass & Canary, 1991; Sauer, 1997). Using struc-
tural equation techniques, Bergami and Bagozzi (2000), Gautam, van Dick, and
Wagner (in press), Mael and Tetrick (1992), and van Knippenberg and Sleebos
(2001) found that AOC and OI fit a two-factor model better than a single-factor
model. Nevertheless, the correlation between the two constructs was strong in these
studies (rs > .50). This begs the question of whether the distinctiveness of AOC and
OI matters for any theoretically or practically significant purposes—for example,
when the goal is to explain or predict work outcomes.
One way of addressing this question is to compare the correlates of OI and AOC.
If no differences can be detected, or if they occurred only on correlates that were of
minor importance for research and practice, it would be wise to abandon one of
these terms to avoid conceptual confusion and redundant research efforts (see Sass
& Canary, 1991; for an empirical study based on this reasoning). Accordingly, in
the present study, the overlap in the phenomena tapped by OI and AOC measures
was examined by comparing their correlates. Given the heterogeneity of OI mea-
sures, some OI measures may be empirically distinct from AOC measures while oth-
ers are not. Furthermore, the same OI measure may have discriminant validity from
the ACS but not from the OCQ or vice versa. To account for this, I conducted sep-
arate analyses for the most often used OI measures on the one hand and for the ACS
and the OCQ on the other. In the following, I will give a meta-analytic overview of
the correlates of OI and then compare the results with findings from previous meta-
analyses of AOC research.
3. Meta-analysis of research on organizational identification
3.1. Study collection and coding of study characteristics
All empirical studies were of potential relevance that: (a) explicitly dealt with OI
and (b) operationalized it with a measure labeled OI. Both published and unpub-
lished studies were considered. To identify published studies, I searched the keyword
‘‘organizational identification’’ in the electronic databases PsycINFO, Business
Source Premier, and Social Sciences Citation Index (all databases updated last time
in May 2003) and on the Internet with several search engines. To identify unpub-
lished studies, I sent e-mails to 21 researchers who either had published on OI or
whom I knew to have conducted unpublished studies on OI (e.g., from the research-
ersÕ Internet homepages). I asked those researchers whether they had unpublished
data on OI and if yes, whether they would provide me with them. Eleven researchers
responded, and 8 provided usable unpublished data. The references in every research
report obtained were examined to identify other relevant studies.
The final meta-analysis was conducted on data from 96 independent samples (to-
tal N = 20,905; see References for the single studies). For each sample, the following
information was coded: sample size, name, and source the OI measure used, corre-
lations of OI with other variables (called correlates in the following), and reliabilities
of the measures involved in the coded correlations (i.e., internal consistencies or
split-half reliabilities). Only those correlates were considered in the final analysis
for which data from at least five independent samples were available (see first column
of Table 1 for these correlates). Data referring to four of these correlates (job scope/
challenge, organizational prestige, in-role performance, and extra-role performance)
were coded independently by me and a graduate student who received detailed cod-
ing instructions. Intercoder agreement was at least 95% for the correlations and reli-
ability estimates in each category. Inconsistencies were resolved by discussion. Given
this high agreement, I coded all other information by myself. A list of the analyzed
samples with their coded characteristics is available from the author.
3.2. Computations
The computational procedure followed the meta-analytic approach of Hunter and

Schmidt (1990). Hunter and Schmidt suggested that a meta-analysis aggregate data
across studies and correct the data for artifacts as far as possible. The current meta-
analysis controlled for the artifacts of sampling and measurement error.
To correct for measurement error, I disattenuated each coded correlation, that is,
divided it by the product of the square-roots of the reliability estimates of the in-
volved variables. With self-report data (except factor scores), the internal consisten-
cies or split-half reliabilities coded for the respective sample were used as reliability
estimates. Where no such estimate was coded, the N weighted average of all esti-
mates coded for the focal variable was inserted. In computing these averages, I
substituted the value 500 for all sample sizes larger than 500 to avoid that single
studies with extraordinarily large sample sizes overly dominated the weighted
Table 1
Correlates of organizational identification
Correlate k n rc SDrc SDrho % v2 CI
Demographic variables
Organizational tenure 25 5305 .13 .17 .11 30 83.67*** .06 .32
Mael scale 6 1230 .16 .14 .12 27 22.39*** .05 .36
Age 21 4802 .12 .17 .15 18 117.40*** .13 .38
Mael scale 5 751 .07a .09 .03 90 5.55 .02 .12
OIQ 2 225 .60 .23 .22 8 24.75*** .24 .96
Job level 5 708 .24 .13 .09 46 1.94* .11 .42
Mael scale 2 428 .31 .11 .08 40 5.00* .17 .44
Female gender 18 4331 .04a .11 .09 41 43.89*** .18 .11
Mael scale 7 1100 .11a .16 .14 29 24.31*** .34 .11
Education 5 549 .06a .12 .05 80 6.31 .15 .02
Mael scale 3 330 .01a .09 .00 100 2.23 — —
Work-related attitudes
AOC 16 4263 .78 .46 .46 1 2499.48*** .02 1.54
Mael scale 3 500 .79 .08 .06 47 6.44* .69 .89
OIQ 3 1022 .90 .08 .07 9 33.88*** .78 1.02
AOC measured with OCQ 8 2228 .79 .45 .45 1 1182.22*** .06 1.53
OIQ 2 572 .94 .01 .00 100 0.69 — —
AOC measured with ACS 7 1791 .71 .12 .12 11 61.82*** .53 .90
Mael scale 2 237 .74 .10 .07 53 3.75 .62 .85
Occupational attachment 13 2445 .47 .25 .24 9 145.73*** .08 .87
Mael scale 3 534 .39 .12 .09 43 7.02* .24 .54
Work group attachment 20 3867 .52 .24 .23 8 240.73*** .14 .91
Mael scale 4 446 .46 .17 .13 39 10.24* .25 .67
OIQ 2 238 .92 .09 .08 16 12.29*** .78 1.05
Job satisfaction 38 8759 .54 .20 .20 9 415.71*** .22 .86
Mael scale 7 1298 .47 .20 .19 12 56.56*** .16 .77
OIQ 8 1283 .68 .19 .18 9 85.27*** .38 .97
Organizational satisfaction 6 1530 .59 .17 .16 8 76.97*** .33 .86
Mael scale 2 297 .52 .05 .01 96 2.09 .50 .54
Job involvement 12 2837 .61 .13 .11 26 46.09*** .44 .79
Mael scale 6 1703 .60 .11 .09 39 15.26** .46 .75
Context characteristics
Job scope/challenge 10 1699 .33 .11 .07 61 16.42 .12 .54
Mael scale 2 342 .26 .08 .00 100 1.99 — —
Organizational prestige 16 5257 .56 .13 .12 19 82.57*** .37 .75
Mael scale 8 2423 .55 .13 .11 22 35.89*** .37 .73
Work-related intentions and behaviors

Intention to leave 34 7243 .48 .20 .19 13 277.27*** .79 .17
Mael scale 9 2055 .35 .15 .12 28 31.89*** .55 .15
OIQ 6 991 .64 .13 .12 20 3.19*** .84 .44
In-role behavior 16 3009 .17 .14 .11 44 36.53** .01 .35
Mael scale 5 891 .17 .17 .14 34 14.52** .06 .40
Extra-role behavior 25 6644 .35 .19 .17 15 163.35*** .07 .63
Mael scale 9 2475 .39 .15 .13 22 4.23*** .18 .61
Absenteeism 6 1581 .01a .06 .00 100 5.22 — —
Table 1 (continued)
Notes. Results for the Mael scale and the OIQ are reported only if k > 1. k, number of analyzed corre-
lations. n, sum of the sizes of the analyzed samples. rc, mean correlation corrected for sampling error and
attenuation. SDrc, observed standard deviation of the disattenuated correlations. SDrho, estimated stan-
dard deviation of the mean population correlation. %, percent variance of the observed correlations
attributable to artifacts. v2, result of the significance test for unaccounted variance, with df = k 1. CI:
90% credibility interval around rc. Mael scale, studies using the scale developed by Mael (see Mael and
Tetrick, 1992). OIQ, studies using the Organizational Identification Questionnaire (Cheney, 1983). AOC,
attitudinal organizational commitment. ACS, Affective Commitment Scale (Allen and Meyer, 1990).
OCQ, Organizational Commitment Questionnaire (Mowday et al., 1979).
a
The 95% confidence interval around rc includes zero.
*
p 6 .05.
**
p 6 .01.
***
p 6 .001.
average. The average reliability estimates for all relevant variables were as follows:
AOC .79, extra-role performance (self-rating) .80, in-role performance (self-rating)
.83, intent to leave .81, job involvement .75, job satisfaction .84, job scope/challenge
.76, occupational attachment .83, OI .84, organizational prestige .79, organizational
satisfaction .86, and work group attachment .82. With other-reported data (here,
peer or supervisor ratings of in-role or extra-role performance), I used estimates
of interrater reliability rather than internal consistencies or split-half reliabilities in
disattenuating. Using the latter estimates would have underestimated true correla-
tions because a type of measurement error that is specific to othersÕ ratings (namely,
construct independent variation in responses due to rater idiosyncracies) would have
been uncorrected. Because no study included in the present meta-analysis reported
interrater reliabilities, Viswesvaran, Ones, and SchmidtÕs (1996) meta-analytical esti-
mates of the interrater reliability of supervisor ratings (.52) and peer ratings (.42)
were used to correct correlations computed from such ratings. Finally, so-called hard
data (here: age, gender, education, organizational tenure, job level, absenteeism, and
objective indicators of in-role or extra-role performance such as sales figures and
numbers of contributions to suggestion programs) and factor scores of self-report
data were assigned reliability estimates of 1.00 in the disattenuation procedure
because (a) the forms of measurement error corrected in the other data were not
relevant here and (b) no other relevant reliability information for those data were
available from the analyzed studies.
In the next step, I averaged the disattenuated correlation coefficients for each var-
iable across independent samples. Following the recommendations of Hunter and
Schmidt (1990), to correct for sampling error, I weighted every corrected correlation
coefficient with the product of sample size and the reliability coefficients for the two
correlated variables. Sample size weights were again limited to 500. The average
weighted correlation coefficients are estimates of the population correlations. Note
that these estimates are necessarily flawed by all artifacts not corrected for, here,
all artifacts besides measurement and sampling error. To test the corrected mean cor-
relations for significance, I computed 95% confidence intervals around them using
the formulae by Hunter and Schmidt (1990). In the following, mean correlation coef-
ficients are called significant only if the confidence interval excludes zero.
Another parameter of interest was the variance of the population correlations.

The estimate recommended by Hunter and Schmidt (1990) was used, i.e., the differ-
ence between the variance of the corrected correlation coefficients and their average
squared standard error. The latter term is an estimate of the variance attributable to
the corrected artifacts. The estimated variance of the population correlations was
tested for significance with the v2 test of Hunter and Schmidt (1990). A significant
result would indicate that there is more than one population correlation, that is, that
moderators of the correlation exist. Further, 90% credibility intervals around mean
corrected correlations were computed by multiplying the square-root of the esti-
mated variance of the population correlations with the appropriate z value.
3.3. Results
Table 1 displays the results of the meta-analysis. Both overall results and results
for the the Mael scale and the OIQ are presented. These two measures were used in
their original form or a shortened version (i.e., a version consisting of at least half of
the items of the original scale) in 26 and 11 of the analyzed samples, respectively. All
other measures were used in less than eight samples. (A complete list of the measures
and their use frequencies is available from the author.) All ps reported in the follow-
ing are two-tailed. Results with p 6 .05, p 6 .10, and p > .10 are labeled significant,
marginally significant, and nonsignificant, respectively.
3.3.1. Overall analysis

For ease of presentation, the correlates were subdivided in four categories: demo-
graphic variables, work-related attitudes, context characteristics, and work-related
intentions and behaviors. In the demographic variables category, organizational ten-
ure, age, and job level related significantly to OI. The correlations were positive and
small to medium according to J. CohenÕs (1988) effect-size classification (rs = .13, .12,
and .24, respectively). The correlations with gender and education were nonsignifi-
cant (rs = .04 and .06, respectively). The variables in the work-related attitudes
category all correlated significantly and positively with OI. The highest mean corre-
lation emerged between OI and AOC (r = .78). The mean correlations were similarly
strong for studies using the ACS and studies using the OCQ (rs = .71 and .79, respec-
tively; difference: z < 1). Further, OI correlated strongly with attachment to (i.e.,
either commitment to or identification with) oneÕs occupation and oneÕs work group
(rs = .47 and .52, respectively). Moreover, there was a strong overlap with job and
organizational satisfaction and job involvement (rs = .54, .59, and .61, respectively).
Likewise, the correlations with the two context characteristics considered (job scope/
challenge and organizational prestige) were significant and positive. The correla-
tions were medium to large in size (rs = .33, and .56, respectively). Finally, of the
work-related intentions and behaviors considered, intention to leave correlated
strongly and negatively with OI (r = .48), and in-role and extra-role performance
correlated weakly and moderately with OI (rs = .17 and .35, respectively). All cor-
relations were significant. The fourth variable in this category, absenteeism, was
unrelated to OI (r = .01).
Note that the results of the analyzed OI studies were heterogeneous. For all vari-
ables except education, job scope/challenge, and absenteeism, (a) the v2 test lead to
the rejection of the null hypothesis that the individual correlations were equal (with
all ps < .01), and (b) more than half of the variance of the observed individual cor-
relations remained unexplained. Together, this indicates the likely existence of mod-
erators for these variables. The following section deals with the question of whether
the operationalization of OI is a potential moderator. For this purpose, all studies
using either the OIQ or the Mael scale were analyzed separately.
3.3.2. Analyses for single OI measures

Separate correlations for the Mael scale and the OIQ were computed for every
correlate for which at least two samples using the respective measure were available.
In these analyses, all studies were considered that used either the original versions of
these measures or shortened versions that included at least half of the items in the
original scales. To test for the moderating influence of using one of these scales in-
stead of any other scale, I conducted Hunter and SchmidtÕs (1990) z test on the dif-
ferences between samples in which the Mael scale (OIQ) was used and all
independent samples in which other scales were used.
With regard to the Mael scale, data were available for all correlates considered in
the overall analyses except for OCQ and absenteeism. In general, the results paral-
leled the results from both the total analysis and the analysis for studies that used
another OI scale (see Table 1 for details). The only significant difference emerged
in the category work-related intentions and behavior. The Mael scale correlated signif-
icantly less strongly with intent to leave (r = .35) than did the other OI measures
(r = .53, difference: z = 2.66, p = .01; all other zs < 1.57).
Correlations for the OIQ could be computed for six correlates only (see Table 1
for details). On four of them, significant differences between the OIQ and the other
measures emerged. In particular, the OIQ correlated significantly more strongly with
age (r = .60), work group attachment (r = .92), job satisfaction (r = .68), and intent
to leave (r = .64) than the other measures (rs = .09, .59, .42, and .45, respectively;
zs > 2.21, ps < .04). Further, the OIQ correlated somewhat more strongly with AOC
in total (r = .90) and with the OCQ in particular (r = .94) than did the other OI mea-
sures (rs = .73 and .74, respectively). However, these differences were nonsignificant
(zs < 1.49).
3.4. Discussion
In interpreting the data, one should bear in mind that the correlations were very
heterogeneous in general. Thus, any conclusions drawn from the overall analysis
may not apply to single settings or measures in particular. Hence, the following
arguments apply to the fictitious average OI study. A particular real study may have
yielded quite different outcomes or may do so in the future.
The highest mean correlation obtained was the one with AOC (r = .78). This sug-
gests that the average OI study has focused on a construct that overlapped largely
with what is measured by AOC measures. However, the correlation is far from per-
fect; it corresponds to only 62% shared variance. Thus, OI as measured in the typical
OI study may well have unique features as compared with AOC in its common oper-
ational form. The next section deals with this possibility more in detail. Further, the
correlations with all other considered constructs to which OI is conceptually related
(job and organizational satisfaction, job involvement, and occupational and work
group attachment) were markedly lower (shared variances below 40%). This suggests
that OI relates to but is distinct from these constructs.
The present analyses also revealed significant mean correlations with a range of
variables that may be either antecedents or consequences of OI, including tenure,
job scope/challenge, organizational prestige, intent to leave, and in-role and extra-
role performance. Because only correlational studies were analyzed, no attempt is
made here to interpret the results in terms of causality. The interested reader can find
detailed models of the causal relation of OI with most of the correlates considered
herein elsewhere (e.g., Bergami & Bagozzi, 2000; Cheney, 1983; Mael & Ashforth,
1992; van Knippenberg, 2000; Wan-Huggins et al., 1998). For present purposes, it
seems sufficient to note that the directions of all significant mean correlations re-
ported above are both intuitively plausible and consistent with the common causal
assumptions of OI researchers. From a more practical perspective, the significant
and partly strong correlations with those variables indicate that OI may be a useful
predictor of several variables that are relevant to organizational practice.
In light of the mentioned heterogeneity of the results of OI studies, it is revealing
to look at the results for single measures. The Mael scale was used in the analyzed
samples most often. The results of the studies using this scale were close to the results
for studies using other measures as well as for the results for all measures. Hence, the
Mael scale seems to be the most representative OI measure with regard to its empir-
ical outcomes. Another interesting finding is that the correlations involving the Mael
scale showed much less variation than the correlations involving all OI measures.
For example, for several correlates, the correlations from all OI studies were strongly
significantly heterogeneous (ps < .001) but the correlations from studies using the
Mael scale were not (i.e., with regard to the correlates age, AOC as measured with
the ACS, and organizational satisfaction). Moreover, the credibility intervals were
typically narrower for correlations computed from studies using the Mael scale than
for the correlations computed from all studies. This suggests that (a) in general, the
operationalization of OI does matter for the correlations obtained (at least for the
variation in these correlations) and (b) the studies using the Mael scale comprise a
relatively homogenous subgroup within OI research with regard to their findings.
The second most often used measure in the samples analyzed was the OIQ.
Although this scale is the most common OI measure in communication research
(see Miller, Allen, Casey, & Johnson, 2000), few studies have been reported correla-
tions with the OIQ that pertained to the same variables. In the studies analyzed here,
more than two correlations involving the OIQ were available only for six correlates.
Yet, it is striking that all six correlations deviated markedly (and four of them sig-
nificantly) from the average across the studies that used another measure. Especially
important, the OIQ was more strongly related to three constructs that are conceptu-
ally similar to OI: AOC, work group attachment, and job satisfaction. Thus, the OIQ
seems to have less discriminant validity with regard to AOC scales than do the other
OI measures in total and the Mael scale in particular. Specifically, the very strong
correlation with AOC (r = .90) suggests that the OIQ is almost interchangeable with
AOC measures, especially with the OCQ (r = .94). In this respect, the present re-
search is in accordance with the conclusions that Sass and Canary (1991) drew from
their study, namely that OIQ and OCQ refer to the ‘‘same cluster of attitudes’’ (p.
275). Further, the OIQ correlated more strongly with intent to leave and age than
did the other OI measures, including the Mael scale. The former correlation might
point to a higher usefulness of the OIQ as compared with the average other OI mea-
sure when the goal is to predict intent to leave. However, because the individual cor-
relations were strongly significantly heterogenous in all of these analyses (ps < .001),
the aforementioned conclusions regarding the OIQ are only tentative.
Because OI is strongly correlated with AOC, one may wonder whether it differs
from this construct with respect to its correlates. In other words, does it make a dif-
ference whether one uses an OI scale or an AOC scale in predicting or explaining
other variables? The following section addresses this question.
4. Meta-analytic comparison between organizational identification and attitudinal

organizational commitment
As mentioned above, OI and AOC would prove distinct if they differed in the
strength of their respective correlations with same variables. To explore whether this
is the case, I compared the results reported above with results from previous meta-
analyses of AOC studies. Because the vast majority of AOC studies used either the
ACS or OCQ, I considered only studies that used one of these measures. This had
the advantage that the analyzed studies were rather homogenous with respect to
operationalization of AOC. Yet, because ACS and OCQ partly differ in the contents
of their items (see above), I conducted all comparisons separately for the ACS and
the OCQ. In addition, to account for the heterogeneity in the operationalization of
OI, I conducted separate analyses for the Mael scale and the OIQ alongside overall
analyses, which referred to all OI measures.
4.1. Study search and computations
To identify meta-analyses that reported correlations involving either the ACS or

the OCQ, I searched the keywords ‘‘commitment’’ and ‘‘meta-analysis’’ in the dat-
abases Business Source Premier, PsycINFO, and Social Sciences Citation Index
(all databases updated last time in May 2003). I considered only those meta-analyses
that (a) reported mean correlations corrected for sampling error and unreliability,
(b) dealt with at least one correlate that was considered in the above analyses for
OI, and (c) reported, or allowed for the computation of, separate correlations for
the ACS and the OCQ. When results referring to the same correlate were available
from two or more meta-analyses, I considered only the result based the largest num-
ber of studies. The following meta-analyses met all these criteria and contributed at
least one mean correlation to the analyses reported below: Aven, Parker, and McE-
voy (1993), Cohen (1993), Meyer, Stanley, Herscovitch, and Topolnytsky (2002),
Riketta (2002), Tett and Meyer (1993), and Wallace (1993). Because Riketta
(2002) and Wallace (1993) did not report separate results for ACS or OCQ with re-
gard to the correlates that were of interest in the present study, I re-analyzed their
datasets. In doing so, I included only the studies that used the OCQ (either in its ori-
ginal 15-item or a shortened version, as long as the latter contained at least eight
items of the long version). The data allowed for comparisons between OI and
AOC on 11 correlates.
4.2. Results
Table 2 shows the results for the ACS or the OCQ. To enable a quick comparison
with the results of the above analyses for OI, Table 2 displays the corresponding cor-
rected mean correlations from Table 1 in addition (i.e., correlations for OI scales in
Table 2
Comparison of findings from research on organizational identification (OI) and attitudinal organizational
commitment
Variable rc from OI studies Studies using the ACS Studies using the OCQ
Total Mael OIQ k n rc SDrc k n rc SDrc
Organizational .13 .16 — 51 18630 .16 .13 48 26175 .10 .09
tenure
Age .12B .07a,B .60A,B 53 21446 .15 .10 53 26402 .20 .10
Female gender .04 .11 — 32 11764 .03 .10 16 8979 .01 .12
Education .06 .01 — 32 11491 .02 .12 —
Occupational .47 .39 — 13 3599 .51 .11 16 6414 .42 .15
attachment
Job satisfaction .54A .47A .68 69 23656 .65 .14 47 14597 .68 —
Job involvement .61a .60 — 16 3625 .53 .14 —
Intent to leave .48a .35A .64 51 17282 .56 .22 24 5403 .53 —
In-role .17 .17 — 25 5938 .16 .12 47 11072 .18 .13
performance
Extra-role .35B .39B — 22 6277 .32 .15 12 3446 .23 .08
performance
Absenteeism .01A — — 22 3543 .15 .09 —
Notes. The data referring to the OCQ are from the following sources: Cohen (1993) for age and tenure;
Aven et al. (1993) for female gender. Tett and Meyer (1993) for intent to leave and job satisfaction
(sample-size weighted averages of the results for OCQ-15 and OCQ-9; results for intent to leave are the
results that Tett and Meyer reported for both turnover intention and withdrawal cognition); Riketta
(2002) for in-role and extra-role performance (re-analyzed); Wallace (1993) for occupational attachment
(re-analyzed). The data referring to the ACS are from Meyer et al. (2002). rc, mean correlation corrected
for sampling error and attenuation. ACS, Affective Commitment Scale (Allen and Meyer, 1990). OCQ,
Organizational Commitment Questionnaire (Mowday et al., 1979). k, number of analyzed correlations. n,
sum of the sizes of the analyzed samples. SDrc, observed standard deviation of the disattenuated corre-
lations. Total, all OI studies. Mael, studies using the Mael scale. OIQ, studies using the OIQ.
a
Differs at p 6 .10 from the corresponding rc for the ACS.
A
Differs at p 6 .05 from the corresponding rc for the ACS.
B
Differs at p 6 .05 from the corresponding rc for the OCQ.
total, the Mael scale, and the OIQ). As far as the necessary data were available, all
possible differences in the correlations for all OI scales in total, the Mael scale, and
the OIQ on the one hand and ACS and OCQ on the other were tested for significance
with Hunter and SchmidtÕs (1990) z test.
The analyses for all OI measures revealed at least marginally significant differ-
ences on 6 of the 11 correlates. In particular, OI measures in total differed from
the ACS at least marginally significantly on the correlations with job satisfaction
(r = .54 vs. .65 for OI measures in total and the ACS, respectively; difference:
z = 2.96, p < .01), job involvement (r = .61 vs. .53, z = 1.65, p = .10), intent to leave
(r = .48 vs. .56, z = 1.77, p = .08), and absenteeism (r = .01 vs. .15, z = 4.59,
p < .001). Further, OI measures in total differed from the OCQ significantly on the
correlations with age (r = .12 vs. .20, z = 1.98, p = .05) and extra-role performance
(r = .35 vs. .23, z = 2.59, p = .01). Another marked difference between OI measures
in total and the OCQ and emerged with regard to job satisfaction (r = .54 vs. .68).
Significance testing was not possible in this case due to missing information. (All
other zs < 1.)
Results for the Mael scale were very similar. At least marginally significant differ-
ences occurred on four of the ten correlates for which comparison data were avail-
able. In particular, the Mael scale differed from the ACS at least marginally
significantly on the correlations with age (rs = .07 vs. .15, z = 1.71, p = .09), job sat-
isfaction (r = .47 vs. .65, z = 2.36, p = .02), and intent to leave (r = .35 vs. .56,
z = 3.67, p < .001). The difference from job involvement was in the same order as
in the analyses for OI measures but nonsignificant (r = .60 vs. .53, z = 1.22). The
Mael scale differed from the OCQ significantly on the correlations with age
(rs = .07 vs. .20, z = 2.85, p < .01) and extra-role performance (r = .39 vs. .23,
z = 2.93, p < .01). Further, like OI measures in total, the Mael scale differed mark-
edly from the OCQ on the correlation with job satisfaction (r = .47 vs. .68). Again,
no significance testing was possible in this case. (All other zs < 1.60.)
The available data permitted comparisons for the OIQ on only three corre-
lates. The OIQ differed from both ACS and OCQ significantly on the correlation
with age (r = .60 vs. .15 and .20, z = 2.81, p < .01, and z = 2.45, p = .01, respec-
tively). Note that the direction of the difference is opposite to the one for all
OI scales in total and the Mael scale in particular. Further, the OIQ showed
(nearly) the same correlation with job satisfaction as did the ACS and the
OCQ (r = .68 vs. .65 and .68, z < 1 and no significance testing possible, respec-
tively). Finally, the OIQ tended to show a stronger correlation with intent to
leave than did the ACS and the OCQ (r = .64 vs. .56 and .53, respectively),
although the differences were nonsignificant (z = 1.28) and not testable for signif-
icance, respectively.
4.3. Discussion
Despite the heterogeneity of the findings of OI research, this analysis revealed sev-
eral statistically significant differences in the average outcomes of OI and AOC re-
search. The two most important findings can be summarized as follows. First, OI
seems to overlap less strongly with job satisfaction and more strongly with job
involvement than does AOC. This hints to substantive differences between the con-
structs of OI and AOC. Second, OI seems to differ from AOC in the correlations
with some work-related behaviors and intentions. Specifically, OI was less strongly
related to absenteeism and intent to leave than was AOC but was more strongly re-
lated to extra-role performance.
All mentioned differences except the one for job involvement were significant also
in the analyses for studies that used the Mael scale. This confirms the above conclu-
sion that the Mael scale is a rather typical or representative OI measure. By contrast,
the OIQ proved more similar to the AOC measures than to the typical OI measure.
Specifically, like the ACS and OCQ, the OIQ correlated more strongly with age, job
satisfaction, and intent to leave than did all OI measures in total and the Mael scale
in particular. The correlation with job satisfaction was even virtually identical to the
correlations of the ACS and OCQ with this variable. Thus, the present findings are
consistent with Miller et al.Õs (2000) and Sass and CanaryÕs (1991) conclusion that the
OIQ is almost interchangeable with AOC measures.
How can one explain the observed differences between OI in general and the Mael
scale in particular on the one hand and the ACS and the OCQ on the other? In gen-
eral, these differences seem to be consistent with the fact that many OI scales and
especially the Mael scale were designed to measure a more narrowly defined con-
struct than AOC. On the one hand, this fact can explain the relatively low correla-
tion of OI with job satisfaction, absenteeism, and intent to leave. These variables can
have many other causes apart from identification. Some of these causes may be cap-
tured by AOC scales better than by OI scales. For example, some items of the OCQ
and the ACS refer to willingness to stay with the organization and to such evalua-
tions of the organization that are not necessarily related to identification but may
indicate or contribute to job satisfaction, such as, respectively, ‘‘I would be very hap-
py to spend the rest of my career with this organization’’ (ACS) or ‘‘There is not too
much to be gained by sticking with this organization indefinitely’’ (OCQ) and ‘‘For
me this is the best of all possible organizations to work for’’ (OCQ) or ‘‘I enjoy dis-
cussing my organization with people outside it’’ (ACS). Some OI scales, among them
the widely used Mael scale (and also, e.g., the scales by Bergami & Bagozzi, 2000;
Lee, 1969; OÕReilly & Chatman, 1986), do not include such items. This can explain
why OI scales in total and the Mael scale in particular relate less strongly to those
variables than do AOC measures.
On the other hand, the narrower focus of OI measures can explain why the cor-
relation of OI with extra-role behavior and job involvement is relatively high. Extra-
role behavior is defined as voluntary behavior that is beneficial to the organization
(Organ, 1988). Among others, the motivation for such a behavior may stem from
internalization of organizational norms and emotional attachment to the
organization (van Knippenberg, 2000). These two variables, however, are at the core
of most definitions and measures of OI. Thus, in general, OI measures may be better
predictors of extra-role behaviors than are AOC measures because the former focus
more narrowly on these crucial causes of extra-role behaviors than do the latter,
which may also refer to variables that are not relevant to extra-role behaviors. Fur-
ther, job involvement overlaps conceptually with intrinsic motivation. Because OI as

commonly defined is likely to produce such an intrinsic motivation (van Knippen-
berg, 2000), OI should be an important predictor of job involvement. Thus, similarly
as in the case of extra-role behavior, the narrower focus of OI measures on factors
that contribute to intrinsic motivation as compared with AOC measures can explain
the formersÕ stronger correlation with job involvement.
In light of this reasoning, the finding that the OIQ proved empirically almost
interchangeable with AOC scales in so many respects becomes understandable. Un-
like other OI measures (including the Mael scale), the OIQ was based on a broad def-
inition of OI (namely, PatchenÕs, 1970). Consequently, some items of the scale
obviously refer to variables that are conceptually distinct from OI (e.g., ‘‘I would
be willing to spend the rest of my career with the organization’’; ‘‘I am glad to have
chosen to work for the organization rather than another company’’; ‘‘I feel that the
organization cares about me’’). Thus, scores of the OIQ may reflect a similar broad
mixture of satisfaction, work-related intentions, and identification as do scores of
AOC measures. Hence, it should not be surprising that the OIQ has similar empirical
qualities as AOC scales (for similar conclusions, see Miller et al., 2000; Sass &
Canary, 1991).
5. General discussion
5.1. Correlates and distinctiveness of organizational identification
The results presented in the two foregoing main sections provided at least two
important insights. First, they showed that OI as it has been measured in previous
research correlates with a number of practically or theoretically interesting variables.
Some of them can plausibly be interpreted as either antecedents or consequences of
OI, for example, organizational tenure, job scope/challenge, organizational prestige,
intent to leave, and in-role and extra-role performance. Thus, the present analysis
pointed to several variables that should be considered in research aimed at exploring
the nomological network around OI. Moreover, the results suggest that OI may be a
useful predictor for many practically relevant variables.
Second, the analyses indicated that OI as it has been measured in previous re-
search is empirically distinct from conceptually related constructs. For one, it proved
distinct from its closest conceptual neighbor, AOC (cf. Mael & Tetrick, 1992; Pratt,
1998; van Dick, 2001). Although the empirical overlap was large (61% shared vari-
ance), OI proved distinct from AOC with respect to its correlates. In particular, OI
correlated less strongly with job satisfaction, intent to stay, and absenteeism, and
more strongly with extra-role behavior and job involvement than did AOC. The
results of these analyses suggest that empirical OI research has indeed dealt with a
construct that is more specific or homogeneous than is AOC (e.g., Ashforth & Mael,
1989; van Knippenberg & Sleebos, 2001). Moreover, the correlations with three
other conceptually related constructs, job and organizational satisfaction and job
involvement, indicated that OI was related to but nonetheless distinct from them
(less than 37% shared variance). Thus, and consistent with most current OI research-
ersÕ claims, OI as operationalized in the average previous OI study had unique empir-
ical qualities.
5.2. The role of the operationalization of organizational identification
The aforementioned conclusions were based on the overall analyses for all OI
studies. Individual studies may yield outcomes that are not compatible with the
above conclusions, that is, moderators may exist. The existence of moderators is very
likely because for most correlates considered herein, the observed variation in indi-
vidual correlations was statistically significant and remained substantial even after
correcting for sampling error and unreliability (i.e., these corrections mostly reduced
the observed variance by less than 50%).
Because many different OI measures have been used in previous research, the
operationalization of OI suggests itself as such a moderator. To explore whether this
is the case, I conducted separate analyses for studies that used one of the two most
common OI measures, the Mael scale and the OIQ. It turned out that the Mael scale,
which is the most often used OI scale, yielded results that were close to the average
across studies that used other scales and were relatively homogenous. It follows that
the above conclusions may apply particularly to studies using the Mael scale.
Quite differently, the OIQ produced results that (a) differed from the average re-
sults of studies that used other scales and (b) were especially heterogeneous. More-
over, the OIQ seems to be particularly similar to an AOC scale. This is perhaps most
clearly indicated by the facts that the OIQ correlated very strongly with AOC scales
(r = .90, 81% common variance) and correlated similarly strongly with job satisfac-
tion as did AOC (r = .68, 46% common variance). Thus, the conclusions from the
foregoing section might not apply to studies using the OIQ.
Because the results of OIQ studies were very heterogeneous and separate analyses
were possible only for few correlates, it may not be justified to draw specific conclu-
sions about the OIQ on the basis of the available studies. Nevertheless, the results do
suggest that results obtained with the OIQ are not equivalent to results obtained with
the Mael scale. This has at least two implications. First, reviewers of the OI literature
should be more cautious with generalizing across OI measures than it has been the
case in previous reviews of OI research. Second, empirical OI researchers who want
to minimize redundancy of their findings with findings from AOC research should
use the Mael scale rather than the OIQ.
Overall, the results speak to the usefulness of the Mael scale. Given the additional
advantages of this scale (short and easy to administer; widely used; many results
demonstrating its reliability and construct validity), it appears to be the best OI mea-
sure available to date.
5.3. Limitations
Several limitations of the present studies should be mentioned. For one, the num-
ber of analyzed studies was small for several correlates. Thus, in theory, a small num-
ber of future studies could markedly change the results for those correlates. This is
especially true for the analyses regarding the Mael scale and the OIQ, in which all
sample sizes were below 10. Less than 10 samples were available also with regard
to the following correlates: job level, education, organizational satisfaction, and
absenteeism. Thus, the results referring to these variables should be considered
rather preliminary.
A general problem of meta-analyses is that the observed effects of moderator vari-
ables may actually be due to variables confounded with the moderator. In particular,
it cannot be ruled out that the observed differences between OI and AOC research
reported herein reflect (also) other differences between these branches of research
rather than (only) differences between OI and AOC measures. For example, on aver-
age, OI and AOC research might differ also with respect to design, populations, or
operationalization of correlates. These variables may affect the results in addition to
the focal variable (OI vs. AOC). In general, too few studies were available for the
present analyses to systematically consider these variables. However, it is important
to note that the present results are partly in line with studies that compared the cor-
relates of OI and AOC within the same sample and hence held the mentioned fea-
tures constant. For example, also Mael and Tetrick (1992) and van Knippenberg
and Sleebos (2001) found that OI as measured with the Mael scale related less
strongly to job satisfaction than did AOC; in addition, like the present research,
van Knippenberg and Sleebos found a weaker relation of OI with intent to leave.
Further, as shown above, especially the observed differences with respect to extra-
role behavior, intention to leave, job satisfaction, and job involvement can be plau-
sibly linked to common differences between OI and AOC measures (namely, the
usually narrower focus of the former ones on norm internalization or emotional
attachment). Thus, it appears likely that the observed differences reflect unique fea-
tures of OI and AOC measures rather than only other features of the analyzed
studies.
Finally, one should note that OI and AOC were only studied in their operational
form. The present analyses dealt only with the question of whether empirical OI re-
search has indeed studied something that is not identical to what has been empiri-
cally studied under the rubric of AOC. The present analyses did not address the
question of whether the OI measures used in the analyzed studies were adequate
in light of specific definitions of OI. For example, some items of the Mael scale seem
to refer to affective components of OI (e.g., ‘‘When someone praises this organiza-
tion, it feels like a personal compliment’’) whereas the OI definition underlying the
development of this scale explicitly excluded affective components (cf. Ashforth &
Mael, 1989; Mael & Tetrick, 1992). Hence, in studies using this scale, there might
be a mismatch between the conceptualization and operationalization of OI (for this
criticism, see also van Dick, 2001). In general, this shows that the results reported
herein cannot be taken as evidence for the adequacy or the success of particular
OI concepts. Thus, although the results of the present analyses point to differences
between OI and AOC at the empirical level, the precise nature of this difference
and its implications for theorizing on OI still have to be explored (e.g., what the Mael
scale taps that AOC scales do not tap and vice versa).
5.4. Research recommendations
The present findings point to several promising areas for future research. For
example, the findings suggest that OI generally correlates with some variables that
play a crucial role in theorizing on OI, such as organizational prestige, intent to
leave, and in-role and extra-role performance (for relevant models, see, e.g., Duke-
rich, Golden, & Shortell, 2002; Mael & Ashforth, 1992; Wan-Huggins et al., 1998;
van Knippenberg, 2000). Hence, in the next step, it would be necessary to explore
the causal relation of OI with these variables and thus to test the extant theories.
Moreover, in light of the heterogeneity of the results for almost all of the correlates
studied herein, research on moderators seems necessary. The present meta-analysis
studied only the operationalization of OI as moderator. Several other moderators
are conceivable. For example, regarding the relation between OI and work out-
comes, research on AOC or on social identification suggest moderators such as
accountability (Barreto & Ellemers, 2000), economic dependency on the job (Brett,
Cron, & Slocum, 1995), intergroup comparisons (Ouwerkerk et al., 1999; Tajfel &
Turner, 1986), or organizational tenure (Wright & Bonett, 2002). These moderators
could not be examined meta-analytically herein due to small numbers of studies and/
or lack of information in the original reports.
Another important task for future research is to avoid a further fragmentation of
the field with respect to OI measurement. In more than a half of the samples included
in this meta-analysis, an OI scale was used that was constructed ad hoc. This heter-
ogeneity in the operationalization of OI makes it difficult to compare results between
studies. Further, the present analyses suggested that the choice of the OI measure
may influence the size of the correlations obtained (cf. the separate analyses for
the OIQ). Thus, the heterogeneity at the measurement level may have substantially
contributed to the large heterogeneity of the observed correlations. Hence, rather
than introducing new measures, future research should use already established mea-
sures whenever possible. As explained above, the Mael scale currently appears the
preferable measure. The use of another existing measure may be justified if the re-
search goal is to replicate or extend findings obtained with that measure. However,
even in this case, it seems advisable to include the Mael scale (or other OI scales) in
addition to explore the generality of the findings across operationalizations of OI.
This is not meant to say that OI research should not address conceptual issues any
more. For specific purposes, more differentiated conceptualizations of OI may prove
useful. For example, several studies have shown that different subtypes of identifica-
tion (e.g., affective vs. cognitive) relate differentially to work outcomes (see Bergami
& Bagozzi, 2000; van Dick, Wagner, Stellmacher, & Christ, 2004; also Ellemers,
Kortekaas, & Ouwerkerk, 1999). Thus, further research on the dimensions of OI
does seem desirable. Yet, before developing completely new scales that measure
OI on these dimensions, researchers should explore to which extent existing scales,
for example the Mael scale, tap one or more of those dimensions and whether they
can either be decomposed in subscales or combined with each other to measure those
dimensions. This way, researchers could promote a more differentiated look at OI
while ensuring some comparability with previous research. However, if the develop-
ment of a new OI scale appears necessary for some theoretical reason, researchers
should demonstrate the incremental validity of the new scale over existing ones to
justify its use. In any case, before developing another OI measure, researchers should
carefully weight the potential benefits of doing so against the potential costs (e.g.,
noncomparability of the own findings with previous ones).
A final issue refers to the distinction between OI and AOC in future research. This
distinction has a methodological and a theoretical aspect. The methodological aspect
can be expressed in the question: When should one use an OI scale rather than an
AOC scale and vice versa? The present findings suggest that either type of scale
can have unique advantages when the research is merely to predict other variables.
Specifically, OI scales (especially the Mael scale) may be superior in predicting extra-
role behavior whereas AOC scales may be superior in predicting absenteeism and in-
tent to stay. However, when the research goal is to test theoretical models, some OI
scales (e.g., the Mael scale) appear basically superior to AOC scales because they
represent a more narrowly and precisely defined construct. Thus, they may enable
a more fine-grained empirical analysis of psychological processes than do AOC
scales. For example, several models clearly distinguish between (affective and/or cog-
nitive) OI and its evaluative and motivational consequences (such as job satisfaction
and intent to stay; e.g., Ashforth & Mael, 1989; van Dick, 2001). A test of these mod-
els requires that these constructs are measured separately. AOC scales, which obvi-
ously tap a mixture of (components of) OI and evaluative and motivational
variables, appear less suited to meet this requirement than do certain OI scales with
their more specific focus.
A second and more theoretically important question in the context of the OI–
AOC distinction is: To what extent should researchers interested in OI take into
account findings referring to AOC and vice versa? In my view, the answer is: to
the greatest extent possible. There are at least two reasons for this. One reason
is that this can avoid redundant research efforts. The use of particular scales is of-
ten a matter of personal habit or a striving to appear consistent. A researcher who
has used the same AOC or OI scale in all of his or her previous studies will likely
use and may be expected to use this scale also in the future. Thus, AOC scales may
often have been used in studies for which an OI scale would have been more
appropriate and vice versa. Moreover, even if the focal construct (OI vs. AOC)
has been deliberately selected, the results may be relevant to research on both con-
structs because in many (albeit evidently not all) research contexts, OI and AOC
scales may fulfill almost identical roles. For example, the above analyses suggest
that OI and AOC may be similarly suited for the prediction of in-role behavior.
In all mentioned cases, the findings are relevant to research on both OI and
AOC, irrespective of the rubric under which they have been reported. Hence, an
OI researcher who plans to test (what he or she believes to be) an innovative
hypothesis should examine whether this hypothesis has already been tested with
regard to AOC and vice versa. If this is the case, a test of the hypothesis may
not be much more than a replication, and it may be more appropriate to consider
the hypothesis as already confirmed and use it as a starting-point for addressing a
more sophisticated issue.
A second reason why OI researchers should be open-minded to AOC research and

vice versa is that researchers in both fields seem to be able to provide each other with
valuable theoretical input. For one, AOC researchers may profit from several models
and theories that have been used in OI research and that do not have parallels in
AOC research, for example, social identity theory (Tajfel & Turner, 1986), Elsbach’s
(1999), taxonomy of identification types, RousseauÕs (1998) distinction between sta-
ble and temporary identification, and models of the dimensions of OI (Ellemers
et al., 1999; van Dick et al., 2004). Conversely, OI researchers may profit from ideas
that originated in AOC research and have not yet been taken up in OI research, for
example, assumptions concerning moderators of the AOC–work behavior relation
such as economic dependency on the job (Brett et al., 1995) and tenure (Wright &
Bonett, 2002). Further, OI and AOC researchers seem to have enough common
ground to engage in fruitful discussions about theoretical issues. Thus, they may ben-
efit not only from transfer of theoretical models but also from mutual criticism.
References1
*Abrams, D., Ando, K., & Hinkle, S. (1998). Psychological attachment to the group. Personality and
Social Psychology Bulletin, 24, 1027–1039.
Albert, S., Ashforth, B. E., & Dutton, J. E. (Eds.) (2000). Organizational identity and identification
[Special issue]. Academy of Management Journal, 25.
Allen, N. J., & Meyer, J. P. (1990). The measurement and antecedents of affective, continuance and
normative commitment to the organization. Journal of Occupational Psychology, 63, 1–18.
Ashforth, B. E., & Mael, F. (1989). Social identity and the organization. Academy of Management Review,
14, 20–39.
Aven, F. F., Parker, B., & McEvoy, G. M. (1993). Gender and attitudinal commitment to organizations: A
meta-analysis. Journal of Business Research, 26, 63–73.
*Barker, J. R., & Tompkins, P. K. (1994). Identification in the self-managing organization. Human
Communication Research, 21, 223–240.
Barreto, M., & Ellemers, N. (2000). You canÕt always do what you want: Social identity and self-
presentational determinants of the choice to work for a low-status group. Personality and Social
Psychology Bulletin, 26, 891–906.
*Beerman, L., (1987). Einstellungen und Werthaltungen zur Arbeit und Freizeit als Interaktion zwischen
Person und Organisation [Attitudes and values toward work and leisuretime as interaction of person
and organisation]. Unpublished diploma thesis, University of Munich, Germany. [Cited after Stengel,
1987].
*Benkhoff, B. (1997a). A test of the HRM model: Good for employers and employees. Human Resource
Management Journal, 7(4), 44–60.
*Bergami, M., & Bagozzi, R. P. (2000). Self-categorization, affective commitment and group self-esteem as
distinct predictors of social identity in the organization. British Journal of Social Psychology, 39,
555–577.
Brett, J. F., Cron, W. L., & Slocum, J. W. (1995). Economic dependency on work. Academy of
Management Journal, 95, 261–271.
Brown, M. E. (1969). Identification and some conditions of organizational involvement. Administrative
Science Quarterly, 14, 346–355.
Brown, R., & Williams, J. (1984). Group identification: The same thing to all people? Human Relations, 37,
547–564.
1
References marked with an asterisk indicate studies included in the meta-analysis for OI.
*Buchanan, B. (1974). Building organizational commitment: The socialization of managers in work

organizations. Academy of Management Journal, 19, 533–546.
Bullis, C. A., & Tompkins, P. K. (1989). The Forest Ranger revisited: A study of control practices and
identification. Communication Monographs, 56, 287–306.
*Cheney, G. (1983). On the various and changing meanings of organizational membership: A field study
of organizational identification. Communication Monographs, 50, 343–362.
Cohen, A. (1992). Antecedents of organizational commitment across occupational groups: A meta-
analysis. Journal of Organizational Behavior, 13, 539–558.
Cohen, A. (1993). Age and tenure in relation to organizational commitment: A meta-analysis. Basic and
Applied Social Psychology, 14, 143–159.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). New York: Academic
Press.
*Dukerich, J. M., Golden, B. R., & Shortell, S. M. (2002). Beauty is in the eye of the beholder: The impact
of organizational identification, identity, and image on the cooperative behaviors of physicians.
Administrative Science Quarterly, 47, 507–533.
Dutton, J. E., Dukerich, J. M., & Harquail, C. V. (1994). Organizational images and member
identification. Administrative Science Quarterly, 39, 239–263.
Edwards, M.R. (2002). Disentangling organisational identification from organisational commitment. Paper
presented at the British Academy of Management Annual Conference.
*Edwards, M.R. (2003). Disentangling organizational identification: A case study of an NHS trust.
Unpublished PhD thesis.
Ellemers, N., de Gilder, D., & van den Heuvel, H. (1998). Career-oriented versus team-oriented
commitment and behavior at work. Journal of Applied Psychology, 83, 717–730.
Ellemers, N., Kortekaas, P., & Ouwerkerk, J. W. (1999). Self-categorisation, commitment to the group
and group self-esteem as related but distinct aspects of social identity. European Journal of Social
Psychology, 29, 371–389.
Elsbach, K. D. (1999). An expanded model of organizational identification. Research in Organizational
Behavior, 21, 163–200.
Fontenot, J., & Scott, C. R. (2000). A meta-analysis of the organizational identification literature. Paper
presented at International Communication Association Convention, Acapulco, Mexico.
Fontenot, J. C., & Scott, C .R., (2003). Correlates of organizational identification moderated by
measurement scale, organizational type, and publication date: A meta-analysis. Unpublished manuscript,
University of Texas at Austin.
*Gautam, T., van Dick, R., & Wagner, U. (in press). Organizational identification and organizational
commitment: Distinct aspects of related concepts. Asian Journal of Social Psychology.
Griffin, R. W., & Bateman, T. S. (1986). Job satisfaction and organizational commitment. In C. L. Cooper
& I. Robertson (Eds.), International review of industrial and organizational psychology (pp. 157–188).
New York: Wiley.
*Hall, D. T., Schneider, B., & Nygren, H. T. (1970). Personal factors in organizational identification.
Administrative Science Quarterly, 15, 176–190.
Harquail, C. V. (1998). Organizational identification and the ‘‘whole person’’: Considering affect,
behavior, and cognition. In D. A. Whetten & P. C. Godfrey (Eds.), Identity in organizations
(pp. 223–231). Thousand Oaks, CA: Sage.
Haslam, S. A. (2001). Psychology in organizations: The social identity approach. London: Sage.
Hunter, J. E., & Schmidt, F. L. (1990). Methods of meta-analysis: Correcting error and bias in research
findings. Newbury Park, CA: Sage.
*Iyer, V., Bamber, E. M., & Barefield, R. M. (1997). Identification of accounting firm alumni with their
former firm: Antecedents and outcomes. Accounting, Organizations, and Society, 22, 315–336.
Kelman, H. C. (1961). Processes of opinion change. Public Opinion Quarterly, 25, 57–78.
Lee, S. M. (1969). Organizational identification of scientists. Academy of Management Journal, 12,
327–337.
*Mael, F. A., & Ashforth, B. E. (1992). Alumni and their alma mater: A partial test of the reformulated
model of organizational identification. Journal of Organizational Behavior, 13, 103–123.
*Mael, F. A., & Tetrick, L. E. (1992). Identifying organizational identification. Educational and
Psychological Measurement, 52, 813–824.
March, J. G., & Simon, H. A. (1958). Organizations. New York: Wiley.
Mathieu, J. E., & Zajac, D. M. (1990). A review and meta-analysis of the antecedents, correlates, and
consequences of organizational commitment. Psychological Bulletin, 108, 171–194.
Meyer, J. P., Stanley, D. J., Herscovitch, L., & Topolnytsky, L. (2002). Affective, continuance and
normative commitment to the organization: A meta-analysis of antecedents, correlates and
consequences. Journal of Vocational Behavior, 61, 20–52.
*Miller, V. D., Allen, M., Casey, M. K., & Johnson, J. R. (2000). Reconsidering the Organizational
Identification Questionnaire. Management Communication Quarterly, 13, 626–658.
Mottola, G. R., Bachman, B. A., Gaertner, S. L., & Dovidio, J. F. (1997). How groups merge: The effects
of merger integration patterns on anticipated commitment to the merged organization. Journal of
Applied Social Psychology, 27, 1335–1358.
Mowday, R. T., Steers, R. M., & Porter, L. W. (1979). The measurement of organizational commitment.
Journal of Vocational Behavior, 14, 224–247.
*OÕReilly III, C., & Chatman, J. (1986). Organizational commitment and psychological attachment: The
effects of compliance, identification, and internalization on prosocial behavior. Journal of Applied
Organ, D. W. (1988). Organizational citizenship behavior: The good soldier syndrome. Toronto: Lexington.
Ouwerkerk, J. W., Ellemers, N., & de Gilder, D. (1999). Group commitment and individual effort in
experimental and organizational contexts. In N. Ellemers, R. Spears, & B. Doosje (Eds.), Social
identity (pp. 185–204). Oxford: Blackwell.
*Padaki, R., & Gandhi, P. (1981). Organizational climate and identification with work and organization.
Managerial Psychology, 2, 46–60.
Patchen, M. (1970). Participation, achievement, and involvement on the job. Englewood Cliffs, NJ: Prentice
Hall.
Porter, L. W., Steers, R. M., Mowday, R. T., & Boulian, P. V. (1974). Organizational commitment, job
satisfaction, and turnover among psychiatric technicians. Journal of Applied Psychology, 59, 603–609.
*Potvin, T. C. (1991). Employee organizational commitment: An examination of its relationship to
communication satisfaction and an evaluation of questionnaires designed to measure the construct.
Unpublished doctoral thesis, University of Kansas.
Pratt, M. G. (1998). To be or not to be: Central questions in organizational identification. In D. A.
Whetten & P. C. Godfrey (Eds.), Identity in organizations (pp. 171–207). Thousand Oaks, CA: Sage.
Riketta, M. (2002). Attitudinal organizational commitment and job performance: A meta-analysis.
Journal of Organizational Behavior, 23, 257–266.
*Rotondi, T. (1975a). Organizational identification: Issues and implications. Organizational Behavior and
Human Performance, 13, 95–109.
*Rotondi, T. (1975b). Organizational identification and group involvement. Academy of Management
Journal, 18, 892–897.
Rousseau, D. M. (1998). Why workers still identify with organizations. Journal of Organizational Behavior,
19, 217–233.
*Sass, J. E., & Canary, D. J. (1991). Organizational commitment and identification: An examination of
conceptual and operational convergence. Western Journal of Speech Communication, 55, 275–293.
*Sauer, B. (1997). Organisationale Identifikation: Entwicklung und Erprobung eines Fragebogens [Orga-
nizational identification: Development and test of a questionnaire]. Unpublished diploma thesis,
Catholic University of Eichstätt, Germany.
Stengel, M. (1987). Identifikationsbereitschaft, Identifikation, Verbundenheit mit einer Organisation oder
ihren Zielen [Readiness for identification, identification with, and commitment to, an organization or
its goals]. Zeitschrift fuer Arbeits- und Organisationspsychologie, 31, 152–166.
Tajfel, H. (Ed.). (1978). Differentiation between social groups. Studies in the social psychology of intergroup
relations. London: Academic Press.
Tajfel, H., & Turner, J. C. (1986). The social identity theory of intergroup behavior. In S. Worchel & W.
G. Austin (Eds.), Psychology of intergroup relations (2nd ed.) (pp. 7–24). Chicago: Nelson-Hall.
Tett, R. P., & Meyer, J. P. (1993). Job satisfaction, organizational commitment, turnover intention, and
turnover: Path analyses based on meta-analytical findings. Personnel Psychology, 46, 259–293.
Tyler, T. (1999). Why people cooperate with organizations: An identity-based perspective. Research in
Organizational Behavior, 21, 201–246.
van Dick, R. (2001). Identification in organizational contexts: Linking theory and research from social and
organizational psychology. International Journal of Management Reviews, 3, 265–283.
van Dick, R. (2004). My job is my castle: Identification in organizational contexts. In C. L. Cooper & I. T.
Robertson (Eds.). International review of industrial and organizational psychology (Vol. 19,
pp. 171–204). Chichester: Wiley.
*van Dick, R., Wagner, U., & Gautam, T. (2002). Identifikation in Organisationen: Theoretische
Zusammenhaenge und empirische Befunde [Identification in organizations: theoretical relations and
empirical findings]. In E. H. Witte (Ed.), Sozialpsychologie wirtschaftlicher Prozesse (pp. 147–173).
Lengerich: Pabst.
*van Dick, R., Wagner, U., Lemmer, G., & Britt Lima, C. (2002). Mitarbeiterbefragung im Rahmen einer
Fusion zweier Kliniken [Employee survey in the context of the merger of two hospitals]. Unpublished
manuscript, University of Marburg, Germany.
*van Dick, R., Wagner, U., Stellmacher, J., & Christ, O. (2004). The utility of a broader conceptualization
of organizational identification: Which aspects really matter? Journal of Occupational and Organiza-
tional Psychology, 77, 171–191.
van Knippenberg, D. (2000). Work motivation and performance: A social identity perspective. Applied
Psychology: An International Review, 49, 357–371.
van Knippenberg, D., & Hogg, M. (Eds.) (2002). Social identity processes in organizations [Special issue].
Group Processes and Intergroup Relations, 4.
*van Knippenberg, D., & Sleebos, E. (2001). Further exploration of the organizational identification
concept: Identification versus commitment. Unpublished manuscript, University of Amsterdam.
*van Knippenberg, D., & van Schie, E. C. M. (2000). Foci and correlates of organizational identification.
Journal of Occupational and Organizational Psychology, 73, 137–147.
Viswesvaran, C., Ones, D. S., & Schmidt, F. L. (1996). Comparative analysis of the reliability of job
performance ratings. Journal of Applied Psychology, 81, 557–574.
Wallace, J. E. (1993). Professional and organizational commitment: Compatible or incompatible? Journal
of Vocational Behavior, 42, 333–349.
*Wan-Huggins, V., Riordan, C. M., & Griffeth, R. W. (1998). The development and longitudinal test of a
model of organizational identification. Journal of Applied Social Psychology, 28, 724–749.
Wright, T. A., & Bonett, D. G. (2002). The moderating effects of employee tenure on the relation between
organizational commitment and job performance: A meta-analysis. Journal of Applied Psychology, 87,
1183–1190.
Further reading
*Ahearne, M., Bhattacharya, C. B., & Gruen, T. (2003). Antecedents and consequences of customer–
company identification: Expanding the role of relationship marketing. Unpublished manuscript,
University of Connecticut.
*Ahlswede, O., & Grubba, C. (2002). Zusammenhänge zwischen Arbeitszufriedenheit, Teamklima,
Identifikation und Engagement im Bankbereich [Relations between job satisfaction, team climate,
identification and engagement in banks]. Unpublished diploma thesis, University of Marburg,
Germany.
*Ashforth, B. E., & Saks, A. M. (1996). Socialization tactics: Longitudinal effects on newcomer
adjustment. Academy of Management Journal, 39, 149–178.
*Bagozzi, R. P., & Bergami, M. (2003). Identification mediates the effects of prestige and stereotypes on
commitment, job and work involvement, and satisfaction in the organization. Unpublished manuscript,
University of Michigan.
*Balfour, D. L., & Wechsler, B. (1991). Commitment, performance, and productivity in public
organizations. Public Productivity and Management Review, 19, 355–367.
*Bartel, C. A. (2001). Social comparisons in boundary-spanning work: Effects of community outreach
on membersÕ organizational identity and identification. Administrative Science Quarterly, 46,
379–413.
*Becker, T. E. (1992). Foci and basis of commitment: Are they distinctions worth making?. Academy of
Management Journal, 35, 232–244.
*Becker, T. E., Billings, R. S., Eveleth, D. M., & Gilbert, N. L. (1996). Foci and bases of employee
commitment: Implications for job performance. Academy of Management Journal, 39, 464–482.
*Bell, S. J., & Menguc, B. (2002). The employee-organization relationship, organizational citizenship
behaviors, and superior service quality. Journal of Retailing, 78, 131–146.
*Benkhoff, B. (1997b). Better performance through organisational identification: A test of outcomes and
antecedents based on social identity theory. In J. Wickham (Ed.), The search for competitiveness and its
implications for employment (pp. 159–179). Dublin: Oak Tree Press.
*Bhattacharya, C. B., Rao, H., & Glynn, M. A. (1995). Understanding the bond of identification: An
investigation of its correlates among art museum members. Journal of Marketing, 59(October), 46–57.
*Brady, M. K., Noble, C. H., Utter, D. J., & Smith, G. E. (2002). How to give and receive: An exploratory
study of charitable hybrids. Psychology and Marketing, 19, 919–944.
*Cable, D. M., & DeRue, D. S. (2002). The convergent and discriminant validity of subjective fit
perceptions. Journal of Applied Psychology, 87, 875–884.
*Cook, J., & Wall, T. (1980). New work attitude measures of trust, organizational commitment and
personal need non-fulfilment. Journal of Occupational Psychology, 53, 399–452.
*De Ley, F. (2003). Effect of the remote working base on social identity with work related groups.
Unpublished thesis, London Metropolitan University.
*Efraty, D., & Wolfe, D. M. (1988). The effect of organizational identification on employee affective and
performance responses. Journal of Business and Psychology, 3, 105–112.
*Gould, S., & Werbel, J. D. (1983). Work involvement: A comparison of dual wage earner and single wage
earner families. Journal of Applied Psychology, 68, 313–319.
*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.
*Hauptmeier, M. (2002). Facetten und Zusammenhänge zu arbeitsrelevanten Einstellungen und Verhal-
tensweisen [Facets and relations of work related attitudes and behaviors]. Unpublished diploma thesis,
University of Marburg, Germany.
*Hennessy, J., & West, M. A. (1999). Intergroup behavior in organizations: A field test of social identity
theory. Small Group Research, 30, 361–382.
*Hoehfeld, C. (2001). Teamklima und Arbeitszufriedenheit [Team climate and job satisfaction]. Unpub-
lished diploma thesis, University of Marburg, Germany.
*Jetten, J., OÕBrien, A., & Trindall, N. (2002). Changing identity: Predicting adjustment to organizational
restructure as a function of subgroup and subordinate identification. British Journal of Social
*Johnson, J. R., Bernhagen, M. J., Miller, V., & Allen, M. (1996). The role of communication in managing
reductions in work force. Journal of Applied Communication Research, 24, 139–164.
*Koch, S., Kubat, A., Kruse, L., & Thimm, C. (2001). Teamkommunikation am Arbeitsplatz: Entwicklung
der Kodierschemata und erste Ergebnisse [Telecommunication at the workplace: Development of the
coding schemas and first results]. Unpublished research report, project ‘‘Die kommunikative
Konstruktion von Geschlecht in beruflichen Settings’’, University of Heidelberg, Germany.
*Kollar, R., & Stengel, M. (1990). Sind Berufsorientierungen und organisationale Identifikation Chimären
der Forschung? [Are job orientations and organizational identification chimeras of research?].
Zeitschrift fuer Arbeits- und Organisationspsychologie, 34, 74–84.
*Kreiner, G. E., & Ashforth, B. E. (2004). Evidence toward an expanded model of organizational
identification. Journal of Organizational Behavior, 25, 1–27.
*Lehr, J. K., & Rice, R. E. (2002). Motivation, organizational identification, and experiences of the quality
examiner. Quality Management Journal, 9, 63–90.
*Lipponen, J. (2001). Organizational identification: Antecedents and consequences of identifications in a

shipyard context. Helsinki, Finland: Department of Social Psychology, University of Helsinki.
*Mael, F. A. (1991). Desire for career upward mobility and workplace adaptation. The International
Journal of Career Management, 4, 10–16.
*Mael, F. A., Waldman, D. A., & Mulqueen, C. (2001). From scientific work to organizational leadership:
Predictors of management aspiration among technical personnel. Journal of Vocational Behavior, 59,
132–148.
*Mael, F. A., & Alderks, C. E. (1993). Leadership team cohesion and subordinate work unit morale and
performance. Military Psychology, 5, 141–158.
*Martin, R., & Epitropaki, O. (2001). Role of organizational identification on implicit leadership theories
(ILTs), transformational leadership and work attitudes. Group Processes and Intergroup Relations, 4,
247–262.
*Millward, L. J., & Brewerton, P. M. (1999). Contractors and their psychological contracts. British
Journal of Management, 10, 253–274.
*Moltzen, K. (2002). Eine empirische Untersuchung zu Arbeitszufriedenheit, Arbeitsmotivation, Teamklima,
Identifikation und OCB in der Praxis [An empirical study of job satisfaction, work motivation, team
climate, identification and OCB in organizational practice]. Unpublished diploma thesis, University of
Marburg, Germany.
Mowday, R. T., Porter, L. W., & Steers, R. M. (1982). Employee-organization linkages: The psychology of
commitment, absenteeism, and turnover. New York: Academic.
*Moye, N. A., & Bartol, K. M. (2003). The role of the psychological contract in motivating employee
identity: A social exchange perspective. Unpublished manuscript, Vanderbilt University.
*Oliver, N. (1990). Rewards, investments, alternatives and organizational commitment: Empirical
evidence and theoretical development. Journal of Occupational Psychology, 63, 19–31.
Prakrash, A., & Misra, S. (1986). Organizational outcome factors in relation to personal values
of employees in banking organizations. Indian Journal of Current Psychological Research, 1,
112–120.
*Reynolds, S. J. (2003). Perceptions of organizational ethicality: Do indeed perceptions of self lead to
inflated perceptions of the organization. Journal of Business Ethics, 42, 253–266.
*Roccas, S. (2003). Identification and status revisited: The moderating role of self-enhancement and self-
transcendence values. Personality and Social Psychology Bulletin, 29, 726–736.
*Russo, T. C. (1998). Organizational and professional identification. Management Communication
Quarterly, 12, 72–111.
*Schneider, B., Hall, D. T., & Nygren, H. T. (1971). Self image and job characteristics as correlates of
changing organizational identification. Human Relations, 24, 397–416.
*Scott, C. R., & Timmerman, C. E. (1999). Communication technology use and multiple workplace
identifications among organizational teleworkers with varied degrees of virtuality. IEEE Transactions
on Professional Communication, 42, 240–260.
Scott, C. R. (1997). Identification with multiple targets in a geographically dispersed organization.
Management Communication Quarterly, 10, 491–522.
*Siders, M. A., George, G., & Dharwadkar, R. (2001). The relationship of internal and external
commitment foci to objective job performance measures. Academy of Management Journal, 44,
570–579.
*Smidts, A., Pruyn, A. T. H., & van Riel, C. B. M. (2001). The impact of employee communication and
perceived external image on organizational identification. Academy of Management Journal, 44,
1051–1062.
*Stierle, C. (2002). Eignungskriterien und Erfolgsfaktoren für international tätige Mitarbeiter [Qualification
criteria and factors for the success of expatriates]. Unpublished diploma thesis, University of Marburg,
Germany.
Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell, M. S. (1987). Rediscovering the social
group: A self-categorization theory. Oxford: Basil Blackwell.
*Tyagi, P. K., & Wotruba, T. R. (1993). An exploratory study of reverse causality relationships among
sales force turnover variables. Journal of the Academy of Marketing Science, 21, 143–153.
*van Knippenberg, D., van Knippenberg, B., Monden, L., & de Lima, F. (2002). Organizational
identification after a merger: A social identity perspective. British Journal of Social Psychology, 41,
233–252.
*Vandenberg, R. J., Self, R. M., & Seo, J. H. (1994). A critical examination of the internalization,
identification, and compliance commitment measures. Journal of Management, 20, 123–140.
*Wegge, J., & Neuhaus, L. (2002). Emotionen bei der Bueroarbeit am PC: Ein Test der affective events
Theorie [Emotions and office work with computers: A test of affective events theory]. Zeitschrift fuer
Arbeits- und Organisationspsychologie, 46, 173–184.

A Meta Analysis of Organizational Identification

Uploaded by

Copyright:

Available Formats

A Meta Analysis of Organizational Identification

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

A Meta Analysis of Organizational Identification

Uploaded by

Copyright:

Available Formats

Journal of Vocational Behavior 66 (2005) 358–384

Organizational identiﬁcation: A meta-analysisq

Received 22 December 2003

Keywords: Organizational identiﬁcation; Organizational commitment; Work behavior; Meta-analysis

Compared with other psychological variables assumed to be relevant to work

2. Organizational identiﬁcation and attitudinal organizational commitment

Many deﬁnitions of OI have been proposed. Most of them conceptualize OI as a

3. Meta-analysis of research on organizational identiﬁcation

3.1. Study collection and coding of study characteristics

The computational procedure followed the meta-analytic approach of Hunter and

Work-related intentions and behaviors

Another parameter of interest was the variance of the population correlations.

3.3.1. Overall analysis

3.3.2. Analyses for single OI measures

4. Meta-analytic comparison between organizational identiﬁcation and attitudinal

4.1. Study search and computations

To identify meta-analyses that reported correlations involving either the ACS or

ther, job involvement overlaps conceptually with intrinsic motivation. Because OI as

5.1. Correlates and distinctiveness of organizational identiﬁcation

5.2. The role of the operationalization of organizational identiﬁcation

5.4. Research recommendations

A second reason why OI researchers should be open-minded to AOC research and

*Buchanan, B. (1974). Building organizational commitment: The socialization of managers in work

*Lipponen, J. (2001). Organizational identiﬁcation: Antecedents and consequences of identiﬁcations in a

You might also like