Integrative Tests of a Multidimensional Model of Organizational Identification

S. Boro5 et al.: Integrative Tests of Organizational IdentificationSocial Psychology 2011; Vol. 42(2):111–123© 2011 Hogrefe Publishing

Original Article

Integrative Testsof a Multidimensional Model

of Organizational IdentificationSmaranda Boros1, Petru Lucian Curseu1, and Mircea Miclea2

1Department of Organization Studies, Tilburg University, The Netherlands2Department of Psychology, Babe5-Bolyai University, Cluj-Napoca, Romania

Abstract. Organizational identification works on the cognitive, affective, and evaluative levels. The present study, like previous research,conceptualizes identification as a multidimensional concept. We test a model of identification developed in experimental settings usingan empirical study carried out in three Romanian organizations. We investigate relevant antecedents and consequences of cognitive,affective, and evaluative identification for organizations. The findings shed light on the conditions in which the impact of cognitiveidentification on organizational outcomes is mediated by the affective attachment to the organization. They also indicate that evaluativeidentification has a smaller impact in organizational settings than experimental studies have previously shown, contradicting predictionsof the self-enhancement hypothesis. Based on these findings, we draw out the practical implications for effective strategies targetingorganizational identification enhancement.

Keywords: organizational identification, turnover intentions, job satisfaction, organizational citizenship behavior, affective commitment

Organizations have an important place in the life of an in-dividual. People identify with their employing organizationboth at the cognitive and affective level, and enhance theirself-esteem through this identification. Although the threedimensions of identification (cognitive, affective, and eval-uative) have been discussed from the earliest theoreticalmodels of social identity (Tajfel, 1978) and extensivelytested in experimental studies, empirical studies that testthis model in organizations in an integrative manner arestill rare (Bergami & Bagozzi, 2000; Ellemers, Kortekaas,& Ouwerkerk, 1999; Johnson & Morgeson, 2005). In aneffort to provide empirical support for a multidimensionalconcept of organizational identification, we test a model ofidentification developed in experimental settings (Ellemerset al., 1999) in a field study carried out in three organiza-tions.

We address three limitations of previous studies. First,we investigate in an integrative model the relations be-tween the three dimensions of identification. While theo-retical reviews have proposed the explanatory mechanismswhich make cognitive identification the primary base ofidentification, with affective and evaluative identificationfollowing on from it (Ellemers, De Gilder, & Haslam,2004), empirical research in organizational settings has notextensively tested these propositions. Second, we add tothe very small number of studies (e.g., Johnson & Morge-son, 2005) that assess antecedents and consequences of or-

ganizational identification at the same time, thus advancinga comprehensive model of identification. Third, we addressthe divide in the identification correlates used in fundamen-tal SIT experiments and those used in empirical organiza-tion studies. The variables that we test in this model arerelevant both to SIT fundamental literature and for organi-zational life.

Cognitive, Affective, and EvaluativeIdentification

Starting from Tajfel’s (1978) classical approach, Ellemerset al. (1999) redefined social identity using three dimen-sions: a cognitive component (self-categorization: a cogni-tive awareness of individual membership in a social group),an evaluative component (group self-esteem: a positive ornegative value connotation attached to this group member-ship), and an emotional component (affective commitment:a sense of emotional involvement with the group). Theo-retical developments point to the underlying mechanismsthat explain how the identification process takes place foreach of these three dimensions (Ellemers et al., 2004).

However, the number of empirical studies investigatingthis multidimensional model of identification remainsscant. Some of Ellemers’ studies (Doosje, Ellemers, &

DOI: 10.1027/1864-9335/a000051© 2011 Hogrefe Publishing Social Psychology 2011; Vol. 42(2):111–123

Spears, 1995; Ellemers et al., 1999) investigated the differ-ential relations of several antecedents of the three forms ofidentification. All these studies were experimental andused the minimal group paradigm. Most researchers whoinvestigated this model in organizational settings (Bergami& Bagozzi, 2000; Johnson & Morgeson, 2005) kept onlytwo components of the original definition: the cognitiveand the affective. To our knowledge, there is only one studythat tests the three-dimensional model of identification inorganizational settings (van Dick, Wagner, Stellmacher, &Christ, 2004). Furthermore, the relationship between thethree dimensions of identification is barely tackled in em-pirical research. Bergami and Bagozzi (2000) advance amodel in which affective identification is a mediator be-tween cognitive identification and organizational attitudes.Therefore, we set out to explore the relationship betweenthe three dimensions of identification in organizational set-tings and to test if the predictions of this model built onexperimental data remain the same in organizational set-tings.

Using the principle of Occam’s razor (the simplest ex-planation), we argue that while, theoretically, identity andcommitment are distinguishable concepts, there are boththeoretical as well as empirical grounds for using affectivecommitment as an indicator of the affective aspect of iden-tification in organizations. We align ourselves here with anumber of researchers (Bergami & Bagozzi, 2000; Riketta& Van Dick, 2005), including those who proposed the the-oretical model we are testing here (Ellemers et al., 1999,2004), who overlap the affective aspect of identificationand affective commitment. Our main theoretical argumentis that with few exceptions (Johnson & Morgeson, 2005,Meyer, Becker, & van Dick, 2006), researchers that differ-entiated between identification and commitment focusedon the cognitive side of identification and the affective sideof commitment (e.g., Ashforth & Mael, 1989; Riketta,2005). In these definitions, identification refers only to self-categorization, while commitment is the affective link thatsupplies the motivational force for attachment to the organ-ization (Bergami & Bagozzi, 2000; Ellemers et al., 1999).Researchers who differentiated between the cognitive andaffective dimensions of both identification and commit-ment have not elaborated on the differences between affec-tive identification and commitment (Meyer et al., 2006).Therefore, in organizational practice no instruments thatcan make a clear-cut differentiation between affective iden-tification and commitment have yet been validated.

The theoretical differentiation between the affectivesides of identification and commitment is lost in the empir-ical instruments that measure the two. For instance, Elle-mers’ scale of group commitment (GC), which supposedlyassesses the emotional side of identification is composedof three items that are very similar to the ones in Meyer andAllen’s (1991) Affective Commitment Scale (ACS): Iwould like to continue working with this group (GC) – Iwould be happy to spend the rest of my career with thisorganization (ACS); I dislike being a member of this group

(GC) – I do not feel “emotionally attached” to this organ-ization (ACS); I would rather belong to another group(GC) – I think that I could easily become as attached toanother organization as I am to this one (ACS). Further-more, the rest of the items in the ACS target aspects thatare incorporated in the definitions of affective identifica-tion, as proposed by authors that distinguish between thistype of identification and commitment. For instance, John-son and Morgeson (2005) as well as Meyer et al. (2006)define affective identification as the feelings individualsexperience about themselves in relation to the social refer-ent and the value they place on that social identity (ACS 6.I do not feel “emotionally attached” to this organization;ACS 7. This organization has a great deal of personalmeaning for me). They also state that “affective identifica-tion” is associated with positive feelings about individualmembership, including pride (ACS 2. I enjoy discussing myorganization with people outside it), enthusiasm, and asense of affiliation or “belongingness” with others (ACS 5.I do not feel like “part of the family” at my organization;ACS 8. I do not feel a strong sense of belonging to myorganization) (Johnson & Morgenson, 2005, p. 2).

Although we are aware of the need to distinguish be-tween affective identification and commitment, the pur-pose of this research is to test an integrative model of iden-tification in organizational settings, and not to develop ameasurement for the affective side of identification thatempirically distinguishes it from affective commitment.Therefore, we prefer using scales already validated andlargely recognized in organizational settings. For these rea-sons, in the present study we shall henceforth refer to theaffective side of identification as affective commitment andmeasure it using Meyer and Allen’s Affective CommitmentScale. Since no organizational measures have been elabo-rated yet to assess the evaluative side of identification, weadopted Ellemers’s et al. (1999) Group Self-Esteem Scalefor organizational settings.

Articulation of the Working Model

Given the nature of our study, the choice of our variableswas circumscribed by the criteria of both theoretical (SIT)and organizational relevance. Hence, we chose two ante-cedents, one pertaining to the individual (need for identifi-cation) and one related to the organization (perceived or-ganizational prestige). The latter is an organization-leveltranslation of the high vs. low status of groups as manipu-lated in experimental SIT research. We also chose four con-sequences, two behavioral (perceived performance and or-ganizational citizenship behavior – OCB) and two attitudi-nal (job satisfaction and turnover intention). Again,perceived performance and OCB are organizational prox-ies for individual vs. collective action, in SIT terms, whileturnover intention is an approximate illustration for socialmobility. The work definitions of these variables can befound in Table 1.

112 S. Boro5 et al.: Integrative Tests of Organizational Identification

Social Psychology 2011; Vol. 42(2):111–123 © 2011 Hogrefe Publishing

Perceived Organizational Prestige

The self-enhancement explanation of identification pre-dicts that the more prestigious the organization, the greaterthe potential to boost members’ self-esteem through iden-tification (Mael & Ashforth, 1992; Tajfel, 1978): Highprestige helps an individual to assimilate the superiority ofthe organization and “bask in its reflected glory” (Cialdiniet al., 1976). This view of the positive relationship betweenorganizational prestige and identification was consistentlysupported over time and across different contexts. Per-ceived prestige of the profession and perceived organiza-tional prestige were important determinants of organiza-tional identification among a group of scientists in a divi-sion of the Federal Public Health Service (Lee, 1971). Thelevel of identification of college alumni with their formercollege (Mael & Ashforth, 1992) and of accountant alumniwith their former employer (Iyer, Bamber, & Barefield,1997) increased with perceived organizational prestige.Ahearne, Bhattacharya and Gruen (2005) found that whena customer sees the construed external image of a companyas attractive (i.e., customers believe that the attributes thatdistinguish the company are positive and socially valuedby relevant others), customer identification with that com-pany is strengthened. Furthermore, the most important out-come of prestige is the increased affective commitment to-ward the organization, which is then reflected in altruistic

behavior toward colleagues (Carmeli, 2005). The connec-tion with the organization is also expressed in the form oflow turnover intention for internal stakeholders or brandfidelity for clients (Wan Huggins, Riordan, & Griffeth,1998).

Need for Identification

Identification is not just a response to external stimuli, it isalso dependent on individual differences (Ashforth &Mael, 1989). Glynn (1998) advances the idea that, althoughall individuals are at least somewhat receptive to identifi-cation as a means of fostering a sense of belonging and self,“individuals might differ in their propensity to identify withsocial objects” (Glynn, 1998, p. 238) – a propensity thatshe defined as the need for organizational identification.Need for identification is an individual disposition, mani-fested as the desire to be “imprinted upon” (Glynn, 1998,p. 240) by an organization and receptivity to socialization,and negatively associated with a desire for separatenessfrom the organization. Individuals with a high need foridentification have a marked tendency to publicly identifythemselves as organizational members so as to extend theself to incorporate the organization, as part of a broadersearch for meaning (Glynn, 1998). Therefore, this need toidentify will be reflected both in stronger identification

Table 1. List of variables and scales used in the study: definitions and descriptive statistics

Scale No. items Source Likert M SD αPerceived organizational image refers to members’ beliefs aboutoutsiders’ perceptions of the organization (Bergami & Bagozzi,2000; Smidts, Pruyn, & Van Riel, 2001).

8 Mael & Ashforth, 1992 1–5 3.85 .79 .78

Need for organizational identification refers to the propensity toidentify with social objects, the desire to be “imprinted upon” byan organization, and receptivity to socialization, and negatively as-sociated with a desire for separateness from the organization(Glynn, 1998).

5 Kreiner & Ashforth, 2004 1–5 3.64 .73 .71

Organizational identification is the perception of unity with or be-longing to an organization (Ashforth & Mael, 1989)

6 Mael & Ashforth, 1992 1–5 3.58 .92 .87

Group self-esteem refers to the positive or negative value connota-tion one attaches to his/her group (organizational) membership(Ellemers, Kortekaas, & Ouwerkerk, 1999).

4 Ellemers, Kortekaas, &Ouwerkerk, 1999

1–7 6.06 1.11 .68

Affective commitment refers to one’s emotional attachment to his orher organization (Meyer & Allen, 1991).

8 Meyer & Allen, 1991 1–7 5.21 1.00 .73

Job satisfaction is defined as “a pleasurable or positive emotionalstate resulting from the appraisal of one’s job or job experiences”(Locke, 1976, p. 1300).

5 Price, 1997 1–5 4.11 .64 .75

Turnover intention refers to the desire to cease organizational mem-bership (van Knippenberg, van Knippenberg, Monden, & de Lima,2002).

5 Price, 1997 1–5 2.12 .65 .72

Perceived performance refers to the self-perceived performance inthe task.

9 Abramis, 1994 1–5 4.59 .49 .83

Organizational citizenship behavior is defined as “employee behav-ior that is above and beyond the call of duty and is therefore discre-tionary and not rewarded in the context of an organization’s formalreward structure” (Konovsky & Pugh, 1994, p. 656, cit. in Dalal,2005).

19 Moorman & Blakely, 1995 1–7 5.39 .83 .89

S. Boro5 et al.: Integrative Tests of Organizational Identification 113

© 2011 Hogrefe Publishing Social Psychology 2011; Vol. 42(2):111–123

(subjective outcome), as well as in manifested behavior in-tended for the benefit of the organization (such as organi-zational citizenship behavior).

To summarize, both antecedents (organizational pres-tige and need for identification) are expected to impact pos-itively on all dimensions of organizational identification.We also expect a bidirectional association between needfor identification and organizational prestige (Glynn,1998; Kreiner & Ashforth, 2004).

On the one hand, the greater the perceived organization-al prestige, the greater likelihood that individuals wouldexperience the need to identify with that organization. Lessprestigious organizations create an internal conflict for theindividual, since the need for self-enhancement is not beingmet by identifying with that particular organization. Pres-tigious organizations, however, meet both underlying mo-tives for organizational identification: They both provideindividuals with uncertainty reduction (Hogg, 1996) andenhance their self-esteem as a result of the association withthe organization. Therefore, need for identification has noimpediments in being fully experienced and declared inprestigious organizations.

On the other hand, the greater the need to identify withan organization, the more an individual needs to view hisor her organization in positive terms, in order to resolve thecognitive dissonance that results from the need to identifywith a less prestigious organization.

Furthermore, based on the empirical evidence thus far,we expect strong direct links between perceived organiza-tional prestige and turnover intention (negative relation),and between need for identification and organizational cit-izenship behavior (positive relationship). Therefore, ourfirst hypotheses propose that:

H1. Perceived organizational prestige positively impactscognitive, affective, and evaluative identification.

H2. Need for identification positively impacts cognitive,affective, and evaluative identification.

H3. Perceived organizational prestige negatively impactsturnover intention.

H4. Need for identification positively impacts organiza-tional citizenship behavior.

In-Role and Extra-Role Behavior

People who strongly identify with an organization are like-ly to focus on tasks that benefit the whole organization rath-er than those purely motivated by self-interest. Cross-sec-tional as well as longitudinal research carried out in orga-nizational settings (Ellemers, Van Rijswijk, & De Gilder,1998; Haslam, Powell, & Turner, 2000; Ouwerkerk, Elle-mers, & De Gilder, 1999; Tyler, 1999; van Dick, Grojean,Christ, & Wieseke, 2006; van Knippenberg, 2000) has in-

dicated strong positive correlations between organizationalidentification and organizational citizenship behavior. VanKnippenberg and Sleebos (2006) argue that the more peo-ple identify with an organization, the more the values,norms, and interests of the organization are incorporatedinto the individual’s conception of self. Collective interestis experienced as self-interest (i.e., the collective self-inter-est), and individuals are therefore intrinsically motivated tocontribute to the collective (Ashforth & Mael, 1989; Dut-ton, Dukerich, & Harquail, 1994; van Knippenberg & Elle-mers, 2003).

Several studies have pointed to the stronger connectionbetween organizational identification and the display of ex-tra-role behavior, that is, organizational citizenship behav-ior (OCB) rather than in-role behavior. In-role behavior isusually assessed as job performance and the relation be-tween identification and job performance is proved to de-pend upon several factors, such as possessed knowledgeand skills. Furthermore, performance is related to identifi-cation only to the extent that the identity is salient and per-formance is specified in the group (organizational) norms(van Knippenberg & Ellemers, 2003). Extra-role behavioris more under the individual’s control than task perfor-mance, since behavior such as interpersonal help dependsmore upon personal attitude than on knowledge, skills andexisting resources (van Knippenberg, 2000).

Withdrawal Behavior and Intentions

Identification is expected to be negatively related to turn-over intentions, because people are reluctant to give up thepart of their self-definition that is tied to organizationalmembership (van Knippenberg, van Knippenberg, Mon-den, & de Lima, 2002), and indeed research has establisheda negative relationship between identification and turnoverintentions (van Dick, Christ et al., 2004). The desire tomaintain organizational membership is not as central to theorganizational identification concept as it is to the organi-zational commitment concept, however, and in severalstudies, commitment has proved to be more strongly asso-ciated with turnover intentions than is identification (e.g.,Cole & Bruch, 2006; Harris & Cameron, 2005; van Knip-penberg & Sleebos, 2006). In a more fine-grained analysisof the impact of identification and commitment, Harris andCameron (2005) point to the importance of the affectiveelements of both identification and commitment for pre-dicting turnover intentions. This contrasts with the cogni-tive centrality of the organization (cognitive identification)or normative or continuance commitment, which does notuniquely predict turnover intentions.

Job Satisfaction

Commitment is more attitudinal in nature than identifica-tion and is rooted in social exchange processes and employ-

114 S. Boro5 et al.: Integrative Tests of Organizational Identification

Social Psychology 2011; Vol. 42(2):111–123 © 2011 Hogrefe Publishing

ees’ attitude toward the job and the organization (Meyer etal., 2006; van Knippenerg & Sleebos, 2006), so it followsthat commitment should be more closely aligned withmembers’ attitudes toward the organization, such as jobsatisfaction (Pratt, 1998; Reichers, 1985). Job satisfactionwas defined as “a pleasurable or positive emotional stateresulting from the appraisal of one’s job or job experiences”(Locke, 1976, p. 1300). Based on this definition, a majorset of correlates of commitment is found in variables relat-ed to the extent to which the job itself is enjoyable andchallenging (Mathieu & Zajac, 1990; Meyer & Allen,1997). Because people tend to think positively about thingsassociated with the self, identification too may be positive-ly related to job satisfaction (van Knippenberg & vanSchie, 2000), but this should be a more distal associationthan for commitment and satisfaction. Correlations be-tween commitment and job satisfaction, organizational sat-isfaction, and job involvement are consistently higher thanbetween identification and these measures (Mael & Tetrick,1992).

To summarize, evidence of the impact of organizationalidentification on organizational-related attitudes and be-havior is mixed (Johnson & Morgeson, 2005; Musgrove &McGarty, 2008; Stierle, van Dick, & Wagner, 2002). In hismeta-analysis, Riketta (2005) found organizational identi-fication (defined as cognitive identification) to be morestrongly correlated to extra-role behavior than organiza-tional commitment, while commitment was a better predic-

tor than identification for job satisfaction, turnover inten-tion and absenteeism. However, Johnson and Morgeson(2005) found that commitment, OCB, organizational in-volvement, and professional development were indepen-dently predicted by both cognitive and affective identifica-tion, while job satisfaction was predicted by affective iden-tification alone (Johnson & Morgeson, 2005). Van Dickand collaborators (van Dick, Christ et al., 2004) providesempirical data on four samples that supports the strong im-pact of cognitive identification on job satisfaction, whichin turn impacts turnover intention, hence creating a medi-ation path.

Research based on the model of Ellemers et al. generallypoints to the importance of the affective side of identifica-tion for subsequent collective behavior (Bergami & Bagoz-zi, 2000; Ellemers et al., 1998, 1999). The argument is thatit is a sense of emotional involvement with the group (af-fective commitment), rather than the cognitive (self-cate-gorization) or evaluative (group self-esteem) component ofin-group identification which predisposes people to showin-group favoritism (Ellemers et al., 1999). Similarly, Ber-gami and Bagozzi (2000) argued that self-categorizationprovides a cognitive basis for performance of citizenshipbehavior, while affective identification (group commit-ment, in Ellemers’ terms) and group self-esteem supply themotivational force.

In consideration of these findings, we conceptualizedthe relation between the three forms of identification as one

Figure 1. An integrated model of organizational identification.

S. Boro5 et al.: Integrative Tests of Organizational Identification 115

© 2011 Hogrefe Publishing Social Psychology 2011; Vol. 42(2):111–123

of mediation. We argue that the cognitive side of identifi-cation is the primary form of identification, generating theother two forms (affective and evaluative). To respect theoriginal model, we assumed this mediation to be partial.

Thus, we hypothesize that:

H5. The impact of cognitive identification on subsequentvariables is partially mediated by affective and evaluativeidentification.

Furthermore, based on the reviewed research, we expectthat:

H6. Cognitive identification has a stronger direct positiveimpact on organizational-related behaviors (such as OCBand perceived performance) than affective or evaluativeidentification.

H7. Affective identification has a stronger direct impact onorganization-related attitudes (job satisfaction and turnoverintentions) than cognitive or evaluative identification.

Extrapolating from SIT experimental research, we wouldexpect that:

H8. Evaluative identification has a strong negative impacton turnover intention and positive on organization-relatedbehavior.

The hypothesized relations are summarized in the modeldepicted in Figure 1.

Method

Respondents and Procedure

Respondents

We carried out the present study in three Romanian orga-nizations (262 respondents), all of which were from thesame city (Cluj-Napoca): Two were production units(clothing and electronics) and one was a medical unit (aspecial unit of ambulance services). Our sample comprised200 men and 63 women, with a mean age of 31.47 years(ranging from 21 to 54). Of these, 134 were high-schoolgraduates, 31 had attended college, and 98 had a universitydegree. Their tenure on the job varied from 3 months to 27years, with an average of 4.5 years (SD = 4.8). Table 2presents the distributions across the three organizations.Since this distribution is uneven for the medical organiza-tion, we performed a multivariate analysis of the independ-ent variables, to check for possible clusterings within theunits. Differences were nonsignificant with respect to bothperceived organizational prestige (F = .62, p = .53) and theneed for identification (F = .70, p = .49).

Procedure

The questionnaires were anonymous and filled in at theworkplace. One of the four research assistants distributedthem by hand at the beginning of the day and asked forthem to be filled in and dropped in a black box left at theentrance of the organization by the end of the day. Respon-dents were informed that these questionnaires were part ofthe research conducted by the research assistants for theirMasters’ thesis, and they were reassured their question-naires would be used only for scientific purposes. The re-sponse rate was 90%. This high response rate is in part dueto managers being positive and supportive about the ideaof running the research in their organizations, but mainlyto the research assistants being present throughout the dayof the survey in the organization and checking with em-ployees at the end of the day whether they had filled in thequestionnaires.

Instruments

Starting from Ellemer’s model of identification, we em-ployed Mael and Ashforth’s (1992) Organizational Identi-fication Scale (OIS) for the cognitive component of iden-tification, Meyer and Allen’s (1991) Affective Commit-ment Scale (ACS) for the affective dimension ofidentification, and Ellemers et al.’s (1999) Group Self-Es-teem Scale (GSE) for the evaluative dimension. We did notoverlap the identification and commitment constructs, butthe very high correlations between the scales used to assessthem (e.g., .70, reported by Gautam, van Dick, & Wagner,2004), along with confirmatory factor analysis that sup-ports the separate factor loading of the two, allowed us touse them as instruments to assess separate dimensions ofthe same construct. The choice of these three scales for theoperationalization of the three dimensions of identificationis supported by the results of a study on 456 respondents(Boro5, 2008). The study tested the convergent and dis-criminant validity of two different organizational identifi-cation scales and of three different organizational commit-ment scales. Ashforth and Mael’s OIS had the best psycho-metric indices (Cronbach’s α= .85) for the organizational

Table 2. Respondents’ distribution across organizations

Medical-military

Production(electronics)

Production(clothing)

Gender Male 165 23 12

Female 21 24 18

Education High school 112 9 13

College 23 7 1

Universitydegree

51 31 16

Age M (SD) 30.17 (6.71) 35.57 (7.82) 34.8 (7.93)

Tenure M (SD) 4.64 (5.52) 4.59 (2.53) 3.69 (2.41)

116 S. Boro5 et al.: Integrative Tests of Organizational Identification

Social Psychology 2011; Vol. 42(2):111–123 © 2011 Hogrefe Publishing

identification scales and Meyer and Allen’s ACS (Cron-bach’s α = .83) for organizational commitment scales. Fur-thermore, following intercorrelations and confirmatoryfactor analyses, the three chosen scales had the best fit formeasuring Ellemers et al.’s three-dimensional model. Theobtained intercorrelations were (OI-GSE = .46, p < .001;OI-ACS = .58, p < .001, ACS-GSE = .59, p < .001). Whenrunning the confirmatory factor analysis, the model imply-ing three dimensions of the same construct as measured bythese scales, had a better fit (RMSEA = .07; CFI = .87; NFI= .86; PNFI = .74) than a one-dimensional model (RMSEA= .12; CFI = .71; NFI = .69; PNFI = .61) measuring themore general psychological attachment construct.

The scales assessing the organizational correlates are in-troduced in Table 1. Since these scales used different end-points (1–7 or 1–5 Likert scales, respecting the original ver-sion of the scale – see Table 1 for details), we used stan-dardized Z measures in our analyses.

Data Analysis

In order to test our model, we opted for structural equationmodeling using a mixed structural and measurement mod-el. The structural equation modeling (SEM) analysis ofthe integrative model was undertaken using the AMOS(16.0) software. SEM with latent and measurement com-ponents was selected as a statistical methodology because

of its several advantages over regression modeling, in-cluding use of confirmatory factor analysis to reduce mea-surement error by having multiple indicators per latentvariable, better model visualization through its graphicalmodeling interface, the desirability of testing modelsoverall rather than coefficients individually, the ability totest models with multiple dependents, as well as the abil-ity to model mediating variables rather than be restrictedto an additive model as in regression (Garson, 2009). Theresults are reported according to the conventions devel-oped by the American Psychological Association (2009)and by McDonald and Ho (2002). We also checked for theassumptions needed to perform SEM: sample size (263respondents), interval data, and multivariate normality.With one exception, the variables in the model were with-in the range that characterizes normal distributions: Skew-ness varied between –2.05 and .51, and kurtosis between–.17 and 2.06 (Schumaker & Lomax, 2004). The one ex-ception was the kurtosis of evaluative identification (mea-sured by the GSE Scale), which was 5.11. This irregularityappeared to impact on our basic goodness-of-fit measure(χ²). χ² has been noted to be extremely sensitive to thesmallest violations of the assumption of multivariate nor-mality. While in this case it had been advisable to use theSatorra-Bentler χ², which penalizes for the amount of kur-tosis in the data, this index is not available in Amos.Therefore, we computed relative χ², reported in Table 4in the column CMIN/df. Schumaker and Lomax argue that

Figure 2. Standardized structural coefficients of the integrative model of organizational identification.

S. Boro5 et al.: Integrative Tests of Organizational Identification 117

© 2011 Hogrefe Publishing Social Psychology 2011; Vol. 42(2):111–123

Table 3. Matrix of correlations of the variables employed in the model

Mean SD 1. 2. 3. 4. 5. 6. 7. 8. 9.

1. Need for identification 3.61 .74 1 .46** .53** .51** .52** –.36** .15* .42** .50**

2. Perceived org. prestige 3.96 .74 1 .52** .30** .54** –.18** .13* .27** .38**

3. Cognitive identification 3.66 .96 1 .49** .37** –.28** .21** .32** .53**

4. Affective commitment 5.36 .90 1 .33** –.42** .14* .48** .44**

5. Evaluative identification 6.20 .99 1 –.23** .04 .35** .25**

6. Turnover intention 2.02 .59 1 –.41** –.60** –.52**

7. Perceived performance 4.58 .52 1 .32** .44**

8. Job satisfaction 4.24 .58 1 .47**

9. Org. citizenship behavior 5.39 .83 1

Notes. **Correlation is significant at the .01 level (2-tailed). *Correlation is significant at the .05 level (2-tailed).

Table 4. Fit indices for the tested path models

Model χ² df Sig. CMIN/df RMSEA NFI CFI PNFI

Mixed model:Step 1 – CFA of latent variables .83 1 .36 .83 .00 .99 1.00 .16

Step 2 – mixed model 32.08 15 .00 2.13 .06 .96 .98 .40

Concurrent mixed model 1:commitment mediating evaluative identification

31.5 15 .00 2.01 .06 .96 .98 .40

Concurrent mixed model 2:evaluative identification as an independent dimension

32.3 16 .00 2.18 .06 .96 .98 .42

Observable models:Behavioral outcomes 5.71 5 .33 .98 .00 .99 1.00 .28

Attitudinal outcomes 7.50 5 .18 1.50 .04 .98 .99 .23

Table 5. Unstandardized structural coefficients of the proposed mixed model

Regressions Estimate S.E. C.R. p

Org. identification ← Perceived org. prestige .46 .07 6.45 ***

Org. identification ← Need for identification .48 .07 6.79 ***

Affective commitment ← Need for identification .41 .07 5.73 ***

Affective commitment ← Perceived org. prestige –.04 .07 –5.36 .59

Evaluative identification ← Perceived org. prestige .52 .07 6.75 ***

Evaluative identification ← Need for identification .47 .07 6.06 ***

Affective commitment ← Org. identification .29 .05 5.24 ***

Evaluative identification ← Org. identification –.02 .06 –.43 .66

Organization-related attitudes ← Affective commitment .23 .03 6.33 ***

Organization-related attitudes ← Evaluative identification .09 .03 3.01 **

Organization-related attitudes ← Org. identification .05 .03 1.65 .09

Organization-related behaviors ← Evaluative identification .15 .04 3.47 ***

Organization-related behaviors ← Org. identification .21 .04 4.56 ***

Organization-related behaviors ← Organization-related attitudes 1.05 .15 7.02 ***

Organization-related behaviors ← Affective commitment .10 .05 1.87 *

Job satisfaction ← Organization-related attitudes 1.00

Turnover intention ← Organization-related attitudes –1.09 .10 –10.52 ***

Perceived performance ← Organization-related behaviors .51 .07 7.17 ***

Org. citizenship behaviors ← Organization-related behaviors 1.00

Org. citizenship behaviors ← Need for identification .29 .06 4.89 ***

Turnover intention ← Perceived org. prestige .08 .04 2.03 *

Notes. ***Correlation is significant at the .001 level (2-tailed); **correlation is significant at the .01 level (2-tailed); *correlation is significantat the .05 level (2-tailed).

118 S. Boro5 et al.: Integrative Tests of Organizational Identification

Social Psychology 2011; Vol. 42(2):111–123 © 2011 Hogrefe Publishing

a value of 5 or less reflects an acceptable fit, while othersinsist it should be 3 or less (Garson, 2009).

According to Garson (2009), a violation of the multivar-iate normality assumption is not reflected in the path indi-ces of the model so much as in the goodness-of-fit indices.Therefore, because of theoretical considerations, we didnot delete evaluative identification from our model, buttook into account this the kurtosis of this variable whenjudging the fit indices of the tested model. The proposedmodel is presented in Figure 2.

Results

Before addressing the main results, note the descriptive sta-tistics for the variables in the model and the matrix of in-tercorrelations presented in Table 3.

Since we employed latent and observed variables in ourfirst model, we performed the two-step modeling of mixedmodels, as suggested by the work of McDonald and Ho(2002). In the first step, we checked the model fit for theconfirmatory factor analysis for the latent variables. Thismodel presents a very good model fit, thus allowing fortesting of the total, mixed model. The fit indices for bothsteps are presented in Table 4. Except for χ², all reportedindices range within optimal value intervals. We have ex-plained in the previous section why χ² was affected, inde-pendently from the fitness of the model. As specified, wealso computed relative χ². Its value (2.13) is within the lim-its that indicates a fit model (< 3, according to Garson,2009). Considering all the fit indices, we were thus able toassume we had a well fitting model and proceeded with thepath analyses.

We shall discuss the path indices in three clusters: orga-nizational identification antecedents, links between the di-mensions of identification, and the impact of identificationon organizational outcomes. Table 5 presents the unstan-dardized structural coefficients of the model. The standard-ized structural coefficients are represented with arrows inFigure 2.

As predicted (H1), perceived organizational prestige had

a large impact on cognitive (.35) and evaluative (.40) iden-tification, but not on affective commitment (–.03). Needfor identification impacted all three dimensions (H2): .37for cognitive identification, .36 on evaluative identifica-tion, and .34 on affective commitment. As expected, thetwo antecedents were strongly correlated (.47), and needfor identification positively impacted organizational citi-zenship behavior (.27) – H4, but organizational prestige didnot have a significant impact on turnover intentions (.10)– H3.

The impact of cognitive identification on subsequent or-ganizational attitudes and behaviors is partially mediated byaffective commitment (see Table 6 for indirect effects aswell), but not by evaluative identification, as proposed in H5.

As reported in previous research as well as proposed inH6 and H7, cognitive identification has a reduced impacton organization-related attitudes (.12), and a strong impacton organizational behavior (.36). Affective commitmenthas a lesser direct impact on organizational behavior (.16),but a strong impact on organizational attitudes (.46). How-ever, affective commitment impacts behavior as well, albe-it indirectly, mediated by organizational attitudes (see Fig-ure 2 and Table 6). Evaluative identification has a moderateeffect both on organizational behavior (.25) and attitudes(.20) – H8.

Discussion

This study explores an integrative multidimensional modelof organizational identification. We begin by discussing themechanisms through which cognitive identification and af-fective commitment impact on organizational behavior andattitudes. We then focus on the general impact of evaluativeidentification in organizational settings. Throughout thesediscussions, we analyze the impact of personal and organi-zational antecedents on identification.

Our study points to two different mechanisms underly-ing the impact of organizational identification on behavior-al outcomes: On the one hand, the cognitive and evaluativesides of identification have a direct impact on behavior. As

Table 6. Standardized indirect effects for model 1 (presented in Figure 2)

Org. prestige Need foridentif.

Org.identif.

Evaluativeidentif.

Affectivecommit.

Org.attitudes

Org.behaviors

Organizational identification .00 .00 .00 .00 .00 .00 .00

Evaluative identification –.01 –.01 .00 .00 .00 .00 .00

Affective commitment .11 .12 .00 .00 .00 .00 .00

Organizational attitudes .15 .33 .14 .00 .00 .00 .00

Organizational behaviors .14 .23 .16 .15 .37 .00 .00

Organizational citizenship behavior .09 .16 .37 –.06 .14 .56 .00

Perceived performance .08 .13 .30 –.05 .11 .46 .00

Turnover intention –.13 –.27 –.21 –.16 –.38 .00 .00

Job satisfaction .12 .25 .20 .15 .35 .00 .00

S. Boro5 et al.: Integrative Tests of Organizational Identification 119

© 2011 Hogrefe Publishing Social Psychology 2011; Vol. 42(2):111–123

expected, cognitive identification leads to behavior that isbeneficial for the organization. However, this impact isstronger for extra-role than for in-role behavior. In line withprevious research (van Knippenberg & Ellemers, 2003),neither dimension of identification has a relevant direct im-pact on perceived performance. Such a relation is expectedonly to the extent that the identity is salient and perfor-mance is specified in the group norms of the organization(van Knippenberg & Ellemers, 2003). However, strong di-rect and indirect links were evidenced between cognitiveidentification and organizational citizenship behavior. Thissupports classical SIT assumptions postulating that self-categorization, that is, defining oneself in terms of belong-ing to a social group, leads to exerting effort in the nameof that group (i. e. collective action) (Voci, 2006).

On the other hand, affective commitment did not signifi-cantly impact organizational behavior directly. This appears tocontradict previous research, which consistently points to therole of commitment for pro-organizational behavior (Mathieu& Zajac, 1990; Meyer et al., 2006). Instead, our results pointto the operation of commitment through a different mecha-nism. While cognitive identification impactsbehaviordirectly,commitment influences behavior via attitudes. Higher com-mitment changes the attitudes toward the organization: Themore committed an employee is, the more motivated that per-son is (Meyer, Becker, & Vandenberghe, 2004), the more sat-isfied with the job (Mathieu & Zajac, 1990), and the lesser hisintent to leave the organization (Mathieu & Zajac, 1990). Thestrong impact of organizational attitudes on organization-re-lated behavior link therefore commitment to behavior in anindirect manner. This mediation might explain both the smalldirect impact of commitment on organizational behaviors aswell as the consistently reported strong impact of commitmenton organizational outcomes (Mathieu & Zajac, 1990; Meyeret al., 2004). Consequently, the attitudinal-based change inbehavior might be strongerand more enduring, as documentedby fundamental research on behavioral change (Kraus, 1995).

A second differentiation of mechanisms between cogni-tive identification and affective commitment was apparentfor attitudinal outcomes. In this case, affective commitmentimpacts attitudes directly, while cognitive identificationimpacts attitudinal outcomes only when mediated by affec-tive commitment. In other words, cognitive identificationsuffices to exert behavior in the name of the group, but tochange individual attitudes toward the organization, deeperaffective ties have to be present. This finding is in line withprevious studies (Bergami & Bagozzi, 2000) that demon-strate the mediatory role of affective commitment betweencognitive identification and organizational outcomes.

Thus, evidence in this study suggests that:1. Cognitive identification has a direct effect on organiza-

tional behavior.2. Cognitive identification has an indirect effect on organi-

zational attitudes, mediated by affective commitment.3. Affective commitment has a strong impact on organiza-

tional attitudes, and impacts behavior via this attitudinalroute.

4. Hence, cognitive identification has a double route to im-pact organizational behavior: a direct one, and an indi-rect one, via affective commitment and attitudes.

Another interesting set of findings refer to the evaluativedimension of identification. Our findings on evaluativeidentification are limited by the distribution curve for thisvariable, which presented a high kurtosis. The right-skewed distribution of this measure indicated that no mat-ter the strength of cognitive identification or affective com-mitment, the respondents declared a high evaluative iden-tification. This simple descriptive measure in itself is aconfirmation of the need for the self-enhancement hypoth-esis. People need to derive a positive sense of self fromtheir affiliation with an organization. What is interesting tonotice though is the minimal impact of this evaluative iden-tification on subsequent behavior and attitudes.

Evaluative identification has no major behavioral benefitsfor the group, in that it neither leads to group-benefiting be-havior nor to a better performance. The impact on turnoverintentions or job satisfaction is also moderate at best. Thisevidence contradicts the predictions of the personal self-in-terest explanation for identification. This widespread (Bet-tencourt, Dorr, Charlton, & Hume, 2001; Tajfel, 1972) yetarguable (Brown, 2000; Hogg, 1996) perspective maintainsthat group members gain self-enhancement due to groupmembership. In other words, the primary reason why peopleidentify with groups is to boost their own self-esteem throughthe connection with a high-status group. Yet, although ourresults prove there is indeed a strong relationship betweenperceived organizational prestige and evaluative identifica-tion (group self-esteem), this does not lead to the expectedbehavioral or attitudinal outcomes. This finding is relevantfor managerial strategic practices, by underlining the fact thatpolicies meant to improve organizational image do not al-ways reap all the expected benefits in terms of consequentemployee behavior and attitudes. In other words, maintaininga positive organizational image is not the panacea some stud-ies suggest (Dukerich, Golden, & Shortell, 2002).

The findings related to our other antecedent support thisview of identification. Need for organizational identifica-tion presented just as strong links with the three dimensionsof identification as perceived organizational prestige. Interms of underlying mechanisms, this supports the idea thatidentification is indeed more related to the core of humanfunctioning in society than the self-enhancement hypothe-sis implies. Identification does not occur only when it isconvenient for a person to be affiliated to a high prestigegroup. Defining ourselves in terms of group belonging is amore stable and ubiquitous need, independent of the targetgroup. As Glynn stated, people are “looking for somethingto identify with as a part of a broader search for meaning”(Glynn, 1998, p. 240). This finding is in line with assump-tions of the uncertainty reduction model (Hogg, 1996) andthe meaning-making perspective as the underlying motiva-tion for social identification.

120 S. Boro5 et al.: Integrative Tests of Organizational Identification

Social Psychology 2011; Vol. 42(2):111–123 © 2011 Hogrefe Publishing

Limitations and Future Research

The most apparent limitation of this study is the use ofcross-sectional survey to test a multiple mediation model.Using data collected at the same point in time from thesame persons employing the same method (self-report)limits our claims of causality. Although a preliminarycross-sectional investigation was needed before proceed-ing further, the model proposed here should be tested in alongitudinal design in future research.

For a more accurate testing of the theoretical model weproposed, future studies should also develop a measure-ment for the affective side of identification that empiricallydistinguishes it from affective commitment. For the pur-pose of the present research, we opted for validated andlargely recognized scales in organizational settings in orderto eliminate validity concerns. Therefore, we had to overlapin measurement affective identification and commitment.Future studies, however, need to investigate empirically thedistinction between affective identification and commit-ment and elaborate instruments that discriminate betweenthe two. Only then could the relation between identificationand commitment be properly tested in integrative models.

Conclusions

Our study points to a number of particularities of how iden-tification takes place in organizational settings. First, ourresults bring more clarity to previous contradictory find-ings related to the impact of cognitive identification andaffective commitment. Our evidence suggests that (1) af-fective commitment does indeed mediate the relationshipbetween cognitive identification and attitudinal outcomes;(2) organizational attitudes mediate the impact of affectivecommitment on organizational behaviors; (3) cognitiveidentification has a double route to impact organizationalbehavior: a direct one and an indirect one, via affectivecommitment and attitudes. Consequently, the differentia-tion between behavioral and attitudinal outcomes is neededfor the accurate prediction of organizational identification.

Second, we advance counterevidence of the self-en-hancement hypothesis. When moving beyond the minimalgroup paradigm, evaluative identification (group self-es-teem) largely loses its impact. It is not reflected in outcomesrelevant for the organization, such as extra-role behavior orintention to remain in the organization. We suggest that inorganizational settings, identification might be more a mat-ter of search for meaning than a search for self-esteem.

Acknowledgments

The authors thank N. Ellemers for her valuable suggestionsand comments on earlier drafts of this paper. Also, we thankC. Mure5an, G. tomlea, A. Jeler, and A. Madar for their

hard fieldwork in gathering the data, and C. Ermiya forproofreading the manuscript. Special thanks go to GuidoHertel and two anonymous reviewers for the rich and in-formed feedback that helped us shape the previous versionsof this paper.

References

Abramis, D. J. (1994). Work role ambiguity, job satisfaction, andjob performance: Meta-analyses and review. PsychologicalReports, 75, 1411–1433.

Ahearne, M., Bhattacharya, C. B., & Gruen, T. (2005). Anteced-ents and consequences of customer company identification:Expanding the role of relationship marketing. Journal of Ap-plied Psychology, 90, 574–585.

American Psychological Association. (2009). Publication manu-al of the American Psychological Association (6th ed.). Wash-ington, DC: Author.

Ashforth, B. E., & Mael, F. A. (1989). Social identity theory andthe organization. Academy of Management Review, 14, 20–39.

Bergami, M., & Bagozzi, R. P. (2000). Self-categorization, affec-tive commitment and group self-esteem as distinct aspects ofsocial identity in the organization. British Journal of SocialPsychology, 39, 555–577.

Bettencourt, B. A., Dorr, N., Charlton, K., & Hume, D. L. (2001).Status differences and in-group bias: A meta-analytic exami-nation of the effects of status stability, status legitimacy, andgroup permeability. Psychological Bulletin, 127, 520–542.

Boro5, S. (2008). Organizational identification: Theoretical andempirical analyses of competing conceptualizations. Cognitie,Creier, Comportament (Cognition, Brain, Behavior), 1, 1–28.

Brown, R. (2000). Social identity theory: Past achievements, cur-rent problems and future challenges. European Journal of So-cial Psychology, 29, 634–667.

Carmeli, A. (2005). Perceived external prestige, affective com-mitment, and citizenship behaviors. Organization Studies, 26,443–464.

Cialdini, R. B., Borden, R. J., Thorne, A., Walker, M. R., Freeman,S., & Sloan, L. R. (1976). Basking in reflected glory: Three(football) field studies. Journal of Personality and Social Psy-chology, 34, 366–375.

Cole, M. S., & Bruch, H. (2006) Organizational identity strength,identification, and commitment and their relationships to turn-over intention: Does organizational hierarchy matter? Journalof Organizational Behavior, 25, 585–605.

Dalal, R. S. (2005). A meta-analysis of the relationship between or-ganizational citizenship behavior and counterproductive workbehavior. Journal of Applied Psychology, 90, 1241–1255.

Doosje, B., Ellemers, N., & Spears, R., (1995). Perceived intra-group variability as a function of group status and identifica-tion. Journal of Experimental Social Psychology, 31, 410–436.

Dukerich, J. M., Golden, B. R., & Shortell, S. M. (2002). Beautyis in the eye of the beholder: The impact of organizational iden-tification, identity, and image on the cooperative behavior ofphysicians. Administrative Science Quarterly, 47, 507–533.

Dutton, J. E., Dukerich, J. M., & Harquail, C. V. (1994). Organi-zational images and member identification. AdministrativeScience Quarterly, 39, 239–263.

S. Boro5 et al.: Integrative Tests of Organizational Identification 121

© 2011 Hogrefe Publishing Social Psychology 2011; Vol. 42(2):111–123

Ellemers, N., De Gilder, D., & Haslam, S. A. (2004). Motivatingindividuals and groups at work: A social identity perspectiveon leadership and group performance. Academy of Manage-ment Review, 29, 459–478.

Ellemers, N., Kortekaas, P., & Ouwerkerk, J. W. (1999). Self-cat-egorization, commitment to the group and group self-esteemas related but distinct aspects of social identity. European Jour-nal of Social Psychology, 29, 371–389.

Ellemers, N., Van Rijswijk, W., & De Gilder, D. (1998). Groupcommitment as a moderator of attributional and behavioral re-sponses to power use. European Journal of Social Psychology,28, 555–573.

Garson, D. G. (2009). Structural equation modeling. Retrievedfrom http://faculty.chass.ncsu.edu/garson/PA765/struc-tur.htm#faq

Gautam, T., van Dick, R., & Wagner, U. (2004). Organizationalidentification and organizational commitment: Distinct aspectsof two related concepts. Asian Journal of Social Psychology,7, 301–315.

Glynn, M. A. (1998). Individuals need for organizational identi-fication (nOID): Speculations on individual differences in thepropensity to identify. In D. A. Whetten & P. C. Godfrey(Eds.), Identity in organizations: Building theory through con-versations (pp. 238–244). Thousand Oaks, CA: Sage.

Harris, G. E., & Cameron, J. E. (2005). Multiple dimensions oforganizational identification and commitment as predictors ofturnover intentions and psychological well-being. CanadianJournal of Behavioral Science, 37, 159–169.

Haslam, S. A., Powell, C., & Turner, J. C. (2000). Social identity,self-categorization, and work motivation: Rethinking the con-tribution of the group to positive and sustainable organization-al outcomes. Applied Psychology: An International Review,49, 319–339.

Hogg, M. A. (1996). Social identity, self-categorization, and thesmall group. In E. H. Witte & J. H. Davis (Eds.), Understand-ing group behavior: Vol. 2. Small group processes and inter-personal relations (pp. 227–253). Mahwah, NJ: Erlbaum.

Iyer, V. M., Bamber, E. M., & Barefield, R. M. (1997). Identifica-tion of accounting firm alumni with their former firm: Ante-cedents and outcomes. Accounting, Organizations and Society,22, 315–336.

Johnson, M. D., & Morgeson, F. P. (2005, August). Cognitive andaffective identification in organizational settings. Paper pre-sented at the 65th Annual Meeting of the Academy of Man-agement, Honolulu, HI, USA.

Kraus, S. J. (1995). Attitudes and the prediction of behavior: Ameta-analysis of the empirical literature. Personality and So-cial Psychology Bulletin, 21, 58–75.

Kreiner, G. E., & Ashforth, B. E. (2004). Evidence toward an ex-panded model of organizational identification. Journal of Or-ganizational Behavior, 25, 1–27.

Lee, S. M. (1971). An empirical analysis of organizational iden-tification. Academy of Management Journal, 14, 213–226.

Locke, E. A. (1976). The nature and causes of job satisfaction. InM. D. Dunnette (Ed.), Handbook of industrial and organiza-tional psychology (pp. 1297–1349). Chicago, IL: Rand Mc-Nally.

Mael, F. A., & Ashforth, B. E. (1992). Alumni and their alma mat-er: A partial test of the reformulated model of organizationalidentification. Journal of Organizational Behavior, 13,103–123.

Mael, F. A., & Tetrick, L. E. (1992). Identifying organizationalidentification. Educational and Psychological Measurement,52, 813–824.

Mathieu, J. E., & Zajac, D. M. (1990). A review and meta-analysisof the antecedents, correlates, and consequences of organiza-tional commitment. Psychological Bulletin, 108, 171–194.

McDonald, R. P., & Ho, M. H. (2002). Principles and practice inreporting structural equation analyses. Psychological Meth-ods, 7, 64–82.

Meyer, J. P., & Allen, N. J. (1991). A three-component conceptu-alization of organizational commitment. Human ResourceManagement Review, 1, 61–89.

Meyer, J. P., & Allen, N. J. (1997). Commitment in the workplace:Theory, research and applications. Thousand Oaks, CA: Sage.

Meyer, J. P., Becker, T. E., & van Dick, R. (2006). Social identitiesand commitments at work: Toward an integrative model. Jour-nal of Organizational Behavior, 27, 665–683.

Meyer, J. P., Becker, T. E., & Vandenberghe, C. (2004). Employeecommitment and motivation: A conceptual analysis and inte-grative model. Journal of Applied Psychology, 89, 991–1007.

Moorman, R. H., & Blakely, G. L. (1995). Individualism-collec-tivism as an individual difference predictor of organizationalcitizenship behavior. Journal of Organizational Behavior, 16,127–142.

Musgrove, L., & McGarty, C. A. (2008). Opinion-based groupmembership as a predictor of collective emotional responsesand support for pro- and anti-war action. Social Psychology,39, 37–47.

Ouwerkerk, J. W., Ellemers, N., & De Gilder, D. (1999). Groupcommitment and individual effort in experimental and organi-zational contexts. In N. Ellemers, R. Spears, & B. Doosje(Eds.), Social identity (pp. 185–204). Oxford: Blackwell.

Pratt, M. G. (1998). To be or not to be? Central questions in or-ganizational identification. In D. A. Whetten, & P. C. Godfrey(Eds.), Identity in organizations: Building theory through con-versations (pp. 171–207). Thousand Oaks, CA: Sage.

Price, J. L. (1997). Handbook of organizational measurement. In-ternational Journal of Manpower, 8, 305–558.

Reichers, A. E. (1985). A review and reconceptualization of or-ganizational commitment. Academy of Management Review,10, 465–476.

Riketta, M. (2005). Organizational identification: A meta-analy-sis. Journal of Vocational Behavior, 66, 358–384.

Riketta, M., & van Dick, R. (2005). Foci of attachment in orga-nizations: A meta-analytic comparison of the strength and cor-relates of workgroup versus organizational identification andcommitment. Journal of Vocational Behavior, 67, 490–510.

Schumacker, R. E., & Lomax, R. G. (2004). A beginner’s guide tostructural equation modeling, Second edition. Mahwah, NJ:Erlbaum.

Smidts, A., Pruyn, A. H., & Van Riel, C. B. M. (2001). The impactof employee communication and perceived external prestigeon organizational identification. Academy of ManagementJournal, 49, 1051–1062.

Stierle, C., van Dick, R., & Wagner, U. (2002). Success or failure?Personality, family, and intercultural orientation as determi-nants of expatriate managers’ success. Zeitschrift für Sozial-psychologie, 33, 209–218.

Tajfel, H. (1972). La categorization sociale [Social categoriza-tion]. In S. Moscovici (Ed.), Introduction à la psychologie so-ciale (Vol. 1, pp. 272–302). Paris: Larousse.

122 S. Boro5 et al.: Integrative Tests of Organizational Identification

Social Psychology 2011; Vol. 42(2):111–123 © 2011 Hogrefe Publishing

Tajfel, H. (Ed.). (1978). Differentiation between social groups.Studies in the social psychology of intergroup relations. Lon-don: Academic Press.

Tyler, T. (1999). Why people cooperate with organizations: Anidentity-based perspective. Research in Organizational Behav-ior, 21, 201–246.

van Dick, R., Christ, O., Stellmacher, J., Wagner, U., Ahlswede,O., Grubba, C., . . . Tissington, P. A. (2004). Should I stay orshould I go? Explaining turnover intentions with organization-al identification and job satisfaction. British Journal of Man-agement, 15, 351–360.

van Dick, R., Grojean, M. W., Christ, O., & Wieseke, J. (2006).Identity and the extra-mile: Relationships between organiza-tional identification and organizational citizenship behavior.British Journal of Management, 17, 283–301.

van Dick, R., Wagner, U., Stellmacher, J., & Christ, O. (2004).The utility of a broader conceptualization of organizationalidentification: Which aspects really matter? Journal of Occu-pational and Organizational Psychology, 77, 171–191.

van Knippenberg, D. (2000). Work motivation and performance:A social identity perspective. Applied Psychology: An Inter-national Review, 49, 357–371.

van Knippenberg, D., & Ellemers, N. (2003). Social identity andgroup performance: Identification as the key to group-orientedefforts. In S. A. Haslam, D. van Knippenberg, M. J. Platow, &N. Ellemers (Eds.), Social identity at work: Developing theoryfor organizational practice (pp. 29–42). New York: Psycholo-gy Press.

van Knippenberg, D., & Sleebos, E. (2006). Organizational iden-tification versus organizational commitment: Self-definition,social exchange, and job attitudes. Journal of OrganizationalBehavior, 27, 571–584.

van Knippenberg, D., van Knippenberg, B., Monden, L., & deLima, F. (2002). Organizational identification after a merger:A social identity perspective. British Journal of Social Psy-chology, 41, 233–252.

van Knippenberg, D., & van Schie, E. C. M. (2000). Foci and cor-relates of organizational identification. Journal of Occupation-al and Organizational Psychology, 73, 137–147.

Voci, A. (2006). Relevance of social categories, depersonalizationand group processes: Two field tests of self-categorization the-ory. European Journal of Social Psychology, 36, 73–90.

Wan-Huggins, V., Riordan, C. M., & Griffeth, R. W. (1998). Thedevelopment and longitudinal test of a model of organizationalidentification. Journal of Applied Social Psychology, 28,724–749.

Received September 22, 2009Final revision received March 4, 2010Accepted April 2, 2010

Smaranda Boro5

Department of Organization StudiesTilburg UniversityRoom P3.108Warandelaan 2PO Box 901535000 LE TilburgThe NetherlandsE-mail [email protected]

S. Boro5 et al.: Integrative Tests of Organizational Identification 123

© 2011 Hogrefe Publishing Social Psychology 2011; Vol. 42(2):111–123