motivating individuals and groups at work: a social ... · ment of individual goals or outcomes...

MOTIVATING INDIVIDUALS AND GROUPS ATWORK: A SOCIAL IDENTITY PERSPECTIVE ON

LEADERSHIP AND GROUP PERFORMANCE

NAOMI ELLEMERSLeiden University

DICK DE GILDERFree University, Amsterdam

S. ALEXANDER HASLAMUniversity of Exeter

We argue that additional understanding of work motivation can be gained by incor-porating current insights concerning self-categorization and social identity processesand by examining the way in which these processes influence the motivation andbehavior of individuals and groups at work. This theoretical perspective that focuseson the conditions determining different self-definitions allows us to show how indi-vidual and group processes interact to determine work motivation. To illustrate theadded value of this approach, we develop some specific propositions concerningmotivational processes underpinning leadership and group performance.

In theoretical accounts of work motivation,scholars examine the factors that energize, di-rect, and sustain work-related behavior (e.g.,Pinder, 1998). They aim to understand (1) whichconditions encourage people to invest behav-ioral energy in their work (energize), (2) whichactivities people are likely to focus their effortson (direction), and (3) what makes people persistin such efforts over time (persistence). This hasresulted in the development of a range of workmotivation models (see Steers, Porter, & Bigley,1996). These show how the different aspects ofmotivation operate, as well as how they areinterrelated.

Some of these models primarily address ener-gizing factors, describing the needs that may befulfilled by work-related behavior (e.g., Maslow,1943), and specify how workers may be moti-vated by appealing to particular needs (e.g.,Hackman & Oldham, 1976). Others provide in-sight into the direction work-related efforts arelikely to take by examining the cognitive pro-cesses that underlie behavioral choices (e.g., ex-pectancy theory [Vroom, 1964], equity theory[Mowday, 1979]). Finally, reinforcement theories(e.g., Komaki, Coombs, & Schepman, 1996),based on psychological learning principles (e.g.,operant conditioning), help us understand whycertain behaviors are more likely to be sus-

tained than others. Some theories also addressmultiple components of the motivation pro-cess—in particular, goal-setting theory (e.g.,Locke & Latham, 1990), which relates to bothmotivational direction and persistence.

Over the years, empirical research has pro-vided support for the validity of each of thesemotivational processes (Ambrose & Kulik, 1999)and has demonstrated their relevance to work-related behavior (see also Mitchell, 1982). Yetdespite the different focus of each of these ap-proaches to work motivation, one striking com-monality is that, to date, they have been usedmainly to understand processes underlying thebehavior of individual workers as separateagents. That is, theory and research in workmotivation have focused mainly on the individ-ual needs people may have, their own indepen-dent goals and expectations, or the personaloutcomes they find rewarding. At the same time,developments in the workplace have created arange of situations in which the function of in-dividual needs, goals, expectations, or rewardsis less clear, not least because individual work-ers have to function in concert and cannot al-ways be seen as representing independent en-tities. As a result, workers are not necessarilydriven by personal considerations only; instead,individual motivation is projected on, informed

� Academy of Management Review2004, Vol. 29, No. 3, 459–478.

459

by, and adapted to the needs, goals, expecta-tions, or rewards of the team or organization inwhich individuals work.

For instance, with the increasing proportion ofworkers involved in exchange of knowledge orprovision of services, instead of production ofgoods (Cascio, 1995; Gutek, 1995), it has becomemore difficult to define individual work perfor-mance (see also Brief & Motowidlo, 1986) or toassess individual productivity unambiguously.As a result, it is not always clear how insightswith respect to individual goal setting or rein-forcement might apply in these situations (Kohn,1993; Pearce, 1987). Additionally, people nowa-days tend to work more in (self-managed) teamsthan before (Parker, 1993; Smith, 1997), requiringthem to support each other to achieve commongoals (e.g., in multidisciplinary project teams),instead of focusing only on the achievement ofindividual outcomes (Schaubroeck & Ganster,1991).

While motivational processes may apply tothe achievement of collective goals or outcomesin exactly the same way they do individualgoals or outcomes, we do not know the implica-tions of such a shift from the individual to thecollective, since this has not constituted a sys-tematic topic of research (see also Ambrose &Kulik, 1999, and Wegge & Haslam, 2003). Further-more, it is unclear how these motivational prin-ciples operate when personal goals or expecta-tions (e.g., achieving individual performancetargets) are incompatible with collective goalsor expectations (e.g., helping new colleaguesadapt).

Finally, given that lifetime employment now-adays is exceptional and that organizations of-fer less security than before (Smith, 1997), long-term exchange relationships betweenindividual workers and organizations have be-come less viable as a reinforcement tool. Thechallenge for motivation theorists, therefore, isto cater to these contemporary work situations—for instance, by helping us understand (1) howpeople are energized to engage in behaviorsthat are significant primarily at a collectivelevel, such as “service provision” or organiza-tional citizenship behavior (Brief & Motowidlo,1986; Organ, 1988); (2) how people direct theiractivities toward individual as well as collec-tive goals, particularly when these seem incom-patible; and (3) how people sustain behavioraleffort on behalf of the collective through organ-

izational changes or in the face of insecure jobprospects (see also Meyer & Allen, 1997).

In this article we consider more explicitly howcurrent insights into work motivation may bedeveloped to incorporate these more complexsituations. In doing this, we use social identityand self-categorization principles—that is, thesocial identity approach (Tajfel, 1978; Tajfel &Turner, 1979; Turner, Hogg, Oakes, Reicher, &Whetherell, 1987; see also Ashforth & Mael,1989)—to develop a metatheoretical perspectivethat can help define different behavioral mo-tives in individual as well as in collective terms(e.g., applying to the team or organization). Wepropose that a social identity approach can helpspecify the circumstances under which workersare likely to conceive of themselves either asseparate individuals or as part of a collective(Ellemers, Spears, & Doosje, 2002). Furthermore,we review some initial evidence showing howthis can help us understand behavioral motiva-tion in contemporary work settings (see alsoHaslam, 2001; Haslam & Ellemers, in press;Haslam, Van Knippenberg, Platow, & Ellemers,2003; Hogg & Terry, 2000; Van Dick, 2001), and wedevelop some specific predictions with respectto leadership and group performance that followfrom this theoretical perspective.

INDIVIDUALS AND GROUPS AT WORK

As we argued above, an important definingcharacteristic of contemporary work situationsis that they often require individuals toalign—at least to some extent—with a collec-tive, such as their work team or the organizationas a whole. As a result, workers are expected toadopt converging goals and to sacrifice (short-term) individual interests (e.g., by working over-time) in order to achieve (more long-term) collec-tive outcomes (e.g., attracting new business).Accordingly, others before us have concludedthat further developments in motivation theoryshould focus on its applicability to teams aswell as individuals (e.g., Erez, Kleinbeck, & Thi-erry, 2001; Sussmann & Vecchio, 1982). However,in their recent review of over 200 empirical stud-ies on work motivation, Ambrose and Kulik(1999) conclude that relatively little is knownabout motivation in workgroups. At the sametime they maintain, “As organizations continueto move toward group-based systems, research

460 JulyAcademy of Management Review

on motivation within groups is increasingly im-portant” (1999: 274).

The traditional approach to this problem hasbeen to adopt an exchange orientation and tocreate situations in which the ultimate achieve-ment of individual goals or outcomes (e.g., get-ting a pay raise) depends on the attainment ofcollective goals (e.g., an increase in organiza-tional profits). Thus, even when the aim is tomotivate workers to exert themselves on behalfof the collective, traditional approaches to workmotivation tend to locate the primary motivatingmechanism in the individual as a separate en-tity. That is, the desired behavior is reinforcedby pointing to the interdependence betweenpersonal and collective outcomes. Essentially,this implies that the motivation to achieve acollective performance is regarded as derivedfrom individual concerns and motives (e.g., Ilgen& Sheppard, 2001). In other words, the commonassumption underlying previous work on groupmotivation is that people tend to behave in waysthat seem to be rewarding from an individualpoint of view, without systematically consider-ing how individual behavioral preferences maybe adapted to align with collective concerns orjoint goals (see Shamir, 1991, for a similar obser-vation).

Although we would not dispute that the ca-pacity of a workgroup or organization to providerewards or other desired outcomes may consti-tute a powerful motivating force for individualworkers, we think that our understanding of therange of motivational processes that may oper-ate in workgroups could be enriched by alsoconsidering the ways in which groups maycome, in and of themselves, to represent inter-nalized values and important identities (seealso Tyler, 2002). Indeed, to the extent that cur-rent theories of work motivation focus on theindividual as the primary or sole source of self-conception (e.g. Brief & Aldag, 1981), all expec-tations, goals, and outcomes that relate to theworkgroup or organization are considered to beextrinsic to the self. However, the propositionthat there are circumstances in which peoplemay come to adopt a primary definition of theself in collective terms opens up the possibilitythat group-based expectations, goals, or out-comes are sometimes regarded as intrinsicsources of motivation.

This shift from a conception of self in individ-ual terms to a conception of self in collective

terms and the resulting redefinition of motiva-tional forces as external or internal to the selfare relevant for issues of work motivation, im-plying that, compared to the motivation to worktoward common goals that is derived from per-ceived interdependence of individuals or froman exchange relationship between the individ-ual and the group, a concern with the collectiveself provides a much broader and more powerfulsource of group-based motivation (see alsoCoates, 1994, and Lembke & Wilson, 1998). In-deed, a self-conception in collective termswould energize people to exert themselves onbehalf of the group, facilitate the direction ofefforts toward collective (instead of individual)outcomes, and help workers sustain their loyaltyto the team or organization through times inwhich this is not individually rewarding.

As a result, when the definition of self shiftsfrom being personal (“I”) to collective (“we”), ex-actly the same motivational processes that ap-ply to the individual self may come to apply tothe collective self. Thus, whereas needs, goals,or expected outcomes are still likely to motivatethe behavior of individual workers, when theyconceive of themselves in collective terms, theseare needs of the group, collective goals, andexpected group outcomes. In this article we pro-pose that an analysis of the circumstances un-der which the self tends to be defined in collec-tive instead of individual terms can help predictin which situations the group may come to rep-resent an intrinsic source of motivation, or whenit is more likely to remain extrinsic to the self.

SELF-CATEGORIZATION ANDSOCIAL IDENTITY

The central assumption underlying socialidentity theory (Tajfel, 1974, 1975, 1978) is thatwhile in some social situations people think ofthemselves as independent individuals who in-teract with each other on the basis of personalcharacteristics or preferences (e.g., in friendshipgroups), there are many social settings in whichpeople primarily think of themselves and othersin terms of particular group memberships (e.g.,in terms of their professional roles). In earlyversions of social identity theory, Tajfel (1978)and Tajfel and Turner (1979) specified three in-trapsychological processes that underlie suchgroup-based social interaction—namely, socialcategorization, social comparison, and social

2004 461Ellemers, de Gilder, and Haslam

identification. In later elaborations of the socialidentity approach, which are subsumed underthe term self-categorization theory (Turner et al.,1987), researchers specified the conditions underwhich different definitions of self are likely tobecome salient (focusing on category accessibil-ity and category fit) and detailed the conse-quences of those different definitions for socialperception and social behavior (e.g., Oakes,Haslam, & Turner, 1994; Spears, Oakes, Ellemers,& Haslam, 1997). A more detailed discussion ofthis theoretical perspective is beyond the scopeof this article and can be found elsewhere (e.g.,Ellemers, Haslam, Platow, & Van Knippenberg,2003; Haslam, 2001; Haslam & Ellemers, in press).Here, though, we briefly outline the psychologi-cal mechanisms that are most pertinent to ourcurrent analysis.

Social categorization refers to the notion thatin many situations people organize social infor-mation by categorizing individuals into groups.This enables them to focus on collective proper-ties that are relevant to the situation at hand(e.g., students versus teachers), while neglectingthe “noise” of other variations (e.g., differencesin age or clothing style) that occur among indi-viduals within the same group. Generally, aparticular categorization is more likely to beused when group memberships are relativelyinvariable over time, whereas any category be-comes less useful as an information-organizingprinciple to the extent that individuals are likelyto change from one group to another (groupboundary permeability [e.g., Ellemers, 1993]). Forinstance, when people work in a career systemwhere they are only judged on the basis of indi-vidual merit, this encourages a conception ofself in individual terms and makes employeesfocus on individualistic motives (e.g., self-development, career progress). However, a worksituation in which people are systematically ex-cluded from certain rewards or opportunities onthe basis of their category membership (e.g.,their age, gender, or ethnic background) inducesthem to think of themselves in terms of thatcategorization (Schmitt, Ellemers, & Brans-combe, 2003), with the result that their categorymembership becomes more cognitively accessi-ble (Oakes, 1987).

Social comparison is the process by which asocial categorization is invested with meaning.While people may have a relatively clear idea ofthe range of properties that apply to a particular

group, proponents of the social identity ap-proach maintain that social comparisons withother groups (e.g., sales persons versus custom-ers in a store/sales persons versus productionworkers in the organization) determine whichfeatures or behavioral norms help to define thegroup in a particular situation. Generally, thesefeatures are those that distinguish the groupfrom relevant comparison groups (e.g., Spears,Doosje, & Ellemers, 1997; Van Rijswijk & Ellem-ers, 2002). Thus, which different possible groupmemberships will become salient depends onthe so-called comparative and normative fit of aparticular categorization to the situation athand (Haslam & Turner, 1992; Oakes, 1987;Oakes et al., 1994).

For instance, when production workers andsales representatives try to improve the logisticsof a production process, differences betweenthem are likely to become salient, not only be-cause the individuals belonging to these twogroups have systematically different work expe-riences (comparative fit) but also because thenature of the problems they are likely to encoun-ter depends meaningfully on the content of theirwork (normative fit). However, when these sameindividuals are concerned with the developmentof an affirmative action program, a categoriza-tion in terms of ethnic or gender identity willprovide a better comparative and normative fitand, hence, will constitute a more appropriateguide for defining their position in relation toothers than distinctions based on professionalroles. As a result, what defines members of thegroup may differ from one situation to the next,depending on the comparative context and theways in which group members are distinct fromothers in that context (e.g., Barreto & Ellemers,2001).

Social identification is the process by whichinformation about social groups is related to theself. That is, it refers to the inclination of a par-ticular individual to perceive himself or herselfas representative of a particular group, whichmakes the individual perceive characteristicgroup features as self-descriptive and leads himor her to adopt distinctive group norms as guide-lines for his or her own behavior. While most ofus belong to multiple groups simultaneously,the relative degree to which we see each ofthese different identities as self-descriptive in aparticular situation or at a given point in timewill determine the extent to which these identi-


ties tend to affect our motivated behavior in thatcontext. A well-known phenomenon reflectingthe operation of this process is that people arerelatively willing to identify with groups thatseem to contribute to a positive sense of self,such as high-status or high-power groups (Elle-mers, 1993; Haslam, Powell, & Turner, 2000;Spears et al., 1997).

However, additional concerns may moderateor even override such identity enhancement mo-tives (see also Ashforth & Kreiner, 1999, and El-lemers & Barreto, 2000). For instance, as a resultof the search for distinctive group features (seealso Mlicki & Ellemers, 1996) when groups haveequal status, members of minority groups gen-erally identify more strongly with their groupthan members of majority groups (see Brewer,1991; Ellemers & Van Rijswijk, 1997; Simon &Brown, 1987). Furthermore, people are ready toidentify with groups that compare unfavorablyto other groups (e.g., low-status groups), to theextent that they believe in the potential of thegroup to improve its plight—that is, where inter-group differences are unstable (e.g., Doosje,Spears, & Ellemers, 2002; Ellemers, Van Dyck,Hinkle, & Jacobs, 2000)— or perceive theirgroup’s disadvantage as unjust—that is, whereintergroup differences are illegitimate (e.g., El-lemers, 2001a).

The cognitive tool of social categorization andthe evaluative implications of social compari-son processes can elicit a person’s emotionalinvolvement with a particular social group (Taj-fel, 1978; see also Ellemers, Kortekaas, & Ou-werkerk, 1999): their sense of social identifica-tion. In this way, the social identity approachaddresses the possibility that definitions of selfvary across different situations or over time, andit specifies the conditions under which a partic-ular self-definition or social identity is likely tobecome salient. That is, it enables us to predictin any given situation whether people are likelyto define themselves as individuals or as partsof a collective and to understand when particu-lar group memberships will tend to becomemore powerful determinants of behavior thanothers (Haslam, Postmes, & Ellemers, 2003).

While scholars have previously documentedthe general relevance of situational features tothe development of a collective identity, the ap-plication of this thinking to understand themechanisms involved in work organizations isrelatively novel. In fact, some of the conditions

social identity theory elaborates on refer to in-dividual instrumentality and interdependenceprinciples (resulting, for instance, in the greaterinclination to identify with groups that offer ac-cess to status, power, or other desirable out-comes) and, thus, converge with current insightsin organizational psychology.

Additionally, however, the social identity ap-proach enables us to formulate further predic-tions about work conditions that encourage feel-ings of identification, since this approach alsoinforms us about the circumstances under whichpeople tend to identify with the collective in theabsence of interdependence or individual in-strumentality considerations. Specifically, theassumption that people tend to focus on catego-ries that offer a distinct identity would implythat they are less likely to identify as groupsbecome larger and more inclusive. This reason-ing is consistent with observations (Ellemers,2003; Terry & Callan, 1998) that people tend toresist organizational changes (e.g., mergers ormoves toward privatization in the public sector)when they see these changes as underminingthe distinctiveness of their professional identi-ties, whereas the adoption of such changes isfacilitated when groups of workers can some-how maintain their distinct identity within thenew structure (e.g., Hornsey & Hogg, 1999; Jetten,O’Brien, & Trindall, 2002; Van Leeuwen, VanKnippenberg, & Ellemers, 2003).

Proposition 1: People will identifymore with a particular collective (e.g.,their work team) to the extent that itmeaningfully distinguishes them fromother relevant collectives (e.g., otherteams in the organization).

Proposition 2: In a given comparativecontext, people are more likely toidentify with more distinctive collec-tives (e.g., smaller units such as theirwork team) than with more inclusivecollectives (the larger organization inwhich they work).

Additionally, social identity theory and re-search indicate that the conviction the currentstanding of the group can be improved (becausethe outcomes of the group are unstable or ille-gitimate) fosters group identification, sincethese conditions help maintain people’s beliefsin the value of their group, even when the group


has little to offer to the individual at present. Ina similar vein, the threat of position loss (due toinstability or illegitimacy of current intergrouprelations) challenges people to affirm thevalue of their group and, hence, increasesidentification.

Proposition 3: In the absence of collec-tive success, individuals’ identifica-tion with the collective (e.g., the or-ganization) will be stronger to theextent that external circumstances(e.g., market developments) or collec-tive practices (e.g., human resourcemanagement) make it seem morelikely the collective will be successfulin the future.

Proposition 4: When a collective (e.g.,a work team or organization) is cur-rently successful, individuals’ identifi-cation with this collective will be en-hanced when external circumstancesthreaten this success.

SOCIAL IDENTIFICATION ANDWORK MOTIVATION

Having examined some of the situationalcharacteristics that may lead people to identifyin collective instead of individual terms, we nowconsider the implications of social identificationfor motivated behavior in work situations (seealso Haslam et al., 2000). A first assumption thatfollows from a social identity approach to moti-vation is that when people think of themselvesas part of a collective, they are energized bydifferent experiences or events than when theyidentify themselves as separate individuals (El-lemers et al., 2002). That is, we propose thatidentification as a member of a particular groupimplies people are activated by situations thatchallenge their inclusion in that group. In con-trast, when they are less inclined to identifywith a group, those same people are energizedto undertake action when they are being treatedas indistinct from other group members (see alsoBarreto & Ellemers, 2002). For example, newcom-ers who are proud of their membership in theorganization are prompted into action to under-line their collective identity when they are notrecognized as such (e.g., by their coworkers orclients). However, to the extent that they con-ceive of themselves in terms of specific personal

achievements or abilities, these same workersshould be inclined to enact their individualidentity when management proposes installingteam rewards instead of individual rewards.

As a result, the direction of the resulting effortis expected to differ, again depending on theextent to which the situation induces a defini-tion of the self either as separate from the groupor as part of a collective. That is, while scholarspredict group identification leads individuals todemonstrate loyalty to the group and inducesadherence to group norms, we argue that situa-tions in which individuals are led to conceive ofthemselves as separate from the group shouldmake them behave in ways that show how theydiffer from other group members. For example,when female workers aim for advancement intheir career, they tend to adopt a typically mas-culine behavioral style while continuing to em-phasize feminine traits of other women in theorganization (Ellemers, 2001a; Schmitt et al.,2003).

Finally, we expect that when circumstancesinduce individuals to identify with the group,they are more likely to sustain their efforts onbehalf of the group across changing circum-stances, whereas situational features that en-courage a conception of self in individual termsshould lead people to adapt their group-relatedefforts, depending on the extent to which theseseem to be individually rewarding. For instance,those who do not feel emotionally involved withthe group are only induced to direct their effortstoward the achievement of collective goals ifthey are likely to be personally sanctioned forfailing to do so (i.e., in public situations but notin private), while those who identify stronglywith the group consistently work for their group,regardless of whether their behavior is open toscrutiny from others (Barreto & Ellemers, 2001).This is in line with our previous contention thata self-definition in collective terms may helppeople internalize group goals as intrinsicallymotivating, whereas a self-definition as a sepa-rate individual implies that displays of group-oriented behavior depend on the presence orabsence of external pressure to do so (in thiscase, public accountability [see also Barreto &Ellemers, 2002, and Barreto & Ellemers, 2003]).

Proposition 5: When situational fea-tures induce workers to identify in col-lective terms, they will be energized


when their inclusion in the collectiveis not acknowledged, they will bringtheir behavior in line with what is dis-tinctive for the collective, and theywill sustain a concern with collectivegoals across different situations andover time.

Proposition 6: When the situationleads workers to disidentify with thecollective (because they identify ei-ther as individuals or with some othercollective), they will be energized toexpress this lack of identificationwhen they are treated as part of thecollective, they will direct their be-havior in ways that show how theydiffer from the collective, and theywill only sustain a concern with col-lective goals in situations where thisis individually rewarding (in the caseof individual identification) or whenthese converge with the goals of an-other collective (in situations wherethey prefer to identify with that othercollective).

IDENTIFICATION VERSUS COMMITMENT INORGANIZATIONS

The general idea that identification in collec-tive terms helps people orient their behavioralefforts toward collective goals seems consistentwith insights on organizational commitment(Mowday, Steers, & Porter, 1979) maintainingthat feelings of commitment can motivate indi-vidual workers to behave in accordance withorganizational goals. Indeed, high levels ofcommitment are accompanied by low levels ofindividual “withdrawal” behavior, as indicatedby empirical research on absenteeism, tardi-ness, and turnover (for a meta-analysis seeMathieu & Zajac, 1990). Commitment is also pre-dictive of a general willingness to engage indiscretionary work-related effort, such as organ-izational citizenship behavior (e.g., Meyer &Allen, 1997). Therefore, we now examine how theconcepts of identification and commitment re-late to each other, and in what way a conceptu-alization in terms of social identity might pro-vide additional insights into work motivationthat would be difficult to derive from currentknowledge about organizational commitment.

In order to interpret previous attempts to di-rectly compare the value of organizational iden-tity and organizational commitment as predic-tors of work-related behavior, we should notethat, in these studies, identification is conceptu-alized as the cognitive/perceptual awarenessthat the self constitutes part of the organization,while the term commitment is used to refer tothe affective ties between the individual and thegroup (e.g., Mael & Tetrick, 1992). However, so-cial identification (Tajfel, 1978; see also Hinkle,Taylor, Fox-Cardamone, & Crook, 1989), as wellas organizational commitment (Allen & Meyer,1996; Meyer & Allen, 1991), is commonly definedas a multidimensional construct, and both theseconstructs carry a reference to the cognitiveawareness of some interdependence, as well asincorporate a sense of emotional involvementwith the collective.

Indeed, studies that have taken these differentaspects of organizational commitment or socialidentification into account have yielded consis-tent results in the sense that, for both constructs,affective involvement of the individual with thegroup emerges as a relevant predictor of group-oriented efforts (Allen & Meyer, 1996; Burke &Reitzes, 1991; Ellemers et al., 1999; Ellemers,Spears, & Doosje, 1997). In contrast, perceivedinterdependence may tie individuals to the or-ganization (so that they are less likely to leave)but fails to induce optimal work behavior (as isthe case with continuance commitment [Meyer &Allen, 1997]). Thus, it seems that both work onorganizational commitment and research withinthe social identity tradition are consistent withthe general notion that, independent of moreindividually instrumental considerations re-flecting perceived interdependence between in-dividual outcomes and collective outcomes, theaffective sense of emotional involvement of theself with the group under consideration can mo-tivate individuals to direct their efforts towardgroup goals.

Nonetheless, the question remains whethertheorizing about social identity instead of organ-izational commitment yields novel insights. Wewould argue that the added value of thinkingabout organizational commitment as a form ofsocial identification with the work organizationis that it opens up the possibility of applyingadditional knowledge about conditions thatmay foster a concern with collective rather thanindividual conceptions of self. As a result, we


believe that the adoption of a social identityperspective may deepen our understanding ofthe psychological processes that may either ele-vate or depress resulting feelings of commit-ment, as well as help predict which alternativepossible source of collective self-definition islikely to emerge as primary in any givensituation.

Previous researchers of organizational com-mitment have acknowledged that a sense of in-volvement can be derived from different organ-izational constituencies or can have multiplefoci (Becker, 1992; Reichers, 1986), but a socialidentity analysis helps to specify when a partic-ular focus of commitment tends to become morerelevant than others, as well as what the likelymotivational consequences are of such feelingsof commitment. As we argued above, propo-nents of the social identity approach maintainthat people are likely to consider themselvesand others in terms of groups that help themdistinguish in meaningful ways between thosepresent in the situation at hand. For instance, inwork situations that imply interactions with rep-resentatives of other organizations or with ex-ternal customers, people should be inclined toperceive the organization as a salient entity,since this distinction provides them with a rele-vant behavioral guideline. However, when thesesame individuals interact with coworkers withinthe organization, a conception of the self andothers as organizational members is less infor-mative, since this is the identity they all share.Instead, they are more likely to focus on a cate-gorization that distinguishes between differentcoworkers, resulting, for instance, in the workteam becoming the relevant focus of commit-ment (Ellemers, de Gilder, & Van den Heuvel,1998; see also Van Knippenberg & Van Schie,2000).

Being able to understand and anticipate suchshifts in people’s use of different categories isrelevant to predicting the behavioral conse-quences of the resulting feelings of commitment.That is, although commitment to the organiza-tion as well as commitment to the work teammay motivate people to pursue collective goalsinstead of focus on their individual outcomes(Ellemers, de Gilder, & Van den Heuvel, 1998), itis important to note that team goals are notnecessarily aligned with broader organizationalgoals. For instance, when workers primarilyidentify as team members, they are less likely to

share information with other work teams, al-though the exchange of this information couldcontribute to the success of the organization as awhole (see also Haslam, 2001, and Postmes,2003).

Thus, the application of a social identity ap-proach enables us to consider identification as adynamic outcome of situational features, in-stead of as a property that emerges consistentlyin particular individuals or cultures (such asindividualism versus collectivism [Hofstede,1980; Triandis, 1995]). On the one hand, this im-plies that we should not view the tendency toidentify with a collective as a generic inclina-tion but, rather, as group specific. That is, wherepeople can be seen as belonging to multiplegroups, in any one situation they may opt todefine themselves in terms of particular catego-ries—for example, their work team—while theyare much less inclined to identify with others—for example, the organization as a whole (seealso Ellemers, de Gilder, & Van den Heuvel,1998). On the other hand, it means that, whenfocusing on a particular group, we should notregard the willingness or reluctance to identifywith that group as a stable predisposition of theindividual in question but, rather, as contextdependent. For instance, while a female doctormay try to avoid being seen as a member of hergender group when at work, she may be per-fectly happy to act as a representative of womenin a discussion on neighborhood provisions.

Indeed, the added value of the social identityapproach is that it helps us understand howissues of collective motivation apply in thesemore complex situations where multiple (andpossibly conflicting) group memberships oper-ate simultaneously. It does this by providing theconceptual tools to specify the psychologicalprocesses that operate in such situations, aswell as delineating the factors that determinethe relative salience of one identity over otheralternative identities in any given situation.

LEADERSHIP

In the previous sections we examined the sit-uational features that may contribute to a per-son’s self-concept being defined in individual orin collective terms, and we assessed the likelyconsequences of these different self-definitionsfor work-related behavior. We now illustratesome implications of this social identity ap-


proach to work motivation by examining howprocesses of leadership are affected by self-categorization and social identity. This is rele-vant to work motivation, since it helps us under-stand who is most likely to be accepted as amotivating force by others and under which con-ditions they will be most successful in mobiliz-ing their followers.

We propose that the potential of leaders ormanagers to communicate and create a sense ofshared identity is an important determinant ofthe likelihood that their attempts to energize,direct, and sustain particular work-related be-haviors in their followers will be successful (seealso Reicher & Hopkins, 1996). In doing this, wemove beyond approaches to leadership that at-tribute the emergence and effectiveness of lead-ers to specific behavioral styles (e.g., Kirkpatrick& Locke, 1991), to their talent to inspire others(e.g., charismatic or transformational leadership[Bass, 1985]), or to special interpersonal relation-ships they develop with their followers (Graen &Scandura, 1987). Instead, we focus on the waysin which leadership acceptance is contingent onsituational features (Hollander, 1964) that affectthe likelihood followers will either focus on theidentity they share with the leader or considerthe ways in which the leader is distinct fromthem.

For a range of practical reasons, leaders can-not always behave in ways that are individuallyrewarding for their followers, not least becausetheir role requires that they supervise and cor-rect the work carried out by those who fall undertheir responsibility. However, in view of the gen-eral (identity-enhancing) tendency to evaluatecharacteristics and behaviors of ingroup mem-bers more positively than those of outgroupmembers (e.g., Tajfel & Turner, 1979), followersshould generally be more inclined to hold posi-tive expectations about the underlying motivesand intentions of their leaders, to the extent thatthese are perceived as ingroup (rather than out-group) members (Duck & Fielding, 1999). Oneimportant consequence of this phenomenon isthat deviations from the expected pattern tendto be “explained away”—for instance, by attrib-uting negative leadership behaviors to externalpressures in the case of a leader who is consid-ered to be an ingroup member (Haslam, Mc-Garty, Brown, Eggins, Morrison, & Reynolds,2001). In contrast, people see events that confirmprevious expectancies (negative behavior in the

case of a leader who is regarded as an outgroupmember, or positive behavior in the case of aleader who is perceived as an ingroup member)as more diagnostic of the true nature and inten-tions of their leader (see also Hewstone, 1990).

Such ingroup-favoring biases in the interpre-tation of leadership behavior may constitute animportant mechanism that increases (when theleader is perceived as an ingroup member) orreduces (when the leader is perceived as anoutgroup member) the ability of leaders to ener-gize, direct, and sustain work-related effortsamong their followers. Indeed, empirical evi-dence supports our contention that identicalleadership behavior is interpreted differentlydepending on whether it is enacted by an in-group or outgroup member (Ellemers, VanRijswijk, Bruins, & de Gilder, 1998; see also Duck& Fielding, 1999). That is, as a result of attribu-tional differences, subordinates tend to remainloyal to an ingroup leader, despite displays ofundesirable leadership behavior (see also Bru-ins, Ellemers, & de Gilder, 1999). In contrast,subordinates’ willingness to cooperate with anoutgroup leader depends on whether this leaderhas treated them positively in the past (Ellem-ers, Van Rijswijk, Bruins, & de Gilder, 1998).

This implies that the extent to which followersperceive their leaders as sharing the same iden-tity has important consequences for the motivat-ing mechanisms that the leader can use effec-tively. That is, while the motivation to cooperatewith a leader who is seen as an outgroup mem-ber depends on how rewarding the exchangerelationship is for the subordinate, loyalty to aningroup leader emerges more unconditionally.This general idea is consistent with insightsthat, compared to transactional leadership,charismatic or transformational leadership (e.g.,Bass, 1985) is more broadly effective, as speci-fied, for instance, in leader-member exchange(LMX) theory (Graen & Scandura, 1987). However,these types of theories approach the issue ofleadership in individual terms, in the sense thatthey focus on individual properties (of leaders,of their followers, or both) that enhance the like-lihood leaders and followers will develop a spe-cial relationship with each other (Hogg & Mar-tin, 2003). In contrast, the application of insightsfrom the social identity approach allows us tosee leadership as a group phenomenon and toconsider the situational features that may en-


able leaders to draw on their followers’ sense ofshared identity (see also Haslam et al., 2001).

Proposition 7: To the extent that fol-lowers perceive their leader to share acommon identity with them, positiveleadership behavior is more likely tobe seen as indicative of the “true self”of the leader than negative leadershipbehavior, whereas the reverse patternof behavioral attributions will emergewhen the leader is perceived to be anoutgroup member.

We now apply the social identity perspectiveto help answer the question of whose guidelinesare most likely to be accepted as a motivationalforce by others. A crucial concern here is theextent to which a (prospective) leader is seen torepresent the group’s distinct identity—that is,the extent to which the leader is perceived asprototypical for the group. However, a socialidentity approach also implies that the per-ceived prototypicality of a leader is context de-pendent (Hogg, Hains, & Mason, 1998). In otherwords, depending on other groups present in thesituation and the ways in which the ingroup isdistinct from those particular outgroups, mem-bers may come to see different properties asprototypical for the group and, hence, desirablefor the group leader (Turner & Haslam, 2001). Thenovel contribution of this way of thinking is thatit enables us to predict how—in an intergroupcontext—the willingness to follow particularleaders and not others depends on whetherthose leaders represent the characteristics orfeatures that help distinguish the ingroup fromother groups.

One implication of this reasoning is that, insome cases, leadership acceptance depends oncharacteristics that would be quite difficult, ifnot impossible, to derive from a more traditionalanalysis, based on individual processes. For in-stance, in a recent series of experiments,Haslam, Turner, and Oakes (1999) and Turnerand Haslam (2001) demonstrated how the selec-tion of a group leader depended on the per-ceived characteristics of the outgroup againstwhich the group had to compete. When theleader of the other group excelled in terms ofintelligence, people tended to endorse an in-group leader who was unintelligent (but consid-erate). Presumably, this would help distinguishthe ingroup from the outgroup in a meaningful

way, in this particular context. In a similar vein,additional research demonstrated that a leaderwho favored ingroup members that opposed theoutgroup generally received more support andwas better able to mobilize individual effortsthan a leader who treated all ingroup membersequally (Haslam & Platow, 2001a,b; Platow,Hoar, Reid, Harley, & Morrison, 1997). Again, thisis consistent with the notion that it is not thedesirability of the leader’s behavior per se thatdetermines acceptance by his or her followersbut the extent to which the behavior of theleader represents the distinct meaning of theirshared identity compared to other groups in thatsituation.

The ability of the leaders to motivate theirfollowers, in this research, would be difficult toexplain from principles that assume the en-dorsement of leaders depends on their ability toshow superior individual qualities (e.g., intelli-gence or fairness). This is not to say that tradi-tional approaches to leadership do not provideuseful insights. However, whether it is useful tothink about leadership in terms of individualqualities or in terms of group processes dependson whether the situation induces people to iden-tify as separate individuals or as parts of acollective. Indeed, research has shown thatwhen people conceive of themselves and othersprimarily as independent individuals, it is pos-sible to identify specific characteristics or be-haviors that define an attractive leader, such asfairness (Platow & Van Knippenberg, 2001). How-ever, when participants conceive of the situationin intergroup terms, they are more inclined toendorse a leader who is prototypical for the in-group or favors the ingroup (Platow & Van Knip-penberg, 2001).

Proposition 8: When group membersdefine a situation in intergroup terms,they are most likely to endorse asleaders those who most clearly repre-sent ways in which the ingroup can bepositively distinguished from relevantcomparison groups.

It follows from the above arguments that lead-ers may engender greater loyalty and coopera-tiveness, to the extent that followers perceivethem as ingroup members, while a failure toestablish a sense of shared social identity willmean that leadership effectiveness depends onleaders’ being seen as instrumental for the


achievement of individual goals. Again, an im-portant contribution of the social identity ap-proach to this line of reasoning is that the per-ception of a leader as representative of thegroup may vary across situations or over time,depending on whether specific circumstances orevents enhance the salience of the identity thatthe leader shares with the group, or draw atten-tion to differences between the leader and therest of the group.

As an example of such processes, we predictthe presence of a salient outgroup or “commonenemy” (e.g., a competing organization) leadspeople to focus on their shared organizationalidentity (see also Rabbie & Bekkers, 1978). As aresult, management should be able to draw onthis sense of common identity as “entrepreneursof identity” (Reicher & Hopkins, 1996) when at-tempting to motivate workers to make personalsacrifices (e.g., work overtime, accept lowerraises) for the benefit of the organization as awhole. However, when these same organiza-tional members are induced (e.g., by rewardstructures) to categorize themselves and othersat a different level of inclusiveness (e.g., asmembers of competing work teams), similar at-tempts to induce organizational citizenship be-havior may be much less effective, since peopleare more likely to turn to the leader who embod-ies team goals that are not necessarily compat-ible with those of the broader organization.

Now that we have examined how circum-stances that lead people to think in terms ofparticular categorizations instead of others mayinfluence the effectiveness of leaders represent-ing these different categorizations, we turn tosome less than evident consequences of thisphenomenon. Again, this enables us to demon-strate how the application of insights about self-categorization and social identity leads us tomake predictions that differ from more con-ventional insights on leadership derived fromindividual-level analysis. Specifically, we drawattention to the problem that while pointing outthat a leader has exceptional skills may helplegitimize his or her position, the downside ofsuch a focus on distinctive individual qualitiesis that it may effectively set the leader apartfrom the rest of the group. An intriguing conse-quence is that while choosing a leader withsuperior individual abilities is often desirablefor other reasons, this may not always be thebest way to instill a sense of common identity in

the group. An experiment designed to test thevalidity of this reasoning demonstrated that aleader who was randomly selected from thegroup was more successful in motivating groupmembers to work together on a joint task than aleader who stood out from the group in terms ofindividual competence (Haslam, McGarty,Brown, Eggins, Morrison, & Reynolds, 1998).

A similar mechanism may come into playwhen reward structures clearly differentiate be-tween leaders and team members. Again, allo-cating more benefits to those in leadership po-sitions makes perfect sense as long as we thinkof this issue in individual terms. That is, in orderto retain equity and motivate leaders, leadersshould receive greater rewards, to the extentthat they are expected to carry more responsi-bilities, fulfill stricter requirements, or workharder than their subordinates. However, whensuch differences in rewards become too large, orwhen an appeal to workers to curb their re-quests for salary raises occurs while manage-ment receives huge bonuses, the adverse effectmay be that the feeling that leaders and follow-ers share a common identity is undermined (seeDrucker, 1986). Haslam, Brown, McGarty, andReynolds (1998) confirmed this possibility in re-search showing that while a differentiated re-ward structure may serve to motivate leaders,group members working under such a regimeactually report less enthusiasm and display lesseffort on a collective task than they do underconditions where leaders and followers receiveequal rewards.

Proposition 9: Circumstances that en-hance a sense of shared group identityfacilitate a leader’s attempts to moti-vate his or her followers, whereas fac-tors that set the leader apart from fol-lowers (in ways that do not enhancegroup identity or performance) canundermine leadership effectiveness.

In summary, the application of insights fromsocial identity theory to issues of leadership em-phasizes the point that the secret of successfulleadership lies in the capacity of the leader toinduce followers to perceive him or her as theembodiment of a positive social identity thatthey have in common and that distinguishesthem from others.


GROUP PERFORMANCE

During the past few decades the proportion ofpeople who work in groups or (self-managing)teams has steadily increased. This is not onlyconsistent with popular beliefs about the “syn-ergy” that may emerge when working in groupsbut also seems in line with scientific knowledgeon organizational psychology. Relevant consid-erations are that work teams may offer opportu-nities for job enrichment, may accommodate theneed for autonomy of workers, may decrease theworkload of supervisors, may increase perfor-mance on tasks that are too complicated for in-dividuals, and so on. However, given the varietyof reasons teams are used as an organizingprinciple, as well as differences in the circum-stances under which they work, clearly it is dif-ficult to draw conclusions about the general ef-fects of teamwork (see Buchanan, 2000). Apertinent question, thus, is whether we can pre-dict the conditions under which teams are likelyto perform successfully and understand whythis is the case.

Alongside the expected advantages of team-work, relevant insights also point to a possibledrawback of working in teams (Steiner, 1972).The so-called Ringelmann effect indicates thatpeople expend less effort when they perform acollective task than when they work on the sametask individually (Kravitz & Martin, 1986), pre-sumably because of motivation losses. This phe-nomenon, typically called “social loafing” (seeKarau & Williams, 1993), offers a rather pessi-mistic view of teamwork, since it suggests thatpeople are generally less willing to exert them-selves in group settings than when working in-dividually. Accordingly, typical solutions to thesocial loafing problem all revolve around rec-ommendations to make the work situation lesssocial by treating team members more as indi-viduals—for instance, by making the contribu-tions of individual team members identifiable orshowing how the contribution to group goalsmay help them obtain personally valued out-comes (Karau & Williams, 1995).

These solutions may be valid in some situa-tions, but they run counter to current develop-ments in the workplace, and, indeed, in manycases they are impossible to apply. A definingfeature of self-managing teams, for instance, isthat team members contribute to a common goaland motivate themselves and each other to do

so. An important question, therefore, is whetherthe expected benefits of installing workgroupsand teams are undermined by these traditionalmeasures to avoid social loafing. We proposethat a social identity analysis not only offersinsights that may help develop alternative waysto avoid social loafing that are more in keepingwith the essence of teamwork but also may con-tribute to our understanding of relevant factorsthat are likely to enhance (instead of impair) theperformance of the group compared to the per-formance of its individual members (sociallaboring).

A social identity analysis would predict thatworkers who identify with the group in questionshould be energized to act in terms of theirgroup membership, instead of in terms of whatseems individually rewarding. Accordingly, re-search has revealed that groups of close friendsor teammates display less social loafing thangroups composed of strangers or mere acquain-tances (Williams, Karau, & Bourgeois, 1993). Jehnand Shah (1997) have shown this effect is alsocaused by the higher levels of commitment ob-served in friendship groups. Thus, in addition totraditional remedies to social loafing that focuson workers as separate individuals, measuresthat enhance the salience of a collective identitycan also contribute to the motivation to achievecollective goals and, hence, to avoid motivationlosses in group performance situations (see alsoEllemers, 2001b, and Tyler & Blader, 2000).

However, work globalization and technologi-cal progress imply that the collaboration inwork teams often is virtual, with team interac-tions mainly occurring via the exchange of writ-ten information through computer networks. Ac-cording to traditional approaches to groupperformance, the comparative anonymity ofteam members that is likely to result wouldseem to encourage social loafing. At the sametime, collaboration in virtual teams is not par-ticularly conducive to the development of theinterpersonal interactions or friendships thathave been found to elicit feelings of commit-ment. Here again, it becomes apparent how asocial identity approach can extend insights de-rived from individual-level processes. That is,according to this perspective, identification as agroup member is not only derived from interper-sonal ties between group members but also isfacilitated by situational factors that enhancethe salience of the categorization. Consistent


with this reasoning, research suggests thatwhen members of multiple groups are present,computer-mediated communication can facili-tate (instead of hinder) a definition of the situa-tion in group terms (Lea, Spears, & Rogers, 2003;Postmes, Spears, & Lea, 1999). In other words,this form of interaction discourages team mem-bers from attending to individual differencesand helps them focus on the group membershipof the people they communicate with (see alsoWorchel, Rothgerber, Day, Hart, & Butemeyer,1998).

Proposition 10: Collectives that are notbound together by interpersonal ties(e.g., work teams) can nevertheless beenergized to work on joint tasks whenthe circumstances under which theywork enhance the salience of theircommon identity and prevent themfrom focusing on interpersonal dis-tinctions.

As we have argued in previous sections, theperception of a common identity and the result-ing feelings of identification with the workgroupconstitute an important factor that motivatesgroup members to work toward collective goals.However, whether the resulting behavior actu-ally enhances or diminishes the group’s perfor-mance further depends on the relevant compar-ative context and its implications for the group’sdistinct identity (see also Van Knippenberg &Ellemers, 2003). For instance, when examininggroup performance on an orange-picking task,Erev, Bornstein, and Galili (1993) observed thatthe introduction of an intergroup competition(encouraging a definition of the situation ingroup terms instead of individual terms) effec-tively ruled out the occurrence of social loafing.But the greatest effort on the group task wasobserved when the competing teams were sim-ilar to each other and could only establish adistinct identity by showing a superior teamperformance.

This example illustrates how the search for adistinct identity may direct group members’ ef-forts toward behaviors that set them apart fromother groups, but it is important to note that theadoption of behavioral norms that characterizethe group as distinct from other groups will notnecessarily result in greater group productivity.Indeed, systematic underperformance or exces-sive absence can also result when workers di-

rect their behavior toward specific group normsthat are perceived to be undesirable from amanagerial point of view—as in the case of so-called soldiering, where a group sets norms forunderperformance (Taylor, 1911; see also Gel-latly & Luchak, 1998).

A counterintuitive consequence of this pro-cess is that enhanced group identification caneven increase the amount of effort directed atthe achievement of individual goals when dis-tinctive group norms prescribe individualisticbehavior (Barreto & Ellemers, 2001)—for in-stance, when the organizational culture empha-sizes individual competitiveness. Conversely,when group members establish their collectiveidentity by setting distinct goals for the group,this not only fosters their sense of identificationwith the group but also increases their efforts toachieve these group goals (see Wegge &Haslam, 2003).

Proposition 11: The emergence of col-lective identification directs workers’efforts toward the enhancement oftheir joint performance when thishelps achieve or maintain a distinctcollective identity. However, collec-tive identification will diminish jointperformance when the distinctivenorm is for collective underperfor-mance.

When aiming to establish the circumstancesunder which group members will sustain theirefforts on behalf of the group, it is important toconsider whether the group members consider achange in their collective performance, whichcan imply either improvement or deterioration,to be a realistic prospect (Tajfel & Turner, 1979).In line with the social identity approach, thisdepends on the comparative context in whichgroup members find themselves. That is, whenthe group is currently performing worse thanrelevant comparison groups, the awareness thatother groups have achieved a higher perfor-mance level makes a performance improvementof the group seem feasible and helps groupmembers actually achieve a superior group per-formance (Ouwerkerk, de Gilder, & De Vries,2000; Ouwerkerk, Ellemers, & de Gilder, 1999).According to social identity theory, this is par-ticularly likely when differences in the relativestanding of the groups seem unstable (Doosje etal., 2002) or illegitimate (Ellemers, Wilke, & Van


Knippenberg, 1993). Likewise, whereas theknowledge that the group consistently outper-forms relevant other groups elicits satisfactionwith the group’s achievements, resulting in asense of complacency (Ouwerkerk & Ellemers,2002), the awareness that the group is losing itscompetitive edge (e.g., because other groups areimproving their relative performance) can helpsustain a high level of collective effort (Ou-werkerk et al., 1999).

Proposition 12: Individuals will sus-tain their efforts on behalf of a collec-tive either when they consider collec-tive performance improvement to be arealistic prospect or when they areconcerned with the possibility of col-lective position loss.

In sum, the application of a social identityperspective enables us to explain how groupperformance can be optimized when circum-stances prevent the application of more tradi-tional remedies to social loafing (e.g., becauseindividual contributions cannot be monitored, orwhen contributions to collective performanceclearly are not individually rewarding). At thesame time, we have established that providinggroups with a sense of collective identity is onlya first step toward achieving optimal group per-formance. That is, in addition to energizing in-dividuals to work toward collective goals, groupnorms should direct members’ efforts to achiev-ing superior performance. However, these ef-forts will only be sustained when collective per-formance improvement seems feasible, or underthe threat of collective position loss (see alsoEllemers, 2001b, and Van Knippenberg & Ellem-ers, 2003).

CONCLUSIONS AND DIRECTIONS FORFUTURE RESEARCH

In this article we have argued that our under-standing of motivated behavior in the work-place can be enriched by considering the possi-bility that the self can be defined in differentways. In many situations individual consider-ations may have a role to play (e.g., in theachievement of individual goals and rewards, orthe avoidance of sanctions). However, in manyother situations people may be motivated to be-have in ways that express or support a socialidentity that is shared with others in the work

situation (see also Tyler, 2002). The social iden-tity approach provides an interesting perspec-tive on these alternative sources of motivation,implying that neither of these mechanismsshould be seen as more important or more validthan the other but specifying the conditions un-der which each is likely to operate (see alsoBarreto & Ellemers, 2001, and Haslam, 2001).

When the situation fosters a definition of theself in individual terms, individually instrumen-tal considerations are crucial determinants ofwork motivation. If, however, the situation in-duces workers to identify as parts of a collective,they are more likely to be concerned with theenhancement of that collective identity—for in-stance, by pursuing shared goals or behaving inways that are normative for that identity. Thus,there is no a priori reason to privilege one formof identification over another, or to see one formof self-definition as deriving straightforwardlyfrom the other, since the same individuals mayperceive themselves in other terms and behavedifferently from one situation to the next.

Indeed, whether a particular identity is rele-vant to understand (or to change) work motiva-tion depends on the focus of the motivation oneaims to address and on the forms of behaviorone wishes to predict (or induce). When the aimis to examine why a broad range of people areleaving an organization, or what is motivatingthem to stay, the organization represents an ap-propriate level of inclusiveness at which identi-fication (or lack of it) should be assessed. How-ever, when the intention is to promote people’sefforts directed toward a particular team perfor-mance, one should focus on the extent to whichthey identify with that team, instead of with theorganization as a whole. At the same time, thecommon practice of treating workers as sepa-rate individuals, in the hope that their efforts tofulfill their individual ambitions will benefit theorganization as a whole, seems less worthwhilefrom this point of view.

In the domain of leadership, we have arguedthat the perception of a common identity withthe leader is crucial for the leader’s effective-ness in mobilizing individual efforts toward col-lective goals. In line with this reasoning, wehave reviewed research findings that would bedifficult to predict on the basis of traditionalthinking about leadership. We have consideredhow contextual characteristics may inducegroup members to accept or even expect traits or


behaviors that seem prototypical for the group,even when these are not considered attractivefor leaders in a more individualized context. Ad-ditionally, we have illustrated how some com-mon practices in organizations (e.g., emphasiz-ing superior qualities of leaders, or introducingsubstantial reward advantages for managers)may, in fact, interfere with leadership effective-ness, to the extent that they foster a conceptionof the leader as someone who stands outside thegroup rather than being part of it.

Turning to group performance, we have ar-gued that enhancement of collective identityconstitutes a significant way of avoiding socialloafing and optimizing collective performance.It is important to note that this principle is notnecessarily incompatible with contemporary de-mands for workers to collaborate in virtual en-vironments (e.g., ones that are based on com-puter-mediated communications). At the sametime, we have argued that enhancing a sense ofcollective identity only constitutes a first steptoward the achievement of an optimal groupperformance, for even when people identify asgroup members and are motivated to exert efforton behalf of the group, the way their behaviorwill actually be directed depends on specificfeatures of the social context. This is for tworeasons: (1) because contextual features deter-mine the nature of salient group goals and dis-tinctive group norms, and these can also elicitless desirable outcomes (e.g., when they lead tosoldiering), and (2) because social contextualfeatures determine whether or not collectiveperformance improvement seems feasible oreven desirable.

While we have tried to demonstrate some pos-sible consequences of social identity processesthat seem difficult to predict or understand froma more individualistic perspective on work mo-tivation, this is not intended to imply that indi-vidual needs, goals, outcome comparisons, orreinforcement mechanisms are unimportant.However, we do wish to emphasize that thesesame principles of motivation may have funda-mentally different implications when applied ata collective level. Furthermore, we think of thetendency to define the self primarily in individ-ual or in collective terms (or more in terms of oneparticular group than as a member of anothergroup) as an adaptive response to the situationat hand, instead of as a stable property that isdetermined by the individual’s disposition or bycultural norms. For this reason, it becomes im-portant to establish whether evaluation and re-ward structures reinforce a self-definition in in-dividual or collective terms, whether relevantwork goals apply to individuals or to groups,and whether equity considerations derive frominterpersonal comparisons or from intergroupcomparisons.

Indeed, we have developed some concretepropositions derived from the general predictionthat individual needs, goals, or comparisons arethe primary source of motivation in work situa-tions that foster a conception of the self as anindependent individual, while collective needs,goals, and comparisons are likely to predomi-nate in situations that facilitate a definition ofthe self in collective terms. In Table 1 we havesummarized how these propositions address (1)issues of self-definition (Propositions 1–4), (2) sit-uational influences (Propositions 5 and 6), (3)

TABLE 1Collective Identification and Work Motivation: Overview of Propositions

Self-Definition(Propositions 1–4)

Situational Influences(Propositions 5 and 6)

Acceptance of Leadership(Propositions 7–9)

Performance Consequences(Propositions 10–12)

When do people define theself as part of a particularcollective?

How does the situation inducea particular motivationalfocus?

Who can mobilize themotivation to directindividual efforts towardcollective goals?

How does collective motivationimpact group performance?

● Comparative distinctiveness● Relative inclusiveness● Current success and future

expectations

● Compatibility of internaland external definitions ofself

● Expression of individual vs.collective identity

● Social identity–basedexpectations

● Embodiment of groupdistinctiveness

● Shared social identity

● Salience of shared socialidentity

● Distinctive group norms● Future prospects


acceptance of leaders (Propositions 7–9), and (4)consequences for work performance (Proposi-tions 10–12).

At the practical level, our approach thus im-plies that there is no one best way to motivatepeople at work and that there is no quick andeasy solution for problems of motivation. In-stead, it is crucial to first establish who shouldbe motivated to work toward which goal, be-fore the work situation can be geared towardaddressing the definition of self that is rele-vant to that goal. Such measures should not berestricted to formal features of the work situa-tion (such as the nature of the reward struc-ture) but should also encompass more infor-mal aspects of the organizational culture, aswell as the culture’s enactment by manage-ment. Indeed, the effectiveness of specificmeasures intended to motivate people to en-gage with collective goals is likely to be un-dermined when the broader organizationalstructure and culture continue to foster a con-sideration of the self as a separate individual.We think that, from a managerial point ofview, this is, in fact, an interesting notion, inthe sense that it offers scope to encourage theoperation of either individual or collective mo-tives by adapting salient features of the worksituation, such as the reward system or pro-motion opportunities. Conversely, to the extentthat organizational practices resist such change(e.g., because they are legally anchored), manage-ment should be aware that this is likely to limit theeffectiveness of its attempts to influence the focusof workers’ motivation and effort.

We have tried to show that the social identityapproach may provide a useful analyticalframework for understanding motivational pro-cesses of individuals and groups at work. How-ever, the research we have reviewed in supportof our argument consists largely of experimen-tal work focusing on these psychological pro-cesses as they operate in relatively contrivedsituations. Thus, while we think this provides asolid basis for our reasoning and results in pre-dictions that are highly relevant to issues ofwork motivation, we believe further research isclearly necessary to examine the concrete impli-cations of our arguments in particular work set-tings. We hope the present contribution mayinspire researchers in the field to address themechanisms and variables we have describedin their future studies.

REFERENCES

Allen, N. J., & Meyer, J. P. 1996. Affective, continuance, andnormative commitment to the organization: An exami-nation of construct validity. Journal of Vocational Behav-ior, 49: 252–276.

Ambrose, M. L., & Kulik, C. T. 1999. Old friends, new faces:Motivation research in the 1990s. Journal of Manage-ment, 25: 231–292.

Ashforth, B. E., & Kreiner, G. 1999. “How can you do it?”: Dirtywork and the dilemma of identity. Academy of Manage-ment Review, 24: 413–434.

Ashforth, B. E., & Mael, F. 1989. Social identity theory and theorganization. Academy of Management Review, 14: 20–39.

Barreto, M., & Ellemers, N. 2001. You can’t always do whatyou want: Social identity and self-presentational deter-minants of the choice to work for a low status group.Personality and Social Psychology Bulletin, 26: 891–906.

Barreto, M., & Ellemers, N. 2002. The impact of respect vs.neglect of self-identities on group identification andgroup loyalty. Personality and Social Psychology Bulle-tin, 28: 629–639.

Barreto, M., & Ellemers, N. 2003. The effects of being cat-egorised: The interplay between internal and externalsocial identities. European Review of Social Psychology,14: 139–170.

Bass, B. M. 1985. Leadership and performance beyond expec-tations. New York: Free Press.

Becker, T. E. 1992. Foci and bases of commitment: Are theydistinctions worth making? Academy of ManagementJournal, 35: 232–244.

Brewer, M. B. 1991. The social self: On being the same anddifferent at the same time. Personality and Social Psy-chology Bulletin, 17: 472–482.

Brief, A. P., & Aldag, R. J. 1981. The “self” in work organiza-tions: A conceptual review. Academy of ManagementReview, 6: 75–88.

Brief, A. P., & Motowidlo, S. J. 1986. Prosocial organizationalbehaviors. Academy of Management Review, 11: 710–725.

Bruins, J. J., Ellemers, N., & de Gilder, D. 1999. Power use andstatus differences as determinants of subordinates’evaluative and behavioural responses in simulated or-ganizations. European Journal of Social Psychology, 29:843–870.

Buchanan, D. A. 2000. An eager and enduring embrace: Theongoing rediscovery of teamworking as a managementidea. In S. Procter & F. Mueller (Eds.), Teamworking:25–43. London: Macmillan.

Burke, P. J., & Reitzes, D. C. 1991. An identity approach tocommitment. Social Psychology Quarterly, 54: 239–251.

Cascio, W. F. 1995. Whither industrial and organizationalpsychology in a changing world of work? American Psy-chologist, 50: 928–939.

Coates, G. 1994. Motivation theories’ lacklustre performance:Identity at work as explanation for expended effort. Ed-


ucational and Training Technology International, 31:26–30.

Doosje, B., Spears, R., & Ellemers, N. 2002. Social identity asboth cause and effect: The development of group iden-tification in response to anticipated and actual changesin the intergroup status hierarchy. British Journal of So-cial Psychology, 41: 57–76.

Duck, J. M., & Fielding, K. S. 1999. Leaders and subgroups:One of us or one of them? Group Processes and Inter-group Relations, 2: 203–230.

Drucker, P. F. 1986. The frontiers of management: Wheretomorrow’s decisions are being shaped today. New York:Dutton.

Ellemers, N. 1993. The influence of socio-structural variableson identity enhancement strategies. European Reviewof Social Psychology, 4: 27–57.

Ellemers, N. 2001a. Individual upward mobility and the per-ceived legitimacy of intergroup relations. In J. T. Jost &B. Major (Eds.), The psychology of legitimacy: 205–222.Cambridge: Cambridge University Press.

Ellemers, N. 2001b. Social identity, commitment, and workbehavior. In M. A. Hogg & D. J. Terry (Eds.), Social iden-tity processes in organizational contexts: 101–114. Lon-don: Psychology Press.

Ellemers, N. 2003. Identity, culture, and change in organiza-tions: A social identity analysis and three illustrativecases. In S. A. Haslam, D. Van Knippenberg, M. Platow,& N. Ellemers (Eds.), Social identity at work: Developingtheory for organizational practice: 191–204. New York:Psychology Press.

Ellemers, N., & Barreto, M. 2000. The impact of relative groupstatus: Affective, perceptual and behavioural conse-quences. In R. Brown & S. Gaertner (Eds.), The Blackwellhandbook of social psychology. Volume 4: Intergroupprocesses: 324–343. Oxford: Blackwell.

Ellemers, N., de Gilder, D., & Van den Heuvel, H. 1998.Career-oriented versus team-oriented commitment andbehavior at work. Journal of Applied Psychology, 83:717–730.

Ellemers, N., Haslam, S. A., Platow, M. J., & Van Knippenberg, D.2003. Social identity at work: Developments, debates,directions. In S. A. Haslam, D. Van Knippenberg, M. J.Platow, & N. Ellemers (Eds.), Social identity at work:Developing theory for organizational practice: 3–28. NewYork: Psychology Press.

Ellemers, N., Kortekaas, P., & Ouwerkerk, J. W. 1999. Self-categorization, commitment to the group and social self-esteem as related but distinct aspects of social identity.European Journal of Social Psychology, 29: 371–389.

Ellemers, N., Spears, R., & Doosje, B. 1997. Sticking togetheror falling apart: Ingroup identification as a psychologi-cal determinant of group commitment versus individualmobility. Journal of Personality and Social Psychology,72: 617–626.

Ellemers, N., Spears, R., & Doosje, B. 2002. Self and socialidentity. Annual Review of Psychology, 53: 161–186.

Ellemers, N., Van Dyck, C., Hinkle, S., & Jacobs, A. 2000.Intergroup differentiation in social context: Identity

needs versus audience constraints. Social PsychologyQuarterly, 63: 60–74.

Ellemers, N., & Van Rijswijk, W. 1997. Identity needs versussocial opportunities: The use of group-level and individ-ual-level identity management strategies. Social Psy-chology Quarterly, 60: 52–65.

Ellemers, N., Van Rijswijk, W., Bruins, J., & de Gilder, D. 1998.Group commitment as a moderator of attributional andbehavioural responses to power use. European Journalof Social Psychology, 28: 555–573.

Ellemers, N., Wilke, H., & Van Knippenberg, A. 1993. Effects ofthe legitimacy of low group or individual status on in-dividual and collective identity enhancement strate-gies. Journal of Personality and Social Psychology, 64:766–778.

Erev, I., Bornstein, G., & Galili, R. 1993. Constructive inter-group competition as a solution to the free rider prob-lem: A field experiment. Journal of Experimental SocialPsychology, 29: 463–478.

Erez, M., Kleinbeck, U., & Thierry, H. (Eds.). 2001. Work moti-vation in the context of a globalizing economy. Mahwah,NJ: Lawrence Erlbaum Associates.

Gellatly, I. R., & Luchak, A. A. 1998. Personal and organiza-tional determinants of perceived absence norms. Hu-man Relations, 51: 1085–1102.

Graen, G. B., & Scandura, T. A. 1987. Toward a psychology ofdyadic organizing. Research in Organizational Behav-ior, 9: 175–208.

Gutek, B. A. 1995. The dynamic of service: Reflections on thechanging nature of the service industry. San Francisco:Jossey-Bass.

Hackman, J. R., & Oldham, G. R. 1976. Motivation through thedesign of work: Test of a theory. Organizational Behav-ior and Human Performance, 16: 250–279.

Haslam, S. A. 2001. Psychology in organizations: The socialidentity approach. London: Sage.

Haslam, S. A., Brown, P., McGarty, C., & Reynolds, K. J. 1998.The impact of differential reward on the motivation ofleaders and followers. Working paper, Australian Na-tional University, Canberra.

Haslam, S. A., & Ellemers, N. In press. Social identity inindustrial and organizational psychology: Concepts,controversies, and contributions. International Reviewof Industrial and Organizational Psychology.

Haslam, S. A., McGarty, C., Brown, P. M., Eggins, R. A., Mor-rison, B. E., & Reynolds, K. J. 1998. Inspecting the emper-or’s clothes: Evidence that randomly-selected leaderscan enhance group performance. Group Dynamics: The-ory, Research, and Practice, 2: 168–184.

Haslam, S. A., & Platow, M. J. 2001a. The link between lead-ership and followership: How affirming social identitytranslates vision into action. Personality and Social Psy-chology Bulletin, 27: 1469–1479.

Haslam, S. A., & Platow, M. J. 2001b. Your wish is my com-mand: How a leader’s vision becomes a follower’s task.In M. A. Hogg & D. J. Terry (Eds.), Social identity pro-


cesses in organizational contexts: 213–228. New York:Taylor & Francis.

Haslam, S. A., Platow, M. J., Turner, J. C., Reynolds, K. J.,McGarty, C., Oakes, P. J., Johnson, S., Ryan, M. K., &Veenstra, K. 2001. Social identity and the romance ofleadership: The importance of being seen to be “doing itfor us.” Group Processes and Intergroup Relations, 4:191–205.

Haslam, S. A., Postmes, T., & Ellemers, N. 2003. More than ametaphor: Organizational identity as a sine qua non oforganizational life. British Journal of Management, 14:357–369.

Haslam, S. A., Powell, C., & Turner, J. C. 2000. Social identity,self-categorization, and work motivation: Rethinking thecontribution of the group to positive and sustainableorganisational outcomes. Applied Psychology: An Inter-national Review, 49: 319–339.

Haslam, S. A., & Turner, J. C. 1992. Context-dependent vari-ation in social stereotyping 2: The relationship betweenframe of reference, self-categorization and accentua-tion. European Journal of Social Psychology, 22: 251–277.

Haslam, S. A., Turner, J. C., & Oakes, P. J. 1999. Contextualvariations in leader prototypes. Working paper, Austra-lian National University, Canberra.

Haslam, S. A., Van Knippenberg, D., Platow, M., & Ellemers, N.(Eds.). 2003. Social identity at work: Developing theoryfor organizational practice. New York: Psychology Press.

Hewstone, M. 1990. The “ultimate attribution error”? A reviewof the literature on intergroup causal attribution. Euro-pean Journal of Social Psychology, 20: 311–336.

Hinkle, S., Taylor, L. A., Fox-Cardamone, D. L., & Crook, K. F.1989. Intragroup identification and intergroup differen-tiation: A multi-component approach. British Journal ofSocial Psychology, 28: 305–317.

Hofstede, G. 1980. Culture’s consequences: International dif-ferences in work-related values. Beverly Hills, CA: Sage.

Hogg, M. A., Hains, S. C., & Mason, I. 1998. Identification andleadership in small groups: Salience, frame of reference,and leader stereotypicality effects on leader evalua-tions. Journal of Personality and Social Psychology, 75:1248–1263.

Hogg, M. A., & Martin, R. 2003. Social identity analysis ofleader-member relations: Reconciling self-categoriza-tion and leader-member exchange theories of leader-ship. In S. A. Haslam, D. Van Knippenberg, M. J. Platow,& N. Ellemers (Eds.), Social identity at work: Developingtheory for organizational practice: 139–154. New York:Psychology Press.

Hogg, M. A., & Terry, D. J. 2000. Social identity and self-categorization processes in organizational contexts.Academy of Management Review, 25: 121–140.

Hollander, E. P. 1964. Leaders, groups, and influence. NewYork: Oxford University Press.

Hornsey, M. J., & Hogg, M. A. 1999. Subgroup differentiationas a response to an overly-inclusive group: A test ofoptimal distinctiveness theory. European Journal of So-cial Psychology, 29: 543–550.

Ilgen, D. R., & Sheppard, L. 2001. Motivation in work teams. InM. Erez, U. Kleinbeck, & H. Thierry (Eds.), Work motiva-tion in the context of a globalizing economy: 169–179.Mahwah, NJ: Lawrence Erlbaum Associates.

Jehn, K. A., & Shah, P. P. 1997. Interpersonal relationshipsand task performance: An examination of mediatingprocesses in friendship and acquaintance groups. Jour-nal of Personality and Social Psychology, 72: 775–790.

Jetten, J., O’Brien, A., & Trindall, N. 2002. Changing identity:Predicting adjustment to organizational restructure as afunction of subgroup and superordinate identification.British Journal of Social Psychology, 41: 281–297.

Karau, S. J., & Williams, K. D. 1993. Social loafing: A meta-analytical review and theoretical integration. Journal ofPersonality and Social Psychology, 65: 681–706.

Karau, S. J., & Williams, K. D. 1995. Social loafing: Researchfindings, implications, and future directions. Current Di-rections in Psychological Science, 4: 134–140.

Kirkpatrick, S. A., & Locke, E. A. 1991. Leadership: Do traitsmatter? Academy of Management Executive, 5(2): 48–60.

Kohn, A. 1993. Why incentive plans cannot work. HarvardBusiness Review, 71(5): 54–63.

Komaki, J., Coombs, T., & Schepman, S. 1996. Motivationalimplications of reinforcement theory. In R. M. Steers,L. W. Porter, & G. A. Bigley (Eds.), Motivation and lead-ership at work (6th ed.): 34–52. New York: McGraw-Hill.

Kravitz, D., & Martin, B. 1986. Ringelmann revisited: Alterna-tive explanations for the social loafing effect. Personal-ity and Social Psychology Bulletin, 50: 936–941.

Lea, M., Spears, R., & Rogers, P. 2003. Social processes inelectronic team work: The central issue of identity. InS. A. Haslam, D. Van Knippenberg, M. Platow, & N. Elle-mers (Eds.), Social identity at work: Developing theoryfor organizational practice: 99–116. New York: Psychol-ogy Press.

Lembke, S., & Wilson, M. G. 1998. Putting the “team” intoteamwork: Alternative theoretical contributions for con-temporary management practice. Human Relations, 51:927–944.

Locke, E. A., & Latham, G. P. 1990. A theory of goal setting andtask performance. Englewood Cliffs, NJ: Prentice-Hall.

Mael, F., & Tetrick, L. E. 1992. Identifying organizationalidentification. Educational and Psychological Measure-ment, 52: 813–824.

Maslow, A. H. 1943. A theory of motivation. PsychologicalReview, 50: 370–396.

Mathieu, J. E., & Zajac, D. 1990. A review and meta-analysisof the antecedents, correlates, and consequences of or-ganizational commitment. Psychological Bulletin, 108:171–194.

Meyer, J. P., & Allen, N. J. 1991. A three-component conceptu-alization of organizational commitment. Human Re-source Management Review, 1: 61–89.

Meyer, J. P., & Allen, N. J. 1997. Commitment in the workplace:Theory, research, and application. London: Sage.

Mitchell, T. R. 1982. Motivation: New directions for theory,


research, and practice. Academy of Management Re-view, 7: 80–88.

Mlicki, P., & Ellemers, N. 1996. Being different or being bet-ter? National stereotypes and identifications of Polishand Dutch students. European Journal of Social Psychol-ogy, 26: 97–114.

Mowday, R. T. 1979. Equity theory predictions of behavior inorganizations. In R. M. Steers & L. W. Porter (Eds.), Mo-tivation and work behavior: 124 –146. New York:McGraw-Hill.

Mowday, R. T., Steers, R., & Porter, L. W. 1979. The measure-ment of organizational commitment. Journal of Voca-tional Behavior, 14: 224–247.

Oakes, P. J. 1987. The salience of social categories. In J. C.Turner, M. A. Hogg, P. J. Oakes, S. D. Reicher, & M. S.Whetherell (Eds.), Rediscovering the social group: A self-categorization theory: 117–141. Oxford: Blackwell.

Oakes, P. J., Haslam, S. A., & Turner, J. C. 1994. Stereotypingand social reality. Oxford: Blackwell.

Organ, D. W. 1988. Organizational citizenship behavior: Thegood soldier syndrome. Lexington, MA: Lexington Books.

Ouwerkerk, J. W., de Gilder, D., & De Vries, N. K. 2000. Whenthe going gets tough, the tough get going: Social iden-tification and individual effort in intergroup competi-tion. Personality and Social Psychology Bulletin, 26:1550–1559.

Ouwerkerk, J. W., & Ellemers, N. 2002. The benefits of beingdisadvantaged: Performance-related circumstances andconsequences of intergroup comparisons. EuropeanJournal of Social Psychology, 32: 73–91.

Ouwerkerk, J. W., Ellemers, N., & de Gilder, D. 1999. Groupcommitment and individual effort in experimental andorganizational contexts. In N. Ellemers, R. Spears, &B. Doosje (Eds.), Social identity: Context, commitment,content: 184–204. Oxford: Blackwell.

Parker, M. 1993. Industrial relations myth and shop floorreality: The team concept in the auto industry. InN. Lichtenstein & J. H. Howell (Eds.), Industrial democ-racy in America: 249–274. Cambridge: Cambridge Uni-versity Press.

Pearce, J. L. 1987. Why merit pay doesn’t work: Implicationsfrom organizational theory. In D. B. Balkin & L. R. Gomez-Meija (Eds.), New perspectives on compensation: 169–178. Englewood Cliffs, NJ: Prentice-Hall.

Pinder, C. C. 1998. Work motivation in organizational behav-ior. Upper Saddle River, NJ: Prentice-Hall.

Platow, M. J., Hoar, S., Reid, S., Harley, K., & Morrison, D. 1997.Endorsement of distributively fair or unfair leaders ininterpersonal and intergroup situations. European Jour-nal of Social Psychology, 27: 465–494.

Platow, M. J., & Van Knippenberg, D. 2001. A social identityanalysis of leadership endorsement: The effects ofleader ingroup prototypicality and distributive inter-group fairness. Personality and Social Psychology Bul-letin, 27: 1508–1519.

Postmes, T. 2003. A social identity approach to communica-tion in organizations. In S. A. Haslam, D. Van Knippen-

berg, M. J. Platow, & N. Ellemers (Eds.), Social identity atwork: Developing theory for organizational practice: 81–97. Philadelphia: Taylor & Francis.

Postmes, T., Spears, R., & Lea, M. 1999. Social identity: Nor-mative content, and “deindividuation.” In N. Ellemers,R. Spears, & B. Doosje (Eds.), Social identity: Context,commitment, content: 164–183. Oxford: Blackwell.

Rabbie, J. M., & Beckers, F. 1978. Threatened leadership andintergroup competition. European Journal of Social Psy-chology, 8: 9–20.

Reicher, S. D., & Hopkins, N. 1996. Self-category constructionsin political rhetoric: An analysis of Thatcher’s and Kin-nock’s speeches concerning the British Miners’ Strike(1984–5). European Journal of Social Psychology, 26: 353–372.

Reichers, A. E. 1986. Conflict and organizational commit-ments. Journal of Applied Psychology, 71: 508–514.

Shaubroeck, J., & Ganster, D. C. 1991. Beyond the call of duty:A field study of extra-role behavior in voluntary organi-zations. Human Relations, 44: 569–582.

Schmitt, M. T., Ellemers, N., & Branscombe, N. R. 2003. Per-ceiving and responding to gender discrimination atwork. In S. A. Haslam, D. Van Knippenberg, M. J. Platow,& N. Ellemers (Eds.), Social identity at work: Developingtheory for organizational practice: 277–292. New York:Psychology Press.

Shamir, B. 1991. Meaning, self and motivation in organiza-tions. Organization Studies, 12: 405–424.

Simon, B., & Brown, R. J. 1987. Perceived intragroup homoge-neity in minority-majority contexts. Journal of Personal-ity and Social Psychology, 53: 703–711.

Smith, V. 1997. New forms of work organization. Annual Re-view of Sociology, 23: 315–339.

Spears, R., Doosje, B., & Ellemers, N. 1997. Self-stereotypingin the face of threats to group status and distinctiveness:The role of group identification. Personality and SocialPsychology Bulletin, 23: 538–553.

Spears, R., Oakes, P. J., Ellemers, N., & Haslam, S. A. (Eds.).1997. The social psychology of stereotyping and grouplife. Oxford: Blackwell.

Steers, R. M., Porter, L. W., & Bigley, G. A. (Eds.). 1996. Moti-vation and leadership at work (6th ed.). New York:McGraw-Hill.

Steiner, I. D. 1972. Group process and productivity. New York:Academic Press.

Sussmann, M., & Vecchio, R. P. 1982. A social influence in-terpretation of worker motivation. Academy of Manage-ment Review, 7: 177–186.

Tajfel, H. 1974. Social identity and intergroup behavior. So-cial Science Information, 14: 65–93.

Tajfel, H. 1975. The exit of social mobility and the voice ofsocial change. Social Science Information, 15: 101–118.

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

Tajfel, H., & Turner, J. C. 1979. An integrative theory of inter-


group conflict. In W. G. Austin & S. Worchel (Eds.), Thesocial psychology of intergroup relations: 7–24.Monterey, CA: Brooks/Cole.

Taylor, F. W. 1911. Principles of scientific management. NewYork: Harper.

Terry, D. J., & Callan, V. J. 1998. Ingroup bias in response toan organizational merger. Group Dynamics: Theory, Re-search, and Practice, 2: 67–81.

Triandis, H. C. 1995. Individualism and collectivism. Boulder,CO: Westview.

Turner, J. C., & Haslam, S. A. 2001. Social identity, organiza-tions, and leadership. In M. E. Turner (Ed.), Groups atwork: Advances in theory and research: 25–65. Hillsdale,NJ: Lawrence Erlbaum Associates.

Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., &Whetherell, M. S. 1987. Rediscovering the social group: Aself-categorization theory. Oxford: Blackwell.

Tyler, T. R. 2002. Leadership and cooperation in groups.American Behavioral Scientist, 45: 769–782.

Tyler, T. R., & Blader, S. 2000. Cooperation in groups. London:Psychology Press.

Van Dick, R. 2001. Identification in organizational contexts:Linking theory and research from social and organiza-tional psychology. International Journal of ManagementReview, 4: 265–284.

Van Knippenberg, D., & Ellemers, N. 2003. Social identity andgroup performance: Identification as the key to collec-tive effort. In S. A. Haslam, D. Van Knippenberg, M. Pla-tow, & N. Ellemers (Eds.), Social identity at work: Devel-

oping theory for organizational practice: 29–42. NewYork: Psychology Press.

Van Knippenberg, D., & Van Schie, E. C. M. 2000. Foci andcorrelates of organizational identification. Journal of Oc-cupational and Organizational Psychology, 73: 137–147.

Van Leeuwen, E., Van Knippenberg, D., & Ellemers, N. 2003.Continuing and changing group identities: The effectsof merging on social identification and ingroup bias.Personality and Social Psychology Bulletin, 29: 679–690.

Van Rijswijk, W., & Ellemers, N. 2002. Context effects on theapplication of stereotype content to multiple categoriz-able targets. Personality and Social Psychology Bulletin,28: 90–101.

Vroom, V. H. 1964. Work and motivation. New York: Wiley.

Wegge, J., & Haslam, S. A. 2003. Group goal setting, socialidentity and self-categorization: Engaging the collectiveself to enhance group performance and organizationaloutcomes. In S. A. Haslam, D. Van Knippenberg, M. Pla-tow, & N. Ellemers (Eds.), Social identity at work: Devel-oping theory for organizational practice: 43–60. NewYork: Psychology Press.

Williams, K. D., Karau, S. J., & Bourgeois, M. 1993. Working oncollective tasks: Social loafing and social compensa-tion. In M. A. Hogg & D. Abrams (Eds.), Group motivation:Social psychological perspectives: 130–148. New York:Harvester Wheatsheaf.

Worchel, S., Rothgerber, H., Day, E. A., Hart, D., & Butemeyer, J.1998. Social identity and individual productivity withingroups. British Journal of Social Psychology, 37: 389–413.

Naomi Ellemers currently is professor of social and organizational psychology atLeiden University. She studied at the University of California at Berkeley and at theUniversity of Groningen, the Netherlands, where she received her Ph.D. Her researchfocuses on the role of social identity in group interactions and intergroup relations.

Dick de Gilder currently is assistant professor of public administration and organi-zation at the Free University in Amsterdam. He studied at the University of Amsterdamand received his Ph.D. from the University of Groningen, the Netherlands. His researchinterests include organizational commitment, error management, and research man-agement.

S. Alexander Haslam currently is professor of social and organizational psychology atthe University of Exeter, where he investigates a range of social and organizationaltopics. He studied at Macquarie University, Emory University, and the University of St.Andrews, where he received his Ph.D. He recently published a book on the socialidentity approach to psychology in organization.


motivating individuals and groups at work: a social ... · ment of individual goals or outcomes...

Documents