Diversity in goal orientation, team reflexivity, and team performance

Organizational Behavior and Human Decision Processes 114 (2011) 153–164

Contents lists available at ScienceDirect

Organizational Behavior and Human Decision Processes

journal homepage: www.elsevier .com/ locate/obhdp

Diversity in goal orientation, team reflexivity, and team performance

Anne Nederveen Pieterse a,⇑, Daan van Knippenberg b, Wendy P. van Ginkel b

a University of Groningen, The Netherlandsb Rotterdam School of Management, Erasmus University Rotterdam, The Netherlands

a r t i c l e i n f o a b s t r a c t

Article history:Received 17 July 2009Accepted 15 November 2010Available online 15 December 2010Accepted by Paul Levy

Keywords:Goal orientationTeam performanceTeam processesDiversityReflexivity

0749-5978/$ - see front matter � 2010 Elsevier Inc. Adoi:10.1016/j.obhdp.2010.11.003

⇑ Corresponding author. Address: Department of H& Organizational Behavior, Faculty of EconomicsGroningen, Nettelbosje 2, 9747 AE Groningen, The Ne

E-mail address: [email protected] (A. Ne

Although recent research highlights the role of team member goal orientation in team functioning,research has neglected the effects of diversity in goal orientation. In a laboratory study with groups work-ing on a problem-solving task, we show that diversity in learning and performance orientation are relatedto decreased group performance. Moreover, we find that the effect of diversity in learning orientation ismediated by group information elaboration and the effect of diversity in performance orientation bygroup efficiency. In addition, we demonstrate that team reflexivity can counteract the negative effectsof diversity in goal orientation. These results suggest that models of goal orientation in groups shouldincorporate the effects of diversity in goal orientation.

� 2010 Elsevier Inc. All rights reserved.

Introduction

Much of the behavior at work is goal-directed. Accordingly, dif-ferences in goal orientation – preferred goals in achievement situa-tions (Dweck, 1986; Dweck & Leggett, 1988) – have been shown toexert a powerful influence on individual behavior and performanceat work (e.g., Payne, Youngcourt, & Beaubien, 2007; Porath &Bateman, 2006). Given that teams and workgroups are often theprimary unit of organization (DeShon, Kozlowski, Schmidt, Milner,& Wiechmann, 2004; Kozlowski & Bell, 2003) and the abundantevidence for the influence of goal orientation at the individual level,the question arises how goal orientation plays out in a group context.Only recently researchers set foot in this underdeveloped area(LePine, 2005; Porter, 2005), showing that mean levels of goal orien-tation affect group member attitudes and behavior. However, theshift from the individual level of analysis to goal orientation as afactor in team composition introduces another dimension yet unex-plored in goal orientation research – diversity in goal orientation(i.e., differences between team members in goal orientation). In viewof the evidence that diversity on a host of dimensions affects groupprocess and performance (van Knippenberg & Schippers, 2007;Williams & O’Reilly, 1998), it is important for our understanding ofgoal orientation in teams to also study the influence of goal orienta-tion diversity. Moreover, in view of the evidence that main effects areunable to capture the effects of team diversity (van Knippenberg &

ll rights reserved.

uman Resource Managementand Business, University oftherlands.derveen Pieterse).

Schippers, 2007), we develop our analysis of diversity in goalorientations so that we are able to identify a moderator of itsinfluence – team reflexivity (West, 1996).

Thus, we contribute to the literature in several important ways.Our main contribution lies in the demonstration of the importanceof goal orientation diversity for team performance. We, thereby,further develop the goal orientation framework as well as extendresearch in diversity in at least two ways. By integrating insightsfrom research in socially shared cognition with more traditionalperspectives on diversity we extend research in diversity concep-tually (cf. van Knippenberg & Schippers, 2007). Furthermore, ourfindings suggest that diversity in individual differences more prox-imal to behavior in achievement settings may be more influentialthan the personality factors most often studied in this area (i.e.,the Big Five; Bell, 2007). Finally, by identifying team reflexivityas a moderator of the relationships between goal orientation diver-sity and group performance we both substantiate the shared cog-nition perspective underlying our analysis and point to amoderator variable that provides clear opportunities for the man-agement of goal orientation diversity.

Goal orientation

Goal orientation is a predisposition to adopt and pursue certaingoals in achievement contexts (Dweck & Leggett, 1988; Payneet al., 2007; VandeWalle, 1997). In this respect, a distinction ismade between learning orientation and performance orientation(Dweck, 1986). Learning and performance orientation differ inthe standards used to evaluate competence. Learning orientation

http://dx.doi.org/10.1016/j.obhdp.2010.11.003

mailto:[email protected]

http://dx.doi.org/10.1016/j.obhdp.2010.11.003

http://www.sciencedirect.com/science/journal/07495978

http://www.elsevier.com/locate/obhdp

154 A. Nederveen Pieterse et al. / Organizational Behavior and Human Decision Processes 114 (2011) 153–164

makes use of an absolute or intrapersonal standard (i.e., improveown past performance) and performance orientation draws on nor-mative comparisons (i.e., outperform others; Elliot & McGregor,2001; Elliot & Murayama, 2008). Goal orientation is mostly seenas a relatively stable trait that may be influenced by situationalcharacteristics (Button, Mathieu, & Zajac, 1996; Murayama & Elliot,2009). Although learning and performance orientation were origi-nally seen as opposing poles (Dweck, 1986), researchers have ar-gued that individuals often have multiple competing goals(Button et al., 1996). Indeed, research has shown that learning ori-entation and performance orientation are best portrayed as sepa-rate and largely independent dimensions1 (Button et al., 1996;Payne et al., 2007). Thus, people can be high (or low) in both learningand performance orientation.

Goal orientation has received a tremendous amount of atten-tion of researchers at the individual level (Payne et al., 2007), butresearch has only recently started to explore effects of team com-position in goal orientation on team functioning. Studies haveshown that mean levels (i.e., the average of individual members’goal orientation) of both learning and performance orientationare related to team functioning as evident in relationships withsuch concepts as team efficacy, backing-up behavior, team com-mitment, and team adaptation (LePine, 2005; Porter, 2005). How-ever, these studies of mean level effects were unable to establisha relationship between team composition in goal orientation andteam performance.

Previous studies of goal orientation in teams have essentiallyextrapolated hypotheses at the individual level to the team level,taking mean goal orientation as the team-level analogue of individ-ual goal orientation. While the rationale for this approach is clearand supported by both research in goal orientation (DeShon et al.,2004) and research in individual differences in teams more gener-ally (Barrick, Stewart, Neubert, & Mount, 1998; Bell, 2007), a focuson goal orientation in teams also introduces a dimension that hasno analogue at the individual level of analysis: differences in goalorientation between team members – goal orientation diversity.The effects of diversity in goal orientation have been disregardedso far. In sharp contrast with this lack of attention, there is an abun-dance of evidence that diversity on a variety of dimensions rangingfrom demographic to individual differences may affect group pro-cess and performance (van Knippenberg & Schippers, 2007;Williams & O’Reilly, 1998). This literature, especially in combinationwith literature on sharedness and diversity in group members’ taskrepresentations and task mental models, clearly hints at the possi-bility that diversity in goal orientation influences group perfor-mance (Marks, Zaccaro, & Mathieu, 2000; Mohammed & Ringseis,2001; van Ginkel, Tindale, & van Knippenberg, 2009). This suggeststhat to develop our understanding of the role of team composition ingoal orientation, we need to pay attention to goal orientationdiversity. As noted in the previous, trait learning and performanceorientation are best portrayed as independent dimensions insteadof as opposing poles. Accordingly, the study of diversity in trait goalorientation revolves around diversity in each dimension of goalorientation.

1 Goal orientation research has also adopted a subdivision between performanceprove and performance avoid orientation – the former reflecting the disposition toprove one’s competence, the latter reflecting the disposition to avoid displayingincompetence (Elliot & Harackiewicz, 1996; VandeWalle, 1997). As different goalorientations are orthogonal, however, they can be studied without necessarilyincorporating all dimensions (cf. Bunderson & Sutcliffe, 2003). Consistent with theoriginal formulation of goal orientation theory (Dweck, 1986) and with previousresearch on team composition in goal orientation that was the direct inspiration forthe current study (LePine, 2005; Porter, 2005), we focused the present analysis on theprimary distinction between learning and performance orientation (an issue werevisit in ‘‘Discussion’’).

Diversity in goal orientation

Diversity in goal orientation may be classified as concerningdeep-level diversity (e.g., personality and individual differences,attitudes and values; Harrison, Price, & Bell, 1998; van Knippen-berg & Schippers, 2007). One of the major perspectives withinthe diversity literature, the social categorization perspective, seesdiversity as a potential source of ‘us vs. them’ distinctions, wheredissimilar others are categorized as ‘outgroup’. This may disruptgroup process and performance. While such categorizations arenot limited to demographic categories, they are more likely for cat-egories for which individuals hold well-developed stereotypes(Fiske, 1998), which is a condition that would seem highly unlikelyto hold for deep-level differences in goal orientation. The socialcategorization perspective thus seems less suited to understandthe influence of goal orientation diversity.

The similarity/attraction perspective holds that individuals aremore attracted to similar others. As a consequence, people in teamsare more willing to collaborate with others similar to themselvesand as a result interact more smoothly with these others, renderinghomogeneity more conducive to group performance than diversity.The similarity/attraction perspective should readily apply to deep-level similarities and differences (Byrne, 1971), which suggeststhat diversity in goal orientation disrupts group performance byreducing the quality of team member interaction (i.e., communica-tion and coordination). This prediction is well aligned with findingsof positive relationships between deep-level similarity and com-munication and interaction in teams (Harrison et al., 1998;Mohammed & Angell, 2003).

The information/decision making perspective, in contrast,points to the potentially positive effects of diversity. Starting pointfor this perspective is the notion that differences between teammembers may be associated with valuable task-relevant differ-ences in knowledge, expertise, and perspectives, which may ex-pand the available information in diverse teams. While suchdifferences in task-relevant knowledge are unlikely to be limitedto functional dimensions of diversity (van Knippenberg, De Dreu,& Homan, 2004), they may be less prevalent for deep-level diver-sity and accordingly may only play a relatively modest role inthe effects of goal orientation diversity. Moreover, a more recentintegration of the different perspectives on diversity (van Knippen-berg et al., 2004) suggests that social categorization and similarity/attraction processes may disrupt elaboration (i.e., exchange andintegration) of diverse information and perspectives. This suggeststhat similarity/attraction processes associated with goal orienta-tion diversity would render it less likely that groups would benefitfrom any task-relevant insights associated with diversity in goalorientation.

The social categorization, similarity/attraction, and informa-tion/decision making perspectives are the main perspectives onwork group diversity and performance (van Knippenberg &Schippers, 2007; Williams & O’Reilly, 1998). Yet, there is anotherperspective on deep-level diversity in particular that is typicallynot recognized as a perspective on diversity (van Knippenberg &Schippers, 2007): the socially shared cognition perspective.Interestingly and importantly, we propose that this perspective isthe most relevant to understand the influence of goal orientationdiversity. Research in socially shared cognition puts group mem-bers’ mental representation of the task and the team center-stagein understanding team performance (DeChurch & Mesmer-Magnus, 2010; Marks et al., 2000; Mathieu, Heffner, Goodwin,Salas, & Cannon-Bowers, 2000; Mohammed, Ferzandi, & Hamilton,2010; van Ginkel & van Knippenberg, 2008). Guiding principle hereis the proposition that members’ mental representations of thetask and the team guide team member behavior in group interac-tion and group task performance. Task representations (Newell &

A. Nederveen Pieterse et al. / Organizational Behavior and Human Decision Processes 114 (2011) 153–164 155

Simon, 1972) or mental models (Rouse & Morris, 1986) representpeople’s understanding of the task and how to best perform it.The importance of the concept of task representations lies in theproposition that individual’s engagement with a task is not a directfunction of objective characteristics of the task but rather the con-sequence of their mental representation of the task.

Research in socially shared cognition has highlighted the impor-tance of sharedness of representations/mental models (i.e., similar-ity between group members; cf. diversity) and how sharedness ofrepresentations/mental models may feed into team performance(e.g., Marks et al., 2000; Rentsch & Klimoski, 2001; van Ginkel &van Knippenberg, 2008). More specifically, low levels of sharednessof representations (i.e., diversity of representations/mental mod-els) have been associated with breakdowns in coordination (e.g.,Marks, Sabella, Burke, & Zaccaro, 2002; Mathieu, Heffner, Goodwin,Cannon-bowers, & Salas, 2005; Mathieu et al., 2000) and commu-nication (e.g., Marks et al., 2000; Mathieu et al., 2000, 2005), result-ing in lower team performance.

We propose that the key insight for the study of goal orientationdiversity is that there are important linkages between goal orienta-tions and task representations. Goal orientations reflect and followfrom individuals’ understanding of the nature of ability, taskachievement, and performance (Dweck & Leggett, 1988; Elliot &McGregor, 2001). As such, goal orientations are mental frameworksthat represent what people value in achievement situations andtherefore affect how they approach tasks and how they interpretand respond to achievement situations (Dweck & Leggett, 1988;VandeWalle, Brown, Cron, & Slocum, 1999). Much like the perspec-tive on task representations, goal orientation theory thus suggeststhat task cognitions and subjective task goals guide task execution.

Learning orientation engenders task strategies conducive toattaining an understanding of the task (cf. Dweck & Leggett,1988). As a result, high learning orientation expresses itself indeep-level information processing (e.g., Dupeyrat & Mariné,2005; Harackiewicz, Barron, Tauer, Carter, & Elliot, 2000; cf.Chaiken & Trope, 1999). Performance orientation fosters task-execution strategies that maximize the chance of demonstratinghigh ability (cf. Dweck & Leggett, 1988). As a result, individualshigher in performance orientation focus on those aspects of thetask that are perceived to be related to immediate performance(cf. Elliot & McGregor, 1999; Elliot, Shell, Henry, & Maier, 2005;Ford, Smith, Weissbein, Gully, & Salas, 1998). Moreover, perfor-mance-oriented individuals understand performance as flowingfrom ability rather than effort, and thus are more likely to favorlow-effort task-execution strategies than individuals lower inperformance orientation (Fisher & Ford, 1998).

Goal orientation theory and research has not conceptualizedthese differences in achievement goals and task strategies in termsof task representations (i.e., which also emphasize goals and strat-egies), but the parallels are clear. Accordingly, we may conceptual-ize goal orientation diversity as a dimension of deep-level diversityakin to diversity in task representations. We propose that diversityin goal orientation is associated with impaired communication andcoordination, and thus with lower team performance. As it issharedness (and not content) of task representations which is ofconcern here, we expect diversity in learning orientation and diver-sity in performance orientation to have similar effects even thoughthey may relate to very different task representations. This is inline with the similarity–attraction perspective, where the nature(content) of the differences between team members also doesnot play a role in predicting the consequences of these differences.

In sum, drawing from literatures on deep-level diversity, sharedcognition, and goal orientation, we argue that diversity in learningorientation as well as diversity in performance orientation havethe potential to disrupt group dynamics and performance. Fromthese literatures difficulties with coordination and communication

arise as the key aspects of team functioning affected by goal orien-tation diversity. More specifically, we see these processes as beingprimarily reflected in group information elaboration (cf. effectivetask-relevant communication) and group process efficiency (cf.efficient coordination).

Group information elaboration is defined as the exchange, dis-cussion, and integration of task-relevant information and perspec-tives (van Knippenberg et al., 2004) and has been identified as akey mediating process in the relationship between diversity andgroup performance (e.g., Homan, van Knippenberg, Van Kleef, & DeDreu, 2007; Kearney & Gebert, 2009; van Ginkel & van Knippenberg,2008, 2009; cf. Dahlin, Weingart, & Hinds, 2005; Mesmer-Magnus &DeChurch, 2009). Information elaboration is a key process in anyteam task to the extent that the task has information processing,problem-solving, and decision making components. It is anespecially salient concern in diverse groups, both because socialcategorization and similarity/attraction processes associated withdiversity may disrupt elaboration and because information elabora-tion is the key process involved in mobilizing and using diversity asan informational resource (van Knippenberg et al., 2004).

As argued in the previous, diversity in learning orientation andperformance orientation are associated with differences in taskstrategies and attention to information. That is, differences in goalorientation may lead group members to emphasize different infor-mation, to interpret information in reference to different goals, andto differ in the extent to which they desire to engage in in-depth pro-cessing. Research in shared task representations/mental modelsshows that such differences in task approach may impair communi-cation between group members (e.g., Marks et al., 2000; Mathieuet al., 2005). The process of group information elaboration relies onthe collective effort of group members to exchange, discuss, andintegrate information. When different group members emphasizedifferent types of information with different goals in mind (i.e., a fo-cus on in-depth processing to reach a deep understanding vs. a focuson informational cues to apply decision heuristics), group memberswill be less likely to follow up on each others’ remarks and observa-tions because they seem less relevant to the own informational per-spective and processing goals. This in turn will discourage furtherattempts at exchanging and discussing viewpoints, because it sig-nals that contributions to the discussion are less appreciated bythe group (Edmondson, 1999). In effect, goal orientation diversitythus lowers group information elaboration, and as a consequencegroup performance.

The second group process that we propose is involved in the ef-fects of goal orientation diversity is group process efficiency. Col-laborative task performance requires coordination of efforts andcontributions (i.e., including the integration of contributions;Kozlowski & Bell, 2003). When coordination is suboptimal it re-sults in process losses (Steiner, 1972), which are associated withlowered performance (e.g., Brodbeck & Greitemeyer, 2000).Research in mental models shows that sharedness results in great-er efficiency in coordination (Marks et al., 2000; Mathieu et al.,2005). Having a common frame of references in task executionallows group members to better anticipate each other’s behavior,which makes it easier to understand each other and align and inte-grate each others’ contributions (Cannon-Bowers & Salas, 2001). Incontrast, in teams lacking shared representations coordination isoften less smooth, requiring members to further explain, ask for,and clarify issues (Cannon-Bowers, Salas, & Converse, 1993;Espinosa, Lerch, & Kraut, 2004; Salas & Fiore, 2004). As this is likelyto result in a more time-consuming group interaction (cf.Hambrick, Cho, & Chen, 1996), time to accomplish a task is a goodproxy for group process efficiency (Brodbeck & Greitemeyer, 2000).To illustrate this in reference to the process of informationelaboration, group process efficiency reflects the time needed toaccomplish a given level of elaboration.


Thus, as diversity in learning and performance orientation im-plies that contributions to collective performance are made fromthe backdrop of differences in mental frameworks, the coordina-tion of contributions and efforts are more challenging, and thusless time-efficient. In any task context where time is limited (asusual for teams in organizations), lowered group process efficiency– conceptualized as time-efficiency – will go at the expense of thequality or quantity of group performance or both (cf. Jehn, 1997).Accordingly, we propose that group process efficiency is a secondprocess mediating the relationship between diversity in learningand performance orientation, and performance.

Team reflexivity as a moderator

Diversity research suggests that diversity effects are highly con-tingent in nature and that a focus on ‘main effects’ is unlikely toadequately capture diversity’s influence (van Knippenberg &Schippers, 2007). We expect this to also hold for diversity in goalorientation. This is consistent with research in individual differ-ences that argues that situational influences may inhibit, as wellas invite, trait expression (Mischel, 1977; Tett & Burnett, 2003).An achievement situation may invite the expression of individualdifferences in goal orientation, but situational influences may alsohelp align resulting differences and thereby prevent negative ef-fects of goal orientation diversity. Thus, to further develop our con-ceptual analysis we focused on identifying a relevant moderator ofthe effects of goal orientation diversity. The additional benefit ofthis focus on a situational moderator is that it helps to identifyopportunities for managerial intervention, increasing the appliedrelevance of our research.

Our analysis suggests that factors that may stimulate groups tocome to a more shared understanding of the task should moderatethe effects of dispositional differences in goal orientation. There-fore, we focused on a factor that has been argued to be particularlyinstrumental in creating shared understanding: team reflexivity(van Ginkel et al., 2009; West, 1996). Team reflexivity is definedas ‘the extent to which group members overtly reflect upon, andcommunicate about the group’s objectives, strategies, and pro-cesses, and adapt them to current or anticipated circumstances’(West, Garrod, & Carletta, 1997, p. 296). Reflexivity thus differsfrom information elaboration, as reflexivity is focused on discuss-ing ‘metalevel’ issues (i.e., taking a step back to evaluate group pro-cess, strategies, and objectives) and information elaboration isfocused on processing task-relevant information.

Collectively reflecting on goals and strategies has been shown tobe valuable for team functioning and found to relate to satisfaction,commitment, performance, and team innovation (e.g., Carter &West, 1998; Müller, Herbig, & Petrovic, 2009; Schippers, DenHartog, Koopman, & Wienk, 2003; Tjosvold, Tang, & West, 2004).Researchers have argued that teams have a natural tendency tolimit their reflexivity as they are inclined to keep to their custom-ary routines (cf. Gray, 2007; West, 1996). However, reflexivity mayvary widely across teams in organizations and is affected by con-textual variables like team leadership or team training (Hirst,Mann, Bain, Pirola-Merlo, & Richter, 2004; Okhuysen & Eisenhardt,2002; Schippers, Den Hartog, Koopman, & van Knippenberg, 2008).

Reflexivity (i.e., discussing objectives, strategies, and processes)has been argued to be instrumental in surfacing and clarifying dif-ferences in task representations between team members. As differ-ences in understanding of the task may go unnoticed and continueto negatively affect group processes, by bringing these differencesto light reflexivity may help groups to reach a more shared under-standing of task strategies and goals (van Ginkel et al., 2009). Thus,reflexivity seems particularly suited to overcome those issues thatwe identified as the root cause of the negative effects of goal orien-tation diversity. Building on our earlier proposition that diversity

in learning orientation and diversity in performance orientationlower group performance because they imply lower shared under-standing of the task situation at hand, we thus advance the hypoth-esis that team reflexivity attenuates this negative relationship. Byhelping to reconcile different perspectives on the task, team reflex-ivity may diminish goal orientation diversity’s negative relation-ship with group information elaboration and group efficiency.That is, we expect that information elaboration and group effi-ciency mediate the interactions of diversity in learning orientationand diversity in performance orientation with team reflexivity ongroup performance. In sum, we advance the following hypotheses.

Hypothesis 1a. Diversity in learning orientation is moderated byreflexivity in its relationship with group performance, such that itis more strongly negatively related to performance for non-reflexive groups than for reflexive groups.

Hypothesis 1b and 1c. The interaction between diversity in learn-ing orientation and reflexivity on group performance is mediatedby (b) group information elaboration and (c) group processefficiency.

Hypothesis 2a. Diversity in performance orientation is moderatedby reflexivity in its relationship with group performance, such thatit is more strongly negatively related to performance for non-reflexive groups than for reflexive groups.

Hypothesis 2b and 1c. The interaction between diversity in per-formance orientation and reflexivity on group performance is med-iated by (b) group information elaboration and (c) group processefficiency.

We tested these hypotheses in an experiment with three-per-son groups working on relatively complex problem-solving tasks.This set-up allowed us to manipulate team reflexivity and thusto establish causality in the proposed role of reflexivity. Moreover,the controlled set-up made it possible to assess group processesthrough relatively objective behavioral measures with high inter-nal validity (Weingart, 1997).

Method

Participants and design

Participants were 147 students from a university in The Nether-lands, assigned to 49 three-person groups. One group had to beeliminated from the study due to logistical errors. The mean ageof the participants was 20 (SD = 1.83) and 66.7% were male. Themajority of these participants were enrolled in business adminis-tration or economics (97%). A compensation of 15 € (approximately18 USD) was paid out to participants. Groups were randomly as-signed to either the reflexive or the non-reflexive condition andmeasures of learning orientation diversity and performance orien-tation diversity were added to the design as continuous variables.

Task

Groups worked on a collective rule induction task (Laughlin &Hollingshead, 1995). We selected this task, because it is a relativelycomplex problem-solving task that entails both performance andlearning elements. Moreover, not only performance but also com-petence increase is feasible over the period in the lab, making bothperformance and learning orientation (a focus on competence in-crease) realistic for participants. Therefore, both learning and per-formance orientation should play a role in the task, which makes it

2 As the task also has disjunctive elements we also tested whether the maximumlevel of learning and performance orientation exerted an influence on group processor performance. No significant effects of maximum goal orientation were found. Asmaximum levels of goal orientation correlated very highly with both mean anddiversity in goal orientation we left them out of our final model to avoid issues ofmulticollinearity.


possible for diversity in learning and performance orientation toplay out. Moreover, for an adequate test of reflexivity as a moder-ator, a task with multiple rounds was necessary. Collective induc-tion entails a cooperative search for rules and principles, byobservations of patterns and relationships and by testing and revis-ing hypotheses (Laughlin & Hollingshead, 1995; cf. for example thework of scientific research teams and specialized medical teams).

Groups were to discover a series of card sorting rules using a stan-dard deck of 52 playing cards (e.g., Laughlin & Hollingshead, 1995;Laughlin, VanderStoep, & Hollingshead, 1991). Instructions ex-plained that each sorting rule could be based on any attribute ofthe cards (e.g., suit, color, numerical and logical sorting rules, etc.)and several examples were given (e.g., red–black–red). The taskconsisted of four games, where in each game participants had to dis-cover a new sorting rule. At the onset of each game the experimenterpresented the group with a card that was in correspondence with thesorting rule. Each round started with participants thinking individu-ally what they thought the sorting rule might be. Next, they were tocome up with a group hypothesis about the rule through group dis-cussion. Then, the group had to choose a card from the deck on whichthey wanted to receive feedback whether it fitted the sorting rule.Each round ended with feedback on the card by the experimenter.A maximum of 10 rounds (i.e., consisting of individual hypotheses,group hypothesis, playing a card, receiving feedback) and 10 minwere given per sorting rule. After the 10th round or when the10 min were up, groups were to write down their final group hypoth-esis. After this, the experimenter informed the groups about the cor-rect rule. Then, groups started with the next sorting rule, until fourrules were played.

Goal orientation

Goal orientation was measured using the validated 16-item ques-tionnaire of Button et al. (1996), with eight items measuring perfor-mance orientation (a = .71, M = 3.77, SD = .49) and eight itemsmeasuring learning orientation (a = .65, M = 3.98, SD = .38). Sampleitems include ‘‘The things I enjoy the most are the things I do thebest’’ (performance orientation) and ‘‘The opportunity to learnnew things is important to me’’ (learning orientation). The itemswere rated on a 5-point scale, ranging from 1 (completely disagree)to 5 (completely agree). Confirmatory factor analysis showed theintended two-factor solution had adequate fit to the data(v2 = 189.54, df = 103, GFI = .85, RMSEA = .08; Browne & Cudeck,1993; MacCallum, Browne, & Sugawara, 1996). Moreover, this mod-el had a significant better fit than a one-factor solution; v2 = 283.47,df = 104, GFI = .78, RMSEA = .11; Dv2 = 93.93, p < .001) or an inde-pendence model v2 = 454.62, df = 120, Dv2 = 265.08, p < .001. Todetermine diversity in learning and performance orientation withinteams, we calculated the standard deviations of the learning and per-formance orientation scores of all team members within each team(Harrison & Klein, 2007).

Experimental manipulation of reflexivity

Reflexivity was manipulated through written instructions.Reflexive groups were informed that to do well on the task itwas important to discuss as much as possible how they are doing.Before starting on the task they were given 2 min to discuss thebest approach to the task. In addition after each rule reflexivegroups were given 1 min to discuss their team work, whether theyused the right approach to the task, what caused mistakes to occur,and how they could do better. They were encouraged to use the gi-ven time fully for this purpose, and to continue to reflect during thetask. In the non-reflexive condition groups were given no extrainstructions. These groups were given 2 min waiting time beforethe first rule and 1 min after each rule.

Group performance, group information elaboration and group processefficiency

Group performance was operationalized as the total number ofcorrect final group hypotheses, ranging from 0 (no correct finalhypotheses) to 4 (all final hypotheses correct).

Group information elaboration was measured using audio–vi-deo recordings of 44 groups (four groups had to be omitted dueto technical problems). A coding scheme was adjusted from vanGinkel and van Knippenberg (2008), rooted in concrete behavioralanchors, such as sharing possible solutions, discussing evidence foror against possible solutions, explaining trains of thought to othergroup members, remarks inviting information elaboration (e.g.,‘‘let’s keep our options open’’, ‘‘what else could it be?’’), and elab-oration avoidant remarks (e.g., ‘‘just do whatever’’, ‘‘it doesn’t mat-ter anyway’’). Two coders blind to the conditions rated the groupson information elaboration giving a score from 1 to 5, where ahigher score represented more information elaboration. One coderrated all groups, and the second coder rated a subset of 30% of thegroups to determine interrater reliability (M = 2.99, SD = 1.17,r = .92). As reliability was sufficiently high the ratings of the firstcoder were used in analyses.

Group process efficiency was operationalized followingBrodbeck and Greitemeyer (2000) as the time (in seconds) takenfor group discussion before deciding which hypothesis to writedown and which card to play. Longer time reflects lower efficiency.Previous research suggests the independence of time-efficiencyand information elaboration in group work (cf. Scholten, vanKnippenberg, Nijstad, & De Dreu, 2007). Indeed, in the presentstudy information elaboration and group efficiency were not re-lated (r = �.12, ns). Moreover, our analyses enabled us to controlfor information elaboration when examining group efficiency andvice versa.

Manipulation check and controls

As a manipulation check for the reflexivity manipulation fourquestions were adjusted from Schippers, Den Hartog, and Koopman(2007; M = 3.25, SD = .71, a = .61). A sample item is ‘‘As a group wereviewed various approaches to the task’’ 1 (completely disagree)to 5 (completely agree).

In addition, we included a number of controls. We measuredtask liking using three items on a 5-point scale ranging from 1(completely disagree) to 5 (completely agree). As task liking mayvary greatly between people and may affect performance indepen-dently of our predicted effects, controlling for it allows us to filterout its influence on performance. A sample item is ‘‘I liked workingon the task’’ (M = 3.87, SD = .67, a = .66). For exploratory purposes,we included a more and a less difficult version of the task. Preli-minary analysis indicated that while performance was lower onthe difficult version of the task, task difficulty did not moderateany of the relationships of interest, rendering it essentially irrele-vant to the current purposes. We included a dummy representingtask difficulty as a control to reduce error variance. We also con-trolled for gender diversity to make sure our findings for goal ori-entation diversity were unrelated to potential gender differences.

We also added mean learning and performance orientation tothe model. Previous studies have indicated that goal orientation isrelated to several team level variables and we were interested in ef-fects of diversity in goal orientation independent of mean levels.2

Table 1Descriptive statistics and correlations among the variables.

Variable M SD 1 2 3 4 5 6 7 8 9 10

1. Task difficulty – –2. Task liking 3.87 .44 �.35**

3. Gender diversity .32 .20 .12 .104. Mean learning orientation 3.97 .22 �.07 .24 �.075. Mean performance orientation 3.77 .28 �.17 .22 �.01 .146. Diversity in learning orientation .34 .16 .12 �.05 �.04 .04 �.247. Diversity in performance orientation .43 .23 �.08 .09 .13 .14 .08 �.278. Reflexivity – – .05 .08 .28* .02 .24 �.07 .129. Group process efficiency 29.73 5.24 .01 �.20 .01 �.07 .03 �.05 .17 �.1810. Group information elaboration 2.99 1.17 �.05 .43** �.12 �.12 .25 .02 �.10 �.07 �.1211. Team performance 1.56 .99 �.42** .52** �.08 �.02 �.08 �.06 �.17 �.11 �.41** .58**

N = 48, for correlations with group efficiency and information elaboration N = 44.* p < .05.

** p < .01.


Moreover, inclusion of the mean learning and mean performance ori-entation also adds to the comparability with previous study findings(LePine, 2005; Porter, 2005).

Procedure

On arrival in the laboratory participants were seated separatelyand asked to fill out a questionnaire including a measure of goal ori-entation. When finished they read the instructions for the grouptask and in the reflexivity condition they were given the reflexivityinstructions (see above). When all group members were finishedreading the instructions, they were seated at one table as a groupand the experimenter repeated the basic task instructions. In thereflexivity condition the reflexivity instruction was repeated andthey were given 2 min to discuss how to approach the task. In thenon-reflexive condition groups received no extra instructions andwere given their first exemplar after 2 min. Then groups startedon the task as described above. After 5 min groups were warned thatthey had 5 min left. After each rule the reflexive groups were given1 min to reflect and the non-reflexive groups waited for 1 min forthe next game to start. After the final rule group members wereseated separately again and were given a final questionnaire includ-ing manipulation checks and controls after which they were de-briefed, thanked, and paid out for their participation. The entireexperiment took approximately one and a half hours.

Results

Preliminary analyses and manipulation check

Table 1 displays all correlations. No significant correlationswere found between the independent variables and control vari-ables. Group information elaboration, group efficiency, task diffi-culty, and task liking were related to team performance.

Awg(1) values were calculated to check whether agreement onthe manipulation check of reflexivity warranted analysis at thegroup level. Awg(1) index has been proposed to overcome severallimitations attached to the Rwg index (Brown & Hauenstein,2005). A value of .85, well above the threshold of .70 (Brown &Hauenstein, 2005) was found, justifying group-level analysis. Inaddition, ICC values indicated group-level analysis was warranted(ICC1 = .18; ICC2 = .65). A regression analysis of the reflexivitymanipulation check on all variables was performed. Only a maineffect of the reflexivity manipulation was found (b = .37, b = .36,p < .05), such that groups in the reflexive condition scored higheron the measure than groups in the non-reflexive condition. Nointeractions or other main effects were found. These findings indi-cate the effectiveness of the reflexivity manipulation.

Hypothesis testing

We used hierarchical multiple regression to test the hypotheses.Centered variables were used, following Aiken and West (1991). Inthe first step the regression model included the control variables,reflexivity (dummy coded �.5 and +.5), diversity in learning orien-tation, and diversity in performance orientation. Task liking was, asexpected, positively related to group performance. No other maineffects were found. In the second step the interactions of reflexivitywith diversity in learning orientation and diversity in performanceorientation were added. The second step had a significant addedvalue over the model in the first step. Table 2 shows the resultsof these analyses.

As expected (Hypothesis 1a) we found moderation by reflexiv-ity of the relationship between diversity in learning orientationand group performance (see Fig. 1). Next we performed simpleslopes analysis, following Aiken and West (1991). For groups inthe non-reflexive condition a negative relationship of diversity inlearning orientation with group performance was found(b = �3.49, b = �.58, p < .01). For the reflexive groups a positiverelationship between diversity in learning orientation and perfor-mance was found (b = 2.15, b = .36, p < .05). Moreover, for highlydiverse groups (+1 SD) reflexivity had a positive relationship withperformance (b = .78, b = .40, p < .05). Unexpectedly we foundreflexivity to have a negative relationship with group performancefor more homogeneous (�1 SD) learning oriented groups(b = �1.00, b = �.51, p < .01).

In line with Hypothesis 2a, an interaction was found betweenreflexivity and diversity in performance orientation on groupperformance (see Fig. 2). Simple slopes analysis showed that fornon-reflexive groups diversity in performance orientation had anegative relationship with group performance (b = �2.73,b = �.63, p < .001). For the reflexive groups no relationship be-tween diversity in performance orientation and performance wasfound (b = .70, b = .16, ns). Also, we found reflexivity to have a po-sitive relationship with performance for groups highly diverse inperformance orientation (b = .64, b = .33, p < .05). Finally, forgroups with low diversity in performance orientation we found anegative relationship between reflexivity and group performance(b = �.86, b = �.44, p < .01).

Until recently little attention had been paid to the methodologyof simultaneous testing of multiple mediators (Preacher & Hayes,2008). An obvious reason is the complexity of these models. How-ever, simultaneous testing has some clear advantages over thetesting of several simple mediations; it reduces parameter biasand it is possible to test whether specific mediators mediate the ef-fect of x on y conditional on other mediators in the model. There-fore, Preacher and Hayes (2008) argue against using the causal

Table 2Hierarchical regressions.

Variable Step 1 Step 2

b SE b b t b SE b b t

Task difficulty �.52 .26 �.27 �2.02 �.73 .23 �.37 �3.25**

Task liking 1.12 .31 .50 3.66** 1.51 .27 .68 5.66**

Gender diversity �.28 .64 �.06 �.43 �.51 .54 �.10 �.94Mean learning orientation �.49 .57 �.11 �.86 �.81 .48 .18 �1.69Mean performance orientation �.26 .48 �.07 �.55 �.59 .41 �.17 �1.46Diversity in learning orientation �.52 .79 �.09 �.66 �.56 .65 .09 �.86Diversity in performance orientation �1.00 .57 �.23 �1.75 �.94 .48 �.22 �1.96Reflexivity �.16 .26 �.08 �.63 �.11 .22 �.06 �.51Diversity in learning orientation* reflexivity 5.41 1.41 .45 3.84**

Diversity in performance orientation* reflexivity 3.29 .97 .38 3.41**

R2 = .43** for Step 1; DR2 = .20** for Step 2.N = 48.

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

0.5

1

1.5

2

2.5

highlowDiversity in learning orientation

Team

per

form

ance

non-reflexive groups

reflexive groups

Fig. 1. The interaction between diversity in learning orientation and teamreflexivity on team performance.

0.5

1

1.5

2

2.5

highlow

Diversity in performance orientation

Team

per

form

ance


reflexive groups

Fig. 2. The interaction between diversity in performance orientation and teamreflexivity on team performance.

1,8

2,3

2,8

3,3

3,8

low high

Gro

up in

form

atio

n el

abor

atio

n

Diversity in learning orientation


reflexive groups

Fig. 3. The interaction between diversity in learning orientation and teamreflexivity on group information elaboration.


steps method by Barron and Kenny (1986) for testing models withmultiple mediators. In addition, the use of bootstrapping tech-niques is generally most appropriate for testing indirect effects,as the sampling distribution is rarely normal or symmetrical, vio-lating the assumptions of the normal-theory tests for mediation(e.g., Preacher & Hayes, 2008; Shrout & Bolger, 2002). Bootstrap-ping is a non-parametric test, which does not require assumptionsof normality of the sampling distribution. It involves repeatedresampling from the dataset to estimate an empirical approxima-tion of the sampling distribution of the indirect effects. Bootstrap-ping has been frequently used in former research to test formediation (e.g., Giessner & van Knippenberg, 2008). Therefore,we used the Preacher and Hayes method for multiple mediation

to test whether information elaboration and group efficiency med-iated the interactions of reflexivity with diversity in learning orien-tation and diversity in performance orientation on groupperformance (Hypotheses 1b, 1c, 2b, and 2c). The method allowsfor simultaneous testing of direct and indirect (through mediators)effects of independent variables on the dependent variable. Fullmediation can be concluded when the specific indirect effect ofthe interaction on the dependent variable through the mediatordiffers from 0 and the total (indirect + direct) effect of the interac-tion on the dependent variable differs from 0, but the direct effectof the interaction on its own does not differ from 0.

The specific indirect effect of the interaction of diversity inlearning orientation with reflexivity on performance throughinformation elaboration was significant (point estimate = 1.54,SE = 1.04, 95% CI: .15–4.37) in line with Hypothesis 1b (seeFig. 3). However, the specific indirect effect through group effi-ciency was not significant (point estimate = .85, SE = .70, 95%CI = �.03 to 2.84), opposing Hypothesis 1c. The total effect (indi-rect + direct) of the interaction of diversity in learning orientationwith reflexivity on performance differed from 0 (point esti-mate = 5.84, SE = 1.46, t = 4.01, p < .001) and was stronger thanthe direct effect alone (point estimate = 3.45, SE = 1.39, t = 2.49,p < .05), showing partial mediation.

Against expectations (Hypothesis 2b) no specific indirect effectof the interaction between diversity in performance orientationand reflexivity on group performance through information elabo-ration was found (point estimate = .37, SE = .50, 95% CI = �.45 to1.68). However, the specific indirect effect through group efficiencywas significant (see Fig. 4), in line with Hypothesis 2c (point

highlow

Gro

up e

ffic

ienc

y

Diversity in performance orientation


reflexive groups

Fig. 4. The interaction between diversity in performance orientation and teamreflexivity on group process efficiency.


estimate = .57, SE = .43, 95% CI = .002–1.98). The total effect(indirect + direct) of the interaction on performance also differedfrom 0 (point estimate = 3.11, SE = .95, t = 3.26, p < .01) and wasstronger than the direct effect on its own (point estimate = 2.17,SE = .85, t = 2.54, p < .05), showing partial mediation.

Discussion

Recent findings in team research suggest that goal orientationnot only has relevance for individual performance but also as ateam composition variable (LePine, 2005; Porter, 2005). We addan important new dimension to this emerging field of researchby demonstrating the influence of goal orientation diversity inteams. Results confirmed our proposition that diversity in learningorientation as well as diversity in performance orientation has anegative relationship with group performance, which can be coun-teracted by reflexivity. Also, results give insight into the underlyingprocesses by showing that the interaction for diversity in learningorientation was mediated by group information elaboration andthe interaction for diversity in performance orientation was medi-ated by group efficiency.

Theoretical implications

Our results demonstrate that diversity in goal orientation in-deed plays a role in team functioning, and – in contrast to meanlevels of goal orientation – also in team performance. Previous re-search has shown the value of mean goal orientation for groupmember attitudes and behavior, but no relationship with groupperformance was found (e.g., Porter, 2005).3 The fact that we founddiversity in goal orientation to be related to group performancewhen mean goal orientation was not, tentatively suggests that goalorientation diversity may be a more important influence on teamperformance than the specifics of this goal orientation per se. Clearly,future replications and extensions are in order to bolster this conclu-sion. Even so, the present findings indicate that one cannot simplyextrapolate goal orientation research from the individual level tothe team level, as on the team level additional issues come into playthat are fundamentally different from the individual level due to dif-ferences between team members. Thus, the findings seem to put a

3 Bunderson and Sutcliffe (2003) examined team learning orientation as a teamclimate variable and found a curvilinear relationship with team performance. Whiletheir focus on team climate is distinct from the current focus on individual differences– including the mean of these individual differences – we explored the possibility thatmean learning orientation had a curvilinear relationship with performance. Thisrelationship was not significant.

premium on the study of goal orientation in teams with attentionto the influence of goal orientation diversity.

An interesting unpredicted outcome of the present study is thatdiversity in learning orientation was related to group performancethrough information elaboration, while diversity in performanceorientation affected group performance through process efficiency.Thus, even though diversity in both orientations had similar rela-tionships with performance, the disparity in mental frameworksbetween the orientations did play out in differing underlying pro-cesses. A possible explanation for these differential pathways maybe found in differences between learning orientation and perfor-mance orientation at the individual level of analysis. Learning ori-entation more than performance orientation has been linked todifferences in information processing, both in terms of the depthof processing and (by implication) in terms of the information towhich individuals are likely to attend (e.g., informational cues thatcan be used in decision heuristics). We outlined in the introductionhow such differences in emphasis would result in lower elabora-tion, and indeed this proposition was supported. While the basisfor decisions may thus be poorer as a consequence of more shallowinformation elaboration, this need not express itself in loweredefficiency in coming to an agreement about task-relevant issues(cf. Kooij-De Bode, van Knippenberg, & van Ginkel, 2008; Scholtenet al., 2007). Differences in performance orientation, in contrast,are less related to information processing but more to strategypreferences (e.g., Harackiewicz et al., 2000). Such differences maylower group process efficiency by increasing the difficulties inaligning different contributions, but they do not necessarily stimu-late or impede discussion of decision-relevant information. That is,the differences in task approach associated with different levels ofperformance orientation are less likely to express themselves inthe processing of information, and accordingly less likely to affectgroup information elaboration. Even so, they may reduce groupprocess efficiency when they result in less successful or slowerintegration of contributions made from different strategicperspectives.

Diversity in learning orientation was found to relate positivelyto performance when groups reflected on their processes, goals,and strategies. This finding is in line with the information/decisionmaking perspective, which argues that diversity may have positiveconsequences. Moreover, in a recent integration of the diversity lit-erature researchers have proposed that diversity may contributeuseful insights, but only when barriers to information elaborationhave been removed (van Knippenberg et al., 2004). In our studythe creation of a shared understanding through team reflexivityserved this purpose. Interestingly, this positive relationship wasonly obtained for learning orientation diversity. This is consistentwith the finding that only learning orientation diversity was re-lated to information elaboration and with the proposition thatinformation elaboration is the key process mediating the positiveeffects of diversity (van Knippenberg et al., 2004).

The finding that diversity in goal orientation plays an importantrole in group functioning demonstrates the role of goal orientationin social interactions. Relatively little attention has been paid tothis area of research, and the present research adds to claims ofthe importance of this relatively new area in the literature (e.g.,Darnon, Butera, & Harackiewicz, 2007; Janssen & van Yperen,2004). Our study shows that goal orientation is not merely an in-tra-psychic phenomenon with effects occurring only within indi-viduals, but has distinct intragroup effects. Interestingly andimportantly, however, these effects do not obtain from goal orien-tation per se (i.e., mean goal orientation) but from differences ingoal orientation.

The moderating effect of reflexivity suggests support for ourclaim that aligning task representations may help groups deal withdiversity in goal orientation. Although we did not measure


sharedness of task representations, previous research has under-lined the relationship between reflexivity and shared understand-ing of the task (van Ginkel et al., 2009). This implies that othervariables that affect sharedness of task representations may alsohelp in dealing with diversity in goal orientation. For example,variables that have been related to shared mental models may beof influence here, such as team size or team experience (Rentsch& Klimoski, 2001). In addition, leaders may instigate a shared men-tal model (cf. Kozlowski, Gully, McHugh, Salas, & Cannon-Bowers,1996; van Ginkel & van Knippenberg, in press) and align task strat-egies, possibly decreasing effects of diversity in goal orientation.Also, as research has shown that goal orientation can be inducedby situational characteristics (Button et al., 1996), stimulating sim-ilar (state) goal orientation within a team may be beneficial. Final-ly, the moderating effect of reflexivity implies that teams insituations which provide limited opportunities for discussion orinteraction, such as virtual teams, may be more prone to sufferthe negative consequences of goal orientation diversity.

With homogeneity in learning orientation as well as homogene-ity in performance orientation, reflexivity was less beneficial. Thisis in line with previous research that unexpectedly found negativeeffects for reflexivity (De Dreu, 2002). De Dreu speculated thatreflexivity may not affect team performance itself, but may serveas a moderator either causing constructive controversy and explo-ration of differing viewpoints leading to positive outcomes, orcausing concurrence seeking and limited breadth of informationsearch leading to negative consequences. Our study seems not onlyto validate his argument, but to take a step further by determiningwhat circumstances may determine these alternative effects. Itseems that discussing similar goals, strategies, and processesmay have limited added value. Moreover, reflecting on similargoals and strategies may overly strengthen team members’ posi-tions causing processes in line with groupthink, which diminishteam performance (De Dreu, 2002; Janis, 1972). Positive effectsof reflexivity may instead only occur when it is necessary to aligngroup members, when constructive controversy over differingviewpoints may improve performance. Thus, in other words the ef-fects of reflexivity may be moderated by ‘need for reflexivity’. Thismay be an important extension in the reflexivity literature.

The results of our study may also have implications for researchin diversity in personality and individual differences more gener-ally. Results in this domain have been remarkably inconclusive(Bell, 2007). A reason for this may be that the variables studiedwere often quite distal to behavior in achievement settings (i.e.,primarily the ‘‘Big Five’’ factors of personality). Diversity of individ-ual difference variables more proximal to behavior in achievementsettings such as goal orientation may therefore have stronger ef-fects. Accordingly, it may be worthwhile for future research in thisdomain to focus on such variables more proximal to behavior inorganizations.

Limitations and suggestions for future research

The present study is not without limitations. Our study madeuse of a laboratory setting with a student sample. Although obvi-ous benefits are increased control, evidence for causality, and rela-tively objective measurement of team process (i.e., throughobservation rather than retrospective self-rating) and performance,the use of a laboratory setting may raise questions concerning gen-eralizability. However, previous studies have demonstrated thatfindings in the laboratory are often replicated in the field and thereis no reason to expect students to differ from other populations intheir behavior in achievement settings (e.g., Brown & Lord, 1999;Dipboye, 1990; Locke, 1986). The goal of the present study wasnot to demonstrate external validity, but to study underlying rela-tionships for which a laboratory setting is most appropriate (cf.

Brown & Lord, 1999; Mook, 1983). However, to address concernssome may have with issues of generalizability, replicating the pres-ent findings in the field would be valuable.

While the measure of goal orientation we employed (Buttonet al., 1996) is thoroughly validated and used by much recent re-search on the topic (e.g., DeShon et al., 2004; Porter, 2005), perfor-mance orientation may be subdivided into prove (or approach) andavoid dimensions (Elliot & Harackiewicz, 1996; VandeWalle, 1997;some argue the same holds for learning orientation, Elliot &McGregor, 2001, but this is considerably more controversial, e.g.,DeShon & Gillespie, 2005). Our study does not speak to this distinc-tion. A recent meta-analysis suggests that the measure of perfor-mance orientation we employed mainly concerns performanceprove/approach orientation (Payne et al., 2007). Our study thuscan be taken to speak to performance prove orientation, but doesnot speak to the influence of performance avoid orientation – thedisposition to avoid being perceived as incompetent. The inspira-tion for our study was recent work on goal orientation in teamswhich worked from the two-dimensional distinction betweenlearning and performance orientation (DeShon et al., 2004; LePine,2005; Porter, 2005). Previous research found mainly this distinc-tion to play a role in interpersonal settings (e.g., Poortvliet, Janssen,Van Yperen, & Van de Vliert, 2007). Moreover, our analysis focusedon task representations as the focal underlying mechanism wheredifferences in achievement standards are the focal concern, not dif-ferences in approach or avoidance motivation. As the dimensionsof goal orientation are orthogonal, not including a performanceavoid measure does not disqualify the current findings for learningorientation or performance (prove) orientation.

Nevertheless, examining effects of diversity in performanceavoid orientation would have added value to research on goal ori-entation in teams. We would predict that our analysis extends toperformance avoid orientation, as diversity in performance avoidorientation should also relate to differences in task representa-tions. Therefore, diversity in performance avoid orientation maysimilarly relate to diminished group performance – unless teamscreate greater sharedness through reflexivity. Moreover, as perfor-mance avoid orientation holds similar achievement standards asperformance approach orientation (i.e., normative), but like learn-ing orientation has been related to deep information processing(but negatively; Elliot, McGregor, & Gable, 1999) both informationelaboration and efficiency may be relevant mediating processes.Future research including the performance avoid dimension maynot only validate this suggestion, but also examine whether ourinitial evidence for the greater impact of diversity compared tomean effects of team composition in goal orientation holds for allcomponents. Considering the extensive negative consequences ofperformance avoid orientation, mean (or maximum) compositioneffects might be as large as or even overshadow diversity effectson this dimension. As previous research into team composition ingoal orientation has also only included the main distinction be-tween learning and performance orientation, studying all dimen-sions in future research seems desirable.

We developed our theoretical analysis around an integration ofresearch in goal orientation and research in socially shared cogni-tion. While shared cognition typically is not recognized as reflectingdiversity (or rather, the reverse), the present support for our analysiscorroborates the usefulness of conceptualizing deep-level dimen-sions of diversity like goal orientation in terms of task representa-tions. We acknowledge that we did not establish this link betweengoal orientation diversity and (shared) task representations empiri-cally, but derived it from the strong linkages between goal orienta-tion theory and theories of task representations and mentalmodels. Therefore, it would be valuable if further research wouldempirically establish the link between goal orientation and task rep-resentations, not only because it would provide further validation


for the current analysis, but also because it would more firmly openthe door for an integration of research in diversity and research in so-cially shared cognition. In this respect, it is important to realize thatlearning orientation and performance orientation capture differenttask representations, and accordingly that operationalizationsshould reflect this. An interesting, related question is whether a con-ceptualization in terms of task representations is also useful inunderstanding the influence of diversity in other deep-level dimen-sions, and potentially also of other types of diversity such as func-tional diversity which may be associated with different taskrepresentations through training and education rather thandisposition.

Finally, the collective induction task employed in this study wasselected because it contained both performance and learning ele-ments. Although collective induction is an intellective task, it re-quires the generation of hypotheses and thus creative problem-solving plays a role in task performance. Previous research has ar-gued that diversity of opinions, expertise, and perspectives may bemore helpful on these tasks (van Knippenberg et al., 2004; Wil-liams & O’Reilly, 1998). Therefore, the positive effect of diversityin learning orientation might not have occurred on a task with lessproblem-solving requirements. Moreover, the detrimental effectsof diversity in goal orientation may be more pronounced on differ-ent kinds of tasks. In contrast, effects of mean goal orientation maybe more pronounced on tasks where either a high performance orhigh learning orientation is more clearly advantageous (e.g., a taskfor which adjusting to changed circumstances is highly relevant forteam performance; cf. LePine, 2005) and we should be careful notto discard the possibility that mean levels of goal orientation mayunder specific circumstances affect team performance.

Managerial implications

Several studies have advanced the suggestion of selectingemployees on the basis of goal orientation (e.g., VandeWalleet al., 1999). Based on the present study we would add that orga-nizations may be well-advised to take into account the goal orien-tation of other team members when making selection decisions.When forming teams selecting team members with similar levelsof goal orientation may be worthwhile. This is especially true whenit is unlikely that the team will engage in high levels of reflexivity.Results of the present study may imply more generally that it isuseful to base the profile of the ideal applicant (at least partly)on the team members he or she will be working with.

Moreover, our results show that reflexivity may counter the det-rimental effects of diversity in goal orientation. Thus by stimulatingteam reflexivity organizations may attenuate the negative influenceof diversity in goal orientation. This may be achieved through train-ing teams to foster team reflexivity (Okhuysen & Eisenhardt, 2002)or by aligning task strategies by other means, for example by givingclear directions for the best strategy. In addition, although goal ori-entation is a relatively stable trait, it can be affected by situationalfactors (Button et al., 1996). Therefore, inducing a shared state goalorientation may help teams be more effective, for example byemphasizing the importance of learning or performance and compe-tition, or by suitable compensation and feedback systems.

Conclusion

As organizations make more and more use of teams as their ba-sic units, the study of what affects team functioning and perfor-mance is gaining in importance. Recent research has shown theimportant role of team composition in goal orientation for teamfunctioning. The present study extends and develops this emergingliterature by establishing the influence of diversity in goal orienta-tion on team processes and performance. In addition, the identifi-

cation of a means to counteract these effects opens up promisingfuture research opportunities.

References

Aiken, J. S., & West, S. G. (1991). Multiple regression: Testing and interpretinginteractions. New York: Sage.

Barrick, M. R., Stewart, G. L., Neubert, M. J., & Mount, M. K. (1998). Relating memberability and personality to work-team processes and team effectiveness. Journalof Applied Psychology, 83, 377–391.

Barron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction insocial psychological research: Conceptual, strategic, and statisticalconsiderations. Journal of Personality and Social Psychology, 51, 1173–1182.

Bell, S. T. (2007). Deep-level composition variables as predictors of teamperformance: A meta-analysis. Journal of Applied Psychology, 92, 595–615.

Brodbeck, F. C., & Greitemeyer, T. (2000). A dynamic model of group performance:Considering the group members’ capacity to learn. Group Processes andIntergroup Relations, 3, 159–182.

Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit. In K. A.Bollen & J. S. Long (Eds.), Testing structural equation models (pp. 136–162).Beverly Hills, CA: Sage.

Brown, R. D., & Hauenstein, N. M. A. (2005). Interrater agreement reconsidered: Analternative to the rwg indices. Organizational Research Methods, 8, 165–184.

Brown, D. J., & Lord, R. G. (1999). The utility of experimental research in the study oftransformational/charismatic leadership. The Leadership Quarterly, 10, 531–539.

Bunderson, J. S., & Sutcliffe, K. M. (2003). Management team learning orientationand business unit performance. Journal of Applied Psychology, 88, 552–560.

Button, S. B., Mathieu, J. E., & Zajac, D. M. (1996). Goal orientation in organizationalresearch: A conceptual and empirical foundation. Organizational Behavior andHuman Decision Processes, 67, 26–48.

Byrne, D. (1971). The attraction paradigm. New York: Academic Press.Cannon-Bowers, J. A., & Salas, E. (2001). Reflections on shared cognition. Journal of

Organizational Behavior, 22, 195–202.Cannon-Bowers, J. A., Salas, E., & Converse, S. (1993). Shared mental models in

expert team decision making. In N. J. J. Castellan (Ed.), Individual and groupdecision making: Current issue (pp. 221–246). Hillsdale, NJ: Erlbaum.

Carter, S. M., & West, M. A. (1998). Reflexivity, effectiveness and mental health inBBC production teams. Small Group Research, 29, 583–601.

Chaiken, S., & Trope, Y. (1999). Dual process theories in social psychology. New York:Guilford Press.

Dahlin, K. B., Weingart, L. R., & Hinds, P. J. (2005). Team diversity and informationuse. Academy of Management Journal, 48, 1107–1124.

Darnon, C., Butera, F., & Harackiewicz, J. M. (2007). Achievement goals in socialinteractions: Learning with mastery vs. performance goals. Motivation andEmotion, 31, 61–70.

De Dreu, C. K. W. (2002). Team innovation and team effectiveness: The importanceof minority dissent and reflexivity. European Journal of Work and Occupationaland Organizational Psychology, 70, 285–298.

DeChurch, L. A., & Mesmer-Magnus, J. R. (2010). The cognitive underpinnings ofeffective teamwork: A meta-analysis. Journal of Applied Psychology, 95, 32–53.

DeShon, R. P., & Gillespie, J. Z. (2005). A motivated action theory account of goalorientation. Journal of Applied Psychology, 90, 1096–1127.

DeShon, R. P., Kozlowski, S. W. J., Schmidt, A. M., Milner, K. R., & Wiechmann, D.(2004). A multiple-goal, multilevel model of feedback effects on the regulationof individual and team performance. Journal of Applied Psychology, 89,1035–1056.

Dipboye, R. L. (1990). Laboratory vs. field research in industrial and organizationalpsychology. In C. L. Cooper & I. T. Robertson (Eds.). International review ofindustrial and organizational psychology (Vol. 5). Chichester, UK: Wiley.

Dupeyrat, C., & Mariné, C. (2005). Implicit theories of intelligence, goal orientation,cognitive engagement, and achievement: A test of Dweck’s model withreturning to school adults. Contemporary Educational Psychology, 30, 43–59.

Dweck, C. (1986). Motivational processes affecting learning. American Psychologist,41, 1040–1048.

Dweck, C. S., & Leggett, E. L. (1988). A social-cognitive approach to motivation andpersonality. Psychological Review, 95, 245–273.

Edmondson, A. (1999). Psychological safety and learning behavior in work teams.Administrative Science Quarterly, 44, 350–383.

Elliot, A. J., & Harackiewicz, J. M. (1996). Approach and avoidance achievement goalsand intrinsic motivation: A mediational analysis. Journal of Personality and SocialPsychology, 70, 968–980.

Elliot, A. J., & McGregor, H. A. (1999). Test anxiety and the hierarchical model ofapproach and avoidance achievement motivation. Journal of Personality andSocial Psychology, 76, 628–644.

Elliot, A. J., & McGregor, H. A. (2001). A 2 � 2 achievement goal framework. Journalof Personality and Social Psychology, 80, 501–519.

Elliot, A. J., McGregor, H. A., & Gable, S. (1999). Achievement goals, study strategies,and exam performance: A meditational analysis. Journal of EducationalPsychology, 91, 549–563.

Elliot, A. J., & Murayama, K. (2008). On the measurement of achievement goals:Critique, illustration, and application. Journal of Educational Psychology, 100,613–628.

Elliot, A. J., Shell, M. M., Henry, K., & Maier, M. (2005). Achievement goals,performance contingencies, and performance attainment: An experimental test.Journal of Educational Psychology, 97, 630–640.


Espinosa, J. A., Lerch, F. J., & Kraut, R. (2004). Explicit vs. implicit coordinationmechanisms and task dependencies: One size does not fit all. In E. Salas & S. M.Fiore (Eds.), Team cognition: Understanding the factors that drive process andperformance (pp. 107–130). Washington, DC: APA.

Fisher, S. L., & Ford, J. K. (1998). Differential effects of learner effort and goalorientation on two learning outcomes. Personnel Psychology, 51, 397–420.

Fiske, S. T. (1998). Stereotyping, prejudice and discrimination. In D. T. Gilbert, S. T.Fiske, & G. Lindzey (Eds.), The handbook of social psychology (pp. 357–411).Boston: McGraw-Hill.

Ford, J. K., Smith, E. M., Weissbein, D. A., Gully, S. M., & Salas, E. (1998). Relationshipsof goal orientation, metacognitive activity, and practice strategies with learningoutcomes and transfer. Journal of Applied Psychology, 83, 218–233.

Giessner, S. R., & van Knippenberg, D. (2008). ‘‘License to fail’’: Goal definition,leader group prototypicality, and perceptions of leadership effectivenessafter leader failure. Organizational Behavior and Human Decision Processes,105, 14–35.

Gray, D. E. (2007). Facilitating management learning: Developing critical reflectionthrough reflective tools. Management Learning, 38, 495–527.

Hambrick, D. C., Cho, T. S., & Chen, M. J. (1996). The influence of top managementteam heterogeneity on firms’ competitive moves. Administrative ScienceQuarterly, 41, 659–684.

Harackiewicz, J. M., Barron, K. E., Tauer, J. M., Carter, S. M., & Elliot, A. J. (2000).Short-term and long-term consequences of achievement goals: Predictinginterest and performance over time. Journal of Educational Psychology, 92,316–330.

Harrison, D. A., Price, K. H., & Bell, M. P. (1998). Beyond relational demography:Time and the effect of surface- versus deep-level diversity on groupcohesiveness. Academy of Management Journal, 41, 96–107.

Harrison, D. A., & Klein, K. J. (2007). What’s the difference? Diversity constructs asseparation, variety, or disparity in organizations. Academy of ManagementReview, 32, 1199–1228.

Hirst, G., Mann, L., Bain, P., Pirola-Merlo, A., & Richter, A. (2004). Climbing thelearning curve: Leadership learning the relationship with adaptive decisionprocesses, leader role performance and project performance. LeadershipQuarterly, 15(3), 311–327.

Homan, A. C., van Knippenberg, D., Van Kleef, G. A., & De Dreu, C. K. W. (2007).Bridging faultlines by valuing diversity: The effects of diversity beliefs oninformation elaboration and performance in diverse work groups. Journal ofApplied Psychology, 92, 1189–1199.

Janis, I. (1972). Victims of groupthink. Boston: Houghton.Janssen, O., & van Yperen, N. W. (2004). Employees’ goal orientations, the quality of

leader–member exchange, and the outcomes of job performance and jobsatisfaction. Academy of Management Journal, 47, 368–384.

Jehn, K. A. (1997). A qualitative analysis of conflict types and dimensions inorganizational groups. Administrative Science Quarterly, 42, 530–557.

Kearney, E., & Gebert, D. (2009). Managing diversity and enhancing team outcomes:The promise of transformational leadership. Journal of Applied Psychology, 94,77–89.

Kooij-De Bode, H. J. M., van Knippenberg, D., & van Ginkel, W. P. (2008). Ethnicdiversity and distributed information in group decision making: Theimportance of information elaboration. Group Dynamics, 12, 307–320.

Kozlowski, S. W. J., & Bell, B. S. (2003). Work groups and teams in organizations. InW. C. Borman, D. R. Ilgen, & R. J. Klimoski (Eds.). Handbook of psychology:Industrial and organizational psychology (Vol. 12, pp. 333–375). London: Wiley.

Kozlowski, S. W. J., Gully, S. M., McHugh, P. P., Salas, E., & Cannon-Bowers, J. A.(1996). A dynamic theory of leadership and team effectiveness: Developmentaland task contingent leader roles. Research in Personnel and Human ResourcesManagement, 4, 253–305.

Laughlin, P. R., & Hollingshead, A. B. (1995). A theory of collective induction.Organizational Behavior and Human Decision Processes, 61, 94–107.

Laughlin, P. R., VanderStoep, S. W., & Hollingshead, A. B. (1991). Collectiveversus individual induction: Recognition of truth, rejection of error, andcollective information processing. Journal of Personality and Social Psychology,61, 50–67.

LePine, J. A. (2005). Adaptation of teams in response to unforeseen change: Effectsof goal difficulty and team composition in terms of cognitive ability and goalorientation. Journal of Applied Psychology, 90, 1153–1167.

Locke, E. A. (1986). Generalizing from laboratory to field: Ecological validity orabstraction of essential elements. In E. A. Locke (Ed.), Generalizing fromlaboratory to field settings (pp. 3–9). Lexington, MA: Lexington Books.

MacCallum, R. C., Browne, M. W., & Sugawara, H. M. (1996). Power analysis anddetermination of sample size for covariance structure modeling. PsychologicalMethods, 1, 130–149.

Marks, M. A., Sabella, M. J., Burke, C. S., & Zaccaro, S. J. (2002). The impact of cross-training on team effectiveness. Journal of Applied Psychology, 87, 3–13.

Marks, M. A., Zaccaro, S. J., & Mathieu, J. E. (2000). Performance implications ofleader briefings and team interaction training for team adaptation to novelenvironments. Journal of Applied Psychology, 85, 971–986.

Mathieu, J. E., Heffner, T. S., Goodwin, G. F., Salas, E., & Cannon-Bowers, J. A. (2000).The influence of shared mental models on team processes and performance.Journal of Applied Psychology, 85, 273–283.

Mathieu, J. E., Heffner, T. S., Goodwin, G. F., Cannon-bowers, J. A., & Salas, E. (2005).Scaling the quality of teammates’ mental models: Equifinality and normativecomparisons. Journal of Organizational Behavior, 26, 37–56.

Mesmer-Magnus, J. R., & DeChurch, L. A. (2009). Information sharing and teamperformance: A meta-analysis. Journal of Applied Psychology, 94, 535.

Mischel, W. (1977). The interaction of person and situation. In D. Magnusson & N. S.Endler (Eds.), Personality at the crossroads: Current issues in interactionalpsychology (pp. 333–352). Hillsdale, NJ: Erlbaum.

Mohammed, S., & Angell, L. (2003). Personality heterogeneity in teams: Whichdifferences make a difference for team performance? Small Group Research, 34,651–677.

Mohammed, S., & Ringseis, E. (2001). Cognitive diversity and consensus in groupdecision making: The role of inputs, processes, and outcomes. OrganizationalBehavior and Human Decision Processes, 85, 310–335.

Mohammed, S., Ferzandi, L., & Hamilton, K. (2010). Metaphor no more: A 15-yearreview of the team mental model construct. Journal of Management, 36(4),876–910.

Mook, D. G. (1983). In defense of external invalidity. American Psychologist, 38,379–387.

Müller, A., Herbig, B., & Petrovic, K. (2009). The explication of implicit teamknowledge and its supporting effect on team processes and technicalinnovations: An action regulation perspective on team reflexivity. Small GroupResearch, 40, 28–51.

Murayama, K., & Elliot, A. J. (2009). The joint influence of personal achievementgoals and classroom goal structures on achievement-related outcomes. Journalof Educational Psychology, 101, 432–447.

Newell, A., & Simon, H. A. (Eds.). (1972). Human problem solving. Engelwood Cliffs,NJ: Prentice Hall.

Okhuysen, G. A., & Eisenhardt, K. M. (2002). Integrating knowledge in groups: Howsimple formal interventions help. Organization Science, 13, 370–386.

Payne, S. C., Youngcourt, S. S., & Beaubien, J. M. (2007). A meta-analytic examinationof the goal orientation nomological net. Journal of Applied Psychology, 92,128–150.

Poortvliet, P. M., Janssen, O., Van Yperen, N. W., & Van de Vliert, W. (2007).Achievement goals and interpersonal behavior: How mastery and performancegoals shape information exchange. Personality and Social Psychology Bulletin, 33,1436–1447.

Porath, C. L., & Bateman, T. S. (2006). Self-regulation: From goal orientation to jobperformance. Journal of Applied Psychology, 91, 185–192.

Porter, C. O. L. H. (2005). Goal orientation: Effects on backing up behavior,performance, efficacy, and commitment in teams. Journal of Applied Psychology,90, 811–818.

Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies forassessing and comparing indirect effects in multiple mediator models. BehaviorResearch Methods, 40, 879–891.

Rentsch, J. R., & Klimoski, R. J. (2001). Why do great minds think alike? Antecedentsof team member schema agreement. Journal of Organizational Behavior, 22,107–120.

Rouse, W. B., & Morris, N. M. (1986). On looking into the black box: Prospects andlimits in the search for mental models. Psychological Bulletin, 100, 349–363.

Salas, E., & Fiore, S. M. (2004). Why team cognition? In E. Salas & S. M. Fiore (Eds.),Team cognition: Understanding the factors that drive process and performance(pp. 3–8). Washington, DC: APA.

Schippers, M. C., Den Hartog, D. N., Koopman, P. L., & Wienk, J. A. (2003). Diversityand team outcomes: The moderating effects of outcome interdependence andgroup longevity, and the mediating effect of reflexivity. Journal of OrganizationalBehavior, 24, 779–802.

Schippers, M. C., Den Hartog, D. N., & Koopman, P. L. (2007). Reflexivity in teams: Ameasure and correlates. Applied Psychology: An International Review, 56,189–211.

Schippers, M. C., Den Hartog, D. N., Koopman, P. L., & van Knippenberg, D. (2008).The role of transformational leadership in enhancing team reflexivity. HumanRelations, 61, 1593–1616.

Scholten, L., van Knippenberg, D., Nijstad, B. A., & De Dreu, C. K. W. (2007).Motivated information processing and group decision-making: Effects ofprocess accountability on information processing and decision quality. Journalof Experimental Social Psychology, 43, 539–552.

Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and nonexperimentalstudies: New procedures and recommendations. Psychological Methods, 7,422–445.

Steiner, I. D. (1972). Group processes and productivity. New York: Academic Press.Tett, R. P., & Burnett, D. D. (2003). A personality trait-based interactionist model of

job performance. Journal of Applied Psychology, 88, 500–517.Tjosvold, D., Tang, M. M. L., & West, M. (2004). Reflexivity for team innovation in

China: The contribution of goal interdependence. Group and OrganizationalManagement, 29, 540–559.

VandeWalle, D. (1997). Development and validation of a work domain goalorientation instrument. Educational and Psychological Measurement, 57,995–1015.

VandeWalle, D., Brown, S. P., Cron, B., & Slocum, J. (1999). The influence of goalorientation and self-regulation tactics on sales performance: A longitudinalfield test. Journal of Applied Psychology, 84, 249–259.

van Ginkel, W. P., Tindale, R. S., & van Knippenberg, D. (2009). Team reflexivity,development of shared task representations, and the use of distributedinformation in group decision making. Group Dynamics, 13, 265–287.

van Ginkel, W. P., & van Knippenberg, D. (2008). Group information elaboration andgroup decision making: The role of shared task representations. OrganizationalBehavior and Human Decision Processes, 105, 82–97.

van Ginkel, W. P., & van Knippenberg, D. (2009). Knowledge about the distributionof information and group decision making: When and why does it work?Organizational Behavior and Human Decision Processes, 108, 218–229.


van Ginkel, W. P., & van Knippenberg, D. (in press). Group leadership and sharedtask representations in decision-making groups. Leadership Quarterly.

van Knippenberg, D., De Dreu, C. K. W., & Homan, A. C. (2004). Work group diversityand group performance: An integrative model and research agenda. Journal ofApplied Psychology, 89, 1008–1022.

van Knippenberg, D., & Schippers, M. C. (2007). Work group diversity. Annual Reviewof Psychology, 58, 515–541.

Weingart, L. R. (1997). How did they do that? The ways and means of studyinggroup process. Research in Organizational Behavior, 19, 189–239.

West, M. A. (1996). Reflexivity and work group effectiveness: A conceptualintegration. In M. A. West (Ed.), Handbook of work group psychology(pp. 555–579). Chichester, UK: Wiley.

West, M. A., Garrod, S., & Carletta, J. (1997). Group decision-making and effectiveness:Unexplored boundaries. In C. L. Cooper & S. E. Jackson (Eds.), Creating tomorrow’sorganizations: A handbook for future research in organizational behavior(pp. 293–316). Chichester, UK: Wiley.

Williams, K. Y., & O’Reilly, C. A. (1998). Demography and diversity in organizations:A review of 40 years of research. Research in Organizational Behavior, 20, 77–140.

Documents

Diversity in goal orientation, team reflexivity, and team performance