Group leadership and shared task representations in decision making groups

The Leadership Quarterly 23 (2012) 94–106

Contents lists available at SciVerse ScienceDirect

The Leadership Quarterly journal homepage: www.elsevier.com/locate/leaqua

Group leadership and shared task representations in decision making groups Wendy P. van Ginkel ⁎, Daan van Knippenberg Rotterdam School of Management, Erasmus University Rotterdam, PO Box 1738, 3000 DR Rotterdam, The Netherlands

a r t i c l e

i n f o

Available online 23 November 2011 Keywords: Team leadership Shared cognition Shared task representations Shared mental models Information elaboration Hidden profile Group decision making

a b s t r a c t The development of adequate shared understanding of the task is of critical importance to group functioning. Group leaders play an important role in this respect, as a key function of leadership is to shape group members' understanding of their job. In the present study we focus on decision making groups with distributed information and examine how group leaders shape members' mental representations of the group decision task through leadership behavior rooted in their own representations of the task. We propose that the extent to which the group leader has task representations that emphasize information exchange and integration affects group members' task representations, group information elaboration, and decision quality. We tested these hypotheses in an experiment (N = 94 three-person groups) in which we manipulated whether a group leader was present and whether this leader held representations emphasizing information elaboration. Results supported the hypotheses, and suggest that team leaders may play an important role in creating a socially shared understanding of team tasks. © 2011 Elsevier Inc. All rights reserved.

1. Introduction It has long been recognized that a key role of leadership is to develop employees' understanding of their task and to provide guidance and coaching in task performance (Fiedler, 1967; House, 1971; Yukl, 1998). Yet, as team-based structures in organizations are becoming more and more prevalent and work groups and teams have become the basic building blocks of organizations (Ilgen, Hollenbeck, Johnson, & Jundt, 2005), it becomes increasingly apparent that leadership research has largely neglected leadership's role in developing team members' understanding of team work — their understanding of the team, the team task, and the requirements of working interactively and interdependently as a team (Kozlowski, 1998; Zaccaro, Rittman, & Marks, 2001). This is in sharp contrast with accumulating evidence for the importance of a shared and task-adaptive understanding of the team task (i.e., shared mental models, shared task representations; Cannon-Bowers, Salas, & Converse, 1993; Kerr & Tindale, 2004). Moreover, there are many reasons to believe that team leaders can be of particular importance in creating such a shared and task-adaptive understanding (Hackman & Wageman, 2005; Hill & Levenhagen, 1995; Kozlowski, Gully, Salas, & CannonBowers, 1996; Weick, 2001). This then points to the question of whether and how leaders can influence group member task representations as an issue of key importance to our understanding of leadership of groups and teams. In the present study, we address this issue by examining how group leaders can develop shared and task-adaptive understanding in groups. One of the core reasons to organize work in groups and teams is that ‘two heads know more than one’: Group performance may benefit from the different knowledge and perspectives of group members. This requires, however, that this distributed knowledge is exchanged, discussed, and integrated, and research in group decision making with distributed information in particular has highlighted that such a process of group information elaboration (van Knippenberg, De Dreu, & Homan, 2004) often only takes place to a modest degree (Stasser, 1999; Wittenbaum & Stasser, 1996). Here, then, lies an important challenge for group leadership, and moreover a challenge that recent research has linked to group members' task representations: Groups' use of

⁎ Corresponding author. E-mail addresses: [email protected] (W.P. van Ginkel), [email protected] (D. van Knippenberg). 1048-9843/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.leaqua.2011.11.008

W.P. van Ginkel, D. van Knippenberg / The Leadership Quarterly 23 (2012) 94–106

95

distributed information is often impeded by an under-appreciation of the information processing requirements of the task, and groups with distributed information perform substantially better when they develop a shared understanding of the task that emphasizes the need for information elaboration (van Ginkel & van Knippenberg, 2008, 2009). Accordingly, we focus on the role of group leadership in creating shared and task-adaptive task representations in decision making groups with distributed information to zoom in on leaders' role in this core process in the performance of groups and teams. 1.1. Shared task representations and group information elaboration Before we focus on the role of group leadership, we first briefly outline the importance and problematic nature of the use of distributed information in groups, and the role of shared task representations in this respect. As we already noted, a key assumption underlying the team-based organization of work is that groups may benefit from the knowledge distributed among their membership (Ilgen et al., 2005). Nowhere perhaps is this more evident than in context of decision-making groups where a key advantage of groups is seen to lie in the decision-relevant informational resources available to them (Cohen & Levinthal, 1990; Kozlowski & Bell, 2003; Madhavan & Grover, 1998; Tindale, Kameda, & Hinsz, 2001). While the advantage of distributed information is clear in theory, research has shown that groups' informational resources are often not used to the full — distributed information is often not sufficiently exchanged and processed (Gigone & Hastie, 1993; Gruenfeld, Mannix, Williams, & Neale, 1996; Scholten, van Knippenberg, Nijstad, & De Dreu, 2007; Stasser & Titus, 1985; Winquist & Larson, 1998; cf. van Knippenberg & Schippers, 2007). Accordingly, motivated by the notion that groups may reach higher-quality decisions when they are able to exchange and integrate the diversity of information and perspectives held by their members (De Dreu, Nijstad, & van Knippenberg, 2008; Hinsz, Tindale, & Vollrath, 1997; van Knippenberg, van Knippenberg, De Cremer, & Hogg, 2004), research in group decision making has invested substantial effort in understanding groups' use of distributed information. Recent work by van Ginkel and van Knippenberg (2008, 2009) (van Ginkel, Tindale, & van Knippenberg, 2009; cf. Kooij-De Bode, van Knippenberg, & van Ginkel, 2008) identified group members' understanding of the group task – their cognitive task representations – as a fundamental influence on groups' use of their informational resources. People's engagement with a task is guided by their subjective understanding of how to perform the task — their task representations (Newell & Simon, 1972; Tindale, Smith, Thomas, Filkins, & Sheffey, 1996). Accordingly, different people may try to perform the same task in different ways contingent on their task representations, and some task representations are more task-adaptive – stimulate more effective task performance – than others. From this perspective an important and fundamental factor in human performance are taskadaptive task representations. Insights in shared cognition in teams show that this observation holds not only for individual performance, but also for group performance (e.g., Tindale et al., 1996), and van Ginkel and van Knippenberg linked differences in shared task representations to differences in groups' use of distributed information. Their line of research shows that group members often fail to recognize the importance of the elaboration (i.e., exchange and integration; van Knippenberg et al., 2004) of distributed information. Rather they seem preoccupied with reaching an agreement at the expense of information elaboration. van Ginkel and van Knippenberg (2008), for instance, show in an experiment that when groups were stimulated through experimental instructions to form task representations emphasizing information elaboration, they discussed more distributed information and made higher-quality decision than groups that were not stimulated to form such task representations. In the present study, we use these recent insights to analyze the role group leadership may play in stimulating groups to use their distributed information in decision making. Specifically, we propose that an important role of group leadership is to influence and change group members' task representations, rendering them more attuned to the importance of information elaboration. 1.2. Group leadership and shared task representations Although a number of researchers have pointed to the potential role of leaders in the development of (shared) task cognition (Kozlowski, 1998; Kozlowski et al., 1996; Zaccaro et al., 2001; cf. Marks, Zaccaro, & Mathieu, 2000), empirical analyses have been lagging behind. While the vast majority of work groups and teams function within some kind of leadership structure, research on shared cognition more or less exclusively focused on leaderless groups (Mathieu, Heffner, Goodwin, Cannon-bowers, & Salas, 2005; Mathieu et al., 2000; Marks et al., 2000). Moreover, studies examining the development of task-appropriate shared cognition more generally are rare (Mohammed & Ringseis, 2001; van Ginkel et al., 2009). As a consequence, little is still known about the ways in which group leaders can facilitate the development of shared cognition. Even though empirical evidence concerning the role of leadership in the development of shared task representations thus is scarce, we can identify ways in which group leaders may influence members' task representations based on an integration of theory in leadership and shared cognition. To ground this analysis firmly in the current focus on leaders' role in engendering shared task representations that foster information elaboration, we center our discussion around the two alternative task representations highlighted by van Ginkel and van Knippenberg (2008, 2009) (van Ginkel et al., 2009): task representations emphasizing elaboration and task representations emphasizing a search for common ground. While the former clearly are the more task-adaptive representations in groups with distributed information, the latter seem to be the more common ones. In the context of groups dealing with distributed information, leadership's role in shaping task-adaptive representations can thus be seen as engendering representations emphasizing elaboration rather than common ground. Perhaps the most direct way for leaders to influence members' task representations is by giving them directions on how to approach the task (Marks et al., 2000; cf. van Ginkel & van Knippenberg, 2008). Although the emphasis in leadership research

96

W.P. van Ginkel, D. van Knippenberg / The Leadership Quarterly 23 (2012) 94–106

in the last decades primarily has been on leaders' effectiveness in mobilizing and motivating followers (e.g., Bass & Avolio, 1993; Conger & Kanungo, 1987; Lord & Brown, 2004; Shamir, House, & Arthur, 1993; van Knippenberg et al., 2004), task-directed leadership behaviors are still regarded as an important and necessary aspect of leadership. Several researchers have identified clarifying task requirements and making sense of ambiguous situations as important leader roles (Hackman & Wageman, 2005; House, 1971; Reicher & Hopkins, 2001; van Knippenberg & Hogg, 2003; Yukl, 1998). That is, group leaders should give directions on how to approach the group task to develop group members' understanding of the task at hand. Indeed, by virtue of their leadership role, group leaders are both expected and entitled to exert such influence (cf. Berger, Wagner, & Zelditch, 1985; French & Raven, 1959; van Knippenberg, van Knippenberg, & van Dijk, 2000). In fostering task representations emphasizing elaboration, one way for leaders to do so is thus to explain the importance of information elaboration and to actively guide the information elaboration process by instructing group members to exchange, discuss, and integrate information, developing their understanding of the process and its consequences. If the leader would instead advocate the often more common notion that group members need to search for common ground in order to reach agreement, the leader would promote less task-appropriate representations. Group leaders are likely to influence members' representations also by means of their own approach to task performance. Leaders are likely to convey cues about how to work together on a task through role-modeling certain behaviors: By displaying a certain behavior themselves, leaders may invite followers to behave likewise. The key process underlying the influence of role modeling is social learning (Bandura, 1977). By observing the role model displaying the behavior as well as the behavior's effects, observers come to understand the appropriateness and effects of the behavior – that is, develop cognitive representations identifying the behavior as appropriate – and are invited to behave likewise. Analyses of charismatic leadership have confirmed that role-modeling desired behaviors may indeed be a powerful way in which leaders influence followers (Conger & Kanungo, 1987). Leaders may thus foster task representations emphasizing elaboration by engaging in the process of elaboration themselves, exchanging and discussing information, repeating information (cf. Larson, Christensen, Abbott, & Franz, 1996), and soliciting information and clarification from group members. By doing so, they convey the value of information elaboration as well as clarify the actions required to successfully engage in elaboration. In contrast, leaders who are more focused on the exchange of preferences (cf. Hastie & Pennington, 1991; van Ginkel & van Knippenberg, 2008) or who for instance publicly change their mind in order to reach agreement or ask group members to do so would invite or bolster task representations emphasizing common ground. The guidance leaders give to group members in task performance is likely follow for a great part from their mental representations of the task. Similarly, their own engagement with the task (cf. role-modeling) is likely to be guided by their task representations. That is, group leaders are likely to both voice and enact their own task representations in group task performance, and thus shape group members' task representations in the process. In group interaction group members are thus likely to develop task representations that are similar to their leader's task representations. This means that group leader directions and rolemodeling will only engender group member task representations emphasizing elaboration to the extent that the leaders' own task representations emphasize elaboration. The effectiveness of group leaders in engendering group information elaboration and high-quality decisions therefore should be critically contingent on the leader's task representations. Whereas leaders with task representations that emphasize the elaboration of task-relevant information may substantially improve groups' use of distributed information, leader task representations that mainly endorse finding common ground would, if anything, actually harm rather than benefit group information use and decision making performance. In sum then, we propose the following hypotheses (also see Fig. 1). Hypothesis 1a. Groups with a leader with task representations that emphasize information elaboration elaborate more on decision-relevant information than groups with a leader with representations that emphasize seeking common ground.

Leader behavior focusing on elaboration

Members’ representations focusing on elaboration

Information elaboration

Group leader representations

Leader behavior focusing on common ground

Group decision quality

Members’ representations focusing on common ground

Fig. 1. Group decision quality as a function of information elaboration, task representations, group leader behavior, and group leader representations.

W.P. van Ginkel, D. van Knippenberg / The Leadership Quarterly 23 (2012) 94–106

97

Hypothesis 1b. Groups with a leader with task representations that emphasize information elaboration reach higher-quality decisions than groups with a leader with representations that emphasize seeking common ground. Our analysis suggests that it is the leadership role that renders the leaders' task representations particularly ‘visible’ to group members and particularly influential in affecting group members' task representations. That is, the influence of an individual's task representations on other group members' task representations, group process, and performance, should be stronger when the individual is the group leader than when the individual is not. To substantiate this proposition, we also included a comparison with a situation in which task representations emphasizing elaboration where held by one of the group members rather than the group leader. Hypothesis 2a. Groups with a leader with task representations that emphasize information elaboration elaborate more on decision-relevant information than groups without a leader. Hypothesis 2b. Groups with a leader with task representations that emphasize information elaboration reach higher quality decisions than groups without a group leader. Our analysis also highlights three variables that sequentially mediate (i.e., the one leading to the other in sequence) the effects of group leader task representations on group decision making performance. Hypothesis 3. The effect of group leaders' task representations on decision making performance is mediated sequentially by (a) leader behavior (advocating elaboration or the search for common ground), (b) member task representations (emphasizing elaboration or the search for common ground), and (c) group information elaboration. We tested these hypotheses in an experiment with small interactive groups with distributed information (Gruenfeld et al., 1996; Stasser & Titus, 1985; van Ginkel & van Knippenberg, 2008). We manipulated whether or not a group leader was present and whether the group leader held task representations emphasizing information elaboration or task representations emphasizing finding common ground. This experimental set-up not only allowed us to reach conclusions about causality, but also to rely on behavioral coding of audio–video data to assess leadership behavior and group information elaboration. Audio–video data are more reliable and provide a richer source of information about group processes than the self-report data commonly used in field research (Weingart, 1997), and should thus increase the confidence in the conclusions based on our findings. 2. Method 2.1. Design and participants The experiment had a one-factor design with three levels (group leader with task representations that emphasize elaboration, group leader with task representations that emphasize finding common ground, no group leader). A total of 285 business and economy students were assigned to 95 three-person groups. One group was excluded from the analyses, because the person randomly selected to be the group leader indicated not to be willing to fulfill this role, leaving 282 participants (90 women and 192 men) and 94 groups. Groups were randomly assigned to one of the three conditions (31 in the leader elaboration condition, 31 in the leader seeking common ground condition, and 32 in the no-leader condition). For four of these groups, due to technical problems no audio–video data to assess elaboration and leadership behavior were available. Because there are no reasons to expect these groups to differ from the other groups, the groups were kept in the analyses for the analyses that did not involve audio– video data (listwise deletion was used for the analyses). The majority of participants were business administration students (87%). Their mean age was 21.3 (SD = 3.6). For their participation they received a compensation of 10 euro (approximately 13 USD). 2.2. Experimental task The task used in this study was a cooperative decision making task inspired by the Towers Market task (Weingart, Bennet, & Brett, 1993) and adapted by van Ginkel and van Knippenberg (2008) to study group decision making with distributed information. The task concerns the organization of a small market center that contains a bakery, a florist, and a vegetable market. Group members are told that they will function as an independent advisory committee that is to aid the three stores in making three interrelated decisions about the temperature for the market center, the division of maintenance costs between three stores, and the organization of store space. To do this all participants were given information on the preferences of three stores and on the relative importance of the three issues to the stores. In addition they were told to take the interests of all three stores into account. Group members received background information about the three stores. Following prior research in group decision making, part of the decision-relevant information was given to all group members and part of the information was given to only one of the group members (Gruenfeld et al., 1996; Stasser & Titus, 1985). Items of information necessary to make high-quality decisions were distributed over group members, so that for groups had to exchange these items and integrate them with the other

98

W.P. van Ginkel, D. van Knippenberg / The Leadership Quarterly 23 (2012) 94–106

information available to the group to arrive at an optimal decision. For each decision issue there were always three items of information that were crucial for reaching an optimal decision. Each of these was uniquely assigned to one of the group members. For each decision issue group members received this crucial, unshared information on a different store (e.g., group member 1 received unshared information about the bakery on item 1, about the florist on item 2, etc., while group member 2 received crucial information about the florist on item 1 and about the vegetable market on item 2, etc.). To make the task more complex, in addition to information necessary to make a correct decision some irrelevant information about the three stores was also included (this information was always given to all members). As is commonly done in the present paradigm, groups were informed that the information each member received might differ (Stasser & Titus, 1985). For each issue, groups could choose from a limited number of options. Based on all available information a hierarchy in the quality of the decision options existed (i.e., some combinations of decision options served the interests of all stores better than others). 2.3. Experimental manipulations 2.3.1. Group leadership Group leaders were randomly selected from the group. The person selected to be the group leader and the other group members were informed about this by the experimenter. The group and the designated leader were told that the leader would participate together with the other members in the group task and that he or she would be responsible for the way in which the group worked together. The leader's main task was to ensure the decision making process proceeded in an optimal fashion and to take action whenever he or she believed this was necessary. While the leader was given instructions with respect to the specific leader role, the leader did not receive more information on the group task than any of the group members. To make sure that the designated leader was capable of fulfilling the role, he or she was given additional instructions. This instruction included behavioral descriptions of ways in which the leader could affect the group process (e.g., ask team members for contributions, encourage more quiet members to contribute, keep track of the time, etc.). The designated leader was informed that if he or she felt that he or she would not be able to fulfill the role, he or she could inform the experimenter and the group could simply continue without a leader. The person selected as a leader was ensured that this would not be a problem or hinder the experiment, because for other purposes groups without a leader were also necessary (only one person indicated that he did not want to be a leader). Finally, before groups started with the decision making task, the experimenter talked to the group leader in private to answer any questions he or she might have and subsequently reminded the group as a whole about the fact that one person was selected to fulfill the group leader role. 2.3.2. Task representations To manipulate task representations, written instructions were used following earlier work by van Ginkel and van Knippenberg (2008, 2009) (van Ginkel et al., 2009). Task instructions were designed such that they suggested different ways to approach the group task — different task representations. The representations were based on research in group decision making as described in the Introduction section. One instruction focused on the importance of the exchange of individual members' information and critical discussion of this information. It outlined how discussing each member's input and avoiding prematurely reaching consensus before all information is discussed is important to reach high-quality decisions. Another instruction focused more on a seeking common ground approach to group-decision making. It discussed how not being too rigid in your choice and being willing to change your mind or to make compromises for the sake of a group consensus is important to reach high-quality decisions. Rather than only providing the leader with a task representation, all members' were provided with task representations to avoid confounds in the experimental design. In this way, the task representations available to the group as a whole where identical in all conditions. The only differences were whether or not there was a group leader, and if so, whether the leader was assigned task representations emphasizing elaboration or common ground. As a consequence, all differences between conditions can be attributed to the leadership manipulations and not to the assignment of task representations per se. (Note that, if anything, this yields a more conservative test of our hypotheses, because when members already have clearly articulated task representations they may be less likely to adopt alternative representations adopted by the leader than when they do not yet have clearly defined ideas about the task.) Two individuals were always presented with representations emphasizing finding common ground and one person with representations emphasizing elaboration. In the group leader elaboration condition, the group member with elaboration representations was selected as the group leader. In the group leader common ground condition, one of the members with common ground representations was selected as the leader. We opted to assign two members representations emphasizing common ground and only one representations emphasizing elaboration, because of the evidence that members of decision making groups more often hold common ground than elaboration representations (van Ginkel & van Knippenberg, 2008, 2009) — indeed, our research was in part inspired by the observation that a key role of group leadership in groups with distributed information is to shift group members understanding of the task from an emphasis on seeking common ground to an emphasis on information elaboration. This implies that the leader elaboration condition differs from the leader common ground condition not only in the representation the leader has, but also in the distribution of representations among the non-leader members (two members have similar common ground representations in the leader elaboration condition, and one member has common ground representations and the other elaboration representations in the leader common ground condition). Again, if anything this results in a conservative test of our hypotheses, because the sharedness of the two members' common ground representations in the leader elaboration condition

W.P. van Ginkel, D. van Knippenberg / The Leadership Quarterly 23 (2012) 94–106

Leader - elaboration condition

Common ground

Member Elaboration

Leader

99

Leader – common ground condition

Common ground

Common ground

Member

Member

Elaboration

Member Common ground

Leader

Fig. 2. Graphic representation of task representations held by members and leader within the leadership conditions.

may only make it less likely that members in this condition converge on the leaders' elaboration representation. See Fig. 2 for a graphic representation of this manipulation. 2.4. Measures Task representations were measured using questionnaires. Participants were asked to rate the extent to which they believed several actions would either harm or help performance on the group decision task on a 7-point scale, ranging from − 3 (harms performance considerably) to 3 (helps performance considerably). This way of measuring mental representations has been successfully employed before (Webber, Chen, Payne, Marsh, & Zaccaro, 2000). Seven statements were used to assess task representations emphasizing information elaboration. A few examples of statements are “basing a decision on all available information”, “continue with discussing information even when we all agreed”, “exchanging information”, and “end the discussion when we all agreed” (R) (α = .60). To measure task representations for seeking common ground eight statements were used. A few examples of statements are “quickly trying to reach an agreement”, “changing your mind with the purpose of reaching an agreement”, “making sacrifices with the purpose of reaching an agreement”, and “making sure all parties make some compromises” (α = .63). Leader behavior was measured using audio–video coding. Two leader behavior measures were developed. One set of behaviors focuses on advocating information elaboration, while the other set of behavior focuses on advocating seeking commonground. The behaviors were chosen after carefully examination of all audio–video recordings and were based on a conceptual analysis of the task representation constructs (e.g., van Ginkel & van Knippenberg, 2008). Two coders blind to the experimental conditions counted how often the leadership behaviors were displayed by both the group leader and the group members (i.e., in order to establish that these behaviors were displayed primarily by leaders rather than by all members). For behavior focusing on information elaboration we counted how often leaders/members asked others for information, repeated information, and made remarks about the value of information elaboration. A focus on common ground involves being preoccupied with reaching agreement, which is often expressed in a preoccupation with pooling group members preferences rather than pooling the underlying information (Hastie & Pennington, 1991; van Ginkel & van Knippenberg, 2008), For behavior focusing on advocating seeking common ground coders counted how often leaders/members asked for preferences, asked members about agreement, and made remarks about the value of seeking common ground (κ = .69). Behaviors were coded per decision issue and a mean score over the three decision issues was computed for both behavioral categories. Audio–video recordings were also used to assess group information elaboration following a coding scheme developed by van Ginkel and van Knippenberg (2008) anchored with specific behavioral standards observed in the audio–video recordings (cf. Weingart, 1997) and culminating in an overall elaboration score on a 7-point scale (also see van Ginkel et al., 2009; van Ginkel & van Knippenberg, 2009; van Knippenberg et al., 2010). Because information elaboration entails both sharing information and discussing and integrating it (van Knippenberg et al., 2004), high levels of information elaboration can only be attained when group members engage in not only the sharing, but also the discussion and integration of information. This is reflected in our coding scheme in which a lower score is given when overall less elaboration is displayed. Two coders blind to experimental conditions rated group information elaboration (κ = .78). A score of “1” was given when information was completely ignored by all three group members and the group immediately started with exchanging preferences. A score of “2” was given when one member mentioned an unshared item of information, but none of the other members reacted to it (either by saying something or by nodding or by clearly looking at the person that mentioned it) or used the item in making a decision. A score of “3” was given when one member mentioned an unshared item of information and at least one other member reacted to it, but after this the group still failed to integrate it with the other information. A score of “4” was given when one unshared piece of information was mentioned by at least one group member, with the other members clearly reacting to the information and integrating it (for instance by asking a question about it, by combining it

100

W.P. van Ginkel, D. van Knippenberg / The Leadership Quarterly 23 (2012) 94–106

with another piece of information, or by drawing a conclusion from it with regard to what the best decision option would be in light of this information) or when two pieces were clearly mentioned by at least one member, but when the group as a whole failed to integrate both items of information in coming to a decision. A score of “5” was given when one unshared item of information got fully discussed by all members and integrated with other information and at least one other unshared item of information was clearly mentioned, however without the mentioning of the unshared information influencing the use of that item of information by the group as a whole. A score of “6” was given when at least two pieces of unshared information were fully discussed by the whole group and integrated with the other information. A score of “7” was given when all three crucial unshared items of information were clearly discussed by at least two of the three members and clearly integrated. 3. Results 3.1. Preliminary analysis To determine the level of analysis mean awg(1) values were calculated for each variable (Brown & Hauenstein, 2005). Cut-off scores of .60 or .70 have been reported, with scores above the cut-off point suggesting analysis at the group level (Brown & Hauenstein, 2005). The mean awg(1) value for task representations emphasizing information elaboration and task representations emphasizing seeking common ground was both above these cut-off points (respectively .91 and .90). These variables were therefore analyzed at the group-level. Furthermore, because we were interested in how group members' task representations were influenced by their leader, in the leader conditions we omitted the leader's representations from the aggregation, because including the leader's representation would make it impossible to determine whether any increase or decrease would occur because of the influence the leader has on the group or because of the sheer level of the leaders' own task representations. 3.2. Manipulation checks To check the manipulation of task representations participants were asked to briefly describe what they believed would be the best way to go about the group task. These descriptions were then compared with the instructions that had been presented to them by two coders blind to experimental conditions (κ = .91). Of all participants who were asked to describe their ideas about the task 93% reported task representations similar to the instruction they had been given. The remainder of the participants either did not write down anything, described another representation than the one they had been presented with, or wrote down something completely different. Because the task representations of the group leaders were of particular importance in the present study, we also looked at the percentage of leaders that correctly described the task representations provided to them. Here we found that only two of the 62 group leaders (both in the information elaboration condition) failed to correctly describe the representations provided to them. 1 This implies that the manipulation of task representations was successful. 3.3. Leader behavior To be able to compare leadership behavior over the three conditions, we first aggregated leader behavior over groups, not taking into account which member displayed the leader behavior. We then used planned contrast analyses to compare the three conditions. Results showed that in the leader elaboration condition more behavior to encourage elaboration was displayed than in the leader common ground, t(87) = 9.69, p b .001, or the control condition, t(87) = 9.75, p b .001. No differences were found between the leader common ground and the control condition, t(87) = −.10, p = .92. In addition, we found that more behavior encouraging common ground was displayed in the leader common ground condition than the leader elaboration condition, t(87) = −7.01, p b .001, or the control condition, t(87) = − 5.80, p b .001, while there were no differences between the leader elaboration and the control condition, t(87) = −1.13, p = .26. To test whether it really were the group leaders rather than the non-leader members that engaged in the specific leader behaviors, we examined how often the leader versus the non-leader members engaged in the behavior using ANOVA. Results showed that group leaders indeed engaged in more behaviors encouraging information elaboration, F(1,182) = 67.26, p b .001, and more behaviors encouraging seeking common ground, F(1,182) = 87.47, p b .001, than non-leader members. 3.4. Task representations First, group members' task representations were aggregated by simply taking the mean (note that the awgj value warranted group-level analysis) over group members' representations. In the two group leader conditions only the task representations of the two members (not including the leader) were aggregated. To keep the number of members whose task representations were aggregated in the control condition similar to the number in the two leader conditions, here we also conducted the analyses over only two of the three group members (the two members not sitting in the chair in which the leader sat in the leader conditions). Differences in group members' representations emphasizing information elaboration and seeking common ground between the three experimental conditions were tested. As expected, groups in the leader elaboration condition scored higher on 1 We examined whether our findings were in any way affected by these two leaders that failed to correctly describe the representation by leaving the two groups out of our analysis. Because this did not affect the results, we chose to report the results for all groups, including these two groups.

W.P. van Ginkel, D. van Knippenberg / The Leadership Quarterly 23 (2012) 94–106

101

representations emphasizing information elaboration than groups in the leader common ground condition, t(91) = 3.09, p b .01, and groups in the control condition, t(91) = 2.68, p b .01. Moreover, groups in the leader elaboration condition also scored significantly lower on task representations emphasizing seeking common ground than groups in the leader common ground condition, t(91) = −2.98, p b .05, and groups in the control condition, t(91) = −2.57, p b .05. No significant differences were found between the leader seeking common ground condition and the control condition on either task representations emphasizing information elaboration, t(91) = .44, p = .66, or task representations emphasizing seeking common ground, t(91) = −.43, p = .67. See Table 1 for means, standard deviations, and confidence intervals. 3.5. Group information elaboration Planned contrasts were computed to test differences between conditions in information elaboration. Hypothesis 1a predicted that groups in the leader elaboration condition elaborate more on information than groups with a leader with task representations emphasizing seeking common ground. In line with Hypothesis 1b, groups in the leader elaboration condition scored higher on group information elaboration than groups in the leader common ground condition, t(87) = 4.64, p b .01. Hypothesis 2a predicted that groups in the leader elaboration condition elaborate more on information than groups without a leader. Results showed that indeed groups elaborated more in the leader elaboration condition than in the control condition, t(87) = 5.17, p b .01. This supports Hypothesis 2a. Finally, no significant differences in information elaboration were found between groups in the control condition and groups in the leader common ground condition, t(87) = −.57, p = .57. 3.6. Group decision making performance Planned contrast analysis showed that groups in the leader elaboration condition performed better than groups in the leader common ground condition, t(91) = 4.68, p b .01. This is in line with Hypothesis 1b. In addition, groups in the leader elaboration condition performed better than groups in the control condition, t(91) = 2.79, p b .01, which supports Hypothesis 2b. Finally, groups without a leader performed marginally better than groups in the leader common ground condition, t(91) = 1.93, p = .06. 3.7. Mediation analysis Hypothesis 3 predicted that the effects of the leadership manipulation on performance are sequentially mediated by (a) group leader behavior, (b) group members' shared representations of the task, and (c) information elaboration. To test Hypothesis 3 we used path analysis, in which we tested all the mediated relations suggested by our causal model by means of a combination of regression analyses and bootstrapping (see Fig. 1) (Edwards & Lambert, 2007). Each of the proposed indirect effects was computed relying on bootstrap samples and constructing a bias-corrected confidence interval. Bootstrap samples were created by drawing 1000 random samples with replacement from the full sample. Two dummy variables representing the differences between the three experimental conditions were created. Because from a theoretical point of view the comparison with the leader elaboration condition seems the most interesting we will report the findings for the comparison with this condition. Note that we did not predict significant indirect effects for the effect of the dummy variable representing the difference between the leader elaboration and the control condition on decision quality, as we did not predict differences on leader behavior focusing on common ground between these two conditions. As can be seen in Table 2 all predicted indirect effects were significant. In addition, we examined the total causal model. Because we chose to focus on the comparison of both the control and the leader common ground condition with the leader elaboration condition, this implies that there are three interesting causal paths to test. That is, 1) the path from the dummy variables representing the difference between the leader elaboration condition and the control condition through leader behavior encouraging elaboration, through representations emphasizing elaboration, through information elaboration on decision quality, 2) the path from the dummy variable representing the difference between the leader elaboration condition and the leader common-ground condition, through leader behavior encouraging elaboration, through representations emphasizing elaboration, through information elaboration on decision quality, and 3) the path from the dummy variable representing the difference between the leader elaboration condition and the leader common-ground Table 1 Means, standard deviations, and 95% confidence intervals (CI) for task representations emphasizing information elaboration (TR elaboration), task representation emphasizing seeking common ground (TR common ground), information elaboration, and performance per experimental condition. Leader elaboration M

SD

No leader CI

M

SD

CI

Lower

.44a .49a 5.76a 8.71a

.56 .48 1.05 .94

.23 .31 5.36 8.37

M

SD

Lower

Upper TR elaboration TR common ground Information elaboration Performance score

Leader agreement-seeking

Upper .64 .66 6.15 9.05

.07 b .77b 4.11b 7.81b

.53 .40 1.30 1.40

−.12 .62 3.63 7.31

Note. Different subscripts within rows indicate that values differ from each other at the p b .05 level.

CI Lower Upper

.26 .92 4.60 8.32

.01b .82b 4.29b 7.19b

.55 .42 1.33 1.42

−.19 .66 3.80 6.67

.21 .97 4.78 7.72

102

W.P. van Ginkel, D. van Knippenberg / The Leadership Quarterly 23 (2012) 94–106

Table 2 Overview of direct and indirect effect. Unstandardized regression coefficients. Direct effects

Dummy1 Dummy2 Leader behavior elaboration Leader behavior common-ground Representations elaboration Representations common-ground Information elaboration

Leader behavior elab (LBE)

Leader behavior common-ground (LBC)

Representations elab (TRE)

Representations common-ground (TRC)

Information elaboration (IE)

Decision quality (DQ)

− 1.03⁎ − 1.04⁎

.10 .57⁎

−.37⁎⁎ −.43⁎⁎ .19⁎ −.32⁎

.28⁎⁎ .33⁎ −.15⁎ .27⁎

− 1.63⁎⁎ − 1.48⁎⁎ 1.57⁎⁎ −.89⁎ .84⁎⁎ − 1.21⁎⁎

−.90⁎⁎ − 1.42⁎⁎ 1.17⁎⁎ − 1.43⁎⁎ .97⁎⁎ − 1.13⁎⁎ .64⁎⁎

Indirect effects 95% bias-corrected confidence interval

Dummy 1 on TRE through LBE Dummy 2 on TRE through LBE Dummy 2 on TRC through LBC LBE on IE through TRE LBC on IE through TRC TRE on DQ through IE TRC on DQ through IE

−.20 −.20 .15 .18 −.32 .54 −.78

Lower bound

Upper bound

−.39 −.39 .01 .01 −.67 .23 − 1.19

−.03 −.03 .30 .34 −.03 .84 −.43

Dummy 1 represents the difference between the leader elaboration condition and the control condition and dummy 2 represents the difference between the leader common ground and leader elaboration condition. ⁎ p b .05. ⁎⁎ p b .01.

condition, through leader behavior encouraging seeking common ground, through representations emphasizing common ground, through information elaboration on decision quality. Note that we do not predict that leader behavior emphasizing common ground mediates in the path between the dummy variable representing the difference between the leader elaboration condition and the control condition and decision quality, because we do not expect differences in leader behavior emphasizing commonground between these experimental conditions (see above). Using the regression coefficients presented in Table 2 we computed the indirect effect from the dummy variable representing the difference between the leader elaboration condition and the control condition, through leader behavior encouraging elaboration, through representations emphasizing elaboration, through information elaboration on decision quality by taking the product of all regression weights which was −.11. The confidence interval excluded zero (−.24, −.01), indicating significant mediation. In similar means we computed the indirect effect from the dummy variable representing the difference between the leader elaboration condition and the leader common-ground condition, through leader behavior encouraging elaboration, through representations emphasizing elaboration, through information elaboration on decision quality, which also was −.11. The confidence interval again excluded zero (−.23, −.01), indicating significant mediation. Finally, the indirect effect from the dummy variable representing the difference between the leader elaboration condition and the leader common-ground condition, through leader behavior encouraging seeking common ground, through representations emphasizing common ground, through information elaboration on decision quality was −.02. The confidence interval excluded zero (− 1.61, −.06), indicating significant mediation of the path by the mediators. Results thus supported Hypothesis 3.

4. Discussion Analyses of the importance of shared task cognition in teams begged the question of whether and how team leadership may be instrumental in changing and developing shared and adaptive task representations in work groups and teams. We addressed this very issue focusing on the role of team leadership in developing group members' understanding of the role of information elaboration, a core process in work groups and teams with distributed information. We showed that group leaders' own task representation influenced the degree to which leaders encouraged group members to elaborate on information or to seek common ground, and thus influenced group members' task representations. Through this process, group leaders with task representations emphasizing elaboration were more effective in engendering group information elaboration and high-quality decisions than group leaders with less adaptive task representations. Moreover, comparison with leaderless groups showed that this is an influence particular to leadership which does not obtain if a regular group member holds such adaptive task representations. The present findings highlight leaders' ability to change group member task representations as important in effective leadership and identify the leadership process through which leaders' may influence follower task representations.

W.P. van Ginkel, D. van Knippenberg / The Leadership Quarterly 23 (2012) 94–106

103

4.1. Theoretical implications One of the main reasons for the team-based organization of work is the distributed informational resources available to teams. For this reason, we put group's use of distributed information central-stage in the current analysis. The leadership implications of our analysis should by no means be limited to distributed information and the process of group information elaboration, however. Whatever the task facing the team, research in shared cognition in teams suggests that shared and accurate cognitive representations of the task are conducive to smoother team process and superior team performance (e.g., Cannon-Bowers et al., 1993; Klimoski & Mohammed, 1994; Marks, Sabella, Burke, & Zaccaro, 2002; Mathieu, Goodwin, Heffner, Salas, & Cannon-Bowers, 2000). Accordingly, the importance of leaders' task representations and their ability to convey these representations should not be limited to groups with distributed information, and it would be worthwhile if further developments of the current analysis would focus on different tasks that may engender task representations emphasizing different processes. Even so, it stands to reason that for simpler, more routine tasks group member task representations will be less of an issue because they should be more likely to be adaptive and shared. Leaders' role in shaping group member task representations should thus be more evident in groups dealing with more complex, non-routine tasks. In mapping leaders' role in influencing group member task representations, we focused on relatively direct ways in which leaders may exert this influence — conveying own task representations through instructions and role-modeling. Leaders may also more indirectly foster task-adaptive representations, however. An important issue in shared cognition is that group members typically do not discuss their underlying task cognition (Cronin & Weingart, 2007; Hollingshead 1998; Mohammed & Ringseis, 2001). Differences in understanding thus are often not discovered and resolved, and group members miss out on opportunities to adapt their understanding to others' more insightful understanding of the task. A process of team reflexivity (West, 1996), which refers to the process of discussing the group task, goals, and the way in which those goals can be reached is particularly useful in aligning and developing group members' understanding of their task (van Ginkel et al., 2009). Importantly, groups do not habitually engage in team reflexivity, but leadership may play an important role in engendering such a process of collective reflection (Schippers, Den Hartog, Koopman, & van Knippenberg, 2008). In further developing our understanding of the influence of leadership in developing shared and adaptive task representations, we may thus also fruitfully explore the role of such more indirect ways in which leaders may influence task representations. Importantly, however, as research by van Ginkel and van Knippenberg (2009) shows, such a process of reflection only helps developing more adaptive task representations with the right input. A focus on team reflexivity too would thus put a premium on the accuracy of leaders' own task representations in guiding the reflexivity process. As expected, only marginal performance differences and no elaboration differences were found between leaderless situations and situations with a group leader with task representations emphasizing seeking common ground. Because one of the individuals in the group (i.e., the leader or a regular group member) always held task-adaptive representations emphasizing elaboration, these findings testify to the greater influence of group leaders as compared to regular group members. Neither in groups without a leader nor in groups with a leader with task representations emphasizing seeking common ground were group members with task-representations emphasizing elaboration able to realize the same level of elaboration and decision-making performance as group leaders endorsing elaboration were. These findings point to the influential role of group leaders in determining group members' understanding of the task and use of informational resources in group decision making. While from the perspective of leadership research the greater influence of leaders will not come as a surprise, these findings are important to research in socially shared cognition that has mainly concerned leaderless groups. In trying to understand how a shared and accurate understanding of the team task comes about, this research would do well to pay more head to the role of leadership. In the present study we focused on the influence of leaders internal to the group. While many work groups and teams in organizations have such an internal leader, not all do. Which is not to say that such self-managing teams (Manz & Sims, 1993) are entirely leaderless. They are typically subject to an external leadership structure (Kozlowski et al., 1996; Yukl, 1998). Such external leaders may potentially also influence group members' task representations, but as external leaders are likely to interact less with their followers the influence of external leaders may be smaller than the influence of internal leaders. Leader influence on follower task representations would seem to materialize through leader–team interaction, and accordingly the frequency of leader–team interaction may affect leaders' ability to influence group member task representations. The frequency of leader–team interaction is likely to be lower for external than for internal leaders, and as a consequence, the influence of external leadership in this respect is likely to be smaller too. However, it is also seems likely that the effect of internal versus external leadership is further qualified by other factors that determine a leader's influence on followers, like for instance a leader's hierarchical rank. That is, because leaders higher up in the organization may have a bigger impact on followers than leader's lower in the hierarchy (French & Raven, 1959), leaders higher in hierarchical rank may also have a stronger influence on members' representation even when they are not part of the team. Further exploration of this issue may be valuable for theory as well as practice. 4.2. Limitations and future directions The current study is not without its limitations. One potential limitation concerns the use of a laboratory experiment with a student sample, which may make it harder to generalize results to organizational settings. The experimental nature of present study was chosen for its potential to test causal relations and because it allowed us to study group process through the relatively unobtrusive and objective coding of audio–video data, rather than in a quest for external validity (Brown & Lord, 1999; Mook, 1983). Even so, an important question is whether we may observe such relationships in organizations. While previous studies

104

W.P. van Ginkel, D. van Knippenberg / The Leadership Quarterly 23 (2012) 94–106

have found that results of laboratory studies are often replicated in field studies (Dipboye, 1990), the proof of the pudding is in the eating, and it would be valuable to replicate the present results in a field setting using existing teams in organizations. In addition, because little research has been done on the relation between leadership and task representations we concentrated on establishing the relation between these concepts, which undoubtedly has gone at the expense of investigating other potentially interesting variables. For instance, in the present study the leader selection process mostly resulted in leaders having legitimate power (French & Raven, 1959) rather than any other type of power. It might also be interesting to investigate the effects of leadership resulting from different power bases such as expert power or reward power. Furthermore, specific leadership styles may interact with the influence a leader has on members' representations, such that not all leaders may be able to exert a similar amount of influence. Finally, we chose to examine the effects of leadership on group members' representations in a specific kind of setting, a group decision-making situation in which members had unique distributed information. While the finding that leaders influence group members' understanding of the task likely generalizes to other tasks and settings as well, the taskadaptive representations put center-stage in the present study are especially relevant or even specific to situations of distributed information and informational diversity (cf. van Knippenberg et al., 2004). For other types of tasks for which the exchange and integration of information is of less importance, such task representations may have smaller or even detrimental effects on performance, for instance when making a timely decision is much more important than making a decision based on as much available information as possible. Future research might thus extend the present findings in important ways by focusing on different task requirements and different task representations to establish whether the same basic processes obtain in linking leader task representations via leader behavior to group member task representations and performance. 4.3. Implications for practice While we believe it is always sensible not to base too far-reaching implications for practice on laboratory experiments, the present study does point to a number of implications for practice. The finding that group leaders with an appropriate understanding of the tasks at hand can benefit and develop the group, while group leaders lacking this understanding will, if anything, hold the group back, points to the importance of training group leaders, both to develop an appropriate understanding of the task and to develop their ability to convey their understanding of the task to followers. In this respect, note that the latter is not selfevident. Research in socially shared cognition in teams points to the fact that people's understanding of the task at hand typically goes undiscussed — it often requires a more reflective kind of discussion that people do not habitually engage in (cf. Mohammed & Ringseis, 2001). Likewise, we cannot take for granted that leaders will communicate their task representations. It might require explicit efforts in leadership development to not only create an accurate understanding of the task at hand on the part of the leader, but to also ensure leaders' awareness of the need to communicate this understanding as well as to enact it as a role model. As already hinted at earlier in our discussion, in addition to conveying their task representations directly through verbal communication and role-modeling, leaders may also coach groups and teams towards the development of task-adaptive representations by engendering a process of team reflexivity. Team reflexivity too is a process that does not seem to take place habitually (cf. Gray, 2007), and it will often require deliberate efforts of the leader to stimulate the process. Developing leaders' understanding of the benefits of team reflexivity as well as of the role of leadership in stimulating and guiding this process would thus also seem valuable. 4.4. Conclusion Groups and teams are often employed for knowledge-intensive tasks such as complex decision making. The implicit or explicit assumption driving much of this practice seems to be that groups are able to create value from their distributed informational resources. The distributed information paradigm employed in the current study in that sense captures the essence of the challenge facing such decision making groups — how to effectively use distributed informational resources. In an important counterpoint to research in distributed information and informational diversity more generally (van Knippenberg & Schippers, 2007), the present study highlights the role of group members' task representations and group leadership. As such, it invites a call to arms both for research in leadership of work groups and teams to put the study of task representations more center-stage and to the study of group process and team performance to put leadership more in the spotlight. References Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: Prentice-Hall. Bass, B. M., & Avolio, B. J. (1993). Transformational leadership: A response to critiques. In M. M. Chemers, & R. Ayman (Eds.), Leadership theory and research, (pp. 49–80). San Diego, CA: Academic Press. Berger, J., Webster, D., & Zelditch, M. J. (1985). Expectation states theory: Review and assessment. In J. Berger, & M. Zelditch (Eds.), Status, rewards and influence: (pp. 1–72). San Francisco: Jossey-Bass. Brown, R. D., & Hauenstein, N. M. A. (2005). Interrater agreement reconsidered: An alternative to the rwg indices. Organizational Research Methods, 8, 165–184. Brown, R. J., & Lord, R. G. (1999). The utility of experimental research in the study of transformational/charismatic leadership. The Leadership Quarterly, 10, 531–539. 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 group decision making: Current issues (pp. 221–246). Hillsdale, NJ: Erlbaum. Cohen, W. M., & Levinthal, D. A. (1990). Absorptive capacity: A new perspective on learning and innovation. Administrative Science Quarterly, 35, 128–152.

W.P. van Ginkel, D. van Knippenberg / The Leadership Quarterly 23 (2012) 94–106

105

Conger, J. A., & Kanungo, R. N. (1987). Towards a behavioral theory of charismatic leadership in organizational settings. Academy of Management Review, 12, 637–647. Cronin, M. A., & Weingart, L. R. (2007). Representational gaps, information processing, and conflict in functionally diverse teams. Academy of Management Review, 32, 761–773. De Dreu, C. K. W., Nijstad, B. A., & van Knippenberg, D. (2008). Motivated information processing in group judgment and decision making. Personality and Social Psychology Review, 12, 22–49. Dipboye, R. L. (1990). Laboratory vs. field research in industrial and organizational psychology. In C. L. Cooper, & I. T. Robertson (Eds.), International review of industrial and organizational psychology, Vol. 5. (pp. 1–34)Chichester, UK: Wiley. Edwards, J. R., & Lambert, L. S. (2007). Methods for integrating moderation and mediation: A general analytic framework using moderated path analysis. Psychological Methods, 12, 1–22. Fiedler, F. E. (1967). A theory of leadership effectiveness. New York: McGraw-Hill. French, J. R. P., & Raven, B. (1959). The bases of social power. In D. Cartwright (Ed.), Studies in social power (pp. 150–167). Oxford, Ann Arbor: University of Michigan. Gigone, D., & Hastie, R. (1993). The common knowledge effect: Information sharing and group judgment. Journal of Personality and Social Psychology, 65, 959–974. Gray, D. E. (2007). Facilitating management learning: Developing critical reflection through reflective tools. Management Learning, 38, 495–527. Gruenfeld, D. H., Mannix, E. A., Williams, K. Y., & Neale, M. A. (1996). Group composition and decision making: How member familiarity and information distribution affect process and performance. Organizational Behavior and Human Decision Processes, 67, 1–15. Hackman, J. R., & Wageman, R. (2005). When and how team leaders matter. Research in Organizational Behavior, 26, 37–74. Hastie, R., & Pennington, N. (1991). Cognitive and social processes in decision making. In L. B. Resnick, & J. M. Levine (Eds.), Perspectives on socially shared task representations (pp. 308–327). Washington, DC: American Psychological Association. Hinsz, V. B., Tindale, R. S., & Vollrath, D. A. (1997). The emerging conceptualization of groups as information processors. Psychological Bulletin, 121, 43–64. Hill, R. C., & Levenhagen, M. (1995). Metaphors and mental models: Sensemaking and sensegiving in innovative and entrepreneurial activities. Journal of Management, 21, 1057–1074. Hollingshead, A. B. (1998). Group and individual training: The impact of practice on performance. Small Group Research, 29, 254–280. House, R. J. (1971). A path goal theory of leadership effectiveness. Administrative Science Quarterly, 16, 321–338. Ilgen, D. R., Hollenbeck, J. R., Johnson, M., & Jundt, D. (2005). Teams in organizations: from input-process-output models to IMOI models. Annual Review of Psychology, 56, 517–543. Kerr, N. L., & Tindale, R. S. (2004). Group performance and decision making. Annual Review of Psychology, 55, 2201–2232. Klimoski, R., & Mohammed, S. (1994). Team mental model: Construct or metaphor? Journal of Management, 20, 403–437. Kooij-De Bode, H. J. M., van Knippenberg, D., & van Ginkel, W. P. (2008). Ethnic diversity and distributed information in group decision making: The importance of information elaboration. Group Dynamics, 12, 307–320. Kozlowski, S. W. J. (1998). Training and developing adaptive teams: Theory, principles, and research. In J. A. Cannon-Bowers, & E. Salas (Eds.), Making decisions under stress: Implications for individual and team training (pp. 115–153). Washington, DC: American Psychological Association. Kozlowski, S. W. J., & Bell, B. S. (2003). Work groups and teams in organizations. In W. C. Borman, & D. R. Ilgen (Eds.), Handbook of psychology: Industrial and organizational psychology (Vol. 12, pp. 333-375). New York: Wiley. Kozlowski, S. W. J., Gully, S. M., Salas, E., & Cannon-Bowers, J. A. (1996). Team leadership and development: Theory, principles, and guidelines for training leaders and teams. In M. M. Beyerlein, D. A. Johnson, & S. T. Beyerlein (Eds.), Advances in interdisciplinary studies of work teams: Team leadership, Vol. 3. (pp. 253–291)Greenwich, CT: JAI. Larson, J. R., Christensen, C., Abbott, A. S., & Franz, T. M. (1996). Diagnosing groups: Charting the flow of information in medical decision-making teams. Journal of Personality and Social Psychology, 71, 315–330. Lord, R. G., & Brown, D. J. (2004). Leadership processes and follower identity. Mahwah, NJ: Erlbaum. Madhavan, R., & Grover, R. (1998). From embedded knowledge to embodied knowledge: New product development as knowledge management. The Journal of Marketing, 62, 1–12. Manz, C. C., & Sims, H. P., Jr. (1993). Business without bosses: How self-managing teams are building high-performance companies. New York: Wiley. 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 of leader briefings and team interaction training for team adaptation to novel environments. Journal of Applied Psychology, 85, 971–986. Mathieu, J. E., Goodwin, G. F., Heffner, T. S., 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 normative comparisons. Journal of Organizational Behavior, 26, 37–56. Mohammed, S., & Ringseis, E. (2001). Cognitive diversity and consensus in group decision making: The role of inputs, processes, and outcomes. Organizational Behavior and Human Decision Processes, 85, 310–335. Mook, D. G. (1983). In defense of external invalidity. American Psychologist, 5, 379–387. Newell, A., & Simon, H. A. (1972). Human problem solving. Englewood Cliffs, NJ: Prentice Hall. Reicher, S. D., & Hopkins, N. (2001). Self and nation. London: Sage. Schippers, M. C., Den Hartog, D. N., Koopman, P. L., & van Knippenberg, D. (2008). The role of transformational leadership in enhancing team reflexivity. Human Relations, 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 of process accountability on information sharing and decision quality. Journal of Experimental Social Psychology, 43, 539–552. Shamir, B., House, R., & Arthur, M. B. (1993). The motivational effects of charismatic leadership: A self-concept based theory. Organization Science, 4, 577–594. Stasser, G. (1999). The uncertain role of unshared information in collective choice. In L. L. Thompson, J. M. Levine, & D. M. Messick (Eds.), Shared cognition in organizations: The management of knowledge (pp. 49–69). Mahwah, NJ: Erlbaum. Stasser, G., & Titus, W. (1985). Pooling unshared information in group decision making: Biased information sampling during discussion. Journal of Personality and Social Psychology, 48, 1467–1478. Tindale, R. S., Kameda, T., & Hinsz, V. B. (2001). Group decision-making. In M. A. Hogg, & J. Cooper (Eds.), Sage handbook of social psychology (pp. 381–403). London: Sage. Tindale, R. S., Smith, C. M., Thomas, L. S., Filkins, J., & Sheffey, S. (1996). Shared representations and asymmetric social influence processes in small groups. In E. Witte, & J. H. Davis (Eds.), Understanding group behavior: Consensual action by small groups (Vol 1). Hillsdale, NJ: Erlbaum. van Ginkel, W. P., & van Knippenberg, D. (2008). Group information elaboration and group decision making: The role of shared task representations. Organizational Behavior and Human Decision Processes, 105, 82–97. van Ginkel, W. P., & van Knippenberg, D. (2009). Knowledge about the distribution of information and group decision making: When and why does it work? Organizational Behavior and Human Decision Processes, 108, 218–229. van Ginkel, W., Tindale, R. S., & van Knippenberg, D. (2009). Team reflexivity, development of shared task representations, and the use of distributed information in group decision making. Group Dynamics, 13, 265–280. van Knippenberg, D., De Dreu, C. K. W., & Homan, A. C. (2004). Work group diversity and group performance: An integrative model and research agenda. Journal of Applied Psychology, 89, 1008–1022. van Knippenberg, D., & Hogg, M. A. (2003). A social identity model of leadership effectiveness in organizations. Research in Organizational Behavior, 25, 243–295. van Knippenberg, D., & Schippers, M. C. (2007). Work group diversity. Annual Review of Psychology, 58, 515–541.

106

W.P. van Ginkel, D. van Knippenberg / The Leadership Quarterly 23 (2012) 94–106

van Knippenberg, D., van Knippenberg, B., De Cremer, D., & Hogg, M. A. (2004). Leadership, self, and identity: A review and research agenda. The Leadership Quarterly, 15, 825–856. van Knippenberg, D., van Knippenberg, B., & van Dijk, E. (2000). Who takes the lead in risky decision making? Effects of group members' individual riskiness and prototypicality. Organizational Behavior and Human Decision Processes, 83, 213–234. van Knippenberg, D., Kooij-De Bode, H. J. M., & van Ginkel, W. P. (2010). The interactive effects of mood and trait negative affect in group decision making. Organization Science, 21, 731–744. Webber, S. S., Chen, G., Payne, S. C., Marsh, S. M., & Zaccaro, S. J. (2000). Enhancing team mental model measurement with performance appraisal practices. Organizational Research Methods, 3, 307–322. Weick, K. E. (2001). Making sense of the organization. Oxford, UK: Blackwell. Weingart, L. R. (1997). How did they do that? The ways and means of studying group processes. Research in Organizational Behavior, 19, 189–240. Weingart, L. R., Bennet, R. J., & Brett, J. M. (1993). The impact of consideration of issues and motivational orientation on group negotiation process and outcome. Journal of Applied Psychology, 78, 504–517. West, M. A. (1996). Reflexivity and work group effectiveness: A conceptual integration. In M. A. West (Ed.), Handbook of work group psychology (pp. 555–579). Chichester, UK: Wiley. Winquist, J. R., & Larson, J. R., Jr. (1998). Information pooling: When it impacts group decision making. Journal of Personality and Social Psychology, 74, 371–377. Wittenbaum, G. M., & Stasser, G. (1996). Management of information in small groups. In J. L. Nye, & A. R. Brower (Eds.), What's social about social task representations? Research on socially shared task representations in small groups (pp. 3–28). Thousand Oaks, CA: Sage. Yukl, G. (1998). Leadership in organizations (4th ed.). Upper Saddle River, NJ: Prentice-Hall. Zaccaro, S. J., Rittman, A. L., & Marks, M. A. (2001). Team leadership. The Leadership Quarterly, 12, 451–483.