hey

The Leadership Quarterly 18 (2007) 105–120 www.elsevier.com/locate/leaqua

The role of leaders in shaping formal team norms

Simon Taggar a,⁎, Robert Ellis b,1

a Wilfrid Laurier University, Waterloo, Ontario Canada N2L 3C5 b University of Northern British Columbia, Prince George, BC Canada V2N 4Z9

Abstract

We developed and tested a multilevel model of norm formation. Both leader expectations and staff expectations had a significant impact on the norms that teams adopted around collaborative problem solving. In addition, there was an interaction between the expectations of the leader and the staff. We found that when staff initially held low expectations of collaborative problem solving behaviors, leaders who held high expectations of such behaviors were able to significantly raise the collaborative problem solving norms established by their teams. Team problem solving norms significantly influenced individual team members' problem solving behaviors. © 2007 Elsevier Inc. All rights reserved.

Keywords: Team norms; Norm formation; Collaborative problem solving

It is clear that team norms, the shared beliefs regarding the social behaviors expected of its members, may guide behavior, exert a powerful form of social control, and impact performance (Barker, 1993; Bettenhausen & Murnighan, 1985, 1991; Steiner, 1972; Vroom, 1969), particularly in groups that are interdependent (Wageman, 1995). Yet, norm formation has been studied only rarely (Bettenhausen & Murnighan, 1985, 1991; Levine, Higgins, & Choi, 2000), despite periodic calls for more systematic empirical attention (Cialdini & Trost, 1998; Feldman, 1984; Hackman, 1976). Indeed, in their review of the social influence literature, Cialdini and Trost (p. 154) note: “although social scientists have used norms as explanatory constructs throughout the twentieth century, the empirical literature specifically studying the emergence and transmission of social norms is exceedingly small.” The present investigation sought to understand the role of leaders in shaping the norms that are developed in teams. Feldman (1984) suggested that leaders may have a strong impact on the development of team norms, but this notion has not been tested empirically.

In order to encourage effective team member behaviors, academics and practitioners have suggested that teams set formal (prescriptive) norms at the beginning of team interaction (e.g., Argots, 1989; Feldman, 1984; Spich & Keleman, 1985). The norms would provide members with information about the “group's reality and afforded standards against which to compare a person's behavior” (Colman & Carron, 2001; p. 201). In conditions of high reciprocal interdependence, the absence of strong norms supporting task accomplishment detracts from team effectiveness, while

⁎ Corresponding author. Tel.: +1 519 884 0710x2614; fax: +1 519 884 0201. E-mail addresses: [email protected] (S. Taggar), [email protected] (R. Ellis).

1 Tel.: +1 250 960 6491; fax: +1 250 960 5544.

1048-9843/$ - see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.leaqua.2007.01.002

Fig. 1. Formation of norms from leader and staff expectations in newly formed teams.

106 S. Taggar, R. Ellis / The Leadership Quarterly 18 (2007) 105–120

the negotiation of common expectations in the form of agreed upon team norms may contribute to team performance (Argots, 1989; Mullen & Copper, 1994; Roethlisberger & Dickson, 1939; Thompson, 1967).

The present study focuses on how emergent leaders and other team members (referred to as staff by LePine, Hollenbeck, Ilgen, & Hedlund, 1997; Taggar, Hackett, & Saha, 1999) influence team norms that are agreed upon at the outset of team interaction. The primary goal of this study was to develop and assess a model that depicts the impact of team leaders and staff on negotiated and agreed upon team norms for collaborative problem solving in newly formed teams. Specifically, we test whether the expectations held by leaders and staff prior to team formation influence team norms. While Stevens & Campion (1994) view collaborative problem solving as fundamental to team effectiveness in general, it should be especially important in teams that complete heuristic (open-ended) problem solving tasks, where the effective use of collective resources is essential to success. The effects of collaborative problem solving norms established by the team on subsequent individual team member behaviors are also examined. Fig. 1 depicts the model tested.

In our theorizing we follow Morgeson & Hofmann's (1999) guidelines for the investigation of the emergence of constructs at the collective level. That is, we start at the most elementary unit of analysis in any social system– individual beliefs and actions (Parsons, 1937). We then seek to understand the interaction of leaders and staff in a configural model (Kozlowski & Klein, 2000) of norm formation. Finally, we consider context in a cross-level direct effect model (Kozlowski & Klein, 2000). Team norms were our contextual variable. In our model we seek to determine the impact of norms on subsequent teamwork behaviors.

1. Leader emergence and team norms

1.1. Emergent leaders

Leaders emerge early in initially leaderless self-managing work teams (Okhuysen, 2001; Taggar et al., 1999), through “peer-directed role-making” negotiations (Seers, Petty, & Cashman, 1995; p. 21) that form a part of a role

107S. Taggar, R. Ellis / The Leadership Quarterly 18 (2007) 105–120

development process (Hollenbeck, Lepine, & Ilgen, 1996; Seers, 1989). The manner in which the leader comes to power — whether formally designated or emerging — may be “unimportant in comparison to the behaviors of the leader” (Firestone, Lichtman, & Colamosca, 1975, p. 347). Schneider & Goktepe (1983) define emergent leaders as team members who exert significant influence over other members of the team although no formal authority has been vested in them. Being perceived as a leader permits an individual to go beyond a formal role in influencing others (Lord & Maher, 1991). The emergent leader's role involves initiating or formulating goals, organizing the team's thinking, clarifying team member responses, and summarizing, generalizing, and formulating conclusions (Bass, 1949, 1961).

LePine et al. (1997) suggest that leaders and staff play separate roles and thus make unique contributions to team performance. They found that high cognitive ability and conscientiousness on the part of both leader and staff were important for team performance, thereby supporting the idea of separate roles being played by the leader and staff. Similarly, Taggar et al. (1999) found that the resources possessed by a team leader (e.g., personality traits) and by staff both predicted team performance. Moreover, they found that in some cases, the attributes possessed by the team leader were more important than the attributes possessed by the staff in predicting team performance, and that there was an interaction between leader and staff contributions to the team. That is, they found that teams performed best when both the emergent leader and other team members were high in leadership; however, the team leader could not fully compensate for poor self-management behavior among staff.

1.2. Norms

Team norms (whether formally or informally derived) emerge, are transmitted, and persist through the beliefs and actions of the members of a team (Feldman, 1984). By examining interactions among team members, and the processes that underlie these interactions, one can understand how and why norms emerge (Morgeson & Hofmann, 1999).

Early group studies by Sherif (1936) and Asch (1956) focused on greater conformity among members resulting from group norms. Both researchers studied responses of participants performing relatively straightforward judgment tasks (e.g., estimating the movement of a dot on the wall, comparing the length of a line to three others), in both individual and group settings. These studies found that judgments in group settings were influenced by norms when compared to judgments made in individual settings. Subsequent theoretical work has focused on types of group norms and their characteristics, while empirical work has examined the impact of norms on group outcomes (Feldman, 1984). There have been a few recent studies providing insight into norm formation, including the influence of demographic heterogeneity on the emergence of cooperative norms (Chatman & Flynn, 2001) and the development of strategic norms (Levine et al., 2000).

Norms are established in the early stages of team development (Tuckman & Jensen, 1977) and are “usually based on an interpersonal agreement” (Friedkin, 2001, p. 169). When individuals join teams, their feelings of uncertainty regarding expected actions are eased as subsequent communication with the team clarifies appropriate behavior (Colman & Carron, 2001). That is, when a team member proposes a norm, social support for that preference is determined through discussion. Consistent with social exchange theory, norms give rise to obligations that form the basis of each person's relationship with others and with the team as a whole (Shore & Barksdale, 1998). As a result, “group members tend to decrease the variance in their behavior” (Vroom, 1969, p. 223) as their behavior converges on the agreed upon norm. Norms apply more or less equally to all team members (Argots, 1989).

Norms can be beneficial to team functioning because they: (a) facilitate team survival, keep the team together, and protect it from other teams (since behaviors that threaten the team are punished); (b) provide regularity and predictability to the behavior expected from team members, and thus help team members to anticipate the actions of other team members; (c) help the team avoid embarrassing interpersonal problems and thus ensure that members' self- images are protected; and (d) express the central values of the team and clarify what is distinctive about the team's identity (Feldman, 1984). However, because norms are often informal and emerge slowly in teams, they may not support the larger strategic goals of the team (Wageman, 1997). A norm does not always arise when the existence of an effective norm would be in the interests of all or most persons (Coleman, 1990). By debating and establishing formal norms, teams are better able to proactively determine behaviors that are best tailored to their needs (Argots, 1989; Feldman, 1984). Accordingly, because of the importance of team norms, some scholars recommend that formal team norms be debated and agreed upon when the team is initially created (e.g., Argots, 1989; Feldman, 1984). Negotiation of common expectations regarding how each team member should behave (i.e., team norms) represents a proactive

108 S. Taggar, R. Ellis / The Leadership Quarterly 18 (2007) 105–120

stance toward dealing with team problems (Wageman, 1997) and should contribute to team performance (Mullen & Copper, 1994; Roethlisberger & Dickson, 1939). Accordingly, in our sample, norms were established at the beginning of team interaction as opposed to developing informally over time.

1.3. Formation of collaborative problem solving norms in newly formed teams

Norms are formed with respect to behaviors having significance for the team, such as those that facilitate task accomplishment (Feldman, 1984). In this study, teams completed heuristic tasks that required collaborative problem solving and thus we focus on collaborative problem solving norms given their salience in this context. A fundamental activity of teams completing a problem solving task is the integration of individual knowledge into collective knowledge to solve a problem (Taggar, 2002). In addition, in newly formed teams with team-based rewards, a primary concern is how much effort each team member should exert in order to maintain “fair” or equitable reciprocal relationships (Taggar & Neubert, 2004). Hence, establishing expectations about the degree of collaboration is a critical first step when individuals have to work together on a project without knowing how collaborative their peers will be. There are many organizational situations in which team members must integrate their knowledge to solve problems, e.g., employees working in multifunctional product development teams who must make strategic choices about their collective business, and teams of factory representatives working on manufacturing process improvements (Ancona & Caldwell, 1992; Eisenhardt, 1989; Tyre & Orlikowski, 1994).

In previous group problem solving research the dominant research tradition has been to identify variables that encourage social interaction between participants and optimize team productivity (e.g., Bettenhausen & Murnighan, 1991; Jehn & Mannix, 2001; Keller, 2001). For example, research has been conducted on training (Bottger & Yetton, 1987; Wheeler & Valacich, 1996), creativity (Amabile, 1996, 1998; Taggar, 2002), team support systems (Dennis, 1996; Dennis & Valacich, 1993), and team size (Gallupe et al., 1992; Hare, 1981). However, there is little research, to our knowledge, on how a leader impacts team member behavior through team norms.

Friedkin (2001) suggests that norms are formed through a process of interpersonal influence in which “members' attitudes and opinions on an issue change as they revise their positions by taking weighted averages of the influential positions of other members” (p. 171). Emergent team leaders may influence discussion leading to norm formation because they are “likely to direct other team members' activities” (De Souza & Klein, 1995, p. 475) and influence subordinate behavior to facilitate goal achievement (Mumford, 1986). Indeed, leadership has been defined in terms of the influence leaders exert over other team members (Schneider & Goktepe, 1983). Moreover, leaders often direct the discussion process or serve an integrative function within the team (Maier, 1967). The impact that team leaders and staff have on problem solving behavioral norms can be determined if initial behavioral intent were assessed prior to the establishment of team norms.

In summary, we propose a relationship between leader expectations and norms based on a leader's influence in team discussions around norms. The social interaction and exchange inherent in forming norms represents a bottom–up process (Kozlowski & Klein, 2000) in which individual member expectations about how they will behave emerge to form the collective norm for how all team members should behave.

Shamir, Zakay, Breinin, & Popper (1998) argued that when leader behaviors are not directed toward specific individuals but toward the group as a whole then that leader behavior should be treated as being at the group level. Gavin & Hofman (2002) explain that the underlying assumption in treating leadership as a team level variable is that the individual members “who have common leaders, are exposed to a similar leadership environment in terms of the behaviors and actions of the leaders. This common leadership environment manifests a shared leadership climate within the unit” (p. 21). Hypothesis 1 is tested at the team level.

H1. Leader expectations for how frequently they would display collaborative problem solving behaviors (time 1) in a forthcoming team task will be positively associated with the negotiated team problem solving norms (time 2).

Although in this study one emergent leader per team was identified, members of self-directed teams are expected to share the management of team functioning (Cannon-Bowers, Oser, & Flanagan, 1992). Therefore, we also expected that staff would influence team norms.

H2. Average staff expectations for how frequently they would display collaborative problem solving behaviors (time 1) will be positively associated with the negotiated team problem solving norms (time 2).

109S. Taggar, R. Ellis / The Leadership Quarterly 18 (2007) 105–120

Above we note that while norms are a collective construct that deal with behavioral expectations at the team level, at the individual level team members enter a team with initial expectations as to how they will behave. How these initial expectations on the part of leader and staff are incorporated into formal team norms is addressed next.

Once established, norms can enable or constrain behavior (Gersick & Hackman, 1990). Thus, a leader should be concerned with ensuring that appropriately high levels of performance-relevant team behaviors are incorporated into norms. This suggests an interaction between staff and leader inputs (cf. Taggar et al., 1999) when staff expectations fall below the leader's expectations. Specifically, if emergent leaders exert significant influence over team members (cf. Schneider & Goktepe, 1983), a leader with high problem solving expectations could be presumed to positively influence staff behavioral expectations.

H3. Leader and average staff collaborative problem solving expectations will interact to predict team norms, such that a leader can ameliorate the low expectations of staff.

2. Collaborative problem solving norms and subsequent behavior

Collaborative norms appear to be the foundation for successful teamwork. For instance, Chatman & Flynn (2001) have shown that cooperative norms mediate team efficiency and effectiveness. Tjosvold and his colleagues demonstrated the central role played by cooperative goals in building strong team relationships and resolving conflict (Alper, Tjosvold, & Law, 1998, 2000; Tjosvold, Poon, & Zi-you, 2005). Thus it was expected that norms for collaborative problem solving would shape a broad range of behaviors in teams. According to the theory of reasoned action, normative beliefs influence attitudes, behavioral intentions, and behaviors (Ajzen & Fishbein, 1980; Fishbein & Ajzen, 1975). Stevens & Campion (1994) suggest that performance-relevant team member behaviors can be grouped into five categories (a) conflict resolution, (b) collaborative problems solving, (c) communication, (d) goal setting/performance management, and (e) planning/task coordination. Accordingly, we expected:

H4. Collaborative problem solving team norms (measured at time 2) will predict individual team member behaviors (measured at time 3) on (a) conflict resolution, (b) collaborative problems solving, (c) communication, (d) goal setting/ performance management, and (e) planning/task coordination.

3. Methods

3.1. Participants and procedure

Participants were 222 third year honors business students enrolled in an organizational behavior course in a southern Ontario university (55% women, mean age=21 years, SD=.49) who were randomly assigned to teams. Of the 66 possible teams, 56 teams that ranged in size from 3 to 5 persons (median=4) were used in this study because all team members agreed to participate and provided complete data. A graduate research assistant, blind to study hypotheses, solicited voluntary participation.

Work teams were created during the second week of classes and remained together for the duration of the 12-week course. First, participants were asked to complete a measure of their individual expectations regarding their contributions to their teams in terms of problem solving behaviors. Second, upon team formation, behavioral norms were established through a 30-minute discussion during which teams were asked to achieve team consensus regarding the “behavioral expectations of team members” around collaborative problem solving. At the conclusion of this discussion, each team member signed a contract containing the team's agreed upon norms to indicate that consensus had been reached. All the participation consent forms and data were sealed for the duration of the course so neither the research assistant nor authors knew who was participating in the study.

Teams were given autonomy in how best to accomplish tasks over the term. In the final week of classes, 10 weeks later, participants completed peer evaluations of each member of their team. They were asked to “assess frequency of behaviors” of their fellow team members based on their observations during the team interactions. In addition, members completed a leadership assessment of their peers.

110 S. Taggar, R. Ellis / The Leadership Quarterly 18 (2007) 105–120

3.1.1. Team tasks Teams worked together on two assignments for 35% of their course grade. Both tasks involved solving applied

problems. The first, worth 15% of their grade, was to make a 15-minute presentation in class on a topic from the course (e.g., expectancy theories of motivation). They were required to summarize the main concepts and identify their practical implications with respect to solving “real world” problems. Teams made presentations during weeks 3, 6 and 7 of the course. The second task, worth 20% of their grade, was a written analysis of a business case that depicted a management problem. Teams submitted their written case analysis in week 10 of the course. All of the work in these two tasks was conducted outside of the classroom. Teams had received their results for the first assignment, but not for the second, at the time they completed peer evaluations of team behavior and leadership.

3.2. Measures

3.2.1. Collaborative problem solving expectations of members in leader and staff roles Team members' individual intended behavior was measured prior to team formation and before the team had an

opportunity to discuss and agree upon collaborative problem solving team norms. We modified Taggar & Brown's (2001) measure which operationalized Stevens & Campion's (1994) collaborative problem solving dimension of teamwork. The measure is in the form of a Behavioral Observation Scale designed for peer performance appraisals. The original measure consists of critical incidents of performance-relevant behavior. It was modified to ask participants to report their “expected contribution to the team” over the 10 week period that their team would complete assigned tasks. Members reported how frequently they expected to display collaborative problem solving behaviors on 5-point Likert- type scales ranging from 1 (below 20% of the time) to 5 (above 80% of the time). The measure consisted of the following 8 items: (a) ask other team members what they think; (b) clarify and explain issues when someone does not understand; (c) offer ideas; (d) ask relevant questions; (e) accept team roles and tasks as required; (f) voice unique ideas; (g) build on the group's ideas by offering solutions; and (h) summarize and organize the group's ideas (Cronbach's alpha=.89).

In this study, we were concerned with (a) the level of collaborative problem solving behavior that staff expected to exhibit, which was assessed prior to team formation, (b) the level of collaborative problem solving behavior an emergent team leader expected to exhibit, also assessed prior to team formation, and (c) how these staff and leader collaborative problem solving expectations are reflected in the agreed upon norm. We defined the leader as the individual with the highest leadership score in the team (discussed below). Thus, we were concerned with the influence of one person per team in the leadership role. Each leader's expectations regarding collaborative problem solving were averaged over the 8 items (the minimum mean score was 3.02 and the maximum mean score was 4.74). However, the staff role was occupied by 2–4 members.

To create one overall expectations score for the 2–4 members in the staff role, staff expectations were averaged across members and items (scores ranged from an average of 2.36 to 4.17). Since expectations were assessed prior to team formation, consensus amongst team members on their expectations cannot be expected.

Characterizing staff as high in expectations would mean that, for the staff as a unit, members display high expectations. This does not imply that all staff members display high expectations for collaborative problem solving, just that there are at least some members whose scores elevate the average for the staff (cf. Neuman & Wright, 1999; p. 30). Chan (1998) argued that an additive model is appropriate when the theoretical interest is the magnitude of an effect and when group member perceptual agreement is not a concern.

3.2.2. Team collaborative problem solving norms There remains considerable disagreement about how best to measure team level constructs such as team norms. Some

researchers have used an approach whereby each team member provides his or her perception of the team level construct (e.g., norm) and then the members' scores are aggregated to form a team score that represents a team level construct (e.g., Earley, 1999; Feltz & Lirgg, 1998). However, aggregation has also been criticized for (a) not truly capturing team level processes because the level of measurement still resides at the individual rather than at the team level of analysis, and (b) not allowing for intrateam interaction in responding to survey items, which limits the potential to capture underlying team processes (Gibson, 1999; Kirkman, Tesluk, & Rosen, 2001). Campion, Medsker, & Higgs (1993) called for a move away from the aggregation of individual level data to assess team level phenomena and a move toward having teams provide consensus ratings to capture the true underlying dynamics of intact work teams. This involves having the entire team meet together and collectively respond using consensus decision making. Accordingly, the team provides only one

111S. Taggar, R. Ellis / The Leadership Quarterly 18 (2007) 105–120

response for each item, and there is no need to aggregate individual responses. Kirkman et al. (2001) compared aggregation of individual team member ratings to a team consensus approach on their ability to predict manager ratings of team effectiveness. They found that the consensus method explained significantly greater variance in team effectiveness than did the aggregation method alone. The consensus method is particularly appropriate for the study of norm formation because it reflects a process of interaction which is the basis of the norm construct (see Bar-Tal, 1990).

Hence, in this study the agreed upon collaborative problem solving team norms (measured at time 2) were determined as a single team response to a series of items (rather than multiple responses from team members) after a period of discussion. Again, we used a modified version of the Taggar & Brown (2001) measure which operationalizes Stevens & Campion's (1994) collaborative problem solving dimension of teamwork. Though the items were identical to those used to assess team members' individual expectations for how they would behave, which was measured prior to team formation, in the case of the norms measure, each team agreed upon the frequency with which team members' were expected to display collaborative problem solving behaviors (consisting of 8 items with average item scores ranging from 3.12 to 4.44) on 5- point Likert-type scales ranging from 1 (below 20% of the time) to 5 (above 80% of the time) (Cronbach's alpha=.86).

3.2.3. Peer evaluations of team behaviors Team members evaluated the frequency of team behaviors of their fellow members (measured at time 3, 10 weeks

later) on the 5 teamwork dimensions described by Taggar & Brown (2001). The knowledge, skills and abilities identified by this typology generalize across different team contexts (Cannon-Bowers, Tannenbaum, Salas, & Volpe, 1995). Thus, this typology provides an appropriate framework with which to examine the specific teamwork behaviors of individuals in self-managing teams. Again, responses were measured on 5-point Likert-type scales ranging from 1 (below 20% of the time) to 5 (above 80% of the time) and averaged. The dimensions were (a) conflict resolution (10 items ranging in average scores from 2.07 to 5.00), (b) collaborative problem solving (8 items ranging in average scores from 1.97 to 5.00), (c) communication (3 items ranging in average scores from 1.89 to 5.00), (d) goal setting/ performance management (16 items ranging in average scores from 1.52 to 3.61), and (e) planning/task coordination (9 items ranging in average scores from 1.64 to 5.00). Dimension Cronbach alphas ranged from .73 to .88. To assess the level of inter-rater agreement on the behavioral ratings, we calculated the rwg coefficient (James, Demaree, & Wolf, 1983) for each participant. The average coefficient value across all ratings and participants was .85 (ranging from .57 to .96). One team was below a .70 cut-off but was retained in the analyses. In conclusion, most team members agreed as to which individuals demonstrated high levels of the behaviors of interest (and which did not).

3.2.4. Peer evaluations of leadership Lord & Maher (1991) note that while “leadership perceptions may not be reality; they are used by perceivers to

evaluate and subsequently distinguish leaders from non-leaders” and “also provide the basis for social power and influence” (p. 98). Team members evaluated their fellow members on leadership (time 3, week 10) using the General Leadership Impression (GLI) measure (Cronshaw & Lord, 1987). This measure is strongly related to objective measures of leadership behaviors and other measures of leadership perceptions (Lord, 1977). The GLI contains six verbally- anchored scales that assessed the amount of exhibited leadership, the extent to which the member was typical of a leader, the amount of leadership behavior, correspondence to the image of a leader, willingness to choose the member as a leader for a future group, and ranking as the leader of the present group. Responses were measured on 5-point scales with lower numbers indicating greater leadership. As in Cronshaw & Lord (1987), these 6 items were averaged to yield an overall index of general leadership impression (Cronbach's alpha=.85). The mean ratings were averaged across peers to assess leadership. The team member with the strongest general leadership impression (lowest score) was defined as the leader (this methodology is consistent with Kent & Moss, 1990, 1994; Taggar et al., 1999; Wentworth & Anderson, 1984), while other team members were defined as staff. Average rwg across all ratings and participants was .81 (range from .66 to 89). Three teams were below a .70 cut-off but were retained in the analyses. ICC(1) was .78 indicating an adequate proportion of the total variance accounted for by team membership. ICC(2) was .93 indicating reliable group level means.

3.3. Analyses

There were 8 teams where the leadership rating of the next highest leader was within a standard error of the leader's score. Analyses conducted after dropping these teams do not substantially change the results or conclusions drawn when all teams are used. Below, we report analyses using all 56 teams.

Table 1 Correlation matrix of study variables at the individual level (N=222)

1 2 3 4 5 6

1. Peer ratings of leader's leadership – 2. Peer ratings of conflict resolution −.27⁎⁎⁎ – 3. Peer ratings of collaborative problem solving −.53⁎⁎⁎ .55⁎⁎⁎ – 4. Peer ratings of communication −.29⁎⁎⁎ .35⁎⁎⁎ .22⁎⁎⁎ – 5. Peer ratings of goal setting/performance management −.50⁎⁎⁎ .51⁎⁎⁎ .63⁎⁎⁎ .11 – 6. Peer ratings of planning/task coordination −.54⁎⁎⁎ .48⁎⁎⁎ .72⁎⁎⁎ .05 .69⁎⁎⁎ – Mean 2.68 3.82 4.16 4.14 3.04 3.84 SD .53 .46 .50 .55 .30 .44

Note: ⁎ pb.05, ⁎⁎ pb.01, ⁎⁎⁎ pb.001. Low leadership scores denote high leadership.

112 S. Taggar, R. Ellis / The Leadership Quarterly 18 (2007) 105–120

Hypothesis 1–3 were tested using moderated multiple regression analyses. In a context where influences on the dependent variable might span multiple levels, multiple regression has a limitation that can be addressed by hierarchical linear modeling (HLM; Bryk & Raudenbush, 1992; Hofmann, 1997). In regression at an individual level, team level variables are assigned to individuals within the team, which can violate the assumption of independence and produce biased estimates of the standard errors of the team level variable and biased significance tests of parameters (Bryk & Raudenbush, 1992; Hofmann, 1997). The advantages of HLM over moderated regression in explicitly modeling both within and between team variance (Hofmann, Griffin, & Gavin, 2000), led us to use HLM to test Hypotheses 4a 4b 4c 4d 4e where norms (team level or level 2 variable) are hypothesized to impact team member behavior (individual level or level 1 variables). For instance, Hypothesis 4a was tested as follows:

Level-1 (individual level):

Y ðConflict resolution behaviorÞ ¼ b0 Level-2 (team level):

b0 ¼ g00 þ g01ðCollaborative problem solving normÞ þ U0

4. Results

A zero-order correlation matrix of key study variables at the individual level of analysis is in Table 1. Team level correlations are in Table 2. While peer ratings of team member behaviors were correlated in Table 1, we decided not to combine teamwork dimensions to form higher order constructs. Past conceptualizations have suggested the teamwork dimensions are distinct from one another (Taggar & Brown, 2001).

Results of regression analysis testing 1–3 are in Table 3. Hypothesis 1 was supported. That is, team problem solving norms (t2) were associated with leader expectations for how frequently they would display collaborative problem solving behaviors (t1) (β=3.84, pb.05). Hypothesis 2 was also supported. Team problem solving norms (t2) were associated with staff expectations for how frequently they would display collaborative problem solving behaviors (t1) (β=2.92, pb.05). Hypothesis 3 was also supported. We found that leader and staff collaborative problem solving expectations interacted (β=−3.78, pb.05) to predict team norms, such that leaders could ameliorate the low expectations of staff (Fig. 2). Using procedures described by Aiken & West (1991), the slope of the simple regression

Table 2 Correlation matrix of study variables at the group level (N=56)

1 2 3

1 Leader collaborative problem solving expectations – 2 Staff collaborative problem solving expectations −.20 – 3 Team problem solving norm .26⁎ .44⁎⁎⁎ –

Means 4.33 3.71 4.13 SD .35 .20 .27

Note: ⁎ pb.05, ⁎⁎ pb.01, ⁎⁎⁎ pb.001.

Table 3 Regression of team collaborative problem solving norms on leader collaborative problem solving expectations, staff collaborative problem solving expectations, and leader×staff collaborative problem solving expectations (N=56 teams)

Dependent variable measured (t2)

Behavioral expectations (independent variables) (t1).

β a T R2

Collaborative problem solving behavioral norms (t2)

Leader collaborative problem solving expectations ( H1) 3.84⁎ 2.22 .38 Mean staff collaborative problem solving expectations ( H2) 2.92⁎ 2.46 Mean staff collaborative problem solving expectations×Leader collaborative problem solving expectations ( H3)

−3.78⁎ −2.02

Note: ⁎ pb.05, ⁎⁎ pb.01, ⁎⁎⁎ pb.001. a Standardized beta.

113S. Taggar, R. Ellis / The Leadership Quarterly 18 (2007) 105–120

line was found to be significantly different from zero when staff expectations were 1 SD below the mean (unstandardized b=.42, t=4.41, pb.05) but not when they were 1 SD above the mean (unstandardized b=.12, t=1.15, p=.26). This finding reveals that although the leader expectations×staff expectations interaction results in incremental explanation of the team collaborative problem solving norms, the effect is only evident at a low level of staff expectations. Simple slopes were also calculated where leader expectations were treated as the moderator in order to determine that the simple slopes are significant outside the .67 to 1.29 values of leader expectations. This analysis indicates that the level of collaborative problem solving norms achieved by high leader/low staff expectation teams was significantly lower than that achieved by high leader/high staff expectation teams when the leadership score is 1 SD above the mean. Thus, we concluded that the leader's expectations could not fully compensate for low staff expectations.

Results of the HLM analyses (Scientific Software Inc., Chicago) used to test Hypotheses 4a 4b 4c 4d 4e are presented in Table 4. The HLM analog to ICC(1), reflecting the percentage of variance that resides between teams, was calculated using the procedures outlined by Gavin & Hofmann (2002; pg. 24). Table 4 shows that a statistically significant and relatively large percentage of variance in team member behaviors resided between teams (24% to 67%), therefore providing a basis for examining a team level predictor of team member behavior (i.e., problem solving norm). Moreover, Table 4 reveals that Hypotheses 4b (t-ratio=2.39, pb.05), 4c=(t-ratio=2.01, pb .05), and 4e (t-ratio=2.95, pb.01) were supported in the HLM analyses.

4.1. Post-hoc analyses

Teams can develop strong and weak norms characterized by the degree to which members' actions are controlled (Barker, 1993). If norms are weak, team members' subsequent behavior (time 3) may be significantly influenced by

Fig. 2. Interaction effects between leader expectations of collaborative problem solving behaviors and staff expectations of collaborative problem solving behaviors on team norms (N=56 teams)a. Note: a Using marginal means (in GLM).

Table 4 Hierarchical linear modeling analyses of collaborative problem solving team norms predicting individual team member behaviors (N=56 teams)

Variable Fixed effect coefficient estimates a (B)

SE t-Ratio Percentage of variance for the outcome variable that resides between groups

Outcome variable is conflict resolution ( H4a) Intercept (γ00) 10.53 .45 4.30⁎⁎⁎ .42 Problem solving norm (γ01) .03 .06 .09 χ

2 (55)=252.07, pb.001

Outcome variable is collaborative problem solving ( H4b) Intercept (γ00) 12.18 .09 117.86⁎⁎⁎ .27 Problem solving norm (γ01) .22 .07 2.39⁎ χ

2 (55)=156.00, pb.001

Outcome variable is communication ( H4c) Intercept (γ00) 4.15 .05 88.71⁎⁎⁎ .67 Problem solving norm (γ01) .09 .04 2.01⁎ χ

2 (55)=108.15, pb.001

Outcome variable is goal setting/performance management ( H4d) Intercept (γ00) 4.14 .05 88.71⁎⁎⁎ .37 Problem solving norm (γ01) .02 .02 1.00 χ

2 (55)=147.08, pb.001

Outcome variable is goal setting/performance management ( H4e) Intercept (γ00) 11.52 .11 104.68⁎⁎⁎ .24 Problem solving norm (γ01) .25 .08 2.95⁎⁎ χ

2 (55)=138.96, pb.001

Note: ⁎ pb.05, ⁎⁎ pb.01, ⁎⁎⁎ pb.001. a Unstandardized.

114 S. Taggar, R. Ellis / The Leadership Quarterly 18 (2007) 105–120

their initial behavioral expectations at the time (time 1) of entering the team. Having confirmed in the above analyses (Table 4) that team member behavior varied both within- and between teams, we tested whether individual level team member expectations upon entering the team were positively related to their behavior. The random coefficient

Table 5 Random-coefficient regression model for associations between member expectations and team member behavior (N=222 individuals in 56 teams)

Dependent variable Independent variables Fixed effect coefficient estimates a (B)

t-ratio Within-group variance in the outcome accounted for by expectations

Conflict resolution For intercept, 80 11.20⁎⁎⁎ 24.09 3.13% Intercept, γ00 For conflict resolution expectations slope, 81 .01 .06 Intercept, γ10

Collaborative problem solving

For intercept, 80 11.45⁎⁎⁎ 12.03 5.21% Intercept, γ00 For collaborative problem solving expectations slope, 81

.03 1.11

Intercept, γ10 Communication For intercept, 80 4.62⁎⁎⁎ 18.43 6.52%

Intercept, γ00 For communication expectations slope, 81 −.05 −1.82 Intercept, γ10

Goal setting/ performance

For intercept, 80 12.20⁎⁎⁎ 127.72 5.49% Intercept, γ00 For goal setting/performance expectations slope, 81

.00 .15

Intercept, γ10 Planning/task coordination

For intercept, 80 11.36⁎⁎⁎ 14.53 5.93% Intercept, γ00 For planning/task coordination expectations slope, 81

.00 .81

Intercept, γ10 a Unstandardized.

115S. Taggar, R. Ellis / The Leadership Quarterly 18 (2007) 105–120

regression HLM procedure used in this test is described by Gavin & Hofmann (2002; pgs. 25–26). Table 5 reports the results. The t tests associated with the γ10 parameters were not significant, suggesting that team members' expectations prior to entering the team were not related to their subsequent behavior. Thus, our results suggest members' actions were controlled by team norms.

Lastly, dispersion in team member expectations may be important in forming team norms. Accordingly, we correlated the variance in collaborative problem solving expectations within teams and the team norms. The correlation coefficient was − .17 and not statistically significant indicating that the dispersion in collaborative problem solving expectations within teams does not predict the level of the collaborative problems solving norms set by the teams.

5. Discussion

Although it is clear that being perceived as a leader permits an individual to go beyond a formal role in influencing others (e.g., Lord & Maher, 1991), how leaders go about exerting this influence has remained unclear. Given the important function of norms, the present study focused on the leader's impact on team norm formation. We proposed that the expectations that leaders and staff bring to newly formed teams each play a role in determining team norms. Most importantly, we proposed that these expectations interact to form team norms, such that team leaders with high expectations for problem solving behaviors within a team can compensate for staff's low expectations. These hypotheses were supported. We also hypothesized and found that the collaborative problem solving norms that are formed impact a variety of individual behaviors within teams. The present research supports the notion that leaders can play a critical role in shaping team norms.

Emergent leaders differ from their peers in that they make more attempts to influence the team (Bass, 1981). Moreover, effective team leaders shape the development of shared mental models in their teams (Zaccaro, Rittman, & Marks, 2001). We found that one way some emergent leaders influenced their teams and the development of shared mental models was through their impact on norms, perhaps through the team leader's role of gathering and weighing each staff member's opinion in coming up with a single overall decision from the team (Hollenbeck et al., 1995). Kenny & Zaccaro (1983) proposed that emergent leaders judge the varying needs of their teams and change their behavior in response to these needs. Consistent with these descriptions of the leader's role, we found that leaders are likely to have greatest impact on team problem solving norms when they believe that staff expectations for norms are too low and need to be raised.

There has been a remarkable resurgence of interest in the study of groups and work teams (Moreland, Hogg, & Hains, 1994; Sanna & Parks, 1997). Two major reviews have recently summarized advances in theory and research on small groups and on teams in organizations (Ilgen, Hollenbeck, Johnson, & Jundt, 2005; Kerr & Tindale, 2004). Although our understanding of groups and teams is rapidly progressing, Levine & Moreland's (1990) observation that the field is badly fragmented remains valid. In particular, there is scant empirical research on the development of norms (Cialdini & Trost, 1998). Moreover, it is worth noting that Heath & Sitkin (2001), in a survey of select Organizational Behavior (OB) researchers and articles in leading OB journals, concluded that the topic of norms was the most under- researched in Organizational Behavior.

To date, little research has explored the development of formally agreed upon team norms, although team norms play a central role in shaping members' relationships and represent an important research area (Guzzo & Dickson, 1996). Norms can inhibit and impair team performance. For instance, Gersick & Hackman (1990) used the crash of Air Florida Flight 90 to demonstrate how norms can constrain behavior and thus result in major system failures. Our findings suggest that a leader can play an important role in introducing or stemming inappropriate behavioral routines before they become norms. Since emergent team leaders appear to have an inordinate influence on team norms when staff expectations are low, it is important that leaders understand (e.g., through training and coaching) what are appropriate team behaviors and possess appropriate expectations for the frequency with which these behaviors should be displayed. Feldman (1984) elaborates on the importance of monitoring whether team norms are functional and addressing counterproductive norms.

Our research findings also speak to the importance of team composition. While emergent leaders were able to ameliorate the low expectations of their peers, they could not fully compensate for low staff expectations. In addition, when staff expectations were high to begin with, the leader demonstrated little influence. Finally, a leader with low expectations did not negatively influence staff with originally high expectations. Therefore, while leaders play a role in forming team norms, it appears that initial staff expectations can significantly influence the team's overall expectations.

116 S. Taggar, R. Ellis / The Leadership Quarterly 18 (2007) 105–120

This is consistent with the notion that team members participate in the leadership process and leadership in teams is not only the purview of one individual who emerges as the team leader (Day, Gronn, & Salas, 2004). Norm formation is one example of this. Norms help align individual needs, goals, and expected outcomes across the team. The process of creating alignment is a major leadership task (Day et al., 2004) which can involve both the emergent leader and staff or just the staff itself. We conclude that it is critical that teams be composed of staff with high teamwork expectations, and failing that, leaders with high expectations be selected or trained to compensate for the low expectations of staff.

5.1. Limitations and future research

Feldman (1984) specifies four mechanisms through which group norms can be created — explicit statements, critical events, primacy, or carry-over from past situations. We focused on explicit statements. However, Barker (1993) demonstrated a similar process of norm negotiation in a field setting where norms emerged naturally over time, suggesting that our findings may generalize to informal norms. However, this requires further study. In future research, a more naturalistic design would be to allow teams to interact naturally and then assess observationally or through self- reports whether norms had formed and what they were.

Although the participants in our study were students, the team problem solving tasks they performed had meaningful personal consequences. In addition, the participants were largely responsible for the results that they obtained. Thus, we believe that our findings will generalize to organizational settings that use autonomous teams with no formally designated leader. Nevertheless, the precise influence of work experience and organizational context on the present results should be explored in future research.

The more extreme members' initial expectations (either positive or negative), the greater the regression to the mean or convergence when the team norm is set. About 56% of the difference between staff mean expectations prior to entering the team and level of the team norm may constitute a regression artifact because the correlation between staff expectations for collaborative problem solving and the collaborative problem solving norm is .44. However, in this study we did not compare mean team member expectations to team norms but were primarily concerned with the nature of the leader expectations×staff expectations interaction. Regression to the mean would not impact the pattern of the leader expectations×staff expectations interaction plotted in Fig. 1 because both high staff expectations and low staff expectations would converge equally to the mean.

In this study, we describe the impact of collaborative problem solving norms on 5 dimensions of teamwork behavior. Future research should also focus on behaviors important at different stages of the problem solving process and their relationships (both linear and nonlinear) to the outcomes associated with these stages (e.g., the number of ideas generated in a divergent phase of problem solving). We focused on collaborative problem solving norms because of their salience to teams confronting a heuristic task. The formation of other important norms should also be investigated. In this regard, Ehrhart & Naumann (2004) recently proposed a multilevel model of the formation of norms for organizational citizenship behaviors in work groups.

Our focus was on leader and staff roles because conceptual and empirical research by LePine et al. (1997) and Taggar et al. (1999) suggested the importance of both roles in contributing to team functioning. These roles are defined in relation to each other at the group level. Accordingly, we took a construct approach to group level measures where we examined the expectations held by the individuals who later assumed these roles. Our results support the importance of distinguishing leader/staff roles when understanding norm formation. We did not anticipate that expectations would show agreement because the data were collected prior to team formation. However, our approach may have masked individual variations in staff members that are meaningful in their own right. Post-hoc analyses revealed that dispersion in collaborative problem solving expectations within teams did not predict the level of the collaborative problem solving norms set by the teams. Nonetheless, future research should look at variations in behavior, perceptions, affect, and negotiations between team members as expectations are shaped into team norms, in order to build on our model of norm formation. In interpreting our findings, one should avoid an incorrect inference of assuming that the magnitude and nature of the staff expectations–norms relationship would be similar to that of the individual staff member expectations–norms relationship.

Rosenthal (2002) recently summarized 40 years of accumulated evidence that the expectations of experimenters may influence research participants, through a process of covert verbal and nonverbal communication. A number of safeguards in the present investigation make such influence unlikely. First, the study was conducted by a graduate research assistant, blind to the hypotheses. Second, neither the research assistant nor the authors knew who was

117S. Taggar, R. Ellis / The Leadership Quarterly 18 (2007) 105–120

participating in the study or who the team leaders were. Nevertheless, further research on norm formation in teams in different contexts is encouraged to confirm and extend our findings.

This study provides insight into how leader and staff expectations are incorporated into behavioral expectations for the team. About 20 years ago, Feldman (1984, p. 52) wrote:

“Empirical research on norm development and enforcement has substantially lagged descriptive and theoretical work. In large part, this may be due to the methodological problems of measuring norms and getting enough data points either across time or across groups. Until such time as empirical work progresses, however, the usefulness of group norms as a predictive concept, rather than as a post-hoc explanatory device, will be severely limited. Moreover, until it is known more concretely why norms develop and why they are strongly enforced, attempts to change group norms will remain haphazard and difficult to accomplish.”

We believe that there remains a substantial need for additional empirical research on how norms are formed, and once formed, how norms are enforced.

This study also suggests some intriguing possibilities regarding the emergence of leaders. Lord & Maher (1991) propose that there are multiple paths to the attribution of leadership. One is the match between the characteristics possessed by an individual and the perceiver's implicit theory of leader attributes. The other is inferred from outcomes of salient events in the development of the team. In the present study we found that some individuals who were perceived as leaders were active in shaping the norms adopted by the team at the outset of team interaction. These individuals may be perceived as leaders by peers because of their interventions at a salient point in the team's development. However, we also found that the expectations of some perceived leaders did not impact the agreed upon team norms. We note that because their peers held relatively high expectations for collaborative problem solving, these leaders did not have reason to intervene. This suggests that these individuals were perceived to be leaders either because (a) they possessed the attributes consistent with peers' implicit theories of leadership, or (b) they intervened at a later salient event. Future research could explore the impact of leaders at key points in the life of a team (e.g., significant norm violation). Gersick (1991) proposed a punctuated equilibrium model of team development whereby teams alternate between period of inertia and revolution in the themes and behaviors which guide them. Our findings imply that leaders may precipitate these transitions when they perceive the need to do so and shape the future direction adopted by their teams. Our model of leaders actively shaping team development could be integrated with Gersick's (1991) work on inertial revolution in teams. The study of a leader's role in establishing, shaping, and enforcing team norms represents an important avenue for future research.

References

Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interaction. Newbury Park: Sage. Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social behavior. Englewood Cliffs, NJ: Prentice–Hall. Alper, S., Tjosvold, D., & Law, K. S. (1998). Independence and controversy in group decision making: Antecedents to effective self-managing teams.

Organizational Behavior and Human Decision Processes, 74, 33−52. Alper, S., Tjosvold, D., & Law, K. S. (2000). Conflict management, efficacy, and performance in organizational teams. Personnel Psychology, 53,

625−642. Amabile, T. M. (1996). Creativity in Context. Boulder, CO: Westview Press. Amabile, T. M. (1998). How to kill creativity. Harvard Business Review, 76, 77−88. Ancona, D. G., & Caldwell, D. F. (1992). Bridging the boundary: External activity and performance in organizational teams. Administrative Science

Quarterly, 37, 634−666. Argots, L. (1989). Agreement about norms and work-unit effectiveness: Evidence from the field. Basic and Applied Social Psychology, 10,

131−140. Asch, S. E. (1956). Studies of independence and conformity: I. A minority of one against a unanimous majority. Psychological Monographs, 70−140. Bar-Tal, D. (1990). Group beliefs: A conception for analyzing group structure, processes, and behavior. New York: Springer–Verlag. Barker, J. R. (1993). Tightening the iron cage: Concertive control in self-managing teams. Administrative Science Quarterly, 38, 408−437. Bass, B. M. (1949). An analysis of the leaderless group discussion. Journal of Applied Psychology, 33, 627−633. Bass, B. M. (1961). Some aspects of attempted, successful, and effective leadership. Journal of Applied Psychology, 45, 120−132. Bass, B. M. (1981). From leaderless group discussions to the cross-national assessment of managers. Journal of Management, 7, 63−76. Bettenhausen, K., & Murnighan, J. K. (1985). The emergence of norms in competitive decision-making groups. Administrative Science Quarterly,

30, 350−372. Bettenhausen, K. L., & Murnighan, J. K. (1991). The development of an intragroup norm and the effects of interpersonal and structural challenges.

Administrative Science Quarterly, 36, 20−35.

118 S. Taggar, R. Ellis / The Leadership Quarterly 18 (2007) 105–120

Bottger, P. C., & Yetton, P. W. (1987). Improving group performance by training in individual problem solving. Journal of Applied Psychology, 72, 651−657.

Bryk, A. S., & Raudenbush, S. W. (1992). Hierarchical linear models. Newbury Park, CA: Sage. Campion, M. A., Medsker, G. J., & Higgs, A. C. (1993). Relations between work group characteristics and effectiveness: Implications for designing

effective work groups. Personnel Psychology, 46, 823−850. Cannon-Bowers, J. A., Oser, R., & Flanagan, D. C. (1992). Work teams in industry: A selected review and proposed framework. In R. W. Swezey &

E. Salas (Eds.), Teams: Their training and performance (pp. 309−330). Norwood, NJ: Ablex Publishing. Cannon-Bowers, J. A., Tannenbaum, S. I., Salas, E., & Volpe, C. E. (1995). Defining competencies and establishing team training requirements. In

R. Guzzo & E. Salas (Eds.), Team effectiveness and decision making in organizations (pp. 333−380). San Francisco: Jossey–Bass. Chan, D. (1998). Functional relations among constructs in the same content domain at different levels of analysis: A typology of composition models.

Journal of Applied Psychology, 83, 234−246. Chatman, J. A., & Flynn, F. J. (2001). The influence of demographic heterogeneity on the emergence and consequences of cooperative norms in work

teams. Academy of Management Journal, 44, 956−974. Cialdini, R. B., & Trost, M. R. (1998). Social influence: Social norms, conformity, and compliance. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (4th ed.).

The handbook of social psychology, Vol. 2 (pp. 151−192). New York: McGraw–Hill. Coleman, J. S. (1990). Foundations of Social Theory. Cambridge: Harvard University Press. Colman, M. M., & Carron, A. V. (2001). The nature of norms in individual sport teams. Small Group Research, 32, 201−222. Cronshaw, S. F., & Lord, R. G. (1987). Effects of categorization, attribution, and encoding processes on leadership perceptions. Journal of Applied

Psychology, 72, 97−106. Day, D. V., Gronn, P., & Salas, E. (2004). Leadership capacity in teams. Leadership Quarterly, 15, 857−880. Dennis, A. R. (1996). Information exchange and use in group decision making: You can lead a group to information, but you can't make it think. MIS

Quarterly, 20, 433−457. Dennis, A. R., & Valacich, J. S. (1993). Computer brainstorms: More heads are better than one. Journal of Applied Psychology, 78, 531−537. De Souza, G., & Klein, H. J. (1995). Emergent leadership in the group goal-setting process. Small Group Research, 26, 475−496. Earley, P. C. (1999). Playing follow the leader: Status-determining traits in relation to collective-efficacy across cultures. Organizational Behavior

and Human Decision Processes, 80, 192−212. Ehrhart, M. G., & Naumann, S. E. (2004). Organizational citizenship behavior in work groups: A group norms approach. Journal of Applied

Psychology, 89, 960−974. Eisenhardt, K. M. (1989). Making fast strategic decisions in high velocity environments. Academy of Management Journal, 32, 543−576. Feldman, D. C. (1984). The development and enforcement of group norms. Academy of Management Review, 9, 47−53. Feltz, D. L., & Lirgg, C. D. (1998). Perceived team and player efficacy in hockey. Journal of Applied Psychology, 83, 557−564. Firestone, I. J., Lichtman, C. M., & Colamosca, J. V. (1975). Leader effectiveness and leadership conferral as determinants of helping in a medical

emergency. Personality and Social Psychology, 31, 343−348. Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention and behavior: An introduction to theory and research. Reading, MA: Addison–Wesley. Friedkin, N. E. (2001). Norm formation in social influence networks. Social Networks, 23, 167−189. Gallupe, R. B., Dennis, A. R., Cooper, W. H., Valacich, J. S., Bastianutti, L. M., & Nunamaker Jr., J. F., (1992). Electronic brainstorming and group

size. Academy of Management Journal, 35, 350−370. Gavin, M. B., & Hofman, D. A. (2002). Using hierarchical linear modeling to investigate the moderating influence of leadership climate. Leadership

Quarterly, 13, 15−33. Gersick, C. J. G. (1991). Revolutionary change theories: A multi-level exploration of the punctuated equilibrium paradigm. Academy of Management

Review, 16, 10−36. Gersick, C. J. G., & Hackman, J. R. (1990). Habitual routines in task-performing groups. Organizational Behavior and Human Decision Processes,

47, 65−97. Gibson, C. B. (1999). Do they do what they believe they can? Group-efficacy and group effectiveness across tasks and cultures. Academy of

Management Journal, 42, 138−152. Guzzo, R. A., & Dickson, M. W. (1996). Teams in organizations: Recent research on performance and effectiveness. Annual Review of Psychology,

47, 307−338. Hackman, J. R. (1976). Group influences on individuals. In M. D. Dunnette (Ed.), Handbook of industrial and organizational psychology

(pp. 1455−1525). Chicago: Rand McNally. Hare, A. P. (1981). Group size. American Behavioural Scientist, 24, 695−708. Heath, C., & Sitkin, S. (2001). Big-B versus Big-O: What is organizational about organizational behavior? Journal of Organizational Behavior, 22,

43−58. Hofmann, D. A. (1997). An overview of the logic and rationale of hierarchical linear models. Journal of Management, 23, 723−744. Hofmann, D. A., Griffin, M. A., & Gavin, M. B. (2000). The application of hierarchical linear modeling to organizational research. In K. J.

Klein & S. W. J. Kozlowski (Eds.), Multilevel theory, research, and methods in organizations: Foundations, extensions, and new directions (pp. 467−511). San Francisco, CA: Jossey–Bass.

Hollenbeck, J. R., Ilgen, D. R., Sego, D. J., Hedlund, J., Major, D. A., & Phillips, J. (1995). Mutlilevel theory of team decision making: Decision performance in teams incorporating distributed expertise. Journal of Applied Psychology, 80, 292−316.

Hollenbeck, J. R., Lepine, J. A., & Ilgen, D. R. (1996). Adapting to roles in decision-making teams. In K. R. Murphy (Ed.), Individual differences and behavior in organizations (pp. 300−333). San Francisco: Jossey–Bass.

Ilgen, D. R., Hollenbeck, J. R., Johnson, M., & Jundt, D. (2005). Teams in organizations: From input–process–output models to IMOI models. In S. T. Fiske, D. L. Schacter & A. E. Kazdin (Eds.), Annual review of psychology, Vol. 56 (pp. 517−543). Palo Alto, CA: Annual Reviews.

119S. Taggar, R. Ellis / The Leadership Quarterly 18 (2007) 105–120

James, L. R., Demaree, R. G., & Wolf, G. (1983). rwg: An assessment of within-group interrater agreement. Journal of Applied Psychology, 78, 306−309.

Jehn, K. A., & Mannix, E. A. (2001). The dynamic nature of conflict: A longitudinal study of intragroup conflict and group performance. Academy of Management Journal, 44, 238−251.

Keller, R. T. (2001). Cross-functional project groups in research and new product development: Diversity, communications, job stress, and outcomes. Academy of Management Journal, 44, 547−555.

Kenny, D. A., & Zaccaro, S. J. (1983). An estimate of variance due to traits in leadership. Journal of Applied Psychology, 68, 678−685. Kent, R. L., & Moss, S. E. (1990). Self-monitoring as a predictor of leader emergence. Psychological Reports, 66, 875−881. Kent, R. L., & Moss, S. E. (1994). Effects of sex and gender-role on leader emergence. Academy of Management Journal, 37, 1335−1346. Kerr, N. L., & Tindale, R. S. (2004). Group performance and decision making. In S. T. Fiske, D. L. Schacter, & C. Zahn-Waxler (Eds.), Annual review

of psychology, Vol. 55 (pp. 623−655). Palo Alto, CA: Annual Reviews. Kirkman, B. L., Tesluk, P. E., & Rosen, B. (2001). Assessing the incremental validity of team consensus ratings over aggregation of individual-level

data in predicting team effectiveness. Personnel Psychology, 54, 645−654. Kozlowski, S. W., & Klein, K. J. (2000). A multilevel approach to theory and research in organizations: Contextual, temporal, and emergent

processes. In K. J. Klein & S. W. Kozlowski (Eds.), Multilevel theory, research and methods in organizations: Foundations, extensions, and new directions (pp. 3−90). San Francisco: Jossey Bass.

LePine, J. A., Hollenbeck, J. R., Ilgen, D. R., & Hedlund, J. (1997). Effects of individual differences on the performance of hierarchical decision- making teams: Much more than g. Journal of Applied Psychology, 82, 803−811.

Levine, J. M., Higgins, E. T., & Choi, H. (2000). Development of strategic norms in groups. Organizational Behavior and Human Decision Processes, 82, 88−101.

Levine, J. M., & Moreland, R. L. (1990). Progress in small group research. In M. R. Rosenzweig & L.W. Porter (Eds.), Annual review of psychology, Vol. 41 (pp. 585−634). Palo Alto, CA: Annual Reviews.

Lord, R. G. (1977). Functional leadership behavior: Measurement and relation to social power and leadership perceptions. Administrative Science Quarterly, 22, 114−133.

Lord, R. G., & Maher, K. J. (1991). Leadership and information processing. London: Routledge. Maier, N. R. (1967). Assets and liabilities in group problem solving: The need for an integrative function. Psychological Review, 74, 239−249. Moreland, R. L., Hogg, M. A., & Hains, S. C. (1994). Back to the future: Social psychological research on groups. Journal of Experimental Social

Psychology, 30, 527−555. Morgeson, F. P., & Hofmann, D. A. (1999). The structure and function of collective constructs: Implications for multilevel research and theory

development. Academy of Management Review, 24, 249−265. Mullen, B., & Copper, C. (1994). The relation between group cohesiveness and performance: An integration. Psychological Bulletin, 115,

210−227. Mumford, M. D. (1986). Leadership in the organizational context: A conceptual approach and its application. Journal of Applied Social Psychology,

16, 212−226. Neuman, G. A., & Wright, J. (1999). Team effectiveness: Beyond skills and cognitive ability. Journal of Applied Psychology, 84, 376−389. Okhuysen, G. A. (2001). Structuring change: Familiarity and formal interventions in problem solving groups. Academy of Management Journal, 44,

794−808. Parsons, T. (1937). The structure of social action. New York: McGraw–Hill. Roethlisberger, F. J., & Dickson, W. J. (1939). Management and the worker. Cambridge, MA: Harvard University Press. Rosenthal, R. (2002). Covert communication in classrooms, clinics, courtrooms, and cubicles. American Psychologist, 47, 839−849. Sanna, L. J., & Parks, C. D. (1997). Group research trends in social and organizational psychology: Whatever happened to intragroup research?

Psychological Science, 8, 261−267. Schneider, C. E., & Goktepe, J. R. (1983). Issues in emergent leadership: The contingency model of leadership, leader sex, leader behavior. In H. H.

Blumberg, A. P. Hare, V. Kent, & M. F. Davies (Eds.), Small groups and social interaction, Vol. 1. Chicester, England: John Wiley. Seers, A. (1989). Team-member exchange quality: A new construct for role-making research. Organizational Behavior and Human Decision

Processes, 43, 118−135. Seers, A., Petty, M. M., & Cashman, J. F. (1995). Team-member exchange under team and traditional management: A naturally occurring quasi-

experiment. Group and Organization Management, 20, 18−38. Shamir, B., Zakay, E., Breinin, E., & Popper, M. (1998). Correlates of charismatic leader behavior in military units: Subordinates' attitudes, unit

characteristics, and superiors' appraisals of leader performance. Academy of Management Journal, 41, 387−410. Sherif, M. (1936). The psychology of social norms. Oxford, England: Harper. Shore, L. M., & Barksdale, K. (1998). Examining degree of balance and level of obligation in the employee relationship: A social exchange approach.

Journal of Organizational Behavior, 19, 731−744. Spich, R. S., & Keleman, K. (1985). Explicit norm structuring process: A strategy for increasing task-group effectiveness. Group and Organization

Studies, 10, 37−59. Steiner, I. (1972). Group process and productivity. New York: Academic Press. Stevens, M. J., & Campion, M. A. (1994). The knowledge skill and ability requirements for teamwork: Implications for human resource management.

Journal of Management, 20, 503−530. Taggar, S. (2002). Individual creativity and group ability to utilize individual creative resource: A multilevel model. Academy of Management

Journal, 45, 315−330. Taggar, S., & Brown, T. C. (2001). Problem-solving team behaviors: Development and validation of BOS and a hierarchical factor structure. Small

Group Research, 32, 698−726.

120 S. Taggar, R. Ellis / The Leadership Quarterly 18 (2007) 105–120

Taggar, S., Hackett, R., & Saha, S. (1999). Leadership emergence in autonomous work teams: Antecedents and outcomes. Personnel Psychology, 52, 899−926.

Taggar, S., & Neubert, M. (2004). The impact of poor performers on team outcomes: An empirical examination of Attribution Theory. Personnel Psychology, 57, 935−968.

Thompson, J. D. (1967). Organizations in action. New York: McGraw–Hill. Tjosvold, D., Poon, M., & Zi-you, Y. (2005). Team effectiveness in China: Cooperative conflict for relationship building. Human Relations, 58,

341−367. Tuckman, B. W., & Jensen, M. C. (1977). Stages of small group development revisited. Group and Organization Studies, 2, 419−427. Tyre, M. J., & Orlikowski, W. J. (1994). Windows of opportunity: Temporal patterns of technological adaptation in organizations. Organizational

Science, 5, 98−118. Vroom, V. (1969). Industrial social psychology. In G. Lindzey & E. Aronson (Eds.), The handbook of social psychology. Reading, MA: Addison–

Wesley. Wageman, R. (1995). Interdependence and group effectiveness. Administrative Science Quarterly, 40, 145−180. Wageman, R. (1997). Case study: Critical success factors for creating superb self-managing teams at Xerox. Compensation and Benefits Review, 29,

31−41. Wentworth, D. K., & Anderson, L. R. (1984). Emergent leadership as a function of sex and task type. Sex Roles, 11, 513−523. Wheeler, B. C., & Valacich, J. S. (1996). Facilitation, GSS, and training as sources of process restrictiveness and guidance for structured group

decision making: an empirical assessment. Information Systems Research, 7, 429−450. Zaccaro, S. J., Rittman, A. L., & Marks, M. A. (2001). Team leadership. Leadership Quarterly, 12, 451−483.