Group & Organization Management http://gom.sagepub.com/ Peer Justice and Teamwork Process Russell Cropanzano, Andrew Li and Lehman Benson III Group & Organization Management 2011 36: 567 originally published online 2 August 2011 DOI: 10.1177/1059601111414561 The online version of this article can be found at: http://gom.sagepub.com/content/36/5/567 Published by: http://www.sagepublications.com Additional services and information for Group & Organization Management can be found at: Email Alerts: http://gom.sagepub.com/cgi/alerts Subscriptions: http://gom.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations: http://gom.sagepub.com/content/36/5/567.refs.html >> Version of Record - Sep 12, 2011 OnlineFirst Version of Record - Aug 2, 2011 What is This? Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 414561 © The Author(s) 2011 GOM36510.1177/1059601111414561Cropa nzano et al.Group & Organization Management Peer Justice and Teamwork Process Group & Organization Management 36(5) 567­–596 © The Author(s) 2011 Reprints and permission: http://www. sagepub.com/journalsPermissions.nav DOI: 10.1177/1059601111414561 http://gom.sagepub.com Russell Cropanzano1, Andrew Li2, and Lehman Benson III1 Abstract We articulate a teamwork process model of peer justice, defined as a shared perception regarding how individuals who work together within the same unit and who do not have formal authority over each other judge the fairness with which they treat one another. We argue that unit-level judgments of procedural and interpersonal fairness may influence team process, characterized by such things as better communication, better coordination, and more mutual support for members. These team processes, in turn, promote higher team performance and unit-level citizenship behaviors. These ideas were tested among teams of business students working on a semester-long class project. Findings generally supported our theoretical model, attesting to the importance of peer justice.We also found that peer procedural justice strength influenced team outcome variables, including performance, through teamwork processes.The theoretical and practical implications of these findings are discussed. Keywords peer justice, organizational justice, team effectiveness, teamwork process 1 University of Arizona, Tucson, AZ, USA West Texas A&M University, Canyon, TX, USA 2 Corresponding Author: Andrew Li, West Texas A&M University, Department of Management, College of Business, WTAMU Box 79016, Canyon, TX 79016, USA Email: ali@wtamu.edu Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 568 Group & Organization Management 36(5) An important consideration for workers is whether they are being treated fairly by their employing organization and by decision makers with authority over them (Cohen-Charash & Mueller, 2007). These perceptions of organizational justice have been found to be of considerable importance. Employees who perceive that they have been treated unfairly report poorer work attitudes (Cohen-Charash & Spector, 2001) and more stress (Cropanzano & Wright, 2011). Organizations are harmed as well. When workers perceive a lower level of justice, they show lower job performance, perform fewer organizational citizenship behaviors (OCBs), and engage in more counterproductive work behaviors (Cohen-Charash & Mueller, 2007). Historically, most organizational justice (the words justice and fairness have been used interchangeably in social science research, see Cropanzano, Stein, & Nadisic, 2011, Chapter 1) research has tended to focus on justice perceptions emanating from authority figures (for evidence, see Ambrose & Schminke, 2001; Tyler & Blader, 2000). This focus is consistent with the dominance of hierarchically structured work units in the workplace that have a manager as the ultimate decision maker with the rest of the employees following his or her lead. However, an increasing number of organizations have adopted the use of work teams that tend to have greater discretion as coworkers work interdependently to accomplish a common goal (Bosch-Sijtsema, Fruchter, Vartiainen, & Ruohomaki, 2011). In the process of doing so, they hold each other accountable for the final outcome (e.g., Banker, Field, Schroeder, & Sinha, 1996; Barker, 1993; Bishop & Scott, 2000). One implication of this change is that teammates are likely to play an important role in shaping an employee’s justice perceptions (Lavelle, Rupp, & Brockner, 2007). This suggests that the traditional justice research, which tends to focus on an authority figure (be it a supervisor or the organization as a whole), may need to be extended to include peers as an alternative source of justice perceptions. Changes of this type have led Ambrose and Schminke (2001) to call for new paradigms of justice research. Specifically, more research is needed on fairness perceptions that concern how peers treat one another. As defined here, peer justice refers to a shared perception regarding how individuals who work together within the same unit and who do not have formal authority over each other judge the fairness with which they treat one another. Previous researchers have referred to these collective judgments as “intraunit justice climate” (e.g., Cropanzano, Li, & James, 2007; Li & Cropanzano, 2009). We prefer the simpler term “peer justice,” in part, because it avoids the unnecessary confusion with the older concept of “justice climate,” as discussed below. Furthermore, a key goal of the present study is to integrate research on collective justice perceptions with related investigations of teamwork processes. Teamwork Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 Cropanzano et al. 569 researchers tend to use the terminologies “peers/coworkers,” which are consistent with the terminology “peer justice” we use in this study. Two key features distinguish peer justice from much, though not all, previous fairness research. First, peer justice pertains to employees who do not have formal authority over each other (Aquino, Tripp, & Bies, 2006). Examples of such relationships abound in many organizations, such as engineers working in a cross-functional team or members of various ad hoc committees in educational institutions. Our definition of peer justice also includes scenarios whereby members of the unit may not be identical in their organizational ranking, but they have no formal authority over each other in the unit. For example, faculty research project teams may involve members who occupy different academic ranks (full, associate, or assistant professors), but they are considered peers within the project team. Peer justice is distinct from justice climate, which, as defined, refers to the shared perceptions among unit members as to how their unit has been treated by an authority figure (Colquitt, Noe, & Jackson, 2002; Ehrhart, 2004; Naumann & Bennett, 2000). Justice climate is explicitly concerned with an organizational authority as the source of justice perceptions. Peer justice is concerned with coworkers as a source of fairness. Research on multifoci justice has shown that individuals can distinguish justice perceptions coming from different sources (Frazier, Johnson, Gavin, Gooty, & Snow, 2010; Liao & Rupp, 2005), and so it is important to distinguish coworkers from supervisors (Lavelle et al., 2007). With respect to peer justice, this was directly tested in the study by Li, Cropanzano, & Benson (2007) who found that peer justice (intraunit justice climate) explains unique variance on outcome variables beyond the effects of justice climate. Second, peer justice is a unit-level construct. We conceptualize peer justice as a facet of organizational climate which refers to a shared perception that members within a unit (organization, department, or team) develop regarding certain aspects of their social environment and is posited to shape unit members’ attitudes and behavior (Kuenzi & Schminke, 2009; Schneider, Ehrhart, & Macey, 2010). It is important to distinguish organizational climate, which is a unit-level construct, from psychological climate, which is an individual-level construct pertaining to individuals’ subjective perceptions of and the meaning that they impute to the work environment (James & James, 1989). In other words, to conceptualize a construct such as peer justice as an organizational climate variable, substantial agreement among members of the same unit (within-unit homogeneity) and variations across units (acrossunit variability) regarding the particular facet of the social environment are required (Kozlowski & Klein, 2000). Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 570 Group & Organization Management 36(5) The appropriate unit of analysis has ignited considerable debate in climate research (Kozlowski & Klein, 2000). As climate scholars point out, groups/ teams, departments, divisions, and organizations are all relevant units of analysis. Although it is possible that individuals may interact with others outside of their formally defined social unit, the bulk of interactions most likely still occur within the boundary of their primary work group (Liao & Rupp, 2005). The unit of analysis that we choose in this study (i.e., team) is consistent with the argument that work teams provide the most prominent social context (Liao & Rupp, 2005). In addition, the interdependent nature of most work teams may result in social interactions among team members which may trigger a collective sense-making process, a prerequisite for the emergence of unit-level constructs. Of course, peer justice may also emerge within the boundary of larger units, such as department, insofar as members within these units work interdependently and develop similar perceptions which can be aggregated to a higher level. A Teamwork Process Model of Peer Justice Morgeson and Hofmann (1999) argued that to justify the scientific legitimacy of a unit-level construct, both its structure and functions need to be clearly articulated. Structurally, scholars need to identify how unit-level constructs emerge. Functionally, scholars need to define the causal effects of these constructs and consider their effects beyond their individual-level counterparts. Following Morgeson and Hofmann (1999), we explore the structure of peer justice and conceptualize peer justice as a collective or unitlevel construct. We then shift our focus to the functional analysis of peer justice by examining why peer justice should predict citizenship behaviors and team performance. We argue that peer justice boosts the quality of member interactions, which, in turn, is the proximal determinant of effective work behavior. Structure of Peer Justice Historically, organizational justice researchers have divided fairness perceptions into three families (for a review, see Cohen-Charash & Spector, 2001). Distributive justice refers to the perceived fairness of the outcomes received, procedural justice refers to the fairness of decision-making processes, and interactional justice refers to the fairness of the interpersonal treatment that one receives from another person. For its part, interactional justice is sometimes subdivided into two subdimensions: interpersonal justice, which concerns Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 Cropanzano et al. 571 the extent to which individuals treat each other with respect and dignity, and informational justice, which concerns the extent to which information is communicated timely and accurately (e.g., Colquitt, 2001). Other scholars have taken a slightly different view, suggesting that perhaps interactional justice could be measured by “a broad range of interpersonal treatment” (Roch & Shanock, 2006, p. 317; see also Bies, 2001). Given that we could not resolve this conceptual debate within the present study, we assessed interactional justice as a single dimension, using a previously validated measure of how coworkers treat each other interpersonally. Our measure of interactional justice measure, developed by Donovan, Drasgow, and Munson (1998), emphasizes the interpersonal subdimension of interactional justice. For precision, we refer to the specific construct as “interpersonal justice” and return to the two-factor model of interactional fairness when interpreting our findings in the Discussion section. Following past research, Li and colleagues (2007) argued that peer justice also falls into the three-factor structure. Specifically, peer distributive justice is defined as the shared perceptions that unit members receive what they deserve based on their contribution. Peer procedural justice is defined as the extent to which unit members use fair procedures in the decision-making process. Peer interactional justice is defined as the interpersonal fairness with which unit members treat one another. Li and colleagues (2007) found empirical evidence in support of the three-factor structure. In the present study, we focus on procedural and the interpersonal dimension of interactional justice because justice research at the unit level tends to focus on these two dimensions (e.g., Colquitt et al., 2002; Liao & Rupp, 2005; Naumann & Bennett, 2000). Of course, our present emphasis on peer procedural and interpersonal justice should not be construed as the lack of importance of other dimensions. This is an empirical question which should be examined in future research. Unit-Level Perceptions of Peer Justice Peer justice is conceptualized as a unit-level variable (see. Ambrose & Schminke, 2007). For this conceptualization to be viable, there needs to be some evidence that individuals within a work team share a collective sense of fairness with regard to how they treat one another. In their discussions of work groups, Hackman (1992) makes a distinction between discretionary stimuli and ambient stimuli. Discretionary stimuli have been widely studied by previous justice scholars. These pertain to how an individual is treated by another person, such as his or her supervisor or a coworker. As it is “discretionary” Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 572 Group & Organization Management 36(5) (i.e., not all people are treated the same way), it exists at the individual level of analysis. Ambient stimuli are those that are experienced by all members of a collective, such as a particular human resource policy. In team settings, ambient stimuli are especially important. Peer justice is an ambient stimulus because it refers to how team members behave toward one another. Such a phenomenon can only exist if the target behaviors occur over a long enough period of time so that they come to characterize the team as a whole (Rupp & Paddock, 2010). Past research (Li & Cropanzano, 2009; Rupp & Paddock, 2010; Schneider & Reichers, 1983) has provided three mechanisms to account for why shared perceptions of fairness are likely to emerge over time. First, according to the attraction-selection-attrition (ASA) model (Schneider, 1987), organizations tend to attract and retain comparable kinds of people who respond in a common fashion to workplace events. The resemblance among workers within a firm could cause them to share similar opinions regarding justice (Naumann & Bennett, 2000). Second, according to the social information processing model, workers may share information and thereby come to a common interpretation of ambiguous events (Salancik & Pfeffer, 1978). Over time, this could cause greater correspondence among employees’ justice judgments (Roberson, 2006). Third, according to the structualist approach, the structure of an organization, such as its size and the centralization of decision-making authority, may also influence the etiology of climates. This argument suggests that certain “objective” features of the work environment may condition unit members’ tendency to (or not to) develop similar perceptions. These theoretical arguments, taken together, suggest that peer justice should be evaluated at the unit level. In the present study, we examine whether peer justice can be aggregated to the unit level. In addition, we also examine whether peer justice explains unique variance on outcome variables beyond the effects of its individual-level counterparts. Peer Justice and Teamwork Process With the structural issues explored, we now turn to our theoretical model, which is presented in Figure 1. Stated generally, we argue that peer procedural justice boosts two important teamwork processes—the manner in which the team performs the task (task teamwork process) and the manner in which team members relate to one another (interpersonal teamwork process). Peer interpersonal justice, which is less directly relevant to task behavior, is only predicted to affect interpersonal teamwork process. The two team process variables, in turn, affect effective work behaviors. Task teamwork Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 573 Cropanzano et al. Task Teamwork process Communicaon Coordinaon Contribuon Peer Procedural Jusce Task Performance Team Cizenship Behavior Peer Interpersonal Jusce Helping Interpersonal Teamwork process Cohesion Effort Loyalty Support Figure 1. Hypothesized Model process enhances performance, whereas interpersonal teamwork process increases citizenship behaviors. Below we explain our model in greater detail, beginning by discussing the nature of teamwork process. Defining and understanding teamwork process. A recent trend in group/team research has been an increased emphasis on teamwork process, which reflects “members’ interdependent acts that convert inputs to outcomes through cognitive, verbal, and behavioral activities directed toward organizing taskwork to achieve collective goals” (Marks, Mathieu, & Zaccaro, 2001, p. 357). Research has uncovered a number of distinct process variables, though they tend to be correlated. To organize these, Hoegl and Gemuenden (2001) have devised an inclusive taxonomy. According to their taxonomy, teamwork processes include six facets which represent how members perform functions allowing them to “solve the objective problem to which the group is committed” and to “build, strengthen, and regulate group life” (Gladstein, 1984, p. 500). The six facets of teamwork process include communication, defined as the amount, frequency, and openness of exchanges among team members; coordination, defined as the synchronization and adjustment of activities and behavior to achieve mutual alignment among team members; balance of member contribution, defined as the extent to which teams allocate tasks to those who have the capacities to perform them; mutual support, defined as the extent to which team members resolve their conflict cooperatively or Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 574 Group & Organization Management 36(5) competitively; effort, defined as the extent to which team members are willing to contribute to the success of the team; cohesion, defined as the extent to which members feel a strong attachment to each other and a desire to remain as part of the team. Previous work has found that these facets cluster into two higher order dimensions. Task teamwork process includes communication, coordination, and contribution. Interpersonal teamwork process includes cohesion, effort, and support (Dayan & Benedetto, 2008). Hoegl and Gemuenden’s (2001) taxonomy has been validated in a number of empirical studies (e.g., Easley, Devaraj, & Crant, 2003; Hoegl, Ernst, & Proserpio, 2007; Hoegl, Parboteeah, & Gemuenden, 2003). As shown in Figure 1, we propose that each dimension of teamwork process predicts a different type of effective work behavior. Task teamwork process, with its emphasis on core performance, should be associated with team performance (Hoegl & Gemuenden, 2001). Interpersonal teamwork process, with its emphasis on interpersonal relationships, should be a significant predictor of team citizenship behaviors (Ehrhart, 2004). We explore these specific predictions next. Peer justice and teamwork process. As shown in Figure 1, we hypothesize that peer procedural justice affects both dimensions of teamwork process. This is because procedural justice affects teamwork in two distinct ways. First, procedures convey whether one is valued and respected by the group (Tyler & Blader, 2000). A high level of peer procedural justice creates a favorable social environment as it induces team members to participate in decision-making processes, apply decision-making rules in a consistent manner, among others (Leventhal, 1976). This may allow them to generate positive feelings such as loyalty and commitment toward each other (Liao & Rupp, 2005; Yang, Mossholder, & Peng, 2007) and set the tone for and reinforce positive interactions among them. In contrast, lower peer procedural justice provides cues to team members that they are not valued members of the community. As such, they are less likely to engage in positive interpersonal processes, such as providing support to each other, putting forth sufficient effort, and engaging in activities aimed at promoting team cohesion. The strengthened interpersonal bond, as a result of a high level of peer procedural justice, may motivate them to engage in OCB to benefit the team (Cardona, Lawrence, & Bentler, 2004). In other words, OCB allows team members to maintain their identification with their team (Tyler & Blader, 2000) and to repay the goodwill gesture shown to them by their teammates (Cropanzano & Mitchell, 2005; Ehrhart, 2004). Hypothesis 1: Peer procedural justice will be positively associated with interpersonal teamwork process. Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 Cropanzano et al. 575 Hypothesis 2: Peer procedural justice will be positively associated with team OCB, but this effect will be mediated by interpersonal teamwork process. Second, procedures provide an institutional framework within which team members can present ideas, hold each other accountable, assign roles, and so on (Tangirala & Ramanujam, 2008). This procedural framework is essential to group performance. A high level of procedural justice creates a mechanism to safeguard against the abuse of power. As such, it provides a psychologically safe environment for team members to share information with others without the fear that the information will be used against them (Edmondson, 1999). The fair climate in the team may also ease team members’ fears of being exploited (Van den bos, 2001). Therefore, they are more likely to align their behavior with others in the team and to contribute to the team effort based on their expertise. We argue that a higher level of task teamwork may contribute to a higher level of team performance for two reasons. First, a high level of peer procedural justice may reduce the likelihood of social loafing (Price, Harrison, & Gavin, 2006). When team members treat each other fairly, they know that their interest will be protected. Second, a higher level of task teamwork may also enhance team efficiency and reduce coordination loss (de Jong & Elfring, 2010). In other words, when they constantly communicate with each other and synchronize each others’ activities, they are more likely to fully capitalize on each others’ strengths and make up for each others’ weaknesses, identify potential speed bumps, and bring team performance to a higher level (Colquitt et al., 2002). Hypothesis 3: Peer procedural justice will be positively associated with task teamwork process. Hypothesis 4: Peer procedural justice will be positively associated with team performance, but this effect will be mediated by task teamwork process. Peer interpersonal justice also has important effects, though we expect that these effects will tend to fall more directly on interpersonal teamwork process. According to the social exchange framework, employees are motivated to establish long-lasting social exchange relationships with those who show them goodwill gestures (Cropanzano & Mitchell, 2005). Such exchange relationships involve social and emotional currencies such as mutual identification, commitment, and support. Past research has suggested that one catalyst of the formation of social exchange relationship is interpersonal justice or the Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 576 Group & Organization Management 36(5) level of respect and dignity that exchange parties show to each other (Rupp & Cropanzano, 2002). When employees feel that they have been treated fairly interpersonally, they may reciprocate through cooperative and discretionary behavior. This argument suggests that interpersonal justice enhances interpersonal transactions (Masterson, Lewis, Goldman, & Taylor, 2000; Tekleab, Takeuchi, & Taylor, 2007). When team members believe that they are treated with respect and dignity by most of their teammates, they are more likely to develop favorable attitudes toward them. Such positive attitudes may motivate them to create and maintain a social bond with their teammates. For example, they may be more willing to provide support to their teammates and use their teams as part of their self-definition. They are also less likely to withhold effort in a self-serving manner. Such strong interpersonal relationships may in turn motivate team members to engage in voluntary citizenship behaviors on behalf of their coworkers. In other words, such voluntary citizenship behavior serves as a way that allows team members to return the favor to their teammates for the favorable treatments they receive and further cement the high-quality interpersonal relationship in the team. From this, follow our final two hypotheses: Hypothesis 5: Peer interpersonal justice will be positively associated with interpersonal teamwork process. Hypothesis 6: Peer interpersonal justice will be positively associated with team OCB, but this effect will be mediated by interpersonal teamwork process. Method Research Participants Following the work of Lavelle et al. (2009, Study 2), we tested our hypotheses with 170 teams of undergraduate students enrolled in the business school of a large state university. Team members were cohorts of students newly admitted into the Business College. As part of the program requirements, they were required to work on several semester-long team projects in their core classes. In all classes, team members were graded based on their collective performance, which created a high level of interdependence among coworkers. Consequently, these were equivalent to real-world teams with practical consequences for their members. Membership was determined by the Business College and remained stable across all the classes. Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 Cropanzano et al. 577 Procedure Participants received an email from the first author soliciting their assistance in this study. Those who indicated their willingness to participate were given two surveys which they filled out and returned though electronic mail to the first author. In exchange for their voluntary participation, students earned extra credits toward their final grade in their management class. The measures paralleled the presumed causal order. Peer justice was assessed at Time I, about 7 weeks after the inception of the team. Eight weeks later, we examined the six dimensions of teamwork process and the two dimensions of citizenship behavior. In addition to the survey, we also obtained team performance data from the instructor (to be described below). These 170 teams include a total of 814 members. Team size ranged from 3 to 6 (M = 4.79; SD = .54). Out of a total of 814 members in these teams, 625 responded to both surveys, representing a response rate of 77%. All 170 teams had at least two members responding to both surveys, resulting in a group-level response rate of 100%. The average group response rate was 78%, ranging from 40% to 100% (SD = .19). Although the response rate is higher than the response rate using organization samples (52.7%, Baruch & Holtom, 2008), it is still relatively low, given that a student sample was used. One explanation for the low response rate is that the second survey was distributed toward the end of the semester. As participants were busy studying for their final exam, some of them might have overlooked the reminder for the survey. We compared the justice ratings of participants who completed both Time I and Time II surveys with those who only completed the Time I survey. No significant difference was found between the two groups. Measures Participants responded to all the measures with a 5-point Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly agree). Although climate scholars have used both the direct consensus model (scale items referencing self-perceptions) and the referent-shift model (scale items referencing the higher level) to measure organizational climate, we follow Kozlowski and Klein’s (2000) recommendation that if possible, the referent-shift model be preferred over the direct consensus model when assessing unitlevel constructs. Peer justice. A five-item Procedural Justice scale was developed based on the criteria proposed by Leventhal (1976). One sample item of the measures is “my teammates are able to express their views and feelings about the way Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 578 Group & Organization Management 36(5) decisions are made in the team.” Interpersonal justice was measured with an existing four-item scale developed and validated by Donovan and colleagues (1998). An example item is “teammates put each other down.” Teamwork process. Teamwork process was assessed with the six-dimension measure designed and validated by Hoegl and Gemuenden (2001). Communication was measured with a 10-item scale. A sample item of the scale is “there is frequent communication within the team.” Coordination was measured with a four-item scale. A sample item of the scale is “the work within the project is closely harmonized.” Balance of contribution was measured with a three-item scale. A sample item of the scale is “team members contribute to the achievement of the team’s goals in accordance with their specific potentials.” Mutual support was measured with a six-item scale. A sample item of the scale is “team members help and support each other as best as they can.” Effort was measured with a four-item scale. A sample item of the scale is “every team member fully pushes the projects.” Cohesion was measured with a ten-item scale. A sample item of the scale is “it is important for the members of our team to be part of these projects.” As discussed in the Introduction, we expected these specific dimensions to be organized into two higher order constructs. Task teamwork process should be composed of communication, coordination, and contribution. Interpersonal teamwork process should be composed of cohesion, effort, and support (Dayan & Benedetto, 2008). Team-level citizenship behavior. Following past research (Van der Vegt, Van de Vliert, & Oosterhof, 2003), we focused on two dimensions of team-level citizenship behavior that are relevant to how teammates treat one another. Helping behavior, which captures team members’ tendency to share each other’s workload, was measured with a four-item scale (Van der Vegt et al., 2003). A sample item of the scale is “team members help out those who have heavy workloads.” Loyal behavior, which denotes team members’ propensity to go above and beyond the call of duties, was measured with a threeitem scale (Van der Vegt et al., 2003). A sample item of the scale is “team members never avoid extra duties and responsibilities for the projects.” Although the specific facets of OCB are sometimes distinguished, a metaanalytic review by LePine, Erez, and Johnson (2002) found that the strong correlations exhibited among OCB dimensions allow researchers to combine them into a higher order construct. Given this, we treated OCB as a superordinate construct. Performance. Out of the 170 teams, 79 were required to each submit a report on human resources planning. The report included a brief description of the product/services that the team intended to bring to the market. In Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 Cropanzano et al. 579 addition, team members were required to identify business functions within the team (such as finance and marketing) and to assign team members to each business function. The report also included their self-assessment of the gaps in their skills/knowledge and action plans to fill the gaps. Finally, they were also asked to convene an advisory board to oversee the operations of the team. The project was worth 50 points of their final grade. The other 91 teams were required to submit a report in which they identified business functions within the team and assigned members to the functions. They were asked to provide justifications for their decision. In the case when no appropriate individuals within the team could perform the designated functions, they were instructed to identify criteria that they would use to select the appropriate candidates externally. This project was worth 100 points for these teams. As different grading scales were used (although all teams were taught by the same professor who graded the team projects), we obtained the standardized score for our subsequent analyses. Results Means, standard deviations, coefficient alphas, and scale intercorrelations appear in Table 1. Considerations of Structure Following Anderson and Gerbing’s (1988) two-step approach, we first considered the structure of our different measures using confirmative factor analysis (CFA). To conserve degrees of freedom, we analyzed peer justice (procedural and interpersonal), teamwork process (task and interpersonal), and OCB (helping and loyalty) separately. As our analysis involves both levels (individual and unit), we provide below the result of CFAs at both levels. Factor structure of peer justice. At the individual level, the predicted twofactor structure of peer justice (χ2 = 106.23, df = 22, χ2 / df = 4.83; Incremental Fit Index [IFI] = .96; Comparative Fit Index [CFI] = .96; Root Mean Square Error of Approximation [RMSEA] = .078), which separated procedural and interpersonal fairness, was superior to the one-factor structure that collapsed both types of peer justice into a single dimension (χ2 = 116.253, df = 23, χ2 / df = 5.054; IFI = .95; CFI = .95; RMSEA = .081). The difference between the two models was significant (Δχ2 = 10.023, df = 1, p < .05). At the team level, the predicted two-factor structure of peer justice (χ2 = 59.30, df = 22, χ2 / df = 2.70; IFI = .95; CFI = .95; RMSEA = .087) was superior to the one-factor structure Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 580 Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 3.96 4.23 3.70 3.70 3.71 3.91 3.36 3.55 3.81 3.38 92.14 0.34 0.40 0.43 0.47 0.50 0.43 0.65 0.47 0.51 0.52 7.08 SD .80 .74** .57** .57** .57** .57** .58** .57** .46** .45** −.01 1 Note: Reliabilities are reported in the diagonals. N = 170. *p < .05. **p < .01. 1. Procedural justice 2. Interpersonal justice 3. Communication 4. Coordination 5. Balance of contribution 6. Mutual support 7. Effort 8. Cohesion 9. Helping behavior 10. Loyalty 11. Performance M .83 .48** .53** .55** .57** .55** .54** .44** .43** .01 2 .89 .85** .81** .82** .84** .79** .75** .70** .12 3 .85 .79** .80** .80** .78** .73** .71** .16** 4 .69 .82** .85** .79** .75** .67** .10 5 .93 .76** .81** .78** .69** .05 6 7 .87 .83** .71** .73** .12 Table 1. Means, Standard Deviations, and Intercorrelations of Measured Variables at the Unit Level .89 .73** .74** .08 8 .93 .73** .06 9 .82 .04 10 — 11 Cropanzano et al. 581 (χ2 = 64.35, df = 23, χ2 / df = 2.80; IFI = .94; CFI = .94; RMSEA = .093). The difference between the two models was significant (Δχ2 = 5.05, df = 1, p < .05). Factor structure of teamwork process. We also examined the factor structure of teamwork process. As mentioned in the Introduction, past research has suggested that the six dimensions of teamwork process can be clustered into two categories: task and interpersonal (Dayan & Benedetto, 2008). As already discussed, the task dimension includes three facets: communication, coordination, and balance of contribution. The interpersonal dimension includes the other three facets: effort, cohesion, and mutual support. At the individual level, the hierarchical model (χ2 = 2,788.47, df = 626, χ2 / df = 4.454; IFI = .97; CFI = .97; RMSEA = .074) was superior to the one-factor model (χ2 = 3,231.931, df = 629, χ2 / df = 5.14; IFI = .97; CFI = .97; RMSEA = .081). The difference between the two models was significant (Δχ2 = 443.461, df = 3, p < .05). At the team level, the hierarchical model (χ2 = 1,750.31, df = 626, χ2 / df = 2.80; IFI = .96; CFI = .96; RMSEA = .082) was superior to the one-factor model (χ2 = 1,839.85, df = 629, χ2 / df = 2.93; IFI = .96; CFI = .96; RMSEA = .087). The difference between the two models was significant (Δχ2 = 89.54, df = 3, p < .05). Factor structure of OCB. As articulated earlier, LePine et al. (2002) treated the facets of OCB as part of a broader dimension. Hoffman, Blair, Meriac, and Woehr (2007) went so far as to state that doing so “presents a more accurate estimate of the relationship between OCB and other variables” (p. 562). With this prior work in mind, we subjected our measure of OCB to a CFA. At the individual level, the hierarchical model of OCB (χ2 = 33.82, df = 13, χ2 / df = 2.60; IFI = .99; CFI = .99; RMSEA = .051) was superior to the one-factor structure (χ2 = 191.805, df = 14, χ2 / df = 13.70; IFI = .92; CFI = .92; RMSEA = .143). The difference between the two models was significant (Δχ2 = 157.985, df = 1, p < .05). At the team level, the hierarchical model of OCB (χ2 = 20.16, df = 13, χ2 / df = 1.55; IFI = .99; CFI = .99; RMSEA = .057) was superior to the one-factor structure (χ2 = 75.71, df = 14, χ2 / df = 5.41; IFI = .93; CFI = .93; RMSEA = .161). The difference between the two models was significant (Δχ2 = 55.55, df = 1, p < .05). Data Aggregation As individual team members provided the scale responses, there must be sufficient within-team agreement to warrant aggregation to the team level. To examine this possibility, we computed the within-group interrater agreement index (rwg(j)), the intraclass coefficient ICC(1), and the group mean reliability ICC(2). The mean rwg(j) values were as follows: procedural justice: .91, interpersonal justice: .90, communication: .93, coordination: .86, balance of Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 582 Group & Organization Management 36(5) contribution: .84, effort: .79, mutual support: .94, cohesion: .93, helping behavior: .88, and loyalty: .76. The mean rwg(j) values for all the variables were greater than .70, a conventional cutoff for acceptable within-unit agreement. The ICC(1) and ICC(2) for the measures were as follows: procedural justice: .12 and .34, interpersonal justice: .20 and .47, communication: .39 and .70, coordination: .25 and .55, balance of contribution: .36 and .67, effort: .39 and .70, mutual support: .33 and .64, cohesion: .39 and .70, helping behavior: .24 and .54, and loyal behavior: .20 and .48. All of the ICC(1) values reported here are greater than .12, traditionally considered as the cutoff value. Our ICC(2) values are relatively low, partly due to the small team size. The low ICC(2) values in our study mean lower reliability, which render our findings conservative (it is worth noting that all the F test results based on one-way analysis of variance [ANOVA] were significant, further buttressing our confidence in the level of agreement within the teams, see Bliese, 1998). Based on these results, we aggregated all the measures to the team level. Incremental Variance of Peer Justice Before we tested our hypotheses, we ran a series of hierarchical linear modeling (HLM) analysis to examine the incremental values of peer justice at the unit level over its counterparts at the individual level using the SAS PROC MIXED program (Singer, 1998). To conduct this analysis, we treated task and interpersonal teamwork processes as individual-level outcome variables of peer justice. Following Hofmann and Gavin (1998), we grand mean centered the Level 1 predictor variables (individual-level peer procedural and interpersonal justice). We ran an unconditional means (null model) to confirm that there was significant cross-group variance with respect to task and interpersonal team processes. Results of the null model indicated that 39% of the variance in task teamwork process and 46% of the variance in interpersonal teamwork process resided between teams. We then ran a random coefficients model with peer procedural justice at the individual level as a Level 1 predictor of task teamwork process. Peer procedural justice at the individual level was a significant predictor of task teamwork process (γ = .44, p < .05), explaining 22% of the variance. We then added peer procedural justice climate as a Level 2 predictor which was also a significant predictor of task teamwork process (γ = .49, p < .05), even when we controlled for its individual-level counterpart. Peer procedural justice climate as a Level 2 predictor explained 20% of the unique variance. Similarly, we ran a random coefficients model with peer procedural justice at the individual level as a Level 1 predictor of interpersonal teamwork process. Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 Cropanzano et al. 583 Peer procedural justice at the individual level was a significant predictor of interpersonal teamwork process (γ = .42, p < .05), explaining 21% of the variance. We then added peer procedural justice climate as a Level 2 predictor which was also a significant predictor of interpersonal teamwork process (γ = .60, p < .05), even when we controlled for its individual-level counterpart. Peer procedural justice climate as a Level 2 predictor explained 23% of the unique variance. We then ran a random coefficients model with peer interpersonal justice at the individual level as a Level 1 predictor of interpersonal teamwork process. Peer interpersonal justice at the individual level was a significant predictor of interpersonal teamwork process (γ = .36, p < .05), explaining 15% of the variance. We then added peer interpersonal justice climate as a Level 2 predictor which was also a significant predictor of interpersonal teamwork process (γ = .48, p < .05), even when we controlled for its individual-level counterpart. Peer interpersonal justice climate as a Level 2 predictor explained 20% of the unique variance. Together, results of these analyses suggest that peer justice at the unit level accounted for unique variance on outcome variables above and beyond their individual-level counterparts. Testing the Full Model Given that we had a larger number of parameters relative to the teams, we followed the advice of Humphrey, Ellis, Conlon, and Tinsley (2004) and Shrout and Bolger (2002), modeling the data using path analysis. Specifically, multiple-item scales were aggregated by calculating the mean and these averages were treated as observed indicators (Hoegl & Gemuenden, 2001). We followed past research to operationalize teamwork process as two higher order constructs (task and interpersonal) indicated by the six facets (Hoegl & Gemuenden, 2001). Results of these analyses are presented in Table 2. We first tested a completely mediated model that does not have any direct paths from the independent variables to the dependent variables (Model 1: χ2 = 85.357, df = 40, p < .05; IFI = .97; CFI = .97; RMSEA = .082). We then tested a model that we hypothesized. Specifically, we added the direct paths from peer procedural justice to performance and citizenship behavior and the direct path from peer interpersonal justice to citizenship behavior (Model 2: χ2 = 78.827, df = 37, p < .05; IFI = .98; CFI = .98; RMSEA = .082). There was no significant difference between these two models (Δχ2 = 6.53, df = 3, ns). We then compared our predicted model with a number of theoretically viable alternative models. First, we added a direct path from peer interpersonal justice to task teamwork process and a direct path from peer interpersonal Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 584 Group & Organization Management 36(5) Table 2. Model Fit Model Model 1: Fully mediated model Model 2: Model 1 plus three links: procedural justice to OCB, procedural justice to performance, and interpersonal justice to OCB Model 3: Model 2 plus two links: interpersonal justice to task teamwork, interpersonal justice to performance Model 4: Model 3 plus one link: task teamwork to OCB Model 5: Model 3 plus one link: interpersonal teamwork to performance χ2 df CFI IFI RMSEA 85.36 40 .97 .97 .082 78.83 37 .98 .98 .082 72.41 35 .98 .98 .080 70.36 34 .98 .98 .080 69.30 34 .98 .98 .078 Note: CFI = Comparative Fit Index; IFI = Incremental Fit Index; RMSEA = root mean square error of approximation; OCB = organizational citizenship behavior. justice to performance (Model 3: χ2 = 72.409, df = 35, p < .05; IFI = .98; CFI = .98; RMSEA = .080). Comparison of Model 2 and Model 3 indicated that the new model (with two paths added) provided a significantly better fit than the hypothesized model (Δχ2 = 6.42, df = 2, p < .05). Model 3 was also significantly better than Model 1, the completely mediated model (Δχ2 = 12.95, df = 5, p < .05). Building on this new model, we added one more path (a direct path from task teamwork process to OCB). This new model (Model 4: χ2 = 70.355, df = 34, p < .05; IFI = .98; CFI = .98; RMSEA = .080) was not significantly better than Model 3 (Δχ2 = 2.05, df = 1, ns). We also tried a fifth model that adds a direct path from interpersonal teamwork process to performance. This new model (Model 5: χ2 = 69.299, df = 34, p < .05; IFI = .98; CFI = .98; RMSEA = .078) was not significantly better than Model 3 (Δχ2 = 3.11, df = 1, ns). Based on these analyses, we use Model 3 to test our hypotheses (see Figure 2). Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 585 Cropanzano et al. Task Teamwork process .11* .45** Communicaon Peer Procedural Jusce Coordinaon Contribuon Task Performance Direct Effect: -.13 Indirect Effect: .09* Direct Effect: -.08 Indirect Effect: .43** .40** .24** Direct Effect: -.01 Indirect Effect: .05* Team Cizenship Behavior Direct Effect: -.08 Indirect Effect: .32** Peer Interpersonal Jusce .30** Cohesion Interpersonal Teamwork process Effort .94** Helping Loyalty Support Figure 2. Final Model Showing Both Indirect and Direct Paths Hypotheses Tests: Peer Procedural Justice Peer procedural justice was positively related to the interpersonal (β = .40, p < .05) dimension of teamwork process, supporting Hypothesis 1. Interpersonal teamwork process was associated with OCB (β = .94, p < .05). The indirect path, which linked peer procedural justice to OCB (through interpersonal teamwork process), was significant (β = .43, p < .05). This supported Hypothesis 2. Peer procedure justice also predicted the task teamwork process (β = .45, p < .05), supporting Hypothesis 3. Task teamwork process was associated with team performance (β = .11, p < .05). The indirect path, which linked peer procedural justice to performance (through task teamwork process), was also significant (β = .09, p < .05). This supported Hypothesis 4. Hypotheses Tests: Peer Interpersonal Justice We found that peer interpersonal justice was significantly related to the interpersonal dimension of teamwork process (β = .30, p < .05), supporting Hypothesis 5. The indirect path linking peer interpersonal justice to OCB (through interpersonal teamwork process) was also significant (β = .32, p < .05), supporting Hypothesis 6. Although we did not make the prediction, we also found that peer interpersonal justice was significantly related to the task dimension of teamwork process (β = .24, p < .05). In addition, the indirect Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 586 Group & Organization Management 36(5) path linking peer interpersonal justice to performance (through task teamwork process) was also significant (β = .05, p < .05). This is a post hoc finding, thereby in need of replication. Auxiliary Analysis At the suggestion of an anonymous reviewer, we investigated the effects of peer justice climate strength in an exploratory manner. Climate strength refers to the level of agreement among unit members regarding the meaning they ascribe to the social environment. A high level of climate strength indicates a greater level of consensus, whereas a low level of climate strength indicates significant variability among unit members (Kuenzi & Schminke, 2009). Justice scholars have argued that variation in justice perceptions among unit members may hinder interpersonal processes and cooperation among unit members (Colquitt et al., 2002). Therefore, we used a series of models to examine the effects of peer justice strength on performance and citizenship behavior through task and interpersonal teamwork processes. Following Colquitt and colleagues (2002), we operationalized climate strength by computing the standard deviation of team members’ justice perceptions in each team and dividing that by the team’s mean level. We then standardized the value and reversed the sign so that high values represented high levels of strength. We reran all the five models that were identical to the ones described earlier, with the exception of using climate strength to replace climate level. The most parsimonious model among the five was Model 2 (χ2 = 84.286, df = 37, p < .05; IFI = .97; CFI = .97; RMSEA = .086). Based on this model, peer procedural justice strength was positively related to the interpersonal dimension (β = .17, p < .05) and the task dimension of teamwork process (β = .20, p < .05). In addition, peer procedural justice climate strength was indirectly related to team performance (β = .03, p < .05, through task teamwork process) and OCB (β = .16, p < .05, through interpersonal teamwork process). Peer interpersonal justice strength, however, was not related significantly to the interpersonal dimension of teamwork process. In addition, we also examined whether peer justice level and strength interact to influence outcome variables. No significant effects were observed in the present study. Discussion The present study examines a teamwork process model of peer justice. We found that individuals were able to form reliable judgments of peer procedural Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 Cropanzano et al. 587 and interpersonal justice. These two types of fairness predicted both task and interpersonal teamwork processes. In addition, we found that peer justice conceptualized at the unit level explained unique variance on outcome variables, even when we controlled for their individual-level counterparts. These findings are consistent with research showing that contextual variables predict outcome variables beyond the effects of individual cognitions (Kuenzi & Schminke, 2009). By way of their impact on teamwork process, both types of peer justice were associated with effective work behaviors. Specifically, task teamwork process engendered higher team performance, measured as grades, whereas interpersonal teamwork process boosted team citizenship behaviors, measured as helping and loyalty. These are extremely promising findings. Our results indicated that individuals form unit-level judgments of group fairness, these judgments then affect team processes, and then the processes result in more or less effective work outcomes. In the sections below, we discuss the implications of our findings for future research and practices. Future Research Directions Structure of peer justice. Although distributive peer justice is not the central focus of the present study, we should note that there are settings where this concept may be quite relevant. For example, faculty committees often vote on the tenure of their colleagues, which can create perceptions of distributive injustice (e.g., Ambrose & Cropanzano, 2003). In settings where team members control one another’s outcomes, it may well be possible to document the existence of peer distributive justice. This would seem to be an important avenue for future inquiry. A second structural issue revolves around the construct of informational justice. Along with interpersonal justice, informational justice is a second potential subdimension of interactional fairness (Colquitt, 2001). As we discussed in the Introduction, the Donovan et al. (1998) scale emphasizes the interpersonal subdimension. As a result, it attends somewhat less to the informational subdimension. If the two-factor model of interactional justice is supported by additional validation research, then informational justice will likely become an important consideration for peer justice researchers. A climate with a high level of informational justice may reinforce team members’ motivation to engage in positive interpersonal teamwork processes such as identifying themselves with the team, providing support to each other when such needs arise, and exerting additional effort to help the team accomplish its goals. This possibility should be evaluated in future research. Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 588 Group & Organization Management 36(5) Antecedents of peer justice: Leadership and “trickle-down” models. Future research should also examine the antecedents of peer justice, with the hope that doing so would bring about fairer work units. Although our present study did not explore antecedents, the literature on ethical leadership offers some interesting parallels. Studies have found that the benefits of ethical leadership “trickle down” to the workforce. For example, Detert, Treviño, Burris, and Andiappan (2007) observed that ethical restaurant managers encountered less food loss at their establishments. In a parallel fashion, fair supervisors may engender fair subordinates. Hence, leaders could set the tone for the work team by modeling appropriate conducts. This argument is consistent with the structualist approach to the etiology of climates. In other words, leadership behavior serves as a structural element that provides the contexts based on which unit members develop similar meanings and perceptions. We can use this same line of thinking to integrate the peer justice construct with the older concept of justice climate. As we have already observed, justice climate pertains to how an organizational authority treats the individuals he or she manages (Mossholder, Bennett, & Martin, 1998). Peer justice, however, pertains to how workers treat one another. Thus, an environment characterized by low justice climate—an unfair boss who mistreats the work team—can create an environment characterized by low peer justice—team members who mistreat their coworkers. If this is so, then justice climate and peer justice could be closely and causally related, as the former is an antecedent of the latter. Moreover, by extension, peer justice may mediate some of the effects of justice climate that have been observed by previous scholars (such as Colquitt, Zapata-Phelan, & Roberson, 2005). This is an exciting possibility because it has the potential to bring greater coherence to the justice literature. The choice of criterion measures. In this study, we selected criterion variables that should be practically important to work organizations—team performance and team citizenship behaviors. We chose these outcome measures so as to demonstrate the impact of peer justice on “bottom-line” performance. As important as fairness may be for employers, it could matter even more for employees. Available research, most of it done at the individual level of analysis, suggests that fair workplaces tend to engender more positive emotions and lower levels of stress (Cropanzano & Wright, 2011). Given the importance of peer influences on workplace adjustment (see. Chiaburu & Harrison, 2008; Riketta & Van Dick, 2005), it is likely that the impact of peer justice on workers’ well-being could be quite strong. Consequently, our model should be extended to predict employee-relevant criterion variables, such as wellness, stress, and mood. Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 Cropanzano et al. 589 Practical Implications Given the importance of workplace fairness (e.g., Cohen-Charash & Spector, 2001), Skarlicki and Latham (2005) have promoted training as a means of fostering workgroup fairness. Evidence supports this possibility. In two studies, Skarlicki and Latham (1996, 1997) found that formal training in fairness promoted more just supervisory behavior. This, in turn, increased OCB within workgroups. These studies have implications for our findings regarding peer justice. Training of supervisors promotes fairness, but given the importance of peer justice, it may also be worthwhile to train employees how to treat their coworkers fairly. Beyond training, organizations can also employ other methods to create a high level of peer justice in teams. One such method is the employment of 360-degree performance feedback (Atwater, Brett, & Charles, 2007). Its premise is that individuals may rate their own behavior favorably due to self-serving biases; as such, additional inputs from one’s peers are necessary to complement self-ratings. Team leaders should encourage team members to evaluate each other in terms of the extent to which they use fair procedures to make decisions in the team and treat each other in an interpersonally sensitive manner. Results of the feedback can be used as the basis for training and development. In addition, in line with the recommendation that a strong foundation in a team may facilitate team effectiveness, it is also important that team members craft a team charter, or the agreed-on way that they interact with each other, at the initial phase of team development (Mathieu & Rapp, 2009). The charter may serve as a norm that sets the tone and provides guidance for team members’ behavior. Strengths and Limitations The present research had a number of strengths. Our participants worked in actual teams, where their behavior was relevant to important outcomes. In addition, we were able to follow team members over the course of an entire semester. This allowed us to survey individuals in waves, aligning our data collection with the presumed causal order. Despite these advantages, it is worth considering at least three limitations that could point to future research needs. External validity. While this was a “real-world” sample from the perspective of the participants, it was also an academic one. Hence, it remains to be seen how well our findings will generalize to an organizational setting. Although we acknowledge the need for future research, we are also optimistic about our findings. The participants in our study worked in an existing team; their performance had important stakes. Hence, this setting is likely to have been Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 590 Group & Organization Management 36(5) psychologically engaging, an essential feature of the work setting. That said, generalizability is a question that can be best answered through replication. Self-report and mono-method bias. Another potential concern is that our predictors and outcomes were sometimes collected from the same individuals. To allay this concern, we followed advice from Podsakoff, MacKenzie, Lee, and Podsakoff (2003). Our criterion measure, team performance, was determined by the group’s project grade. We also separated our measures in time, which should serve to reduce biases due to memory effects. Given these features of our research design, it does not seem likely that mono-method bias could provide a complete explanation of our results, though it could have inflated some of the observed associations. Performance measure. The instructor’s grading of team projects may potentially be influenced by personal biases, prior impression of certain team members, and the interaction with them throughout the semester. For example, the instructor might have formed overall impressions of the team (such as the level of communication and mutual support) based on conversations with certain team members, which may influence the grading of team projects. This possibility, although quite improbable given the large class size and number of teams, cannot be independently verified as the instructor was the only person grading the projects. In future research employing a similar method as we did in this study, it is recommended that two raters independently rate the team project so that the reliability of the ratings can be measured. Conclusions Structural changes in the workplace point to the inadequacy of focusing on authority figures as the only source of justice. To address this limitation, we developed and examined a teamwork process model of peer justice. As a unitlevel construct, peer justice is well suited to understanding team dynamics. Furthermore, peer justice is a good predictor of effective work-unit behavior. Although considerably more research is necessary, the present findings underscore the importance of considering peers in the study of fairness perceptions. Acknowledgments The authors thank the editor, Dr. Yehuda Baruch, and the three anonymous reviewers for their many constructive comments. The authors also thank Wes Friske for his comments on earlier drafts of the manuscript. Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 Cropanzano et al. 591 Funding The author(s) received no financial support for the research, authorship, and/or publication of this article. References Ambrose, M. L., & Cropanzano, R. (2003). A longitudinal analysis of organizational fairness: An examination of reactions to tenure and promotion decisions. Journal of Applied Psychology, 88, 266-275. Ambrose, M. L., & Schminke, M. (2001). Are flexible organizations the death knell for the future of procedural justice? In R. Cropanzano (Ed.), Justice in the workplace: From theory to practice (Vol. 2, pp. 229-244). Mahwah, NJ: Erlbaum. Ambrose, M. L., & Schminke, M. (2007). Examining justice climate: Issues of fit, simplicity, and content. In F. Dansereau & F. J. Yammarino (Eds.), Research in multilevel issues (Vol. 6, pp. 397-412). New York, NY: Elsevier. Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: A review and recommended two-step approach. Psychological Bulletin, 103, 411-423. Aquino, K., Tripp, T. M., & Bies, R. J. (2006). Getting even or moving on? Power, procedural justice, and types of offense as predictors of revenge, forgiveness, reconciliation, and avoidance in organizations. Journal of Applied Psychology, 91, 653-668. Atwater, L. E., Brett, J. F., & Charles, A. C. (2007). Multisource feedback: Lessons learned and implications for practice. Human Resource Management, 46, 285-307. Banker, R. D., Field, J. M., Schroeder, R. G., & Sinha, K. K. (1996). Impact of work teams on manufacturing performance: A longitudinal study. Academy of Management Journal, 39, 867-890. Barker, J. R. (1993). Tightening the iron cage: Concertive control in self-management teams. Administrative Science Quarterly, 38, 408-437. Baruch, Y., & Holtom, B. C. (2008). Survey response rate levels and trends in organizational research. Human Relations, 61, 1139-1160. Bies, R. J. (2001). Interactional (in)justice: The sacred and the profane. In J. Greenberg & R. Cropanzano (Eds.), Advances in organizational justice (pp. 89-118). San Francisco, CA: Stanford University Press. Bishop, J. W., & Scott, K. D. (2000). An examination of organizational and team commitment in a self-directed team environment. Journal of Applied Psychology, 85, 439-450. Bliese, P. D. (1998). Group size, ICC values, and group-level correlations: A simulation. Organizational Research Methods, 1, 355-373. Bosch-Sijtsema, P. M., Fruchter, R., Vartiainen, M., & Ruohomaki, V. (2011). A framework to analyze knowledge work in distributed teams. Group & Organization Management, 36, 275-307. Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 592 Group & Organization Management 36(5) Cardona, P., Lawrence, B. S., & Bentler, P. M. (2004). The influence of social and work exchange relationships on organizational citizenship behavior. Group and Organization Management, 29, 219-247. Chiaburu, D. S., & Harrison, D. A. (2008). Do peers make the place? Conceptual synthesis and meta-analysis of coworker effects on perceptions, attitudes, OCBs, and performance. Journal of Applied Psychology, 93, 1082-1103. Cohen-Charash, Y., & Mueller, J. S. (2007). Does perceived unfairness exacerbate or mitigate interpersonal counterproductive work behaviors related to envy? Journal of Applied Psychology, 92, 666-680. Cohen-Charash, Y., & Spector, P. E. (2001). The role of justice in organizations: A metaanalysis. Organizational Behavior and Human Decision Processes, 86, 278-321. Colquitt, J. A. (2001). On the dimensionality of organizational justice: A construct validation of a measure. Journal of Applied Psychology, 86, 386-400. Colquitt, J. A., Noe, R. A., & Jackson, C. L. (2002). Justice in teams: Antecedents and consequences of procedural justice climate. Personnel Psychology, 55, 83-109. Colquitt, J. A., Zapata-Phelan, C. P., & Roberson, Q. M. (2005). Justice in teams: A review of fairness effects in collective contexts. In J. J. Martocchio (Ed.), Research in personnel and human resources management (Vol. 24, pp. 53-94). Oxford, UK: Elsevier. Cropanzano, R., Li, A., & James, K. (2007). Intraunit justice and interunit justice and the people who experience them. In F. Dansereau & F. J. Yammarino (Eds.), Research in multi-level issues (Vol. 6, pp. 415-437). Oxford, UK: Elsevier. Cropanzano, R., & Mitchell, M. S. (2005). Social exchange theory: An interdisciplinary review. Journal of Management, 31, 874-900. Cropanzano, R., Stein, J. H., & Nadisic, T. (2011). Social justice and the experience of human emotion. New York, NY: Taylor and Francis. Cropanzano, R., & Wright, T. A. (2011). The impact of organizational justice on occupational health. In J. C. Quick & L. E.Tetrick (Eds.), Handbook of occupational health psychology (pp. 205-219). Washington, DC: American Psychological Association. Dayan, M., & Benedetto, A. D. (2008). Procedural and interactional justice perceptions and teamwork quality. Journal of Business & Industrial Marketing, 23, 566-576. De Jong, B. A., & Elfring, T. (2010). How does trust affect the performance of ongoing teams? The mediating role of reflexivity, monitoring and effort. Academy of Management Journal, 53, 535-549. Detert, J. R., Treviño, L. K., Burris, E. R., & Andiappan, M. (2007). Managerial modes of influence and counterproductivity in organizations: A longitudinal business-unitlevel investigation. Journal of Applied Psychology, 92, 993-1005. Donovan, M. A., Drasgow, F., & Munson, L. J. (1998). The Perceptions of Fair Interpersonal Treatment Scale: Development and validation of a measure of interpersonal treatment in the workplace. Journal of Applied Psychology, 83, 683-692. Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 Cropanzano et al. 593 Easley, R. F., Devaraj, S., & Crant, M. (2003). Relating collaborative technology use to teamwork quality and performance: An empirical analysis. Journal of Management Information Systems, 19, 247-268. Edmondson, A. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44, 350-383. Ehrhart, M. G. (2004). Leadership and procedural justice climate as antecedents of unit-level organizational citizenship behavior. Personnel Psychology, 57, 61-94. Frazier, M. L., Johnson, P. D., Gavin, M., Gooty, J., & Snow, D. B. (2010). Organizational justice, trustworthiness, and trust: A multifoci examination. Group & Organization Management, 35, 39-76. Gladstein, D. L. (1984). Groups in context: A model of task group effectiveness. Administrative Science Quarterly, 29, 499-517. Hackman, J. R. (1992). Group influences on individuals in organizations. In M. D. Dunnette & L. M. Hough (Eds.), Handbook of industrial and organizational psychology (Vol. 3, pp. 199-267). Palo Alto, CA: Consulting Psychologists Press. Hoegl, M., Ernst, H., & Proserpio, L. (2007). How teamwork matters more as team member dispersion increases. Journal of Product Innovation Management, 24, 156-165. Hoegl, M., & Gemuenden, H. G. (2001). Teamwork quality and the success of innovative projects: A theoretical concept and empirical evidence. Organization Science, 12, 435-449. Hoegl, M., Parboteeah, K. P., & Gemuenden, H. G. (2003). When teamwork really matters: On the moderating influence of task innovativeness on the teamwork–performance relationship. Journal of Engineering and Technology Management, 20, 281-302. Hoffman, B. J., Blair, C. A., Meriac, J. P., & Woehr, D. J. (2007). Expanding the criterion domain? A quantitative review of the OCB literature. Journal of Applied Psychology, 92, 555-566. Hofmann, D. A., & Gavin, M. B. (1998). Centering decisions in hierarchical linear models: Implications for research in organizations. Journal of Management, 24, 623-641. Humphrey, S. E., Ellis, A. P. J., Conlon, D. E., & Tinsley, C. H. (2004). Understanding customer reactions to brokered ultimatums: Applying negotiation and justice theory. Journal of Applied Psychology, 89, 466-482. James, L. A., & James, L. R. (1989). Integrating work environment perceptions: Exploration into the measurement of meaning. Journal of Applied Psychology, 74, 739-751. Kozlowski, S. W. J., & Klein, K. J. (2000). A multilevel approach to theory and research in organizations: Contextual, temporal, and emergent processes. In K. J. Klein & S. W. J. Kozlowski (Eds.), Multilevel theory, research, and methods in organizations: Foundations, extensions, and new directions (pp. 3-90). San Francisco, CA: Jossey-Bass. Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 594 Group & Organization Management 36(5) Kuenzi, M., & Schminke, M. (2009). Assembling fragments into a lens: A review, critique, and proposed research agenda for the organizational work climate literature. Journal of Management, 35, 634-717. Lavelle, J. J., Brockner, J., Konovsky, M. A., Price, K. H., Henley, A. B., Taneja, A., & Vinekar, V. (2009). Commitment, procedural fairness, and organizational citizenship behavior: A multifoci approach. Journal of Organizational Behavior, 30, 337-357. Lavelle, J. J., Rupp, D. E., & Brockner, J. (2007). Taking a multifoci approach to the study of justice, commitment, and organizational citizenship behavior: The target similarity model. Journal of Management, 33, 841-866. LePine, J. A., Erez, A., & Johnson, D. E. (2002). The nature and dimensionality of organizational citizenship behavior: A critical review and meta-analysis. Journal of Applied Psychology, 87, 52-65. Leventhal, G. S. (1976). The distribution of rewards and resources in groups and organizations. In L. Berkowitz & W. Walster (Eds.), Advances in experimental social psychology (Vol. 9, pp. 91-131). New York, NY: Academic Press. Li, A., & Cropanzano, R. (2009). Fairness at the group level: Interunit and intraunit justice climate. Journal of Management, 35, 564-599. Li, A., Cropanzano, R., & Benson, L. (2007). Intraunit justice climate: Explication and validation of a new construct. Paper presented at the annual meeting of the Academy of Management, Philadelphia, PA. Liao, H., & Rupp, D. E. (2005). The impact of justice climate and justice orientation on work outcomes: A cross-level multifoci framework. Journal of Applied Psychology, 90, 242-256. Marks, M. A., Mathieu, J. E., & Zaccaro, S. J. (2001). A temporally based framework and taxonomy of team processes. Academy of Management Review, 26, 355-376. Masterson, S. S., Lewis, K., Goldman, B. M., & Taylor, M. S. (2000). Integrating justice and social exchange: The differing effects of fair procedures and treatment on work relationships. Academy of Management Journal, 43, 738-748. Mathieu, J. E., & Rapp, T. L. (2009). Laying the foundation for successful team performance trajectories: The roles of team charters and performance strategies. Journal of Applied Psychology, 94, 90-103. 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. Mossholder, K. W., Bennett, N., & Martin, C. L. (1998). A multilevel analysis of procedural justice context. Journal of Organizational Behavior, 19, 131-141. Naumann, S. E., & Bennett, N. (2000). A case for procedural justice climate: Development and test of a multilevel model. Academy of Management Journal, 43, 881-889. Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 Cropanzano et al. 595 Podsakoff, P. M., MacKenzie, S. B., Lee, J.-Y., & Podsakoff, N. P. (2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88, 879-903. Price, K. H., Harrison, D. A., & Gavin, J. H. (2006). Withholding inputs in team contexts: Member composition, interaction processes, evaluation structure, and social loafing. Journal of Applied Psychology, 91, 1375-1384. Riketta, M., & Van Dick, R. (2005). Foci of attachment in organizations: A metaanalytic comparison of the strength and correlates of workgroup versus organizational identification and commitment. Journal of Vocational Behavior, 67, 490-510. Roberson, Q. M. (2006). Justice in teams: The activation and role of sensemaking in the emergence of justice climates. Organizational Behavior and Human Decision Processes, 100, 177-192. Roch, S. G., & Shanock, L. R. (2006). Organizational justice in an exchange framework: Clarifying organizational justice distinctions. Journal of Management, 32, 299-322. Rupp, D. E., & Cropanzano, R. (2002). The mediating effects of social exchange relationships in predicting workplace outcomes from multifoci organizational justice. Organizational Behavior and Human Decision Processes, 89, 925-946. Rupp, D. E., & Paddock, E. L. (2010). From justice events to justice climate: A multilevel temporal model of information aggregation and judgment. In B. Mannix, M. Neal, & E. Mullen (Eds.), Research on managing groups and teams: Fairness and groups (pp. 245-273). Bingley, UK: Emerald. Salancik, G. R., & Pfeffer, J. (1978). A social information processing approach to job attitudes and task design. Administrative Science Quarterly, 23, 224-253. Schneider, B. (1987). The people make the place. Personnel Psychology, 40, 437-454. Schneider, B., Ehrhart, M. G., & Macey, W. H. (2010). Perspectives on organizational climate and culture. In S. Zedeck (Eds.), APA handbook of industrial and organizational psychology (Vol. 1, pp. 373-414). Washington, DC: American Psychological Association. Schneider, B., & Reichers, A. E. (1983). On the etiology of climates. Personnel Psychology, 36, 19-39. Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and nonexperimental studies: New procedures and recommendations. Psychological Methods, 7, 422-445. Singer, J. D. (1998). Using SAS PROC MIXED to fit multilevel models, hierarchical models, and individual growth models. Journal of Educational & Behavioral Statistics, 24, 323-355. Skarlicki, D. P., & Latham, G. P. (1996). Increasing citizenship behavior within a labor union: A test of organizational justice theory. Journal of Applied Psychology, 81, 161-169. Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 596 Group & Organization Management 36(5) Skarlicki, D. P., & Latham, G. P. (1997). Leadership training in organizational justice to increase citizenship behavior within a labor union: A replication. Personnel Psychology, 50, 617-633. Skarlicki, D. P., & Latham, G. P. (2005). How can training be used to foster organizational justice? In J. Greenberg & J. A. Colquitt (Eds.), The handbook of organizational justice (pp. 3-56). Mahwah, NJ: Erlbaum. Tangirala, S., & Ramanujam, R. (2008). Employee silence on critical work issues: The cross level effects of justice climate. Personnel Psychology, 61, 37-68. Tekleab, A. G., Takeuchi, R., & Taylor, M. S. (2007). Extending the chain of relationships among organizational justice, social exchange, and employee reactions: The role of contract violation. Academy of Management Journal, 48, 146-157. Tyler, T. R., & Blader, S. L. (2000). Cooperation in groups: Procedural justice, social identity, and behavioral engagement. Philadelphia, PA: Psychology Press. Van den bos, K. (2001). Uncertainty management: The influence of uncertainty salience on reactions to perceived procedural fairness. Journal of Personality and Social Psychology, 80, 931-941. Van der Vegt, G., Van de Vliert, E., & Oosterhof, A. (2003). Dissimilarity and organizational citizenship behavior: The role of intrateam interdependence and team identification. Academy of Management Journal, 46, 715-727. Yang, J., Mossholder, K. W., & Peng, T. K. (2007). Procedural justice climate and group power distance: An examination of cross-level interaction effects. Journal of Applied Psychology, 92, 681-692. Bios Russell Cropanzano is the Brien Lesk professor of organizational behavior in the University of Arizona’s Eller College of Management, Tucson. His research focuses on perceptions of organizational justice as well as on workplace emotion. He has authored more than 100 scholarly articles and chapters and two books. Andrew Li is the Williams professor of management at the West Texas A&M University, Canyon. He received his PhD from the University of Arizona. His research focuses on organizational justice, work–life balance, team/group, and selection/ assessment. Lehman Benson III is the McCoy/Rogers fellow of management at the University of Arizona. He received his PhD in cognitive psychology from Lund University (in Sweden). His postgraduate education was completed at the University of California at Berkeley, where he was awarded a presidential postdoctoral fellowship. Downloaded from gom.sagepub.com at LIBERTY UNIV LIBRARY on August 17, 2012 301 The British Psychological Journal of Occupational and Organizational Psychology (2010), 83, 301-324 n ^ H V ^ ç • ©2010 The British Psychological Society wlß^^ Society www.bpsjournals.co.uk Proactively performing teams: The role of work design, transformational leadership, and team composition Helen M. Williams'* Sharon K. Parker^ and Nick Turner^ 'Swansea University, UK ^The University of Western Australia, Perth, Australia University of Manitoba, Winnipeg, Canada This study investigated the determinants of team proactive performance amongst 43 shift teams from a UK chemical processing plant. Using external ratings of team proactive performance, the study found that the most proactive teams were those v^ith higher levels of self-management, transformational team leaders, and a higherthan-average level of proactive personality. The relationship between transformational leadership and team proactive performance was mediated by favourable interpersonal norms. In addition, lower diversity of proactive personality amongst team members had an indirect association with team proactive performance via its negative effect on favourable interpersonal norms. 'Teams are remarkably passive and accepting even when given work that is inappropriate for performance by a team, when the design of the team's task is flawed, or when contextual supports for teamwork are unavailable or inadequate ..." Oldham and Hackman (2010, p. 474) As suggested in the above quote, some work teams can be overly passive and adaptive when a more appropriate response might be to take charge and proactively change the situation. Yet, despite considerable research on individual-level proactivity (Bateman & Crant, 1993; Frese & Fay, 2001; Parker, Williams, & Turner, 2006), few studies have focused on what drives this behaviour at the team level. As Oldham and Hackman (2010, p. 474) concluded little is known about the roots of this passivity or what it would take to foster greater team proactivity about such matters. It would be good to know more'. In this study, we focus on the determinants of team proactive performance and the processes through which team proactive performance is achieved. We consider four types of ambient stimuli that could influence the proactivity of the team * Correspondence should be addressed to Dr Helen M. Williams, School of Business and Economics, Swansea University, Swansea, SA2 8PP, UK (e-mail: helen.m.wHliams@swan.ac.uk). DOI: 10.1348/096317910X502494 302 Helert M. Williams et al. (work design, leadership, norms, team composition), and the possible mediating role of favourable interpersonal norms. Our study suggests unique insights into factors that lead teams to take charge and initiate change in their environment or the team. Team proactive performance Most research and theory has considered the concept of proactivity at the individual level. Individual-level proactive behaviour refers to self-starting, future-focused action in which the individual alms to change the external situation, such as improving work methods, or to change some aspect of his/her self, such as improving one's performance by actively seeking feedback from a supervisor (Parker et al., 2006). Such behaviour is more active, change oriented, and future focused than either core task performance or adaptive performance, and, as such, is particularly important in uncertain contexts in which there is a lack of predictability in the inputs, processes, or outputs of work systems (Griffin, Neal, & Parker, 2007). When uncertainty is high, work-roles cannot be formalized precisely; they must emerge dynamically in response to changing conditions and demands. Proficient compliance with specifications is not sufficient; nor is it enough just to adapt and respond to these dynamic changes. Individuals need to take charge of situations, anticipate problems before they arise, and initiate change in the work system and work-roles. Researchers have shown the importance for individual and organizational outcomes of such individual-level proactivity (Fuller & Marier, 2009). Our focus in the current paper is on proactive teams rather than proactive individuals. We propose that team proactive performance is a team-level concept that has theoretical similarity with individual-level proactive performance and thus define proactive team performance as the extent to which a team engages in self-starting, future-focused action that aims to change the external situation or the team itself. Examples of proactive team behaviours include the team introducing new work methods, the team preventing problems rather than only reacting to them, or the team scanning the environment to identify potential opportunities. Such team proactivity is collective in emphasis: it is about the way the team behaves as a group, that is, as an interdependent and goal-directed combination of individuals (Morgeson & Hofmann, 1999). As such, proactive team performance is not the same as the sum of individual team member proactive performances, such as multiple individual team members acting proactively to contribute to individual or team goals (Strauss, Griffin, & Rafferty, 2009). Individuals within a team tnight behave proactively, such as by introducing new methods, but unless this effort is coordinated, the team Itself might not be proactive. Team proactive performance is an emergent property of teams that reflects and shapes team interactions. When a team carries out its tasks, team members interact w^ith each other to plan how they wUl meet their goals, monitor goal achievement, monitor external conditions, and coordinate interdependent activities (Marks, Mathieu, & Zaccaro, 2001). Through these interactions, team members develop shared and enduring ways of responding to challenges in the environment, which then become the team's behavioural patterns. For some teams, these interactions lead to the team trying to meet their goals in proactive ways, such as by platming ahead to prevent future problems. Other teams collectively develop routines for managing these processes of goal setting, monitoring, and performing in more passive ways. Team proactive performance, whilst distinct from individual-level proactive performance in structure (because it is composed of the interactions of team members rather than individual behaviour), is thus similar to individual-level proactive performance in function (Morgeson & Hofmann, 1999). Team proactive performance 303 There are relatively few studies of team proactive behaviour, but those that do exist suggest that proactive teams achieve positive outcomes. Hyatt and Reddy (1997) found that proactive behaviour of maintenance work groups was positively related to the response time of teams. Kirkman and Rosen (1999) found that team proactive behaviour was positively linked to team-level customer service and productivity. Similarly, Tesluk and Mathieu (1999) found that road crews who used highly self-starting strategies to manage performance barriers (e.g., taking advantage of low-workload times to improve methods) were most effective. Finally, Druskat and Kayes (2000) found that team proactivity in problem solving (defined as anticipating and heading off problems through proactive investigation, assessment, and action) predicted team learning and team performance in short-term student project teams. Determinants of team proactive performance Understanding the determinants of team proactive performance is important because one cannot simply assume homology across levels. As Chen, Bliese, and Mathieu (2005) argued: 'if researchers find that relationships are homogolous across levels of analysis, it adds to the parsimony and breadth of theories. In contrast, should relationships not prove to be homogolous across levels, it signals a boundary condition and a need to refine theories and to better understand how the processes operate at each distinct level' (p. 376). Moreover, team proactive performance is distinct from other team-level performance concepts that have had more attention, such as team adaptability. Kirkman and Rosen (1999) identified team empowerment (i.e., the team's collective feelings of meaning and control) as a determinant of team proactive performance, and empowerment in turn was predicted by external team leader behaviours, production/service responsibilities, team-based HR policies, and social structure. Likewise, Tesluk and Mathieu (1999) found that proactive crew management strategies were predicted by self-management, leadership, and teamwork processes. However, the mediating links between the variables of work design, leadership, and team processes were not considered in these studies, and team composition was not examined as a determinant. Our study seeks to expand understanding of the determinants of team proactive performance by considering a broader range of team-level variables than considered thus far (see Figure 1 for model). To identify predictors, we draw on Chen and Kanfer's (2006) categorization of factors that influence team motivation and behaviour. These scholars proposed a reorganization of person and situation factors according to their stimulus characteristics, not their impact. They identified ambient and discretionary inputs to teams. Ambient stimuli refer to team-oriented stimuli that pervade the team as a whole, such as socio-technical aspects of work like work design, whereas discretionary stimuli are those that are directed at or presented to specific team members, such as personalized feedback to individual team members or rewards at an individual level. In the current study, we focus on ambient stimuli because these have stronger and more direct effects on team-level motivation and performance than discretionary stimuli (Chen & Kanfer, 2006). Of the four categories of ambient inputs identified by Chen and Kanfer (2006), we include three categories for which there are clear theoretical reasons to expect associations with proactive performance: work design (i.e., team self-management), leadership (i.e., the transformational leadership of the team leader), and norms (i.e., the extent to which there are favourable interpersonal norms within the team). We expected these stimuli to operate similarly to how they 304 Helen M. Williams et al. Mean level of proactive personality Self-management Transformational leadership H3 Favorable interpersonal norms Proactive personality diversity Team proactive performance H8 H7 Figure I. Hypothesized model. operate at the individual level of analysis. We did not include team feedback, the fourth category of ambient stimuli identified by Chen and Katifer, because we expected feedback to be more important for fostering core, proficient performance rather than proactive performance. In addition to work design, leadership, and norms, we included team composition as a determinant. Specifically, we considered both the mean level of individual proactive personality and the diversity of proactive personality within the team. Although past research has investigated how the proactive personality of individuals affects individuallevel proactivity (e.g., Parker et al., 2006), research has not investigated how team member personality characteristics combine to affect team-level proactivity. Team composition has been found to be a key factor in predicting team effectiveness (e.g., see Ilgen, Hollenbeck, Johnson, & Jundt, 2005; Williams & Allen, 2008, for reviews) and, as we elaborate shortly, there are compelling reasons to consider this attribute of teams in relation to team proactive performance. The inclusion of team composition means we test a hybrid theory of homology (Chen et al., 2005) that identifies certain homologous predictors across levels (e.g., leadership, self-management), but also some predictors that are only meaningful at the group level (team composition). Work design: Team self-mar)agemer)t In terms of work design, our focus is on team self-management. Self-managing teams are interdependent groups of individuals who assume collective responsibility for the dayto-day operations of the team (Goodman, Devadas, & Griffith-Hughson, 1988; Parker & Wall, 1998). Members of self-managing teams typically experience greater variety, feedback, task significance, and task identity as a result of the self-managing design, but the most important feature is the greater collective autonomy that individuals have over their activities. For example, self-managing production teams typically allocate tasks amongst themselves, decide on the timing of their methods, and take responsibility for aspects such as quality and machine maintenance. In essence, team self-management is a form of shared leadership focused around decision-making, with shared leadership Team proactive performance 305 defined as 'an emergent team property that results from the distribution of leadership influence across multiple team members' (Carson, Tesiuk, & Marrone, 2007, p. 1218). At the individual level of analysis, job autonomy has been identified as one of the most consistent determinants of proactive behaviours, such as proactive problem solving and idea implementation (Parker et al., 2006), personal initiative (Frese, Kring, Soose, & Zempel, 1996), voice (LePine & Van Dyne, 1998), and suggesting improvements (Axtell et al., 2000). We suggest the same positive effects of autonomy operate at the team level. Self-managing teams allow team members the control and opportunity to manage their demands (variances) more actively. In essence, team autonomy 'allows' the team to be more proactive. Self-management also increases the team's motivation to be proactive. Thus self-managing teams, through greater task control and engagement in challenging tasks, develop a shared sense of collective efficacy that they can shape their environment in a proactive way. Prior research shows that self-management enhances collective efficacy which in turn drives performance (Chen & Katifer, 2006). Team self-management could also etihance team proactive performance through a cross-level process in which team self-management, because of the greater autonomy it affords individuals as well as teams, results in greater individual proactive motivation which in turn drives individual proactive behaviour. Although we do not test this cross-level process, it provides a further explanation as to why team selfmanagement might affect team proactive performance. In sum, there is good reason to expect that team self-management will be associated with team proactive performance, as indeed shown by two team-level studies (Kirkman, Rosen, Tesiuk, & Gibson, 2004; Tesiuk & Mathieu, 1999). We aimed to replicate these findings. Our hypothesis is: Hypothesis I: Team self-management will be positively related to team proactive performance. Group norms: Favourable interpersonal norms Norms are informal 'rules' present in a group that regulate the behaviour of members belonging to the group and establish a common code of conduct (Feldman, 1984). Chen and Klimoski (2003) argued that interdependent contexts amplify the importance of group norms and climate. Here we focus on interpersonal norms, or the code of conduct by which team members typically treat each other. Favourable interpersonal norms are especially important for proactivity because engaging in proactive behaviour can be interpersonally 'risky' (Parker etal., 2006). At the individual level, it is theorized that individuals weigh up the likely benefits and risks before deciding whether to take charge at work (Morrison & Phelps, 1999). Perceived risks are enhanced when individuals fear they might be put down or not respected by colleagues. According to Dutton, Ashford, Lawrence, and Miner-Rubino (2002), when deciding whether to engage in discretionary behaviour, individuals engage in contextual sensemaking assessing 'whether or not the context is favorable for taking some type of action' (p. 355). In support of this, relationship quality between the individual and the people with whom they were to sell was important in predicting issue selling (Ashford, Rothbard, Piderit, & Dutton, 1998), a proactive behaviour in which individuals try to itifluence the organizational agenda by 'selling' issues to leaders. Further, Parker et al. (2006) found team members' trust in co-workers to be positively related to engaging in individual-level proactive work behaviour. We propose the importance of a favourable interpersonal context at the team level. In deciding whether to suggest ideas or start discussions with other team members 306 Helen M. Williams et al. regarding how an anticipated problem can be overcome, team members will assess the way they work together and decide whether the within-team environment is supportive (or favourable) for taking such action. If the norms include team members' supporting and respecting each other, they will more likely take the risk of being proactive. Moreover, when there are favourable interpersonal norms, one team member putting forward suggestions will start positive discussions amongst team members, thus encouraging more ideas to be put forward by the team. Some support for this idea comes from Tesluk and Mathieu (1999), who found that team work processes of coordination, potency, and familiarity were positively related to problem-management actions and strategies. Likewise, Zárraga and Bonache (2005) found a 'high care' atmosphere (incorporating issues such as respect within the team) facilitated the transfer and creation of knowledge in self-managed teams. Similarly, team psychological safety has been found to be important to team learning behaviour (Edmondson, 1999), and team norms supporting itinovation predict team innovation (Anderson & West, 1998). We therefore suggest that, for the team to be proactive, it requires members to appraise the interpersonal norms as favourable so that they are willing to speak out and challenge the status quo, and are prepared to put forward suggestions and ideas for improvement. We hypothesize that: Hypothesis 2: Favourable interpersonal norms within the team will be positively related to team proactive performance. Leadership: Transformational team leader According to Chen and Kanfer (2006), leadership arguably represents the most important of all contextual factors which might affect individual and team motivation' (p. 40). A key type of ambient leadership, directed towards the team as a whole, is transformational leadership. Transformational leaders motivate teams by transforming the values and priorities of team members and inspiring them to perform beyond expectations (Bass, 1985). Bass's (1985) four components of transformational leadership (idealized influence, inspirational motivation, intellectual stimulation, and individualized consideration) collectively inspire followers to achieve more than was thought possible. Importantly for proactivity, transformational leaders encourage followers to question assumptions and think about ne^v ways of doing tasks. Consistent w^ith these ideas, transformational leadership has been show^n to lead to positive individual and organizational outcomes (Bass & Riggio, 2006; Judge & Bono, 2000). At the team level, research from military units in America (Bass, Avolio, Jung, & Berson, 2003) and Singapore (Um & Ployhart, 2004) has found that unit commanders' transformational leadership predicts unit performance in operational training exercises. Like'wise, Keller (1992) found that transformational leadership in research and development teams predicted superior technical quality of products produced by these teams. In the current study, ^ve focus on team leaders who are 'hands-on' within the team. We suggest that one reason transformational leadership affects team proactivity is because transformational leaders encourage team self-management. Transformational leaders support individual development (Avolio & Gibbson, 1988) and inspire ind...
Purchase answer to see full attachment