Integrating leader-member exchange (LMX) research with role engagement theory (Kahn, 1990) and role system theory (Katz & Kahn, 1978), we propose a multilevel, dual process model to understand the mechanisms through which LMX quality at the individual level and LMX differentiation at the team level simultaneously affect individual and team performance. With regard to LMX differentiation, we introduce a new configural approach focusing on the pattern of LMX differentiation to complement the traditional approach focusing on the degree of LMX differentiation. Results based on multiphase, multisource data from 375 employees of 82 teams revealed that, at the individual level, LMX quality positively contributed to customer-rated employee performance through enhancing employee role engagement. At the team level, LMX differentiation exerted negative influence on teams' financial performance through disrupting team coordination. In particular, teams with the bimodal form of LMX configuration (i.e., teams that split into 2 LMX-based subgroups with comparable size) suffered most in team performance because they experienced greatest difficulty in coordinating members' activities. Furthermore, LMX differentiation strengthened the relationship between LMX quality and role engagement, and team coordination strengthened the relationship between role engagement and employee performance. Theoretical and practical implications of the findings are discussed.

We propose that it is important to take the content of team voice into account when examining its impact on team processes and outcomes. Drawing on regulatory focus theory (Higgins, 1997), we argue that promotive team voice and prohibitive team voice help teams achieve distinct collective outcomes-that is, team productivity performance gains and team safety performance gains, respectively. Further, we identify mechanisms through which promotive and prohibitive team voices uniquely influence team outcomes as well as boundary conditions for such influences. In data collected from 88 production teams, we found that promotive team voice had a positive association with team productivity performance gains. By contrast, prohibitive team voice had a positive association with team safety performance gains. The relationship between promotive team voice and team productivity performance gains was mediated by team innovation, and the relationship between prohibitive team voice and team safety performance gains was mediated by team monitoring. In addition, the indirect effect of prohibitive team voice on team safety performance gains via team monitoring was stronger when prior team safety performance was lower. We discuss the theoretical and practical implications of these findings. (PsycINFO Database Record

Summary To understand how trust in supervisor translates into individual job performance, we hypothesize that trust in supervisor facilitates positive psychological conditions of meaningfulness, safety, and availability, which in turn predict individual job performance. We assert that each of the three mediating paths in our model represents a distinct mechanism by which trust in supervisor contributes to individual job performance. We test our hypotheses with 206 supervisor–subordinate dyads and find that trust in supervisor contributes to job performance through psychological availability and psychological safety but not psychological meaningfulness. By examining three different psychological mechanisms within the same frame, we provide a test that compares and contrasts the uniqueness of the three pathways. Our findings suggest practical ways for managers to build trust with subordinates and guidance for the design of productive work conditions. Copyright © 2012 John Wiley & Sons, Ltd.

Drawing on social hierarchy theory, we develop a contingency model of leader–member exchange (LMX) differentiation in which LMX differentiation is positively and negatively related to group cooperation and group social undermining, respectively, when it is based on the group members’ performance, but the relations are reversed (i.e., negative and positive, respectively) when it stems from a leader's personal liking of the members. In addition, we propose that the moderating effects of the performance and personal liking bases of LMX differentiation are magnified by the levels of reward interdependence. Specifically, under a high (vs. low) level of reward interdependence, LMX differentiation based on performance more strongly relates to high group cooperation and low group social undermining, whereas LMX differentiation with a personal liking basis is more likely to decrease group cooperation and increase group social undermining. Group cooperation and social undermining are then hypothesized to convey the three‐way interactive effects of LMX differentiation, its two bases, and reward interdependence on subsequent group performance. Analyses of data from 328 sales groups of a large retailer support the core part of our contingency model of LMX differentiation.

No abstract