A SOCIAL NETWORK PERSPECTIVE ON WORK GROUP PERFORMANCE

As business environment is rapidly changing, most organizations use groups as their fundamental unit of organizational structure. The question of what factors affect work group performance has become one of the important topics in the organization studies. However, in spite of the relational nature of group phenomenon, previous research on group effectiveness has revealed a lack of network perspectives, while producing mixed empirical findings. We propose that interactions within and outside a group need to be the pivotal focus of group effectiveness research rather than being considered one of the constituting factors of the input-process-output model in the previous group research. Social network perspective focusing on social relationships that link the individual and the social structure can be a fresh theoretical and methodological tool for group effectiveness research. Thus, in this paper, we explore the effects of relational characteristics of internal and external group activities on group effectiveness using a social network perspective. Specifically, we generate propositions concerning three relationships within and outside a group: (1) the vertical relationships between the leader and the members within a group, (2) the horizontal relationships among the members within a group, (3) the group’s external relationships with people outside a group. Regarding intra-group vertical relationships, we suggest that the leader’s LMX ties with the informal leader, the leader’s LMX ties with diverse cliques within a group, and the leader’s accuracy of the group social network perception influence work group effectiveness. We propose that the more central the followers with whom the leader has high LMX in the group, the more the leader connects to diverse cliques through high LMX with the central person in each clique, and the more accurately the leader perceives the social networks in the group, the better performance the group achieves as a whole. For the horizontal relationships among members, we consider social distance (density), structural embeddedness (multiplexity), reciprocity (symmetry), conflict (negative ties and fragmentation), and centralized interaction (centralization). We suggest that the close social distance among group members will have an inverted U-shaped relationship with group performance. We also propose that the more embedded and the more reciprocal group members’ interactions are, the better performance the group achieves as a whole. On the other hand, we expect the worse group performance when the more cliques are connected through negative ties, and group members’ interactions are more centralized. Finally, we turn to the relationships with people outside a group. Linking connections to diverse groups and to central persons outside the group can result in better group performance. Furthermore, we suggest some directions for future research including the joint effect of network characteristics within and outside the group, the relationships between individuals’ attributes and network interactions, the possible moderating factors to affect the relationships between network characteristics and work group performance. We also discuss the possible measurement issues concerning network characteristics and work group performance for the future empirical test of our propositions. In conclusion, we propose that a social network perspective can shed lights on the unanswered questions and unresolved debates on group effectiveness by capturing the dynamics of group advantages inherently derived from the relational nature of the group. Particularly, we attempt to combine various determinants of group performance that have been studied in the separate research areas by focusing on the three relationships around the group. We believe that our arguments can be a complementary theoretical resource to existing group research.