This paper reports on a study of structural antecedents to team learning. In a study of self-managed pharmaceutical research and development teams, we first find that more team-level structure is associated with more internal learning as well as more external learning. We then establish that more organizational-level structure is negatively associated with both internal and external learning. We find that psychological safety mediates the positive relationship between team structure and team learning, and that task autonomy constraints mediate the negative relationship between organizational structure and team learning. Investigating the interaction effect between team and organizational structure, we find, unexpectedly, that organizational structure supports external team learning under conditions of less team structure. Specifically, when teams have less team structure, the relationship between organizational structure and external team learning is positive. This structure substitutability finding suggests that although more organizational structure, on average, hurts external team learning, there are situations in which it helps. An important implication of the study is that multiple levels of structure, and their interactions, should be taken into consideration when assessing structural effects on team learning.

Key words: learning; team learning; structure; group structure; group processes; multilevel research

History: Published online in Articles in Advance September 27, 2012.

Introduction

A growing literature on learning in organizational teams shows that teams perform better when they engage in team learning (Argote et al. 2001). Some scholars have even suggested that team learning lies at the heart of organizational change and renewal (e.g., Edmondson 1999, Senge 1990). The central role of team learning to both teams and organizations, combined with evidence of persistent and considerable variance in learning across teams, highlights a need for more research examining how and under what circumstances teams learn. Accordingly, as the team learning literature has matured, scholars are increasingly interested in identifying antecedents that account for variance in learning across teams.

In existing research on team effectiveness, structure has been identified as one of the most important influences on both processes and outcomes in teams (Hackman 1987, Campion et al. 1993, Cohen and Ledford 1994), highlighting it as a potentially central antecedent to team learning. Structure concerns specialization of tasks, hierarchical arrangements, as well as formalization of objectives and procedures (Bunderson and Boumgarden 2010, Montananari 1979, Weber 1947), thereby inherently affecting a team's task context.

The expected relationship between structure and team learning, however, is not obvious. On the one hand, the classic view of organizational theory argues that more organizational structure is bad for creativity and innovation (Miller and Friesen 1980, Siggelkow 2001) and may therefore hinder team learning. On the other hand, recent work focusing on the team level has shown that more structure can help team learning (Bunderson and Boumgarden 2010). What might account for these different observations, and what should we conclude about the relationship between structure and team learning? Clarification may be found in the fact that structure is a multilevel construct, with the classical studies focusing on the organizational level (henceforth, "organizational structure") and recent work focusing...