Utilizing expertise in work teams: the role of transactive memory systems.

Abstract

Transactive memory system (TMS) theory has attracted a lot of attention in understanding expertise recognition and utilization in work teams in recent years. However, the literature has been mainly focusing on the direct effects of TMS on team performance, leaving the effects on other team outcomes as well as the contingent factors of these effects largely unknown (Lewis, 2003). In addition, according to a recent review by Kozlowski and Ilgen (2006), the literature, featuring an emerging field, needs development on clarifying the definition of the construct and identifying antecedents that contribute to the establishment of TMS. To address these limitations and to advance our understandings of TMS, in this dissertation, I first clarified and synthesized the divergent dimensions of TMS by proposing a two-higher-order-component framework, i.e. TMS structure and TMS manifestation. Based on a revised model, I argued that TMS structure is captured by TMS specialization, sharedness, and accuracy, and TMS manifestation captured by TMS credibility and coordination. Using this framework, I tried to answer three questions: (1) What relationships does TMS structure (specialization, sharedness, accuracy) have with team outcomes (performance, innovation, commitment) and are these relationships mediated by TMS manifestation (credibility and coordination)? (2) Do the relationships between TMS manifestation and team outcomes differ among teams with different task characteristics (i.e. task interdependence, task routineness, and alignment of task assignment)? And (3) Do functional diversity, status characteristics (i.e. average levels of task-related and non-task-related status characteristic cues), and interpersonal

connections (closeness and communication frequency) in teams predict TMS structure and TMS manifestation? Using a sample of 92 charter school boards and 90 school directors, I conducted hierarchical ordinary least square (OLS) regressions. Results showed that, as predicted, TMS credibility and coordination mediated the positive relationships between TMS specialization and board performance and innovation rated by both the charter school directors and the board chairs as well as board commitment. In addition, TMS credibility mediated the positive relationships between TMS sharedness and school director-rated board performance and innovation as well as board commitment. Also as predicted, task interdependence moderated the relationships between TMS credibility and director- and chair-rated performance, director-rated innovation, and board commitment such that the positive relationships were stronger when task interdependence was high than when it was low. Task interdependence moderated the relationships between TMS coordination and director-rated performance and innovation and board commitment in the same fashion. Task routineness moderated the relationships between TMS credibility and director-rated performance and innovation such that the relationships were positive when the tasks were highly routine and negative when the tasks were non-routine. Task routinesness also moderated the relationships between TMS credibility and board commitment in the same fashion as task interdependence does. The average levels of task-related status cues positively predicted TMS specialization, the average levels of non-task-related status cues positively predicted TMS sharedness, and both status characteristics predicted TMS accuracy. Additionally, the levels of interpersonal v closeness among board members were positively related to TMS accuracy as well as TMS credibility and coordination. Contributions, practical implications, limitations and future research are discussed.