Transactive memory systems (TMSs) research offers insights into how specialized knowledge in teams may be effectively coordinated and utilized for optimal task performance. TMS is the group system for encoding, storing, and retrieving expertise based on a shared awareness of where specialized knowledge is located in the team. TMS theory has traditionally been grounded in a group-information processing framework, whereby the focus is on the cognitive aspects involved in encoding, storing, and retrieving information. However, because transactive processes occur through the communication and interpersonal interactions among team members, there have been calls to understand the role of social identification processes in how team members interact with each other for knowledge sharing purposes (Haslam, 2001; Ren & Argote, 2011). This thesis contributes to the TMS literature by providing a social identity lens to understanding TMS building in both typical and multidisciplinary teams.
Chapter 1 provides a general introduction, and Chapter 2 reviews the TMS literature, highlighting the need to move beyond a sole reliance on a group-information processing framework for understanding TMS. Chapter 3 draws on the social identity approach (consisting of social identity theory and self-categorization theory) to consider the social identification processes shaping TMS in typical teams (i.e., teams consisting of members from similar backgrounds) and multidisciplinary teams (i.e., teams formed with members from distinct professional backgrounds). In Chapter 3, a series of hypotheses is presented on the process model for predicting TMS in typical teams. Specifically, the quantity of task-related communication (H1a), and the quality of task related communication (H1b) are hypothesized to positively predict TMS. It is predicted that these relationships are mediated through a collective sense of team identification (H2a and H2b). Furthermore, team identification is predicted to improve team performance (H3), and this relationship is mediated through TMS (H4).
Chapter 4 outlines a programmatic series of three time-ordered studies that aimed to test the hypotheses outlined in Chapter 3. In Studies 1a and 1b, data were collected across three time-points from student teams (N = 53 & N = 46, respectively) working on interdependent team projects over a 12-week period, while in Study 1c, data were collected across four time-points from student teams (N = 47) working on interdependent team projects over a 9-month period. Across all three studies, group-level structural equation modeling demonstrated support for the mediating role of team identification in the positive relationship between communication quality and TMS, although communication quantity was consistently found to be unrelated to team identification and TMS. In Study 1b, TMS fully mediated the positive relationship between team identification and self-ratings of team performance. In Study 1c, TMS also fully mediated the positive relationship between team identification and peer-ratings of team performance, although TMS only partially mediated the positive relationship between team identification and team project grades.
Chapter 5 extends the process model described in Chapter 3 to consider professional identification as an additional social identity process influencing TMS in multidisciplinary teams, wherein both a team identity and professional identity are salient. Drawing on research examining the interplay between shared common identities and unique subgroup identities, professional identification is predicted to strengthen the positive relationship between team identification and TMS (H5). A moderated-mediation hypothesis also is proposed, such that professional identification strengthens the mediating effect of team identification in the positive relationship between communication quality and TMS (H6). Study 2 details a cross-sectional field study with 126 multidisciplinary healthcare teams that, again, aimed to examine the mediating role of team
identification (H2b) in the positive relationship between quality of communication and TMS (H1b), in addition to the interplay between team identification and professional identification on TMS (H5 and H6). Using multi-level structural equation modeling, results demonstrated that communication quality promoted TMS through a collective sense of team identification, and that professional identification moderated the link between team identification and TMS. Although the nature of theinteraction was not as expected (the positive effect of team identification on TMS was only evident for teams with low levels of professional identification), results nevertheless supported the general expectation that professional identification acts as a positive resource to TMS building. Overall, the pattern of findings suggests that professional subgroup identification can compensate for low team identification, and draws attention to the interplay between shared (common) and unshared (subgroup) identities for building TMS in multidisciplinary teams.
In the general discussion (Chapter 6), a summary of the findings and theoretical contributions is provided. The findings draw attention to the importance of making a distinction between the quantity and quality of communication, and the value of understanding TMS from a social identity approach, wherein relevant social identities shape knowledge transfer and coordination practices among team members. Practical implications are provided for improving TMS and team performance in both typical and multidisciplinary teams. Finally, limitations stemming from the empirical studies provide avenues for future research.