Social Networks and Human Agency in Knowledge Sharing and Creation: Lessons from a Public Research University

Effective knowledge management has become important for the survival and prosperity of public research universities in competitive environments. However, theoretically, the existing knowledge management frameworks that adopt methodological individualism or collectivism have some limitations in explaining relational and matching mechanisms of knowledge sharing and coproduction in organizations. Practically, the business-oriented literature on knowledge management does not provide much guidance on how to manage intramural knowledge networks in public research universities. To address these gaps, this study examines the following research question: How is knowledge mobilized and coproduced among faculty in knowledge networks in public research universities? To answer the question, three subquestions are explored: (1) How do network structures affect knowledge sharing and coproduction? (2) How are expressive ties associated with instrumental ties? (3) How do individual attributes affect knowledge sharing and coproduction?.;Multiple methods based on exponential random graph models (ERGMs) are used to analyze data collected from 39 faculty members in a public research university in the United States. ERGM analyses consistently demonstrate that intramural knowledge-sharing networks among faculty in public research universities are self-organizing and that dyadic bonding based on reciprocal benefits and bridging social capital works in the self-organizing process. Such structural effects are strong and robust across expressive and knowledge-sharing relations. ERGM estimation results also provide evidence of homophilic bonding among actors in terms of ethnicity and departmental affiliation in research and teaching knowledge sharing, and in age in research knowledge sharing. This study empirically demonstrates that knowledge sharing is highly influenced by individuals' motivation and perceived values of others' knowledge resources.;One of the important findings of this study is that the mechanisms of knowledge coproduction are not identical to those of knowledge sharing. The knowledge coproduction network is also self-organizing, and bonding social capital embedded in triads is associated with the self-organizing process. The study results also indicate that actors' information on others' knowledge, which is partner-specific, and the perceived value of others' knowledge, which is not partner-specific, are both associated with knowledge coproduction ties. The effect of actors' communication style on knowledge sharing and coproduction was also tested, but the study did not find evidence for this effect.