Recent research suggests that people obtain useful knowledge from others with whom they work closely and frequently (i.e., strong ties). Yet there has been limited empirical work examining why this is so. Moreover, other research suggests that weak ties provide useful knowledge. To help integrate these multiple findings, we propose and test a model of two-party (dyadic) knowledge exchange, with strong support in each of the three companies surveyed. First, the link between strong ties and receipt of useful knowledge (as reported by the knowledge seeker) was mediated by competence-and benevolence-based trust. Second, once we controlled for these two trust dimensions, the structural benefit of weak ties became visible. This latter finding is consistent with prior research suggesting that weak ties provide access to non-redundant information. Third, we found that competence-based trust was especially important for the receipt of tacit knowledge. We discuss implications for theory and practice.
3Promoting knowledge creation and transfer within organizational settings is an increasingly important challenge for managers today (Kogut & Zander, 1992). Organizations that can make full use of their collective expertise and knowledge are likely to be more innovative, efficient, and effective in the marketplace (Grant, 1996;Wernerfelt, 1984). Yet ensuring effective knowledge creation and transfer has proven a difficult challenge. At least three separate literatures-on social networks, trust, and organizational learning/knowledge-have addressed aspects of the knowledge transfer problem. We propose and test empirically a theoretical approach that synthesizes these three streams.
Structural Characteristics of Knowledge TransferSocial network researchers have offered clear evidence of the extent to which knowledge diffusion occurs via social relations (e.g., Rogers, 1995). Work dating to Pelz and Andrews (1968), Mintzberg (1973), andAllen (1977) indicates that people prefer to turn to other people rather than documents for information. For example, Allen (1977) found that engineers and scientists were roughly five times more likely to turn to a person for information than to an impersonal source such as a database or file cabinet. More recently, Cross (2001) found that even people with ready access to well-populated electronic and paper-based sources of information reported seeking information from colleagues significantly more than from these sources. In general, researchers have found relationships to be important for acquiring information (Burt, 1992); learning how to do one's work (Lave & Wenger, 1991); making sense of ambiguous environments or events (Weick, 1979); and solving complex problems (Hutchins, 1991).Social network theorists have focused much of their attention on structural properties of networks (Adler & Kwon, 2002), such as structural holes at the network level (Burt, 1992) and tie strength at the dyadic level (Granovetter, 1973). Tie strength characterizes the closeness of a relationship between two...