Sharing Knowledge:
Institutional and Technological Facilitators and Inhibitors

10.6.2009 SPEAKER: David Lazer, PHD, Associate Professor of Political Science and Computer Science, Northeastern Univ.; Director, Program on Networked Governance, Harvard Kennedy School, Harvard Univ. MODERATOR: Rajiv Gupta, MD, PHD, Director, VCT Lab and Assistant Radiologist, Department of Radiology, MGH

Forum Summary

Social networks facilitate the sharing of knowledge, making it easier for people to avoid making mistakes.  David Lazer, PhD, and a team of researchers recently studied information sharing in three settings.  First, they studied a network of fifty state health officials.  Then they looked at a network of 435 congressional web design teams, and finally they considered all forensic DNA laboratories in the United States.  The researchers interviewed network members to learn about situations conducive to knowledge sharing, and they also assessed social networks quantitatively through surveys asking each participant to identify (from a list) everyone he or she knew within a network.

The researchers found that the network of state health officials was similar to a hub with many spokes.  Some state health officials said that they do not talk to any of their peers whereas a few said that they communicate with many colleagues.  The most connected person in the network was more connected than the twenty-five least connected people combined.  Officials from the same region under the Department of Health and Human Services were more likely to share knowledge, as were officials from adjacent states.  A number of reasons explaining the sparseness of the network emerged over the course of the study.  Some officials found it difficult to locate expertise in the network, and others felt limited by the lack of pre-existing relationships.  The state health officials tended to be embedded in local politics, and many perceived their needs to be exceptional or unique.  When new state health officials trained together, however, they formed many long-distance connections, suggesting that social networks can be rewired. 

Knowledge sharing was limited in the network of congressional web designers.  The most common way that congressional web designers gathered information was simply by looking at the layout of other websites.  Again, perceptions of exceptionalism prevented many teams from communicating productively.  Tenure in the House of Representatives is usually short, so people had difficulty forming relationships.  Even members of the same political party expressed little sense of common purpose.

People at the forensic DNA laboratories were more willing to share knowledge.  The system for analyzing forensic evidence in this country is fairly decentralized, and as one might expect, larger institutions tended to have more connections in the network.  As in the network of state health officials, connections appeared to be influenced by geography, perhaps because crime often crosses state boundaries.  The lack of rapid turnover in the field appeared to contribute to knowledge sharing, as did relationships formed at a training meeting run by the Federal Bureau of Investigation (F.B.I.).  There was some parochialism in the network, but everyone agreed that DNA is the same no matter where in the United States it is found, leading to a sense of common purpose among laboratory personnel.                  

Taken together, the three studies suggest that certain general principles affect knowledge sharing through social networks.  Physical proximity appears to encourage sharing, as does the perception that knowledge is universally applicable.  A sense of common purpose and low employee turnover also encourage sharing.  Exogenous interventions, such as new training programs, can rewire networks to promote the sharing of knowledge.

---

Post a Comment