James Kitts
James Kitts (University of Massachusetts Amherst)
James A. Kitts is professor of sociology and Director of the Computational Social Science Institute at the University of Massachusetts. He earned his Ph.D. from Cornell University in 2001 and previously held faculty appointments at Columbia University, Dartmouth College, and the University of Washington. He models dynamics of social networks within and across groups, cooperation and competition within and across organizations, and social influence conveyed through networks. He is presently Principal Investigator on an NIH R01 grant investigating the diffusion of health behavior on networks of interaction and sentiments among youth.
Rethinking Social Networks in the Age of "Big Brother": Perspectives from Computational Social Science
Social network analysis and theory has shifted our analytical focus from individuals or groups to social relations or ‘ties.’ The ties that constitute these networks have been conceptualized as socially constructed role relations such as friendship or coauthorship; interpersonal sentiments such as liking or hatred; behavioral interactions such as communication or citation; or opportunity structures for resource exchange. These literatures treat the network as a temporally continuous and stable graph, i.e., a latent structure that underlies whatever interaction behavior, emotions, attributions, or exchange opportunities we observe. I will analyze the interplay of these concepts, consider where ties (and non-ties) are likely to correspond across these four domains, and thus assess where we may stretch theories or data from one domain to another. Then I will discuss some lenses emerging from computational social science – wearable sensors, location-aware devices, electronic calendar meetings, logs of phone calls, messages, online transactions, etc. I will ask how these time-stamped event series correspond to the conventional network concepts above, and call for a new approach: Directly theorizing and analyzing the structural-temporal interdependencies of social interaction events redirects our attention from structural patterns to social processes.
This talk draws from a theoretical working paper and an empirical study that I recently published in the American Journal of Sociology.