"Interaction, Ideology, and Identity in the U.S. Senate, 1979-2001"

Abstract

This article reconciles two seemingly incompatible expectations about social interaction and ideological change. One theoretical perspective predicts that an increase in interaction between two actors will promote subsequent convergence in their ideologies, while another anticipates ideological divergence. Integrating network-analytic approaches to social influence with social psychological theories of identification, we argue that interaction between actors who share a salient social identity promotes ideological convergence, while interaction between actors with contrasting social identities leads to divergence. Moreover, the consequences of social identity for influence depend on the local context of interaction. Social identity’s effects on influence are greatest in groups with a limited, rather than extensive, history of prior collaboration and with moderate, rather than low or high, levels of ideological diversity. Empirical support for these propositions comes from analyses of the U.S. Senate from 1979 to 2001. Using two distinct indicators of social identity —party affiliation and region of representation—we demonstrate that, as the level of interaction between senators changed, the ideological distance between them subsequently shifted as a function of their respective social identities and characteristics of committees they served on together. These findings contribute to research on social influence, elite integration and political polarization, and social identity.