Published online by Cambridge University Press: 30 July 2012
The workplace has been hailed as a fruitful context for encountering difference, but other institutions of adult life — notably the church — have been downplayed, as it has been argued that self-selection produces political homogeneity within these environments. At the same time, much of what scholars know about social influence has been based on relatively blunt measures of disagreement, typically included in surveys conducted over multiple-week spans either before or after the actual voting has taken place. Inspired by work on the survival, loci, and democratic consequences of political disagreement, we survey voters at the moment when the literature would suggest they should be most likely to report agreement: Election Day. Wedding exit poll methodology with items that capture the major dimensions of networks and the content of discussion, we re-examine the contexts in which discussions take place and untangle issue-specific patterns of disagreement. We find evidence that church-based networks fulfill important democratic roles relative to other contexts, exposing individuals to cross-cutting discourse while serving as unique sources of information in the midst of broader electoral environments.