Date of this Version
Jackson, J. (2018). Thousands of small battles: A case study on the impact of political discussion networks on vote choice in caucuses. PhD diss. University of Nebraska-Lincoln.
In this dissertation, I seek to refine generalizations about the influence of political discussion networks on voting behavior, mainly developed to explain behavior in general elections, to nomination contests, a comparatively underdeveloped area of inquiry. This study also contributes to a greater understanding of the behavior of Iowa caucus attendees, an understudied area despite Iowa’s importance (along with New Hampshire) in our sequential presidential nominating system. I make several findings affirming theories on social influences on voting behavior within the context of nomination contests. The first is that individuals are reasonably accurate when predicting which candidate a political discussion partner supports despite the multi-candidate nature of the nomination contest and not having party ID as a frame of reference. Second, although there is some evidence for campaign effects in the form of contacts from campaigns or supporters of candidates, political discussion networks have a stronger influence on vote choice in nomination contests than do campaign effects. A unique finding of this study that homophily within a group may trigger strategic voting behavior, in the form of supporting the main rival of the leading candidate within a political discussion network, as well as bandwagoning. Finally, an emergent study of local political elites finds variation between local elected officials, party leaders, and party activists in terms of ideology and the size of political discussion networks. Those findings suggest that local party leaders may play a more important role than local elected officials in diffusing political information in their communities during nomination contests.
Advisor: Elizabeth Theiss-Morse