Abstracts Political Science

The group-ishness of voting: preferences towards group membership, within-group authority, and between-group hierarchy shape and predict the way we vote

by Denise Baron

Attempts to understand and predict voting have often pitted potential explanations against each other: policies versus partisanship, identities versus ideologies. This thesis, instead, suggests a pluralist framework of group-ishness, which highlights the role of group membership, coordination, and competition in our political cognition and behaviour. Instead of putting prominent theories in competition with each other, research on this topic integrates group identities (i.e. national identification and partisan affiliation), orientations towards authority within a group (i.e. authoritarianism), and preferences for the distribution of rights and resources between groups (i.e. egalitarianism). This thesis provides an argument for why group-based preferences are so strongly linked to voting decisions, experimentally tests this framework with a series of survey experiments and validates it with actual election results. The findings indicate that our group-based preferences influence our voting decisions and perceptions of candidates and can also be used to predict election results. The first paper employs a discrete choice experiment and identifies shared group-based preferences as highly influential on voting decisions. Beginning with a broad consideration of social feelings, perceptions, and commitments, the first study in this paper confirms the importance of group-based preferences based on the commitment to a shared group and to principles for distributing power and resources within the shared group as well as between groups. The second paper confirms that shared group-based commitments are underlying voters’ perceptions of similarity with candidates as well as vote intention, more so than shared socio-demographic characteristics. This paper also considers perceived similarity alongside the traditional candidate traits of competence and warmth, and the results indicate perceived similarity is more closely linked to vote intention than candidate warmth or competence. The third paper considers this framework within actual election contexts and explores the relative predictive ability of such a framework for vote choice and election results as compared with traditional predictors of political ideology and demographics. Overall, these findings contribute to the growing literature on the group-based foundations of our political preferences and behaviours, contributing evidence of both causal links and application to actual election contexts.