Explorations of Heterogeneity in Models of Voter Choice

Thumbnail Image



Journal Title

Journal ISSN

Volume Title

Repository Usage Stats



In this dissertation, I examine three sources of heterogeneity in voter choice that fit into two typologies. The first of these is inter-individual heterogeneity, or the idea that different types of individuals use unique models in evaluating candidates. I examine voters’ levels of moral conviction and voters’ partisanship as sources of this type of determinacy of decision criteria. My results indicate that the morally convicted differ in the shape of their utility function for candidate assessment. Instead of having a linear function representing the relationship between the candidate’s distance from their ideal points and how much they like the candidate, the morally convicted have a convex curve. I also find that the criteria on which individuals rate candidates differs based on whether they are partisan or moderate voters. I use the method of eye tracking to find that moderates tend to weigh candidates’ policy stances much less than partisans do.

The second type of heterogeneity explored is intra-individual heterogeneity, or the idea that the same individual uses different models depending on the context she is in. In an experiment, I find this type of heterogeneity is present: it is possible to prime people to use directional or proximity theory based on whether the issue is presented through a proximity-based framework or a directional-based structure.

This dissertation contributes to the literature on heterogeneity by providing a theoretical framework within which to think of different types of heterogeneity. It also provides something new in each chapter: Chapters 2 and 4 offer the first examination of these sources of heterogeneity while Chapter 3 uses a new method in political science.





Jenke, Libby (2018). Explorations of Heterogeneity in Models of Voter Choice. Dissertation, Duke University. Retrieved from https://hdl.handle.net/10161/16892.


Dukes student scholarship is made available to the public using a Creative Commons Attribution / Non-commercial / No derivative (CC-BY-NC-ND) license.