Three Papers on Electoral Geography

Date of Award

Spring 5-15-2013

Author's School

Graduate School of Arts and Sciences

Author's Department

Political Science

Degree Name

Doctor of Philosophy (PhD)

Degree Type

Dissertation

Abstract

This dissertation is composed of three papers on electoral geography. The first, entitled "MCMC Network Voronoi Algorithm for Electoral Districting", develops a sampling routine (along with R code that implements it, available in the Software section) for systematically exploring the space of partitions that define a contiguous districting plan. Leveraging the properties of MCMC sampling, the routine produces statistically useful distributions of quantities of interest (such as partisan bias, or population equality) over the space of possible districting plans. This should allow reviewers of proposed plans to better answer questions like "Does this proposed plan generate an extreme bias in favor of party X?". The second paper, entitled "The Partisan Effects of Distance Induced Voting Costs" evaluates whether the physical location of polling places can induce biases in favor or against certain parties. The seemingly administrative decision of where to place polling stations can effectively induce different costs of voting for different types of voters. Insofar as preferences for certain parties are not uniformly distributed over space, these different costs can affect supporters of one party more than supporters of another. I use geocoded polling place locations in Hungary to assess these differential costs, as well as their impact on partisan biases. Finally, the third paper, entitled "The Policy Effects of Partisan Bias", evaluates the impact of geographically induced partisan biases on policy outcomes. Using a spatial voting formal model, and evaluating the model's comparative statics using government finance data for 25 countries (and, more specifically, national budget compositions), I study the effects of partisan biases, estimated using the approach of King (1990). I find that partisan bias can generate policies that are closer to those pre- ferred by advantaged parties, and provide evidence to this effect using a Dirichlet model.

Language

English (en)

Chair and Committee

Brian F. Crisp

Committee Members

Brian F. Crisp, Marcus Berlian, Randall Calvert, Andrew D. Martin, Jonathan Rodden, Margit Tavits

Comments

Permanent URL: https://doi.org/10.7936/K75X26XF

This document is currently not available here.

Share

COinS