Examining Partisan Advantage In Congressional Maps Using Simulations Based On Election Data

<p>Partisan gerrymandering has been and will continue to be a topic of interest in the coming years. States will soon begin their redistricting process following the 2020 Census. We introduce a method of simulating Congressional elections which provides a new way of examining and visualizing the votes-to-seats relationship for a state Congressional map using past election data. We are able to build upon Mira Bernstein's method of uniformly simulating elections by injecting a data-driven component of variation into the simulations. Additionally, we are able to directly evaluate the accuracy of our simulations using a type of cross-validation. We compare our results from a handful of notable states to other measures of partisan gerrymandering, such as the efficiency gap, and do so in light of recent court cases and other important contexts. </p>

Name thesis:degree_name

Master of Science (MS)

Level thesis:degree_level

Master's

Discipline thesis:degree_discipline

Mathematics and Statistics

Grantor dc:publisher

Eastern Kentucky University

Year

2019

Author dc:creator

• Morgan, Zachary James

• Cross-Validation
• Efficiency Gap
• Gerrymandering
• Partisan Asymmetry
• Redistricting
• Simulations
• American Politics
• Statistics and Probability

Statement dc:rights

• Copyright 2019 Zachary James Morgan

Repository record dc:identifier

https://encompass.eku.edu/etd/578

OAI identifier oai:identifier

oai:encompass.eku.edu:etd-1576