Publication Date

5-2023

Date of Final Oral Examination (Defense)

2-21-2023

Type of Culminating Activity

Thesis

Degree Title

Master of Science in Political Science

Department

Political Science

Supervisory Committee Chair

Charles Hunt, Ph.D.

Supervisory Committee Member

Jaclyn Kettler, Ph.D.

Supervisory Committee Member

Jeffrey Lyons, Ph.D.

Abstract

Since well before the U.S. presidential election of 2020, voter identification laws have been a topic of discussion amongst politicians, voters, the news media, and scholars. Many have questioned the focus and true reason for their creation, their implementation, their effects and potential unintended consequences. Specifically, many have argued that voter identification laws pose too great a barrier to potential voters to be worth the benefits gained in election security. Since the election of 2020, those discussions seemed to magnify. For example, in a May 2021 speech, President Biden repeated similar assertions made in the past by scholars and activists (Amwine and Smeal 2013), but in a much higher-profile fashion. Currently, thirty-five of the fifty U.S. states have voter identification laws, and it is the changes to some of these laws that have received criticism. States like Georgia and Texas have taken steps to further revise their voter identification laws which have resulted in the filing of numerous lawsuits. The recent developments and changes to current voter identification laws have led to a new unanswered question: do these laws adversely affect the turnout of women voters more than men? According to the US Census Bureau, we know that women turn out to vote at higher levels than men. But does the turnout gender gap decrease when voter identification laws are implemented and increase in level of strictness? If so, this could suggest that voter identification laws do adversely affect turnout of women more than men. This is the research question I aim to answer here. To answer my hypothesis, I created a dataset using voter turnout data from the U.S. Census Bureau for presidential election years 2000-2020. I then created an index to measure the strictness levels of voter identification laws in all 50 states. I also created a competitiveness scale to measure the competitiveness of the presidential and senate races for the same election years and collected numerous control variables thought to affect voter turnout. After collecting that data and applying advanced statistical techniques and multivariate regression models using both random and fixed effects, I found that the evidence was largely null and suggestive at best that voter identification laws do adversely affect turnout of women more than men. The descriptive models initially revealed indicative evidence to support the theory; however, after running the advanced regression models that initial evidence did not replicate, as they revealed no statistically significant differences in the turnout gender gap as the voter ID index increased.

DOI

https://doi.org/10.18122/td.2052.boisestate

Download

Share

COinS