Sara Loving

Sara Loving

Mentor

Dr. Daniel A. Smith

College

College of Liberal Arts and Sciences

Major

Political Science and Statistics

Minor

Spanish

Academic Awards

University Scholars Program 2021, President's List, Bright Futures Academic Scholarship 2018, Students Taking Academic Responsibility (STAR) Award

Organizations

Alpha Delta Pi Sorority, Voting and Election Science Team, Acts of Random Kindness Club, Ignite, Undergraduate Research Scholars Program

Volunteering

Inheritance of Hope, Ronald McDonald House

Research Interests

Elections, Voting, Redistricting

Hobbies and Interests

Research Project

Movers and Political Polarization

Bill Bishop’s 2008 book, The Big Sort, advanced a bold claim about polarization in the United States: the residential clustering of like-minded people is creating ideological and homogeneous communities of party uniformity. His work, along with several other scholarly studies that both support and challenge Bishop’s findings, rely primarily on aggregate-level census and presidential voting statistics to make inferences about whether the spatial polarization of the electorate is (or is not) occurring. However, the question of whether individuals are actually changing their party identification when they geographically sort themselves remains unanswered. Using a more sophisticated measure of political behavior—official individual voter registration records in Florida and North Carolina over time—my research investigates whether registered voters who change residences within these states are also more likely to change their party registration upon moving into a more Republican (or Democratic) neighborhood and if so, are more likely to alter their voting behavior.

Using publicly available Florida and North Carolina voter files from all 12 months of 2018 and all twelve months of 2020, I plan to identify registered voters who moved to a new residence at some point during the two election years. I will carry out exploratory data analysis to detect patterns in age, race, gender, and other fixed factors. This will reduce the likelihood of any confounding variables in the model. Next, I will use voting statistics to determine if people who moved were more or less likely to change their party registration, vote in local, primary, and general elections, and change their method of voting compared with non-movers. This will involve employing logit models to predict whether one (a) changed party registration, (b) voted in an election, or (c) voted by mail (rather than in person) with the main predictor variable of whether one changed residence. By answering these questions, I hope to gain a deeper understanding of the effect of moving on various aspects of political polarization and voting behavior.