Partisan Conversion and Partisan Activation: The Behavioral Consequences of Partisan Segregation

Abstract

Increasing political segregation in the United States raises new concerns as to how living in politically homogeneous communities influences voters and divides political parties. This dissertation investigates the behavioral consequences of geographic partisan polarization, analyzing how living in homogeneous political communities influences voters’ partisanship and political participation. To this end, I develop data on residential exposure to Democratic and Republican neighbors for every voter in the United States over the past decade, leveraging precise information on each voter’s residential location, partisan affiliation, and political behavior. With these data I present new evidence on the extent and causes of partisan sorting in the United States and test how where Democrats and Republicans live in relation to one another influences politics. The first study uses data on U.S. voters from 2018 and advances in spatial data computation to measure exposure to Democratic and Republican neighbors for every voter in the country, demonstrating high levels of partisan residential isolation. The second study uses a panel of voters from 2008-2020 in 5 states and an original survey to demonstrate that living near Democrats or Republicans influences voters to change their registration to match neighbors' partisanship. The final study uses data on every registered voter across 30 states from 2012-2021 to measure the effect of partisan exposure on political participation and civic engagement, finding that voters become more active in politics when they live surrounded by neighbors who share their partisanship.