Student Voting Habits in Light of COVID-19 Disruptions

• In Event: Electoral Engagement

Abstract

Although there has been a substantial amount of research on improving turnout among college-aged voters, there has been little research examining where college students choose to vote. This is because identifying the student voting population is a complicated task. Unlike other populations, college students have the option to vote in their hometown or their campus address. This paper takes an alternative approach to the typical survey analysis by using geocoded voter files to identify students based on their age and residence. Specifically, this paper examines the factors which influence college students’ choice to cast their ballot in their college town or hometown in the 2012-2020 general elections in North Carolina.
This study posits that COVID-19 disruptions on campus will impact student voting through three mechanisms. First, students will not be exposed to as many efforts by the community targeted towards lowering the informational and administrative costs for newly eligible voters. Given the reduced foot traffic on campus, it is less likely that students will encounter registration drives, campaign activity, information sessions, and informal conversations with peers about how to register and vote on-campus. Moreover, policies like on-campus polling places and on-campus vote centers will not have the same effectiveness when colleges are at a reduced capacity. Second, campus de-densification policies and quarantines from positive COVID-19 tests have meant that many students did not return to campus for the Fall 2020 semester or were unexpectedly sent home. This has greatly increased the level of residential uncertainty of college students. Third, the disruption of social ties resulting from students relocating unexpectedly and isolating from their community through social distancing has removed many social rewards associated with voting. Although students may be voting in other locations, this study hypothesizes that fewer students will cast a ballot in their campus community.
Using a geocoded North Carolina voter file, preliminary analyses using linear and generalized linear mixed-effects regression models suggest COVID-19 disruptions have decreased student voting in on-campus precincts. COVID-19 de-densification decreased the effectiveness of on-campus polling places in attracting student voters given that they were associated with lower counts of student voters than in previous years. When available, vote centers on campus decrease the proportion of hometown voting.
Taken together, these findings have real political consequences. With fewer students voting on campus, the strength of the student voting bloc and young voters, more generally, was diluted in areas where they have the most consequence (e.g. local and state legislative races). Given the potential for long-term COVID-19 disruptions as new variants emerge, this line of inquiry is ripe for further investigation.

Author

• Nadine Suzanne Gibson, University of North Carolina at Wilmington