(iPoster) Dynamic Voter Behavior: Longitudinal Analysis of Electoral Choice in 2016 and 2020

• In Event: Poster Session: Public Opinion in the United States

Abstract

This study investigates how partisan identification, issue preferences, and candidate evaluations interact to shape voter behavior across two critical U.S. presidential elections (2016 and 2020). Leveraging longitudinal panel data from the American National Election Studies (ANES), the research extends the dynamic simultaneous equation model proposed by Markus and Converse to analyze the evolving nature of voter decision-making in an era of heightened polarization, demographic shifts, and media transformation. The analysis incorporates stable predictors, such as partisan loyalty, alongside more variable elements, including issue salience and candidate evaluations, to model changes in voter behavior over time.
Key findings are anticipated to reveal the persistence of partisan identification as a predictor of vote choice and the temporal variation in issue salience driven by contextual factors (e.g., healthcare, economic recovery). The broader goal of this research is to inform what influences voter decision-making. By adapting a foundational model to contemporary electoral contexts, this research offers new insights into how identity-driven politics, issue dynamics, and candidate perceptions converge to influence electoral outcomes.
The findings contribute to the APSA 2025 theme, "Reimagining Politics, Power, and Peoplehood in Crisis Times," by elucidating how identity-based divisions, misinformation, and global crises shape voter behavior and democratic legitimacy. This study underscores the importance of longitudinal analysis in capturing the dynamic interplay of stability and change in electoral decision-making, advancing our understanding of political participation in polarized and crisis-driven environments.

Authors

• Jessica Grassetti, Rutgers

• George D. Quinn, Rutgers University