JOINT WORKSHOP: POPULATION STUDIES & QUANTITATIVE RESEARCH METHODS
Abstract: Self-reported turnout in public opinion surveys is well known to exceed actual turnout figures, yet scholars disagree about why. Some attribute the gap to social desirability and deliberate overreporting, while others emphasize nonresponse bias or errors introduced during turnout validation. This talk addresses both the measurement and inference challenges that arise when validating turnout without unique identifiers. First, we use a probabilistic record linkage framework to link the American National Election Studies (ANES) to a national voter file. The resulting validated turnout rates closely match actual turnout. Second, we tackle a key but often overlooked issue: how to incorporate linkage uncertainty into downstream analyses. We introduce a general post-linkage regression method. We illustrate its value by showing that overreporting is systematically associated with education, income, age, and political interest.
Ted Enamorado is an Assistant Professor of Political Science at Washington University in St. Louis, where he is affiliated with the Center for the Study of Race, Ethnicity & Equity and the Division of Computational & Data Sciences. He holds a PhD in Politics from Princeton University (Political Economy Program) and specializes in Political Economy and Political Methodology, focusing on probabilistic record linkage and improving probabilistic methods. Substantively, his research examines the impact of criminal justice inequities on political participation. Prior to graduate school, he worked as a Research Fellow at the Inter-American Development Bank and as a Consultant at the World Bank.
This workshop is being hosted jointly by the Population Studies Workshop and the Quantitative Research Methods Workshop at ISPS, and is open to the Yale community.