I am a political methodologist that studies American political institutions. My work exists at the intersection of formal models of political behavior, statistical modelling, and legislative behavior. My work in this area received the John T. Williams Dissertation Prize in 2018 from the Society for Political Methodology (for the best dissertation proposal in the area of political methodology).
My research in political methodology focuses on machine learning, Bayesian statistics, and nonparametric statistical models. In particular, I explore estimation strategies for latent variable models and propose a number of Bayesian nonparametric methods that solve problems related to measurement in the social sciences, such as dimensionality, complex dependence structures in latent variables, and hierarchical clustering within latent constructs.
Theoretically, I combine both formal theoretic and statistical approaches to explore theories of legislative behavior. I am particularly interested in the role that ideal point estimates have played in driving theories of legislative behavior and how the assumptions inherent in roll call scaling approaches have led to incorrect inferences about decision making. This work examines long-held theories related to party control, agenda setting, issue-based voting, gridlock, and polarization in the U.S. legislature and examines how appropriate models of ideal points can lead to different conclusions about legislative voting and representation over the course of U.S. history.