I am a quantitative methodologist and Ph.D. candidate in Educational Psychology at the University of Wisconsin–Madison. My sits at the intersection of causal inference, machine learning, and measurement.
My work focuses on moving beyond average effects to uncover treatment variation and my dissertation, Measurement, Regularization, and the Resolution of Causal Claims, investigates the factors that contribute to heterogeneous treatment effects and how modern methods do (not) distinguish between said factors. Across three simulation studies with empirical applications, my dissertation shows that the estimation objective, the instrument used to measure the outcome, and shrinkage all influence where a model determines where variation exists. I also provide diagnostic tools that applied researchers can use to see these patterns in their analyses.
This work is done in pursuit of two things. The first is a Discover, Characterize, Confirm (DCC) framework for causal machine learning, in which CATE estimation is held to the same standards as measurement of latent traits in individuals. The second is to improve the equity and efficacy of interventions and policies in education, and to identify where modern methods for causal inference and the analysis of heterogeneous treatment effects provide less information or attenuate the results from underrepresented groups.
Education
PhD in Educational Psychology | Expected 2027
University of Wisconsin-MadisonMS in Educational Psychology | 2024
University of Wisconsin-MadisonBS in Interdisciplinary Studies | 2022
North Carolina State University
