Monday, November 18, 2024, 3:30pm to 4:20pm
College of Public Health Building , C217 CPHB
145 North Riverside Drive, Iowa City, IA 52246

Nandita Mitra, PhD

Professor of Biostatistics, Department of Biostatistics, Epidemiology, and Informatics, Perelman School of Medicine

Vice Chair of Education, Department of Biostatistics, Epidemiology, and Informatics, Perelman School of Medicine

Co-Director, Center for Causal Inference, Perelman School of Medicine

Title: A Causal Framework for Evaluating Drivers of Policy Effect Heterogeneity Using Difference-in-Differences

Abstract: In designing and evaluating public policies, policymakers and researchers often hypothesize about the ways in which a policy may differentially affect a population and aim to assess these pathways in practice. For example, when studying unhealthy food or beverage excise taxes, researchers might explore how cross-border shopping (i.e., spillover), economic competition (i.e., interference) and store-level price changes influence sales. However, policy evaluation designs, including the difference-in-differences (DiD) approach, traditionally target the average effect of the intervention rather than the underlying drivers of heterogeneity. Extensions of these approaches to evaluate drivers of policy effect heterogeneity often involve exploratory subgroup analyses or outcome models parameterized by driver-specific variables. However, neither approach investigates potential drivers within a causal framework, limiting the analysis to associative relationships between drivers and outcomes, which may be confounded by differences among sub-populations exposed to varying levels of the drivers. Therefore, rigorous policy evaluation requires robust methods to adjust for confounding and accommodate the interconnected relationship between stores within competitive economic landscapes.

In this talk, I will present a causal framework for evaluating policy drivers by studying the Philadelphia beverage tax. Our approach builds on recent advancements in semiparametric causal effect curve estimators under DiD designs, offering tools and insights for assessing the drivers of policy effect heterogeneity.

This is joint work with my PhD student, Gary Hettinger, at the University of Pennsylvania.

Please be aware that this seminar will not be recorded as Dr. Mitra will be presenting unpublished work. A Zoom link will be provided upon request to non-departmental attendees.

Individuals with disabilities are encouraged to attend all University of Iowa–sponsored events. If you are a person with a disability who requires a reasonable accommodation in order to participate in this program, please contact Emily Stagman in advance at 319-335-1585 or emily-stagman@uiowa.edu.