Social Policy & Evaluation Methods

This course introduces students to the empirical methods economists use to evaluate the effects of social policies. We will ask a simple question — does this policy work? — and discover that answering it rigorously requires careful thinking about causality, data, and research design.

The course draws primarily on Mostly Harmless Econometrics (Angrist and Pischke, 2008) and Causal Inference: The Mixtape (Scott Cunningham), but the technical level is low, no prior knowledge of statistics or econometrics is required. The mathematics stays light — we work with averages, simple regressions, and differences between groups — but the conceptual demands are real. Students will be expected to think carefully about what data can and cannot tell us.

The course is organized in three parts. We will begin with the fundamental problem of causal inference: why it is so hard to know whether a policy actually caused an outcome, and why correlation is not enough. We then move through the main evaluation tools economists use in practice — difference-in-differences, regression discontinuity design, matching, instrumental variables, and randomized experiments — studying each through real policy applications in health, education, labor markets, housing, and finance. In the final part of the course, students apply what they have learned by working through a policy evaluation of their own, from question to design.

*FULL SYLLABUS COMING SOON*