Principles of Randomized Evals

3 Credits · Level: Undergraduate

This course examines the principles underlying randomized evaluations and their analysis. Students learn how randomization enables causal identification, how treatment effects are estimated using regression and likelihood-based methods, and how real-world complications (such as noncompliance, attrition, and missing data) threaten causal inference. Building on prior conceptual exposure to causal inference in PPOL 3000 , we dive into the statistical foundations of experimental analysis, heterogeneous treatment effects, and identifying assumptions, and apply these methods to real data throughout the course. By the end of the course, students will be able to design a randomized experiment, estimate and interpret causal effects using R, assess the validity of their findings, and communicate evidence-based policy conclusions.