Evidence-based policymaking relies on the ability to combine causal methods and work with data. The course introduces the key causal inference methods used to evaluate programs and policies, using the statistical software R. Case studies will be used to emphasize the methods presented. The course will focus on practical applications of evaluation techniques.
We begin with a gentle introduction to R followed by the potential outcomes and causal graph framework to build an understanding of the cause. Topics include experiments (RCTs), difference-in-difference, co-variate adjustment via regressions, and matching.
Session format: Short lecture on the concepts followed by lab work with R
Course methods and materials
Program and policy evaluation requires practice. The in-class lab work, homework assignments, group projects, and exam are designed to help students master the techniques. The concepts will be presented via lectures and case studies, while the lab work with R aims to build data skills. Students are encouraged to work in groups on the assignments. Class attendance and participation are highly encouraged, and unjustified absences will be noted.
Students are required to prepare for class with the assigned readings. Additional materials for those interested includes:
1. Mastering `Metrics: The Path from Cause to Effect (2014) by Joshua Angrist and Jorn Steffen Pischke.
2. Causal Inference: The Mixtape (2021) by Scott Cunningham
3. Quantitative Social Science (2017) by Kosuke Imai
4. Dayal, V., 2015. An introduction to R for quantitative economics. SpringerBriefs in Economics.