Description

This course explores how we can make policy recommendations using data. The overall goal of the course is to provide a survey of the most commonly-used empirical tools for political science and public policy research. Our focus is design-based causal inference, or the use of statistical methods to answer research questions that concern the impact of some cause (e.g., an intervention, a change in institutions, passage of a law, changes in economic conditions, or policies) on a certain outcome (e.g., vote choice, income, election results, levels of violence, political attitudes). We cover a variety of causal inference designs and methods, including experiments, regression, matching, difference-in-differences, and regression discontinuity designs. We will analyze the strengths and weaknesses of these methods using applications from the real world.

Syllabus

Slides

1 Introduction and Course Overview

2 The Potential Outcomes Framework

3 Omitted Variable Bias and Selection Into Treatment

4 Experiments and Randomization

5 Experiments and Inference

6 Spillovers, Noncompliance, and Attrition

7 Interaction Effects and Heterogeneous Treatment Effects

8 Selection on Observables

9 Linear Regression

10 Matching

11 Differences-in-Differences

12 Synthetic Control Methods

13 Regression Discontinuity

14 Instrumental Variables

15 Encouragement Experiments and Fuzzy RD

16 Conclusion

Assignments

Problem Set 1

Problem Set 2

Problem Set 3

Problem Set 4

Problem Set 5

Optional Extra Credit Assignment

Final Paper