by Yiqing Xu

This course explores how we can make policy recommendations using data. The overall goal of this course is to introduce a basic framework for policy evaluation -- what we call design-based causal inference -- essentially, how we can use statistical methods to answer research questions that concern the impact of some cause on certain policy outcomes. We cover the mostly commonly used research designs, including randomized experiments, selection on observables, and difference-in-differences, and analyze the strengths and weaknesses of these methods using applications from the real world.

From a skill-builiding point of view, this course has three objecives:

1. Introduce an analytical framework for policy evaluation and discuss several related methods
2. Introduce the most basic (and some of the most important) statistical concepts
3. Equip students with basic coding skills with R

• Syllabus

• Introduction
• Potential Outcomes Framework
• Randomized Controlled Trials
• Selection on Observables
• Difference-in-Differences
• Final Review

I disccus one real-world application at the beginning of each class.

• Stop and Frisk
• Robots and Jobs
• Seeing Inequality
• Management Can Be Taught
• Fighting Corruption
• Government Responsiveness in China
• Why You Should Sleep More
• Inequality in the United States
• Volunteer for America
• Politicians for Sale
• The Tragedy of Child Soldiering
• Central Bank and Financial Crisis
• Direct Democracy = Better Governance?
• Eradicating Malaria in the Americas