KAFP 6360 - Policy Analysis and Policy Evaluation - Intermediate

Public policy evaluation seeks to estimate the causal effect of policies on relevant outcomes. The extent to which we can rigorously evaluate public policies depends on their design and implementation. This course aims to give students the tools to design and implement policies conducive to their rigorous evaluation. We will introduce five research designs (randomized experiment, encouragement design, selection on observables, regression in discontinuities, and difference-in-differences), discussing the identifying assumptions behind these designs and the conditions under which these assumptions are most likely to hold. To do all of this, we will use mathematical abstraction. We will not use statistical software but focus instead on the intuitions behind research designs.

Throughout, we will use specific examples of policy evaluations to build intuition and encourage future policy-makers to design policy in a way that allows for rigorous evaluation.

Mathilde Emeriau is an Assistant Professor in Political Science at Sciences Po’s médialab. Her research centers on discrimination and immigrant integration, focusing on France. She is currently working on impact evaluations of policies related to immigrant legalization, proximity policing, and social mixing in public schools. She holds an M.A. in Economics from the Paris School of Economics and a Ph.D. in Political Science from Stanford University. Before starting her Ph.D., she worked for JPal on health and cash transfer program evaluations in Cameroon, Uganda, and Liberia.

A basic understanding of probability theory is required to follow the course (random process, random variables, expectation and conditional probability).

• A midterm test ("galop") accounting for 40% of the final grade.
• A group project to be submitted by early December accounting for 60% of the final grade