Quantitative Methods in Program Evaluati

Pre-req: SIPA IA6501 - Quant II. The goal of this course is to provide students with a basic knowledge of how to perform some more advanced statistical methods useful in answering policy questions using observational or experimental data. It will also allow them to more critically review research published that claims to answer causal policy questions. The primary focus is on the challenge of answering causal questions that take the form “Did A cause B?” using data that do not conform to a perfectly controlled randomized study. Examples from real policy studies and quantitative program evaluations will be used throughout the course to illustrate key ideas and methods.

First, we will explore how best to design a study to answer causal questions given the logistical and ethical constraints that exist. We will consider both experimental and quasi-experimental (observational studies) research designs, and then discuss several approaches to drawing causal inferences from observational studies including propensity score matching, interrupted time series designs, instrumental variables, difference in differences, fixed effects models, and regression discontinuity designs.

As this course will focus on quantitative methods, a strong understanding of multivariate regression analysis is a prerequisite for the material covered. Students must have taken two semesters of statistics (IA6500 & IA6501 or the equivalent) and have a good working knowledge of STATA.