PDC23: Propensity Score Methods for Causal Inference in Education Research

Session Type: Professional Development Course

Abstract

Through lectures and hands-on activities, this course will introduce basic concepts of propensity score methods, including matching, subclassification, and weighting, as well as the use of software packages such as R. Packages of propensity score methods in SAS, Stata, and SPSS will also be briefly introduced. This course is appropriate for faculty members, graduate students, and applied researchers. Participants will learn why and when we need propensity score methods for making causal inference in educational research and how to perform propensity score methods in R—as well as in SAS, Stata, and SPSS—using samples of a large-scale data set from the Educational Longitudinal Study of 2002. Instructions for downloading and installing R software and related packages, as well as example datasets, will be provided to participants in advance through a course website. No prior knowledge of propensity score methods or causal inference is needed. However, a basic understanding of research designs, t-tests, and logistic regression is preferable. Participants are encouraged to bring their own laptop computers for hands-on activities.

Sub Unit

• Professional Development and Training Committee

Instructors

• Wei Pan, Duke University

• Haiyan Bai, University of Central Florida

• Christopher M. Swoboda, University of Cincinnati