A 2-Day Seminar Taught by Shenyang Guo, Ph.D.

Propensity score analysis is a relatively new and innovative class of statistical methods that has proven useful for evaluating the effects of treatments or interventions when using nonexperimental or observational data.

To reflect the most recent advances in the field, this Second Course of Propensity Score Analysis reviews the statistical principles and applications of important models recently developed. It illustrates how these models can be used for solving evaluation and casual-inference problems. With a strong focus on practical applications, the course explores various strategies for employing Propensity Score Analysis, discusses the use of Propensity Score Analysis with alternative types of data, and delineates the limitations of Propensity Score Analysis under a variety of constraints. This course will focus on the following topics:

• Overview of the origins, history, and statistical foundations of Propensity Score Analysis
• Handling multilevel data in estimating and using the propensity scores
• Addressing the issue of limited overlap of estimated propensity scores between treatment and control groups
• Conducting propensity score subclassifications
• Newly developed methods analyzing treatment dosage with generalized propensity scores
• Tests of heterogeneous treatment effects and modeling issues
• Running Propensity Score Analysis in conjunction with structural equation modeling