Marginal Mean Weighting Through Stratification: A Generalized Method for Causal Inference

Session Submission Type: Professional Development Course

Abstract

The marginal mean weighting through stratification (MMW-S) method is a non-parametric causal inference strategy suitable for evaluating not only binary treatments but also multiple treatments measured on an ordinal or a nominal scale. It overcomes important limitations of other propensity score based methods. The objective of this course is to equip participants with preliminary knowledge and skills needed for applying the MMW-S method. The course is designed for evaluation researchers and graduate students. Participants will receive related readings, data examples, and SPSS, SAS, STATA, and R command files for analyzing the data. The instructors will also demonstrate in HLM how to apply the MMW-S method to multilevel educational data. Each participant is expected to bring a laptop.

Sub Unit

• Professional Development and Training Committee

Director

• Guanglei Hong, University of Chicago

Instructors

• Yihua Hong, RTI International

• Bing Yu, University of Chicago