Matching Methods for Observational and Experimental Causal Inference - Online Course

ABSTRACT

We will cover how to detect and ameliorate model dependence, where small, indefensible changes in model specification have large, perhaps unintended impacts on our conclusions. Easy-to-use matching methods for both observational and experimental data will be discussed.

In observational data, matching methods can greatly reduce the problem of model dependence and improve the credibility of statistical conclusions.  In experimental data (i.e., when randomization is possible), these methods can drastically reduce financial costs, statistical uncertainty, and the time necessary to complete a study.

Along the way, we will show that propensity score matching, an enormously popular approach, accomplishes the opposite of its intended goal — increasing imbalance, inefficiency, model dependence, and bias — and should be replaced with other matching methods in applications. Fortunately, other matching methods, such as coarsened exact matching, are far easier to use and understand and more powerful statistically.

Easy-to-use software, including CEM, MatchIt, and WhatIf, is available for all methods introduced.

This Distinguished Speaker Series seminar will consist of three hours of lecture and Q&A, held live* via the free video-conferencing software Zoom.

*The video recording of the seminar will be made available to registrants within 24 hours and will be accessible for four weeks thereafter. That means that you can watch all of the class content and discussion even if you cannot participate synchronously. 

Closed captioning is available for all live and recorded sessions. Live captions can be translated to a variety of languages including Spanish, Korean, and Italian. For more information, click here.