A key objective of decomposition analysis is to identify risks or resources (‘mediators’) that contribute to disparities between groups of individuals defined by social characteristics such as race, ethnicity, gender, class, and sexual orientations. In decomposition analysis, a scholarly interest often centers on estimating how much the disparity (e.g., health disparities between Black women and White men) would be reduced/remain if we set the mediator (e.g., education) distribution of one social group equal to another. However, causally identifying disparity reduction and remaining depends on the no omitted mediator-outcome confounding assumption, which is not empirically testable. In this talk, we discuss a flexible way to 1) estimate disparity reduction and remaining and 2) assess the robustness of the estimates to the possible violation of no omitted mediator-outcome confounding. We apply the proposed methods to an empirical example, examining the contribution of education in reducing health disparities across race-gender groups. Our proposed methods are available as open-source software (‘causal.decomp’ R package)

Joint work with Suyeon Kang, Chioun Lee, and Shujie Ma

Dr. Soojin Park