A Regression Framework for Causal Mediation Analysis with Applications to Behavioral Science

• Published in: Multivariate behavioral research Vol. 54; no. 4; pp. 555 - 577
• Main Authors: Saunders, Christina T., Blume, Jeffrey D.
• Format: Journal Article
• Language: English
• Published: United States Routledge 04.07.2019; Taylor & Francis Ltd
• Subjects: Algorithms; Behavioral Sciences; Causal modeling; Data Interpretation, Statistical; direct and indirect effects; Downloading; Humans; Mediation; Models, Statistical; Regression analysis; direct and indirect effects; Causal modeling; mediation
• ISSN: 0027-3171; 1532-7906
• DOI: 10.1080/00273171.2018.1552109

We introduce and extend the classical regression framework for conducting mediation analysis from the fit of only one model. Using the essential mediation components (EMCs) allows us to estimate causal mediation effects and their analytical variance. This single-equation approach reduces computation time and permits the use of a rich suite of regression tools that are not easily implemented on a system of three equations. Additionally, we extend this framework to non-nested mediation systems, provide a joint measure of mediation for complex mediation hypotheses, propose new visualizations for mediation effects, and explain why estimates of the total effect may differ depending on the approach used. Using data from social science studies, we also provide extensive illustrations of the usefulness of this framework and its advantages over traditional approaches to mediation analysis. The example data are freely available for download online and we include the R code necessary to reproduce our results.