Archived Content

The Office of Disease Prevention (ODP) archives materials that are more than 3 years old and no longer being updated. Over time, links and other information may have changed. We cannot guarantee that all of the links in these materials will be current or accurate.

Methods: Mind the Gap

Webinar Series

Connections Between Traditional and Causal Mediation Methods

June 18, 2020
Picture of David P. MacKinnon, Ph.D.
David P. MacKinnon, Ph.D.

Arizona State University

View the Webinar

About the Webinar

This presentation describes mediation analysis and the connections between traditional mediation analysis and recently developed causal mediation analysis. Mediating variables have a long and important history in theoretical and applied research because they describe how and why two variables are related. One common example of applied mediation research is the study of the mediating processes that explain how a prevention/treatment program achieves its effects on an outcome variable. If the intervention’s active ingredients are identified, the intervention can be made more powerful and more efficient. Other applied mediation examples include identifying how a risk factor leads to disease and how early life experiences affect later development. Important recent developments in causal mediation analysis include new counterfactual (potential outcomes) methods that generate accurate estimates for continuous and categorical measures. In general, researchers have been slow to adopt causal mediation methods because of their complexity and the perceived lack of connection between traditional and causal methods. However, understanding connections between traditional and causal mediation increases understanding of both methods. The background for each approach is described, along with questions about traditional mediation and potential outcomes that causal mediation perspectives can help answer. The presentation ends with future directions in mediation theory and statistical analysis. 

About David P. MacKinnon

Dr. David P. MacKinnon, Ph.D., has been developing, evaluating, and applying methods to assess how interventions work for over 30 years. He is a Foundation Professor of Psychology at Arizona State University. He received his undergraduate degree from Harvard University, and his Ph.D. in measurement and psychometrics from UCLA in 1986. In 2011, he received the Nan Tobler Award from the Society for Prevention Research for his book on statistical mediation analysis. Dr. MacKinnon has been Principal Investigator on several National Institutes of Health grants. He received the Method to Extend Research in Time (MERIT) Award from the National Institute on Drug Abuse for his research on mediation analysis. He has given numerous workshops in the United States and Europe, has served on federal grant review committees and as a consulting editor, and is a Web of Science Highly Cited Researcher. He is a Fellow of the Association for Psychological Science, Society for Prevention Research, and American Psychological Association Quantitative and Qualitative Methods Division. He is past president of American Psychological Association Division 5 and is president-elect of the Society of Multivariate Experimental Psychology.

Last updated on