Table of Contents

PART I: MEDIATION ANALYSIS
Chapter 1. Explanation and Mechanism

Chapter 2. Mediation: Introduction and Regression-Based Approaches

Chapter 3. Sensitivity Analysis for Mediation

Chapter 4. Mediation Analysis with Survival Data

Chapter 5. Multiple Mediators

Chapter 6. Mediation Analysis with Time-Varying Exposures and Mediators

Chapter 7. Selected Topics in Mediation Analysis

Chapter 8. Other Topics Related to Intermediates


PART II: INTERACTION ANALYSIS
Chapter 9. An Introduction to Interaction Analysis

Chapter 10. Mechanistic Interaction

Chapter 11. Bias Analysis for Interactions

Chapter 12. Interaction in Genetics: Independence and Boosting Power

Chapter 13. Power and Sample-Size Calculations for Interaction Analysis


PART III: SYNTHESIS AND SPILLOVER EFFECTS
Chapter 14. A Unification of Mediation and Interaction

Chapter 15. Social Interactions and Spillover Effects

Chapter 16. Mediation and Interaction: Future and Context

Appendix. Technical Details and Proofs
References

About the Author

Tyler J. VanderWeele, Ph.D., is a methodologist at Harvard University. He holds degrees in biostatistics, mathematics, finance, philosophy and theology and is currently Professor of Epidemiology in the Departments of Epidemiology and Biostatistics at the Harvard School of Public Health and a faculty affiliate of the Institute of Quantitative Social Science at Harvard University. His empirical research has been in epidemiology, various fields within
the social sciences, and the study of religion and health.

Reviews

"VanderWeele's new book is a landmark achievement in the emerging field of causal inference, pulling together material scattered across dozens of technical journals into a coherent whole. The book blends intuitive explanation, rigor, and scholarship with a clear and friendly writing style, making it broadly accessible and a pleasure to read. It and its subsequent editions should become a standard teaching and reference work for decades to come." --Sander
Greenland, Professor of Epidemiology and Statistics, UCLA
"Yes, mediation is an important topic. It has longed been used in the social sciences especially psychology. Of late there has been interest in many different fields including economics, sociology, epidemiology, political science and education, among other fields. Tyler VanderWeele is very qualified to author this book. He has contributed important work to the development of this topic and is a talented and careful researcherEL I think there is potential for
adoption in graduate courses in the social and biomedical sciences. I also think it could be widely purchased by applied researchers as a reference. I recommend publication." --Luke Keele, Associate
Professor, Department of Political Science, Penn State University
"Mediation is about understanding pathways between a treatment and an outcome that lead to the outcome, i.e., mechanisms. Mechanisms are a central thing in science and statisticians have been providing new principled methods for studying these topics over especially the last 10 years. Especially in the social and behavioral sciences and in epidemiology there has been great interest in these methods, and the methodology the author wants to write about is the new
stuff from the last 10 years [VanderWeele] is the key player in statistical literature these days. He's a good communicatorEL Primary market: applied researchers doing mediation in epidemiology,
social and behavioral sciences. Secondary market: applied statisticians teaching causal inference and/or working in the area. Yes, I think this might get some adoptions, and as the potential outcomes framework becomes more established in disciplines such as epidemiology and psychology, more adoptions."-Michael Sobel, Dept Sociology, Columbia
"Vanderweele has written the first book-length treatment of mediation and interaction (née moderation) using the causal-inference approach. Regardless of discipline, anyone who has a serious interest in the application of statistical methods to estimate, test, and probe causal processes will greatly benefit from this book." --David A. Kenny, Distinguished Board of Trustees and Alumni Professor, University of Connecticut
"Mediation analysis has been a highly active area of research in recent years, and VanderWeele arguably has been the most important contributor to this research. His book provides an accessible but comprehensive presentation of mediation, providing an introduction to both past and cutting-edge research. A must-read. An instant classic." --Christopher Winship, Diker-Tishman Professor of Sociology, Harvard University
"This outstanding book fills a void by describing new methods based on the counterfactual framework. VanderWeele provides a unified approach to several difficult aspects of mediation analysis." --David Mackinnon, Professor of Psychology, Arizona State University
"Have you ever tried to estimate how or why a cause leads to an effect, using methods you learned more than a few years ago? Want to know what you did wrong and how to fix it? Then I recommend you read this important and unusually accessible report from the cutting-edge of statistical science." --Gary King, Albert J. Weatherhead University Professor and Director of the Institute for Quantitative Social Science, Harvard University