1

Reconsidering Baron and Kenny:
Myths and Truths about Mediation Analysis
Xinshu Zhao*
Professor and Director
Center for Research in Journalism and Mass Communication
University of North Carolina and Chair Professor and Dean
School of Communication, Hong Kong Baptist University zhao@unc.edu John G. Lynch, Jr.*
Roy J. Bostock Professor of Marketing
Fuqua School of Business
Duke University
Durham, NC 27708
Tel: (919)-660-7766 john.lynch@duke.edu Qimei Chen*
Shidler Distinguished Professor
Chair/Associate Professor of Marketing
Shidler College of Business
University of Hawaii at Manoa
C303, 2404 Maile Way,
Honolulu, HI 96822
Tel: (808) 956-8921
Fax: (808) 956-9886 qimei@hawaii.edu Manuscript #08-0083-2, re-submitted to Journal of Consumer Research, June 2009
* The authors contributed equally to this article. This study was supported in part by a UNC-CH Research Council
Grant #3-12818, UNC-CH School of Journalism and Mass Communication Summer Grants for Research, 20012007, and grants from NICHD (R24 HD056670, Henderson PI) and UNC-CH Center for AIDS Research (#07-1191,
Brown PI). The authors also wish to thank James R. Bettman, Jane D. Brown, Gavan Fitzsimons, Rhonda Gibson,
Joe Bob Hester, Joel Huber, Laurence W. Jacobs, Chuanshu Ji, Wagner Kamakura, Gary McClelland, Carl Mela,
Andres Musalem, Jonathan Levav, Jason Roos, Woochoel Shin, Stephen Spiller, Rick Staelin, Ning Mena Wang,
William D. Wells, Stacy Wood, and seminar participants at the Duke University for their assistance and comments.
We thank Jon James for assistance with the Monte Carlo simulations. Any errors or omissions are the fault of the authors. 2

ABSTRACT
Baron and Kenny’s (1986) framework for mediation analysis has become a standard part of the consumer researcher’s toolkit: an independent variable X affects some mediator M that in turn affects some