Xinshu Zhao, Mary Junck Research Colloquium

Mary Junck Research Colloquium
UNC School of Journalism and Mass Communication
Feb. 24, 2010

Xinshu Zhao
Professor
School of Journalism and Mass Communication
UNC - Chapel Hill

Dean and Chair Professor
School of Communication
Hong Kong Baptist University

"Reconsidering Baron and Kenny: Myths and Truths about Mediation Analysis"

Baron and Kennys (1986) procedures for determining if an independent variable affects a dependent variable through some mediator are so well known that they are used by authors and requested by reviewers almost reflexively.

Category:

Education

Tags:

• Mary Junck Research Colloquium
• Xinshu Zhao

[vgx2008]

I'm not sure...without reading more about his article, it seems to me his examples deal with situations where a fourth hidden factor is introduced between X and Y (e.g., X'). It is likely that c' is not significant due to the existence of X', when M remains a viable mediator between X and Y (i.e., a x b is significant). If so, B & K mediation tests deal with the effect of X on X’, not Y. In that sense, the assumption of a significant c'' (X --> X') still must hold.

[ellomellowify]

Very insightful!