A Meditation on Mediation

Dawn Iacobucci, Wharton, US
Mediation is frequently of interest to social science researchers. A theoretical premise posits an intervening variable, M, some indicative measure of the process through which an independent variable, X, is thought to impact a dependent variable, Y. The researcher assesses the extent to which the effect of X on Y is direct or indirect, usually via regressions, per Baron and Kenny (JPSP, 1986). We begin by challenging the frequent use of mediation assessments. We then conduct a series of Monte Carlo simulation studies and show that the regression technique never surpasses structural equations modeling.
[ to cite ]:
Dawn Iacobucci (2007) ,"A Meditation on Mediation", in NA - Advances in Consumer Research Volume 34, eds. Gavan Fitzsimons and Vicki Morwitz, Duluth, MN : Association for Consumer Research, Pages: 536-538.