Mediation Variables and Mediation Analysis: What and Why?

While there are various definitions for mediation variables, the one by Baron and Kenny (1986) is considered the most classic definition: “In general terms, a moderator is a qualitative (e.g., sex, race, class) or quantitative (e.g., level of reward) variable that affects the direction and/or strength of the relation between an independent or predictor variable and a dependent or criterion variable.” Mediation variables and their impact are everywhere. Using prevention studies as an example, smoking/cholesterol/blood pressure are the mediation variables when studying death due to myocardial infarction; and self-efficacy/enjoyment/knowledge of behaviors are the mediation variables when studying weekly physical activities. Clearly, ignore mediation variables could lead to misinterpretation of the outcome measures of a study. The mediation analysis is used to examine the relation between the predictor and the criterion variables, the relation between the predictor and the mediator variables, and the relation between the mediator and criterion variables. The relation between predictor and criterion should be reduced (to zero in the case of total mediation) after controlling the relation between the mediator and criterion variables. Once a moderator effect is founded, mediation analysis is used to explain the source of the effect. This presentation will provide a comprehensive review on mediation variables and mediation analysis, including their definitions, historical development, their relationships with similar variables (e.g., “covariate”), their importance in a research study, related research design and the latest development.