Content area
Full text
Soc Indic Res (2008) 87:367392
DOI 10.1007/s11205-007-9143-1
Amery D. Wu Bruno D. Zumbo
Accepted: 14 May 2007 / Published online: 6 June 2007 Springer Science+Business Media B.V. 2007
Abstract Mediation and moderation are two theories for rening and understanding a causal relationship. Empirical investigation of mediators and moderators requires an integrated research design rather than the data analyses driven approach often seen in the literature. This paper described the conceptual foundation, research design, data analysis, as well as inferences involved in a mediation and/or moderation investigation in both experimental and non-experimental (i.e., correlational) contexts. The essential distinctions between the investigation of mediators and moderators were summarized and juxtaposed in an example of a causal relationship between test difculty and test anxiety. In addition, the more elaborate models, moderated mediation and mediated moderation, the use of structural equation models, and the problems with model misspecication were discussed conceptually.
Keywords Mediator Moderator Moderated mediation Mediated moderation Cause and effect Structural equation model Experimental design
The methodology of mediation and moderation is commonly used in social science, health, psychological, educational, and sociological research. The purpose of this paper is to provide an up-to-date review of the concepts, uses, and methodology of mediation and moderation. The examples, notation, and statistical demands are purposefully broad and easily accessible, involving several different disciplines, so as to capture as wide an audience of readers as possible. Likewise, we begin with conceptual matters and matters of research design so as to clear ground of many of these confusing foundational ideas before moving to the statistical ideas that build from these foundations.
Mediation and moderation are theories for rening and understanding a causal relationship. They, in essence, are researchers hypotheses about how a cause leads to an
A. D. Wu B. D. Zumbo (&)
Department of ECPS, University of British Columbia, Scarfe Building, 2125 Main Mall, Vancouver, BC, Canada V6T 1Z4e-mail: [email protected]
Understanding and Using Mediators and Moderators
123
368 A. D. Wu, B. D. Zumbo
effect. The investigation of mediation and moderation effects demands an integrated research plan from articulating the theoretical rationale, choosing a research design, analyzing the data, to drawing conclusions. Unfortunately, mediation and moderation effects have been widely misunderstood and misused as data analytical tools and statistical hypothesis testing with...