Testing Moderator and Mediator Effects in Counseling Psychology Research Patricia A. Frazier University of Minnesota Andrew P. Tix Augsburg College Kenneth E. Barron James Madison University The goals of this article are to (a) describe differences between moderator and mediator effects; (b) provide nontechnical descriptions of how to examine each type of effect, including study design, analysis, and interpretation of results; (c) demonstrate how to analyze each type of effect; and (d) provide suggestions for further reading. The authors focus on the use of multiple regression because it is an accessible data-analytic technique contained in major statistical packages. When appropriate, they also note limitations of using regression to detect moderator and mediator effects and describe alternative procedures, particularly structural equation modeling. Finally, to illustrate areas of confusion in coun- seling psychology research, they review research testing moderation and mediation that was published in the Journal of Counseling Psychology during 2001. If you ask students or colleagues to describe the differences between moderator and mediator effects in counseling psychology research, their eyes are likely to glaze over. Confusion over the meaning of, and differences between, these terms is evident in counseling psychology research as well as research in other areas of psychology (Baron & Kenny, 1986; Holmbeck, 1997; James & Brett, 1984). This is unfortunate, because both types of effects hold much potential for furthering our understanding of a variety of psychological phenomena of interest to counseling psychologists. Given this, our goals here are to (a) describe differences be- tween moderator and mediator effects; (b) provide nontechnical, step-by-step descriptions of how to examine each type of effect, including issues related to study design, analysis, and interpreta- tion of results; (c) demonstrate how to analyze each type of effect through the use of detailed examples; and (d) provide suggestions and references for further reading. We focus on the use of multiple regression because it is an accessible data-analytic technique con- tained in major statistical packages that can be used to examine both moderator and mediator effects (Aiken & West, 1991; Baron & Kenny, 1986; Cohen, Cohen, West, & Aiken, 2003; Jaccard, Turrisi, & Wan, 1990). When appropriate, however, we also note limitations of using multiple regression to detect moderator and mediator effects and describe alternative procedures, particularly structural equation modeling (SEM). In addition, to illustrate areas of confusion in counseling psychology research, we review re- search testing moderation and mediation that was published in the Journal of Counseling Psychology (JCP) during 2001. Finally, we want to stress that, although our goal was to summarize informa- tion on current best practices in analyzing moderator and mediator effects, we strongly encourage readers to consult the primary sources we reference (and new sources as they emerge) to gain a better understanding of the issues involved in conducting such tests. DIFFERENCES BETWEEN MODERATOR AND MEDIATOR EFFECTS Consider, for a moment, your primary area of research interest. More than likely, whatever domain you identify includes research questions of the form “Does variable X predict or cause variable Y?” 1 Clearly, questions of this form are foundational to counseling psychology. Examples include correlational questions such as “What client factors are related to counseling outcomes?” as well as causal questions such as “Does a certain counseling intervention (e.g., cognitive therapy) increase well-being?” (see Figure 1A for a diagram). However, to advance counseling theory, research, and practice, it is important to move beyond these basic questions. One 1 For the sake of simplicity, we generally use the term predictor variable to refer to both a predictor variable in correlational research and an independent variable in experimental research. Likewise, we generally use the term outcome variable to refer to both an outcome variable in corre- lational research and a dependent variable in experimental research. Patricia A. Frazier, Department of Psychology, University of Minne- sota; Andrew P. Tix, Department of Psychology, Augsburg College; Ken- neth E. Barron, Department of Psychology, James Madison University. We thank Michele Kielty Briggs, Bryan Dik, Richard Lee, Heather Mortensen, Jason Steward, Ty Tashiro, and Missy West for their comments on an earlier version of this article and David Herring for his assistance with the simulated data. Correspondence concerning this article should be addressed to Patricia A. Frazier, Department of Psychology, University of Minnesota, N218 Elliott Hall, 75 East River Road, Minneapolis, MN 55455. E-mail: pfraz@umn.edu Journal of Counseling Psychology Copyright 2004 by the American Psychological Association, Inc. 2004, Vol. 51, No. 1, 115–134 0022-0167/04/$12.00 DOI: 10.1037/0022-0167.51.1.115 115 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.