Power Anomalies in Testing Mediation

Two rather surprising anomalies relating to statistical power occur in testing mediation. First, in a model with no direct effect for which the total effect and indirect effect are identical, the power for the test of the total effect can be dramatically smaller than the power for the test of the indirect effect. Second, when there is a direct effect of a causal variable on the outcome controlling for the mediator, the power of the test of the indirect effect is often considerably greater than the power of the test of the direct effect, even when the two are of the same magnitude. We try to explain the reasons for these anomalies and how they affect practice.

[1]  D. A. Kenny,et al.  Process Analysis , 1981 .

[2]  D. A. Kenny,et al.  Data Analysis in Social Psychology: Recent and Recurring Issues , 2010 .

[3]  Kristopher J Preacher,et al.  Effect size measures for mediation models: quantitative strategies for communicating indirect effects. , 2011, Psychological methods.

[4]  A. Hayes.  Beyond Baron and Kenny: Statistical Mediation Analysis in the New Millennium , 2009 .

[5]  David R. Cox,et al.  Regression Analysis When There is Prior Information About Supplementary Variables , 1960 .

[6]  D. A. Kenny,et al.  A reanalysis of a behavioral intervention to prevent incident HIV infections: Including indirect effects in modeling outcomes of Project EXPLORE , 2013, AIDS care.

[7]  John G Bullock,et al.  Yes, But What's the Mechanism? (Don't Expect an Easy Answer) , 2010, Journal of personality and social psychology.

[8]  Matthew S. Fritz,et al.  PSYCHOLOGICAL SCIENCE Research Article Required Sample Size to Detect the Mediated Effect , 2022 .

[9]  S. West,et al.  A comparison of methods to test mediation and other intervening variable effects. , 2002, Psychological methods.

[10]  D. A. Kenny,et al.  The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. , 1986, Journal of personality and social psychology.

[11]  Kristopher J Preacher,et al.  Mediation Analysis in Social Psychology: Current Practices and New Recommendations , 2011 .

[12]  P. Shrout,et al.  Mediation in experimental and nonexperimental studies: new procedures and recommendations. , 2002, Psychological methods.

[13]  Michael Scharkow,et al.  The Relative Trustworthiness of Inferential Tests of the Indirect Effect in Statistical Mediation Analysis , 2013, Psychological science.