Keywords

data analysis, statistical, data interpretation, statistical, mediation, models, statistical

 

Authors

  1. Krause, Melanie R.
  2. Serlin, Ronald C.
  3. Ward, Sandra E.
  4. Rony, (Rachel) Yaffa Zisk
  5. Ezenwa, Miriam O.
  6. Naab, Florence

Abstract

Background: R. M. Baron and D. A. Kenny (1986) defined mediation and described how to perform statistical tests of mediation hypotheses. Their approach to testing mediation has been used extensively in the nursing literature. However, many statisticians have identified problems with the Baron and Kenny approach.

 

Purpose: The aim of this paper is to critically evaluate alternative approaches to testing mediation.

 

Approach: The Baron and Kenny approach and its shortcomings are briefly reviewed. A critical analysis of 17 alternate methods in three categories is then presented: (a)causal steps, (b) difference in coefficients, and (c) product of coefficients. The evaluation focuses on Type I error rate control, power, ease of computation, and versatility of use.

 

Results: Of the methods that control Type I error rate adequately, the joint significance test of [alpha] and [beta], the asymmetric distribution of products test, and the test of the products using the percentile bootstrap method are the most powerful tests of mediation. Of these three, the joint significance test of [alpha] and [beta] is superior due to its computational ease and versatility of use.

 

Discussion: Knowledge development in nursing will benefit from continued research testing mediation models. Nurse researchers could move beyond the Baron and Kenny approach to utilize more robust tests of mediation.