Criteria for selecting and reporting mediation effects sizes

Purpose This paper aims to provide guidance on applying, assessing and reporting mediation effect sizes within the context of Partial Least Squares Structural Equation Modeling (PLS-SEM) for social science researchers. Despite the growing popularity of PLS-SEM and mediation analysis in the social sciences in general and organizational research specifically, existing literature indicates inadequate reporting of mediation analysis results. Design/methodology/approach To address a methodology gap, the authors present a collection of guidelines for selecting and assessing mediation effect sizes to enhance the reporting of PLS-SEM mediation results. These guidelines are exemplified through an empirical application to illustrate how future studies can enhance the reporting of mediation models. Findings By aligning post hoc analysis of indirect and direct paths with research goals, researchers can achieve more nuanced insights into their mediation models. This research reviewed the types of mediation, goals of PLS-SEM models and mediation effect sizes to produce guidelines that allow researchers to determine which effect sizes are appropriate and useful to report. Originality/value To the best of the authors’ knowledge, this paper is the first to present a collection of mediation effect sizes in a PLS-SEM context to explain how research should apply effect sizes that align with both the scope and goal of the mediation model.