The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations.¶
Why this mattered¶
Baron and Kenny’s 1986 paper mattered because it gave social psychologists a shared vocabulary and statistical workflow for separating two questions that had often been blurred: when an effect changes and how an effect occurs. A moderator specified conditions under which a relation varied; a mediator specified an intervening process through which a predictor influenced an outcome. By making this distinction explicit, the paper shifted the field from asking only whether variables were associated to asking more theory-bearing questions about boundary conditions and mechanisms.
The paper also made mechanism-testing practically available to a broad empirical community. Its regression-based steps for testing mediation and interaction-based treatment of moderation became a common template for designing studies, analyzing data, and writing theoretical claims in social, personality, clinical, organizational, communication, and health psychology. Even when later methodologists criticized aspects of the “Baron and Kenny” mediation test, especially its low power and reliance on a sequence of significance tests, the paper had already established mediation and moderation as central analytic targets rather than optional interpretive add-ons.
Subsequent breakthroughs built directly on this framework. Later work replaced or refined the original procedures with path analysis, structural equation modeling, bootstrapped indirect effects, counterfactual causal mediation analysis, multilevel mediation, moderated mediation, and conditional process models. Those advances changed the statistical machinery, but they retained the conceptual architecture Baron and Kenny popularized: empirical theories should specify not only whether a cause matters, but through what process, for whom, and under what circumstances.
Abstract¶
(no abstract available)
Related¶
- enables → Psychological Safety and Learning Behavior in Work Teams — Baron and Kenny's moderator-mediator distinction enabled Edmondson to model psychological safety as an intervening team process affecting learning behavior.
- cite ← Psychological Safety and Learning Behavior in Work Teams — Edmondson cites Baron and Kenny's moderator-mediator distinction to frame how team psychological safety mediates or moderates learning behavior.