Mediation research, when combined with strong methods, is uniquely valuable in understanding theoretical pathways and thus in providing a sound basis for policy decisions pertaining to people management. However, our comprehensive review shows that current mediation research in human resources, organizational behavior, and other areas of management has serious deficiencies that frequently result in incomplete, or worse, invalid conclusions and policy recommendations. The most fundamental problem in mediation research is the prevalent exclusive focus on the indirect effect while ignoring the direct and/or total effect (97.8% in our sample of studies). This problem is most serious when mediation is inconsistent (i.e., the sign of the direct effect is opposite to that of the indirect effect), which we find is common (occurring in 26.4% of sampled mediation studies), but almost always ignored. Focusing only on the indirect effect is especially problematic in inconsistent mediation when the direct effect is larger than the indirect effect, resulting in the total effect and indirect effect having opposite signs. Even when mediation is consistent, a sole focus on the indirect effect fails to satisfy a fundamental effect size requirement unique to mediation research: the need to quantify the size of the indirect effect relative to the size of the total effect. We provide a set of recommendations to address these concerns and to improve mediation research going forward. One key recommendation is to report and interpret not only the indirect effect, but also the direct effect and total effect. A second key recommendation is to report PM, as well as a form of PM, Absolute PM, that has been largely overlooked, but which our analyses demonstrate has crucial advantages (smaller sampling variability, robustness to inconsistent mediation) over PM alone.
