Review and Recommendations for Integrating Mediation and Moderation

Abstract

Usage of models integrating mediation and moderation is on the rise in the organizational sciences. While moderation and mediation are fairly well understood by themselves, additional complexities emerge when combining them. Some guidance exists regarding the empirical testing of such models, but this guidance is widely misunderstood. Furthermore, very little guidance exists regarding the theoretical justification of such models. This article offers a checklist of recommendations for the presentation, justification, and testing of models integrating mediation and moderation and compares these to what is actually being done via a review of empirical papers in top-tier journals.

Prenatal risk and physical aggression during the first years of life: The gender‐specific role of inhibitory control

“…First, it has been recommended by Holland et al. () to examine the specific pathway of the mediation model which would be moderated by gender. Therefore, hierarchical regression analysis was used to test whether gender specifically moderated the effect of prenatal risk on inhibitory control (path a) or the effect of inhibitory control on physical aggression (path b).…”

“…Age of the children at the data waves was also examined as potential covariate. The moderated mediation model was tested following the recommendations of Hayes (2015) and Holland, Shore, and Cortina (2017). The first hypothesis was tested using regression analysis to examine whether prenatal risk predicted children's physical aggression.…”

“…For the third hypothesis, two steps were carried out to test whether gender moderated the mediation model involving prenatal risk, inhibitory control, and physical aggression. First, it has been recommended by Holland et al (2017) to examine the specific pathway of the mediation model which would be moderated by gender. Therefore, hierarchical regression analysis was used to test whether gender specifically moderated the effect of prenatal risk on inhibitory control (path a) or the effect of inhibitory control on physical aggression (path b).…”

The Mediating Role of Sleep Quality and Quantity in the Link Between Food Insecurity and Obesity Across Race and Ethnicity

“…In this model, we first framed a series of hypotheses on individual linkages in a sequence to explore independent and direct associations. We then integrated mediation and moderation (). We assumed partial mediation via sleep quality and quantity, as other mediators of the relationship between food insecurity and obesity have been established ().…”

Restricted Variance Interaction Effects: What They Are and Why They Are Your Friends

“…All of the models discussed in Edwards and Lambert (2007) share the same X->M->Y core but vary in the location of the moderation. Holland et al (2017) found that the most common is the first stage (X->M) moderation model. Consider as an example a portion of the model presented in Zhang and Bartol (2010).…”