A Tutorial on Dyadic Growth Curve Mediation Modeling: An Applied Example Testing Pathways of Fathers' Depression Vulnerability to Destructive Marital Conflict.
Sarah Hoegler Dennis, Lee T Gettler, Lijuan Wang, Patrick Davies, E Mark Cummings
Abstract
Open AccessResearchers within and beyond the psychology of men and masculinities (PMM) have underscored the importance of identifying pathways underlying men's mental health in the context of dyadic relationships. Accordingly, there is an emerging need for advancement and accessible application of corresponding longitudinal quantitative methods. Although both Dyadic Growth Curve Modeling and Growth Curve Mediation Modeling have each been used increasingly throughout related fields such as developmental and family sciences, they have not yet been widely applied by researchers in PMM. Moreover, there is a lack of applied examples involving the integration of these two methods. Our goal is therefore to provide an accessible tutorial on longitudinal mediation analysis with dyadic data using latent growth curve modeling. Specifically, we present an illustration of the use of Dyadic Linear Growth Curve Mediation Modeling as an extension of the Dyadic Linear Growth Curve Model, using a real dataset and guided code in R and Mplus to test the mediated effects of destructive marital conflict on depression in father-mother dyads. Results of our example Dyadic Linear Growth Curve Mediation Model revealed that higher levels of destructive conflict predicted higher paternal depression at year 1, and predicted lower levels of fathers' and mothers' marital quality at year 1 and year 2. A significant mediation pathway was also identified, such that higher levels of destructive conflict predicted lower levels of fathers' marital quality at year 2, which in turn predicted higher depression in fathers at year 3.