Retesting Intervention Effects in the Building Strong Families (BSF) Dataset

Abstract

There is a discrepancy between the emphasis on relational partners participating in couple relationship education (CRE) programs together (i.e., joint couple attendance; Arnold & Beelmann, 2019; Cowan & Cowan, 2019; Johnson, 2012; Markman et al., 2022) and the principles of family systems theory and the spillover hypothesis (Cox & Paley, 1997, 2003; Engfer, 1988; Erel & Burman, 1995). Specifically, the joint couple attendance emphasis of CRE programs implies that individuals participating in CRE without their relational partners may lead to detrimental couple and family functioning outcomes. In contrast, family systems theory and the spillover hypothesis suggest that individual and dyadic participation in CRE programs should have equivalent outcomes due to interactions and spillover between members of the family system. I explored this discrepancy with data from the Building Strong Families (BSF) project. The BSF CRE program provided relationship skills education focused on communication and conflict management, affection and intimacy, and the parent-infant relationship, and served a sample consisting of primarily racial-ethnic minority, lower income, unmarried couples transitioning to parenthood (Hershey et al., 2008). Using data from couples in the intervention condition of the BSF project (N = 2,553 couples) and conducting analyses via structural equation modeling and actor-partner interdependence modeling, I examined the associations between the difference in mothers’ and fathers’ BSF intervention dosage (i.e., an indication of whether and the extent to which individuals participated in the BSF CRE program without their relational partners) and outcomes critical for couple and family functioning.In the first paper (Chapter II), I investigated whether the difference in mothers’ and fathers’ BSF intervention dosage was associated with mothers’ and fathers’ destructive conflict 15 months later. Additionally, given the historical and sociocultural factors that have the potential to uniquely affect relationship and family functioning for Black individuals, I examined mothers’ and fathers’ race – comparing those identifying as Black to all others in the sample – as moderators of these associations. Results indicated that the difference in intervention dosage (i.e., mothers attending more of the intervention than fathers) was associated with greater mothers’ destructive conflict at 15 months. In contrast, the difference in intervention dosage was not associated with fathers’ destructive conflict at 15 months. Lastly, neither mothers’ nor fathers’ race moderated these associations. In the second paper (Chapter III), I evaluated whether the difference in mothers’ and fathers’ BSF intervention dosage was associated with mothers’ and fathers’ coparenting relationship quality 36 months later. Further, I tested mothers’ and fathers’ destructive conflict, which was measured at 15 months, as mediators of these associations (i.e., difference in intervention dosage → destructive conflict → coparenting). Results indicated that mothers attending more of the intervention than fathers was associated with worse coparenting at 36 months as reported by both mothers and fathers. These associations were mediated by mothers’ destructive conflict (i.e., mothers attending more of the intervention than fathers → greater mothers’ destructive conflict → worse coparenting). In contrast, fathers’ destructive conflict was not a significant mediator of the associations between the difference in mothers’ and fathers’ BSF intervention dosage and coparenting. In the third paper (Chapter IV), I assessed the difference in mothers’ and fathers’ BSF intervention dosage, the proportion of BSF intervention hours attended as a couple, and destructive conflict as predictors of mothers’ and fathers’ depressive symptoms at 36 months. Results indicated that the difference in mothers’ and fathers’ BSF intervention dosage was not associated with mothers’ or fathers’ depressive symptoms at 36 months. Similarly, the proportion of BSF intervention hours attended as a couple was not associated with mothers’ or fathers’ depressive symptoms at 36 months. Lastly, greater mothers’ and fathers’ destructive conflict at 15 months was associated with higher levels of their own depressive symptoms at 36 months, but was not associated with their partners’ depressive symptoms at 36 months. The findings across these three papers are mixed. Some results support the joint couple attendance emphasis of CRE programs as the difference in mothers’ and fathers’ BSF intervention dosage was associated with maladaptive outcomes (i.e., greater mothers’ destructive conflict, worse mothers’ coparenting, worse fathers’ coparenting). Conversely, some results align with family systems theory and the spillover hypothesis as the difference in mothers’ and fathers’ BSF intervention was not associated with other outcomes (i.e., fathers’ destructive conflict, mothers’ depressive symptoms, fathers’ depressive symptoms). Future research is needed to continue to elucidate the contexts in which, the participants for whom, and the outcomes for which the difference in relational partners’ CRE dosage matters. Nevertheless, the findings of this dissertation demonstrate the importance of examining outcomes of the process by which CRE is delivered or received (e.g., the difference in relational partners’ CRE program dosage). Moreover, the results provide guidance for future CRE programs designed to improve the couple relationship for unmarried parents, highlight the importance of using dyadic techniques to assess the effects of such programs, and suggest meaningful directions for future research.