Longitudinal and Bidirectional Relations between Coparenting Relationship Quality and Father Engagement Among Diverse Low-Income Families

Background and Purpose: Few studies of low-income, unmarried parents of young children have examined the longitudinal and bidirectional relations between coparenting relationship quality and father involvement. Further, available studies show mixed findings, with some demonstrating a unidirectional relation (Jia & Schoppe-Sullivan, 2011) and others showing a bidirectional relation (Fagan & Cabrera, 2012) or no relation (Fagan & Palkovitz, 2011) between coparenting relationship quality and father engagement. Thus, we examined whether low-income parents’ coparenting relationship quality and father engagement share a bidirectional relation across the first three years of their child’s life. Based on the Family System Theory (Cox, Paley, & Harter, 2001), we hypothesized that coparenting relationship quality and father engagement would demonstrate longitudinally bidirectional relations where early coparenting relationship quality predicts later father engagement and early father engagement predicts later coparenting relationship quality.

Methods:

Data and sample: Data are from Building Strong Families (BSF), a study of over 5,000 low-income, unmarried couples who enrolled in a relationship strengthening program at or near the birth of their child. Subsequent reports indicated that the BSF program had no effects on partner relationship quality, coparenting, and father involvement (Clarkwest, Killewald, & Wood, 2015; Wood, Moore, Clarkwest, & Killewald, 2014). Our analytic sample consisted of 2,098 families whose fathers were consistently residential all or most of the time. Most fathers were Black (46.3%) or Hispanic (18.9%), had a high school diploma (61.9%), and were working for pay (79.6%).

Measures: The Parenting Alliance Inventory (PAI; Abidin & Brunner, 1995) measured positive aspects of coparenting relationship quality. We used both fathers’ and mothers’ reports of coparenting relationship quality when the child was 15 months (T1) and 36 months old (T2). Father engagement was measured with items assessing fathers’ self reports of engagement in caregiving (e.g., helped child get dressed) and play activities (e.g., played with games or toys) with the child in the past month.

Analysis: We conducted a cross-lagged analysis using latent variables of couple level coparenting relationship quality and father engagement at T1 and T2. analysis controlled for key sociodemographic factors, multiple-partner fertility, and intimate partner violence.

Results: Cross-lagged analysis demonstrated that the unidirectional coparenting relationship model best fit the data, χ²(1813) = 3836.424, p < .001, RMSEA = .023, 90% CIs [.022, .024], CFI = .96, TLI = .95, SRMR = .037. Only the path going from coparenting to father engagement in play activities was statistically significant (β = .18, p < .05). Couple level coparenting relationship quality predicted low-income residential fathers’ increased engagement in cognitively and socially stimulating play activities, but not caregiving. Neither father engagement in caregiving nor play activities predicted changes in coparenting relationship quality over time.

Conclusions and Implications: Results provide support for a unidirectional instead of a bidirectional hypothesis of the Family Systems Theory. Findings suggest that targeting low-income parents’ coparenting relationship may be a particularly effective intervention strategy for promoting low-income residential fathers’ ongoing engagement in cognitively and socially stimulating play activities—which are linked with positive child outcomes (Milteer & Ginsburg, 2012)—with their children.