Relationship Quality and Changes in Depressive Symptoms among Urban, Married African Americans, Hispanics, and Whites
Fagan, Jay, Family Relations
Bivariate analyses showed that continuously married urban African American, non-Hispanic White, and Hispanic fathers and mothers reporting greater marital support and less relational control experienced a decrease in depressive symptoms. Multiple regression showed a stronger association between concurrent marital support and decreased depressive symptoms for mothers than fathers. African American and Hispanic fathers reporting higher levels of spousal relationship support when children were infants reported a larger decrease in depressive symptoms when children were age 3 compared to non-Hispanic Whites. Relationship control at age 3 was positively related to increased depressive symptoms among all groups except African American fathers and White mothers. African American and Hispanic fathers with marital problems may need additional support services.
Key Words: African American, depressive symptoms. Hispanic, marriage, relationship quality.
A growing body of research literature suggests that married couples experience fewer symptoms of depression, less stress, better health, and a higher sense of well-being than couples who are unmarried (e.g., Freeh & Williams, 2007). These findings are particularly important given the current trend for young couples to form cohabiting relationships or to remain single rather than marry, even following the birth of children (Seltzer, 2000). A significant gap in the literature on marriage and health is the lack of research on the association between these variables among lowincome individuals in different racial/ethnic groups (Whisman, 2001). Knowledge about the association between marriage and health across racial/ethnic groups is important for policy makers and practitioners involved in programs designed to enhance couple relationships. This study examined the association between fathers' and mothers' perceptions of relationship quality and changes in their depressive symptoms during the first several years following the birth of a child among low-income African American, Hispanic, and non-Hispanic White continuously married couples using the Fragile Families and Child Wellbeing (FFCW) data (McLanahan & Garfinkel, 2000). The transition to parenthood is a critical time because couples frequently experience a decline in marital relationship quality and are at heightened risk for increased depressive symptoms (Lawrence, Rothman, Cobb, Rothman, & Bradbury, 2008).
Background and Significance
There is ample evidence supporting the association between marital quality and depression in the general population. Whisman's (2001) meta-analysis of 26 cross-sectional studies examining the association between these variables found an effect size of -.42 for women and -.37 for men. Similar findings have been obtained from longitudinal studies of marital quality and subsequent depression (Beach, Katz, Kim, & Brody, 2003; Dehle & Weiss, 1998). However, many of the available studies lack ethnic diversity in their samples (Tucker & Crouter, 2008). Accordingly, Whisman suggested that different racial/ethnic groups may show different patterns of association between marital quality and depression, and factors that influence their association may also be culturally based (see also Marks et al., 2008). The present study focused on the two largest ethnic minority groups in the United States - African Americans and Hispanics. The marriage literature on these groups is reviewed in an effort to demonstrate that married African Americans with low-quality marital relationships are more likely than married Hispanics and non-Hispanic Whites to report increased depressive symptoms.
A significant shortcoming of studies on marital quality and depressive symptoms is that the vast majority are cross-sectional (Whisman, 2001). Longitudinal studies are better suited to disentangle the associations between marital problems and depressive symptoms because the temporal aspect of these variables can be addressed (Williams, 2003). …