Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2012 Nov 12.
Published in final edited form as: J Consult Clin Psychol. 2011 Oct 17;79(6):814–825. doi: 10.1037/a0025874

Predicting Relationship Stability Among Midlife African American Couples

Carolyn E Cutrona 1, Daniel W Russell 2, Rebecca G Burzette 3, Kristin A Wesner 4, Chalandra M Bryant 5
PMCID: PMC3495553  NIHMSID: NIHMS417054  PMID: 22004304

Abstract

Objective

This study examined predictors of relationship stability over 5 years among heterosexual cohabiting and married African American couples raising an elementary-school-age child. The vulnerability–stress–adaptation model of relationships (Karney & Bradbury, 1995) guided the investigation. Contextual variables were conceptualized as important determinants of education and income, which in turn influence family structure, stress, and relationship quality and stability. Religiosity was tested as a resource variable that enhances relationship stability.

Method

Couples (N = 207) were drawn from the Family and Community Health Study. Variables assessed at Wave 1 (education, income, religiosity, biological vs. stepfamily status, marital status, financial strain, and relationship quality) were used to predict relationship stability 5 years later.

Results

Higher levels of education were associated with higher income, lower financial strain, and family structures that research has shown to be more stable (marriage rather than cohabitation and biological-family rather than stepfamily status; Bumpass & Lu, 2000). These variables, in turn, influenced relationship quality and stability. Religiosity, an important resource in the lives of African Americans, promoted relationship stability through its association with marriage, biological-family status, and women’s relationship quality.

Conclusions

Enhancing the stability of African American couples’ relationships will require changes in societal conditions that limit opportunities for education and income and weaken relationship bonds. Programs to assist couples with blended families are needed, and incorporation of spirituality into culturally sensitive relationship interventions for African American couples may also prove beneficial.

Keywords: African Americans, couples, cohabitating, marriage, relationship quality

Relatively little is known about factors that contribute to relationship satisfaction and stability among African American couples. Most studies of African American couples have been comparative across racial and ethnic groups. Notably lacking are studies examining predictors of relationship stability within samples of African Americans (Bryant et al., 2010). An understanding of factors that promote relationship stability among African American couples is important for guiding policies and intervention strategies to avoid the pain of dissolution for adults and to maximize the continuity of child-rearing contexts for African American children (Bumpass & Lu, 2000).

Approximately 40% of cohabitating couples have children in the home; among African Americans, this figure is approximately 54% (Simmons & O’Connell, 2003). Thus, cohabitation has become a particularly important context for child rearing among African American families. The current study investigated predictors of relationship stability among heterosexual married and cohabiting African American couples with at least one elementary-school-age child in the home.

The vulnerability–stress–adaptation model of marriage (Karney & Bradbury, 1995) posits three major influences on relationship quality and stability: enduring vulnerabilities, stressful events, and adaptive processes. Enduring vulnerabilities are characteristics (e.g., low education level) that influence people’s susceptibility to stressful life events and their ability to interact effectively in their relationships. Stressful events comprise the challenges of the current context in which couples live their daily lives, such as unemployment or steep medical bills. Enduring vulnerabilities multiply the impact of adverse events in daily life because the resources to cope effectively are not available. Adaptive processes are the ways that individuals function in their relationships, including, for example, their approaches to resolving conflict (Karney & Bradbury, 1995). These processes emerge in the daily interactions of couples and influence how successfully they deal with the stressors and contextual strains of their lives.

Of particular interest in the current study was the cascading influence of enduring vulnerability factors. In our previous work, we showed a significant link between neighborhood-level economic disadvantage, couples’ financial strain, and marital quality (Cutrona et al., 2003). In our broad theoretical framework, we emphasize the impact of societal and community contexts on relationships and health (Cutrona et al., 2003, 2005). In the current study, we did not explicitly include community characteristics in our model but examined the impact on relationship stability of education and income, which are closely tied to neighborhood economic disadvantage (Vartanian & Buck, 2005). We examined the associations over time of education and income on (a) family structure, (b) financial strain (a current stressor), (c) relationship quality (adaptive processes), and (d) relationship stability. We also investigated the potential protective effects of religiosity, an enduring resource of particular importance to African American families (Shorter-Gooden, 2004). The model we tested is shown in Figure 1.

Figure 1.

Figure 1

Open in a new tab

Theoretical model.

Enduring Vulnerability Factors

In our model, limited education provides the context for a range of conditions that weaken relationships. A low level of education renders individuals vulnerable to stress through its association with unemployment, low pay, and limited opportunities to accrue resources (Day & Newburger, 2002). Individuals with limited financial resources experience more negative life events (Thoits, 2010) and subjective financial strain, the perception that income is inadequate to meet one’s needs (Conger et al., 1990). The effects of subjective financial strain are different and often greater than those of objective income level (Conger et al., 1990). Financial strain imposes a substantial burden on couples, which leads to tension and instability in their relationships (Bodenmann et al., 2007; Conger & Donellen, 2007). Conger’s model points to emotional distress as a key mediator between financial strain and deterioration in the quality of marital interactions (Conger et al., 1990). Evidence in support of this model was recently reported for African American families (Conger et al., 2002). We predicted that financial strain would decrease relationship stability through the mediation of relationship quality.

Education and income are associated with specific relationship types. Lower income predicts cohabitation rather than marriage (Bumpass & Lu, 2000). There is growing evidence that among low-income individuals, marriage is viewed as something that should be postponed until economic security has been achieved (Gibson-Davis, Edin, & McLanahan, 2005). Cohabitation does not have the same economic requirements. Thus, we predicted that lower income would predict cohabitation rather than marriage. Studies have shown differences in the relationship quality of married versus cohabiting couples, including less frequent disagreements, greater perceived fairness and supportiveness, and higher relationship satisfaction among married couples (Brown, 2003; Stafford, Kline, & Rankin, 2004). Research has further shown that cohabiting relationships are significantly less stable than marital relationships (Bumpass & Lu, 2000), although the stability difference is smaller among African Americans than European Americans (Osborne, Manning, & Smock, 2007). We hypothesized that marriage would predict higher relationship quality, which in turn, would predict higher relationship stability.

Both education and employment status predict biological-father presence in the home (Futris, Nielson, & Olmstead, 2010). The role of breadwinner is highly valued for African American men, despite the challenges faced by these individuals in communities that lack educational and economic opportunities (Futris et al., 2010). African American fathers who lack financial resources may view themselves as failures and withdraw from their family (Toth & Xu, 1999). Low-income mothers are more likely to remain in a relationship with their child’s father if he is employed than if he is unemployed (Futris et al., 2010). Thus, we predicted that education and income would predict the presence of the biological father and mother in the home.

Evolutionary psychology suggests another reason that couples who head blended families may have less stable relationships. According to the theory, nonbiological parents have less commitment to raising their partner’s offspring and may be more willing to leave a relationship in which they do not share biological offspring (Buss & Kenrick, 1988). Thus, we predicted a direct effect of biological family status on stability. We also predicted an indirect effect through relationship quality. Stepfamilies face a unique set of stressors. When children from previous unions are in the home, family members must deal with the complexities of ambiguous roles, financial obligations to multiple families, and juggling relationships with nonresident biological parents and children (Sweeney, 2010). We hypothesized that such stressors would erode relationship quality and, hence, stability.

Resources That Promote Stability

Marital status and biological-family status may be combined in various configurations. Couples may be married or cohabiting, and both, one, or none of the index child’s biological parents may be in the home (e.g., the child may be raised by grandparents, adoptive parents, or foster parents). In addition to the separate associations of marital and biological-family status with stability, we examined their joint associations in an attempt to understand whether any specific family structures are most strongly associated with relationship stability.

Religion and spirituality are central coping strategies for African Americans, especially women (Shorter-Gooden, 2004). Compared to other racial and ethnic groups, African Americans more frequently employ religious and spiritual coping when confronting adversity (Mattis, 2002) and derive a sense of community, shared values, and the strength to accept adversity (Brodsky, 2000). Religiosity and church attendance are associated with attitudes favorable to commitment and family stability (Wilcox & Wolfinger, 2007). We predicted that religious involvement, which we construed as an enduring resource variable, would be associated with marriage rather than cohabitation and, through the mediation of marriage, with relationship stability. We similarly predicted that religious involvement would be associated with biological-family status rather than stepfamily status.

The third component of the vulnerability–stress–adaption model highlights the ways that individuals function in their relationships. African Americans are more egalitarian and critical of traditional gender roles than are European Americans (Broman, 2005; Hunter & Sellers, 1998). In one study, husbands’ participation in household responsibilities and child care was more important to African American wives than to European American wives (Orbuch & Eyster, 1997) and served as a protective factor against divorce (Orbuch, Veroff, Hassan, & Horrocks, 2002). In the same study, low conflict was more strongly related to happiness among African American than European American couples (Veroff, Douvan, Orbuch, & Acitelli, 1998), and fighting and arguing were rated as a good reason for divorce more often by African American than European American couples (Orbuch, Veroff, & Hunter, 1999). Partner supportiveness and warmth are also strong predictors of the relationship stability of African American couples (Orbuch et al., 2002). Consequently, egalitarianism, low conflict, and social support were identified as especially important to relationship outcomes among African Americans. Our assessment of relationship quality included multiple components: satisfaction with partner contributions to domestic responsibilities, overall relationship satisfaction, conflict, and support. We predicted that higher relationship quality would predict greater stability.

To summarize, we tested a theoretical model of relationship stability in African American couples with children, in which lack of education and financial resources render couples more vulnerable to dissolution through multiple pathways (see Figure 1). We predicted that lack of education and financial resources would be associated with financial strain, which would influence stability primarily through its erosive influence on relationship quality. Furthermore, individuals with lower levels of education and income are more likely to live in family configurations that research has shown to be less stable (cohabiting rather than married and stepparent rather than biological-parent families), which contribute to dissolution both directly and indirectly through lower relationship quality. By contrast, those with the enduring resource factor of religious involvement are expected to have more stable relationships because religious individuals are likely to be committed to marriage and to be in biological families rather than stepfamilies.

Method

Participants

Data are from the Family and Community Health Study (FACHS), a longitudinal investigation of resiliency and vulnerability factors in rural African American families (Cutrona, Russell, Hessling, Brown, & Murry, 2000). All procedures were approved by the Iowa State University and University of Georgia Institutional Review Boards. Written informed consent was obtained from all participants.

All families in FACHS had a 9- to 11-year-old target child at the beginning of the study. We recruited families from census tracts identified in both Iowa and Georgia that were outside major metropolitan areas, in which the proportion of African American families was sufficiently high (10% or greater) to make recruitment feasible, and that spanned a wide range of income levels. We completed initial interviews with 71% of the eligible families whom we were able to locate, for an initial sample of 889 families.

In the current study we included couples who were married or cohabiting at Wave 1 (1997–1998) in which both partners participated in the first wave of data collection (N = 242; 200 married and 42 cohabitating). We narrowed the sample to include individuals who also participated in the third wave of data collection, 5 years after the initial interview (N = 207; 171 married and 36 cohabiting). The mean age was 37.3 years (SD = 7.2) among the women and 39.8 years (SD = 8.9) among the men at Wave 1. The median household income in 1997–1998 was $42,000 (SD = $25,440) for married and $30,000 (SD = $17,613) for cohabiting couples. The very large standard deviations for income reflect our deliberate sampling of participants from low, moderate, and high income neighborhoods. Eight percent of married couples and 23% of cohabiting couples were below the poverty line. The mean number of minor children residing in the home was 2.66 (SD = 1.3). The couples who participated in both Waves 1 and 3 did not differ from those who did not on any of the study variables.

Measures

All predictor variables were administered at Wave 1. The primary outcome variable was relationship stability (dissolution = 0, intact = 1), which was assessed during the Wave 3 interview, 5 years after the Wave 1 interview.

Enduring vulnerability and resiliency factors

We assessed number of years of education at Wave 1 for each partner, which served as separate manifest predictor variables in our model (his education and her education). Measures of income were based on each partner’s reports of his or her own personal income in thousands of dollars over the previous year from all sources (e.g., wages, self-owned business proceeds, child support, government programs); these two measures were also specified as separate manifest predictor variables (his income and her income) in the model. It should be noted that we had complete data on all measures except annual per capita income for the 207 couples. Seventy-three of the men and 80 of the women did not report their annual income (discussed further below).

In the domain of family structure, marriage at Wave 1 was coded such that married = 1 and cohabiting (“living with someone in a steady, marriage-like relationship”) = 0. Biological-family versus step-family status was determined by the nature of the relationship of the male and female partners to the target 9- to 11-year-old child (i.e., both biological parents or not, scored as 1 or 0, respectively). Eighty-three percent were married at the time of the first wave of interviews, and both biological parents of the target child were present in 55% of the families. In all cases, when the biological father was in the household, the biological mother was also present (n = 113; 106 married and 7 cohabiting). In addition, there were 74 biological mother–stepfather couples (45 married and 29 cohabiting). In a final group of 20 couples, neither individual was the biological parent of the target child (i.e., 9 grandparents; 1 aunt and uncle; 10 foster or adoptive parents). Although the variety of family configurations complicated analyses, we retained all eligible couples in the analyses because they represent the complexity of modern child-rearing families in the United States (Sweeney, 2010).

A latent religiosity variable for each partner was assessed with a seven-item measure (Simons, 1997) that was created for this study; validity of the scale was demonstrated by significant correlations with the Multidimensional Measure of Religious Involvement for African Americans (Levin, Taylor, & Chatters, 1995). Participants responded to each item on a 1-to-5 response scale (α = .77 and .81 for women and men, respectively). Two items assessed spirituality (the importance of religious or spiritual beliefs in daily life), and five items assessed frequency of church attendance and involvement in church activities. Three parcel measures were formed from the seven-item measure and used to operationalize the latent religiosity variable for each member of the couple (see discussion by Russell, Kahn, Spoth, & Altmaier, 1998).

Financial strain

Four measures developed by Conger and Elder (1994) were administered to both members of the couple and used to form a latent perceived financial strain variable for each partner. The first scale, Unmet Financial Needs, taps specific needs that cannot be met due to financial hardship (e.g., “Not enough money to buy the food we need”; α = .75 and .87 for women and men, respectively). The second scale, Can’t Make Ends Meet, consists of two items that tap the general perception that financial resources are insufficient (e.g., “During the last 12 months, how much difficulty have you had paying your bills?”; α = .70 and .69 for women and men, respectively). Financial Adjustments is an 11-item scale that taps specific ways the family has tried to economize to lessen their financial problems (e.g., “Reduced or eliminated medical insurance because of financial need”; α = .77 and .82 for women and men, respectively). Negative Financial Life Events is a checklist of 11 specific negative financial events in the previous 12 months (e.g., “cut in wages”; “laid off from work”). The four scales were used to specify a latent financial strain variable for each member of the couple.

Adaptive processes

A latent relationship quality variable was created for each partner using four measured indicators: relationship satisfaction, satisfaction with partner’s contributions to household and child care, partner warmth/support, and partner hostility. Relationship satisfaction was assessed with two items (“How happy [satisfied] are you with your relationship?”). Coefficient alpha was .84 for women and .81 for men. Level of satisfaction with the partner’s contributions to household responsibilities was assessed with two items: “How satisfied are you with [partner name]’s contribution to completing household chores, such as doing the laundry, cleaning, preparing meals, and so on?” and “How satisfied are you with [partner’s] contributions to raising [target child]?” Response options ranged from 1 (not at all satisfied) to 4 (very satisfied). These items were averaged to create a measure of satisfaction with partner contributions to household responsibilities. Coefficient alpha was .57 for women and .41 for men. The low reliability of this measure was due to the fact that the two items correlated only .23 for women and .17 for men. Apparently, satisfaction with housework and with child-care contributions may be quite different. Additional analyses were conducted treating these as individual measured variables in specifying the latent relationship quality variables; results were the same as those when using the composite measure.

Partner warmth/support and hostility were assessed with scales developed by Conger and Elder (1994). Each member of the couple reported perceptions of his or her partner’s warmth/support (e.g., “During the past 12 months, how often did your partner let you know he/she really cares about you?”) and hostility (e.g., “During the past 12 months, how often did your partner get angry at you?”). Response options ranged from 1 (never) to 4 (always). Coefficient alpha for Partner Warmth (9 items) was .91 for women and .90 for men. For Partner Hostility (12 items) alpha was .82 for women and .87 for men. Items for Warmth and Hostility served as the third and fourth indicators of the relationship quality variables.

Data Analysis Plan

The first step in testing the model shown in Figure 1 was to evaluate the measurement model. The analyses were conducted based on the covariances among the variables. Three of the variables in the model (religiosity, financial strain, and relationship quality) were specified as latent variables. The remaining variables in the model (education, income, biological family status, marital status, and relationship stability) were specified as measured or manifest variables. Most of the measures (education, income, religiosity, financial strain, and relationship quality) were specified as separate variables for each member of the couple; however, three variables (biological family status, marital status, and relationship stability) were characteristics of the couple. All were dichotomous and were treated as categorical measures in the analysis.

The measurement model was tested with the maximum likelihood estimation procedure in the Mplus 6.11 program. It should be noted that both cases with and without data for income were included in the analyses; the expectation-maximization (EM) algorithm was employed by Mplus to estimate model parameters involving the income variable (see discussion by Enders, 2010). Loadings for the measured variables on the latent religiosity, financial strain, and relationship quality variables were constrained to equality for women and men to ensure that these constructs were defined identically for both members of the couple. Also included in the model were correlated error terms for the measures that were completed by both partners to account for possible nonindependence in the data. Because the model implied mediation effects (e.g., the pathway from financial strain to relationship stability through relationship quality), we tested for hypothesized indirect effects. Finally, we conducted follow-up analyses of specific subgroups of couples to compare their relationship quality and stability.

Results

Characteristics of the Sample

Seventy-five percent of the couples remained intact over the 5-year period. Table 1 presents comparisons of men and women on all variables. Men reported significantly higher income, and women reported significantly higher levels of religiosity. Men generally perceived the relationship more positively than did the women, reporting significantly higher levels of relationship satisfaction, satisfaction with partner contributions, and partner warmth. Surprisingly, they also reported significantly higher levels of partner hostility than the women.

Table 1.

Comparisons of Men and Women on Study Variables

Variable Women
Men
95% CI of the difference
t
M SD M SD LL UL
Education 13.18 2.07 12.95 2.24 −.057 .511 1.57
Income $19,819 $11,823 $28,410 $15,706 −$12,369 −$4,812 −4.52***
Religiosity 2.60 0.54 2.47 0.63 .035 .206 2.76**
Financial strain −0.00 0.74 0.00 0.76 −.099 .099 0.00
Relationship satisfaction 4.84 0.86 5.01 0.87 −.307 −.050 −2.70**
Satisfaction with partner contribution 3.35 0.65 3.79 0.36 −.545 −.349 −9.10***
Partner warmth 3.08 0.62 3.21 0.63 −.228 −.045 −2.94**
Partner hostility 1.46 0.33 1.56 0.45 −.163 −.040 −3.27**
Open in a new tab

Note. df = 206. CI = confidence interval; LL = lower limit; UL = upper limit.

**

p < .01.

***

p < .001.

Testing the Causal Model

Results indicated that the measurement model provided an adequate fit to the data, χ2(303, N = 207) = 370.36, p = .005, comparative fit index (CFI) = .94, root-mean-square error of approximation (RMSEA) = .033. Loadings of the measured variables on the latent variables are shown in Table 2; all of the loadings were highly significant. The correlations among the measured and latent variables included in the model are presented in Table 3. Relationships among these variables were generally consistent with predictions. As expected, being married and the presence of both biological parents correlated positively and financial strain correlated negatively with stability. A few gender differences emerged. Only her relationship quality and religiosity and his education and income were significantly associated with relationship stability. Unexpected findings included a statistically significant negative relationship between the presence of both biological parents and the woman’s relationship quality.

Table 2.

Loadings of the Measured Variables on the Latent Variables

Variable His religion Her religion His financial strain Her financial strain His relationship quality Her relationship quality
Religion Parcel 1 .868 .896
Religion Parcel 2 .665 .629
Religion Parcel 3 .776 .720
Unmet Financial Needs .655 .711
Can’t Make Ends Meet .805 .675
Financial Adjustments .582 .547
Negative Financial Events .593 .556
Relationship satisfaction .814 .831
Role performance .646 .358
Partner warmth .653 .680
Partner hostility −.541 −.744
Open in a new tab

Note. The standardized loadings of the measured variables on the latent variables are shown; the nonstandardized loadings of each measured variable on the latent variable were constrained to be equal for men and women.

Table 3.

Correlations Among the Measured and Latent Variables

Variable 1 2 3 4 5 6 7 8 9 10 11 12
1. Her education —
2. His education .54* —
3. Her income .669* .38* —
4. His income .34* .51* .213 —
5. Her religion .25* .20* .15 .26* —
6. His religion .14* .23* .08 .30* .53* —
7. Biological family .24* .20* .13 .21* .16 .02 —
8. Her financial strain −.30* −.24* −.51* −.45* −.08 −.12 −.12 —
9. His financial strain −.25* −.27* −.37* −.58* −.19* −.24* −.16 .65* —
10. Married .37* .36* .15 .57* .55* .29* .58* −.28* −.23* —
11. Her relationship quality .02 .00 −.12 .05 .05 .18* −.19* −.37* 3−.26* .18 —
12. His relationship quality −.03 −.06 .04 .05 .17* .24* −.08 −.22* −.29* .21* .68* —
Relationship stability .17 .22* .10 .34* .29* .15 .62* −.32* −.28* .50* .26* .18
Open in a new tab
*

p < .05, two-tailed.

The next set of analyses evaluated the fit of the hypothesized structural equation model shown in Figure 1 to the data. This model was also found to provide an adequate fit to the data, χ2(347, N = 207) = 428.51, p = .002, CFI = .93, RMSEA = .034. Standardized coefficients for the paths included in the model are shown in Figure 2. As expected, there was a strong relationship between education and income; education accounted for 40% of the variation in his income and 44% of the variation in her income. Also as expected, religiosity was significantly related to marital status; however, only her religiosity attained significance. The opposite was true for the impact of income on marital status; only his income was significant. In combination, religiosity and income accounted for 40% of the variation in marital status. His income was a significant positive predictor of biological-family status, whereas her income was not. Her religiosity was also a significant positive predictor of biological-family status. By contrast, his religiosity was a significant negative predictor of biological family status, although the zero-order correlation between these two variables was only .02. Predictor variables accounted for 25% of the variance in biological-family status.

Figure 2.

Figure 2

Open in a new tab

Results of the structural equation modeling analysis. *p = .05.

Both partners’ annual incomes were significantly related to their own and their partner’s financial strain, accounting for 22% of the variance in hers and 30% of the variance in his financial strain. His income appeared to be somewhat more important than hers as a predictor of financial strain for both members of the couple.

The woman’s assessment of relationship quality was predicted by marital status, biological family status, and her financial strain. Higher quality was associated with being married; lower quality was associated with biological-family status and her financial strain. These variables accounted for 30% of the variation in her relationship quality. Similarly, the man’s relationship satisfaction was significantly predicted by being married, biological-family status, and his financial strain; these variables accounted for 24% of the variance in his latent relationship quality.

The presence of both biological parents was a significant positive predictor of relationship stability; however, being married was not. It should be noted that marital status and biological-family status were highly correlated (r = .58), and thus multiple collinearity may explain the failure of marital status to attain significance, especially given its sizable zero-order correlation with stability (r = .50). Her relationship quality was a statistically significant predictor of stability, although, surprisingly, his relationship quality was not. Fifty-seven percent of the variation in relationship stability was explained by this set of predictors.

A number of mediation relationships between predictor variables and relationship stability are specified in the model shown in Figure 1. The statistical significance of these hypothesized indirect effects on relationship stability was tested with the bias-corrected bootstrap sampling procedure that is available in the Mplus program. Total indirect effects of the variables on relationship stability are presented in Table 4. As can be seen, five of the 10 expected indirect effects of predictor variables on relationship stability were statistically significant. His education and income, her religiosity and financial strain, and biological family status all showed significant indirect effects in the prediction of relationship stability. For all variables except her financial strain, there was more than one mediating path that accounted for the overall indirect effects that were observed. All specific statistically significant mediational paths are listed in Table 4 (paths a to j).

Table 4.

Statistically Significant Indirect Effects of the Predictor Variables on Relationship Stability

Predictor Effect estimate 95% CI Standardized effect estimate
His education .149 [0.045, 0.230] .193
 a. His education > his income > her financial strain > her quality > stability (+)
 b. His education >his income > married >her quality >stability (+)
 c. His education > his income > biofamily > stability (+)
 d. His education > his income > biofamily > her quality > stability (−)
His income .029 [0.010, 0.044] .305
 Same paths as a through d above, beginning at the second step.
Her religiosity .742 [0.197, 1.380] .238
 e. Her religiosity > married > her quality > stability (+)
 f. Her religiosity > biofamily > stability (+)
 g. Her religiosity > biofamily > her quality > stability (−)
Biological family −.259 [−0.598, −0.011] −.195
 h. Biofamily > stability (+)
 i. Biofamily > her quality > stability (−)
Her financial strain −.680 [−1.336, −0.149] −.183
 j. Her financial strain > her quality > stability
Open in a new tab

Note. The 95% confidence interval (CI) shows the lower and upper 2.5% of the bootstrap estimates for the indirect effects. Cases where the confidence interval does not include zero are statistically significant. Specific significant indirect pathways are listed for each predictor. The sign of the net effect of each pathway is provided in parentheses.

Stability and specific family configurations

To clarify the pattern of results regarding specific family configurations and relationship quality and stability, we calculated the stability and relationship quality of couples in the five family configurations found in the sample: married biological parents, cohabiting biological parents, married stepparents, cohabiting stepparents, and married nonparents (nonparent relatives, adoptive or foster parents). The four measures that served as indicators of the latent relationship-quality measure were aggregated to form a single score for each member of the couple, using the weights (factor loadings) from the measurement model. As shown in Table 5, the most stable couples were the married nonparents (95% intact), followed by the married biological parents (91% intact), and the cohabiting biological parents (86% intact). Far less stable were the married stepfamily couples (51% intact) and the cohabiting step-family couples (38% intact). A logistic regression analysis predicting stability was performed using dummy-coded variables to represent each of the groups (married biological families served as the reference group). It indicated that the relationship stability of these five groups was significantly different, χ2(4, N = 207) = 52.57, p < .001. Post hoc analyses indicated that the two stepfamily groups of parents were significantly less stable than the other three groups but did not differ from one another in stability. The remaining three groups of parents did not differ significantly from one another. A very different pattern emerged for perceived relationship quality. As shown in Table 5, the lowest levels of his and her relationship quality at Wave 1 were reported by the unmarried biological parents. A multiple regression analysis predicting relationship quality comparing this group of parents to the other four groups of parents found that they reported significantly lower levels of his relationship quality, t(204) = −2.20, p = .029, and her relationship quality, t(204) = −2.42, p = .016, whereas the other four groups of parents did not differ significantly from one another on his or her relationship quality.

Table 5.

Relationship Stability and Quality by Family Configuration

Variable n His relationship quality
Her relationship quality
Remain together
M SD M SD %
Bioparents unmarried 7 6.17 1.57 5.74 1.50 .86
Bioparents married 106 7.21 1.28 6.78 1.33 .91
Stepfamily unmarried 29 7.08 1.50 6.75 1.10 .38
Stepfamily married 45 7.35 1.09 7.17 1.31 .51
Nonparents married 20 7.45 1.25 7.51 0.94 .95
Open in a new tab

Relationship quality was relatively high among all but the cohabiting biological parents at Wave 1. Nevertheless, the two groups of stepfamily couples were much more likely to dissolve their relationships than the other groups 5 years later. Initial relationship quality was not an accurate predictor of relationship stability for stepfamily couples; their relatively high initial level of relationship quality did not protect them from high rates of breakup. Similarly, initial relationship quality was not an accurate predictor of stability for cohabiting biological parents; their relatively low level of initial relationship quality was not associated with high rates of dissolution 5 years later. These results are considered further below.

Discussion

Theories of human development emphasize the cascading influence of conditions at one point in time on outcomes in the future (Cox, Mills-Koonce, Propper, & Gariépy, 2010). This perspective is incorporated into the vulnerability–stress–adaptation model (Karney & Bradbury, 1995). The social ecological contexts in which people live influence their opportunities, attitudes, behaviors, and relationships across time (Masten et al., 2005). The current investigation examined level of education and religiosity as the starting points in a cascade of events that influence the relationship stability of African American couples who are parenting an elementary-school-age child.

Lack of education was viewed as an enduring vulnerability factor in our model. Education strongly influenced income and, through a variety of pathways, significantly predicted the stability of African American couples’ intimate relationships. Many African Americans face barriers to education, including economically disadvantaged neighborhoods, underfunded schools, lack of mentors and successful role models, and limited family resources (Crowder & South, 2011; Grogan-Kaylor & Woolley, 2010). Because characteristics of neighborhoods and schools are influential predictors of educational attainment, the importance of intervening at the community level to improve the educational outcomes of African Americans and other low-income minority groups cannot be overemphasized (Crowder & South, 2011). A range of interventions has been explored to give low-income minority families access to better quality schools, including housing vouchers to move to higher income neighborhoods, school vouchers, open enrollment, and specialized magnet and charter schools in low-income neighborhoods; however, most such programs have shown disappointing results (DeLuca & Dayton, 2009). Improving the educational opportunities of low-income minority populations must become a higher local and national priority because the penalties that accompany low levels of education in terms of lifetime earnings are becoming ever greater (Day & Newburger, 2002).

Some of the pathways linking education to relationship outcomes were the same and some were different for men versus women. Men’s education and income were generally more influential than those of women, as discussed below. For both men and women, education showed a significant indirect effect on relationship quality through its association with higher income and lower perceived financial strain. Lower financial strain was associated with higher relationship quality, consistent with prior research (Conger & Donellen, 2007). Notably, only men’s education had a significant indirect effect on relationship stability, through its association with income, women’s financial strain, and relationship quality (path a in Table 4) as well as through its association with family structures that research shows to be more stable: marriage rather than cohabitation and biological rather than step-family status.

Men with higher education and income were more likely to be married than in a cohabiting relationship. The indirect path from education through income to marriage, to her relationship quality, to relationship stability was statistically significant (path b in Table 4). Research suggests that African Americans place greater emphasis on economic security and upward mobility as requirements for marriage than do European Americans (James, 1998; Tucker, 2000). Opportunities for educational and occupational achievement must be increased to increase rates of marriage among African American men.

Men with higher education and income were also more likely to be in couples where both partners were the biological parents of the index elementary-school-age child rather than in stepfather–biological mother couples (path c in Table 4). Analyses showed that stepfather–biological mother couples were the least stable. Second and higher order marriages tend to be less stable than first marriages (Bumpass & Raley, 2007). Remarried individuals tend to have lower wealth and income (Morrison & Ritualo, 2000) and to value autonomy to a greater extent than do continuously married couples (Allen, Baucom, Burnett, Epstein, & Rankin-Esquer, 2001). Becoming a stepparent is complicated by a lack of clear guidelines for the role of the stepparent in children’s lives and the complexity of meeting parenting obligations that span multiple households (Hetherington & Kelly, 2002). All of these factors may contribute to the lower stability of stepfamily couples.

It has been proposed that because African American families have a history of considering people with no biological or legal ties as members of the family (“fictive kin”), the transition to stepfamily status may be easier than among other ethnic and racial groups (Stewart, 2007). African American resident stepfathers’ level of involvement with their stepchildren was similar to that of resident biological fathers in a study of Head Start families (Fagan, 1998). Another study found that racial minority stepfathers reported higher satisfaction in their experiences with stepchildren than did European American stepfathers (MacDonald & DeMaris, 1996). Despite these potential strengths, African American stepfamilies are subject to greater stigmatization than European American stepfamilies (Berger, 1998). African American stepfathers, in particular, are viewed negatively because of stereotypes that they are transient in their stepchildren’s lives and more likely to be abusive, although objective data reveal no racial differences in rates of abuse among stepfathers (Rasheed, 1999; U.S. Department of Human Services, 2005). This stigmatization may be an additional stressor that undermines stepfather adjustment and commitment in African American families (Stewart, 2007).

An unexpected finding was that although biological-family status was associated with higher relationship stability, it was associated with lower relationship quality. As noted previously, the lowest mean level of relationship satisfaction was found among cohabiting biological-family couples, even though these couples were highly stable. This small group of unhappy but committed couples influenced the association between biological-family status and relationship quality. A second contributor to this association was the generally high level of relationship quality reported by the married and cohabiting stepfather–biological mother couples at Wave 1. Nevertheless, biological mother–stepfather couples were much more likely to dissolve their relationships than any other subgroup of couples. This is consistent with prior research that has found little difference in cross-sectional comparisons of the relationship quality of first-married versus remarried couples, despite higher rates of divorce in remarried couples (Amato, Booth, Johnson, & Rogers, 2007). In our sample, the stepfather–biological mother couples had been together for many fewer years (M = 3.74, SD = 3.26) than either the biological-family couples (M = 12.63, SD = 5.41) or the nonparent couples (M = 25.05 years, SD = 11.67). Although stability increases with relationship duration, relationship satisfaction erodes over time (e.g., VanLaningham, Johnson, & Amato, 2001); the stepfather couples may simply have been earlier in the erosion process at the initial wave of data collection. It is also possible that attrition of the least stable biological-family couples occurred before our study began, leaving only those with high levels of commitment. Finally, the relationship quality among the stepfather–biological mother couples was undoubtedly eroded over the 5-year study period by the range of stressors that are encountered by stepfamilies. Interventions tailored to assist African American couples who are striving to raise children from multiple unions are clearly needed (Stewart, 2007).

The high stability of the 20 couples in which neither individual was the parent of the target child merits comment. In the African American community, when neither biological parent is able to care for a child, a member of the extended family often takes on child-rearing responsibility (Stewart, 2007). Families may “screen” such substitute caregivers, selecting the individuals who can provide the most stable home, which may be one reason that the nonparent couples were all married and had very stable relationships. Similar formal screening occurs when couples seek to become adoptive or foster parents. It seems likely that family and nonfamily caregivers were selected for high levels of adjustment and stability.

Contrary to expectation, only women’s relationship quality significantly predicted stability. This is consistent with a meta-analysis conducted by Karney and Bradbury (1995), although the gender difference they found in predictive strength was small. By contrast, a recent long-term prospective study found that only men’s relationship satisfaction predicted stability (Hirschberger, Srivastava, Marsh, Cowan, & Cowan, 2009). Further research is needed on potential gender differences in relationship dynamics among African American couples.

Religious involvement was viewed as an enduring resource variable that has long-term protective influence on intimate relationships. Consistent with previous research (Lichter & Carmalt, 2009), religiosity was more influential among African American women than among men in our study. One significant pathway led from women’s religiosity to marital status, to her relationship quality, to relationship stability (path e in Table 4). Religious women were more likely to marry than to cohabit, and married women rated their relationships more positively than did those who were cohabiting; women’s relationship quality was associated with stability. The link between religious involvement and marriage has been documented in nationally representative samples (Lehrer, 2004) and among low-income African American couples (Wilcox & Wolfinger, 2007). We did not find a significant direct association between religiosity and relationship quality. A stronger link to stability than to quality is consistent with findings from a meta-analysis of 94 studies that examined links between religious involvement and marital outcomes (Mahoney, Pargament, Tarakeshwar, & Swank, 2001), which showed a weaker link from religiosity to marital quality (d = 0.05) than to marital commitment (d = 0.19) and stability (d = 0.125).

Religious women were more likely to be in couples that included the target child’s biological father, which predicted relationship stability (path f in Table 4). As noted previously, biological-family status predicted lower relationship quality among women, which was associated with lower stability (path g in Table 4). Once again, the low relationship quality among cohabiting biological-family couples, the high initial relationship quality among stepfather–biological mother couples, and the longer relationship duration of biological-family couples probably account for the lower relationship quality reported by women in biological-family couples.

There is a need for culturally sensitive relationship-enhancement programs for African American couples. Given the importance of religion to many African Americans, the inclusion of a spiritual or religious component in relationship interventions may be beneficial (Karney, Garvan & Thomas, 2003). A recent randomized clinical trial found that African American couples showed significant improvement in marital quality relative to a control group when they participated in a modification of the widely used PREP marital enhancement program (Markman, Stanley, & Blumberg, 2010) that incorporated spirituality and prayer into the communication skills-oriented PREP program (Beach, Hurt, Fincham, McNair, & Stanley, 2011). Moreover, African American women responded with greater increases in marital quality to the prayer- and spirituality-focused version of the PREP program than they did to a culturally sensitive PREP program that was designed to be Afrocentric but did not include a spiritual component.

Limitations

The sample is not nationally representative of African American families with children; it was drawn from only two states and excludes residents of major metropolitan areas. The couples in this study were all parenting a 9- to 11-year old child and most were at least 35 years old; thus, findings may not generalize to couples who are at earlier stages in the life cycle. Most couples in our sample had been together for at least five years; as a consequence, individuals with poorer quality relationships who broke up early in their relationship were not represented in the sample. Our model tests only a limited number of vulnerability and resource factors; others, including exposure to race-related stressors, should also be investigated. An additional limitation is that the proposed model implies a series of events that unfold over time, yet all of our predictor variables were assessed at a single time point (Wave 1). We were thus unable to assess the directionality of influence and errors of measurement over time. All of the measures used in the current investigation were self-reported, which leads to shared method variance. Future studies should also include direct observation of couples. Counterbalancing these liabilities, the sample included couples with a wide range of income and education levels, thereby avoiding the restriction of range in socioeconomic status that characterizes most studies of African Americans, in which only poor inner-city families are sampled.

Additional studies that follow African American couples from the beginning of their relationships are needed to test models of relationship quality and stability. Our results strongly suggest that contextual factors, such as educational and economic opportunities, and cultural factors, such as religious involvement, are highly influential in the relationship stability of African American couples. Solutions to the challenges that face African American couples may be found in school improvement and economic development in poor communities as much as in the design of relationship interventions. Our findings also suggest that interventions that help stepfamily couples cope with the challenges of multiple roles and family obligations may be helpful for African American couples. Finally, culturally sensitive relationship interventions that incorporate religion into empirically validated programs may be better received and more effective than programs that do not include this resource, which is very important in the lives of many African Americans.

Acknowledgments

This study was supported by National Institute of Mental Health Grant MH6266 to Carolyn E. Cutrona.

Contributor Information

Carolyn E. Cutrona, Department of Psychology, Iowa State University

Daniel W. Russell, Department of Human Development and Family Studies, Iowa State University

Rebecca G. Burzette, Department of Psychology, Iowa State University

Kristin A. Wesner, Department of Psychology, Clarke College

Chalandra M. Bryant, Department of Human Development and Family Studies, University of Georgia

References

  1. Allen ES, Baucom DH, Burnett CK, Epstein N, Rankin-Esquer LA. Decision-making power, autonomy, and communication in remarried spouses compared with first-married spouses. Family Relations. 2001;50:326–334. doi: 10.1111/j.1741-3729.2001.00326.x. [DOI] [Google Scholar]
  2. Amato PR, Booth A, Johnson DR, Rogers SJ. Alone together: How marriage in America is changing. Cambridge, MA: Harvard University Press; 2007. [Google Scholar]
  3. Beach SRH, Hurt TR, Fincham FD, McNair LM, Stanley SM. Enhancing marital enrichment through spirituality: Efficacy data for prayer focused relationship enhancement. Psychology of Religion and Spirituality. 2011;3:201–216. doi: 10.1037/a0022207. [DOI] [Google Scholar]
  4. Berger R. Stepfamilies: A multi-dimensional perspective. New York, NY: Haworth Press; 1998. [Google Scholar]
  5. Bodenmann G, Charvoz L, Bradbury TN, Bertoni A, Iafrate R, Guiliani C, Behling J. The role of stress in divorce: A three-nation retrospective study. Journal of Social and Personal Relationships. 2007;24:707–728. doi: 10.1177/0265407507081456. [DOI] [Google Scholar]
  6. Brodsky AE. The role of religion in the lives of resilient, urban, African American single mothers. Journal of Community Psychology. 2000;28:199–219. doi: 10.1002/(SICI)1520-6629(200003)28:2&#x0003c;199:: AID-JCOP7&#x0003e;3.0.CO;2-3. [DOI] [Google Scholar]
  7. Broman CL. Marital quality in Black and White marriages. Journal of Family Issues. 2005;26:431–441. doi: 10.1177/0192513X04272439. [DOI] [Google Scholar]
  8. Brown SL. Relationship quality dynamics of cohabiting unions. Journal of Family Issues. 2003;24:583–601. doi: 10.1177/0192513X03024005001. [DOI] [Google Scholar]
  9. Bryant CM, Wickrama KAS, Bolland JM, Brant BM, Cutrona CE, Stanik CE. Race matters, even in marriage: Identifying factors linked to marital outcomes for African Americans. Journal of Family Theory & Review. 2010;2:157–174. doi: 10.1111/j.1756-2589.2010.00051.x. [DOI] [Google Scholar]
  10. Bumpass LL, Lu H-H. Trends in cohabitation and implications for children’s family contexts in the United States. Population Studies. 2000;54:29–41. doi: 10.1080/713779060. [DOI] [PubMed] [Google Scholar]
  11. Bumpass L, Raley K. Measuring separation and divorce. In: Hofferth SL, Casper LM, editors. Handbook of measurement issues in family research. Mahwah, NJ: Erlbaum; 2007. pp. 125–143. [Google Scholar]
  12. Buss DM, Kenrick D. Evolutionary social psychology. In: Gilbert DL, Fiske S, Lindzey G, editors. Handbook of social psychology. New York, NY: McGraw-Hill; 1988. pp. 982–1026. [Google Scholar]
  13. Conger RD, Donellen MB. An interactionist perspective on the socioeconomic context of human development. Annual Review of Psychology. 2007;58:175–199. doi: 10.1146/annurev.psych.58.110405.085551. [DOI] [PubMed] [Google Scholar]
  14. Conger RD, Elder GH. Families in troubled times: Adapting to change in rural America. New York, NY: Aldine de Gruyter; 1994. [Google Scholar]
  15. Conger RD, Elder GH, Jr, Lorenz FO, Conger K, Simons R, Whitbeck L. Linking economic hardship to marital quality and instability. Journal of Marriage and the Family. 1990;52:643–656. 869BF2EB-DF92–E28C-8C68–1C5A7513EB50. [Google Scholar]
  16. Conger RD, Wallace LE, Sun Y, Simons RL, McLoyd VC, Brody GH. Economic pressure in African American families: A replication and extension of the family stress model. Developmental Psychology. 2002;38:179–193. 869BF2EB-DF92–E28C-8C68–1C5A7513EB50. [PubMed] [Google Scholar]
  17. Cox MJ, Mills-Koonce R, Propper C, Gariépy J-L. Systems theory and cascades in developmental psychopathology. Development and Psychopathology. 2010;22:497–506. doi: 10.1017/S0954579410000234. [DOI] [PubMed] [Google Scholar]
  18. Crowder K, South SJ. Spatial and temporal dimensions of neighborhood effects on high school graduation. Social Science Research. 2011;40:87–106. doi: 10.1016/j.ssresearch.2010.04.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Cutrona CE, Russell DW, Abraham WT, Gardner KA, Melby JN, Bryant C, Conger RD. Neighborhood context and financial strain as predictors of marital interaction and marital quality in African American couples. Personal Relationships. 2003;10:389–409. doi: 10.1111/1475-6811.00056. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Cutrona CE, Russell DW, Brown PA, Clark LA, Hessling RM, Gardner KA. Neighborhood context, personality, and stressful life events as predictors of depression among African American women. Journal of Abnormal Psychology. 2005;114:3–15. doi: 10.1037/0021-843X.114.1.3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Cutrona CE, Russell DW, Hessling RM, Brown PA, Murry V. Direct and moderating effects of community context on the psychological well-being of African American women. Journal of Personality and Social Psychology. 2000;79:1088–1101. doi: 10.1037/0022-3514.79.6.1088. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Day JC, Newburger EC. The big payoff: Educational attainment and synthetic estimates of work-life earnings. Current Population Reports. 2002 Retrieved from www.census.gov/prod/2002pubs/p23-210.pdf.
  23. DeLuca S, Dayton E. Switching social contexts: The effects of housing mobility and school choice programs on youth outcomes. Annual Review of Sociology. 2009;35:457–491. doi: 10.1146/annurev-soc-070308-120032. [DOI] [Google Scholar]
  24. Enders CK. Applied missing data analysis. New York, NY: Guilford Press; 2010. [Google Scholar]
  25. Fagan J. Correlates of low-income African American and Puerto Rican fathers’ involvement with their children. Journal of Black Psychology. 1998;24:351–367. doi: 10.1177/00957984980243006. [DOI] [Google Scholar]
  26. Futris TG, Nielson RB, Olmstead SB. No degree, no job: Adolescent mothers’ perceptions of the impact that adolescent fathers’ human capital has on paternal financial and social capital. Child and Adolescent Social Work Journal. 2010;27:1–20. doi: 10.1007/s10560-009-0187-8. [DOI] [Google Scholar]
  27. Gibson-Davis CM, Edin K, McLanahan S. High hopes but even higher expectations: The retreat from marriage among low-income couples. Journal of Marriage and Family. 2005;67:1301–1312. doi: 10.1111/j.1741-3737.2005.00218.x. [DOI] [Google Scholar]
  28. Grogan-Kaylor A, Woolley ME. The social ecology of race and ethnicity school achievement gaps: Economic, neighborhood, school, and family factors. Journal of Human Behavior in the Social Environment. 2010;20:875–896. doi: 10.1080/10911359.2010.494927. [DOI] [Google Scholar]
  29. Hetherington EM, Kelly J. For better or for worse: Divorce reconsidered. New York, NY: Norton; 2002. [Google Scholar]
  30. Hirschberger G, Srivastava S, Marsh P, Cowan CP, Cowan PA. Attachment, marital satisfaction, and divorce during the first fifteen years of parenthood. Personal Relationships. 2009;16:401–420. doi: 10.1111/j.1475-6811.2009.01230.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Hunter AG, Sellers SL. Feminist attitudes among African American women and men. Gender & Society. 1998;12:81–99. doi: 10.1177/089124398012001005. [DOI] [Google Scholar]
  32. James AD. What’s love got to do with it? Economic viability and the likelihood of marriage among African American men. Journal of Comparative Family Studies. 1998;29:373–386. [Google Scholar]
  33. Karney BR, Bradbury TN. The longitudinal course of marital quality and stability: A review of theory, method, and research. Psychological Bulletin. 1995;118:3–34. doi: 10.1037/0033-2909.118.1.3. [DOI] [PubMed] [Google Scholar]
  34. Karney BR, Garvan C, Thomas M. Family formation in Florida 2003: Baseline survey of attitudes, beliefs, and demographics relating to marriage and family formation. Gainesville: University of Florida; 2003. [Google Scholar]
  35. Lehrer EL. The role of religion in union formation: An economic perspective. Population Research and Policy Review. 2004;23:161–185. doi: 10.1023/B:POPU.0000019917.00860.ba. [DOI] [Google Scholar]
  36. Levin JS, Taylor RJ, Chatters LM. A multidimensional measure of religious involvement for African Americans. Sociological Quarterly. 1995;36:157–173. doi: 10.1111/j.1533-8525.1995.tb02325.x. [DOI] [Google Scholar]
  37. Lichter DT, Carmalt JH. Religion and marital quality among low-income couples. Social Science Research. 2009;38:168–187. doi: 10.1016/j.ssresearch.2008.07.003. [DOI] [Google Scholar]
  38. MacDonald WL, DeMaris A. Parenting stepchildren and biological children. Journal of Family Issues. 1996;17:5–25. doi: 10.1177/019251396017001002. [DOI] [Google Scholar]
  39. Mahoney A, Pargament KI, Tarakeshwar N, Swank AB. Religion in the home in the 1980s and 1990s: A meta-analytic review and conceptual analysis of links between religion, marriage, and parenting. Journal of Family Psychology. 2001;15:559–596. doi: 10.1037/0893-3200.15.4.559. [DOI] [PubMed] [Google Scholar]
  40. Markman HJ, Stanley SM, Blumberg SL. Fighting for your marriage. 2. New York, NY: Jossey-Bass; 2010. [Google Scholar]
  41. Masten AS, Roisman GI, Long JD, Burt KB, Obradovic J, Riley JR, Tellegen A. Developmental cascades: Linking academic achievement and externalizing and internalizing symptoms over 20 years. Developmental Psychology. 2005;41:733–746. doi: 10.1037/0012-1649.41.5.733. [DOI] [PubMed] [Google Scholar]
  42. Mattis JS. Religion and spirituality in the meaning-making and coping experiences of African American women: A qualitative analysis. Psychology of Women Quarterly. 2002;26:309–321. doi: 10.1111/1471-6402.t01&#x02013;2-00070. [DOI] [Google Scholar]
  43. Morrison DR, Ritualo A. Routes to children’s economic recovery after divorce: Are cohabitation and remarriage equivalent? American Sociological Review. 2000;65:560–580. doi: 10.2307/2657383. [DOI] [Google Scholar]
  44. Orbuch TL, Eyster SL. Division of household labor among Black couples and White couples. Social Forces. 1997;76:301–332. doi: 10.2307/2580327. [DOI] [Google Scholar]
  45. Orbuch TL, Veroff J, Hassan H, Horrocks J. Who will divorce: A 14-year study of Black couples and White couples. Journal of Social and Personal Relationships. 2002;19:179–202. doi: 10.1177/0265407502192002. [DOI] [Google Scholar]
  46. Orbuch TL, Veroff J, Hunter AG. Black couples, White couples: The early years of marriage. In: Hetherington EM, editor. Coping with divorce, single parenting, and remarriage. Mahwah, NJ: Erlbaum; 1999. pp. 23–43. [Google Scholar]
  47. Osborne C, Manning WD, Smock PJ. Married and cohabiting partners’ relationship stability: A focus on race and ethnicity. Journal of Marriage and Family. 2007;69:1345–1366. doi: 10.1111/j.1741-3737.2007.00451.x. [DOI] [Google Scholar]
  48. Rasheed JM. Obstacles to the role of inner-city, low-income, noncustodial African American fathers. Journal of African American Men. 1999;4:9–23. doi: 10.1007/s12111-999-1006-1. [DOI] [Google Scholar]
  49. Russell DW, Kahn JH, Spoth R, Altmaier EM. Analyzing data from experimental studies: A latent variable structural equation modeling approach. Journal of Counseling Psychology. 1998;45:18–29. doi: 10.1037/0022-0167.45.1.18. [DOI] [Google Scholar]
  50. Shorter-Gooden K. Multiple resistance strategies: How African American women cope with racism and sexism. Journal of Black Psychology. 2004;30:406–425. doi: 10.1177/0095798404266050. [DOI] [Google Scholar]
  51. Simmons T, O’Connell M. Married-couple and unmarried-couple partner households: 2000. Washington, DC: U.S. Census Bureau; 2003. [Google Scholar]
  52. Simons R. Unpublished manuscript. Department of Sociology, University of Georgia; Athens: 1997. A scale to assess religious involvement. [Google Scholar]
  53. Stafford L, Kline SL, Rankin CT. Married individuals, cohabitors, and cohabitors who marry: A longitudinal study of relational and individual well-being. Journal of Social and Personal Relationships. 2004;21:231–248. doi: 10.1177/0265407504041385. [DOI] [Google Scholar]
  54. Stewart SD. Brave new stepfamilies: Diverse paths toward stepfamily living. Thousand Oaks, CA: Sage; 2007. [Google Scholar]
  55. Sweeney MM. Remarriage and stepfamilies: Strategic sites for family scholarship in the 21st century. Journal of Marriage and Family. 2010;72:667–684. doi: 10.1111/j.1741-3737.2010.00724.x. [DOI] [Google Scholar]
  56. Thoits PA. Stress and health. Journal of Health and Social Behavior. 2010;51(Suppl 1):S41–S53. doi: 10.1177/0022146510383499. [DOI] [PubMed] [Google Scholar]
  57. Toth JF, Jr, Xu X. Ethnic and cultural diversity in father’s involvement: A racial/ethnic comparison of African American, His-panic, and White fathers. Youth & Society. 1999;31:76–99. doi: 10.1177/0044118X99031001004. [DOI] [Google Scholar]
  58. Tucker MB. Marital values and expectations in context: Results from a 21-city survey. In: Waite LJ, editor. The ties that bind. New York, NY: Aldine de Gruyter; 2000. pp. 166–187. [Google Scholar]
  59. U.S. Department of Human Services. Male perpetrators of child maltreatment: Findings from NCANDS. 2005 Retrieved from http://purl.access.gpo.gov/GPO/LPS77745.
  60. VanLaningham J, Johnson DR, Amato PR. Marital happiness, marital duration, and the U-shaped curve: Evidence from a five-wave panel study. Social Forces. 2001;79:1313–1341. doi: 10.1353/sof.2001.0055. [DOI] [Google Scholar]
  61. Vartanian TP, Buck PW. Childhood and adolescent neighborhood effects on adult income: Using siblings to examine differences in ordinary least squares and fixed-effects models. Social Service Review. 2005;79:60–94. doi: 10.1086/426718. [DOI] [Google Scholar]
  62. Veroff J, Douvan E, Orbuch TL, Acitelli LK. Happiness in stable marriages: The early years. In: Bradbury TN, editor. The developmental course of marital dysfunction. New York, NY: Cambridge University Press; 1998. pp. 152–179. [Google Scholar]
  63. Wilcox WB, Wolfinger NH. Then comes marriage? Religion, race, and marriage in urban America. Social Science Research. 2007;36:569–589. doi: 10.1016/j.ssresearch.2006.02.005. [DOI] [Google Scholar]

RESOURCES