African American Couples’ Coparenting Satisfaction and Marital Characteristics in the First Two Decades of Marriage

Abstract

This study charted the trajectory of coparenting satisfaction during the first and second decades of marriage and examined links between marital characteristics and coparenting satisfaction. Data came from a 3-year study of 145 African American mother-father dyads with pre-to late- adolescent-age offspring. Multi-level growth curves revealed an inverted U-shaped pattern of change in coparenting satisfaction; this effect was qualified by youth age such that the quadratic pattern was evident in families with older but not younger offspring. Controlling for cross-time averages of marital characteristics, changes in marital love were positively related, and changes in marital conflict were negatively related to changes in coparenting satisfaction, with stronger links for fathers than mothers. Inter-parental incongruence in childrearing attitudes moderated the effects of love, such that parents with more incongruent attitudes and lower levels of love reported the lowest levels of coparenting satisfaction.

Coparenting, defined as the ways in which parental figures work together in their parenting roles (Feinberg, 2003) is reflected in feelings of satisfaction about shared parenting (Van Egeren, 2004). As a family systems construct, coparenting satisfaction is affected by developmental changes within the family (e.g., Cox & Paley, 2002). Consistent with this tenet is evidence of declines in coparenting satisfaction during the transition to parenthood, as parents negotiate coordinated parenting for the first time (Davis, Schoppe-Sullivan, Mangelsdorf, & Brown, 2009). We know little, however, about coparenting satisfaction during later developmental periods, when offspring’s transition to adolescence may pose new sets of challenges for mothers and fathers. Coparenting research is also limited to studies of White two-parent families. Although a small body of research has explored coparenting of single African American mothers and non-marital coparents, such as grandparents (Dorsey, Forehand, & Brody, 2007; Jones, Forehand, Dorsey, Foster, & Brody, 2005), inter-generational coparenting may differ in important ways from marital coparenting. To advance understanding of the developmental course of coparenting satisfaction, we charted its trajectory of change during the first two decades of marriage in two-parent African American families. Given close links between coparenting and marital quality from prior short term studies of White couples (Schoppe-Sullivan, Mangelsdorf, Brown, & Sokolowski, 2007), we also tested whether and how changes in marital love and conflict corresponded to changes in African American couples’ coparenting satisfaction across this broad period of marriage.

In addition to coordinating parenting behaviors, effective parenting also involves coordination of mothers’ and fathers’ attitudes and values about childrearing. Mate selection (Epstein & Gutterman, 1984) and social exchange theories (Kelly & Thibaut, 1978) offer support for the idea that inter-parental congruence in attitudes and personality characteristics is linked to happiness and satisfaction in close relationships. Congruence in attitudes toward childrearing roles may be of particular importance in understanding coparenting satisfaction to the extent that attitudes have implications for men’s and women’s family roles and role satisfaction (Lye & Biblarz, 1993). Further, racial variations in patterns of employment and family formation may mean that gendered attitudes towards childrearing have special importance for African American men and women (for a review, see Kane, 2000). Although there is some ambiguity as to whether African American mothers and fathers tend to hold more traditional versus egalitarian childrearing attitudes, parents do not always complement their partner’s childrearing beliefs. In turn, parents’ incongruent attitudes towards who does what in childrearing could reflect coparenting disagreement, putting their satisfaction with coordinated parenting efforts at risk. Extending research on attitude congruence in non-White families, we examined the role of inter-parental congruence in attitudes about childrearing roles for coparenting satisfaction, including whether attitude congruence moderated the links between marital love and conflict and coparenting satisfaction.

Developmental Transitions and the Family System

A cornerstone of family systems theory is the idea that relationships change over time and are affected by developmental changes in the family. For example, both European American and African American parents report lower marital satisfaction and more frequent marital conflict after the transition to parenthood compared to before (Bradbury, Fincham, & Beach, 2000). We know less about changes in marital quality during later stages of offspring’s development, but one study by Whiteman and colleagues (2007) showed declines in marital satisfaction and love through the second decade of marriage in European American families which were partially explained by offspring’s pubertal transition. This work suggests that offspring’s adolescence had negative implications for marital relationships and added to the literature by documenting “child effects” on marriage later in the family life cycle.

Research on the developmental trajectory of coparenting is more limited than research on global marital change, and focuses on White families during the early parenting years. Prior work showed that coparenting satisfaction during the first six months of parenthood was relatively stable (Van Egeren, 2004), whereas challenges surrounding child development were related to declines in coparenting support from infancy to early childhood (Davis, Schoppe-Sullivan, Mangelsdorf, & Brown, 2009). Although scholars agree that later transitions, for example, when youth become adolescents or leave home in young adulthood, are periods of major flux in families (e.g., Petersen, 1988; Paikoff & Brooks-Gunn, 1991), these periods have received little attention in the coparenting literature. Offspring’s adolescence, in particular, brings about a new set of parenting goals and demands, which may require mothers and fathers to renegotiate their coparenting relationship. To the extent that adolescence imposes new parenting challenges, mothers and fathers may adapt to their changing roles in different ways and experience decreased satisfaction with coparenting.

In contrast to the idea that the transition to adolescence challenges coparenting in a negative way, some scholars have argued instead that coparenting should improve over time as parents gain experience and establish coordinated parenting strategies (Van Egeren, 2004). Based on this perspective, we would predict that parents will become increasingly satisfied with coparenting over the course of their marriage. In this paper we tested these competing ideas by examining changes in coparenting satisfaction across two decades of marriage and by exploring the moderating role of youth age in patterns of change. Given that husbands often report more positive coparenting experiences than their wives (Margolin, et al., 2001; Van Egeren, 2004), we also tested whether levels and/or patterns of change in coparenting satisfaction differed for mothers versus fathers.

Associations between Marital Quality and Coparenting Satisfaction

Although there is substantial overlap between characteristics of coparenting and marital relationships, theory and empirical evidence suggest that coordinated parenting is distinct from other dimensions of marriage (Abidin & Brunner, 1995). In part this is because coparenting encompasses characteristics and behaviors of both parents and offspring, and as a result, captures processes that are distinct from both marital dyads and parent-child interactions. Further, past work shows that features of couple relationships that pertain to childrearing are more strongly related to parenting and child adjustment than is general marital quality (McHale & Rasmussen, 1998), and that coparenting can be positive even in the face of marital distress (McHale, et al., 2000).

An emotional spillover perspective highlights the transfer of mood and affect from one domain of family functioning to another (Bolger, DeLongis, Kessler, & Wethington, 1989). Research on spillover has examined the links between qualities of dyadic marital relationships and coparenting characteristics and shows that high marital love and low marital conflict were associated with more positive coparenting among parents of young children (Margolin et al., 2001). From a family systems perspective marital and coparenting processes are not static, but are expected to change over time in conjunction with family circumstances. The extent to which and ways in which changes in marital quality are related to corresponding changes in coparenting satisfaction, especially through later points in family development when marital and coparenting relationships may be challenged in different ways, has yet to be addressed by empirical research. Taking the perspective that marital and coparenting relationships are mutually inter-related yet distinct family dynamics, we examined whether changes in marital quality, specifically marital love and marital conflict, were linked to changes in coparenting satisfaction across time. We included a control for individuals’ average marital love and conflict across time to understand the implications of change in marital quality above and beyond stable individual differences.

The Role of Inter-Parental Incongruence in Childrearing Attitudes

The idea that inter-parental attitude congruence is linked to positive family relationships has been supported in clinical and empirical work (Deal, Wampler, & Halverson, 1992). In agreement with a central tenet of mate selection theory, partners with similar attitudes (Hendrick, 1981) reported more marital happiness and satisfaction, whereas couples with incongruent attitudes about family roles had poorer marital quality (Bowen & Orthner, 1983). Research on the role of attitude congruence for coparenting is much more limited. Belsky, Gable, and Crnic, (1995) failed to find evidence of a link between discrepant childrearing attitudes and coparenting support in families of toddlers, but noted that this relation might become significant over time as parents develop different tactics and the coparenting system is challenged by child development.

Extending research on attitude congruence to coparenting in families of adolescents, we explored whether parents with more congruent childrearing attitudes were more satisfied with coparenting. Taking a cumulative risk perspective, we also tested whether incongruent childrearing attitudes exacerbated the negative impact of low marital warmth or high marital conflict on coparenting satisfaction; it is also possible that congruent childrearing attitudes play a protective role, such that couples with more similar attitudes report higher coparenting satisfaction, even in the face of lower marital love or higher marital conflict. We tested these possibilities by examining whether the degree of congruence between parents’ childrearing attitudes moderated the links between marital quality and coparenting satisfaction.

The Present Study

In sum, the goals of this study were to: (1) chart the trajectory of coparenting satisfaction during the first two decades of marriage in two-parent African American families with adolescent offspring, (2) test the longitudinal links between marital love and marital conflict and coparenting satisfaction, and (3) examine whether inter-parental congruence in childrearing attitudes predicted coparenting satisfaction or moderated the links between marital love, marital conflict, and coparenting satisfaction. With respect to our first study goal, given competing theories and a lack of prior research we did not specify a hypothesis about the pattern of change in coparenting satisfaction, but based on prior research, we expected that fathers would be more satisfied than mothers. To gain insights into the potential challenges of adolescent development for coparenting, we also tested whether offspring age moderated the pattern of change over time in coparenting satisfaction. Our hypotheses relative to our second study goal were grounded in an emotional spillover perspective. We expected that changes in marital love would be positively associated and changes in marital conflict negatively associated with changes in coparenting satisfaction. Prior research showing that fathers’ parenting is more susceptible to marital problems than mothers (Kolak & Volling, 2007), led us to predict further, that these marital-coparenting linkages would be more evident for fathers than mothers. Finally, with respect to our third study goal of examining the implications of parents’ attitudes towards childrearing, we predicted that incongruent attitudes towards childrearing roles would be linked to lower coparenting satisfaction and would also exacerbate the negative effects of low love and high conflict on coparenting satisfaction, whereas attitude congruence would protect couples with low marital love and/or high marital conflict from declines in coparenting satisfaction.

Method

Participants

The data came from 145 married mothers and fathers of adolescents who participated in a three-year longitudinal study of family relationships in two-parent African American families. Given the goals of the larger study, we targeted families that self-identified as Black or African American and that included a mother figure and father figure who were living together with a child in pre- or early adolescence and an older sibling. Recruitment took place in two urban centers in the northeast with substantial African American populations, and we used two strategies to generate the sample (see author for more details). First, we hired African Americans residing in targeted communities to recruit families by advertising in businesses, churches, and at community events. Approximately half of the sample was recruited in this way. To recruit the rest of the sample, we purchased a marketing list of names and addresses of families with offspring in grades 4–7. We sent letters describing the study, and interested and eligible families called a toll-free number or returned a postcard. Of 1,796 letters sent, 131 were returned as undeliverable. Of the 142 families who responded to letters with interest, 93 were eligible to participate. Of those eligible, 86 families were interviewed and 7 could not be located based on the information they provided or they were too busy.

To increase the homogeneity of the sample for these analyses we omitted 8 sets of parents who were not in a couple relationship, (e.g., mother and grandfather pairs), 24 families in which parents separated or divorced during the study period, 15 families in which couples were living together but not married, 7 families in which couples were married for more than 24 years, and 3 families in which couples had been married for fewer than 4 years (the latter two criteria due to insufficient sample size at the tail ends of the distribution of marriage duration). Of the 145 families in the present analyses, at Time 1, mean ages of mothers and fathers were 40.84 (SD = 5.00), and 43.48 (SD = 6.21), respectively, and families were generally working and middle class. Most parents were employed (90% of mothers and 96% of fathers). On average, mothers’ education was 14.84 years (SD = 1.70) and fathers’ education, 14.53 years (SD = 2.31), scores that correspond to vocational school/some college. Fathers worked 45.19 hours per week (SD = 16.78), and mothers worked 33.46 hours per week (SD = 16.53), on average. Mothers’ job prestige averaged 48.37 (SD = 11.76), and fathers’ averaged 47.34 (SD = 13.72, scores that correspond to jobs such as real estate agent and law enforcement officer (Nakao & Treas, 1994). The mean household income for this primarily dual earner sample was $95, 059 (SD = $56,335). With respect to couple relationships, all 145 mothers and fathers were married for the duration of the study period, for an average of 14.02 years (SD = 5.02) at Time 1. Marital duration ranged from 4–24 years. Most offspring were biologically related to mothers (97%) and fathers (86%). Adolescents ranged in age from 8–15 and were about 10.34 (SD = 1.07) years old at Time 1. There were approximately equal numbers of boys (n = 69) and girls (n = 76). Sample attrition was minimal. In addition to the 57 couples that were omitted to reduce heterogeneity, 7 families withdrew over the study period, but they did not differ from other families on demographics at Time 1.

Procedures

Mothers and fathers were interviewed annually in their homes by a team of two interviewers, almost all of whom were African American. Parents reported on relationship experiences and on their personal characteristics and well-being during the past year. Interviews generally lasted two to three hours, and families received a $200 honorarium each year.

Measures

All measures, with the exception of childrearing attitudes, were collected at all three time points. Measures have been validated with African American samples.

Coparenting satisfaction was measured at all three time points using items from the Domains of Marriage scale (Huston, McHale, & Crouter, 1986). On a scale ranging from 1 = extremely dissatisfied to 9 = extremely satisfied, mothers and fathers responded to three questions: (1) “How satisfied are you with your partner’s fundamental principles about how to bring up children (e.g., values, discipline, etc.)” (2) “How satisfied are you with the extent to which your partner supports your decisions about rules and discipline, makes you feel good about the kind of parent you are, etc.?” (3) “How satisfied are you with the way parenting decisions in your family get made and the level of influence you have in those decisions?” Alphas ranged from .84 to .89. Higher scores reflected more satisfaction.

Marital relationship quality was measured at each time. Love was assessed using 9 items from the Relationships Questionnaire (Braiker & Kelley, 1979). Using a scale where 1 = not at all to 9 = very much mothers and fathers reported on marital relationship love during the past year, e.g., “To what extent do you have a sense of ‘belonging’ with your partner?” Alphas ranged from .89 to .95 across study years and parents. Conflict was assessed using 5 items. On the same 9-point scale, mothers and fathers rated their marital conflict during the past year, e.g., “How often do you and your partner argue with one another?” Alphas ranged from .74 to .82. Higher scores reflected more love and more conflict.

Attitudes towards childrearing were assessed at Time 1 using the 7 items from Hoffman and Kloska’s (1995) Gender-Based Attitudes towards Childrearing Scale. Mothers and fathers used a 4-point rating scale, where 1 = strongly agree and 4 = strongly disagree, to rate how well the item described them over the past year, e.g., “It is more important to raise a son to be strong and independent than to raise a daughter that way.” Higher scores reflected more traditional attitudes towards childrearing. Alphas were .80 for mothers and .83 for fathers. Given that we had no predictions about the extent to which different types of incongruence would be important for coparenting satisfaction (i.e., mothers more traditional than fathers; fathers more traditional than mothers), we used the absolute value of the difference between mothers’ and fathers’ attitudes towards childrearing in the analyses. Higher absolute difference reflected greater incongruence between mothers and fathers.

Background and demographic information including offspring ages and gender, household size, mothers’ and fathers’ income, highest education completed, and biological relatedness to offspring was collected at each time point. These variables were included as controls.

Results

Analytic Strategy

To examine the trajectory of coparenting satisfaction as a function of couple relationship duration, we used a multi-level modeling (MLM) strategy. This approach extends multiple regression to account for dependencies in the data, i.e., within person over time, between parents in the same family. Another advantage of MLM is that it provides for the use of unbalanced data, meaning that it is not necessary for every individual to be assessed at the same point in time. This feature of MLM allowed us to use relationship duration as the index of time, despite differences in these durations at each time of measurement. In doing so we were able to detect patterns of change that might have been obscured if we had relied on year of study as the time metric. An MLM framework provides for the use of cases with one or more observations missing at random (Raudenbush, & Byrk, 2002), allowing us to include all 145 families in these analyses.

We began by testing a three level model for coparenting satisfaction using SAS Proc Mixed, version 9.2. The Level 1, or within-person model, captures changes in coparenting satisfaction in relation to time-varying covariates. To describe patterns of change in coparenting satisfaction as a function of marital duration, we included a marital time polynomial at Level 1. Here we also explored whether patterns of change vary according to gender of parents and/or offspring, and biological relatedness. To understand the moderating role of adolescent development for marital-duration related changes in coparenting satisfaction, we included the adolescent age X marital time interaction term, also at Level 1. To examine whether changes in marital love and marital conflict were related to changes in coparenting satisfaction, we also included time-varying covariates at Level 1. These were group-mean centered, or centered around each individual’s cross-time mean.

At Level 2 we included predictors at the between parent, or within-family, level. The Level 2 model accounts for dependencies between members of the same family. By including both mothers and fathers in the same analysis we were able to test the moderating role of parent gender, that is, whether the trajectory of coparenting satisfaction change was different for mothers versus fathers; thus we included the cross-level interaction between the quadratic time polynomial and parent gender to determine whether mothers and fathers differed in their patterns of changes. The variables included at Level 2 were time-invariant characteristics that differed for mothers and fathers (e.g., mothers’ and fathers’ individual attitudes towards childrearing). In addition, to document how changes in parents’ marital experiences were linked to changes in coparenting satisfaction, we included the cross-time means for each individual on the time-varying predictors (included at Level 1). That is, because the Level 2 cross-time means reflect all between-individual variation, controlling for this variable limits the time-varying version of the variable to explaining within-individual variation over time, beyond stable individual differences (Jacobs et al., 2002; Raudenbush & Byrk, 2002).

Level 3 estimates are between-family characteristics, that is, characteristics that are shared by mothers and fathers and do not change over time. Here we included the dyadic score for incongruent attitudes towards childrearing and offspring gender.

In the case of significant interactions by parent gender, to test the slopes we reran the same models treating mothers as the reference group (i.e., dummy coding instead of effect coding). We followed up significant interactions involving two continuous variables using procedures outlined by Aiken and West (1991). Specifically, we distinguished high (1 SD above the mean) versus low (1 SD below the mean) levels of the moderator variable and ran separate models for parents who were more incongruent versus more similar in their attitudes towards childrearing. Simple slopes tests were used to determine the significance of effects at high versus low levels of the moderator (see Aiken and West, 1991 for more details). In reporting the results we focus on significant effects at p < .05, however, we consider trend-level effects, p < .10, when they were consistent with hypotheses and results from prior research. We begin by describing the results of preliminary analyses. The remainder of the results is organized around our research goals.

Preliminary Analyses

Means, standard deviations, and stability coefficients for time-varying variables are shown in Table 1. Of note, coparenting satisfaction and marital love were high, well above the midpoint of these 9-point scales, and marital conflict was relatively low. We found substantial cross-time stability in coparenting satisfaction for both parents across time, and correlations between mothers’ and fathers’ reports of coparenting satisfaction were moderate. Mothers’ and fathers’ reports of marital quality were moderately correlated.

Table 1.

Stability Coefficients and Means (and SDs) for Coparenting Satisfaction and Time-Varying Covariates for Mothers (Above Diagonal), Father (Below Diagonal), and Correlations Between Parents’ Reports of Coparenting Satisfaction (on Diagonal)

Variable	Time	Time 1 Age:10.38	Time 2 Age: 11.43	Time 3 Age: 12.55	Mother M (SD)	Father M (SD)
Coparenting satisfaction	1	.38**	.52**	.49**	7.23 (1.64)	7.46 (1.57)
	2	.67**	.33**	.57**	7.26 (1.59)	7.49 (1.47)
	3	.53**	.35**	.46**	7.16 (1.83)	7.26 (1.61)
Marital love	1	.47**	.62**	.65**	7.82 (1.18)	7.98 ( .98)
	2	.68**	.49**	.72**	7.77 (1.34)	7.87 (1.19)
	3	.64**	.69**	.61**	7.66 (1.39)	7.79 (1.24)
Marital conflict	1	.44*	.62**	.61**	4.29 (1.41)	3.99 (1.50)
	2	.61**	.35	.69**	3.95 (1.57)	3.59 (1.52)
	3	.53**	.54**	.50**	4.10 (1.55)	3.66 (1.49)

Given potential confounds we examined correlations between marital duration and offspring age and family background characteristics. At wave 1, marital duration and offspring age were significantly, but not highly correlated, r = .18, p < .01. With respect to background characteristics, marital duration was positively related to mothers’ education, r = .31, p < .01 and fathers’ education, r = .22, p < .01, family income, r = .19, p < .01, and negatively associated with family size, r = −.11, p < .05. There were no significant associations between marital duration and offspring age or parents’ work hours.

Longitudinal Changes in Coparenting Satisfaction

A preliminary series of MLMs was tested to examine the overall growth curve of coparenting satisfaction as a function of marital duration. All results are shown in Table 1. Models were tested controlling for family background characteristics (i.e., parents’ education, family income, and family size). These controls were non-significant for coparenting satisfaction, and thus, were not included in final models.

The index of time (i.e., marital duration) was centered at 15 years, the mean marital duration across couples, across all time points. Based on AIC and BIC fit statistics and the significance of variance components, a model with a random intercept and fixed quadratic term was chosen as the initial model for coparenting satisfaction. The quadratic term was significant, γ = −.007, SE = .002, p < .01, suggesting that coparenting satisfaction increased through the first decade of couples’ marriages, followed by a period of stability from about year 12 to 18, and then declined (see Figure 1). Neither the parent gender effect nor the gender x quadratic change effect was significant meaning that this pattern of change did not differ for mothers versus fathers. Further, there were no main effects or interactions involving offspring gender or biological relatedness to mothers or to fathers.

Figure 1.

Quadratic change in coparenting satisfaction as a function of marital duration

Note. Intercept and slope for change in coparenting satisfaction are not significantly different for mothers and fathers.

^†p < .10* p < .05, ** p < .01

We next tested whether offspring age moderated the pattern of change in coparenting satisfaction. This model revealed a significant quadratic term X offspring age interaction (shown in Table 3, Model A). To follow-up, we divided couples into two groups: those with offspring younger than 11.5 years (the average age of offspring in the sample across all time points) and those with offspring 11.5 years and older, and ran the analyses separately for each group. These results showed that the significant quadratic pattern for coparenting satisfaction emerged only for couples with older offspring, γ = −.01, SE = .003, p < .01; in contrast, for couples with younger offspring, there was no significant change in coparenting satisfaction, γ = .001, SE = .003, ns (Figure 2).

Figure 2.

Interaction between quadratic effect of marital duration and offspring age for change in coparenting satisfaction

^†p < .10* p < .05, ** p < .01

Links between Marital Quality and Coparenting Satisfaction

Our next step was to test whether changes in marital love and marital conflict were linked with changes in coparenting satisfaction. Findings are shown in Table 3, Model B (love) and Model C (conflict). Beginning with marital love, in line with our second hypothesis, increases in love at the within person level (Level 1) were positively related to changes in coparenting satisfaction, meaning that, beyond average levels of marital love, at times when parents experienced more love in their marriage, they also reported greater satisfaction with coparenting. At Level 2, average marital love was also positively related to coparenting satisfaction, but a significant interaction indicated that this effect differed by parent gender. Follow-up analyses revealed that fathers who reported more marital love were more satisfied with coparenting on average, γ = .84, SE = .08, p < .01. The positive link between marital love and coparenting satisfaction was significant, but not as strong for mothers, γ = .63, SE = .07, p < .01 (Figure 3a).

Figure 3.

Figure 3a. Interaction between marital love (Level 2) and parent gender for coparenting satisfaction

^†p < .10* p < .05, ** p < .01

Figure 3b. Interaction between marital conflict (Level 2) and parent gender for coparenting satisfaction

^†p < .10* p < .05, ** p < .01

Turning to marital conflict, also consistent with our hypothesis, time-varying change in marital conflict at the within-person level was negatively related to coparenting satisfaction: At times when parents experienced greater marital conflict than usual, they were less satisfied with coparenting. In addition, the average level of marital conflict at Level 2 was inversely related to coparenting satisfaction, and this link was moderated by parent gender. Similar to findings for marital love, follow-up tests showed that the negative link between marital conflict and coparenting satisfaction was significant for both parents, but was slightly stronger for fathers, γ = −.61, SE = .07, p < .01, than mothers, γ = −.43, SE = .07, p < .01 (Figure 3b).

The Moderating Role of Inter-Parental Incongruence in Childrearing Attitudes

The final step was to test the main effect and moderating role of inter-parental incongruence in childrearing attitudes for links between marital quality and coparenting satisfaction (Table 1, Model B and Model C). There was no main effect of either parents’ own childrearing attitudes or incongruent attitudes for coparenting satisfaction. Controlling for parents’ individual attitudes toward childrearing, inter-parental incongruence moderated the links between marital love (at Level 1) and coparenting satisfaction, but did not moderate marital conflict – coparenting associations. Follow up analyses revealed that the time-varying association between marital love and coparenting satisfaction was positive, significant, and stronger for parents who were more different in their childrearing attitudes, γ = .89, SE = .11, p < .01, compared to parents whose attitudes were more similar, γ = .49, SE = .04, p < .01 (Figure 4).

Figure 4.

Interaction between marital love (Level 1) and incongruent attitudes towards childrearing for coparenting satisfaction

^†p < .10* p < .05, ** p < .01

Discussion

A growing literature has established that coparenting is a fundamental family systems dynamic that is affected by developmental transitions within the family. The findings from this study are consistent with past work and suggest that coparenting satisfaction changes across the first two decades of marriage, but that the trajectory of change is dependent on the age of adolescent offspring. Findings from this study also confirmed our hypothesis that changes in marital love and conflict correspond to changes in coparenting satisfaction across time, and that the association between marital love and coparenting satisfaction was especially strong for couples who held more incongruent attitudes towards childrearing. This study extends past research on coparenting in two important ways: By examining change in coparenting satisfaction across later years of development, and by examining these changes in a sample of married African American mothers and fathers of adolescents, a population that is largely underrepresented in research on normative family dynamics. Below, we review the findings in greater detail, highlighting the ways in which this study advances understanding of coparenting satisfaction during later phases of family life, and family systems dynamics, more generally.

Longitudinal Change in Coparenting Satisfaction

Coparenting satisfaction increased through the first decade of marriage, peaking at around 10–12 years, leveled off, and then declined in the second half of the second decade of marriage. This effect, however, was qualified by offspring age. In families of youth in mid-to-late adolescence, the inverted-U pattern of change was significant; in contrast, there was no significant change in coparenting satisfaction in families of young adolescents. Couples in relatively shorter marriages (i.e. married fewer than 15 years) but who were coparenting an older adolescent became increasingly satisfied in the first decade of their marriage. This pattern may mean that learning to work together for the first time during a challenging period in offspring’s development promoted more positive coparenting, and in turn, coparenting satisfaction, among these new coparents. For parents with a longer marital duration who presumably had established coparenting dynamics in the early stages of offspring’s development, declines in coparenting satisfaction seemed to coincide with offspring’s adolescence. As a time in development when youth begin to assert their autonomy and play a more salient role in family dynamics (Steinberg, 2001), these findings suggest that the challenges associated with adolescence interfered with established coparenting dynamics.

In contrast to parents of older adolescents, coparents of pre-adolescents were more satisfied overall and experienced no significant change in coparenting satisfaction across the years of marriage that we examined. One possible explanation is simply that their relatively shorter marriages may mean that coparenting dynamics are less well-established among this group and thus that the impact of offspring’s development was not as salient. A second explanation is grounded in the fact that parents of younger offspring in this sample had significantly smaller households, relative to parents of older offspring. Larger family size is a strain on family economic and social resources (Downey, 1995), so the potential stressors of parenting an adolescent may have been reduced among coparents in this group. As noted, however, family size was not significant as a control variable, making this explanation less likely.

Taken together, such findings lend support to the idea that the challenges adolescence imposes on mothers’ and fathers’ coparenting may depend, in part, on the length of parents’ relationship and family structure characteristics, such as household size, that reduce or exacerbate family stresses. More generally, these findings illustrate the family systems idea that offspring’s development reverberated in other family subsystems. Future research should consider the developmental status of youth for other family systems processes, including marital and sibling relationships. It should also be noted that, inconsistent with past research showing that fathers are more satisfied than mothers with their coparenting relationship, we found no parent gender differences in either average level or change in coparenting satisfaction. Further research is needed to understand whether gender differences in coparenting become less evident in families of adolescents, or whether African American mothers and fathers are more similar in their evaluations of coparenting relative to more typically studied European American coparents.

Marital Quality, Incongruent Childrearing Attitudes, and Coparenting Satisfaction

Consistent with expectations regarding emotional spillover, changes in marital quality were associated with corresponding changes in coparenting satisfaction, and there were significant mother-father differences in these links. Marital love was positively related and marital conflict, negatively related to coparenting satisfaction for mothers and fathers, and as predicted, links were somewhat stronger for fathers’ evaluations of coparenting satisfaction. The salience of marital quality for coparenting satisfaction of African American men may have important applied implications. These findings corroborate past research with White families, showing that fathers’ parenting is more susceptible than mothers’ to problems in the marriage (Kolak & Volling, 2007) as well as to positive marital experiences (Coiro & Emery, 1998; Kerig, Cowan, & Cowan, 1993). Mothers’ family roles have been described as more scripted than fathers (Brody, Arias, & Fincham, 1996), which may mean that mothers are less susceptible to negative emotional spillover from other family relationships. However it should be noted that, although significantly different, linkages were evident for both parents.

Our hypothesis regarding the role of incongruent childrearing attitudes as an additional risk factor for coparenting satisfaction was partially supported. Consistent with a cumulative risk model, couples who had incongruent attitudes and reported lower levels of marital love reported the lowest level of satisfaction with coparenting, suggesting that high levels of love are protective for couples who differ in their childrearing attitudes. In contrast to our expectations, however, marital conflict was negatively linked to coparenting satisfaction regardless of the degree of similarity in parents’ attitudes. Taken together, these findings are consistent with the idea that, in the face of specific areas of disagreement, marital love can promote satisfaction (Bradbury et al., 2000). They show further, however, that experiencing areas of agreement is insufficient to overcome more general negativity in the marriage.

Strengths, Limitations, and Study Implications

Extending what is known about coparenting, the findings from this study lend support to the idea that coparenting satisfaction is not a static family process, but continues to develop over the course of marriage. Further, changes in coparenting satisfaction appear to depend, in part, on offspring’s developmental status. Indeed, a key strength of this study was our ability to study the roles of both marital duration and offspring’s age in longitudinal changes in coparenting satisfaction. Taken together, findings from this study take an important first step in documenting the development of coparenting through later phases of family life.

Although this research advanced understanding of coparenting, it is not without limitations. Like most studies on coparenting evaluations (e.g., Margolin et al., 2001) we relied on parents’ reports of their coparenting satisfaction and marital quality, which may be subject to self-reporter biases. Further, even though there was a range in income and education, this sample was relatively advantaged and came from a circumscribed geographic location, meaning that these findings are not generalizable to the larger population of African American married parents. In addition, coparenting was not measured with reference to a focal child. Just as parenting is not the same for all children in the family, it will be important to understand differences in mothers’ and fathers’ coparenting evaluations vis-à-vis each child in the family. Finally, our findings supported a family systems perspective that family subsystems are inter-related and family processes do not occur in isolation. Thus, it is important to recognize that other family dynamics, including parent-child relationships, may help to explain changes in coparenting satisfaction. Examining other time-varying family relationship processes as correlates of coparenting is an important step for future research.

Despite these limitations, this study makes an important contribution to the literature on family systems processes. Further, our findings have important clinical implications. For one, practitioners should be alerted to the possibility that changes in coparenting satisfaction depend on the intersection of marital duration and age of offspring. Indeed, having older adolescent offspring may be a stressor for coparenting in newer marriages. Further, the idea that discrepant gender role attitudes are a greater risk factor in marriages characterized by lower levels of love may be important in the development of targeted intervention programs. Together, findings advance our understanding of families in showing that changes in coparenting and adolescent development do not occur in a vacuum, but influence qualities of other family dynamics. This study also speaks more generally to a growing body of work on normative family relationship processes within African American families. In sum, results highlight the significance of examining coparenting satisfaction across a broad swath of couples’ marriages while also taking other family dynamics into consideration.

Table 2.

Coefficients for change in coparenting satisfaction as a function of marital love and conflict, and moderated by incongruent attitudes towards childrearing.

	Unconditional Model		Model A: Longitudinal Change		Model B: Marital Love		Model C: Marital Conflict
Fixed effects	γ	SE	γ	SE	γ	SE	γ	SE
Intercept	7.51**	.12	7.34**	.15	7.38**	.12	7.40**	.12
Level 1:
Marital duration (Linear)	−.005	.02	.002	.02	.02	.02	.01	.02
Marital duration (Quadratic)	−.006*	.002	−.005	.003	−.003	.003	−.002	.002
Youth age			.05	.06	.02	.05	−.09*	.04
Youth age X Linear			−.01	.01	−.01	.01	−.02*	.01
Youth age X Quadratic			−.003*	.001	−.003†	.001
Love					.69**	.06
Conflict							−.34**	.05
Level 2:
Parent gender					−.10	.10	.01	.12
Average love					.84**	.08
Average conflict							−.58**	.07
Love X Parent gender					−.17*	.09
Conflict X Parent gender							.19*	.09
Childrearing attitudes					.01	.02
Level 3:
Youth gender			.15	.19	.09	.16	.09	.16
Incongruent attitudes					−.02	.03
Cross-Level Interaction:
Incongruent attitudes X L1 Love					.06**	.03
Random effects
Level 3 intercept:	.89**		.87 **		.47**		.46**
Level 2 intercept:	.56**		.55**		.33**		.50**
Residual	1.28**		1.26 **		1.00**		1.12 **

Note.

^aFathers are the reference group.

^†p < .10

^*p < .05

^**p < .01