Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Jun 7.
Published in final edited form as: Fam Process. 2010 Mar;49(1):59–73. doi: 10.1111/j.1545-5300.2010.01308.x

Pathways Between Marriage and Parenting for Wives and Husbands: The Role of Coparenting1

Melinda Morrill 1
PMCID: PMC4896491  NIHMSID: NIHMS777711  PMID: 20377635

Abstract

As family systems research has expanded, so have investigations into how marital partners coparent together. Although coparenting research has increasingly found support for the influential role of coparenting on both marital relationships and parenting practices, coparenting has traditionally been investigated as part of an indirect system which begins with marital health, is mediated by coparenting processes, and then culminates in each partner's parenting. The field has not tested how this traditional model compares to the equally plausible alternative model in which coparenting simultaneously predicts both marital relationships and parenting practices. Furthermore, statistical and practical limitations have typically resulted in only one parent being analyzed in these models. This study used model-fitting analyses to include both wives and husbands in a test of these two alternative models of the role of coparenting in the family system. Our data suggested that both the traditional indirect model (marital health to coparenting to parenting practices), and the alternative predictor model where coparenting alliance directly and simultaneously predicts marital health and parenting practices, fit for both spouses. This suggests that dynamic and multiple roles may be played by coparenting in the overall family system, and raises important practical implications for family clinicians.

One of the most influential processes that has emerged amidst recent systemic studies of marriages and parenting is coparenting (Feinberg, 2003; McHale, 1995). Coparenting has been defined as a unique component of the marital relationship in which parents work together, or, alternatively, struggle against each other when it comes to child rearing (McHale, 2007). Although coparenting was first discussed in the context of postdivorce couples continuing to coordinate their parenting (Ahrons, 1981), more recently it has been recognized as also playing a prominent role in intact two-parent family systems (Margolin, Gordis, & John, 2001). Coparenting is conceptualized as a broad construct composed of several factors including parenting alliance and support, antagonism and undermining between parents, division of childcare labor, the extent to which each partner is engaged in the day-to-day organizing of the lives of their children, and triangulation (Margolin et al., 2001; McHale, 2007). Several of these aspects of coparenting have been investigated separately due to their unique characteristics. For example, division of childcare labor has been studied in association with gender and power inequalities, as findings reveal that egalitarian ideals are often not being realized in heterosexual romantic relationships (Knudson-Martin & Mahoney, 2009). Another aspect of coparenting, coparenting alliance, has been one of the most widely used operationalizations of coparenting when examining how family subsytems influence each other, as a couple's alliance has been shown to be strongly linked to parent-child relationships (Bonds & Gondoli, 2007; Feinberg & Kan, 2008; Floyd, Gilliom, & Costigan, 1998). Weissman and Cohen (1985) listed the following four characteristics of a sound coparenting alliance: (1) Both parents’ investment in the child; (2) Valuing each other's involvement with the child; (3) Respect for each other's judgment about child rearing; and (4) Having a desire to communicate child related information. Each of these elements of coparenting alliance demonstrates the interconnectedness of coparenting to both the marriage and each partner's parenting.

Furthermore, studies of coparenting gain in importance as researchers increasingly focus on both mothers’ and fathers’ roles in the family system (Feinberg, 2002). Although family systems theorists argue for the need for inclusion of both wives and husbands in research with heterosexual parents, many previous studies have noted the lack of data from both spouses in family research (Bonds & Gondoli, 2007; Kolak & Volling, 2007; Margolin et al., 1996). Reasons for this paradox include the logistical challenges of gathering data from both partners, and consistent differences that have been found between wives and husbands, making it difficult to study both concurrently. For example, previous research has found that fathers display more marriage-related disruptions in parenting than do mothers (Belsky, 1990; Kitzmann, 2000; Kolak & Volling, 2007). One theory explaining this is that it may be more difficult for men to maintain interpersonal boundaries between the marital and parent-child subsystems, especially when stress upsets the balance between family relationships (Minuchin, 1974). Alternatively, women's parenting role may not be viewed as “optional” in our society, as mothers have greater expectations to continue their responsibilities as parents regardless of the functioning of their marital relationship. Nonetheless, when it comes to family research and interventions, it is not always practical to consider mothers and fathers separately. Furthermore, overall family processes could be functioning similarly for both fathers and mothers. Indeed, some recent research suggests that the influence of gender on parenting roles is decreasing in importance (Coltrane & Adams, 2008). Therefore, it remains vitally important to include both spouses in studies of the family system, especially given the implications for translational couple research.

The integral role of coparenting in families made up of fathers, mothers and children has been strongly supported by empirical findings from recent research (Bonds & Gondoli, 2007; Feinberg, Kan, & Hetherington, 2007; Floyd et al., 1998; Margolin et al., 2001). One of the most consistent findings in coparenting research is the positive association between marital health and coparenting quality (Kolak & Volling, 2007; Lindahl, Clements & Markman, 1997; McHale, 2007). Certain aspects of the marital relationship appear to be particularly influential on coparenting relationships, such as collaboration, team-work, and family warmth (McHale et al., 2004). In a bidirectional fashion, coparenting is also hypothesized to influence the marital relationship through actions like spousal social support, which is strongly associated with marital health (Feinberg, 2002, 2003). Furthermore, coparenting relationships have been found to strongly influence parenting behavior (Bonds & Gondoli, 2007; Cowan & Cowan, 2005; Feinberg et al., 2007; Floyd et al., 1998; Margolin et al., 2001). Coparenting cohesion is thought to equip parents with resources to parent effectively and to coordinate their parenting roles in ways that benefit their children (Behenke et al., 2008; Feinberg, 2002; Gable, Belsky & Crnic, 1995). In contrast, couples with a weak coparenting alliance are more vulnerable to hostility and competition which can lead to divergent expectations for their children, disagreement about their children's caretaking, and tension when assisting their children (Kitzmann, 2000).

Significantly, the growing data about coparenting consistently indicate that the coparental relationship is more proximally related to both marital quality and parent-child relationships than marriages and parenting are to each other (Abidin & Brunner, 1995; Feinberg et al., 2007). Several recent studies have found that any direct effect between marriages and parenting decreases or disappears after accounting for the mediating effect of coparenting (Bonds & Gondoli, 2007; Floyd et al., 1998; Margolin et al., 2001). The widely used concept of “domain specificity” (Grych & Fincham, 1993) provides one explanation for this greater proximity of coparenting alliance to parenting: Marital interactions concerning child rearing have a greater impact on parenting than other aspects of the marital relationship.

Until recently, coparenting has traditionally been empirically investigated as an intermediary mechanism in an indirect family system that begins with the marital relationship and concludes with parenting practices (Bonds & Gondoli, 2007; Margolin et al., 2001; McHale et al., 2004; but see, for exception, Feinberg et al., 2007). This time-ordered direction of effects makes theoretical sense given that marital relationships are typically formed first, and so are conceptualized as the “executive subsystem” in the family (Minuchin, 1985). This ordering suggests that partners’ identification as a married couple would influence their interactions as coparents and then as parents. However, an alternative perspective is that once married partners become established as parents and coparents, coparenting takes over as the direct “driver” of the marital and parenting subsystems for some families. This new model follows from recent findings about the proximal and pivotal role of coparenting and would predict that the way parents perceive their coparenting directly influences both the well-being of their marriages and the effectiveness of their parenting practices. Although coparenting has been hypothesized to have a reciprocal, bidirectional interaction with marital functioning (Feinberg, 2002), the model where coparenting simultaneously predicts both marital health and parenting practices had not been empirically tested.

Previous studies have paved the way for this question quite explicitly. Erel and Burman (1995) hypothesized that “the link between the marital and the parent-child relationship may be due to the impact of the marriage on the parent-child relations, of parent-child relations on the marriage, or of a third factor on both the marital and parent child relationship” (p.128). Feinberg et al. (2007) later implicated coparenting as a potentially central predictor:

Research on coparenting has demonstrated to date that coparenting is more predictive of parenting and child outcomes than is general marital quality, that coparenting accounts for variance in parenting and child outcome after controlling for individual parent characteristics, and that coparenting predicts marital quality but not vice versa (p.687).

However, to our knowledge, no studies have empirically tested whether this predictor model in which coparenting alliance directly and simultaneously predicts marital relationships and parenting practices and the traditional indirect model (marital relationship to coparenting to parenting practices) could both be well-fitting models for mothers and fathers. The goal of this study was to test both the traditional indirect model and the alternative predictor model using an analytic strategy that allows both wives and husbands to be considered simultaneously. If both models fit, this would add empirical data supporting the flexible and multiple roles that coparenting can play in overall family systems.

METHOD

Participants

This study investigated 76 married, heterosexual parents (152 individuals) who were part of a longitudinal, randomized study of marital relationships, the Marriage Checkup (Córdova et al., 2008). The Marriage Checkup (MC) is a two-session, preventative marital intervention that utilizes integrative couple therapy and motivational interviewing techniques with couples at-risk for marital deterioration. Couples were assessed approximately one month prior to receiving the MC intervention and the pre-treatment data from couples raising children under 18 years old were included in this study.

Couples were recruited by means of advertisements and articles in local newspapers as well as online communities. The mean age of the mothers and fathers in this sample was 42.63 years old (SD = 6.74) and 45.00 years (SD = 7.84) respectively; the average age of the children included in the study was 10.16 years (SD = 4.73). The average length of marriage was 12.71 years (SD = 10.53). The modal total number of children per couple was two. This was a highly educated sample, with 67.8% of the sample having completed a college or graduate education, 22.3% having completed some college, and 9.9% with a high school diploma or less. Likewise, the sample was of a moderate to high socioeconomic status, with 44.1% having a household income over $100,000, 42.1% from $50,000-99,999 and 11.9% with an income less than $49,999 (three participants, 1.9%, did not provide this data). Racially, 92.8% of the sample was White; 5.3% Black and 2.0% Asian.

Measures

Participants in the study were administered measures of marital quality, coparenting alliance, and parenting practices as part of a larger battery of questionnaires assessing their marital health.

Marital Quality

Marital quality was measured through the Quality of Marriage Index (QMI; Norton, 1983). The QMI is a 6-item comprehensive self-report measure where participants respond to each item on a 7-point Likert scale ranging from very strongly disagree (1) to very strongly agree (7). Sample items include, “My relationship with my partner is very stable” and “My relationship with my partner makes me happy.” High scores indicate overall satisfaction with one's marriage. In the current study, internal consistency reliability for the QMI was very high (Cronbach's alpha = .98).

Coparenting

Coparenting alliance was assessed using the Parenting Alliance Measure (PAM; Abidin & Konold, 1999), a 20-item self-report instrument used for parents of children from 1-19 years old. A slightly altered version of the questionnaire was used for this study, in that one item was omitted prior to beginning the study due to inapplicability. Mothers and fathers responded to the items on the PAM using a 5-point Likert scale, ranging from strongly disagree (1) to strongly agree (5), with higher scores reflecting stronger coparental alliance. The PAM is based on Weissman and Cohen's (1985) four characteristics of a sound parenting alliance described above. This measure has been normed over large samples and has been found to measure the same factors for fathers and mothers (Konold & Abidin, 2001). Sample items include, “Talking to my child's other parent about our child is something I look forward to” and “My child's other parent and I are a good team.” The PAM has been found to have a high degree of internal consistency, with alphas of .97 for mothers, .96 for fathers, and .97 for a combined sample (Abidin & Konold, 1999). Our 19-item version also had a high degree of internal consistency in this study (Cronbach's alpha = .95).

Parenting Practices

To measure parenting practices, the parent self-report version of the Alabama Parenting Questionnaire was used (APQ; Frick, 1991). The validity and reliability of the APQ has been established in both clinic and community samples (Shelton et al., 1996), and it has also been shown to have moderate to high internal consistency (Dadds et al., 2003; Hawes & Dadds 2006). The APQ consists of five subscales corresponding to different dimensions of parenting: (1) Positive parenting, (2) Parental involvement, (3) Corporal punishment, (4) Inconsistent discipline and (5) Poor parental monitoring and supervision. To avoid a negative bias toward the corporal punishment items, seven items measured discipline practices other than corporal punishment (Shelton et al., 1996). Respondents rate 42 items on a 5-point scale (Never, Almost Never, Sometimes, Often, Always). Negatively worded items are reverse scored and high scores indicate more effective parenting practices. Items include, “You have a friendly talk with your child” and “Your child is not punished when he/she has done something wrong.”

The APQ measure had high reliability (Cronbach's alpha = .83) in this study, although this was only possible to measure for the 41.4% of participants who completed all of the items (some of the questions were not applicable to children of different age ranges). After using mean replacement for the missing items that were not applicable to children of all ages, the APQ measure continued to display high reliability (Cronbach's alpha = .85).

Data Analysis Plan

A barrier to the inclusion of both sexes in empirical studies of couples is the inherent statistical difficulties of analyzing nonindependent data from members of the same couple (Kenny, Kashy, & Cook, 2006). Analyses of nonindependent data is complicated since a key goal of statistical analysis is independent replication. Ignoring nonindependence leads to statistical problems such as biased significance tests. Additionally, the practice of treating husbands and wives as if they are two different samples presumes differences between the sexes and can result in a loss of power by not combining the results (Kenny et al., 2006). Yet the interpersonal nature of relationships necessitates theories, research methods and data analyses that can accommodate the complexities of family functioning using models that include both parents.

Structural equation modeling (SEM) is a statistical analysis technique that can assess indirect effects between associated variables and is particularly useful in the analysis of dyadic data (Kenny et al., 2006; Meyers, Gamst, & Guarino, 2006). SEM with observed variables is a model fitting strategy that can be applied to path analysis to assess the impact of variables on other variables (Meyers et al., 2006). The process of model fitting can generate estimates of path parameters through the maximum likelihood technique, an iterative process that simultaneously analyzes all the variables and generates coefficients estimating the likelihood of obtaining the observed data based on the hypothesized model. Model fitting also includes error variances for the dependent variables, providing a structural model that includes both the predicted variance and the error variance (Meyers et al., 2006). Significance testing and model fit indexes are then calculated to determine whether the collected data fits the hypothesized model. The adequate fit of a model is most persuasive when alternative models compared against it are shown to not fit as well (Thompson, 2000).

Using modeling to conduct path analyses for nonindependent, distinguishable dyads is particularly beneficial when analyzing data from both partners of married couples. Parallel path models are created for each member of the dyad and correlations across members are inserted to allow for covariance and nonindependence. Therefore, model fitting can be used to effectively and simply conduct a path analysis for both husbands and wives concurrently (Kenny et al., 2006).

In this study, the hypothesized models of the associations between marital quality, coparenting alliance, and parenting practices for both husbands and wives were evaluated using AMOS 16.0. The traditional chi-square statistic was evaluated for statistical significance; a non-significant result indicates a good fit and is considered an adequate measure of fit for models with a relatively small number of cases (Meyers et al., 2006). Additionally, we examined other highly recommended fit indexes including the comparative fit index (CFI), root mean square error of approximation (RMSEA) and chi square over degrees of freedom ratio (χ2 / df) (Bui, Peplau, & Hill, 1996; Du Rocher Schudlich & Cummings, 2007). Researchers advise that the CFI should be the fit statistic of choice in SEM research, especially when using relatively small sample sizes (Meyers et al., 2006). A CFI of .95 or above indicates a good fit and below .90 indicates a poor fit. Additionally, Byrne (1998) described the RMSEA as “one of the most informative criteria in covariance structure modeling” (p.84). An RMSEA of less than .08 indicates a good fit and between .08 and .10 indicates a moderate fit. The χ2 / df ratio was also examined to adjust for model complexity, with values between 1-3 indicating an acceptable fit (Du Rocher Schudlich & Cummings, 2007). Finally, the path coefficients were assessed for statistical significance at p < .001 or p < .01. Therefore, the overall adequacy of the fit of the models in this study was assessed using a combination of traditional significance testing and data from absolute, relative and parsimonious fit measures.

RESULTS

Correlations

As a first step in investigating the associations between the marital relationships, coparenting alliances, and parenting practices of these couples, the Pearson product-moment correlations for these variables were computed along with the means and standard deviations. As shown in Table 1, the means for the marital quality, coparenting and parenting practices indicate that this sample is reporting a fairly high level of marital satisfaction, coparenting alliance, and effective parenting practices. Looking at the associations between these variables, a small to medium-sized significant positive correlation was found for both husbands and wives between their marital quality (QMI) and their coparenting alliance (PAM). This supports prior conceptualizations of coparenting alliance as a significantly related, but still distinct, construct from the marital relationship (Feinberg, 2003). Additionally, wives’ and husbands’ coparenting alliance (PAM) ratings were significantly positively correlated with their own parenting practices (APQ), also what was expected given the previous literature (Margolin et al., 2001). The more robust correlations for each partner between their own coparenting alliance and marital quality, and their coparenting alliance and parenting practices, fit with the conceptualizations of the more proximal role of coparenting to each of these subsystems than they are to each other. Additionally, although both husbands’ and wives’ coparenting is significantly associated with both their marital satisfaction and parenting practices, these variables appear to be more strongly associated for husbands than for wives. This also supports the research discussed above that fathers’ perception of the quality of their marriage and their coparental alliance contributes more to their coparenting and parenting, respectively, than does mothers’. On the other hand, neither husbands’ nor wives’ marital quality was correlated with their parenting practices, consistent with our expectation that the indirect effects through coparenting would be more robust than the direct effects from the marriage to parenting. These results demonstrate the utility of testing the overall models, as a systems conceptualization would expect an inclusive model to most accurately reflect the functioning of the entire family system.

Table 1.

Intercorrelations, Means, and Standard Deviations for Scores on Marital Quality, Coparenting and Parenting Practices for Wives and Husbands

Measure H-QMI W-PAM H-PAM W-APQ H-APQ M SD
Wives marital quality (W-QMI) .51** .28* .26* .18 .10 5.48 1.54
Husbands marital quality (H-QMI) .17 .46** −.09 .22 5.51 1.58
Wives coparenting (W-PAM) .45** .36** .38** 4.10 .75
Husbands coparenting (H-PAM) .27* .52** 4.27 .70
Wives parenting (W-APQ) .37** 3.97 .28
Husbands parenting (H-APQ) 3.85 .37
Open in a new tab

Note. W=Wives; H=Husbands; QMI = Quality of Marriage Index; PAM = Parenting Alliance Measure; APQ = Alabama Parenting Questionnaire.

N=76 couples (152 participants).

*

p <.05

**

p < .01.

Path Analysis Using Model Fitting

Indirect Path

Figure 1 presents the results of the traditional indirect path model where marital quality predicts coparenting alliance which then predicts parenting practices. The chi-square test was not significant, χ2 (8, N = 76) = 12.52, p > .05, and the results yielded acceptably high goodness-of-fit indexes, indicating that the hypothesized model fit the observed data (see Table 2). The CFI yielded a .96 and the RMSEA value was .09, indicating a moderate fit of the model. All the path coefficients demonstrated both statistical significance (p < .01) and practical significance (β ≥ .3). Marital quality accounted for 9% of the variance of wives’ coparenting alliance and 23% of husbands’. Coparenting alliance accounted for 28% of the variance of husbands’ parenting practices and 14% of wives’. The χ2 / df ratio was 1.56. The model for wives and the model for husbands were found to differ when the paths were constrained to be equal, so the two models were left free to vary (Kenny et al., 2006). The alternative models of all other possible orders of the variables were poor fits to the data, as shown in Table 2. The results, therefore, supported the model that marital quality predicted coparenting alliance which in turn predicted parenting practices for both spouses.

Figure1.

Figure1

Open in a new tab

Standardized path coefficients for the traditional indirect path model from marital quality to coparenting alliance to parenting practices for wives and husbands. *p < .01, ** p < .001.

Table 2.

Fit Statistics for Figure 1 – Indirect Path – and Alternative Models

Model df χ 2 χ2/df CFI RMSEA
Model 1: Marriage to Coparenting to Parenting (QMI-PAM-APQ) 8 12.52 1.56 .96 .09
Alternative Model 1: QMI-APQ-PAM 8 31.00*** 3.87 .79 .20
Alternative Model 2: APQ-PAM-QMI 8 16.95* 2.12 .92 .12
Alternative Model 3: APQ-QMI-PAM 8 33.98*** 4.25 .76 .21
Alternative Model 4: PAM-QMI-APQ 8 36.25*** 4.53 .74 .22
Alternative Model 5: PAM-APQ-QMI 8 26.88** 3.36 .83 .18
Open in a new tab

Note. CFI = comparative fit index; RMSEA = root mean square error of approximation.

*

p < .05

**

p < .01

***

p < .001.

Predictor Model

The alternative model was then tested to investigate whether coparenting simultaneously predicted marital quality and parenting practices for wives and husbands (see Figure 2). The model also had a nonsignificant chi-square, χ2 (8, N = 76) = 13.83, p > .05, and the results also yielded acceptable goodness-of-fit indexes (see Table 3), with a CFI = .95 and an RMSEA = .10, indicating a moderate fit to the data. All of the path coefficients demonstrated both statistical significance (p < .01) and practical significance (β > .3). In this model, coparenting accounted for 20% of the variance of husbands’ marital health and 9% of wives’. Coparenting also accounted for 14% of the variance in wives’ parenting practices and 29% of husbands’. The χ2 / df ratio was 1.73, indicating adequate complexity of the model. Again, the model for wives and the model for husbands were found to differ when the paths were constrained to be equal, so the model where the path coefficients were free to vary was used. Alternative models of all other possible orders of the variables were poor fits to our data, as shown in Table 3. These findings also supported the model that coparenting alliance simultaneously predicts marital health and parenting practices for wives and husbands.

Figure 2.

Figure 2

Open in a new tab

Standardized coefficients for the alternative predictor path model from coparenting to marital quality and parenting practices simultaneously for wives and husbands. ** p < .01.

Table 3.

Fit Statistics for Figure 2 – Predictor Path – and Alternative Models

Model df χ 2 χ2/df CFI RMSEA
Model 1: Coparenting to Marriage and Parenting (PAM-QMI + PAM-APQ) 8 13.83 1.73 .95 .10
Alternative Model 1: QMI-PAM + QMI-APQ 8 34.94** 4.37 .75 .21
Alternative Model 2: APQ-PAM + APQ-QMI 8 30.01** 3.75 .80 .19
Open in a new tab

Note. CFI = comparative fit index; RMSEA = root mean square error of approximation.

**

p < .01

DISCUSSION

Weissman and Cohen (1985) described coparenting as “the center around which the family process evolves” (p.24). This study empirically tested two different models in which coparenting serves as “centers” of the family process for both spouses: Either as part of an indirect chain beginning with the marital relationship and concluding at parenting practices, or as a simultaneous driver of both marital relationships and parenting practices. Our findings suggest that both models fit the data equally well.

The results from the indirect path model support the hypothesis that marital quality affects coparenting alliance which in turn affects parenting practices for fathers and mothers. Bolstering this finding, all five alternative orders of indirect models of the variables were poor fits to the data. It is easy to imagine a real-world situation that this indirect model describes. Consider a scenario that begins with a married couple spending time together in the living room enjoying each other's company and feeling emotionally connected. In this context we have the marital subsystem working particularly well. The context shifts, however, as their child enters the room needing help with his or her homework. At this point, the coparenting subsystem is activated, with the positive affective and collaborative qualities of the marital subsystem influencing or “spilling over” into the coparental subsystem. The couple's positive affect and collaborative behaviors enable them to easily coordinate their assistance to the child, either individually or as a team. Emerging almost immediately is the parenting subsystem as the relationship between the child and each parent is played out around the interaction the child experiences in that moment from each parent. Their ability to coordinate and support each other enables them to empathically respond to their child, to spend positive time with their child, and to sensitively intuit his or her developmental needs. On the other hand, if the couple started off in that living room feeling distant, resentful or conflicted, the inquiring child may instead be confronted with continuing tension between the parents or competition about how best to complete the homework. One parent may leave the room, while the other may be distracted or tersely blurt out the answer to the child. Thus, in this model, each domain spills over into the subsequent domain, positively or negatively, in a “chained” fashion.

Our findings also supported the alternative model in which coparenting alliance directly predicted both marital quality and parenting practices simultaneously for wives and husbands. Further supporting this, the other predictor models in which marital quality or parenting practices predicted the other variables were poor fits, suggesting that coparenting but not marital quality nor parenting practices acted in this driving, simultaneous-predictor role. To picture this, consider another imagined family with both parents and their two children together in the living room. The wife wants to help one child with her homework, but the husband does not wish to give the other child a bath, and they argue. This ineffective coparenting decreases the quality of each child's experience with his or her parent; the father may angrily leave the room feeling guilty, the mother may make a spiteful comment to the daughter about her dad. At the same time, neither parent is feeling positive about or intimately connected to their spouse, an emotional state that will likely persist through the evening. On the other hand, if these parents had started off collaborating and supporting each other throughout the evening with their children, it is easy to imagine a much different experience with each child, as well as greater feelings of satisfaction in their marital relationship. In this model, coparenting simultaneously predicts each domain.

The result that both models fit the data equally well has important theoretical implications. The first consideration is how both models could represent the functioning of real fathers and mothers. We propose that there are at least three possible ways in which both models could explain the family functioning of our couples. First, it may be that each of the types of scenarios described above plays out repeatedly within all families depending on whether any particular interaction begins with the marital or coparenting subsystem. Literally which scenario comes first temporally may either start off the indirect chain from the marital relationship or may activate coparenting to begin simultaneously influencing the other systems. In other words, it could be that all couples live out both models at different times and the couples in our sample represent the naturally occurring mix of these moments. Another possibility could be that some couples with children, by virtue of the way they prioritize and conceptualize their day-to-day lives as a family, may experience themselves most frequently as coparents and only secondarily as an intimate couple. For other couples, they may primarily self-identify as an intimate couple first and secondarily as coparents. Thus, in our sample, we may have represented both “types” of couples, with the indirect model fitting best for some couples and the alternative “predictor” model fitting best for other couples. A third explanation could be linked to the developmental stage of the child or children in the family: Couples may be more “marriage centered” or “parenting centered” at different periods in their child's development. As our sample represented children aged 1-18 years old, our findings may reflect the variable functioning during those diverse age ranges. Further investigation of these hypotheses would be needed to increase our understanding of how coparenting is functioning in such a way that both models accurately map the territory of family life.

The finding that both models fit the data also has several important practical implications. Family clinicians often treat couples with children who are either distressed in their marital commitment, their coparental relationship, or in their parenting investment, which is often reinforced by powerful family dynamics. After assessing the family's strengths and weaknesses, knowledge of these different models could provide useful direction about which subsystem to target. For example, if the couple is primarily struggling with their parenting practices, it may be more effective to focus on their coparental cohesion (in addition to parenting training) than it would be to treat their marital relationship. Alternatively, if each partner is parenting effectively as individuals but they are struggling in their coparental synchronization, it may be more effective to focus on their marital functioning (in addition to coparenting interventions).

Furthermore, the viability of the alternative predictor model for at least some couples suggests that targeting couple's’ coparenting alliance could be an efficient and effective tool to influence both their romantic relationship and parenting practices. In other words, coparenting interventions could have the potential to contribute to both marital satisfaction and to each partner's parenting practices, in effect “killing two birds with one stone.” Prior research has demonstrated that coparenting alliance is indeed a malleable construct, making such interventions feasible and practical (Feinberg & Kan, 2008). Additionally, this enhanced utility of coparenting interventions could benefit highly conflicted couples who may be more willing to receive help with their coparenting for the good of the children than they would be to seek marital interventions (Margolin et al., 2001). On the other hand, this amplified influence of coparenting underscores the risks of leaving ineffective coparenting unaddressed, as coparenting dynamics have been shown to remain remarkably stable over time without intervention (McHale et al., 2004). Given the systems focus of the family psychology field, future family interventions like coparenting treatments may increasingly be developed to address multiple subsystems simultaneously.

Future studies could aim to address some of the limitations in this study. First, as this sample is quite homogeneous racially and socioeconomically, and reported being fairly satisfied in each of the family domains, it is important to investigate these effects in parents from more diverse and more highly distressed populations. This is especially crucial given that both socioeconomic and relational stress have been found to negatively influence coparental cohesion and parenting practices (Behnke et al., 2008). Additionally, the nature of this cross-sectional data cannot explicitly identify directional effects or causal links, while longitudinal data could provide stronger evidence of directionality or causality. A larger sample size would also permit model fitting using latent variables, allowing for more precision. Furthermore, it is of note that husbands and wives reported their marital quality, coparenting and parenting practices differently enough that we could not constrain the parameter paths to equal each other in the models. Although we believe that a family-systems approach benefits from analytical methods such as these that incorporate both partners simultaneously, it undoubtedly remains important to also investigate gender differences. For instance, our finding that fathers’ family subsystems were more highly correlated and accounted for more of the variance in their other subsystems than mothers’ implies that gender differences are at play in these processes. Further research could gather additional information about these gender differences in addition to identifying characteristics of couples who fit into the traditional model or the newer, alternative model.

Family systems research has taken many historical twists and turns and continues to evolve. This is the first study to our knowledge that compares two different arrangements of path models of the interrelationships between marital quality, coparenting alliance and parenting practices side-by-side using model fitting. The fact that both models were viable demonstrates the need for new conceptualizations of the role of the coparenting subsystem to continue expanding our understanding of families. Researching the many roles of the coparenting process for fathers and mothers has theoretical and clinical importance that could contribute to the progress of family research.

Footnotes

1

The project described was supported by Award Number R01HD045281 from the Eunice Kennedy Shriver National Institute Of Child Health & Human Development. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Eunice Kennedy Shriver National Institute Of Child Health & Human Development or the National Institutes of Health.

REFERENCES

  1. Abidin RR, Brunner JR. Development of a Parenting Alliance Inventory. Journal of Clinical Child Psychology. 1995;24:31–40. [Google Scholar]
  2. Abidin RR, Konold TR. Parenting Alliance Measure professional manual. Psychological Assessment Resources; Odessa, FL: 1999. [Google Scholar]
  3. Ahrons CR. The continuing coparental relationship between divorced spouses. American Journal of Orthopsychiatry. 1981;51:415–428. doi: 10.1111/j.1939-0025.1981.tb01390.x. [DOI] [PubMed] [Google Scholar]
  4. Bearss KE, Eyberg S. A test of the parenting alliance theory. Early Education and Development. 1998;9:179–185. [Google Scholar]
  5. Behnke AO, MacDermid SM, Coltrane SL, Parke RD, Duffy S, Widaman KF. Family cohesion in the lives of Mexican American and European American parents. Journal of Marriage and Family. 2008;70:1045–1059. [Google Scholar]
  6. Belsky J. Children and marriage. In: Fincham F, Bradbury T, editors. The psychology of marriage. Guilford; New York: 1990. pp. 172–200. [Google Scholar]
  7. Belsky J, Pasco Fearon RM. Exploring marriage-parenting typologies and their contextual antecedents and developmental sequelae. Development and Psychopathology. 2004;16:501–523. doi: 10.1017/s095457940400464x. [DOI] [PubMed] [Google Scholar]
  8. Belsky J, Youngblade L, Rovine M, Volling B. Patterns of marital change and parent-child interaction. Journal of Marriage and the Family. 1991;53:487–498. [Google Scholar]
  9. Bonds DD, Gondoli DM. Examining the process by which marital adjustment affects maternal warmth: The role of coparenting support as a mediator. Journal of Family Psychology. 2007;21:288–296. doi: 10.1037/0893-3200.21.2.288. [DOI] [PubMed] [Google Scholar]
  10. Byrne BM. Structural equation modeling with AMOS. Lawrence Erlbaum Associates; New Jersey: 2001. [Google Scholar]
  11. Bui K-V,T, Peplau LA, Hill CT. Testing the Rusbult model of relationship commitment and stability in a 15-year study of heterosexual couples. Personality and Social Psychology Bulletin. 1996;22:1244–1257. [Google Scholar]
  12. Camara KA, Resnick G. Styles of conflict resolution and cooperation between divorced parents: Effects on child behavior and adjustment. American Journal of Orthopsychiatry. 1989;59:560–575. doi: 10.1111/j.1939-0025.1989.tb02747.x. [DOI] [PubMed] [Google Scholar]
  13. Coltrane S, Adams M. Gender and families. Rowan & Littlefield; New York: 2008. [Google Scholar]
  14. Córdova JV, Morrill MI, Wachs KT, Meade AE, Blair J, Harp AG. The Marriage Checkup: A motivational interviewing approach to the promotion of marital health.. In: Córdova JV, editor. Relational health check-ups for marriage, family, and parenting; Symposium conducted at the annual convention of the American Psychological Association; Boston, MA. August, 2008.2008. [Google Scholar]
  15. Cowan CP, Cowan PA. Two central roles for couple relationships: Breaking negative intergenerational patterns and enhancing children's adaptation. Sexual and Relationship Therapy. 2005;20:275–288. [Google Scholar]
  16. Cummings EM, Faircloth WB, Mitchell PM, Cummings JS, Schermerhorn AC. Evaluating a brief prevention program for improving marital conflict in community families. Journal of Family Psychology. 2008;22:193–202. doi: 10.1037/0893-3200.22.2.193. [DOI] [PubMed] [Google Scholar]
  17. Dadds MR, Maujean A, Fraser JA. Parenting and conduct problems in children: Australian data and psychometric properties of the Alabama Parenting Questionnaire. Australian Psychologist. 2003;38:238–241. [Google Scholar]
  18. Du Rocher Schudlich TD, Cummings EM. Parental dysphoria and children's adjustment: Marital conflict styles, children's emotional security, and parenting as mediators of risk. Journal of Abnormal Child Psychology. 2007;35:627–639. doi: 10.1007/s10802-007-9118-3. [DOI] [PubMed] [Google Scholar]
  19. Erel O, Burman B. Interrelatedness of marital relations and parent-child relations: A meta-analytic review. Psychology Bulletin. 1995;118:108–132. doi: 10.1037/0033-2909.118.1.108. [DOI] [PubMed] [Google Scholar]
  20. Feinberg ME. Coparenting and the transition to parenthood: A framework for prevention. Clinical Child and Family Psychology Review. 2002;5:173–195. doi: 10.1023/a:1019695015110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Feinberg ME. The internal structure and ecological context of coparenting: A framework for research and intervention. Parenting Science and Practice. 2003;3:95–131. doi: 10.1207/S15327922PAR0302_01. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Feinberg ME, Kan ML, Hetherington EM. The longitudinal influence of coparenting conflict on parental negativity and adolescent maladjustment. Journal of Marriage and Family. 2007;69:687–702. doi: 10.1111/j.1741-3737.2007.00400.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Feinberg ME, Kan ML. Establishing family foundations: Intervention effects on coparenting, parent/infant well-being and parent-child relations. Journal of Family Psychology. 2008;22:253–263. doi: 10.1037/0893-3200.22.2.253. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Fincham FD. Child development and marital relations. Child Development. 1998;69:543–574. [PubMed] [Google Scholar]
  25. Floyd FJ, Gilliom LA, Costigan CL. Marriage and the parenting alliance: Longitudinal prediction of change in parenting perceptions and behaviors. Child Development. 1998;69:1461–1479. [PubMed] [Google Scholar]
  26. Frick PJ. The Alabama Parenting Questionnaire. University of Alabama; 1991. Unpublished instrument. [Google Scholar]
  27. Frosch CA, Mangelsdorf SC, McHale JL. Marital behavior and the security of preschooler-parent attachment relationships. Journal of Family Psychology. 2000;14:144–161. doi: 10.1037//0893-3200.14.1.144. [DOI] [PubMed] [Google Scholar]
  28. Gable S, Belskky J, Crnic K. Marriage, parenting, and child development: Progress and prospects. Journal of Family Psychology. 1992;5:276–294. [Google Scholar]
  29. Grych JH, Fincham FD. Children's appraisals of marital conflict: Initial investigations of the cognitive-contextual framework. Child Development. 1993;64:215–230. doi: 10.1111/j.1467-8624.1993.tb02905.x. [DOI] [PubMed] [Google Scholar]
  30. Hawes DJ, Dadds MR. Assessing parenting practices through parent-report and direct observation during parent-training. Journal of Child and Family Studies. 2006;15:555–568. [Google Scholar]
  31. Kenny DA, Kashy DA, Cook WL. Dyadic data analysis. Guilford; New York: 2006. [Google Scholar]
  32. Kitzmann KM. Effects of marital conflict on subsequent triadic family interactions and parenting. Developmental Psychology. 2000;36:3–13. doi: 10.1037//0012-1649.36.1.3. [DOI] [PubMed] [Google Scholar]
  33. Knudson-Martin C, Mahoney AR. Couples, gender, and power: Creating change in intimate relationships. Springer Publishing Company; New York: 2009. [Google Scholar]
  34. Kolak AM, Volling BL. Parental expressiveness as moderator of coparenting and marital relationship quality. Family Relations. 2007;56:47–478. doi: 10.1111/j.1741-3729.2007.00474.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Konold T, Abidin R. Parenting alliance: A multifactor perspective. Assessment. 2001;8:47–65. doi: 10.1177/107319110100800105. [DOI] [PubMed] [Google Scholar]
  36. Lindhal KM, Clements M, Markman H. Predicting marital and parenting functioning in dyads and triads: A longitudinal investigation of marital processes. Journal of Family Psychology. 1997;11:139–151. [Google Scholar]
  37. Margolin G, Christensen A, John RS. The continuance and spillover of everyday tensions in distressed and nondistressed families. Journal of Family Psychology. 1996;10:304–321. [Google Scholar]
  38. Margolin G, Gordis EB, John RS. Coparenting: A link between marital conflict and parenting in two-parent families. Journal of Family Psychology. 2001;15:3–21. doi: 10.1037//0893-3200.15.1.3. [DOI] [PubMed] [Google Scholar]
  39. McHale JP. Coparenting and triadic interactions during infancy: The roles of marital distress and child gender. Developmental Psychology. 1995;31:985–996. [Google Scholar]
  40. McHale JP. Charting the bumpy road of coparenthood: Understanding the challenges of family life. Zero to Three; Washington D.C.: 2007. [Google Scholar]
  41. McHale JP, Kazali C, Rotman T, Talbot J, Carleton M, Lieberson R. The transition to coparenthood: Parents’ prebirth expectations and early coparental adjustment at 3 months postpartum. Development and Psychopathology. 2004;16:711–733. doi: 10.1017/s0954579404004742. [DOI] [PubMed] [Google Scholar]
  42. Meyers LS, Gamst G, Guarino AJ. Applied multivariate research design and interpretation. Sage; Thousand Oaks: 2006. [Google Scholar]
  43. Minuchin P. Families and individual development: Provocations from the field of family therapy. Child Development. 1985;56:289–302. [PubMed] [Google Scholar]
  44. Minuchin S. Families and family therapy. Harvard University Press; Cambridge, MA: 1974. [Google Scholar]
  45. Norton R. Measuring marital quality: A critical look at the dependent variable. Journal of Marriage and the Family. 1983;45:141–151. [Google Scholar]
  46. Schoppe-Sullivan SJ, Schermerhorn AC, Cummings EM. Marital conflict and children's adjustment: Evaluation of the parenting process model. Journal of Marriage and Family. 2007;69:1118–1134. [Google Scholar]
  47. Shelton KK, Frick PJ, Wootton J. Assessment of parenting practices in families of elementary school-age children. Journal of Clinical Child Psychology. 1996;25:317–329. [Google Scholar]
  48. Talbot JA, McHale JP. Individual parental adjustment moderates the relationship between marital and coparenting quality. Journal of Adult Development. 2004;11:191–205. doi: 10.1023/B:JADE.0000035627.26870.f8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Thompson B. Ten commandments of structural equation modeling. In: Grimm LG, Yarnold PR, editors. Reading and understanding more multivariate statistics. American Psychological Association; Washington, DC: 2000. pp. 261–283. [Google Scholar]
  50. Weissman S, Cohen R. The parenting alliance and adolescence. Annals of the American Society for Adolescent Psychiatry. 1985;12:24–45. [PubMed] [Google Scholar]

RESOURCES