Feeling caught as a mediator of co-parental communication and young adult children’s mental health and relational satisfaction with parents

Abstract

This study tested young adults’ feelings of being caught between their parents as mediators of co-parental communication (i.e., supportive and antagonistic communication) and young adults’ mental health and relational satisfaction with parents. Participants included 493 young adult children from intact and divorced families. For participants from intact families, feeling caught mediated indirect effects for both supportive and antagonistic co-parental communication on satisfaction with mothers and mental health symptoms, though direct effects for supportive co-parental communication on all three outcomes remained. For participants from divorced families, however, feeling caught emerged as a mediator only of antagonistic co-parental communication and mental health symptoms. Collectively, the results suggest that feeling caught may function as a mechanism linking co-parental communication to children’s adjustment.

Keywords

Co-parental communication co-parenting divorce feeling caught mental health parent–child satisfaction

Perhaps no other family relationship is more critical to adolescent and young adult children’s adjustment than the co-parenting relationship. A co-parenting relationship exists when “at least two individuals are expected by mutual agreement or societal norms to have conjoint responsibility for a particular child’s well-being” (Van Egeren & Hawkins, 2004, p. 166). Co-parental communication, in turn, refers not to the individual attempts of a parent to guide and direct the behaviors and activities of his or her child, but to the interaction patterns that emerge as one co-parent supports and/or undermines the parenting attempts of his or her partner. Family scholars have argued that co-parental communication should be conceptualized and studied as distinct from other interparental interactions because of the unique effects that co-parenting may have on family member outcomes (Adamsons & Pasley, 2006; Schrodt, 2010, 2011; Schrodt & Braithwaite, 2011). In intact families, for example, co-parenting is more predictive of parents’ and children’s adjustment than is general marital quality, and it accounts for variance in parenting and child outcomes after controlling individual parent characteristics (Feinberg, Kan, & Hetherington, 2007; Schoppe-Sullivan, Mangelsdorf, Frosch, & McHale, 2004).

Despite the importance of co-parental communication not only to intact families (e.g., Feinberg et al., 2007; Schoppe-Sullivan et al., 2004) but also to postdivorce families (e.g., Ahrons & Tanner, 2003; Ganong, Coleman, Markham, & Rothrauff, 2011; Schrodt, Miller, & Braithwaite, 2011), theoretically, researchers have yet to identify the underlying causal mechanisms linking co-parental communication to young adult children’s adjustment in both family types. That is, questions remain as to why co-parental communication is so impactful for children’s adjustment. Researchers have demonstrated that interparental conflict is harmful to young adult children’s well-being when it involves triangulation, or loyalty conflicts that arise when a covert coalition is formed, uniting one family member with another against a third person (Afifi, 2003; Buchanan, Maccoby, & Dornbusch, 1991; Schrodt & Afifi, 2007). Children who feel caught between their parents often feel “torn,” “put in the middle,” or forced to defend their loyalty to each of their parents (Afifi, 2003; Amato & Afifi, 2006). As one form of interparental conflict, antagonistic co-parental communication is likely to engender young adult children’s feelings of being caught. Feeling caught, in turn, is associated with anxiety, depression, and diminished relationships with parents for young adults in both intact and divorced families (Afifi & Schrodt, 2003; Amato & Afifi, 2006; Buchanan et al., 1991; Schrodt & Afifi, 2007). Thus, feeling caught may operate as one theoretical mechanism linking co-parental communication to young adult children’s adjustment.

This study represents a second investigation from a larger program of research examining interparental conflict, co-parental communication, and young adult children’s adjustment. In our first investigation (Shimkowski & Schrodt, 2012), we demonstrated that co-parental communication represents both a risk indicator and a risk mechanism, linking young adults’ observations of interparental conflict to their personal well-being. While risk indicators are the variables (e.g., antagonistic co-parental communication) statistically associated with a particular outcome (e.g., young adult children’s adjustment), risk mechanisms are the factors that explain the association (Rutter, 1994). In this follow-up study, we tested the degree to which feeling caught mediates the effects of supportive and antagonistic co-parental communication on young adult children’s mental health and relational satisfaction with parents. Consistent with the co-parenting literature, we also explored the extent to which divorce altered the associations in our hypothesized model. Not only are the majority of states in the US now mandating some type of parent education for divorcing parents, but most of these programs also attempt to increase awareness of the effects of parental conflict on children and enhance co-parental communication skills (Pollett & Lombreglia, 2008). Given tremendous variability in young adults’ responses to interparental conflict and divorce, including the degree to which young adults from both family types feel caught between their parents (Amato & Afifi, 2006; Schrodt & Afifi, 2007; Schrodt & Ledbetter, 2007), we tested our hypothesized model for measurement and structural invariance for young adults from divorced and intact families.

Theoretical perspective

Our investigation was informed by a constellation of related, but distinct theoretical perspectives. First, theoretical models of co-parenting in intact families have identified core features of the co-parental alliance, including the degree of solidarity and support between the co-parental partners, any dissonance or antagonism expressed during the adults’ co-parental strivings, and the extent to which both partners participate actively in engaging with and directing the children (Feinberg, 2003; McHale, Lauretti, Talbot, & Pouquette, 2002). Scholars investigating co-parental communication, however, have looked primarily at the degree of support and/or hostility expressed between co-parenting partners (Adamsons & Pasley, 2006). Supportive co-parental communication involves teamwork and collaboration, as (ex)partners elaborate on or complement the child-rearing efforts of their co-parenting partner (Gable, Crnic, & Belsky, 1994). When co-parents undermine each other’s parenting efforts, however, their interactions consist of coldness, displeasure, competition, or anger with one’s partner (Schoppe-Sullivan et al., 2004).

Taken together, both forms of co-parental communication are posited to function as risk mechanisms in families, essentially magnifying (or mitigating) the stress and anxiety associated with witnessing interparental conflict (Margolin, Christensen, & John, 1996). This is critical, given that interparental conflict often spills over into the co-parenting relationship that, in turn, affects the amount of stress experienced in the parent–child relationship (Margolin, Gordis, & John, 2001). By definition, co-parental communication includes each co-parent’s evaluation of their partner’s relationship to the child(ren). Thus, it stands to reason that witnessing hostile co-parental communication would be particularly damaging to young adult children’s adjustment and to their relationships with both parents.

Second, our model was informed by social cognitive theories of marital conflict. Specifically, Grych and Fincham’s (1990) cognitive–contextual framework posits that the context of marital (or ex-spousal) conflict and variations in children’s cognitive appraisals of their parents’ discord mediates the associations between the marital (or ex-spousal) relationship and children’s adjustment. Accordingly, young adults’ understanding and interpretation of their parents’ disputes likely mediate the effects of witnessing interparental conflict on their adjustment. Adult children who feel threatened by co-parental conflict may fear becoming involved in their parents’ disputes, subsequently heightening their feelings of being torn between their parents and eroding relational quality with both parents. In fact, Davies and Cummings’s (1994) emotional security hypothesis posits that destructive marital conflict reduces children’s emotional security, which in turn negatively influences their relationships with other family members. In line with their hypothesis, supportive co-parental communication is likely to enhance children’s emotional security as it strengthens perceptions of the marital bond and presents a unified front. As Cowan and McHale (1996) observed, co-parenting “is uniquely characterized by the parents’ connection as parents (a connection that endures even if the marriage does not). A well-functioning co-parental relationship accomplishes parenting tasks but also conveys to the child a sense of solidarity and common purpose” (p. 99).

Moreover, these associations are likely to persist as children age into adolescence and young adulthood. For example, Aquilino (1997) found a high degree of continuity between parent–child relationships in adolescence and adulthood. Apter (2001) argued that, from the ages of 18 to 22 years, most individuals are not ready for the responsibilities of adulthood. Although the content of co-parenting messages and the issues at stake may change over time, similar patterns of co-parental communication and family member adjustment should exist in young adult children given that “emerging adults” (i.e., young adults in their early to mid-20s) are often dependent on their parents’ emotional and financial support for many years (Apter, 2001; Arnett, 2004). In fact, Beaton, Doherty, and Wenger (2012) recently called for more research on co-parenting with young adults, “as more children are living longer at home (or leaving home and returning again); therefore, many mothers and fathers are co-parenting their children for many more years than in the recent past” (p. 236). Thus, we advanced our first set of predictions that are depicted in our hypothesized model (see Figure 1):

Figure 1.

Hypothesized model of feeling caught as a mediator of COPAR and young adults’ mental health and relational satisfaction with parents. COPAR: co-parental communication.

H1: Young adult children’s perceptions of supportive co-parental communication positively predict their relational satisfaction with their parents.

H2: Young adult children’s perceptions of supportive co-parental communication positively predict their mental health.

Antagonistic co-parental communication, on the other hand, is likely to undermine young adult children’s emotional security by threatening the perceived stability of the spousal relationship and/or prolonging interparental conflict. Using longitudinal data, Feinberg et al. (2007) demonstrated that co-parenting conflict predicted more unique variance in adolescent antisocial behavior than did marital quality and disagreement combined. Gasper, Stolberg, Macie, and Williams (2008) reported that co-parental hostility and cooperation partially mediated the effects of marital status on young adults’ mental health and self-esteem. These lines of research suggest that young adult children may be more responsive to their parents’ co-parental communication patterns, and thus, more susceptible to antagonistic co-parenting, than they are to their parents’ marital communication skills. Consequently, we advanced our second set of hypotheses (see Figure 1):

H3: Young adult children’s perceptions of antagonistic co-parental communication negatively predict their relational satisfaction with their parents.

H4: Young adult children’s perceptions of antagonistic co-parental communication negatively predict their mental health.

Although perceptions of co-parental communication are likely to have direct effects on satisfaction with parents and mental health, family systems theory (Minuchin, 1974) would suggest that such perceptions may also have indirect effects on both outcomes through feelings of being caught between parents. For instance, feeling caught often emerges as a function of the interdependence that exists between parents and their young adult children, as each parent attempts to reconcile their dispute with their partner (or not), while maintaining (or undermining) their (ex)partner’s relationships with their offspring (Afifi, Schrodt, & McManus, 2009). This, of course, raises the possibility that some parents may seek to strengthen their relationships with their young adult children, while simultaneously working through their marital (or postdivorce) problems with their (ex)spouse. Family systems theory also suggests that the family influences and is influenced by the other subsystems, such that interactions occurring within one subsystem (e.g., the co-parental subsystem) affect interactions in others (e.g., the parent–child subsystem; Adamsons & Pasley, 2006). As Afifi (2003) noted, (ex)spouses often engender young adults’ feelings of being caught when they elevate the young adult child to the role of peer or co-parent in the family, when they inappropriately disclose about the other parent to the young adult child, and/or when they use the young adult child as a messenger or mediator.

Schrodt and Afifi (2007) demonstrated that young adults’ reports of their parents’ demand/withdraw patterns, aggression, and negative disclosures about the other parent were all positively associated with feelings of being caught. Although their findings do not speak directly to the perceptions of co-parental communication, they do provide indirect evidence to suggest that hostile and antagonistic co-parental communication should be positively associated with feeling caught. Given that antagonistic and supportive co-parental communication are inversely related, it follows that supportive co-parental communication should be negatively associated with feeling caught. To test this line of reasoning, we advanced the following hypotheses (see Figure 1):

H5: Young adult children’s perceptions of supportive co-parental communication negatively predict their feelings of being caught between their parents.

H6: Young adult children’s perceptions of antagonistic co-parental communication positively predict their feelings of being caught between their parents.

Researchers have also found that feeling caught is positively associated with anxiety, perceived stress, depression, and diminished relationships with parents for adolescents and young adults in both intact and divorced families (e.g., Afifi & Schrodt, 2003; Schrodt & Afifi, 2007; Schrodt & Ledbetter, 2007). Therefore, we reasoned that feeling caught should not only negatively predict young adults’ mental health and relational satisfaction with both parents (see Figure 1), but that it should operate as a risk mechanism that explains part of the association between perceptions of co-parental communication and young adult children’s adjustment. To test this, we advanced the following hypothesis (see Figure 1):

H7: Feeling caught mediates the association between young adult children’s perceptions of co-parental communication (i.e., supportive and antagonistic) and their reports of mental health and relational satisfaction with parents.

Finally, we anticipated that divorce might alter the associations in our hypothesized model. Researchers have identified the co-parenting relationship as an important key to understanding children’s reactions to divorce and their well-being following divorce (Buchanan et al., 1991; Maccoby & Mnookin, 1992). Not only do young adults from divorced families report, on average, greater frequencies of interparental conflict than those from intact families, but different patterns of associations between interparental conflict and young adults’ mental health often emerge based on divorce status (Schrodt & Afifi, 2007; Schrodt & Ledbetter, 2007). For instance, young adults from divorced families are more likely than young adults from intact families to feel caught between their parents, yet such feelings are potentially more damaging to the personal and relational well-being of young adults in intact families (Amato & Afifi, 2006; Schrodt & Afifi, 2007). Consequently, in order to account for divorce, we tested our hypothesized model for measurement and structural invariance for young adult children from intact and divorced families.

Method

Participants

The data reported here were collected as part of an investigation of interparental conflict, co-parental communication, and young adults’ adjustment (Shimkowski & Schrodt, 2012). Participants included 493 young adult children from intact (n = 364) and postdivorce families (n = 129) in the southwest region of the US. The sample included 200 men and 293 women with a mean age of 20.3 (SD = 2.9). Most participants were Caucasian (70%), though 10.1% were African American, 8.9% were Hispanic American, and 4.7% were Asian American. Young adult children from intact families reported that their parents had been married for an average of 25.3 years (SD = 4.8), whereas those from divorced families reported an average length of their parents’ marriage (all of which were first marriages) prior to divorce of 12.3 years (SD = 7.5) and an average length of time since their parents’ divorce of 12.8 years (SD = 7.6). When asked who they currently lived with (or who their primary caretakers were when they lived at home), most participants reported living with both their mother and father (68%, n = 335), though 15.8% (n = 78) reported living primarily with their mother, 6.1% (n = 30) with their mother and stepfather, 3.2% (n = 16) with their father, 1.0% (n = 5) with their father and stepmother, and the remaining 5.9% (n = 29) reporting “other.” Finally, when asked how often they talk with both of their parents during a typical week, participants reported talking an average of 4.37 hours a week (SD = 2.39) with their mother and 3.11 hours a week (SD = 2.37) with their father. However, participants in divorced families reported talking less often with their mother in a typical week (M = 3.97 h) than participants from intact families (M = 4.52 h), F(1, 397) = 4.22, p < .05, η² = .01. They also reported talking less often with their father in a typical week (M = 2.10 h) than participants from intact families (M = 3.11 h), F(1, 399) = 14.90, p < .01, η² = .04.

Procedures

Upon securing human subjects approval, we solicited direct participation from a variety of undergraduate students at two mid-sized universities. After obtaining informed consent, participants completed an online questionnaire on a volunteer basis. In classes where instructors granted permission, students were awarded minimal class credit (less than 2%) for participation in the research. All participation took place outside of regular class time and participants completed the questionnaire anonymously. The survey took approximately 25 min to complete, after which respondents were thanked for their participation and debriefed.

Measures

Perceptions of co-parental communication

Participants’ perceptions of their parents’ co-parental communication were measured using Stright and Bales’ (2003) 12-item quality of co-parenting questionnaire, which was developed based on Belsky and colleagues’ co-parenting coding system (Belsky, Crnic, & Gable, 1992). After reading the prompt “My parents,” participants reported how often their parents engaged in certain co-parenting behaviors (e.g., “worked well together raising me,” “gave me conflicting messages when parenting me,” “supported each other’s parenting,” and “criticized each other’s parenting”), including six items each that measured both supportive and antagonistic co-parental communication. Responses were solicited using a five-point frequency scale that ranged from (1) Never to (5) Always. In this study, the scale produced excellent internal reliability with alpha coefficients of .90 and .92 for antagonistic and supportive co-parental communication, respectively.

Feeling caught

Young adults’ feelings of being caught between their parents were measured using Buchanan et al.’s (1991) feeling caught scale. The scale consists of 11 items asking participants to indicate the frequency with which they experienced certain behaviors and feelings (e.g., “How often do you feel caught between your mother and father?” and “How often do you feel like if you are loyal to one parent, you are being disloyal to your other parent?”). Responses were solicited using a five-point frequency scale that ranged from 1 (Never) to 5 (Very often). The reliability and validity of Buchanan et al.’s (1991) measure are well-established (Afifi & Schrodt, 2003; Schrodt & Afifi, 2007; Schrodt & Ledbetter, 2007), and in this study, the feeling caught scale produced an alpha coefficient of .92.

Relational satisfaction

Participants’ reports of relational satisfaction with their parents were operationalized using a modified version of the Marital Opinion Questionnaire (Huston, McHale, & Crouter, 1986). The scale consisted of 10 items measuring satisfaction with seven-point semantic differential scales (e.g., “miserable–enjoyable”) and an additional global satisfaction item that ranged from 1 (Completely dissatisfied) to 7 (Completely satisfied). Participants were asked to report satisfaction with each parent over the last month. Researchers have previously demonstrated the validity and reliability of using the modified version to measure relational satisfaction (e.g., Afifi & Schrodt, 2003; Schrodt & Afifi, 2007). In this study, the 11-item measure produced strong reliability with Cronbach’s alpha coefficients of .96 and .98 for children’s reports of satisfaction with mothers and fathers, respectively.

Mental health

Participants’ mental health was operationalized using the nine-item mental health subscale of Dornbusch, Mont-Reynaud, Ritter, Chen, and Steinburg’s (1991) physical and mental health symptom instrument. Participants indicated how often over the past 2 weeks they had felt certain symptoms (e.g., overtired, nervous, depressed, tense, and without appetite) using a four-point frequency scale that ranged from (1) Never to (4) Three or more times. Higher scores represented greater frequencies of mental health symptoms, and thus, poorer mental health. Previous researchers have demonstrated the reliability and validity of this scale (Schrodt & Afifi, 2007; Schrodt & Ledbetter, 2007), and in this study, the mental health subscale produced an alpha coefficient of .85.

Data analysis

We tested our first six hypotheses using Pearson’s product–moment correlations. For our final hypothesis, we employed structural equation modeling (SEM) with maximum likelihood (ML) estimation using LISREL 8.80 to test the hypothesized associations in Figure 1. Consistent with Kline’s (2005) recommendations, we conducted a confirmatory factor analysis (CFA) of the measurement model to assess the relationships among indicators and their respective latent constructs prior to testing the hypothesized mediation model. The measurement model included six latent constructs: (a) supportive co-parental communication, (b) antagonistic co-parental communication, (c) feeling caught, (d) satisfaction with mother, (e) satisfaction with father, and (f) mental health symptoms. All constructs were formed by parceling each related measurement scale into three parcels, which are “aggregate-level [indicators] comprised of the sum (or average) of two or more items, responses, or behaviors” (Little, Cunningham, Shahar, & Widaman, 2002, p. 152). The parceling technique has several advantages over using items as indicators, including greater reliability, more precise identification of the latent construct, and fewer parameter estimates (Kline, 2005; Little et al., 2002). Given unidimensional measures and no a priori rationale to guide parcel construction, we assigned items to parcels by thirds (e.g., for the mental health scale, items 1, 4, and 7 were assigned to parcel 1, items 2, 5, and 8 were assigned to parcel 2, and items 3, 6, and 9 were assigned to parcel 3).

For both measurement and structural models, we evaluated model fit with the ML chi-squared (χ²) statistic. Due to sensitivity of large sample sizes in the χ² statistic, we also examined the non-normed fit index (NNFI), comparative fit index (CFI), standardized root mean square residual (SRMR), and root mean square error of approximation (RMSEA) to assess model fit. Values greater than .90 for the NNFI and CFI may indicate reasonably good fit (Hu & Bentler, 1999), whereas SRMR estimates less than .10 are generally considered favorable (Kline, 2005). RMSEA estimates less than .05 indicate close model fit, values between .05 and .08 suggest reasonable fit, and values greater than .10 suggest poor fit (Browne & Cudeck, 1993). We standardized all values prior to evaluating the models.

Finally, in order to compare the hypothesized model across divorced and intact families, it is first necessary to establish measurement invariance between both groups (Little, 1997). Following the procedures described by Little (1997), establishing measurement invariance requires a series of sequential model tests including: (a) testing the relationship between the manifest indicators and latent constructs across groups (i.e., configural invariance), (b) testing the equality of the loadings from the indicators to the latent constructs across groups (i.e., loading or weak metric invariance), (c) testing the equality of the means of the manifest indicators across groups (i.e., intercept or strong metric invariance), (d) testing the homogeneity of variances and covariances across groups, (e) testing the equality of correlations between latent constructs, and (f) testing the equality of the means of the latent constructs.

For most models, we identified our constructs by fixing the latent variance in the first group to 1.0 (Kline, 2005). Though this procedure standardizes the variance metric between groups, it does not permit estimation of the latent means in the first group. Therefore, for tests involving the latent mean, we identified the model using an effects coding procedure (Little, Slegers, & Card, 2006). This method involves constraining the loadings of each construct’s manifest indicators to average 1.0 and the mean of each construct’s manifest indicators to average .0. This modification does not alter the estimates of model fit.

Results

Descriptive statistics, including means, SDs, and Pearson’s product–moment correlations for all variables included in the study, are reported in Table 1. At the bivariate level of analysis, perceptions of supportive co-parental communication positively predict relational satisfaction with mothers and fathers, as well as mental health, in both intact and divorced families (H1 and H2 supported). Likewise, perceptions of antagonistic co-parental communication negatively predict satisfaction with both parents and mental health, though only for young adult children from intact families (H3 and H4 partially supported). In both family types, perceptions of supportive co-parental communication negatively predict feelings of being caught between parents, whereas perceptions of antagonistic co-parental communication positively predict such feelings (H5 and H6 supported).

Table 1.

Descriptive statistics and Pearson product–moment correlations for all variables (N = 493).

Variable	M	SD	1	2	3	4	5	6
1. COPAR-SUP	3.71	1.01	–	−.36**	−.25**	.17^†	.39**	−.16^†
2. COPAR-ANT	2.12	.94	−.50**	–	.46**	−.13	−.13	.09
3. Feeling caught	1.68	.77	−.41**	.42**	–	−.17^†	−.17^†	.32**
4. Satisfaction with mother	5.83	1.21	.35**	−.23**	−.31**	–	.02	−.35**
5. Satisfaction with father	5.49	1.55	.55**	−.33**	−.27**	.42**	–	−.21*
6. Mental health symptoms	2.46	.77	−.21**	.19**	.23**	−.32**	−.27**	–

COPAR-SUP: supportive co-parental communication; COPAR-ANT: antagonistic co-parental communication.

Note. Correlations in the lower diagonal are for participants from intact families (n = 364). Correlations in the upper diagonal are for participants from divorced families (n = 129). Responses to the co-parental communication and feeling caught measures were solicited using a five-point scale that ranged from 1 to 5. Responses to the satisfaction with mother and father measures were solicited using a seven-point scale that ranged from 1 to 7. Responses to the mental health symptoms measure were solicited using a four-point scale that ranged from 1 to 4.

^† p < .10; *p < .05; ** p < .01.

Given that parents’ length of marriage or time since the divorce may alter the associations in the hypothesized model, we conducted a series of correlations to determine whether length of marriage or time since the divorce needed to be entered as a control variable. No significant associations emerged and, thus, we proceeded with tests of measurement invariance for the hypothesized structural model.

Measurement models

Prior to tests of metric invariance, we conducted a CFA using the full sample to test the convergent and divergent validity of the six-factor model. This model produced good fit, χ²(120, N = 493) = 252.60, p < .01, NNFI = .99, CFI = .99, SRMR = .03, RMSEA = .049_{(90% confidence interval (CI): .040–.057)}. The modification indices revealed no further modifications that were theoretically defensible and would improve model fit, and each of the indicators loaded well on their hypothesized latent constructs (ranging from .81 to .98 across all constructs).

To test whether strong metric invariance could be established between intact and divorced families, we examined the two-group measurement model with freely estimated model parameters. This model also demonstrated close fit, χ²(240, N = 493) = 387.52, p < .01, NNFI = .98, CFI = .99, SRMR = .03, RMSEA = .049_{(90% CI: .040–.058)}. Next, we tested for weak metric invariance by constraining the loadings of the manifest indicators onto the latent constructs to equality between groups. The ΔCFI and RMSEA model tests (Little, Card, Slegers, & Ledford, 2007) yielded no significant change in model fit (see Table 2), thus indicating that weak metric invariance is tenable. Likewise, the test for strong metric invariance also produced no meaningful change in the CFI or RMSEA fit statistic (see Table 2). These tests indicate strong metric invariance between groups, establishing that the manifest indicators are assessing the same latent constructs for young adults from both intact and divorced families (Little, 1997; see Table 3 for the loading and intercept of each indicator).

Table 2.

Fit indices for the nested sequence in the multiple group confirmatory factor analysis.

Model	χ ²	df	p	Δχ²	Δdf	p	RMSEA	RMSEA 90% CI	NNFI	CFI	Constraint tenable
Configural invariance	387.52	240	<.01	–	–	–	.049	.040–.058	.98	.99	N/A
Loading invariance^a	412.45	252	<.01	–	–	–	.050	.041–.059	.98	.99	Yes
Intercept invariance^a	437.29	264	<.01	–	–	–	.051	.042–.059	.98	.98	Yes
Homogeneity of variances and covariances^b	548.43	273	<.01	135.98	21	<.01	.071	.063–.078	.97	.98	No
Homogeneity of variances^b	486.24	258	<.01	73.79	6	<.01	.064	.056–.072	.98	.98	No
Equality of covariances^b	442.97	265	<.01	30.52	15	<.05	.050	.042–.059	.98	.98	No
Latent mean invariance^c	553.00	270	<.01	115.71	6	<.01	.062	.054–.070	.97	.97	No

RMSEA: root mean square error of approximation; CFI: comparative fit index; CI: confidence interval; NNFI: non-normed fit index.

Note. Each nested model contains its constraints plus the constraints of all previous, tenable models.

^aEvaluated with the ΔCFI and RMSEA model fit tests.

^bEvaluated with the χ² difference test, in comparison to the loading (weak) invariance model.

^cEvaluated with the χ² difference test, in comparison to the intercept (strong) invariance model.

Table 3.

Loading and intercept values, residuals, and R² values for each indicator, and the estimated latent variance from the strong metric invariance model.^a

	Equated LISREL estimates		Standardized	Intact		Divorced
Indicator	Loading (SE)	Intercept (SE)	loading^b	θ^c	R ²	θ^c	R ²
COPAR-SUP: Estimated latent variance (intact = 1.00; divorced = .89)
Parcel 1	.87 (.04)	4.06 (.05)	.86	.24	.76	.33	.67
Parcel 2	.86 (.04)	3.97 (.05)	.89	.19	.81	.24	.77
Parcel 3	.89 (.04)	3.91 (.05)	.88	.24	.76	.23	.77
COPAR-ANT: Estimated latent variance (intact = 1.00; divorced = .98)
Parcel 1	.80 (.04)	2.02 (.05)	.80	.34	.66	.40	.60
Parcel 2	.89 (.04)	1.85 (.05)	.93	.12	.88	.14	.86
Parcel 3	.84 (.04)	1.98 (.05)	.80	.35	.65	.37	.63
Feeling caught: Estimated latent variance (intact = 1.00; divorced = 2.58)
Parcel 1	.61 (.03)	1.53 (.04)	.90	.18	.82	.20	.80
Parcel 2	.55 (.02)	1.48 (.03)	.92	.16	.85	.17	.84
Parcel 3	.54 (.03)	1.45 (.03)	.87	.24	.76	.23	.77
Satisfaction with mother: Estimated latent variance (intact = 1.00; divorced = 1.32)
Parcel 1	1.13 (.05)	6.02 (.06)	.96	.08	.92	.08	.92
Parcel 2	1.09 (.04)	5.98 (.06)	.96	.07	.93	.07	.93
Parcel 3	1.21 (.05)	5.75 (.07)	.92	.15	.85	.16	.84
Satisfaction with father: Estimated latent variance (intact = 1.00; divorced = 2.12)
Parcel 1	1.30 (.05)	5.86 (.07)	.97	.06	.94	.04	.96
Parcel 2	1.28 (.05)	5.84 (.07)	.98	.05	.95	.04	.96
Parcel 3	1.21 (.05)	5.74 (.07)	.93	.14	.86	.15	.85
Mental health symptoms: Estimated latent variance (intact = 1.00; divorced = .88)
Parcel 1	.75 (.04)	2.62 (.04)	.90	.19	.81	.23	.77
Parcel 2	.68 (.04)	2.45 (.04)	.81	.34	.66	.34	.66
Parcel 3	.68 (.04)	2.05 (.04)	.81	.33	.67	.40	.60

COPAR-SUP: supportive co-parental communication; COPAR-ANT: antagonistic co-parental communication.

^aModel identified by fixing latent variance to 1.0.

^bCommon metric completely standardized solution.

^cWithin group completely standardized solution.

After establishing strong metric invariance, an omnibus test evaluated the homogeneity of the variances and covariances. This test failed, indicating significant differences in the variance/covariance matrix between intact and divorced families. As shown in Table 2, a subsequent test of only the homogeneity of variances also failed. The failure of these tests indicates that any associations between the raw latent variables might result either from different variances or from correlational differences. Thus, we used procedures described by Little (1997) to correct this. We created six second-order ‘phantom’ constructs (Rindskopf, 1984) and regressed each first-order construct onto its corresponding second-order construct, fixing the variance of each first-order construct to .0 and the variance of each phantom construct to 1.0. This technique moves the difference in variances to the regression paths between the phantom and first-order constructs, thereby placing the phantom constructs on a similar variance metric and permitting a pure test of the correlation between constructs. Using this method, the equality of correlations test failed, indicating significant differences in the pattern of interrelationships between the latent constructs across divorced and intact groups after controlling for significant differences in variance.

A final series of tests on the latent parameters evaluated whether the construct means differed significantly between intact and divorced families. The omnibus test of all latent means revealed significant differences in the means between groups (see Table 2). Subsequent nested χ² tests on each latent construct indicated significant mean differences across all six latent constructs (see Table 4), with young adults from divorced families reporting less supportive and more antagonistic co-parental communication, greater feelings of being caught, less satisfaction with both parents, and poorer mental health than young adults from intact families.

Table 4.

Results of nested χ² tests for latent mean level differences.

Constructs	Latent mean intact	Latent mean divorced	χ ²	df	p	Δχ²	Δdf	p	Equivalent across groups
Intercept invariance (baseline model)	–	–	437.29	264	<.01	–	–	–	N/A
COPAR-SUP	3.98	2.96	534.07	265	<.01	96.78	1	<.01	No
COPAR-ANT	1.95	2.62	485.25	265	<.01	47.96	1	<.01	No
Feeling caught	1.49	2.20	490.64	265	<.01	53.35	1	<.01	No
M-SAT	5.91	5.53	445.57	265	<.01	8.28	1	<.01	No
F-SAT	5.81	4.55	481.78	265	<.01	44.49	1	<.01	No
MHS	2.38	2.57	444.40	265	<.01	7.11	1	<.01	No

COPAR-SUP: supportive co-parental communication; COPAR-ANT: antagonistic co-parental communication; M-SAT: satisfaction with mother; F-SAT: satisfaction with father; MHS: mental health symptoms; χ²: chi-square.

Note. Model identified by constraining loadings for each indicator to average 1.0 and intercepts for each indicator to average .0.

Structural model and tests of mediation

Since the previous analyses indicated that the variances and correlations were not homogenous between groups, collapsing both groups into a single structural model was not tenable. Therefore, we created separate structural models for each group, testing the hypothesized associations at the phantom construct level to correct the significantly different variances between the latent constructs in both groups (Little, 1997). The saturated model produced good global fit, χ²(252, N = 493) = 412.45, p < .01, NNFI = .98, CFI = .99, SRMR = .06, RMSEA = .050_{(90% CI: .041–.059)}, yet contained several nonsignificant paths (see Figures 2 and 3). To provide a more parsimonious set of models that still fit the data, and consistent with standard procedures for model trimming (Kline, 2005), nonsignificant paths were removed iteratively (beginning with the statistically least significant path) until only significant paths remained in the model for each group. The trimmed model demonstrated good global model fit, χ²(263, N = 493) = 419.06, p < .01, NNFI = .98, CFI = .99, SRMR = .08, RMSEA = .048_{(90% CI: .039–.057)}, and did not produce a significant decline in model fit from the saturated model, Δχ²(11) = 6.62, p > .05. An inspection of Figures 4 and 5 reveals that interrelationships among the constructs in the model differ somewhat for young adult children from intact and divorced families.

Figure 2.

Full model of COPAR, feeling caught, satisfaction with parents, and mental health for young adults from intact families (n = 364). All latent constructs shown are phantom constructs. All parameters are standardized. **p < .01. COPAR: co-parental communication; MHS: mental health symptoms (thus, poorer mental health).

Figure 3.

Full model of COPAR, feeling caught, satisfaction with parents, and mental health for young adults from divorced families (n = 129). All latent constructs shown are phantom constructs. All parameters are standardized. *p < .05. **p < .01. COPAR: co-parental communication; MHS: mental health symptoms (thus, poorer mental health).

Figure 4.

Final model of COPAR, feeling caught, satisfaction with parents, and mental health for young adults from intact families (n = 364). All latent constructs shown are phantom constructs. All parameters are standardized. *p < .05. **p < .01. COPAR: co-parental communication; MHS: mental health symptoms (thus, poorer mental health).

Figure 5.

Final model of COPAR, feeling caught, satisfaction with parents, and mental health for young adults from divorced families (n = 129). All latent constructs shown are phantom constructs. All parameters are standardized. *p < .05. **p < .01. COPAR: co-parental communication; MHS: mental health symptoms (thus, poorer mental health).

The final structural model for intact families indicates that perceptions of supportive co-parental communication positively predict relational satisfaction with both mothers and fathers, and negatively predict feelings of being caught and mental health symptoms. Perceptions of antagonistic co-parental communication, on the other hand, positively predict feelings of being caught between parents. Feeling caught, in turn, negatively predicts satisfaction with mothers but positively predicts mental health symptoms (see Figure 4). When combined, perceptions of co-parental communication account for 24% of the variance in feeling caught, while co-parental communication and feeling caught account for 18% of the variance in satisfaction with mothers, 35% of the variance in satisfaction with fathers, and 7% of the variance in mental health symptoms.

For young adults in divorced families, the final structural model reveals that the perceptions of supportive co-parental communication have a direct, positive effect only on satisfaction with fathers. Similar to the model for intact families, perceptions of antagonistic co-parental communication positively predict feelings of being caught, which, in turn, negatively predict satisfaction with mothers but positively predict mental health symptoms (see Figure 5). When combined, antagonistic co-parental communication accounts for 20% of the variance in feeling caught, while both co-parental communication and feeling caught account for 4% of the variance in satisfaction with mothers, 16% of the variance in satisfaction with fathers, and 11% of the variance in mental health symptoms.

Finally, to determine whether there are significant indirect effects in the model, bootstrapping analyses¹ were conducted using LISREL to acquire standard error estimates, 95% CIs, and effect size estimates from 1000 samples for each indirect effect (Hayes, 2009; Preacher & Kelley, 2011). For young adult children from intact families, the results indicate that feeling caught partially mediates the positive effects of supportive co-parental communication on satisfaction with mothers and mental health, as well as the negative effects of antagonistic co-parental communication on the same two outcomes (see Table 5). For young adult children from divorced families, however, only antagonistic co-parental communication has a negative, indirect effect on mental health through feeling caught² (H7 supported).

Table 5.

Bootstrap analysis of indirect effects.

	Intact families (n = 364)				Divorced families (n = 129)
Indirect effect	β	SE	95% CI (lower, upper)	κ ²	β	SE	95% CI (lower, upper)	κ ²
1. COP-SUP → feeling caught → M-SAT	.068	.031	.013, .133*	.058	.014	.023	−.024, .070	.013
2. COP-SUP → feeling caught → F-SAT	.011	.027	−.047, .064	.010	.009	.017	−.022, .054	.009
3. COP-SUP → feeling caught → MHS	−.053	.024	−.104, −.012*	.046	−.034	.041	−.133, .038	.031
4. COP-ANT → feeling caught → M-SAT	−.060	.032	−.131, −.009*	.051	−.065	.060	−.193, .059	.057
5. COP-ANT → feeling caught → F-SAT	−.011	.025	−.071, .036	.008	−.044	.054	−.164, .067	.035
6. COP-ANT → feeling caught → MHS	.046	.023	.009, .100*	.041	.162	.082	.005, .343*	.131

COP-SUP: supportive co-parental communication; COP-ANT: antagonistic co-parental communication; M-SAT: satisfaction with mother; F-SAT: satisfaction with father; MHS: mental health symptoms; CI: confidence interval.

*Indirect effect is significant at p < .05 (excluding zero).

Discussion

Our primary goal in this study was to explore the degree to which young adults’ feelings of being caught mediates the association between their perceptions of co-parental communication and their mental health and satisfaction with their parents. Overall, the hypothesized model provided a good fit to the data and support for the contention that feeling caught mediates part, though not all, of the effect of witnessing supportive and/or antagonistic co-parental communication on young adults’ satisfaction with their mother and their mental health. Young adults from divorced families reported higher levels of antagonistic co-parental communication, less supportive co-parental communication, less satisfaction with both parents, and poorer mental health than young adults from intact families. Although the measures for each construct were invariant across both family types, these mean-level differences between young adults from intact and divorced families were accompanied by significant differences in the variances and in the pattern of associations among the constructs in the model. In fact, our first investigation demonstrated that antagonistic co-parental communication partially mediated the effects of parental demand/withdraw patterns and mothers’ aggression on young adults’ well-being in both family types (Shimkowski & Schrodt, 2012), yet the results of this study yielded different results for young adults from intact and divorced families, thereby providing at least three implications worth noting.

First, our findings replicate a long line of research documenting the importance of supportive and cooperative co-parental communication to children’s personal and relational adjustment in both intact and divorced families (e.g., Ahrons & Tanner, 2003; Feinberg et al., 2007; Schoppe-Sullivan et al., 2004). When parents actively support each other’s parenting efforts, use similar parenting techniques, and/or calmly discuss any parenting disagreements they might have, young adult children are more likely to be satisfied in their relationships with both parents. One explanation for these findings is that supportive co-parental communication helps facilitate a unified front to children, which in turn communicates a sense of solidarity, a common purpose, and a commitment to the well-being of the children and the family as a whole (Cowan & McHale, 1996). This sense of solidarity may strengthen children’s emotional security, prevent loyalty binds between parents, encourage closer and more satisfying relationships with both parents, and help alleviate any stress or tension that might undermine children’s mental health.

On the other hand, when (ex)spouses (a) give conflicting messages while parenting their children, (b) use parenting techniques that they know the other parent would not want them to use, (c) criticize each other’s parenting, and/or (d) ignore each other’s request for help with parenting the children, such antagonistic co-parenting behaviors may heighten young adults’ feelings of being caught between their parents, diminish their relationships with both parents, and threaten their mental health. As such, our results extend previous research on marital (or ex-spousal) conflict and young adult children’s adjustment by underscoring a particular form of interparental conflict that is likely to persist even if the marriage does not. As Ganong et al. noted (2011), co-parenting involves some form of communication between divorced parents and some interaction with the child by each parent, even if the co-parental communication itself does not occur face-to-face. Indeed, in this study, perceptions of antagonistic co-parental communication were associated with feelings of being caught between parents in both family types, even though young adults from divorced families reported an average length of time since the divorce of 12 years.

A second set of implications to emerge from the present study revolve around the role of feeling caught in mediating the associations among perceptions of co-parental communication behaviors and young adults’ mental health and satisfaction with parents (see Table 5 and Figures 4 and 5). Consistent with theoretical models of co-parenting (e.g., Margolin et al., 1996, 2001), when parents use hostile and antagonistic co-parenting tactics, such behaviors may spill over into the parent–child relationship and heighten the degree to which young adults feel torn between their parents and/or put in the middle of their parents’ disputes. Feeling caught, in turn, may diminish young adults’ mental health by adding additional stress and tension, and by removing much needed social support from parents. Such feelings also negatively predict relational satisfaction with mothers and mediate part of the effects of both supportive and antagonistic co-parental communication on satisfaction with mothers, though the latter only occurred for young adult children from intact families.

It is interesting to note that although supportive co-parental communication produced direct, positive effects on satisfaction with fathers in both family types, the effects on satisfaction with mothers were partially mediated by feeling caught in intact families and no longer significant in divorced families. Of course, this result may simply be a function of multicollinearity given the moderate intercorrelations that emerged among young adults’ perceptions of co-parental communication and feeling caught that were included in the structural model. On the other hand, Margolin et al. (2001) suggested that mothers and fathers may function differently in the co-parenting relationship, as mothers report more overall involvement with their children (Russell & Russell, 1994), and generally carry a heavier load of parenting responsibilities, than do fathers (Floyd & Zmich, 1991). Consequently, children may develop different expectations of their mother and father when it comes to coordinating their parenting efforts and creating a sense of family solidarity and common purpose, thus producing different outcomes for relational satisfaction with both parents.

The third set of implications to emerge from the present study involves the comparisons made between young adults from intact and divorced families. Most notably, the final structural models provided more support for our theoretical line of reasoning in intact as opposed to divorced families, both in terms of the direct and indirect effects of supportive and antagonistic co-parental communication through feelings of being caught. Despite these differences, feeling caught did mediate the effect of antagonistic co-parental communication on young adults’ mental health, and supportive co-parental communication positively predicted satisfaction with fathers, in both groups. Given that supportive co-parental communication positively predicted satisfaction with fathers but not mothers in the divorced group, supporting an ex-spouse’s parenting attempts may be particularly important for nonresidential fathers in postdivorce families. For instance, mothers are more likely to have physical custody of the children, and thus are more likely to act as gatekeepers who restrict or facilitate fathers’ involvement with their children (Kelly, 2007). Likewise, fathers who believe their former wives give them greater support and who have positive attitudes about their own parenting abilities are more involved in co-parenting their children (Madden-Derdich & Leonard, 2000). Coupled with the results of our study, these lines of research underscore the importance of nonresidential fathers developing supportive and cooperative co-parental relationships with their ex-spouses so that they can have access to, and sustain satisfying relationships with, their children. That being said, our results clearly indicate that perceptions of co-parental communication and feeling caught hold greater implications for young adults’ relational satisfaction with their parents in intact families.

Theoretically, the results of this study contribute to models of co-parenting (e.g., Feinberg, 2003) by identifying a risk mechanism (i.e., feeling caught) that links antagonistic co-parental communication to young adults’ mental health. To date, researchers have compared the outcomes of co-parenting with those associated with general marital quality and have concluded that co-parenting is more predictive of children’s adjustment in intact families (Feinberg et al., 2007; Schoppe-Sullivan et al., 2004). Given that co-parental communication focuses specifically on the connection that spouses have established as parents, one possible explanation for this finding may be that supportive and nonantagonistic co-parental communication reduces children’s feelings of being caught between their parents, which in turn enhances their emotional security, mental health, and their relationships with other family members.

Practically speaking, our results also contribute to a growing body of research on the behaviors that exacerbate or mitigate children’s feelings of being caught in both divorced and intact families. For instance, Schrodt and Ledbetter (2012) found that when mothers in divorced families communicate that they value, respect, and cherish their relationships with their children, such parental confirmation helps mitigate the negative effects of feeling caught on family satisfaction. Parents in both intact and divorced families often express desires to promote effective parenting and protect or buffer their children from their own marital (or ex-spousal) disputes (Maccoby & Mnookin, 1992; Schrodt, Baxter, McBride, Braithwaite, & Fine, 2006). Clinical interventions that focus strictly on individual parenting skills, however, do not address the issues that arise when parents undermine each other’s parenting efforts with the children. Based on the results of this study, clinical interventions that promote supportive co-parental communication skills may reduce the tendency for their children to become enmeshed in their marital (or postdivorce) disputes (Margolin et al., 2001). Strengthening the abilities of co-parents to support each other in their parenting attempts, in turn, may reduce the degree to which adolescents and young adults suffer the mental health consequences of feeling caught between their parents.

Despite these contributions, the results should be interpreted with caution given the inherent limitations of the research design. A key limitation to this report is the cross-sectional nature of the data, as well as the mean length of time since the divorce (i.e., 12 years) for the young adults from divorced families. Longitudinal research that tracks changes in interparental conflict and co-parental communication patterns across different family transitions, and at different stages of the divorce process, is needed. Likewise, the use of self-report methods and the homogenous sample warrants caution. Although we theorized that it is young adults’ perceptions of co-parental communication that would influence their reports of mental health, observational studies would reveal additional insights into the congruence (or lack thereof) between parents’ and children’s reports of co-parental communication. Researchers might also consider testing whether or not the amount of time children spend observing their parents communicating about parenting issues moderates the results here, given our use of a global measure of co-parental communication. Tests of mediation and statements of causality based on the results of statistical techniques, such as SEM, must be treated with caution given the correlational nature of the data.

Nevertheless, the results do suggest that young adults’ reports of feeling caught mediate part of the effects that their parents’ co-parental communication patterns have on their mental health and relational satisfaction with parents. Researchers might extend these efforts by examining relational and contextual factors that influence co-parental communication patterns in both intact and divorced families (e.g., disciplining the children, financial support, visitation rights, etc.). Scholars might also explore how co-parental communication alters the degree to which parents’ disclosures about their marital relationship (or the divorce) predict feelings of being caught and children’s adjustment. For instance, Afifi et al.’s (2009) Divorce Disclosure Model posits that the relationship with a former spouse is likely to moderate the direct and indirect effects of parental disclosures on both parents’ and children’s psychological well-being through feeling caught. Although most states now mandate some form of parent education for divorcing parents, it remains unclear what educational approaches and what types of content are most effective for improving co-parents’ attitudes and behaviors in postdivorce families. Hopefully, through these kinds of investigations, researchers can develop an empirically validated yet practical framework useful for parent education and furthering healthy families.

Footnotes

Authors’ note

A previous version of this study was presented at the 2012 National Communication Association’s annual convention in Orlando, FL.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Notes

References

Adamsons

Pasley

(2006). Coparenting following divorce and relationship dissolution. In Fine

M. A.

Harvey

J. H.

(Eds.), Handbook of divorce and relationship dissolution (pp. 241–261). Mahwah, NJ: Erlbaum.

Afifi

T. D.

(2003). ‘Feeling caught’ in stepfamilies: Managing boundary turbulence through appropriate communication privacy rules. Journal of Social and Personal Relationships, 20, 729–755. doi:10.1177/0265407503206002

Afifi

T. D.

Schrodt

(2003). “Feeling caught” as a mediator of adolescents’ and young adults’ avoidance and satisfaction with their parents in divorced and non-divorced households. Communication Monographs, 70, 142–173. doi:10.1080/0363775032000133791

Afifi

T. D.

Schrodt

McManus

(2009). The divorce disclosure model: Why parents disclose negative information about the divorce to their children and its effects. In Afifi

T. D.

Afifi

W. A.

(Eds.), Uncertainty, information management, and disclosure decisions: Theories and applications (pp. 403–425). New York, NY: Routledge.

Ahrons

C. R.

Tanner

J. L.

(2003). Adult children and their fathers: Relationship changes 20 years after parental divorce. Family Relations, 52, 340–351. doi: 10.1111/j.1741-3729.2003.00340.x

Amato

P. R.

Afifi

T. D.

(2006). Feeling caught between parents: Adult children’s relations with parents and subjective well-being. Journal of Marriage and Family, 68, 222–235. doi:10.1111/j.1741-3737.2006.00243.x

Apter

(2001). The myth of maturity. New York, NY: Norton.

Aquilino

(1997). From adolescent to young adult: A prospective study of parent–child relations during the transition to adulthood. Journal of Marriage and the Family, 59, 670–686.

Arnett

J. J.

(2004). Emerging adulthood: The winding road from the late teens through the twenties. New York, NY: Oxford University Press.

10.

Beaton

J. M.

Doherty

W. J.

Wenger

L. M.

(2012). Mothers and fathers coparenting together. In Vangelisti’s

A. L.

(Ed.), Routledge handbook of family communication (2nd ed., pp. 225–240). New York, NY: Routledge.

11.

Belsky

Crnic

Gable

(1992). Manual for coding coparenting. Pennsylvania State Child and Family Development Project, University Park, Pennsylvania State University.

12.

Browne

M. W.

Cudeck

(1993). Alternative ways of assessing model fit. In Bollen

K. A.

Long

J. S.

(Eds.), Testing structural equation models (pp. 136–162). Newbury Park, CA: Sage.

13.

Buchanan

C. M.

Maccoby

E. E.

Dornbusch

S. M.

(1991). Caught between parents: Adolescents’ experience in divorced homes. Child Development, 62, 1008–1029. doi:10.2307/1131149

14.

Cowan

P. A.

McHale

J. P.

(1996). Coparenting in a family context: Emerging achievements, current dilemmas, and future directions. In McHale

J. P.

Cowan

P. A.

(Eds.), Understanding how family-level dynamics affect children’s development: Studies of two-parent families (pp. 93–106). San Francisco, CA: Jossey-Bass.

15.

Davies

P. T.

Cummings

E. M.

(1994). Marital conflict and child adjustment: An emotional security hypothesis. Psychological Bulletin, 116, 387–411. doi:10.1037/0033-2909.116.3.387

16.

Dornbusch

S. M.

Mont-Reynaud

Ritter

P. L.

Chen

Steinberg

(1991). Stressful events and their correlates among adolescents of diverse backgrounds. In Colten

M. E.

Gore

(Eds.), Adolescent stress: Causes and consequences (pp. 111–130). New York, NY: Aldine De Gruyter.

17.

Feinberg

M. E.

(2003). The internal structure and ecological context of coparenting: A framework for research and intervention. Parenting: Science and Practice, 3, 95–131. doi:10.1207/S15327922PAR0302_01

18.

Feinberg

M. E.

Kan

M. L.

Hetherington

E. M.

(2007). The longitudinal influence of coparenting conflict on parental negativity and adolescent maladjustment. Journal of Marriage and Family, 69, 687–702. doi:10.1111/j.1741-3737.2007.00400.x

19.

Floyd

F. J.

Zmich

D. E.

(1991). Marriage and the parenting partnership: Perceptions and interactions of parents with mentally retarded and typically developing children. Child Development, 62, 1434–1448. doi:10.2307/1130817

20.

Gable

Crnic

Belsky

(1994). Coparenting within the family system. Family Relations, 43, 380–387. doi:10.2307/585368

21.

Ganong

L. H.

Coleman

Markham

Rothrauff

(2011). Predicting postdivorce coparental communication. Journal of Divorce & Remarriage, 52, 1–18. doi:10.1080/10502556.2011.534391

22.

Gasper

J. A. F.

Stolberg

A. L.

Macie

K. M.

Williams

L. J.

(2008). Coparenting in intact and divorced families: Its impact on young adult adjustment. Journal of Divorce & Remarriage, 49, 272–290. doi:10.1080/10502550802231924

23.

Grych

J. H.

Fincham

F. D.

(1990). Marital conflict and children’s adjustment: A cognitive–contextual framework. Psychological Bulletin, 108, 267–290. doi:10.1037/0033-2909.108.2.267

24.

Hayes

A. F.

(2009). Beyond Baron and Kenny: Statistical mediation analysis in the new millennium. Communication Monographs, 76, 408–420. doi:10.1080/03637750903310360

25.

Bentler

P. M.

(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6, 1–55. doi:10.1080/10705519909540118

26.

Huston

T. L.

McHale

Crouter

A. C.

(1986). When the honeymoon’s over: Changes in the marriage relationship over the first year. In Gilmour

Duck

(Eds.), The emerging field of personal relationships (pp. 109–132). Hillsdale, NJ: Erlbaum.

27.

Kelly

J. B.

(2007). Children’s living arrangements following separation and divorce: Insights from empirical and clinical research. Family Process, 46, 35–52. doi:10.1111/j.1545-5300.2006.00190.x

28.

Kline

R. B.

(2005). Principles and practice of structural equation modeling (2nd ed.). New York, NY: Guilford.

29.

Little

T. D.

(1997). Mean and covariance structures (MACS) analyses of cross-cultural data: Practical and theoretical issues. Multivariate Behavioral Research, 32, 53–76. doi:10.1207/s15327906mbr3201_3

30.

Little

T. D.

Card

N. A.

Slegers

D. W.

Ledford

E. C.

(2007). Representing contextual effects in multiple-group MACS models. In Little

T. D.

Bovaird

J. A.

Card

N. A.

(Eds.), Modeling contextual effects in longitudinal studies (pp. 121–147). Mahwah, NJ: Erlbaum.

31.

Little

T. D.

Cunningham

W. A.

Shahar

Widaman

K. F.

(2002). To parcel or not to parcel: Exploring the question, weighing the merits. Structural Equation Modeling, 9, 151–173. doi:10.1207/S15328007SEM0902_1

32.

Little

T. D.

Slegers

D. W.

Card

N. A.

(2006). A non-arbitrary method of identifying and scaling latent variables in SEM and MACS models. Structural Equation Modeling, 13, 59–72. doi:10.1207/s15328007sem1301_3

33.

Maccoby

E. E.

Mnookin

R. H.

(1992). Dividing the child: Social and legal dilemmas of custody. Cambridge, MA: Harvard University Press.

34.

Madden-Derdich

D. A.

Leonard

S. A.

(2000). Parental role identity and fathers’ involvement in coparental interaction after divorce: Fathers’ perspectives. Family Relations, 49, 311–318. doi:10.1111/j.1741-3729.2000.00311.x

35.

Margolin

Christensen

John

R. S.

(1996). The continuance and spillover of everyday tensions in distressed and nondistressed families. Journal of Family Psychology, 10, 304–321. doi:10.1037/0893-3200.10.3.304

36.

Margolin

Gordis

E. B.

John

R. S.

(2001). Coparenting: A link between marital conflict and parenting in two-parent families. Journal of Family Psychology, 15, 3–21. doi:10.1037/0893-3200.15.1.3

37.

McHale

Lauretti

Talbot

Pouquette

(2002). Retrospect and prospect in the psychological study of coparenting and family group process. In McHale

Grolnick

(Eds.), Retrospect and prospect in the psychological study of families (pp. 127–165). Mahwah, NJ: Erlbaum.

38.

Minuchin

(1974). Family and family therapy. Cambridge, MA: Harvard Press.

39.

Pollet

S. L.

Lombreglia

(2008). A nationwide survey of mandatory parent education. Family Court Review, 46, 375–394. doi:10.1111/j.1744-1617.2008.00207.x

40.

Preacher

K. J.

Hayes

A. F.

(2008). Contemporary approaches to assessing mediation in communication research. In Hayes

A. F.

Slater

M. D.

Snyder

L. B.

(Eds.), The Sage sourcebook of advanced data analysis methods for communication research (pp. 13–54). Thousand Oaks, CA: Sage.

41.

Preacher

K. J.

Kelley

(2011). Effect size measures for mediation models: Quantitative strategies for communicating indirect effects. Psychological Methods, 16, 93–115. doi:10.1037/a0022658.supp

42.

Rindskopf

(1984). Using phantom and imaginary latent variables to parameterize constraints in linear structural models. Psychometrika, 49, 37–47. doi:10.1007/BF02294204

43.

Russell

(1994). Coparenting early school-age children: An examination of mother-father interdependence within families. Developmental Psychology, 30, 757–770. doi:10.1037/0012-1649.30.5.757

44.

Rutter

(1994). Family discord and conduct disorder: Cause, consequence, or correlate? Journal of Family Psychology, 8, 170–186. doi:10.1037/0893-3200.8.2.170

45.

Schoppe-Sullivan

S. J.

Mangelsdorf

S. C.

Frosch

C. A.

McHale

J. L.

(2004). Associations between coparenting and marital behavior from infancy to the preschool years. Journal of Family Psychology, 18, 194–207. doi:10.1037/0893-3200.18.1.194

46.

Schrodt

(2010). Coparental communication with nonresidential parents as a predictor of couples’ relational satisfaction and mental health in stepfamilies. Western Journal of Communication, 74, 484–503. doi:10.1080/10570314.2010.512282

47.

Schrodt

(2011). Stepparents’ and nonresidential parents’ relational satisfaction as a function of coparental communication in stepfamilies. Journal of Social and Personal Relationships, 28, 983–1004. doi:10.1177/0265407510397990

48.

Schrodt

Afifi

T. D.

(2007). Communication processes that predict young adults’ feelings of being caught and their associations with mental health and family satisfaction. Communication Monographs, 74, 200–228. doi:10.1080/03637750701390085

49.

Schrodt

Baxter

L. A.

McBride

M. C.

Braithwaite

D. O.

Fine

(2006). The divorce decree, communication, and the structuration of co-parenting relationships in stepfamilies. Journal of Social and Personal Relationships, 23, 741–759. doi:10.1177/0265407506068261

50.

Schrodt

Braithwaite

D. O.

(2011). Coparental communication, relational satisfaction, and mental health in stepfamilies. Personal Relationships, 18, 352–369. doi:10.1111/j.1475-6811.2010.01295.x

51.

Schrodt

Ledbetter

A. M.

(2007). Communication processes that mediate family communication patterns and mental well-being: A mean and covariance structures analysis of young adults from divorced and nondivorced families. Human Communication Research, 33, 330–356. doi:10.1111/j.1468-2958.2007.00302.x

52.

Schrodt

Ledbetter

A. M.

(2012). Parental confirmation as a mitigator of feeling caught and family satisfaction. Personal Relationships, 16, 146–161.

53.

Schrodt

Miller

A. E.

Braithwaite

D. O.

(2011). Ex-spouses’ relational satisfaction as a function of coparental communication in stepfamilies. Communication Studies, 62, 272–290. doi:10.1080/10510974.2011.563453

54.

Shimkowski

J. R.

Schrodt

(2012). Coparental communication as a mediator of interparental conflict and young adult children’s mental well-being. Communication Monographs, 79, 48–71. doi:10.1080/03637751.2011.646492

55.

Stright

A. D.

Bales

S. S.

(2003). Coparenting quality: Contributions of child and parent characteristics. Family Relations, 52, 232–240. doi:10.1111/j.1741-3729.2003.00232.x

56.

Van Egeren

L. A.

Hawkins

D. P.

(2004). Coming to terms with coparenting: Implications of definition and measurement. Journal of Adult Development, 11, 165–178. doi:10.1023/B: JADE.0000035625.74672.0b