Abstract
This study examines main effect and interactive models of the relations between marital conflict, divorce, and parent-adult child relationships, and gender differences in these relations. Data were drawn from a longitudinal study of a community sample (N = 585). Parental marital conflict and divorce were measured from age 5 through age 17. Mother-child and father-child relationship quality at age 22 was assessed in terms of Closeness-Support and Conflict-Control. Results indicate that both marital conflict and divorce were associated with poorer quality of parent-adult child relationships. Divorce moderated the link between marital conflict and subsequent negativity in mother-child relationships, with the estimated effects being stronger in continuously married families than in divorced families, especially for women.
Keywords: conflict, divorce, gender, parent-child relations, structural equation modeling
Parent-adult child relationships continue to play a significant role in development during the early adult years (Amato & Sobolewski, 2001). Both the experience of parental divorce and exposure to chronic marital conflict have been found to be associated with low quality of parent-young adult child relationships (e.g., Amato & Afifi, 2006). Other research findings indicate that parental divorce might be problematic for parent-child relationships above and beyond the negative effects of interparental conflict (Riggio, 2004). Divorce often makes it difficult for nonresidential parents to maintain close relationships with their children (Lamb, 1999). A variety of stressful circumstances following divorce can also disrupt the quality of interactions between the residential parent and the child (Hetherington et al., 1992).
A common question that many parents considering divorce ask is whether they should stay together for the sake of their children. Parental marital conflict and divorce are usually viewed as stressful events in a child’s developmental processes. The family stress theory emphasizes the accumulation of negative events, not only a single stressor, which may result in problems for children (Amato, 1993). The notion that some children may benefit from their parents’ divorce, however, is not inconsistent with the stress perspective (Booth & Amato, 2001). Although divorce may usually be viewed as a stressful event in both an adult’s life and a child’s developmental processes, it may also present a new chance for adults to pursue more harmonious, fulfilling relationships, and a new opportunity for children’s personal growth, individuation, and well-being in a new family situation (Hetherington & Stanley-Hagan, 1999). Children of parents who engage in overt, unresolved conflict are at risk for a variety of developmental and emotional problems (Emery, 1999). Under such a condition, when a divorce occurs, these children are freed from a dysfunctional family environment and may genuinely welcome the shift to a calmer single-parent family (Booth & Amato, 2001). In other words, consistent with the stress relief hypothesis (Wheaton, 1990), children who were living with high-conflict parents may experience parental divorce as a stress relief event, which in turn, may lead to a postdivorce improvement in child outcomes (Strohschein, 2005). Research findings from studies on the basis of children’s reports suggest that children who perceive their parents’ marriage as high in conflict demonstrated better long-term adjustment if their parents divorced than those whose parents stayed together (e.g., Amato, Loomis, & Booth, 1995).
Most previous studies that have examined the interaction of parental marital conflict and divorce focused on offspring’s psychological well-being, such as anxiety and depression (Jekielek, 1998), overall happiness and psychological distress (Amato et al., 1995), and antisocial behaviors (Strohschein, 2005). Further, most research considered only short-term effects for children and adolescents. Few prospective longitudinal studies have examined long-term effects (e.g., those persisting through early adulthood) and even fewer have considered interpersonal relationship qualities as outcomes. One study that came close was conducted by Booth and Amato (2001) and focused on the interactive effects of parental marital conflict and divorce on adaptation in multiple relational domains of early adulthood. They found that parental divorce and marital conflict interactively predicted young adult children’s psychological well-being, friend support, and intimate relationships, but not parent-adult child relationships. Parental divorce appeared to have a negative influence on the quality of parent-adult child relationships regardless of the level of conflict preceding the divorce. However, Booth and Amato’s (2001) study assessed only the positive aspects of closeness, affect, and happiness in adults’ interpersonal exchanges with their parents. Their study also did not separately examine mother-child and father-child relationships. In this study, we examined the interactive effect of parental divorce and marital conflict on both the positive (support, involvement, and relationship quality) and negative (conflict, disagreement, and psychological control) aspects of parent-adult child relationships. We also examined these interactive effects separately for mothers and fathers. We hypothesized that the relation between marital conflict and adult children’s relationship qualities with parents would be moderated by divorce. For young adults from nondivorced families, marital conflict is expected to be negatively associated with parent-offspring relationship qualities to a greater degree than for young adults whose parents divorced.
An additional limitation in the studies described above is that the adult child’s gender was treated only as a control variable, not as a factor that might condition or qualify the divorce and conflict interactive effects. Gender differences in parental socialization practices show that girls are more likely to be taught to emphasize relationship-oriented goals, whereas boys are more likely to be taught to emphasize individual-focused goals (Kinsforgel & Grych, 2004). And girls, compared to boys, also have been found to be more directly involved in interparental relationships by assuming a mediating role in parental conflict (Vuchinich, Emery, & Cassidy, 1988). Thus, girls who witness parental conflict may be more sensitive to the potential harm of conflict and may be more likely to form a closer bond with their mothers after parental divorce (e.g., Arditti, 1999; Orbuch, Thornton, & Cancio, 2000). In the current study, we explored possible child gender differences in the moderation of divorce in the association between marital conflict and adult children’s relationships with parents.
An important consideration in research on divorce and child well-being is the timing of the divorce (i.e., age of the child or adolescent when the divorce occurs). However, studies of the long-term effects of parental marital conflict and divorce on the basis of the child’s age have yielded inconsistent results (e.g., Amato & Sobolewski, 2001). Parental remarriage and the quality of the remarriage also may influence adult children’s relationships with their biological parents. Remarried biological parents have been found to report less involvement with their children than parents in first marriage families because they need to invest time and energy in their new partners (Flinn, Leone, & Quinlan, 1999). Therefore, in this study, children’s age at the time of divorce and parental remarriage status were controlled in the major analyses.
A developmental perspective examines whether experiences in the family of origin, such as interparental conflict and divorce, have longlasting effects on individuals’ interpersonal functioning in adulthood (Conger, Cui, Bryant, & Elder, 2000). These effects may depend on the interactions among varied individual, family, and extrafamilial factors (Hetherington & Stanley-Hagan, 1999). On the basis of prior research, we hypothesized that parental marital conflict and divorce would both be independently associated with parent-adult child relationship quality and that marital conflict and divorce may also have interactive effects on parent-adult child relationships.
Method
Participants
The young adults and their families in this study were participants in the Child Development Project, an ongoing, multisite longitudinal study of child development (Pettit, Bates, & Dodge, 1997). Participating families were recruited from Nashville and Knoxville, Tennessee, and Bloomington, Indiana when the children entered kindergarten in 1987 or 1988. Within each site, target schools were selected in consultation with local school administrations. These schools were chosen because they were judged to be broadly representative of the make-up of the schools in each district. At kindergarten preregistration, parents were approached at random and asked if they would participate in a longitudinal study of child development. About 15% of children at the targeted schools did not preregister. These participants were recruited on the first day of school or by letter or telephone. Of those so contacted, approximately 75% agreed to participate. The initial sample of 585 participants was diverse in terms of child gender (52% boys and 48% girls) and ethnicity (81% European American, 17% African American, and 2% other ethnic groups). Although the sample was predominantly middle class, as indicated by an average Hollingshead (1979) score of 40.4 (SD = 14), a range of socioeconomic status was represented, with 9%, 17%, 25%, 33%, and 16% of the families classified in the five classes (from lowest to highest) in the Hollingshead system.
At age 22, 467 participants in the present investigation were contacted, 440 of them completed assessments about their relationships with their biological mothers, and 390 of them completed assessments about their relationships with their biological fathers. About 47.5% (n = 222) of participants were from nondivorced families, 23.3% (n = 109) were from families in which parents did not marry or they divorced before child’s age of 5, 14.8% (n = 69) were in a family in which parents divorced between child’s age of 6 to 10, and 9.9% (n = 46) were in a family in which parents divorced between child’s age of 11 to 17. There were 121 participants who reported living with their mothers currently (44 men and 23 women from nondivorced families, 32 men and 22 women from divorced families); whereas only 82 participants reported living with their fathers currently (42 men and 23 women from nondivorced families, 7 men and 10 women from divorced families). The female participants with divorced parents reported living closer to their mothers (measured by how many miles they lived from their parents; M = 75.05, SD = 262.62) than female participants with nondivorced parents (M = 313.22, SD = 1236.77). However, there was no difference for the male participants with divorced versus nondivorced parents in the distance they lived from their mothers. Further, there was no difference for both male and female participants with divorced versus nondivorced parents in the distance they lived from their fathers.
Participants with divorced parents reported visiting with their fathers less often (measured by 1 = not at all, 2 = once a year, 3 = several times per year, 4 = 1–3 times per month, 5 = once a week, 6 = more than once a week, and 7 = living together; M = 3.61, SD = 2.00 for men; M = 3.70, SD = 1.84 for women), and communicating less often with their fathers (measured by 1 = not at all, 2 = once a year, 3 = several times per year, 4 = 1–3 times per month, 5 = once a week, 6 = more than once a week, and 7 = living together; M = 4.07, SD = 2.08 for men; M = 4.36, SD = 1.88 for women) than those with nondivorced parents (M = 5.37, SD = 1.68 for men and M = 4.82, SD = 1.60 for women regarding how often they visited with their fathers; M = 5.88, SD = 1.14 for men and M = 5.54, SD = 1.19 for women regarding how often they communicated with their fathers). The divorced versus nondivorced groups did not, however, differ on how often they visited and communicated with their mothers.
Measures
Divorce or separation
At child’s age 5, parents reported their marital status. In each subsequent year (until child’s age of 17), mothers were asked if they had divorced or separated from their spouses in the last 12 months. Consistent with most empirical studies (e.g., Amato et al., 1995; Zill, Morrison, & Coiro, 1993), no distinction was drawn between parental divorce and separation. Parental divorce was coded as a dichotomous variable (ever vs. never) to distinguish between children whose parents divorced before age 17 and children whose parents remained married.
Marital conflict
At child’s ages 5, 13, and 16, marital conflict was assessed using a 9-item subscale of the Conflict Tactics Scale (Straus, 1979). Mothers were asked to recall in the last 12 months, except for child’s age 5 when mothers were asked to recall the child’s age 1 to age 4 and the child’s age 4 to age 5, to rate the conflict tactics that they and their spouses used with a 6-point scale ranging from 0 (never) to 6 (almost every day). Items included (a) verbal aggression: argued heatedly but did not yell; yelled, insulted or swore (α = .72 to .87); (b) psychological aggression: sulked or refused to talk about it; stomped out of the room or house; threatened to throw something (α = .69 to .79); and (c) physical aggression: pushed, grabbed, or shoved; and hit (α = .68 to .93). On the basis of the mothers’ reports, the scores for both mother-to-spouse and spouse-to-mother were averaged to create the final marital conflict variables. For parents who remained continuously married, the marital conflict score was calculated as the mean of the scores for child’s age of 1–4, 4–5, 13, and 16 to reflect the general level of conflict in the marriage. In cases of divorce, however, conflict was based on the mean of all conflict scores obtained before the separation. Because of the possibility that the timing between the measure of marital conflict and the measure of parent-adult child relationship outcomes might also have affected associations between parental marital conflict and parent-adult child relationship (e.g., children with parents whose marital conflict was measured more recently were expected to report lower quality of relationships with their parents than children with parents whose marital conflict was measured much earlier), the time at which the last measure of marital conflict was collected was controlled in the major analyses of this study.
Parent-adult child relationships
At age 22, target participants were interviewed about their relationships with their biological parents. Some items were adapted from the National Survey of Families and Households NSFH; (Sweet & Bumpass, 1996). Other items were developed for the CDP study. These items were factor-analyzed to guide scale development (maximum likelihood estimation with oblique rotation). Two latent constructs were identified: Closeness-Support accounted for 29% of the variance in mother-child relationships and 34% of the variance in father-child relationships; Conflict-Control accounted for 36% of the variance in mother-child relationships and 35% of the variance in father-child relationships.
The Closeness-Support construct was measured by three indicators: parental support, involvement, and global parent-child relationship quality. Parental support was indexed by young adults’ responses to three items: “how much does your mother (father) provide for your emotional needs; take care of your practical needs; and act as an advisor/mentor.” Participants rated parental support on a 5-point scale ranging from 1 = never to 5 = a lot of the time. Three items were averaged to produce a young adult’s report of the mother’s (α = .71) and the father’s (α= .81) support score. Parental involvement was measured by young adults’ responses to six items: “how often does your mother (father) talk with you about ordinary daily events in your life; things with which you are happy or satisfied; and problems with which you may be concerned;” and “how often does your mother (father) know about your personal/romantic relationships; your activities at work/school; and when you are sick or have other health problems.” Participants rated parental involvement on a 5-point scale ranging from 1 = never to 5 = very frequently. Six items were averaged to produce a young adult’s report of the mother’s (α = .88) and the father’s (α = .91) involvement score. In addition to these summary measures, young adults were also asked to rate the quality of their relationships with each parent on an 11-point scale ranging from 0 (really bad) to 10 (absolutely perfect).
The Conflict-Control construct was also measured by three indicators: disagreement, conflict, and psychological control. Disagreements between parents and young adult children included whether the parent and adult child had open disagreements over the last three months about 11 issues, such as friends, jobs, staying out at night, etc. Participants rated these 11 items as yes (1) or no (0). These items were summed to produce a young adult’s report of their disagreements with mothers (α = .78) and fathers (α = .72). Conflict between parents and young adult children was measured by one item: “how often do you argue or fight or have a lot of difficulty with your mother/father.” Participants rated this item on a 5-point scale ranging from 1 = not at all to 5 = more than once a week. An index of young adult-reported parental psychological control was derived from three items. These items (my mother or father tries to change how I feel or think about things; makes decisions for me or tells me how to run my life; and brings up my past mistakes when he/she criticizes me) were rated on a 5-point scale (1 = never to 3 = occasionally to 5 = very frequently) and were averaged to produce a young adult’s report of the maternal (α = .71) and paternal (α = .76) psychological control scores.
Covariates
The controlled variables included parents’ socioeconomic status, the children’s ethnicity, parental remarriage status, child age at the time of the divorce, and the time at which the last measure of marital conflict was administered. Parents’ socioeconomic status was measured by the average Hollingshead (1979) scores on the basis of parental education and occupational status across child’s age of 5 to 17. Participants’ ethnicity was a dichotomous variable coded as 1 = African American and 0 = others. Parental remarriage status was measured dichotomously (1 = yes and 0 = no) each year through the mother’s reports about whether she and the child’s biological father had remarried in the last 12 months.
Missing Data and Attrition
There has been 80–84% retention in the most recent data collection waves with little new attrition. We contrasted participants who completed the assessments about their relationships with their biological mothers and fathers at age 22 to those participants who had missing data on those measures. Two significant differences were found: African American participants, and those living with single mothers at age 5, were more likely than European American participants, and the participants living with both parents at age 5, to have missing data on parent-adult child relationship measures at age 22.
Results
Latent variables were constructed for marital conflict, mother-child closeness and support, mother-child conflict and control, father-child closeness and support, and father-child conflict and control. Table 1 shows the descriptive statistics and correlations among main study constructs and control variables. Structural Equation Modeling analyses were conducted to examine the fit of models in which the adult children’s relationship outcomes were predicted by marital conflict, divorce, and the control variables. The residual factor variances of latent variables of Closeness-Support and Conflict-Control in mother-child and father-child relationships were allowed to be correlated with each other to account for the possibility that other factors not explicitly included in the model may affect these endogenous variables and the correlations among them (Hargens, 1988). Models were estimated using the Amos 16 program which uses full information maximum likelihood estimation (FIML) with missing data resulting in unbiased parameter estimates and appropriate standard errors when data are missing at random (MAR). Even when the MAR assumption is not fully met, FIML estimates are generally better than estimates obtained with listwise deletion or other ad hoc methods (Acock, 2005). The full parameter estimations of the models can be found in supporting material published online.
Table 1.
Descriptive Statistics and Correlations among Demographics, Marital Conflict, Divorce, and Parent-Adult Child Relationships
| Variables | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
|---|---|---|---|---|---|---|---|---|---|---|
| 1. SES | - | |||||||||
| 2. Child gender (women) | −.08 | - | ||||||||
| 3. Child ethnicity (Black) | −.34*** | .06 | ||||||||
| 4. Child age at divorce | −.06 | .06 | −.01 | |||||||
| 5. Parental remarriage | −.16** | .08 | .07 | .29*** | - | |||||
| 6. Divorce | −.39*** | .08 | .29*** | .60*** | .56*** | - | ||||
| 7. Verbal conflict | −.08 | .01 | −.09 | .14** | .16** | .20*** | - | |||
| 8. Psychological conflict | −.15** | .02 | .03 | .04 | .22*** | .25*** | .64*** | - | ||
| 9. Physical conflict | −.17** | .01 | .24*** | −.01 | .22*** | .23*** | .42*** | .63*** | - | |
| 10. Relationship quality (M)a | .02 | −.00 | −.04 | −.02 | −.04 | −.08 | −.04 | −.09 | −.06 | - |
| 11. Received support (M) | .08 | .15** | .01 | .04 | −.02 | −.03 | −.03 | −.10 | −.01 | .61*** |
| 12. Positive involvement (M) | .11* | .33*** | −.17*** | .04 | .01 | −.05 | .03 | −.06 | −.09 | .55*** |
| 13. Conflict (M) | −.03 | .04 | −.01 | .03 | .00 | .06 | .05 | .03 | .06 | −.41*** |
| 14. Disagreement (M) | −.08 | −.06 | .11* | −.03 | −.04 | .03 | .02 | .04 | −.02 | −.14** |
| 15. Control (M) | −.05 | .02 | .10* | .04 | −.01 | .07 | .00 | .03 | −.07 | −.35*** |
| 16. Relationship quality (F)b | .20*** | −.07 | −.09 | −.21*** | −.07 | −.42*** | −.12* | −.19** | −.15** | .21*** |
| 17. Received support (F) | .26*** | .01 | −.10* | −.16** | −.14** | −.37*** | −.09 | −.18** | −.19** | .12* |
| 18. Positive involvement (F) | .25*** | .12* | −.21*** | −.12* | −.05 | −.33*** | −.08 | −.17** | −.14** | .13* |
| 19. Conflict (F) | .07 | −.03 | −.07 | −.04 | −.08 | −.16** | .00 | −.09 | −.11 | −.09 |
| 20. Disagreement (F) | .01 | −.06 | −.01 | −.09 | −.16** | −.18** | −.03 | −.04 | −.08 | −.10 |
| 21. Control (F) | .16** | −.07 | −.11* | −.01 | −.04 | −.11* | .01 | −.08 | −.09 | −.11* |
| Mean | 39.83 | .51 | .13 | 6.78 | .24 | .50 | .03 | .02 | .01 | 7.82 |
| SD | 12.36 | .51 | .34 | 2.95 | .43 | .50 | .77 | .60 | .69 | 1.90 |
| Variables | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 10. Relationship quality (M) | |||||||||||
| 11. Received support (M) | - | ||||||||||
| 12. Positive involvement (M) | .58*** | - | |||||||||
| 13. Conflict (M) | −.15** | −.12* | - | ||||||||
| 14. Disagreement (M) | −.01 | −.06 | .40*** | - | |||||||
| 15. Control (M) | −.12* | −.14** | .51*** | .52*** | - | ||||||
| 16. Relationship quality (F) | .03 | .04 | −.16** | −.23*** | −.18*** | - | |||||
| 17. Received support (F) | .35*** | .14** | −.04 | −.12* | −.06 | .67*** | - | ||||
| 18. Positive involvement (F) | .15** | .41*** | −.05 | −.25*** | −.12* | .68*** | .68*** | - | |||
| 19. Conflict (F) | .07 | −.04 | .40*** | .18** | .13* | −.08 | .12* | .07 | - | ||
| 20. Disagreement (F) | −.00 | −.14** | .23*** | .57*** | .23*** | .01 | .10* | −.01 | .46*** | - | |
| 21. Control (F) | −.01 | −.08 | .21*** | .19*** | .40*** | .03 | .16** | .14** | .55*** | .49*** | - |
| Mean | 3.42 | 3.66 | 1.74 | 1.68 | 2.05 | 7.01 | 3.03 | 3.15 | 1.46 | .94 | 1.76 |
| SD | .98 | .89 | .90 | 2.06 | .93 | 2.66 | 1.10 | 1.05 | .75 | 1.54 | .83 |
Notes: ns = 325 to 416 for marital conflict and divorce; ns = 358 to 416 for mother-child relationships; ns = 356 to 363 for father-child relationships.
(M): mother-child relationships.
(F): father-child relationships.
p < .05.
p < .01.
p < .001.
For mother-child relationship outcomes, the model fit was considered acceptable: χ2(60) = 218.08, CFI (Comparative Fit Index) = .95, RMSEA (Root Mean Square Error of Approximation) = .077, and explained 2.8% of the variance in mother-child closeness and support and 2.6% of the variance in mother-child conflict and control. Only marital conflict was marginally related to mother-child closeness and support (β= −.12, p < .10). For father-child relationship outcomes, the model fit was considered good: χ2(60) = 146.91, CFI = .97, RMSEA = .061, and explained 26.6% of the variance in father-child closeness and support and 5.2% of the variance in father-child conflict and control. Parental divorce accounted for a relatively large amount of variance in father-child closeness and support (β = −.70, p <.001), but was not associated with father-child conflict and control. Marital conflict was only marginally associated with less father-child closeness and support (β = −.11, p < .10).
To test whether the hypothesized models in which marital conflict predicted children’s relationship outcomes fit the data equally well across divorced and nondivorced groups, a stacked SEM procedure was used where a model constraining factor loadings and structural paths to be equal across divorced and nondivorced groups was compared with a model in which the factor loadings and structural paths were allowed to vary. In general, the fits of the constrained models were significantly worse than the fits of the unconstrained models: Δχ2 (28) = 340.06, p < .001 for mother-child relationships; Δχ2 (28) = 227.06, p < .001 for father-child relationships, suggesting that the hypothesized models did not fit the data equally well across parents’ marital status groups. It also should be noted that the factor loadings of parents’ marital conflict were found to be different across divorced and nondivorced groups. Verbal aggression consistently showed higher loadings for the nondivorced group than for the divorced group, yet physical aggression consistently showed higher loadings for the divorced group than for the nondivorced group. For mother-child relationship outcomes (see Figure 1 top), higher levels of marital conflict were related to more mother-child conflict and control but only for the nondivorced group and not for the divorced group. Marital conflict was not related to mother-child closeness and support for either group. Subsequent examination of the structural paths parameter by parameter revealed that only the path from marital conflict to mother-child conflict and control was significantly different for nondivorced and divorced groups. For father-child relationship outcomes (see Figure 1 bottom), marital conflict was not related to either father-child closeness and support or father-child conflict and control when conducting these analyses separately by each of the groups. Subsequent examination revealed that the path from marital conflict to father-child closeness and support, and the path from marital conflict to father-child conflict and control were not significantly different for young adults from divorced and nondivorced families.
Figure 1.
Standardized path coefficients of the multi-group structural equation models: Marital conflict predicting mother-child relationships (top, n = 218 for nondivorced group, 205 for divorced group, model fit of unconstrained model: χ2(72) = 203.75, CFI = .90, RMSEA = .066) and father-child relationships (bottom, n = 211 for nondivorced group, 162 for divorced group, model fit of unconstrained model: χ2(72) = 148.79, CFI = .93, RMSEA = .054). Numbers in parentheses refer to coefficients for divorced group. Both models controlled for SES, child ethnicity, child age at divorce, parental remarriage status, and the time at which the last marital conflict measure was collected. *p < .05, **p < .01, ***p < .001.
To test the extent to which the hypothesized models in which marital conflict predicted children’s relationship outcomes fit the data equally across gender and parents’ marital status groups, a series of four-group (men with nondivorced parents, men with divorced parents, women with nondivorced parents, and women with divorced parents) SEM analyses was conducted. In general, the fits of the constrained models were significantly worse than the fits of the unconstrained models: Δχ2 (82) = 501.39, p < .001 for mother-child relationships; Δχ2 (82) = 439.97, p < .001 for father-child relationships, suggesting that the hypothesized models did not fit the data equally well across groups. For mother-child relationship outcomes (see Table 2 left), marital conflict was related to less mother-daughter closeness and support and more mother-daughter conflict and control in nondivorced families. Marital conflict was not associated with either mother-child closeness and support or mother-child conflict and control for the other three groups. Subsequent examination revealed, however, that only the path from marital conflict to mother-daughter conflict and control was significantly different for divorced and nondivorced groups: Δχ2 (1) = 4.05, p = .044. For father-child relationship outcomes (see Table 2 right), marital conflict was not associated with either father-child Closeness-Support or with father-child conflict and control for all four groups. Subsequent examination revealed that neither the path from marital conflict to father-daughter or father-son closeness and support nor the path from marital conflict to father-daughter or father-son conflict and control was significantly different for divorced and nondivorced groups.
Table 2.
Standardized Coefficients Showing Associations between Parents’ Marital Conflict and Young Adult Children’s Relationships with Parents: Men in Nondivorced Families versus Men in Divorced Families; Women in Nondivorced Families versus Women in Divorced Families
| Mother-Child Relationships | Father-Child Relationships | |||||||
|---|---|---|---|---|---|---|---|---|
| Men | Women | Men | Women | |||||
| Nondivorced n = 113 |
Divorced n = 93 |
Nondivorced n = 105 |
Divorced n = 112 |
Nondivorced n = 108 |
Divorced n = 72 |
Nondivorced n = 103 |
Divorced n = 90 |
|
| Factor Loadings | ||||||||
| Marital Conflict | ||||||||
| Verbal | .94 | .65 | .70 | .57 | .90 | .70 | .72 | .59 |
| Psychological | .75 | .87 | .99 | .99 | .79 | .81 | .99 | .99 |
| Physical | .45 | .78 | .54 | .73 | .48 | .66 | .52 | .67 |
| Close-Support | ||||||||
| Relationship quality | .76 | .93 | 73 | .94 | .67 | 70 | .71 | .93 |
| Received support | .76 | .64 | .76 | .77 | .68 | .87 | .72 | .85 |
| Positive involvement | .61 | .56 | .78 | .79 | .64 | .89 | .74 | .88 |
| Conflict-Control | ||||||||
| Conflict | .78 | .72 | .58 | .48 | .71 | .47 | .63 | .94 |
| Disagreement | .81 | .58 | .58 | .53 | .64 | .53 | .78 | .58 |
| Psychological control | .68 | .85 | .86 | .90 | .77 | .88 | .83 | .51 |
| Structural Paths | ||||||||
| Conflict – Close-Support | .01 | −.06 | −23* | −.10 | −.12 | −.10 | −.10 | −.18 |
| Conflict – Conflict-Control | .08 | −.17 | −27* | −.05 | .14 | .02 | −.11 | −.03 |
Notes. All factor loadings were significant. Both models controlled for SES, child ethnicity, child age at divorce, parental remarriage status, and timing at which the last marital conflict measure was collected. Fit indices of unconstrained model for mother-child relationships: χ2 (146) = 276.62, CFI = .90, RMSEA = .046 Fit indices of unconstrained model for father-child relationships: χ2 (146) = 233.85, CFI = .92, RMSEA = .040.
p < .05.
Conclusion
The primary research goals of this study were to examine the main effects and interactive models of the relations between parental marital conflict, divorce, and parent-young adult child relationships, and the potential gender differences in these relations. Consistent with previous research, we found that parents’ marital conflict and divorce were both associated with adult child relationships with parents. These associations were stronger for father-child relationships than for mother-child relationships (e.g., Booth & Amato, 1994). Divorce often makes it difficult for nonresidential parents (90% of them are fathers; Amato, 2001) to maintain close ties with their children (Lamb, 1999). Further, the stressful circumstances following divorce may disrupt the quality of interactions between residential parents and their children, which in turn, may affect the quality of the relationships between parents and adult children (Hetherington et al., 1992).
Results of the study also indicated that divorce moderated the links between parents’ marital conflict and conflict and control in mother-child relationships. Specifically, for young adults whose parents remained in a high-conflict marriage, parents’ marital conflict was found to be associated with young adult children’s feeling of less closeness and support and more conflict and control with their mothers. Conversely, for young adult children from divorced families, parents’ earlier marital conflict was not associated with subsequent conflict and control experienced with their mothers. These findings support the stress relief hypothesis (Wheaton, 1990) that contends that parental divorce, even as a stressful life event, may actually ameliorate some of the negative effects of marital conflict on children’s adjustment if divorce presents escape from the more stressful environment of a high-conflict family. Furthermore, the group differences in factor loadings of parents’ marital conflict suggested that physical violence was more likely to characterize the divorced couples. This result supports the assumption that after leaving a violent or abusive marriage, mothers may be more physically and mentally healthy and thus more able to establish a healthier relationship with their children.
Finally, the moderation effect of divorce on the link between marital conflict and conflict and control in mother-child relationships was found only for women and not for men. This finding is consistent with research showing that divorced mothers and girls in divorced single-mother families are more likely to demonstrate a closer relationship following divorce than are fathers and boys (e.g., Arditti, 1999; Orbuch et al., 2000). On one hand, girls who witness parental conflict may be more sensitive to the potential harm of conflict and may be more likely to form a closer bond with their parents after parental divorce (Arditti, 1999). On the other hand, for fathers, the effect of divorce on their relationships with their children may be exacerbated because of their greater tendency to link their role as father with their role as husband (Kerig, Cowan, & Cowan, 1993). When divorce disrupts their role as husband, fathers may be more likely to withdraw from their children. However, mothers who experience divorce may compensate for the lack of a spousal bond by developing a closer relationship with their children, especially with daughters (Orbuch et al., 2000).
Despite a number of strengths to our study that bolster our confidence in these findings, several limitations warrant discussion. First, this study cannot examine whether or not there are changes in levels of conflict between parents before and after divorce. Future studies would benefit from using data on parental conflict measured consistently at multiple points in time before and after divorce. Second, the sample of the study is predominantly middle class. It will be important for future researchers to extend samples to diverse and low-income families. Third, future research is needed to explore other parent and family characteristics, such as the residential status during childhood, personalities or behavioral tendencies of the parents, and genetic factors, which may also contribute to the parent-child relationship after parental divorce. Finally, although the sample appeared to be representative of the communities from which it was drawn on the basis of demographic comparisons between the sample and the full community, this was not a probability sample, and caution should be used in generalizing the findings.
This study built on and extended previous research to examine main and interactive effects of parental marital conflict and divorce on young adults’ relationships with their parents. Moreover, this study included the role of child’s gender as a factor in the interplay of parents’ marital relationship in adult children’s relationships with their parents. The findings of the study supported a developmental perspective that encompasses the longlasting influences of experiences in the family of origin on individuals’ interpersonal functioning in adulthood (Conger et al., 2000). Furthermore, the findings that divorce moderated the links between parents’ marital conflict and mother-daughter conflict and control also supported the stress relief hypothesis (Wheaton, 1990). Taken together, the findings suggest that parent gender, child gender, and timing of parents' divorce are important considerations in efforts aimed at providing a better understanding of the link between parents’ marital conflict and qualities of parents' subsequent relationships with their adult children.
Acknowledgments
The Child Development Project has been funded by grants MH42498, MH56961, MH57024, and MH57095 from the National Institute of Mental Health, HD30572 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development, and DA016903 from the National Institute on Drug Abuse. We are grateful to the parents, children, and teachers who participated in this research. Appreciation is also expressed to Amy Rauer for comments on an earlier draft of this manuscript.
Contributor Information
Tianyi Yu, Department of Psychology, University of Michigan, Ann Arbor, MI, 2052C East Hall, 530 Church Street, Ann Arbor, MI, 48109, Phone: 734-615-0821, yutianyi@umich.edu.
Gregory S. Pettit, Department of Human Development and Family Studies, Auburn University, Auburn, AL, Phone: 334-844-3228, gpettit@auburn.edu
Jennifer E. Lansford, Center for Child and Family Policy, Duke University, Durham, NC, Phone: 217-586-2870, lansford@duke.edu
Kenneth A. Dodge, Center for Child and Family Policy, Duke University, Durham, NC, Phone: 919-613-9303, dodge@duke.edu
John E. Bates, Department of Psychological and Brain Sciences, Indiana University, Bloomington, IN, Phone: 812-855-8693, batesj@indiana.edu
References
- Acock AC. Working with missing values. Journal of Marriage and Family. 2005;67:1012–1028. [Google Scholar]
- Amato PR. Children’s adjustment to divorce: Theories, hypotheses, and empirical support. Journal of Marriage and the Family. 1993;55:23–38. [Google Scholar]
- Amato PR. Children of divorce in the 1990s: An update of the Amato and Keith (1991) meta-analysis. Journal of Family Psychology. 2001;15:355–370. doi: 10.1037//0893-3200.15.3.355. [DOI] [PubMed] [Google Scholar]
- Amato PR, Afifi TD. Feeling caught between parents: Adult children’s relations with parents and subjective well-being. Journal of Marriage and Family. 2006;68:222–235. [Google Scholar]
- Amato PR, Loomis LS, Booth A. Parental divorce, marital conflict, and offspring well-being during early adulthood. Social Forces. 1995;73:895–915. [Google Scholar]
- Amato PR, Sobolewski JM. The effects of divorce and marital discord on adult children’s psychological well-being. American Sociological Review. 2001;66:900–921. [Google Scholar]
- Arditti JA. Rethinking relationships between divorced mothers and their children: Capitalizing on family strengths. Family Relations. 1999;48:109–119. [Google Scholar]
- Booth A, Amato PR. Parental marital quality, parental divorce, and relations with parents. Journal of Marriage and the Family. 1994;56:21–34. [Google Scholar]
- Booth A, Amato PR. Parental predivorce relations and offspring postdivorce well-being. Journal of Marriage and Family. 2001;63:197–212. [Google Scholar]
- Conger RD, Cui M, Bryant CM, Elder GH. Competence in early adult romantic relationships: A developmental perspective on family influences. Journal of Personality and Social Psychology. 2000;79:224–237. doi: 10.1037//0022-3514.79.2.224. [DOI] [PubMed] [Google Scholar]
- Emery RE. Marriage, divorce, and children's adjustment. 2nd ed. Thousand Oaks, CA: Sage; 1999. [Google Scholar]
- Flinn MV, Leone DV, Quinlan RJ. Growth and fluctuating asymmetry of stepchildren. Evolution and Human Behavior. 1999;20:465–479. [Google Scholar]
- Hargens LL. Estimating multi-equation models with correlated disturbance terms. In: Long JS, editor. Common problems/proper solutions: Avoiding error in quantitative research. Newbury Park, CA: Sage; 1988. pp. 65–83. [Google Scholar]
- Hetherington EM, Clingempeel WG, Anderson ER, Deal J, Stanley-Hagan M, Hollier EA, et al. Coping with marital transitions: A family systems perspective. Monographs of the Society for Research in Child Development. 1992;57:2–3. Serial No. 227. [Google Scholar]
- Hetherington EM, Stanley-Hagan MM. The adjustment of children with divorced parents: A risk and resiliency perspective. Journal of Child Psychology and Psychiatry. 1999;40:129–140. [PubMed] [Google Scholar]
- Hollingshead AB. The Hollingshead four factor index of socioeconomic status. New Haven, CT: Unpublished paper, Yale University; 1979. [Google Scholar]
- Jekielek SM. Parental conflict, marital disruption, and children’s emotional well-being. Social Forces. 1998;76:905–935. [Google Scholar]
- Kerig PK, Cowan PA, Cowan CP. Marital quality and gender differences in parent-child interaction. Developmental Psychology. 1993;29:931–939. [Google Scholar]
- Kinsfogel KM, Grych JH. Interparental conflict and adolescent dating relationships: Integrating cognitive, emotional, and peer influences. Journal of Family Psychology. 2004;18:505–515. doi: 10.1037/0893-3200.18.3.505. [DOI] [PubMed] [Google Scholar]
- Lamb ME. Noncustodial fathers and their impact on the children of divorce. In: Thompson R, Amato P, editors. The postdivorce family: Children, parenting, and society. Thousand Oaks, CA: Sage; 1999. pp. 105–125. [Google Scholar]
- Orbuch TL, Thornton A, Cancio J. The impact of marital quality, divorce, and remarriage on the relationships between parents and their children. Marriage and Family Review. 2000;29:221–246. [Google Scholar]
- Pettit GS, Bates JE, Dodge KA. Supportive parenting, ecological context, and children’s adjustment: A seven-year longitudinal study. Child Development. 1997;68:908–923. doi: 10.1111/j.1467-8624.1997.tb01970.x. [DOI] [PubMed] [Google Scholar]
- Riggio HR. Parental marital conflict and divorce, parent-child relationships, social support, and relationship anxiety in young adulthood. Personal Relationships. 2004;11:99–114. [Google Scholar]
- Straus MA. Measuring intrafamily conflict and violence: The Conflict Tactics (CT) Scales. Journal of Marriage and the Family. 1979;41:75–88. [Google Scholar]
- Strohschein L. Parental divorce and child mental health trajectories. Journal of Marriage and Family. 2005;67:1286–1300. [Google Scholar]
- Sweet JA, Bumpass LL. The National Survey of Families and Households - Waves 1 and 2: Data description and documentation. Center for Demography and Ecology, University of Wisconsin-Madison; 1996. [Google Scholar]
- Vuchinich S, Emery R, Cassidy J. Family members as third parties dyadic family conflict: Strategies, alliances, and outcomes. Child Development. 1988;59:1293–1302. doi: 10.1111/j.1467-8624.1988.tb01498.x. [DOI] [PubMed] [Google Scholar]
- Wheaton B. Life transitions, role histories, and mental health. American Sociological Review. 1990;55:209–223. [Google Scholar]
- Zill N, Morrison DR, Coiro MJ. Long-term effects of parental divorce on parent-child relationships, adjustment, and achievement in young adulthood. Journal of Family Psychology. 1993;7:91–103. [Google Scholar]