Skip over navigation

Journal Issue: Children and Divorce Volume 4 Number 1 Spring/Summer 1994

High-Conflict Divorce
Janet R. Johnston

Children's Adjustment to Divorce-Related Conflict

Effects of Parental Conflict on Children

The studies reviewed in this section are concerned with the general relationship between parental conflict and child functioning in the broader population of divorcing families. Here, we are interested in understanding under what conditions children are affected by interparental conflict, and to what extent and in what ways they are disadvantaged. These are crucial questions with social policy implications. An extensive literature has been generated during the past two decades about the effects of interparental conflict on children, mostly within intact but also within divorcing families (see reviews by Emery; Grych and Fincham; Depner, Leino, and Chun).20-22 A common shortcoming of these numerous studies is that, with few exceptions (for example, Cherlin and colleagues) they have not looked at the long-term outcomes of living with interparental conflict, nor have they looked at how conflict can interfere with the normal developmental course of children.23

Most early studies did not discriminate among conflict domain, tactics, and hostility (for example, Rutter).24 More recently, however, there have been attempts to compare the effects on children of exposure to domestic violence with those of living in highly conflictual but nonviolent homes. Further distinctions are now being made about the effects on children of witnessing violence compared with being directly abused (see review by Jaffe, Wolfe, and Wilson).25 The consistent conclusions of these reviews indicate that interparental hostility and physical aggression are moderately associated with more behavioral problems, emotional difficulties, and reduced social competence in children, compared with nonconflictual families. Where gender differences have been investigated, boys tend to show more overt behavioral disturbance and girls tend to have more covert emotional disturbance. These findings indicate that, in general, children who are exposed to physical aggression between parents are more symptomatic than those who experience nonviolent interparental discord and that this symptomatology is even more pronounced in children who have been directly abused.

It is important to note that, while children from hostile and aggressive families, as a group, have more adjustment problems than normally expected, the range of individual outcomes is broad. Some children do very well despite these adverse environments; others appear to be reactive and negatively affected. What makes for these differences? In studies of divorced families, it has been found that interparental hostility and aggression mostly have indirect effects on children, mediated by the quality of the parent-child relationship, with the child's relationship to the mother being more predictive of child adjustment than is the relationship with the father.7,26,27 This finding implies that a good parent-child relationship can buffer children from interparental conflict. In addition, some characteristics of the individual child (for example, a more adaptable temperament, higher intelligence, and better coping skills) are indicators of more resilience. The problem is, however, that conflict between spouses tends to erode a couple's capacity to cooperate in the care and guidance of children. As a consequence of divided parental authority and lack of respect given to one another, parenting tends to become more problematic: discipline is more coercive, and expectations are more inconsistent, all of which are predictive of more negative and distant parent-child relationships and an increase in children's emotional and behavioral problems.28

Children's Adjustment in High-Conflict Divorced Families

It is argued here that family laws as well as court policies are often justified by research findings from the broad population and are insufficiently backed by studies of the special subgroup of the divorcing population to which they are most frequently applied, that is, to families of high-conflict divorce. High-conflict divorce is characterized by all of the following: intractable legal disputes, ongoing conflict over parenting practices, hostility, physical threats, and intermittent violence. To address social policy with respect to the children of high-conflict divorce, it is necessary to evaluate the extent to which child outcomes are multiply determined and to examine the relative impact of interparental hostility and physical aggression, parent psychopathology, and the custody and visitation arrangements on children's functioning. Only a few such studies can be found scattered throughout the literature. These studies are summarized here in some detail for the first time. Taken together, they begin to paint a picture of a subgroup of children at high risk who have special social policy needs.

In 1988, Brotsky, Steinman, and Zemmelman made clinical observations of 67 children (ages 1 to 15 years) from divorcing families at one, two, and four years after separation.29 Almost half of the parents were referred by the family courts for counseling because of disputed joint physical custody; the remainder were attempting voluntarily to put together a joint custody plan and had requested help. The sample ranged broadly in terms of socioeconomic status and was 80% white. At the one-year follow-up, 12 families were classified as "successful" in implementing a cooperative joint custody plan, 20 were seen as "stressed" by this endeavor, and 15 were rated as having "failed."30

These three groups of couples were distinguished by their capacity to respect and support each other's parental efforts with the child and to keep separate and modulate their own anger and ambivalence toward the ex-spouse. The "successful" and "failed" groups clearly maintained their profile at the four-year followup; four-fifths of the "stressed" group had become more cooperative and satisfied with the coparenting arrangement.

The important finding for social policy is that, at each time of observation, those parents whose joint custody arrangements were court-ordered or court-recommended were more likely to be classified as "failed" or "stressed," and their children were more likely to be symptomatic or at high risk in terms of their behavioral, emotional, and social adjustment. A cautionary note is sounded in that this study employed a small sample and did not report results based on the standardized measures that were used.

In 1989, Schaefer reported on a four-year follow-up of 83 children (ages 1 to 11 years at separation) whose parents had reached a settlement of custody through court-ordered evaluations and/or judicial decree.19 In half of the cases, the fathers had been awarded custody; these were matched with sole mother-custody cases according to age and gender of the child. The socioeconomic status was predominantly middle- to upper-middle-class, and families were predominantly white. The children's adjustment was assessed by the Child Behavior Checklist (CBCL) completed by the custodial parent.31 The parents reported their own symptomatic distress on the Brief Symptom Inventory (BSI) and the amount of coparental communication (see Table 1).9,18

The findings indicated that parents' reports of their own emotional distress were most strongly related to their reports of the children's adjustment. However, it could not be determined whether this was evidence of a true relationship or of the parents' emotional state having biased their perceptions of their children. There were few indications that coparental communication was related to the children's adjustment. The custodial parent reported significantly more internalizing symptomatology (for example, withdrawal, depression, somatic symptoms) when the children had more frequent access to the other parent. There were no significant differences found in either the boys' or girls' adjustment in mother-custody homes compared with father-custody homes on any of the measures. These findings indicate that the custodial parent's own adjustment is the best predictor of child adjustment and that children of both genders do as well in the primary care of their fathers as in the primary care of their mothers, when these arrangements have been made after careful psychological evaluations of the best match between parent and child. The value of frequent access to the noncustodial parent is, however, questionable.

Johnston, Kline, and Tschann studied 100 children (ages 1 to 12 years) whose parents were referred by the family court because of ongoing disputes over their custody and care at the time of litigation and again about two and one-half years later.32 The children were ethnically relatively diverse (62% white) and from low-middle socioeconomic backgrounds. Children's adjustment was measured by an average of both parents' reports on the CBCL.31 Interparental conflict was assessed by both parents' reports on the Conflict Tactics Scale (CTS), in which verbal aggression and physical aggression were combined (see Table 1).1 The custody arrangement was rated in terms of the legal decree, the amount of access (days per month) with the visiting parent, and the number of transitions per week between parents. At the follow-up, 35% of these children were in joint physical custody, 53% in mother custody, and 12% in father custody.

There was no clear evidence that children were better adjusted in either custody type. However, as a group, children who had more shared access to both parents in joint custody arrangements and those who had more frequent visitation with a noncustodial mother or father in sole custody situations were more emotionally and behaviorally disturbed. Specifically, they were more depressed, withdrawn, and/or uncommunicative, had more somatic symptoms, and tended to be more aggressive.

When gender groups were analyzed separately, these findings were significant for girls but, in general, not for boys. This longitudinal study showed that more verbal and physical aggression was generated between parents when children had more frequent access arrangements; consequently, these children tended to get more caught up in the interparental conflict. Whereas girls' adjustment was more adversely affected by more access to both their disputing parents, boys and older children compared to younger ones were more caught up and used in the parental conflict that was generated as a consequence of the more frequent access arrangement, and this, in turn, was related to poorer child outcomes.32

Apart from the small size of this sample, the main limitations of this study were that only averaged parental reports on the children were used as outcomes (possibly obscuring differences between mothers and fathers); that the effects of verbal aggression were not distinguished from those of physical aggression; and that parents' individual dysfunction (an alternative explanation for the findings) was not measured. Consequently, a second study was designed to correct some of these limitations.

In the second study, Johnston reported on the adjustment of 75 children (ages 3 to 12 years) in disputed-custody divorces.12 Their parents were referred from the family court because of interparental violence or because there was ongoing conflict of a nonviolent nature. This sample was diverse in socioeconomic status and 80% white. In addition to the same measures used in the first study, teachers rated the children, as did clinicians.27,32,33 Parents' individual dysfunction was measured by the BSI.18 With respect to the custody of these children, 36% were in joint physical custody, 57% were in sole mother custody, and 7% were in sole father custody.

Boys and girls appeared to differ in their adjustment to the custody and access arrangements. The overall results indicated that girls were functioning better when in the primary care of their mothers in these high-conflict and physically violent families (according to ratings by mothers, fathers, and clinicians, but not those by teachers). There was some weaker evidence that boys did better when they had more access to their fathers (according to fathers' and some teachers' ratings, but not to ratings by mothers or clinicians). However, more frequent access arrangements were associated with more concurrent aggression between parents, which offset some of the benefits of access for boys.

When all factors in this study were simultaneously analyzed, the findings were that a history of physical aggression in the family was strongly and consistently associated with emotional, behavioral, and social problems in children. It was not only directly predictive of more child disturbance, it was also associated with mother's diminished parenting, in that mothers from violent relationships were less warm and more coercive with their children. In addition, the degree of both mothers' and fathers' emotional dysfunction independently predicted child disturbance, both directly and indirectly, as it was associated with less warmth and more coerciveness in parenting. Apart from the small sample size, the weakness of this study is that, because family relationships and children's functioning were assessed at only one point in time, it remains undetermined what is cause and what is effect.

Each of the above studies (conducted by Brotsky and colleagues, by Schaefer, Johnston, and colleagues, and by Johnston) were of children of high-conflict divorce, whose parents had failed mediation, undergone evaluations, or had court-imposed settlements.12,19,29,32 According to Maccoby and Mnookin, these categories represent a small proportion, the top 10th percentile, of legal conflict in the divorcing population.7

To what extent are these children seriously disturbed? In each of the studies where standardized measures of maladjustment were reported, these children scored as significantly more disturbed and were two to four times more likely to have the kinds of adjustment problems typically seen in children being treated for emotional and behavioral disturbance as compared with national norms. In general, boys were more symptomatic than girls.

Caution needs to be used in interpreting and generalizing from these findings. Each study used a relatively small sample of unknown representativeness of the population of high-conflict divorce. Though all but one of these studies were longitudinal in nature, only one used statistical analyses enabling causal directions to be tested.12,32 Only two employed multivariate analysis to examine the joint and relative effects of each factor.12,32 The effects of age of child and ethnic differences were not explored. In sum, these should be viewed as tentative correlational findings to be confirmed by further research. Despite the limitations, however, the findings of all these studies are fairly consistent. Interparental conflict after divorce (for example, verbal and physical aggression, overt hostility, distrust) and the custodial parent's emotional distress are jointly predictive of more problematic parent-child relationships and greater child maladjustment. Court-ordered joint physical custody and frequent visitation arrangements tend to be associated with poorer child outcomes, especially for girls.

It is important to note that the finding of an association between joint custody/ frequent access and poorer child adjustment appears to be confined to that small proportion of families (about one-tenth) of all divorces that are termed "high-conflict" as defined above. Several other studies of the broader population of divorcing families, where custody arrangements are generally made voluntarily by parents, indicate that joint physical custody and frequent visitation are not more detrimental to the majority of children. In some cases, especially where parents are cooperative, they are more beneficial.34-36