Journal of Family Psychology 2003, Vol. 17, No.

2, 169 180

Copyright 2003 by the American Psychological Association, Inc. 0893-3200/03/$12.00 DOI: 10.1037/0893-3200.17.2.169

Family and Legal Indicators of Child Adjustment to Divorce Among Families With Young Children
Marsha Kline Pruett, Tamra Y. Williams, Glendessa Insabella, and Todd D. Little
Yale University
This research used structural equation modeling to examine relations among family dynamics, attorney involvement, and the adjustment of young children (0 6 years) at the time of parental separation. The article presents baseline data (N 102 nonresidential fathers and N 110 primary caretaking mothers) from a larger longitudinal study. Results showed that the effects of parental conict on child outcomes were mediated by paternal involvement, the parent child relationship, and attorney involvement. A scale assessing parental gatekeeping yielded two signicant factors: Spouses Inuence on Parenting and Positive View of Spouse. Paternal involvement was related to childrens adaptive behavior, whereas negative changes in parent child relationships predicted behavior problems. Mothers who experienced greater psychological symptomatology were less likely to utilize an attorney, which in turn predicted greater internalizing problems in their children.

Despite a generation of divorce research providing evidence of the developmental risks faced by children of divorcing families, we still know relatively little about families with very young children. Yet these families represent one of the most vulnerable subgroups of separating and divorcing families. The challenges of infancy, toddlerhood, and preschool developmental eras are heightened by parental separation and divorce, as the childs needs for security, trust, autonomy, and patience that characterize this early period of life meet head on with the exhaustion and stress that often accompany divorce. Qualitative research suggests that divorcing families with young children face signicant challenges and risks, which result in children under age 6 experiencing substantial emotional distress and adjustment problems at the time of the marital dissolution (Wallerstein & Blakeslee, 1989; Wallerstein & Kelly, 1980). Young children are likely to appraise the reasons for the divorce less realistically, to blame themselves more readily, to feel anxiety about abandonment, and to be less likely to utilize the protective resources of other

Marsha Kline Pruett, Tamra Y. Williams, Glendessa Insabella, and Todd D. Little, Division of Law and Psychiatry, School of Medicine, Yale University. Tamra Y. Williams is now at Riverbend Community Health, Concord, New Hampshire, and Todd D. Little is now at the Department of Psychology, University of Kansas. The research for the Collaborative Divorce Project was supported by the Smith-Richardson Foundation, along with the Pritzger Foundation and the Roslyn and Jerry Meyer Foundation. We also wish to acknowledge the collaborative support of the Connecticut Support Services Divisions of the Superior Court, Family Division. Correspondence concerning this article should be addressed to Marsha Kline Pruett, Division of Law and Psychiatry, School of Medicine, Yale University, 34 Park Street, New Haven, Connecticut 06508. E-mail: Marsha.Pruett@po.state.ct.us

people (Hetherington, Bridges, & Insabella, 1998). Furthermore, approximately 55% of the divorced or separated women in the United States with children under 6 years of age live below the poverty line (Teachman & Paasch, 1994). These children spend more hours in day care (of often less than optimal quality) and are less likely to be cared for by their father when their mother is working (Whiteside, 1995). Because parents with young children are often younger themselves, they may be less educated and less mature than their older counterparts. Given the stresses that young children and their divorcing parents face, it is not surprising that divorcing families comprise the largest consumers of courtrelated family services (Depner, Cannata, & Ricci, 1995; M. K. Pruett & Bailey, 1998). However, the long-term sequelae of these stresses are less clear. Recent meta-analytic research of 67 studies conducted throughout the 1990s involving children across a broad range of ages (preschool to college age) reveals that disparities in academic achievement, psychological adjustment, and self-concept between children with divorced and married parents have increased since the 1980s (Amato, 2001). Unlike studies of older children, which have shown consistent negative effects of divorce on childrens and adolescents behavior, social interactions, psychological wellbeing, and academic performance (Amato, 2001; Cherlin et al., 1991; Guidubaldi, Perry, & Nastasi, 1987; McLanahan & Teitler, 1999), the evidence linking marital dissolution to longer term outcomes among younger children is less consistent. On the one hand, younger children may be at greater risk for long-term social and emotional adjustment problems than older children (Zill, Morrison, & Coiro, 1993). Reliable data show elevated clinical symptoms at adolescence among children whose parents divorced when they were young, regardless of the socioeconomic or cultural status of the family (Achenbach & Edelbrock, 1983). On the other hand, Amato and Keith (1991) reported in their metaanalysis that the negative effects of divorce on measures of
169

170

PRUETT, WILLIAMS, INSABELLA, AND LITTLE

academic achievement, behavior, psychological adjustment, self-concept, and social competence appeared to be less for young children than for older ones. A follow-up metaanalysis (Amato, 2001) did not include preschool children but showed stronger negative effects on academic achievement for primary school than secondary school children, and weaker effects for the younger children on psychological adjustment variables. Clarke-Stewart, Vandell, McCartney, Owen, and Booth (2000) reported data from the National Institute of Child Health and Human Development Study of Early Child Care, which assessed over 1,000 children from ages 0 3 years from intact, never married, separated, and divorced families. Parent and child functioning were assessed before and after marital disruption. When maternal education and family income were controlled in statistical analyses, children from divorced, separated, or never married families performed more poorly across age groups only on tests of cognitive ability. Not surprisingly, maternal depression, anxiety, and stimulation/support of the child but not marital statuswere signicant predictors of child adjustment. Concerns about young children and divorce often center on the childs capacity to form secure attachments to both parents in the context of coparenting arrangements. It is possible that young children of divorce are at risk for attachment disorders in early childhood (Solomon & George, 1999). Focusing on the young childs relationship to both parents rather than attachment per se, K. D. Pruett and Pruett (1999) found that children 6 years or younger were concerned about the security and stability of their relationship with both parents and about their own physical and emotional safety. Wolman and Taylors (1991) research suggests that 5 years after divorce disputes, psychological effects on the child can persist with sufcient severity to require psychotherapy or medical treatment. Given the dearth of knowledge about young children in divorcing contexts, it remains imperative to identify the factors that maximize childrens chances of successful coping and adjustment during separation and post-divorce.

tal conict has been associated specically with internalizing and externalizing behavior problems (Buehler et al., 1998), self-blame, and shame in children (Grych & Fincham, 1993). Children from high-conict families also have been found to exhibit signs of stress and fearfulness, poorer interpersonal skills, insecure attachments, and generalized insecurities (Davies & Cummings, 1994; Johnston & Roseby, 1997; K. D. Pruett & M. K. Pruett, 1999). Equally important for young children, marital and parental conict are consistent precursors to poorer parent child relationships after divorce, with less discipline and higher discord noted (Kelly, 1998, 2000; Kline et al., 1991; Tschann, Johnston, Kline, & Wallerstein, 1990). Commonly, parental conict increases when parents are negotiating their rights and obligations (e.g., economic, coparenting) through the legal system (Hetherington, 1993). The divorce process within the institutional context of the judicial system can be described as a culture of litigation in which parents opposing positions become exacerbated through the adversarial process (M. K. Pruett & Bailey, 1998). The system augments family struggles that have neither clear facts nor standards on which to base decisions when two parents have differing viewpoints. The result is a metastatic process in which conict spreads and feeds on itself, sustaining or increasing the negative family dynamics the divorce was intended to dissipate. The role of attorneys cuts both ways during the legal process: Having an attorney can help parents navigate smoothly, or it can contribute to their nancial strains and stresses attendant to the divorce process (M. K. Pruett & Jackson, 1999).

Parent-Child Relationships
Clinical vulnerabilities of parents become clinical realities once family dynamics get entangled in the legal system. Separated and divorced adults consistently report greater psychological distress than married or never married adults (Hope, Power, & Rodgers, 1999; Kitson & Morgan, 1990; Wade & Cairney, 2000). In turn, depression and anxiety can lead to a diminished capacity to parent and poor adjustment in children (Clarke-Stewart & Hayward, 1996; Pett, Wampold, Turner, & Vaughan-Cole, 1999). Disruptions in warm but authoritative parenting capacities or a conicted relationship with the primary residential parent have been associated with lower academic achievement, less social competence, increased internalizing and externalizing problems, and lower self-esteem (DeGarmo & Forgatch, 1999; Kelly, 2000; Tschann et al., 1990).

Predictors of Child Adjustment

It is generally accepted that two critical factors negatively impacting children of divorce are marital and parental conict and destabilized parent child relationships (e.g., see summaries of Amato, 2000; Kelly, 2000). In addition to their inuences on child adjustment, marital and parental conict have been shown to have differential effects on the mother child and father child relationships (Cox, Paley, & Harter, 2001; Frosch, Mangelsdorf, & McHale, 2000).

Paternal Involvement
The role of the nonresidential parent in child adjustment is also believed to be salient, but the ndings show a complex pattern of interconnected factors. Because mothers still tend to be the primary caretakers of children (Demo & Acock, 1993; Lamb & Oppenheim, 1989; Pleck, 1997), researchers have typically equated the status of nonresidential parent with the father. In one study, children from lower conict families reported a better adjustment to the divorce,

Parental Conict
Childrens capacities to adjust to their parents divorce are seriously compromised when the children are exposed to ongoing parental conict (Bolgar, Zweig-Frank, & Parish, 1995; Davies & Cummings, 1994; Emery, 1994; Johnston & Roseby, 1997; Kline, Johnston, & Tschann, 1991; Petersen & Zill, 1986; Vandewater & Lansford, 1998). Paren-

SPECIAL ISSUE: YOUNG CHILDRENS ADJUSTMENT TO DIVORCE

171

and their adjustment was even better when there was a high level of father involvement (Amato & Rezac, 1994). In their meta-analysis of 63 studies of noncustodial fathers, Amato and Gilbreth (1999) found that authoritative parenting style (supportive and nonthreatening parental control) was signicantly associated with better child outcomes such as higher academic achievement and fewer externalizing and internalizing problems. In the divorce literature, diminished fathering has traditionally been the focus of empirical work. Only recently have researchers also examined the depth and breadth of paternal involvement after divorce, not to mention the creativity many men have applied to sustain their relationships with their children. In the general fathering literature, a link is consistently found between active fathering and adaptive child behavior, especially in social and academic domains (see K. D. Pruett, 2000). Thus, a critical dimension of the divorce process remains how to maximize quality parent child relationships and minimize parental conict after divorce, especially during the divorce process itself.

in divorced or separated couples with children. Although the majority of research on gatekeeping has focused on married families, divorce and post-divorce conict may provide fertile ground for furthering our understanding of the family dynamics inherent in gatekeeping. Strict gatekeeping may result in less involvement by the nonresidential parent, more primary parent child conict, and feelings of insecurity in the child regarding his or her relationship with the less seen parent (usually the father) (Doherty, 1998; Kelly, 2000). However, the effects of gatekeeping as discussed in the divorce literature have remained primarily at the theoretical level. In divorcing and separating families, the role played by level of parental conict may be a critical variable for consideration. Primary caretaker parents may use involvement with the child as a reward or punishment depending on how they were treated by their ex-spouses during the marriage. Such ideas have not been expressly tested, nor have the constructs of gatekeeping in separating/divorcing families been empirically investigated as they have been in married, dual earner samples (Allen & Hawkins, 1999).

Gatekeeping
Especially among families with young children, a critical element of parental conict and paternal involvement is the gatekeeping that occurs between parents. Gatekeeping refers both to facilitative and inhibitory functions exercised by one or both parents that determine who will have access to their children, and the nature of that access. The psychological literature has emphasized the primary role of mothers in raising young children, and their role as gatekeepers. Maternal gatekeeping has been dened as a set of beliefs and behaviors that inhibit a collaborative effort between fathers and mothers by limiting the mens opportunity for caring and rearing of their children (Allen & Hawkins, 1999). The mother, in her customary role as primary caretaker, becomes the monitor, permission giver, and controller of the fathers involvement with the child and the form of that involvement (Grossman, Eichler, & Winickoff, 1980). During and after pregnancy, the willingness of the mother to foster or support her husbands or partners involvement with their child is key to his level of involvement. The importance of mothers attitudes in gatekeeping is evident from research showing that 60% 80% of mothers do not want their husbands more involved in childrearing, as such involvement would change the balance of power in the marriage and the important role mothers ascribe to themselves (Lamb & Oppenheim, 1989; Pleck, 1997). In their exploration of the role of maternal attitudes on paternal involvement, Beitel and Parke (1998) found that when mothers perceived their partners as motivated to engage in child care responsibilities and, to a lesser extent, as competent to do so, fathers were more involved in child care. When parents divorce in order to break the spousal connection, the sharing of childrearing provides an opportunity and a responsibility to remain connected. Negotiating who does what with regard to child care is a complicated task, in married or intact relationships (C. P. Cowan & Cowan, 1992) and may be equally difcult, albeit in different ways,

Purpose of the Study

The family law literature is replete with acknowledgments that the current legal system regarding divorce is costly in terms of families (particularly the childrens) emotional and economic well-being (e.g., Medved, 1989; Sarat & Felstiner, 1995). This study was motivated by the view that the custody context provides a unique setting in which to initiate preventive interventions with families during this vulnerable transition. Although review articles and meta-analytic studies summarize current knowledge regarding divorce and child outcomes (Amato, 2000, 2001; Kelly, 2000), the literature lacks a cohesive conceptualization of how specic variables that have been found to be related to specic aspects of child functioning operate among young children. The direct and indirect relations among variables related to child functioning are still undetermined for very young children and their parents. The present study is part of a larger longitudinal investigation examining the effects of a collaborative divorce intervention for families with children ages newborn through 6 years, on child, parent, and legal outcomes. Using structural equation modeling, we offer a model of the relations among key divorce dynamics in families with young children, including parental and legal variables, as well as child outcomes prior to the collaborative intervention. A goal of the study was to test the hypothesis that after controlling for child age and parental distress, parental conict would be predictive of negative changes in parent child relationships, stricter gatekeeping, and less father involvement. We also expected a negative relationship between maternal gatekeeping and paternal involvement. Further, we anticipated that parental symptomatology would be predictive of a poorer parent child relationship. With respect to child outcomes, we expected father involvement to predict child adaptive behavior, whereas parent child distance and discord would predict child behavioral

172

PRUETT, WILLIAMS, INSABELLA, AND LITTLE each assessment. We used information only from the initial assessment for this study because follow-up data are not yet available.

problems. We further expected that parental conict would be associated with the use of attorneys (as a proxy of the adversarial climate) and that the utilization of an attorney would have an additional negative impact on child behavior problems. Finally, since much divorce research has assessed womens reports or mens, but not both, we tested whether the relations among the variables in the model were the same for both mothers and fathers of young children.

Measures Background and Family Climate Variables

Three background variables, length of marriage/relationship, length of separation, and parents socioeconomic status, were examined. However, in the interest of parsimony and because of their lack of inuence in the context of this study, they were not included in the primary model. Age of the child was calculated at the time of the baseline assessment. Parental conict. Conict between parents was scored as the average of multiple measures of parental conict drawn from items in the Acrimony Scale (Emery, 1997), the Discuss and Share Decision-Making Scale (Ahrons, 1981), and the Content of Conict Checklist (Johnston, 1996). From the Acrimony Scale we included the items from the Relationship with Former Spouse subscale that assess areas of potential conict between separated or divorced parents (e.g., Visitation is a problem between me and their mother/father and I feel hostile toward my childrens mother/father). Items are rated on a 4-point scale. Questions from the Discuss and Share Decision-Making Scale address issues of hostility and lack of cooperation between spouses (e.g., How often is the underlying atmosphere [between spouses] one of hostility or anger? and Do you and your spouse have basic differences of opinion about issues related to childrearing?). Items were rated on a 5-point Likert scale. The Content of Conict Checklist is a 17-item measure of the extent of parental disagreement regarding childrearing issues, such as coparenting and access arrangements, discipline, daily care, and trust and support for the other parents involvement with the child. Each item is measured on a 5-point Likert scale (1 false and 5 true). Internal consistency of the composite parental conict construct was high (Cronbachs .90). Parental symptomatology. To assess the levels of parent symptomatology, we employed the 53-item, self-report Brief Symptom Inventory (Derogatis, 1993). The inventory assesses nine primary symptom dimensions, including somatization, obsessive-compulsive thoughts and behaviors, interpersonal sensitivity, depression, anxiety, hostility, phobic anxiety, paranoid ideation, and psychoticism. Parents rated how much discomfort each symptom had caused them in the past 3 months using a 4-point scale ranging from 1 (not at all) to 4 (extremely) ( .97).

Method Participants
The Project was designed to assess the effectiveness of a comprehensive intervention and prevention program for children, ages newborn through 6 years of age, and their families during the initiation of divorce proceedings for married couples or during the development of parenting plans and child support arrangements for nonmarried couples. Court clerks, parents attorneys, the Family Court judges, or Family Services personnel from two Connecticut court districts cooperated with the Collaborative Divorce Project in recruiting participants, as either intervention recipients or comparison families. Eligible families were invited to participate voluntarily in an assessment of an innovative court project designed to help families with young children steer a course through the legal system with less adversarial and expensive legal costs, as well as improved coparenting relationships and childrens psychosocial adjustment. Inclusion criteria included (a) a child 6 years or under in the family who was the biological child of the parties, (b) no substantial history of substance use, and (c) no signicant history of physical spousal or child abuse within the family. Families were enrolled at the beginning of their legal proceedings. The majority of families had just separated, and the others were in the planning phase of doing so. Of the 161 families in our longitudinal sample, we included in this study only those in which mothers were the primary residential parent. The nal sample included data from 102 fathers and 110 mothers (85% Caucasian, 4% Latino or Hispanic, 11% other). Of the children, 59% were boys and 41% were girls. The parents relationships averaged approximately 8 years, ranging from 0.5 to 26 years. Nonmarried couples (n 23 mothers, n 20 fathers) reported themselves as signicantly involved with one another; all but two lived with their partner at the time of conception. Average educational attainment for parents was generally high school graduation, a year of college, or some specialized training. Yearly incomes were measured in discrete intervals. Mothers yearly income averaged between $18,001 and $25,000, whereas fathers income averaged $25,001 to $50,000. Incomes for both mothers and fathers ranged from under $5,000 per year to over $75,000 per year, yielding an economically diverse sample.

Intermediary Variables
Father involvement. To assess level of father involvement, we used Ahronss (1981) 10-item checklist of Nonresidential ParentChild Involvement in this sample of nonresidential fathers. Various parental responsibilities are queried on a 5-point scale (1 never, 2 rarely, 3 sometimes, 4 often, 5 always) to reect how involved the nonresidential father has been with his child. Responsibilities include discipline, dressing and grooming, holiday celebrations, running errands for the child, religious and moral training, and taking the child for recreational activities ( .93). Parents noted when an item was not relevant to them or their child. Negative changes in parent child relationship. To assess changes in the parent child relationship, we used Maccoby, Mnookin, and Depners (1993) 10-item, 5-point Likert measure of the parent child relationship developed for the Stanford Child Custody Study. Parents reported on changes in the emotional

Procedure
Once both parents consented to participate, they completed questionnaires regarding family demographics, parents and childrens adjustment, legal outcomes, custody arrangements, nonresidential parents involvement, and quality of the coparenting and parent child relationships. Mothers and fathers completed each measure, except for the dependent measure of childrens adaptive behavior. Interviews with a trained clinician were conducted over the telephone, typically with custodial parents, to obtain ratings of adaptive behavior with the Vineland Screener (Sparrow, Balla, & Cicchetti, 1984). Parents were paid $50 for their participation at

SPECIAL ISSUE: YOUNG CHILDRENS ADJUSTMENT TO DIVORCE distance in their relationship with their child, as well as changes in expectations for their child, play time, patience, consistency, and the childs compliance since parental separation. Higher scores indicate more distance and difculty between parents and their child ( .84). Use of an attorney. To indicate whether an attorney was used by each parent we used a dichotomous variable, with 0 no attorney involved and 1 attorney involved. Each parent in the study chose either to move through the court process pro se (representing themselves) or to hire an attorney. A total of 64% of fathers and 69% of mothers had retained and utilized an attorney. Attorney involvement was not correlated with socioeconomic status, which thus had no effect on hiring a lawyer. Gatekeeping. Maximum-likelihood extraction with an oblique rotation was used to factor analyze the 19-item measure of the gatekeeping measure developed for the Collaborative Divorce Project. An optimal three-factor solution was retained. Correlations among the three factors ranged from .07 to .36, indicating a distinct set of gatekeeping constructs. Three items comprised each factor. The rst factor, Enhancing Spouses Parenting ( .78), assessed the degree to which parents valued the importance of the other parent and fostered their (ex)-spouses involvement in their childs life. Items that loaded on this factor included the following: It is part of my job as a parent to positively inuence my childs relationship with his/her other parent; It is my job to help my spouse be the best parent he/she can be to our child; and In order to best take care of my child, it is important for me to positively inuence my childs relationship with my spouse. Theoretically, it is assumed that such valuing would be an important determinant in the decision to open the gate to, facilitate, the childs access to the other parent. The second factor, Positive View of Spouse ( .73), contained items measuring the positive feelings about the other parent from the marriage and the effects of affection for a child on a parents view of his or her spouse. Factor items included the following: During my marriage, the closer I felt to my child, the closer I felt to my spouse; During my marriage, my spouse helped me be the best parent I could be; and The closer I feel to my child now, the closer I feel to my spouse. This factor speaks to motivation for gate opening or closing. We theorized that when a parent feels the other parent was a positive force in the marriage and in the childs life (e.g., a positive identication between child and other parent exists), he or she will be more inclined to leave the gate open after divorce. The third factor, Spouses Inuence on Parenting ( .68), measures parents assessment of the support they received from a spouse during the marriage and its impact on their past and present parenting. It contains the items: During my marriage, the support I received from my spouse affected my ability to parent my child; The way my spouse treated me during the marriage (good or bad) affected my condence as a parent; and My feelings about my spouse affect my ability to parent my child. This factor, too, speaks to the motivations for gate opening or closing. When a parent views the other spouse as powerful and as having used that power to support or hinder their own parenting ability or condence, they will be inclined to monitor access to the child after separation and divorce more strictly or openly. There is a potential to repay the other spouse (positively or negatively) for their support during the marriage.

173

as the Vineland Screener (Sparrow et al., 1984). The Vineland Screener ( .97) provides an adaptive behavior composite measure of child functioning in four skill areas: Communication (e.g., speaks in full sentences); Daily Living Skills (e.g., washes and dries face without assistance); Socialization (e.g., follows school and facility rules); and Motor Skills (e.g. walks as a primary means of getting around). Three different versions of the Screener assess age-appropriate skills for infants/toddlers ages newborn to 2 years, preschoolers ages 3 to 5 years, and school-age children (6 to 12 years). Each domain was assessed in the various versions, except for Motor Skills which are not assessed for children 6 years and older. Responses to items are scored as 2 yes, usually, 1 sometimes or partially, 0 no, never, N no opportunity for behavior to be exhibited, and DK dont know. Vineland scores were standardized across all participants. Internalizing and externalizing symptoms. Measures of internalizing behaviors and externalizing behaviors were obtained with the Child Behavior Checklist (CBCL; Achenbach & Edelbrock, 1983) given to each parent. Two versions were used according to the age of the child (2 to 3 years or 4 years and up). CBCL scores were not obtained for the youngest children in our sample. In the toddler version (Achenbach, 1992), internalizing behaviors were summed statements from both the Anxious/Depressed scale of 11 items (e.g., clings to adults or too dependent) and the Withdrawn scale of 14 items (e.g. doesnt get along with other children). Externalizing behaviors reect the sum of two scales, Aggressive Behavior, consisting of 15 items (e.g., hits others), and Destructive Behavior, consisting of 11 items (e.g., gets into everything). In the older child version of the CBCL, internalizing behaviors (n 31 items) are drawn from three scales: Withdrawn (e.g., would rather be alone than with others), Somatic Complaints (e.g., feels dizzy), and Anxious/Depressed (e.g., feels worthless or inferior). Externalizing behaviors include 33 items regarding Delinquent Behavior (e.g., lying or cheating) and Aggressive Behavior (e.g., threatens people). Scores from both problem composites were standardized across participants. These two scales of child behavior problems have high internal consistency (Internalizing, .82; Externalizing, 92) in the current sample and have demonstrated testretest reliability (Achenbach, 1991). It is also known that these two domains of childrens behavior problems are signicantly intercorrelated (r .60).

Analytic Procedures
Missing values for all variables ( 5% of the data) were imputed using a saturated regression equation (i.e., each variable is regressed on the matrix of all possible variables within the data set, except the variable itself). Outliers were also identied using the saturated regression equation (i.e., any data point outside the 99% condence interval surrounding the regression line) and were reweighted to normalize their inuence (for details, see Lopez & Little, 1996; Tabachnick & Fidell, 1989). We employed a two-group (mothers vs. fathers) path analysis with LISREL 8.5 (Joreskog & Sorbom, 2001) to test our substan tive hypotheses. We also examined mean-level differences using means structures analyses with LISREL (Little, 1997). Conducting the analyses separately for mothers and fathers had the dual advantage of identifying any differences between the two parents as well as providing a cross-validation framework for all estimated paths. To evaluate the overall model t, we relied on the maximum likelihood chi-squared statistic for which a nonsignicant value is desired. A nonsignicant chi-square value for the model indicates that the specied path relationships reproduce the observed

Dependent Variables: Child Outcomes

Adaptive behaviors. Vineland scores were obtained from a shortened version of the Vineland Adaptive Behavior Scale known

174

PRUETT, WILLIAMS, INSABELLA, AND LITTLE .08 .26** .17 .33** .03 .05 .05 .20* .13 .08 .70** 12

variance covariance relationships accurately and efciently. Between-group differences (i.e., comparisons between mothers and fathers) were examined using the chi-square difference test. For this test, the signicance of the drop in t is evaluated to determine whether constraining any paths between males and females to be equal in magnitude is warranted. A nonsignicant drop in t indicates that any constrained paths are statistically equivalent.

.11 .28** .30** .30**

.04 .02 .02 .12 .02 .18 .55**

11

.33** .18 .05 .19

.19 .02 .19 .08 .14 .09 .02

10

.03 0.9 0.6 0.5

Results
For the purposes of independent replication of our analyses, the means, standard deviations, and correlations among the variables are presented in Table 1. In general, all constructs were well differentiated and uncorrelated with each other, with the exception of the Internalizing and Externalizing behavior scales from the Child Behavior Checklist (r .60), which have been well-documented to be highly correlated. The purpose of the path analysis was to evaluate how well our hypothesized model predicting child and parental outcomes t the data set and to test for motherfather differences using a two-group path analysis model.

.09 .07 .04 .04 .09 .18 .01

.04 .02 .19 .03

.14 .21* .07 .18

.06 .08 .03 .08 .02 .13 .25**

0.7 0.5 4.1 0.9 3.3 1.7 0.8 0.6 2.0 0.6 1.9 0.6 2.8 0.8 Note: Data for fathers are above the diagonal, and data for mothers are below the diagonal. CBCL *p .05. **p 01. Child Behavior Checklist. 1.8 0.8 2.3 1.0

.16 .10 .30** .03

.21* .35** .19* .02 .03 .04 .01

.19* .01 .13 .05

Table 1 Summary Statistics of Path Model Variables

The means and variances of mothers and fathers selfreports were equivalent for half of the variables, with exceptions noted below. The mean differences offer additional information about the variables for interpretation purposes but had no effect on the actual path analyses. Nonresidential fathers felt they were signicantly more involved with their children than perceived by the mothers. Fathers reported a more negative change in their relationships with their children than did custodial mothers. Nonresidential fathers had higher means on gatekeeping variables than did the mothers across all three domains. Thus, fathers reported that mothers were stricter gatekeepers than vice versa, in accord with our theory. In addition, variances were lower for mothers reports as compared with fathers reports for the negative changes in parent child relationship and the Vineland scores. Again, the slightly lower variances of these two measures did not affect any individual differences relationships. As noted above, the intercorrelations among variables indicated autonomous constructs. Generally, the intercorrelations among variables for mothers and fathers were equal across gender groups.

.14 .03 .32** .22*

.17 .33** .06 .06 .24* .14 .01

.03 .37** .43**

.17 .14 .16 .37** .13 .07 .21* .37**

.07 .24* .07

.39** .19* .21* .05 .17 .13 .01 .08

.06 .23* .44**

.07 .16 .01 .30** .17 .05 .19* .28**

.16 .05 .06 .08 .03 .47** .02 .01

.10 .02 .16

Structural Relations and Gender Differences Among the Variables

The results of the best-tting model assessing the relations among our variables of interest are presented in Figure 1. The nal model revealed an excellent t, 2(98, N 212) 97.5, p .49. As children became older, their involvement with their fathers increased ( .17, p .05), and their adaptive behavior skills increased ( .42, p .01). In addition to the strong direct effect of child age on the Vineland, we

Childs age Parents symptomatology Parental conict Negative changes parentchild relationship 5. Father involvement 6. Enhancing spouses parenting 7. Positive view of spouse 8. Spouses inuence on parenting 9. Use of attorney 10. Vineland Adaptive Behavior 11. CBCL Internalizing 12. CBCL Externalizing

Variable

1. 2. 3. 4.

Fathers M SD Mothers M SD

3.2 1.6

0.8 0.6

1.9 0.6

2.4 0.7

3.5 0.8

4.4 0.7

Mean and Variance-Level Differences

.13 .18 .14 .02 .08 .11 .16

2.6 1.0

2.5 1.0

1.2 0.4

.01 0.7

.00 0.7

.01 0.8

.00 0.8

SPECIAL ISSUE: YOUNG CHILDRENS ADJUSTMENT TO DIVORCE

175

Figure 1. Path relationships among family and legal context variables and child outcomes. Chi square analysis revealed an excellent model t, 2(98, N 42) 97.5, p .49. All beta weights are signicant at the .01 level, with the exception of betas .18 and lower, which were signicant at the .05 level. Correlations within the path model were signicant at the .01 level ( .23) and at the .05 level ( .16). For mean differences listed, negative scores denote fathers reporting higher than mothers on scale. Solid lines reect relations that were signicant and equivalent for both mothers and fathers. Dotted lines depict paths that are signicant only for nonresidential fathers; dashed lines represent signicant effects only for mothers. P-C parent child relationship; CBCL Child Behavior Checklist.

found a mediated effect of age through increased nonresidential father involvement that also enhanced childrens adaptive behaviors ( .15, p .05). These paths were equivalent across both mothers and fathers reports. As hypothesized, parents symptomatology was also an important factor within the model. Greater symptomatology for either parent was associated with negative changes in the parent child relationship ( .39, p .01). In addition, mothers with more symptoms were less likely to use the services of an attorney ( .23, p .01) and more likely to perceive their spouses support or lack thereof as an

inuence on their parenting (Spouses Inuence on Parent.27, p .01). Fathers psychological well-being ing) ( did not affect gatekeeping or the likelihood of attorney involvement. We had expected that after controlling for the childs age and parental symptomatology, parental conict would be predictive of less involvement of nonresidential fathers with their children, gatekeeping, and negative changes in the parent child relationship. Across both mothers and fathers groups, increased levels of parental conict predicted less nonresidential father involvement ( .37, p .01).

176

PRUETT, WILLIAMS, INSABELLA, AND LITTLE

Additionally, higher levels of parental conict predicted a less positive view of spouse ( .24, p .01). Only by fathers accounts did increased levels of parental conict predict difculty between nonresidential fathers and their children ( .43, p .01). Parental conict had no effect on the mother child relationship. As expected, higher levels of marital conict also increased the likelihood of involving an attorney in the divorce or custody proceedings ( .22, p .01). Neither parental conict nor parental symptomatology directly predicted childrens behavior problems. For both mothers and fathers, negative changes in the parent child relationship served as the best predictor of childrens problem behaviors, for both internalizing ( .22, p .01) and externalizing ( .31, p .01) symptoms. For nonresidential fathers, the discord and distance within the father child relationship mediated the effect of parental conict on childrens problem behaviors. Mothers reported fewer child internalizing symptoms, such as anxiety, depression, or somatic complaints, when attorneys were involved in their legal case ( .17, p .05); however, this relationship did not hold true for fathers.

Discussion
This research used structural equation methodology to model possible relations among family and legal variables and outcomes for young children of separating and divorcing families. The proposed model yielded an excellent t to the data and mostly cross-validated across mothers and fathers, with a few notable exceptions (shown in Figure 1). Overall, the models were far more similar than different, indicating family processes that are descriptive from either parents viewpoint. A primary objective of this study was to test the hypothesis that after controlling for child age and parental distress, parental conict would predict less father involvement, stricter gatekeeping, and more negative parent child relationships. Negative changes in the parent child relationship were expected to be predictive of child behavioral problems, whereas parental involvement was expected to be related to childrens adaptive skills. Also, we predicted that maternal gatekeeping would be associated with less father involvement. With regard to use of attorneys, we expected greater use among higher conict couples and, in turn, more behavior problems among the children. The outcomes of our nal model supported most of these premises. First, parental conict was indirectly related to child adjustment through parental involvement, the parent child relationship, and attorney involvement. In general, older children had more involved fathers, which was predictive of better communication, socialization, and daily living skills. This link between father involvement and childrens adaptive skills is consistent with research on involved fathering in intact families (K. D. Pruett, 2000). However, parental conict was an important predictor of father involvement, such that higher conict was associated with less paternal involvement. This nding is in accord with previous research showing that in the face of conict, fathers are more likely to withdraw from their

children than in situations of lower parental conict (Block, Block, & Gjerde, 1986; P. A. Cowan & Cowan, 1987; Howes & Markman, 1989). One pathway in the model traces parental conict to lower father involvement, which in turn was associated with lower overall adaptive functioning by the child. The long shadow of marital conict that affects child development (Kline et al., 1991) is thus evident in the postdivorce parental conict among these families with young children. Parental conict predicts less father involvement, ultimately affecting young childrens early socialization and acquisition of necessary life skills. This link is likely to be especially salient among families with young children, where fathers own withdrawal in the face of parental conict could get exacerbated by maternal inuences on fathers opportunities for regular contact with their children and participation in child-rearing responsibilities (Braver, Wolchik, Sandler, Fogas, & Svetina, 1991). We theorized that more conict might lead mothers to bar fathers from certain kinds of involvement with their children, exercising their gatekeeping functions. Parental conict also was inversely related to one of our three gatekeeping factors: Positive View of Spouse. The more parents fought, the less likely they were to experience an afnity for their partner despite their shared child. This gatekeeping factor seems to capture a positive identication process through which spouses judge the support they received from one another and extend their positive feelings about their child to each other. This kind of connection underlies positive coparenting, as it helps parents look beyond their conicts to the shared role and bond of having a child. Conict undermines this aspect of positive family connection. Parental conict led to reports of negative changes in the parent child relationship, but for fathers only. This nding is not consistent with some previous research comprised largely of older children and focused on mothers decreased warmth and parental authority (e.g., Kline et al., 1991; see also Kelly, 2000) and may be a function of the very young ages of the children. The relatively consistent proximity and intimacy mothers reported to have with their children, and the childrens neediness as commensurate with their developmental levels, may have buffered the mother child relationships, at this point in time, early in the divorcing process. However, the nding is consistent with research indicating that fathers withdraw more from their children under conditions of higher conict (see Kelly, 2000). One aspect of this parent child measure is a question about perceived distance between parent and child since separation. That aspect may be tapping into similar constructs explored in other research. Negative changes in both the mother child and father child relationships were predictive of greater internalizing and externalizing problems in young children, a common nding in studies with older children (Roseby & Johnston, 1998; Tschann et al., 1990; see also Kellys, 1998, discussion). Whereas parental conict negatively impacted the father child relationship, maternal psychological symptomatology was the best predictor of negative changes in the mother

SPECIAL ISSUE: YOUNG CHILDRENS ADJUSTMENT TO DIVORCE

177

child relationship after separation. These parents reported greater difculty in being consistent with their childs demands, maintaining consistent routines, being patient with their child, and making time to play with their child. Hence, it was not the coparenting, but rather mothers own internal well-being that most powerfully affected their parenting capacities. In addition, mothers with greater symptomatology were more likely to endorse the gatekeeping factor measuring their spouses inuence on their parenting, both during the marriage and in its aftermath. This nding could stem from these more vulnerable mothers perceptions that their husbands had more inuence over their parenting. Alternatively, it could suggest an actual behavior in which fathers extend more support and assert more inuence on their spouses parenting in the marriage when they perceive the mothers as less able to parent in a healthy way on their own. Either way, the difculty that less psychologically healthy mothers have in parenting their young children is manifested in reliance on the ex-spouses support and views. For mothers and fathers, parental symptomatology indirectly inuenced child behavioral outcomes through the negative changes in the parent child relationship. This nding is consistent with research which shows that parental distress and vulnerability may lead to diminished parenting capabilities, which in turn is associated with poorer child outcomes (Clarke-Stewart & Hayward, 1996; Pett, Wampold, Turner & Vaughan-Cole, 1999; Tschann et al., 1990).

research with these new gatekeeping factors, we cannot ascertain how the presence of these factors affects child outcomes and under what conditions. Moreover, none of the factors was simply correlated with father involvement within the path model. Thus, it remains for further analyses to better understand how gatekeeping operates in this sample and in what ways the family dynamics tapped herein play out during the separation and divorce process.

Attorney Involvement
Unexpectedly, maternal psychological symptomatology had indirect effects on child behavioral outcomes when mothers used an attorney in their divorce. Mothers with more psychological symptoms were less likely to have hired an attorney, which predicted greater internalizing problems in their children. More healthy mothers may have utilized attorney expertise to foster protective actions regarding their children, helping to organize and contain the mothers stress related to the divorce and/or separation. In contrast, the less healthy mothers may have taken the divorce process upon themselves and, hence, upon their children. This nding held true when socioeconomic status was controlled, indicating that the choice of using an attorney was not solely an economic decision. These results have important implications for the role of attorneys in a collaborative divorce model. Many divorcing parents determine their interests and the interests of their children with the aid of legal counsel. This research suggests that when mothers who experience increased psychological vulnerability work with an attorney during the divorce process, the attorney may help absorb some of the negative effects of that vulnerability on child functioning, especially as the parents stress increases during the legal process. This model demonstrates how marital conict affects young child outcomes in families with low to moderate conict. In addition, it implicates gatekeeping as part of the family dynamic in divorce that also contains changes in father involvement and parent child relationships. However, data are only the rst step in a series needed to better understand the dynamics of families with young children in the legal system.

Operationalizing the Gatekeeping Construct

The hypothesis pertaining to gatekeeping was not borne out in the model. Factor analysis of the gatekeeping measure designed for this study revealed three distinct subcomponents: Enhancing Spouses Parenting, Positive View of Spouse, and Spouses Inuence on Parenting. Allen and Hawkins (1999) discussed three conceptual dimensions of gatekeeping: maternal identity conrmation, differentiated family roles, and standards and responsibility, which they acknowledge are derived from research on intact relationships. Like the latter dimension, the concepts we targeted also were focused on parental, especially maternal, attitudes (see also Beitel & Parke, 1998), but they differ in important ways. We were focused on the dimensions of gatekeeping pertaining specically to the dynamics between separating spouses that might lead to more liberal or stricter gatekeeping. We focused on the balance between negative views of the parent as partner and the potential for maintaining a positive view of the partner as parent to a shared child. We found that Enhancing Spouses Parenting, or belief in the importance of fostering the other parents relationship with the child, did not contribute signicantly to the overall model. Parental conict and maternal psychological symptomatology predicted less Positive View of Spouse and greater Spouses Inuence on Parenting, respectively. Yet these factors did not prove to be related to child outcomes. It is not surprising that the connections between parental conict, maternal symptomatology, and gatekeeping factors do not extend to child outcomes. Until we conduct further

Implications for Application and Public Policy

These ndings lend impetus for movements to improve clinical and psycholegal practice and create social changes in areas of divorce. The results provide information about a relatively unknown population of special importance in the eld of public health and policy our countrys youngest children. Given the common experience of parental separation before marriage or after divorce, these young children will live much of their early life in the shadows of their parents conict, separations, and coparenting. Understanding which family dynamics inuence their positive adaptation will assist in the development of support programs and interventions. With these data, we are beginning to look at processes of psychology and law as they are intertwined during divorce,

178

PRUETT, WILLIAMS, INSABELLA, AND LITTLE

rather than just existing in tandem. How the introduction of an attorney into the process impacts psychological and legal outcomes has important implications for the ways in which people choose to conduct their legal process. Increasing awareness of potential avenues for inuencing the legal system is important to families, mental health professionals, legal professionals, and the public policy leaders who have been calling for change in the current system on the basis of individual anecdotes and collective knowledge and experience (Pruett & Jackson, 1999), but without empirical backing. Empirical backing provides the basis for legislative and social change. We stand ready to examine how the dynamics elucidated in this study change over time, especially in light of an intervention implemented through the legal system. This study presented baseline data from longitudinal research that included a collaborative divorce intervention and a model of family dynamics prior to the intervention. Future analyses from these data will focus on how the collaborative intervention alters family dynamics to affect child and legal outcomes. Indeed, one of the limitations of the present study is that bidirectional relationships between the variables could not be assessed at this point given that only baseline data were analyzed. We did test for different models and these pathways were the only ones borne out by this structural equation model. The model will be tested with the same sample in the future to see whether and how pathways change for the 6-month and 15-month follow-up periods. The intervention will attempt to change the institutional context in which divorce occurs and to examine the impact such changes have on child and family well-being. In so doing, it will help inform policymakers interested in judicial reform about the types of substantive change that are possible and effective within the legal system. In addition to child outcomes, more attention will be given to legal outcomes. Because only cases where the father was the nonresidential parent were analyzed in order to assess maternal gatekeeping, legal outcomes such as custody designations (i.e. joint legal, joint physical vs. joint legal, sole physical custody) were not included in the model. In the next iteration of these data, data from attorneys regarding legal costs, length of divorce, and utilization of court services will be included. Tests of the model for different age groups and each gender will also be considered in forthcoming models. Within the age group sampled in the current study, it is possible that various aspects of the family dynamics may interrelate differently or predict different child outcomes for younger children in comparison with older children. Clearly, this study represents a tip of the iceberg. In future models we hope to examine the impact of frequent overnights and frequent court visits. Step-by-step we begin to shed light on the black hole of young childrens adaptation to parental separation and divorce and on the legal processes that support or hinder parents in their efforts to steer their children through this early life transition.

References
Achenbach, T. M. (1991). Manual for the Child Behavior Checklist/4 18 and 1991 prole. Burlington, VT: University of Vermont, Department of Psychiatry. Achenbach, T. M. (1992). Manual for the Child Behavior Checklist/23 and 1992 prole. Burlington, VT: University of Vermont, Department of Psychiatry. Achenbach, T. M., & Edelbrock, C.S. (1983). Manual for the Child Behavior Checklist and revised child behavioral prole. Burlington: VT: University Associates in Psychiatry. Ahrons, C. R. (1981). The continuing coparental relationship between divorced spouses. American Journal of Orthopsychiatry, 51, 415 429. Allen, S. M., & Hawkins, A. J. (1999). Maternal gatekeeping: Mothers beliefs and behaviors that inhibit greater father involvement in family work. Journal of Marriage and the Family, 61, 199 212. Amato, P. R. (2000). The consequences of divorce for adults and children. Journal of Marriage and Family, 62, 1269 1288. Amato, P. R. (2001). Children of divorce in the 1990s: An update of the Amato and Keith (1991) meta-analysis. Journal of Family Psychology, 15, 355370. Amato, P. R., & Gilbreth, J. G. (1999). Nonresident fathers and childrens well-being: A meta-analysis. Journal of Marriage and the Family, 61, 557573. Amato, P. R., & Keith, B. (1991). Parental divorce and the wellbeing of children: A meta-analysis. Psychological Bulletin, 110, 26 46. Amato, P. R., & Rezac, S. (1994). Contact with residential parents, interparental conict, and childrens behavior. Journal of Family Issues, 15, 191207. Beitel, A. H., & Parke, R. D. (1998). Paternal involvement in infancy: The role of maternal and paternal attitudes. Journal of Family Psychology, 12, 268 288. Block, J., Block, J., & Gjerde, P. (1998). The personality of children prior to divorce: A prospective study. Child Development, 57, 827 840. Bolgar, R., Zweig-Frank, H., & Parish, J. (1995). Childhood antecedents of interpersonal problems in young adult children of divorce. Journal of the American Academy of Child and Adolescent Psychiatry, 34, 143150. Braver, S. H., Wolchik, S. A., Sandler, I. N., Fogas, B. S., & Svetina, D. (1991). Frequency of visitation by divorced fathers. Differences in reports by fathers and mothers. American Journal of Orthopsychiatry, 61, 448 454. Buehler, C., Krishnakumar, A., Stone, G., Anthony, C., Pemberton, S., Gerard, J., & Barber, B. K. (1998). Interparental conict styles and youth problem behaviors: A two-sample replication study. Journal of Marriage and the Family, 60, 119 132. Cherlin, A. J., Furstenberg, F. F., Chase-Lansdale, P. L., Kiernan, K. E., Robins, P. K., Morrison, D. R., & Teitler, J. O. (1991). Longitudinal studies of effects of divorce on children in Great Britain and the United States. Science, 252, 1386 1389. Clarke-Stewart, K. A., & Hayward, C. (1996). Advantages of father custody and contact for the psychological well-being of school-age children. Journal of Applied Developmental Psychology, 17, 239 270. Clarke-Stewart, K. A., Vandell, D. L., McCartney, K., Owen, M. T., & Booth, C. (2000). Effects of parental separation and divorce on very young children. Journal of Family Psychology, 14, 304 326. Cowan, C. P., & Cowan, P. A. (1992). When partners become parents: The big life change for couples. New York: Basic Books.

SPECIAL ISSUE: YOUNG CHILDRENS ADJUSTMENT TO DIVORCE Cowan, P. A., & Cowan, C. P. (1987). Mens involvement in parenthood: Identifying the antecedents and understanding the barriers. In P. Berman & F. Pederson (Eds.), Mens transitions to parenthood: Longitudinal studies of early family experience (pp. 145174). Hillsdale, NJ: Erlbaum. Cox, M. J., Paley, B., & Harter, K. (2001) Interparental conict and parent-child relationships. In J. H. Grych & F. D. Fincham (Eds.), Interparental conict and child development: Theory, research and applications (pp. 249 272). New York, NY: Cambridge University Press. Davies, P. T., & Cummings, E. M. (1994). Marital conict and child adjustment: An emotional security hypothesis. Psychological Bulletin, 116, 387 411. DeGarmo, D. S., & Forgatch, M. S. (1999). Contexts as predictors of changing maternal parenting practices in diverse family structures: A social interactional perspective of risk and resilience. In E. M. Hetherington (Ed.), Coping with divorce, single parenting, and remarriage: A risk and resiliency perspective (pp 227252). Mahwah, NJ: Erlbaum. Demo, D. H., & Acock, A. C. (1993). Family diversity and the division of domestic labor: How much have things really changed? Family Relations, 42, 323331. Depner, C. E., Cannata, K., & Ricci, I. (1995). Report 4: Mediated agreements on child custody and visitation. Family and Conciliation Courts Review, 33, 87109. Derogatis, L. R. (1993). BSI: Brief Symptom Inventory: Administration, scoring, and procedures manual. Minneapolis, MN: National Computer Systems. Doherty, W. J. (1998). Responsible fathering: An overview and conceptual framework. Journal of Marriage and the Family, 60, 277292. Emery, R. E. (1994). Renegotiating family relationships: Divorce, child custody and mediation. New York: Guilford Press. Emery, R. E. (1997). The revised acrimony scale. Unpublished manuscript, University of Virginia, Charlottesville, VA. Frosch, C. A., Mangelsdorf, S. C., & McHale, J. L. (2000). Marital behavior and the security of preschoolerparent attachment relationships. Journal of Family Psychology, 14, 144 161. Grossman, F., Eichler, L., & Winickoff, S. (1980). Pregnancy, birth, and parenthood. San Francisco: Jossey-Bass. Grych, J., & Fincham, F. (1993). Childrens appraisal of marital conict: Initial investigations of the cognitive contextual framework. Child Development, 64, 215230. Guidubaldi, J., Perry, J. D., & Nastasi, B. K. (1987). Growing up in a divorced family: Initial and long-term perspective on childrens adjustment. Annual Review of Applied Social Psychology, 7, 202237. Hetherington, E. M. (1993). An overview of the Virginia Longitudinal Study of Divorce and Remarriage with a focus on early adolescence. Journal of Family Psychology, 7, 39 56. Hetherington, E. M., Bridges, M., & Insabella, G. M. (1998). What matters? What does not? Five perspectives on the association between marital transitions and childrens adjustment. American Psychologist, 53, 167184. Hope, S., Power, C., & Rodgers, B. (1999). Does nancial hardship account for elevated psychological distress in lone mothers? Social Science and Medicine, 29, 381389. Howes, P., & Markman, H. (1989). Marital quality and child functioning: A longitudinal investigation. Child Development, 60, 1044 1051. Johnston, J. R. (1996). Prevention of parent or family abduction of children through early identication of risk factors: Stage II Part B. (Final Report to the Ofce of Juvenile Justice & Delinquency Prevention). Washington, DC: U.S. Department of Justice.

179

Johnston, J. R., & Roseby, V. (1997). In the name of the child: A developmental approach to understanding and helping children of conicted and violent divorce. New York: Free Press. Joreskog, K. G., & Sorbom, D. (2001). LISREL 8.5: Users guide. Chicago: Scientic Software International. Kelly, J. B. (1998). Marital conict, divorce, and childrens adjustment (Child Psychiatric Clinics of North America Monograph). In K. D. Pruett & M. K. Pruett (Eds.), Child custody issues (pp 259 272). New York: W. B. Saunders. Kelly, J. B. (2000). Childrens adjustment in conicted marriage and divorce: A decade review of research. Journal of the American Academy of Child and Adolescent Psychiatry, 39, 963973. Kitson, G. C., & Morgan, L. A. (1990). The multiple consequences of divorce: A decade review. Journal of Marriage and the Family, 52, 913924. Kline, M., Johnston, J. R., & Tschann, J. M. (1991). The long shadow of marital conict: A model of childrens postdivorce adjustment. Journal of Marriage and the Family, 53, 297309. Lamb, M. E., & Oppenheim, D. (1989). Fatherhood and father child relationships: Five years of research. In S. H. Cath & A. Gurwitt (Eds.), Fathers and their families (pp. 1126). Hillsdale, NJ: Analytic Press. Little, T. D. (1997). Mean and covariance structures (MACS) analyses of cross-cultural data: Practical and theoretical issues. Multivariate Behavioral Research, 32, 5376. Lopez, D. F., & Little, T. D. (1996). Childrens action-control beliefs and emotional adjustment in the social domain. Developmental Psychology, 32, 299 312. Maccoby, E. E., Mnookin, R. H., & Depner, C. E. (1993). Stanford Child Custody Study: Child, 1984 1990, Data Set 28 30. Los Altos, CA: Sociometrics. McLanahan, S., & Teitler, J. (1999). The consequences of father absence. In M. E. Lamb (Ed.), Parenting and child development in nontraditional families (pp. 83102). Mahwah, NJ: Erlbaum. Medved, D. (1989). The case against divorce. New York: D. I. Fine. Petersen, J. L., & Zill, N. (1986). Marital disruption, parent child relationships, and behavior problems in children. Journal of Marriage and the Family, 48, 295307. Pett, M. A., Wampold, B. E., Turner, C. W., & Vaughan-Cole, B. (1999). Paths of inuence of divorce on preschool childrens psychosocial adjustment. Journal of Family Psychology, 13, 145164. Pleck, J. H. (1997). Paternal involvement: Levels, sources, and consequences. In M. E. Lamb (Ed.), The role of the father in child development (3rd ed., pp. 66 103). New York: Wiley. Pruett, K. D. (2000). Fatherneed: Why father care is as essential as mother care for your child. New York: Free Press. Pruett, K. D., & Pruett, M. K. (1999). Only God decides: Young childrens perceptions of divorce and the legal system. Journal of the American Academy of Child and Adolescent Psychiatry, 38, 1544 1550. Pruett, M. K., & Bailey, C. (1998). Divorcing families with young children in the family services division: Proles and impact of services (Final report prepared for the Divorce and the Culture of Litigation Project). New Haven, CT: Yale University, Consultation Center. Pruett, M. K., & Jackson, T. (1999). The lawyers role during the divorce process: Perceptions of parents, their young children, and their attorneys. Family Law Quarterly, 33, 283310. Roseby, V. & Johnston, J. R. (1998). Children of Armageddon: Common developmental threats in high-conict divorcing families (Child Psychiatric Clinics of North America Monograph). In

180

PRUETT, WILLIAMS, INSABELLA, AND LITTLE marital transition among mothers: Results from a national panel study. The Journal of Nervous and Mental Diseases, 188, 741 750. Wallerstein, J. S., & Blakeslee, S. (1989). Second chances: Men, women and children a decade after divorce. New York: Basic Books. Wallerstein, J. S., & Kelly, J. B. (1980). Surviving the breakup: How children and parents cope with divorce. New York: Basic Books. Whiteside, M. F. (1995). An integrative review of the literature pertinent to custody of children ve years of age and younger (Executive Summary to the Statewide Ofce of Family Court Services, San Francisco, CA). Ann Arbor, MI: Center for the Family. Wolman, R., & Taylor, K. (1991). Psychological effects of custody disputes on children. Behavioral Sciences and the Law, 9, 399 417. Zill, N., Morrison, D. R., & Coiro, M. J. (1993). Long term effects of parental divorce on parent-child relationships, adjustment, and achievement in young adulthood. Journal of Family Psychology, 7, 91103.

K. D. Pruett & M. K. Pruett (Eds.), Child custody issues (pp. 295310). New York: W. B. Saunders. Sarat, A., & Felstiner, W. L. F. (1995). Divorce lawyers and their clients: Power and meaning in the legal process. New York: Oxford University Press. Solomon, J., & George, C. (1999). The effects on attachment of overnight visitation on divorced and separated families. In J. Solomon and C. George (Eds.), Attachment disorganization (pp. 243264). New York: Guilford Press. Sparrow, S., Balla, D. A., Cicchetti, D. V. (1984). Vineland Adaptive Behavior Scales: Interview edition. Circle Pine, MN: American Guidance Services. Tabachnick, B. G., & Fidell, L. S. (1989). Using multivariate statistics (2nd ed.). New York: Harper Collins. Teachman, J. D., & Paasch, K. M. (1994). Financial impact of divorce on children and their families. The Future of Children, 4, 63 83. Tschann, J. M., Johnston, J. R., Kline, M., & Wallerstein, J. (1990). Conict, loss, change and parent child relationships: Predicting childrens adjustment during divorce. Journal of Divorce and Remarriage, 13, 122. Vandewater, E. A., & Lansford, J. E. (1998). Inuences of family structure and parental conict on childrens well-being. Family Relations, 47, 323330. Wade, T. J., & Cairney, J. (2000). Major depressive disorder and

Received December 21, 2001 Revision received May 23, 2002 Accepted June 7, 2002