marital disruption, parent-child relationships, and ... · marital disruption, parent-child...

Marital Disruption, Parent-ChildRelationships, and Behavior

Problems in Children

JAMES L. PETERSONNICHOLAS ZILLChild Trends, Inc.

This study examines the effects of marital disruption on children's behavior, accountingfor variations in postdisruption living arrangements and the effects of parent-child rela-tionships and marital conflict. The study is based on a 1981 national sample of 1,400 chil-dren aged 12-16. Disruption was associated with a higher incidence of several behaviorproblems, negative effects being greatest with multiple marital transitions. The negative ef-fects are lower if the child lives with the same-sex parent following divorce or maintains agood relationship with one or both parents. High, persistent conflict in intact families is

al{o related to behavior problems.

More than half of all children of elementary-school age interviewed in the recent NationalSurvey of Children said they feel afraid when theirparents have arguments. Similarly, the proportionof children reported to have had a seriously upset-ting experience is highest for those whose parents'marriages have been disrupted or whose parents'relationships are characterized by high conflict(ZiII, 1978). The same data also show that chil-dren's mental health is related to marital disrup-tion and the level of conflict in the family (Zill andPeterson, 1983).

To know why and to what extent marital dis-ruption and parental conflict affect children is of

This paper was presented at the annual meeting of theSociety for Research in Child Development, April 1983.The authors acknowledge research support from GrantsNo. MH-34707 and No. MH-38664, National Instituteof Mental Health, U.S. Department of Health andHuman Services, and from a grimt from the Foundationfor Child Development. Appreciation is also extendedto Frank F. Furstenberg, Jr., a collaborator on the Na-tional Survey of Children, and to the anonymousreferees who provided helpful and thoughtful commentson an earlier version of this article.

Child Trends, Inc., 1990 M Street, N.W., Washington,DC 20036.

practical importance to parents and to clinicianswho seek to help parents in conflict; it is relevantfor those who set policies that affect the function-ing of families; and it relates to some central theo-retical issues in the field of mental health.

A major theoretical concern is the developmentof models that explain the role of family function-ing in the development of behavior disorders inchildren. Research in this area has demonstratedthe usefulness of examining interspousal andparent-child relations as factors influencing thedevelopment of behavior problems in children(Maccoby, 1980). More specifically, marital con-flict and disruption have been found to be sourcesof behavior problems (Hetherington, Cox, andCox, 1977; Wallerstein and Kelly, 1980; Zill andPeterson, 1983; Levinger and Moles, 1979;Levitin, 1979). For example, Hetherington, Cox,and Cox (1977) found that in the year followingdivorce, children became more dependent, dis-obedient, aggressive, and demanding, and less af-fectionate. The effects peaked at about one yearand by two years had diminished for boys andlargely disappeared for girls. Parents' childrearingbehavior changed as well: parents became lessconsistent, less affectionate, and less able to con-trol their children's behavior.

Wallerstein and Kelly (1980) also found nega-tive and more lasting effects of marital disruption

Journal of Marriage and the Family 48 (May 1986): 295-307 295

- -- - - - - - -- - ---

296

on children. Five years after the divorce, one-thirdof the children in their sample were still seriouslydisturbed, and another third were having psycho-logical difficulties. Even ten years later, many ofthe children who were preschoolers at the time ofthe divorce continued to struggle with emotionalissues stemming from the divorce (Wallerstein,1984). More favorable outcomes were found whenthe child had good relationships with bothparents, and the parents were psychologicallyhealthy.

Rutter (1971), studying English children, foundthat marital discord accounted for manydeleterious effects of separation and divorce, atleast for boys. In particular, the longer the dis-cord, the greater the antisocial behavior. Thecessation of discord was followed by a decline inproblem behavior. A good relationship with oneparent mitigated but did not eliminate the effectsof parental discord, a finding also supported byHess and Camara (1979).

Looking at parent-child interactions, Schwarz(1979) suggests that parent-child relationships inmiddle childhood are especially important for ad-justment and personality development. Maritalstrife may seriously upset the way by whichparent-child relationships normally lead to appro-priate and well-adjusted behavior. Specifically,conflict between parents may place the child in adouble approach-avoidance conflict in whichcloseness to one parent introduces the risk of re-jection by the other. Then the child who cannotchoose between the two parents is in danger ofbecoming disordered (Gassner and Murray, 1969;Schwarz and Getter, 1979). For the child whobecomes stably allied with the parent of the samesex, Schwarz hypothesizes that the child becomeschauvinistic and alienated from the opposite sex.In the opposite-sex situation, the child experiencesconflict and has problems with gender identityand lower self-esteem. Finally, the child who re-jects both parents is likely to become antisocialbecause of weak socialization.

Our present research is directed toward specify-ing in more detail how much and in what waysmarital conflict and disruption lead to behavioraldifficulties in children. Specifically, we are testinga series of hypotheses suggested by the research todate, as follows:

1. Child outcomes will be more adverse thehigher the level of parental conflict, and thelonger the duration of conflict.

2. Disruption and conflict will have strongerand more lasting effects on boys, leadingespecially to antisocial behavior. Girls, onthe other hand, will be more likely tobecome depressed or withdrawn.

JOURNAL OF MARRIAGE AND THE FAMILY

3. The effects of conflict and disruption will beworse when the conflict alienates the childfrom both parents. This situation is especial-ly likely to lead to antisocial outcomes.

4. Intermediate effects will result when thechild is able to maintain a strong, positiverelationship with one parent.

5. The best outcomes will occur when the child,despite the disruption and conflict, is able tomaintain a positive relationship with bothparents.

METHODS

Sample

The data on which the analyses are based arefrom the National Surveys of Children, whichcomprise two waves of a longitudinal study ofchildren in the United States. The first wave, con-ducted in the fall and winter of 1976-77, wasbased on a national probability sample of house-holds containing children aged 7-11. Altogether,information was gathered on 2,301 children in1,747 households, this being a completion rate of80070.Interviews were conducted with each eligiblechild and that parent most able to provide infor-mation about the child, usually the mother. Datawere also collected by questionnaires sent to thechildren's teachers.

The second wave of the survey was conductedin the spring and summer of 1981, when the chil-dren were 12 to 16 years old. Limitations in fund-ing precluded reinterviewing the entire sample. Inall, follow-up data were gathered on 1,423 chil-dren, or nearly 80% of those chosen for reinter-view. Further information about the sample andthe study design may be found in previous reports(Furstenberg, Nord, Peterson, and ZiII, 1983; Zilland Peterson, 1983).

Since the focus of this paper is on the role ofparent-child relationships in mediating the effectsof marital disruption, the present analysis islimited to children both of whose parents were liv-ing at the time of the second survey, and who wereliving with at least one biological parent (or withtwo adoptive parents) and had had some contactwith each parent within the last five years. Thesechildren comprise 76% of the sample. Within thisstudy group, 20% were living with only one oftheir biological parents because of a separation ordivorce. The mean time since the initial separationwas eight years; and the mean age of the childrenat separation, six years. Yet much experience ofdisruption was fairly recent. Among childrenwhose parents separated or divorced, one in fourhad this happen in the five-year interval between

--. -----.

DIVORCE AND BEHAVIOR PROBLEMS

the two surveys, and one in two experienced sometransition (first or subsequent) in this period.

Measures

Children were classified into three broad groupson the basis of their living arrangements (seeTable I): those living with both biological parents(or two adoptive parents), those living with theirbiological mothers but not fathers (a step-, foster,or adoptive father might also be in the home), andthose living with their biological fathers but nottheir mothers.

The first group was further subdivided by thelevel of marital conflict, defined by combiningitems measuring marital happiness and maritalarguments. Marriages were characterized as hav-ing low conflict if the parent respondent describedthe marriage as "very happy" and reportedmarital arguments in not more than two of nineareas that were covered. Marriages were charac-terized as having high conflict if the marriage wasdescribed as "not too happy" or if argumentswere reported in five or more of the nine areas.All other situations were classified as havingmoderate conflict. Using the same conflict mea-sure from Wave I, we divided children from high-conflict situations in 1981 into those who ex-perienced high conflict in 1976 as well (persistenthigh conflict) and those who did not (recent highconflict).

Children living with their biological motherswere further cIassifed according to her maritalstatus: single-parent mothers (separated, di-vorced, or never married); remarried; and redis-rupted after a second or subsequent marriage.The number of children in the sample who livedwith their biological father but not mother wastoo small to permit a similar subdivision.

The parent-child relationship was measured byusing the child's responses to a set of four itemsabout each biological parent (whether present inthe house or not). These items were: closeness;frequency of doing things together; amount of af-fection received; and aspiration to be like the kindof person the parent is. The items correlate well(the average correlation among pairs of variablesis .50 for the fathers and .33 for the mothers), in-dicating that the items appear to tap the sameunderlying dimension. To form a summary vari-able of the relationship with both parents, wesummed the items for each parent to form a scalethat was then dichotomized so that at least halfthe sample fell in the upper part of eachdichotomy. The final measure, a cross-classifica-tion of the dichotomized scales, had four cate-gories: having a positive relationship with bothparents; with the mother only; with the father

297

only; or with neither parent. By this measure, halfthe children in the sample have a positive relation-ship with both parents, one in five has a positiverelationship with the mother only; one in five,with neither; and about one in ten, with the fatheronly (see Table I, first row).

Five outcome variables were used. Three ofthese are subscales of a behavior problems indexthat was adapted from a longer checklist devel-oped by Achenbach and Edelbrock (1981) to pre-dict mental and behavior problems that may re-quire therapy. We used three criteria to selectitems from the checklist: reliability; high loadingon one of the subscales that emerged from theiranalyses, and adaptability to an interview situa-tion. Factor analysis of the items in the NationalSurvey of Children yielded four factors, similar toones found in Achenbach's work. Three of thefour factors appear to represent fairly serious be-havior problems. These are depressed/withdrawnbehavior, antisocial behavior, and impulsive / hy-peractive behavior. Each of these subscales iscomposed of six items. (The fourth factor, mea-suring milder and more general behavior prob-lems, is not used in this analysis.)

While factor analysis was used to confirmwhich items belonged together in a subscale, asimple additive index was used to create eachscale. This was done so that the scale score wouldhave a straightforward meaning: namely, thenumber of items (from 0 to 6) that the parentreported as characterizing the child. All threemeasures are skewed, as might be expected. Themeans range from 1.0 (antisocial score for girls) to2.3 (impulsive/hyperactive score for boys).

The remaining two outcome variables are singleitems measuring behavior problems at school.These are whether, in the last four years, the childhas had "any behavior or discipline problems atschool resulting in your receiving a note or beingasked to come in and talk to the teacher or prin-cipal," or has been "suspended, excluded, or ex-pelled from school."

RESULTS

Previous analyses of the National Survey ofChildren have established that marital disruptionand conflict are associated with psychologicalproblems in children, some of which may persistfor several years (ZiII, 1978; Zill and Peterson,1983). The present analysis extends these results toother outcome measures. The aim is to learnwhether these negative outcomes may be ameli-orated by the child's maintaining a positive rela-tionship with one or both parents and, given apositive relationship with one parent, to learnwhether the sex of the parent or thesexof the

-

298

child in relation to that of the parent is of anyconsequence.

Marital Status and Parent-Child Relationships

The disruption of a marriage by separation ordivorce may undermine the relationship a childhas with each parent. These weakened relation-ships, in turn, may contribute to some of thenegative outcomes found among children fromdisrupted families. The drastic curtailment in thelevel of contact between the child and the non-custodial parent may weaken the child's relation-ship with that parent. Indeed, earlier analyses ofdata from this same survey found the level of con-tact to be strongly and positively associated withhow close the child felt to the noncustodialparent. However, nearly half of the children whohad a noncustodial parent had had virtually nocontact with that parent in the year prior to thesurvey, and only one in six had seen him (in over90070of cases the noncustodial parent was thefather) as regularly as once a week (Furstenberg etaI., 1983). As a consequence of such limited con-tact, the quality of the parent-child relationshipsuffered for most of these children.

The relationship with the custodial parent mayalso suffer. In most cases the custodial parent has

JOURNAL OF MARRIAGE AND THE FAMIL Y

to take on an even greater share, if not all, of theburden of childrearing. She may take on or in-crease her level of employment out of economicnecessity, thereby limiting the amount of time thatcan be spent in interaction with children. And theparent's ability to maintain a good relationshipwith her children may deteriorate as a result of thestress often associated with divorce. On balance,however, the relationship with the noncustodialparent may suffer more than the relationship withthe custodial parent, because of the lack of con-tact.

The data in Table I show that marital disrup-tion is associated with poorer parent-child rela-tionships, according to the child-based measurewe used (p < .01, based on x'). Fifty-five percentof the children in intact families maintain a posi-tive relationship with both parents, but only 25.JJoof those living with their mothers and 36% ofthose living with their fathers do so. The relation-ship with the noncustodial parent is especiallylikely to suffer. While 60% of those who live witha custodial mother report a positive relationshipwith her, only 36% of these same children report apositive relationship with the absent father.Similarly, 69% of those living with a custodialfather have a positive relationship with him, com-

TABLE I. RELATIONSHIPWITH PARENTS, BY CHILD'S liVING ARRANGMENTSAND SEX, U.S. CHILDREN AGED 12-16,1981

aThe data are based on children for whom the parent-child relationship variable is defined: namely, those havingboth biological (or adoptive) parents living, having had contact with each parent within the last five years, and hav-ing complete data on the items making up the scale. These children constitute 76070of all the children in the sample.Losses due to incomplete data are minor.

bThe association between living arrangements and parent-child relationships is significant (x'. P < .01).cThe association between sex and parent-child relationships is significant (x'. P < .01).

Percenlage of Children Having Positive Relationship with:Both Mother Father Neither N

Parents Only Only Parent (Unweighted)All childrena 49% 19070 12070 20070 1,080Living arrangementsb

Child lives with bothbiological (or adoptive)parents 55 15 12 18 813Low conflict 62 13 12 13 345Moderate conflict 51 14 12 23 320High conflict, recent 50 20 II 19 96High conflict, persistent 20 29 14 28 52

Child lives with biologicalmother 25 35 II 29 232Separated, divorced, or

never married 26 32 12 30 132Remarried 32 39 9 20 74Redisrupted 8 36 10 46 26

Child lives with biologicalfather 36 22 33 9 28

SexcFemale 41 24 10 25 551Male 57 13 14 16 529

f


pared with 57070having a positive relationshipwith the absent mother.

These last two sets of figures show that the rela-tionship with the mother is more robust in theface of absence. Only a third of children maintaina positive relationship with absent fathers, com-pared with well over half with absent mothers.This difference may stem from the higher levels ofcontact that absent mothers maintain in com-parison with absent fathers.

Within intact marriages the relationship be-tween the child and each parent, especially thefather, suffers as the level of conflict rises.Especially harmful to good relationships is persis-tent high conflict (that is, high conflict at bothsurvey dates). Thus 62070of the children in low-conflict families have a positive relationship withboth parents, compared with only 29070of those inpersistent high-conflict situations. The latterfigure is actually comparable to the proportionsfor children in disrupted families. Since conflictusually attends marital disruption, much of thenegative effect of separation or divorce may be aresult of the associated conflict. Indeed, amongchildren whose parents' marriages were disruptedbetween the two waves of the survey, those in ahigh-conflict situation at Wave I had nearly threetimes the rate of psychological distress betweenthe two waves as those in a low- or moderate-con-flict situation at Wave I (Zill and Peterson, 1983).

After a disruption, the parent-child relationshipappears to be strongly associated with the maritalstatus of the custodial mother. In particular, theremarriage of the mother is associated with amodest improvement in the relationship with themother and, to a lesser extent, the still-absentfather. The disruption of a second marriage, onthe other hand, is associated with markedly lowerproportions having a positive relationship witheither parent. Less than one child in ten from re-disrupted families has a positive relationship withboth parents, compared with about three in tenwhen the mother has either stayed remarried ornot remarried at all. It is unclear from the presentdata whether these differences in proportions arechanges over time induced by marital events (re-marriage, redisruption), are due to the slightlyyounger age at initial disruption of children fromredisrupted families, or are the result of selectionfactors (perhaps mothers more able to maintaingood personal relationships manage to remarryand stay remarried, while those less competentremarry only to divorce or become separatedagain). Certainly, all these factors could con-tribute to the difference simultaneously.

299

Sex and Parent-Child Relationships

We might expect that, through the process ofdeveloping an appropriate sexual identity, girlswill come to ha'{e a closer relationship with theirmothers, and boys, with their fathers. This isclearly the case for girls, as can be seen in Table I.Thus 65% of girls have a positive relationshipwith their mothers, versus 51% with their fathers.The situation for boys is virtually even, with 70%and 71% having a positive relationship with theirmothers and fathers, respectively.

For children from disrupted families, however,the effect of the disruption appears to be morepowerful than whether the parent is of the samesex (data not shown). Both boys and girls who livewith their mothers are more likely to have betterrelationships with their mothers than with theirfathers. For those who live with their fathers, theresults follow a same-sex pattern, boys having bet-ter relationships with their fathers and girls havingslightly better relationships with their mothers.

Overall, girls report poorer relationships withtheir parents than do boys. In both intact anddisrupted families fewer girls than boys reportpositive relationships with both parents, and morereport having a positive relationship with neitherparent. While this may indicate a real sex dif-ference in the quality of parent-child relation-ships, it could also be that boys and girls holdsomewhat different standards for a positive rela-tionship.

Parent-Child Relationships and Child Outcomes

Good parent-child relationships can amelioratethe negative effects of marital disruption only tothe extent to which such relationships are associ-ated with the outcomes in question. If the effectsof positive relationships are at all additive, thenhaving a positive relationship with both parentsshould produce the best results; with one parent,intermediate results: and with neither parent, theleast desirable results. Given a positive relation-ship with only one parent, does it matter whichparent that is? As Schwarz indicated, in the seriesof hypotheses reviewed earlier, it may depend onwhich outcomes one is looking at, and on the sexof the child in relation to that of the parent withwhom the positive relationship is maintained.Whether or not this is the custodial parent mayalso be of importance.

Table 2 presents the data for the five outcomevariables we have examined. The data are pre-sented separately for boys and girls because of sexdifferences in the outcome variables. In thisregard, we note that girls score higher than boyson the depressed/withdrawn scale, but boys score

300

higher on all the other measures. These results areconsistent with suggestions that behavior prob-lems in girls are more likely to take the form ofovercontrolled behavior, whereas in boys theytake the form of undercontrolled behavior. Theresults may also prefigure the higher rates ofdepression of adult women in comparison withmen.

As expected, children who have a positive rela-tionship with both parents are least likely to ex-hibit any of the problems measured. For boys, theworst outcomes usually occur when the child has apositive relationship with neither parent. An ex-ception is the high rate of impulsive/hyperactivebehavior for boys with a positive relationship justwith the father. Given a positive relationship withonly one parent, it makes little difference in threeof the five outcomes which parent the relationshipis with. In the other two cases-impulsive/hyper-active behavior, and being suspended or expelled-the outcome is somewhat worse if the relation-ship is with the father.

The situation for girls is more complex. As withboys, the best outcomes are usually found for girlshaving a positive relationship with both parents.(For suspended/expelled, the "father only" cate-gory is just as low.) And in three of the five out-comes, the worst situation occurs when the girlhas a positive relationshp with neither parent. Butfor antisocial behavior and impulsive/hyperactive

JOURNAL OF MARRIAGE AND THE FAMIL Y

behavior, having a positive relationship only withthe father leads to the worst outcomes. It may bethat since these two behaviors are especially asso-ciated with males (boys' average scores on thesescales are much higher than the girls'), a father'sinfluence, untempered by that of the mother,might increase these behaviors among girls. Yetthe same argument could be made about theschool misbehavior measures, and on these, girlswho have a positive relationship just with thefather score strikingly low.

Another factor may come into play for theseschool variables. Parents, perhaps especiallyfathers, tend to have a double standard for girls'behavior, being much less tolerant of misbehaviorin public than in private. For example, data fromthe National Survey show that girls exhibit prob-lem behavior much less often than boys at school,but are only marginally better than boys onsimilar behaviors at home. If fathers do hold sucha double standard, then girls who have a positiverelationship only with the father may well exhibitmore problem behavior of the sort that parentscan observe, but less of the kind that is likely tocome the attention of school personnel.

Marital Status, Child Outcomes, andParent-Child Relationship

We consider next the associations between themarital situation of the child's family and the

TABLE 2. SOCIALAND PSYCHOLOGICALOUTCOMES, BYSEX AND RELATIONSHIPWITH PARENTS, U.S CHILDREN AGED12-16, 1981

Outcome Measures (parent reports)

FemalesDepressed / withdrawn (mean)bAntisocial behavior (mean)bImpulsive / hyperactive (mean)bSchool note: misbehavior (OJo)CSuspended /expelled (OJo)CN (unweighted)

MalesDepressed/withdrawn (meaQ)bAntisocial behavior (mean)bImpulsive/hyperactive (mean)bSchool note: misbehavior (07o)CSuspended /expelled (%)CN (unweighted)

aSee footnote a, Table I.bScores for the three scales range from 0 to 6, higher scores indicating more problem behavior. The standard

errors of the means range from .06 to .20. The mean scores for "both parents" are all significantly less than those ofthe other three categories (p'::;' .05, using a one-tailed t test) except the impulsive/hyperactive score of males in the"mother only" category.

cThe standard errors of the percentages range from 1.0 to 5.2. The percentages for "both parents" are signifi-cantly less than each of the other categories for males (p .::;. .05, using a one-tailed t test). for females they aresignificantly less than the "neither parent" category for both variables, and the "mother only" category for theschool note variable.

AllChildrena

1.321.001.87

126

551

1.081.352.28

22II

529

Children Having Positive Relationship with:Both Mother father Neither

Parents Only Only Parent

.88

.751.42

53

220

1.311.001.99

156

146

1.761.402.50

83

53

1.861.262.24

23II

132

.871.022.09

134

274

1.171.722.24

291882

1.311.632.89

292278

1.561.972.45

442195

DIVORCE AND BEHAVIOR PROBLEMS 301

TABLE 3. MEAN DEPRESSED/WITHDRAWN SCORES, BY CHILD'S LIVING ARRANGEMENTSAND SEX, U.S. CHILDRENAGED 12-16, 1981

Living Arrangements

All living arrangementsLives with both parentsb

Low conflictModerate conflictHigh conflict, recentHigh conflict, persistent

Lives with biological motherSeparated, divorced, never marriedRemarriedRedisrupted after remarriage

Lives with biological fatherCoefficient for "Living arrangements"

eta (column I); beta (columns 2, 3)Total variation explained

(I)

N(Unweighted) Unadjusted

Males1.080.990.681.091.521.821.652.101.340.930.83

Mean Scores

(2)

Adjusted forControlsa

1.000.671.101.591.871.572.021.230.890.82

0.29212.6070





eta (column I); beta (columns 2, 3) 0.251Total variation explained 6.3070

aControl variables are parent education, family income, race, and child's age.blnc1udes living with two adoptive parents.-These means are significantly different from that for the intact, low-conflict category (p <

tests were calculated on the third column only.

various outcome measures, and the extent towhich these associations are modified by takingthe parent-child relationship into consideration.To examine these associations we made use ofmultiple classification analysis, a form of dummymultiple regression appropriate for use when thedependent variable is an interval scale (or adichotomy) and the predictor variables are cate-gorical. Each of the five outcome variables wasused in turn as the dependent variable in a seriesof analyses. For each dependent variable threeanalyses were done. The first used the child's liv-ing arrangements (defined in terms of the parents'marital status and level of conflict) as the lonepredictor. The second analysis added a set ofbackground control variables-parent education,

5264051761534729

10757361414

5464071681674923

12575381214

0.2938.6070

Females1.321.250.801.641.421.981.531.611.461.291.83

1.220.731.621.492.001.641.651.701.511.77

0.27613.4070

(3)Adjusted forControls andParent/ChildRelationship

1.020.721.10-1.58-1.81-1.48-1.92-1.24-0.670.76

0.26814.0070

1.240.791.60-1.54-I. 90-1.57-1.59-1.64-1.32I. 77-

0.24617.3070

.05). Significance

family income, the child's age, and race-in addi-tion to living arrangements. Finally, the thirdanalysis added the parent-child relationship vari-able to the list of predictors to enable us to deter-mine how much the addition of this variable ex-plained the association between marital conflict ordisruption and child behavior outcomes.

Data from these analyses are presented inTables 3-7, one table for each outcome variable.Column I in each table shows the outcome-vari-able means for each category of the child's livingarrangements, unadjusted for any other predic-tors. Column 2 shows the means adjusted for thecontrol variables; and column 3, adjusted for thecontrols and the parent-child relationship mea-sure. Separate sets of means are presented for

302 JOURNAL OF MARRIAGE AND THE FAMILY

TABLE 4. MEAN ANTISOCIALSCORES, BY CHILD'S LIVING ARRANGEMENTSAND SEX, U.S. CHILDREN AGED 12-16,1981

1.291.011.33.I. 75.2.48·1.68.I.90.1.581.201.22

0.23417.1070

5474081691674923

12575381214

Females1.000.930.761.050.981.601.210.931.651.541.40

0.970.791.091.081.511.050.701.631.391.46

0.980.811.08.1.111.48.1.020.681.60.1.341.43






aControl variables are parent education, family income. race, and child's age.blncludes living with two adoptive parents..These means are significantly different from that for the intact, low-conflict category (p <


0.1999.0070

0.19111.7070

.05). Significance

males and females because of the interactions be-tween sex, parent-child relationships, and livingarrangements.

Depression/withdrawal. Children are leastdepressed and withdrawn when they live withboth parents rather than the biological motheronly (see Table 3, column I). When they live withtheir fathers, an interesting sex difference ap-pears. In this case boys are at least as well off asthose boys living with both parents, whereas girlsare even more depressed and withdrawn thanthose living only with their mothers. However, thesmall sample necessitates caution in interpretingthe means for this group.

For children in intact families, the level ofparental conflict strongly affects the level ofdepression and withdrawal. The depressed/with-

drawn scores rise strongly with level of conflict,the scores for those in a situation of persistentconflict being even higher than for those who livewith just one of their biological parents. Compar-ing girls and boys, it appears that girls may be abit more sensitive to moderate levels of conflict.

Among children living with their biologicalmothers but not fathers, the depressed/with-drawn scores are especially high for those livingwith a single mother (separated, divorced, ornever married). This is especially so for boys,which is consistent with previous research findingson the greater vulnerability of boys to maritaldisruption. In comparison. living with a remarriedmother is associated with lower depressed/with-drawn scores; the difference is especially large forboys. While these data are cross-sectional and do

Mean Scores

(I) (2) (3)Adjusted forControls and

N Adjusted for Parent/ChildLiving Arrangements (Unweighted) Unadjusted Controlsa Relationship

Males

All living arrangements 526 1.35Lives with both parentsb 405 1.23 1.26

Low conflict 176 0.92 0.92Moderate conflict 153 1.26 1.32High conflict, recent 47 1.66 1.75High conflict, persistent 29 2.75 2.62

Lives with biological mother 107 2.01 1.86Separated, divorced, never married 57 2.30 2.09Remarried 36 1.72 1.67Redisrupted after remarriage 14 1.76 1.52

Lives with biological father 14 1.15 1.30Coefficient for "Living arrangements"

eta (column I); beta (columns 2, 3) 0.308 0.280Total variation explained 9.5070 14.1070

DIVORCE AND BEHAVIOR PROBLEMS 303

TABLE 5. MEAN IMPULSIVE/HYPERACTIVE SCORES, BY CHILD'S LIVING ARRANGEMENTSAND SEX, U.S. CHILDRENAGED 12-16, 1981

Living ArrangementsN

(Unweighted)






(I)

5264051761534729

10757361414

Unadjusted

Males2.292.171.662.492.673.262.892.862.902.912.38

Mean Scores

(2)


2.211.732.462.763.282.742.632.972.592.16

0.26116.4070





eta (column I); beta (columns 2, 3) 0.131Total variation explained 1.7%

aControl variables are parent education, family income, race, and child's age.blncludes living with two adoptive parents.-These means are significantly different from that for the intact, low-conflict category (p <


not directly test the causal effect of remarriage,they are consistent with the notion that gaining astepfather improves outcomes for boys (the caseof girls is addressed below). We expected redis-ruption to be associated with even higher levels ofdepression and withdrawal, an expectation notsupported by the data. However, the numbers ofcases in this category were quite small, so themeans are subject to considerable sampling varia-bility.

Adding background variables to the model(Table 3, second column) does not appreciablychange the overall pattern of results. For boys, thestrength of living arrangements as a predictor isunchanged (both beta and eta are .29); for girls, itis actually strengthened slightly. The additionalvariables do, however, raise the proportion of

5474081691674923

12575381214

0.2858.1%

Females1.881.841.621.942.102.502.002.101.772.012.11

1.851.591.942.222.641.962.011.881.892.10

0.14511.9%


2.221.782.45-2.76-3.25-2.71-2.56-3.03-2.472.02

0.24917.5%

1.881.671.942.24-2.55.1.851.931.771.652.03

0.12615.4%

.05). Significance

variation explained in the depressed / withdrawnscore from 9070to 13% for boys, and from 6% to13% for girls.

Introducing controls into the analysis producesone slight yet noteworthy change in the pattern ofresults. The adjusted depressed/withdrawn scorefor girls living with remarried mothers is slightlyhigher than for girls living with a single mother.For the unadjusted scores it is moderately lower.Santrock, Warshak, Lindbergh, and Meadows(1981) noted evidence for negative reactions ofgirls, but not boys, to the remarriage of theirmothers. While our data do not indicate that girlswith remarried mothers are substantially moredepressed and withdrawn than girls with singlemothers (the adjusted score is higher, but not sig-nificantly different), the data do indicate that a

,...

304 JOURNAL OF MARRIAGE AND THE FAMIL Y

TABLE 6. PERCENTAGE RECEIVING SCHOOL NOTE, BY CHILD'S LtVING ARRANGEMENTSAND SEX, U.S. CHILDRENAGED 12-16, 1981

remarriage is more troublesome for girls than forboys, which is consistent with Santrock's find-ings.

We can test the ameliorative effects of a goodparent-child relationship by adding this variableto the model and examining what happens to thepredictive power of the living-arrangements vari-able. When this is done (compare columns 2 and 3of Table 3), the explanatory power of living ar-rangements is reduced, but the magnitude of thereduction is modest at best. The betas drop from.29 to .27 for boys and from .28 to .25 for girls.The overall pattern of results is unchanged; how-ever, the range of differences is somewhat re-duced, consistent with the reduction in the betas.Significance tests, comparing the mean for eachcategory with the mean for those in intact low-

conflict families, show that most means are sig-nificantly higher. The exceptions are children liv-ing with mothers after a redisruption, and boysliving with their fathers.

Antisocial behavior. The results for antisocialbehavior parallel those of depression/withdrawalin several respects (see Table 4). Specifically, anti-social behavior is higher for those living with theirmothers than for those in intact families. Amongthose with their fathers, antisocial behavior iseven higher for girls, whereas it is down to thelevels of intact families for boys. In addition, con-flict in intact families is strongly and positivelyrelated to antisocial behavior; and persistent highconflict is especially detrimental. The overall pat-tern of relationships is little changed by the addi-tion of control variables and the parent-child rela-

ahr

a

h1

- -------- -"-

Mean Scores

(I) (2) (3)Adjusted forControls and

N Adjusted for Parent IChildLiving Arrangements (Unweighted) Unadjusted Controlsa Relationship

Males

All living arrangements 525 22.2Clfo - -t'

Lives with both parentsb 404 19.0 19.7 20.9Low connict 176 18.6 19.7 22.8Moderate connict 152 18.0 18.0 18.3High connict, recent 47 11.7 14.0 13.6High connict, persistent 29 44.6 42.5 37.8

Lives with biological mother 107 37.7 33.9 27.6Separated, divorced, never married 57 39.6 33.0 26.2Remarried 36 38.7 40.2 37.8.Redisrupted after remarriage 14 29.2 22.6 9.6

Lives with biological father 14 33.4 34.5 32.2Coefficient for "Living arrangements" (

eta (column I); beta (columns 2, 3) 0.221 0.189 0.158Total variation explained 4.9070 10.1070 16.5070

Females

All living arrangements 546 11.8070 - - 1\Lives with both parentsb 407 9.9 9.6 9.9

Low connict 169 8.1 7.7 9.3Moderate connict 166 9.5 9.0 8.1High connict, recent 49 10.8 12.8 13.2High connict, persistent 23 29.3 25.9 23.1

Lives with biological mother 125 20.3 21.6 20.0.Separated, divorced, never married 75 16.5 17.8 16.4Remarried 38 32.5 34.4 33.8.Redisrupted after remarriage 12 9.9 9.5 4.9

Lives with biological father 14 3.8 2.6 4.3Coefficient for "Living arrangements" C

eta (column I); beta (columns 2, 3) 0.211 0.222 0.205Total variation explained 4.4Clfo 10.9Clfo 15.1070

aControl variables are parent education, family income, race, and child's age.blnc1udes living with two adoptive parents..These means are significantly different from that for the intact, low-connict category (p < .05). Significance

tests were calculated on the third column only. [(


TABLE 7. PERCENTAGE SUSPENDED OR EXPELLED, BY CHILD'S LIVING ARRANGEMENTSAND SEX, U.S. CHILDRENAGED 12-16, 1981

(I)






aControl variables are parent education, family income, race, and child's age.blncludes living with two adoptive parents..These means are significantly different from that for the intact, low-conflict category (p <


Living ArrangementsN

(Unweighted) Unadjusted

Males11.1%8.85.89.8

11.921.325.024.326.723.24.9






5264051761534729

10757361414

0.2154.6%

Females5.4%3.62.53.65.6

10.712.68.6

19.516.30.0

5464071691664923

12575381214

tionship variable. As before, the addition of thelatter does somewhat weaken the predictive powerof living arrangements, at least for boys, in-dicating that a good relationship does have amodest ameliorative effect in the face of conflictand disruption.

One notable difference between the results forantisocial behavior and depressed/withdrawn be-havior is that girls living with single mothers areno worse off than those in intact, low-conflictfamilies. Perhaps antisocial behavior, more oftena male than a female characteristic, is less likely tooccur in the absence of a father figure in thehousehold.

In contrast, girls with remarried mothers ex-hibit the very highest levels of antisocial behavior.These results are again consistent with Santrock's

Mean Scores

(2)


9.96.2

12.113.717.418.313.927.112.49.9

0.1615.1%

4.53.64.27.58.69.23.7

19.711.6

1.6

0.17713.0%


10.68.2

12.013.814.614.99.7

26.16.36.3

0.13418.8%

4.54.03.77.77.79.23.6

20.1.10.92.7

0.17914.6%

.05). Significance

notion that the remarriage is troublesome for girlsand may also reflect the addition of a male figureto the household. Boys, it can be seen, actuallyfare somewhat better with remarried mothers thanwith single ones.

Impulsive/hyperactive behavior. A similar pat-tern of results emerges once again from theanalysis of impulsive Ihyperactive behavior (seeTable 5). As before, the detrimental effects ofconflict, especially persistent conflict, can be seen.Among children in disrupted families there aredifferential effects according to the sex of thechild in relation to the sex of the custodial parent,with boys doing relatively well if they live withtheir fathers, and girls, if they live with theirmothers. The addition of the parent-child rela-tionship variable to the model again reduces the

-

305

306

effect of living arrangements, but not by verymuch. .

In contrast to the previous two variables, im-pulsive/hyperactive behavior is more commonamong boys living with remarried mothers thanamong those with single mothers, whereas thereverse is true for girls (although the difference issmall).

Significance tests for the full model indicatethat for girls the effect of living arrangements onimpulsive/hyperactive behavior is fairly weak.Only the two categories for high conflict in intactfamilies have mean scores significantly higherthan that for girls in low-conflict intact families.

Schoo/ behavior problems. The results for thetwo variables measuring school behavior prob-lems are less clear-cut (see Tables 6 and 7). Whilehigher levels of conflict in intact families generallyresult in higher proportions with behavior prob-lems, the results are not always as linear or asstrong as with the first three variables (seeespecially the boys). Living in an intact family isgenerally best, except that girls living with theirfathers have especially low rates of behavior prob-lems. This is in sharp contrast to the earlier pat-tern of results. Finally, the rates of behavior prob-lems are especially high for both boys and girls liv-ing with a remarried mother. In most cases, theaddition of the parent-child relationship variabledoes reduce the strength of living arrangements asa predictor and also adds to the explained vari-ance.

The time frame for the two school questions isbroad. Each refers to problems anytime in theperiod between the two waves of the survey-aspan of four years. This vagueness in the timeframe may have muddied the results, as some ofthe marital transitions (though a minority) oc-curred after the school behavior problems. Also,the two school measures, though based on parentreports, are objective measures of actions takenby school personnel. As such they are indicatorsof fairly extreme public behavior. This kind of be-havior may be less predictable from such variablesas conflict, marital disruption, and parent-childrelationships.

DISCUSSION

In line with our hypotheses, the data show thatmarital disruption is associated with a range ofnegative outcomes for children. Both overcon-trolled and undercontrolled behavior are moreprevalent among children who had experiencedsome form of marital disruption. Equally impor-tant, the data demonstrate that marital disruptionis not a simple dichotomy. The postdisruption liv-ing arrangements and marital history make a great

JOURNAL OF MARRIAGE AND THE FAMILY

deal of difference for children's behavior. Maritalconflict in intact homes, especiaIly if persistent,appears to be as harmful as disruption itself.

Furthermore, poor parent-child relationshipslead to more negative child behavior, yet main-taining good relationships with parents can gosome way in reducing the affects of conflict anddisruption. Beyond these findings, which aregenerally consistent with our original hypotheses,the results raise several issues and new researchquestions that merit discussion.

Remarriage of the custodial mother. For girls,living with a remarried mother. rather than onewho remains single, is associated with more be-havior problems on all measures except impul-sive/hyperactive behavior. Why is this so, espe-ciaIly when a remarriage brings the advantages ofa second parent, usually a higher income, and thepromise of some stability in living arrangements?And why shouldn't boys experience the samedegree of difficulty? Remarriage is itself a transi-tion to which the child must adapt. Though even-tuaIly beneficial, the transition may cause short-term disturbances. Indeed, other analyses of thisdata support the notion that in time the behaviorof a child living with a remarried parent is likely toimprove. That girls have more difficulty with thistransition may stem from having lived with thesame-sex parent. These girls may come to identifymore strongly with their mothers, forming a bondthat may be threatened by the arrival of a newfather.

Sex of the custodial parent. The results suggestthat a child living with a parent of the opposite sexis especially prone to problem behavior. This find-ing must be viewed with caution because thenumbers in the sample who are living with theirfathers are quite small. But if the results can becorroborated, the implications for such issues aschild custody are quite important. RecaIl that thechildren in this survey were adolescents, a timewhen the establishment of an appropriate sexualidentity is an especially important aspect ofdevelopment. Perhaps at these ages living with aparent of the opposite sex is detrimental to prog-ress with this developmental stage.

Age may playa minor role in another way. Onthe average, boys living with their fathers tend tobe slightly older than girls or than boys living withtheir mothers. If the behavior problems we mea-sured decline with age, this may account for someof the low rates of behavior problems among boysliving with fathers. Such an age-related decline isclearly evident for impulsive/hyperactive behav-ior but not so much for the other variables.

Sex differences. We expected boys to be morevulnerable to the effects of disruption and con-


flict, especially in areas where boys are particular-ly prone to behavior problems (such as antisocialbehavior). Similarly we expected girls to be mostsusceptible to depression and withdrawal. Thedata are not entirely consistent with thesehypotheses. The unadjusted effects of the con-flict /disruption measure (living arrangements) arestronger for boys in all five areas, includingdepression /withdrawal. However, the effects forthe two school measures are quite close for boysand for girls. Furthermore, once controls, includ-ing the parent-child relationship, are introduced,both school measures show stronger relationshipsto disruption/conflict for girls than for boys.

In part, the stronger effects on boys may be dueto differences introduced by the sex of thecustodial parent. That is, because most boys livein a relatively high-risk living arrangement afterdisruption, compared with girls, the effect of theliving arrangement is bound to be stronger forboys.

The stronger adjusted effect of disruption andconflict on girls' school behavior is more puz-zling. Usually, girls' public behavior (as at school)is more controlled than their private behavior,whereas for boys this difference is minimal.Perhaps under the stress of conflict or disruption,the greater control girls maintain in public breaksdown. Or perhaps teachers hold different stan-dards for girls and that an equivalent rise in mis-behavior by girls and boys is more likely to lead todiscipline for girls.

Parent-child relationships. The data supportour hypothesis that positive relationships withparents can ameliorate the negative effects ofmarital conflict or disruption. But the moderatinginfluence is modest. These moderating effects areabove and beyond any associated with the previ-ously introduced controls for parent background.In fact, the impact of the relationship variable, interms of the reduction in betas, is generally aslarge or larger than that of the entire group ofbackground variables.

We examined the interrelationships amongthese variables more closely by separate analyseswithin each category of the parent-child relation-ship variable. Though small samples cloud theresults, these analyses show reductions in negativeeffects associated with better parent-child rela-tions for children in high-conflict or disruptedfamily situations.

The analysis and discussion has focused on theeffects of conflict and disruption and on sex dif-ferences in effects. However, we must not losesight of the fact that among all the statuses con-sidered, children are clearly better off living in in-tact families with low to moderate levels of con-flict, and that the majority of children enjoy just

such a situation. While the incidence of behavior-al disturbances is higher in cases of conflict,disruption, and poor relationships, in each caseonly a minority exhibited such behavior.

REFERENCES

Achenbach, Thomas and C. Edelbrock. 1981. "Behav-ioral problems and competencies reported by parentsof normal and disturbed children aged four throughsixteen." Monographs of the Society for Research inChild Development, Serial No. 188.

Furstenberg, Frank F., Jr., C. Nord, J. Peterson, andN. Zili. 1983. "The life course of children of divorce:Marital disruption and parental contact." AmericanSociological Review 48 (October): 656-668.

Gassner, Suzanne, and E. Murray. 1969. "Dominanceand connict in the interactions bel ween parents ofnormal and neurotic children." Journal of Abnormaland Social Psychology 74: 33-41.

Hess, Robert, and K. Camara. 1979. "Post-divorcefamily relationships as mediating factors in the conse-quences of divorce for children." Journal of SocialIssues 35: 79-96.

Hetherington, E. Mavis, M. Cox, and R. Cox. 1977."The aftermath of divorce." In J. H. Stevens, Jr.,and M. Matthews (eds.), Mother-Child, Father-ChildRelations. Washington, DC: National Associationfor the Education of Young Children.

Levinger, George, and O. Moles. (eds.). 1979. Divorceand Separation: Context, Causes, and Consequences.New York: Basic Books.

Levitin, Teresa. 1979. "Children of divorce." Journalof Social Issues 35 (Fall): entire issue.

Maccoby, Eleanor. 1980. Social Development: Psycho-logical Growth and the Parent-Child Relationship.New York: Harcourt Brace Jovanovich.

Rutter, Michael. 1971. "Parent-child separation: Psy-chological effects on the children." Journal of ChildPsychology and Psychiatry 12: 233-260.

Santrock, John W., P. Warshak, C. Lindbergh, andL. Meadows. 1981. "Children's and parents" ob-served social behavior in stepfather families." ChildDevelopment 53: 472-4110.

Schwarz, J. Conrad. 1979. "Childhood origins of psy-chopathology." American Psychologist 34: 879-883.

Schwarz, J. Conrad, and H. Getter. 1979. "Parentalfactors in neuroticism: A triple interaction model."Unpublished manuscript.

Wallerstein, Judith. 1984. "Children of divorce: Pre-liminary report of a ten year follow-up of young chil-dren." American Journal of Orthopsychiatry 54:444-458.

Wallerstein, Judith, and J. Kelly. 1980. Surviving theBreak-up: How Children Actually Cope withDivorce. New York: Basic Books.

ZiII, Nicholas. 1978. "Divorce, marital happiness, andthe mental health of children." Paper presented at theNIMH workshop on divorce and children, Bethesda,MD.

ZiII, Nicholas, and J. Peterson. 1983. "Marital disrup-tion and the child's need for psychological help."Report prepared for NIMH.

307

marital disruption, parent-child relationships, and ... · marital disruption, parent-child...

Documents