This study examines whether family processes that predict positive and negative developmental outcomes are the same in intact and remarried families. Surveys.
Journal of Youth and Adolescence, VoL 23, No. 4, 1994
Family Processes and Adolescent Adjustment in Intact and Remarried Families B o n n i e L. Barber I and J a n i c e M. Lyons 2
Received October22, 1992; accepted July 20, 1993
This study examines whether family processes that predict positive and negative developmental outcomes are the same in intact and remarried families. Surveys were administered to 758 tenth graders from intact families and 95 from stepfather families. Measures of cohesion, democratic decision-making style, permissiveness, and conflict were used to predict self-rated depression, worry, and self-esteem. Remarried and intact families provide similar family environments for permissiveness and democratic decision making. Remarried families are more conflictual and less cohesive than intact families. In both family types, conflict had negative effects, and cohesion and democratic decision-making had positive effects on adolescents' adjustment. In remarried families, but not intact, permissiveness was related to higher self-esteern. INTRODUCTION Adolescence is a developmental period characterized by realignment and redefinition of family ties (Hill, 1980). Parent-child relationships are subtly t r a n s f o r m e d as adolescents develop "responsible a u t o n o m y " f r o m their families (Steinberg, 1990). Parents who are warm, democratic, and demanding, a constellation described as authoritative parenting, may be ZAssistant Professor, Human Development and Family Studies, Pennsylvania State University, University Park, Pennsylvania 16802. Received Ph.D in developmental psychology from The University of Michigan. Research interests include family influences on adolescent identity development and the effects of divorce and remarriage on adolescent adjustment. 2Doctoral candidate, Psychology, Pennsylvania State University, University Park, Pennsylvania 16802. Received M.S. in child clinical psychology from Pennsylvania State University. Research interests include family processes in stepfamilies and the impact of family structure on adolescent development. 421 0047-2891/94/0800-0421507.00/0 9 1994 l'lenum Publishing Corporation
422
Barber and Lyons
more adept at negotiating with their adolescent child a more independent role, helping their child to become a self-reliant, behaviorally and academically competent, and mentally healthy young adult (Baumrind, 1991; Maccoby and Martin, 1983; Steinberg, 1990; Steinberg et al., 1992). Over the last two decades it has been well established that authoritative parenting is associated with the development of competence and mental health in white middle-class children and adolescents (Maccoby and Martin, 1983; Steinberg, 1990). Recently, several studies have examined the impact of authoritative parenting styles on the development of psychosocial competence in adolescents from differing socioeconomic, ethnic, and family structure backgrounds (Steinberg et aL, 1991; Lamborn et aL, 1991; Baumrind, 1991). These studies afford the opportunity to examine the ways in which family processes may operate differently in varying family contexts, a research direction suggested by Bronfenbrenner (1979, 1986, 1989; Bronfenbrenner and Crouter, 1983). Steinberg et al. (1991), Lamborn et al. (1991), and Baumrind (1991) have all concluded that authoritative parenting is beneficial, and preferable to authoritarian or permissive parenting, regardless of the adolescent's ecological niche. These studies provide strong evidence that the benefits of authoritative parenting have transcontextual validity in the United States. One area that deserves more careful examination is the relationship between parenting style and adjustment in different family types. The Steinberg et al. (1991) and Baumrind (1991) studies, while carefully examining family ethnicity and socioeconomic status, combined divorced and remarried family structures for analysis. Combining divorced and remarried families into one category, "nonintact," may mask important differences in "process-by-context" interactions (Bronfenbrenner, 1979, 1986, 1989; Bronfenbrenner and Crouter, 1983), as remarried and divorced families are structurally quite different. The Lamborn et aL (1991) study, examining three family structures (intact, mother-only, 3 and stepfather families), found that the adolescent's family structure did not moderate the relations between parenting style and psychosocial development, school achievement, and problem behavior. However, when they examined internalized distress, they found a different pattern. Although there were significant differences in the prevalence of internalized distress as a function of parenting style among adolescents from intact families, there was no relationship between parenting style and internalized distress in either mother-only or stepfather family types. 3Lamborn et aL (1991) did not report whether the mother-only families were composed of divorced mothers and their children or never-married mothers and their children or a combination of the two groups.
Family Processes and Adolescent Adjustment
423
These results suggest that parenting style may operate differently across family structures for some outcomes. Although the general category of authoritative parenting may prove to be beneficial in both single-parent and remarried contexts, it is possible that the relative levels of parental demandingness, warmth, and democracy associated with optimal adjustment may differ by family type. Since certain aspects of family environment may operate differently in remarried families than in intact families, it is important to examine parenting techniques separately rather than as a composite score of authoritative or nonauthoritative. Family status, particularly remarried status, may have a significant moderating effect on the relationship between parenting style and adolescent's self-esteem and related symptoms of depression and worry. In this study, we examine whether democratic decision-making style, permissiveness, cohesiveness, and conflict have different relationships to adolescent's emotional adjustment in stepfamilies than in intact families. Remarried Families
Studies of parental divorce and remarriage generally have reported no relationship or an inconsistent relationship between stepchild status and psychological adjustment (Ganong and Coleman, 1984; Bray, 1988). The conflicting results within the literature have led researchers to focus on process issues in remarried families, as the effects of divorce and remarriage may involve subtle changes in children's relationships with family members (Amato, 1987). Children's adjustment to stepfamilies may be a function of the ability of the family members to adjust their parenting style and the family environment to meet the needs of members in a changed, nontraditional context. Research on parenting styles within remarried families has shown that many families show a context-specific parenting pattern. In the Hetherington (1987) six-year follow-up (age 10), the predominant stepfather parenting style was disengagement, characterized by low-warmth, control, monitoring, and discipline. The most frequent parenting style for mothers was authoritative. Active discipline by stepfathers has been related to behavior problems and, for stepdaughters, lower self-esteem (Bray et al., 1991). This parenting style split may reflect the family members' adjustment to the stepfamily environment, Parent-child relationships are transformed during adolescence as family members renegotiate authority patterns toward mutuality (Grotevant and Cooper, 1986). During adolescence, family decision-making style becomes more democratic and more permissive (Dornbusch et al., 1985;
424
Barber and Lyons
Steinberg~ 1987). In remarried families, the process of gaining autonomy from the family may be complicated. The adolescent's developmental task of individuating from the family may be in conflict with the family's developmental need to strengthen the family culture (Goldner, 1982). The legacy of being in a single-parent family may also affect individuation. Adolescents in single-parent homes are often required to play an active role in family decision-making and household management (Weiss, 1979). Among the adolescents who enjoy this increased responsibility, some may resent the reduction of their role in family decision-making when their parent remarries. Adolescents who enjoyed their expanded role may benefit if they are allowed to retain some of their responsibilities following remarriage. Cohesiveness is likely to be a process that operates differently in remarried families than in intact ones. Stepparents report significantly less frequent activities and engage in fewer positive interactions with their children than do parents in intact families (Thompson et al., 1992; Vuchinich et aL, 1991). Although adolescents do not report a decrease in support from their mothers following remarriage, stepchildren perceive the family as a whole to be less cohesive than do children from never-divorced families (Amato, 1987; Bray, 1988; Garbarino et aL, 1984; Pink and Wampler, 1985). Pasley and Healow (1988) examined the relationship between cohesion/adaptability and self-esteem with a nonclinical sample of 14-19-yearold stepchildren and children from in~act families. Cohesion and adaptability were positively correlated with adolescent self-esteem for intact families. However, in stepfather families, the relationship was inconsistent. The Pasley and Healow (1988) study, like the Lamborn et aL (1991) study that found a relationship between parenting style and internalized distress in intact families but not in stepfather families, suggests that family processes may operate differently in their influence on self-esteem and internalized distress in remarried families than in intact families. Despite popular conceptions of adolescent development, low levels of conf/ict seem to be normative in families with adolescents and may even promote adjustment to developmental change (Steinberg, 1990). However, high levels of conflict may be dysflmctional, particularly when conflict occurs in a context of low cohesiveness (Cooper, 1988; Steinberg, 1990). Stormy or distant parent-adolescent relationships, while not normative, are more prevalent among ztepfamities than intact families (Montemayor, 1986). Although high levels of conflict would be detrimental in both intact and remarried families, stepfamilies may be at greater risk for a level of conflict that contributes to maladjustment. Even when intact families do not differ significantly from remarried families on parenting style, the meaning of these processes within the family may vary as a function of different contexts. Family cohesion may be less
Family Processes and Adolescent Adjustment
425
important to adjustment in a stepfamily than in an intact family, since adolescents may place less value on warm relationships with their stepfathers than they would with biological fathers. The effect of democratic style and permissiveness may be moderated by family context, as adolescents who have lived in a single-parent family may have become more self-reliant, have experienced lower levels of parental control, and may value their independence. With the introduction of a second parent, stepchildren may experience a partial loss of the autonomy they gained in the single-mother family. Since this change would be in opposition to developmental norms for adolescents, lower parental control within the new family system may be beneficial.
METHOD Sample The sample for this study was drawn from the fifth wave of student data in the Michigan Study of Adolescent Life Transitions (MSALT). The fifth wave of MSALT data was gathered from 1492 tenth-grade students in 1988. The students were from 8 middle and lower middle-class communities in southeastern Michigan and represented a range of socioeconomic and educational characteristics. According to 1980 Census information, these eight districts ranged in median family income from $23,200 to $27,500. Students were originally recruited in 1983 with a permission letter distributed in their sixth-grade math classes. The sample is representative of adolescents enrolled in the grades selected for study, and includes a random, normative sample of stepfamilies. Of the students responding at the fifth wave, 28%, or 413, indicated that their parents did not live together. The analyses reported in this paper include only White students, since the sample included only 13 Black or Hispanic students in remarried families. 4 The final sample, with data on all measures, includes 758 children whose parents are married and 95 living with remarried mothers. Of the adolescents living in stepfather families, 69 lived in "simple" stepfamilies, i.e., families including only biological siblings, and 26 lived in "complex" stepfamilies, i.e., families including residential 4Since there were only 13 Black or Hispanic adolescents in remarried families, we were unable to compare across minority and nonminority contexts. Although others have found no process by context interactions for the influence of parenting style on adjustment across race (Steinberg et al. 1991; Lamborn el al., 1991), we were not comfortable pooling minority and nonminority respondents, as remarriage may be different for minority families (Cherlin, 1981).
Barber and Lyons
426
stepsiblings. These two groups were pooled to provide a sample large enough for multivariate analyses.
Measures
Students' perceptions of their family and self-evaluations were assessed using questionnaires consisting of 7-point Likert items. In addition, some family demographic questions were asked. Scales were constructed based on theoretically driven factor analysis. For each scale, a mean was computed when a student had nonmissing data on half or more of the individual items in the scale. Unless otherwise noted, all scales have a minimum value of 1 and a maximum value of 7.
Family Environment Scales The cohesion scale consisted of 5 items about perceived emotional support from family members and frequency of joint family activities. Sample items include "Our family enjoys doing things together" and "Family members are supportive of each other during difficult times." Higher scores indicate closer family relationships. The alpha coefficient for cohesion is .80. The conflict scale consisted of four items about frequency of family arguments and parental criticism of adolescent behavior (alpha = .78). Sample items include "I have lots of fights with my parents about rules and decisions for me" and "My parents criticize/punish me a lot more than I deserve." Higher scores indicate heightened conflict. The permissiveness scale consisted of two items about the adherence to rules in the family (alpha = .53). Items included "When I do something I'm not supposed to and my parents find out, they very often let me get away with it" and "There are rules in the family, but lots of times my parents don't really care if I live up to them." Higher scores indicate minimal parental demandingness regarding adolescent behavior. The democratic decision-making scale consisted of three items about the adolescent's role in family decisions regarding him or herself and the adolescent's perception of parental trust (alpha = .63). Sample items include "My parents encourage me to give my ideas/opinions even if we disagree" and "My parents trust me to do what they expect without checking." Higher scores indicate greater adolescent input into family decisions. All the items included in this scale had a minimum value of 1, and a maximum value of 4.
Family Processes and Adolescent Adjustment
427
Adolescent Adjustment Scales The self-esteem scale consisted of three items about the adolescent's satisfaction with him or herself (alpha = .78). Higher scores indicate greater self-satisfaction. The depression scale consisted of four items about depressed mood, feelings of loneliness, unattractiveness, and loss or increase in appetite (alpha = .70). Higher scores indicate more depressive symptoms. The worry scale consisted of three items about feelings of being overwhelmed by difficulties, and worries about family finances and finding a job in the future (alpha = .69). Higher scores indicate greater concern about these difficulties.
Procedure
The student questionnaire was administered in each of the high schools in April 1988. Those students in each school who had participated in the earlier waves were excused from their regular classes and gathered in a cafeteria or auditorium. Based on pilot testing, the students were given up to 90 minutes to complete the questionnaire, and most finished early. Research staff members were available to answer any questions the participants had.
Plan of Analysis Analyses were conducted in two steps. In the first step, analysis of variance was used to examine differences by family type in mean scores on the family environment scales and on the adjustment scales. In the second step, regression analyses were conducted to test for a moderating influence of family type on the relationship between family environment and adolescent adjustment.
RESULTS Initially, 2 x 2 (Marital Status x Child Gender) analyses of variance were performed to check for interactions between marital status and gender. As there were no interactions for family or adjustment variables, boys and girls were pooled for all analyses.
428
Barber and Lyons
Relation of Family Structure to Family Environment Results of the analyses of family structure differences in family environment are presented in Table I. Contrary to expectations, there were no significant differences between the groups on levels of democratic decisionmaking style or parental permissiveness. There was a trend for remarried families to experience more conflict (p < .08) than intact families. Remarried families were significantly less cohesive (p < .001) than intact families. The experience of family dissolution and reconstitution does not appear to be related to parenting styles in regard to rule setting, but remarriage is associated with the level of family harmony. Relation of Family Structure to Adolescent Adjustment Differences in adjustment were found between adolescents in intact and remarried families (see Table II). Adolescents in remarried families were somewhat more depressed (p < .01), more worried (p < .001), and had lower self-esteem (p < .05) than those in intact families. Thus, on all three measures of adjustment, adolescents in intact families reported more positive and less negative feelings.
Table I. Family Environment Scores for Adolescents from Intact and Remarried Familes Family type Environment
Intact
Remarried
F
3.3 1.4
3.7 1.5
3.3 a
2.7 0.6
2.6 0.7
1.4
3.i 1.4
3.0 1.5
1.1
4.5 1.3
3.9 1.4
12.7 b
Conflict M SD
Democracy M SD
Permissive M SD
Cohesive M SD p
