دانشجویان خانواده های طلاق: کلیدی برای مقاومت آنها

Abstract College students from divorce families are noted for resilience to adjustment problems yet vulnerability to distress. We studied the impact of parental divorce and two personality characteristics (social responsibility and other-direction) on college students' adjustment problems and distress. Participants (N=197) were 105 college students with divorced parents and 92 whose parental marriages were intact. The personality characteristics were measured with the Social Responsibility Scale and the Other-Direction subscale of the Personal Behavior Inventory. Adjustment problems and distress were measured with the Symptom Checklist-90-Revised (SCL-90-R). Analyses indicated that all three predictor variables (parental divorce, social responsibility, and other-direction) were significantly associated with adjustment problems and distress. Parental divorce, however, was also related to higher social responsibility and lower other-direction, and this combination of personality characteristics was shown to reduce the relationship of divorce to adjustment problems but not to experienced distress. Results are discussed in the context of response to the social demands of college life.

1. Introduction College students who were children of divorce are notably resilient because, unlike younger children with divorced parents and adults who experienced parental divorce during their childhoods, they do not manifest more symptoms of adjustment problems than their peers from intact marriages do. Adjustment problems associated with parental divorce have been well documented for childhood, adolescence, and adulthood (see reviews by Amato, 2000 and Kelly, 1998). Yet numerous studies that have expected adjustment problems among late-adolescent college students from families of divorce have failed to find them (e.g., Grant et al., 1993, Nelson et al., 1993 and Weiner et al., 1995). Laumann-Billings and Emery (2000), who specifically noted the resilience of college students from families of divorce, also failed to find more symptoms of adjustment problems among students from divorce families than other college students. They did find, however, that these students reported more distress than their peers from intact parental marriages. Amato and Booth (1991) reported similar findings for distress from a national sample of adults but found that the differences were weak and attributable to those from higher stress parental divorces. The issues over which Laumann-Billings and Emery found distress differences among college students were family issues from both the past and present. They interpreted the greater distress of students from divorce families as evidence that even these resilient youth do manifest subtle negative sequelae of parental divorce. College students from families of divorce may be vulnerable to distress although they are remarkably resilient to adjustment problems. What accounts for the resilience to adjustment problems manifested by college students from divorce families? We wondered if these students have personality characteristics that equip them particularly well for adjustment to the challenges of college life. We also wondered if adjustment problems associated with parental divorce are not observed in comparisons of global adjustment because such (college-pertinent) personality characteristics mask such problems. In other words, college students from divorce families may have adjustment problems associated with their parents' divorces that are not apparent in global adjustment comparisons with their peers because of the advantages such personality characteristics convey to them for adjustment to stresses in college life. Here we explored these questions with a cross-sectional design by searching for personality characteristics both associated with better adjustment among all college students and more frequent among those who had been children of divorce. The literature, in fact, offered little to guide our search. Studies of personality characteristics associated with children's adjustment to parental divorce largely seem to have compared poorly adjusted children to adequately adjusted children in divorce families. That is, they focused on within-divorce group comparisons. Personality characteristics of children of divorce and intact marriages have not been compared. This is the critical comparison if parental divorce leads to increases in particular areas that create resilience. That is, a between-groups comparison is necessary if one suspects that some children of divorce may have personality assets that were actually generated by their experiences associated with parental divorce. The several personality characteristics that have been found to be associated with children's better adjustment within divorce situations, moreover, have not been found to be associated with better adjustment among late adolescents or adults who earlier had experienced their parents' divorce. Among the personality characteristics that have been found to be associated with children's better adjustment to divorce are other-blame (i.e., parental-blame) as contrasted with self-blame (Emery & Forehand, 1994), a greater sense of self-efficacy or of personal ability to cope with demands encountered and emotions (Sandler, Tein, Mehta, Wolchid, & Ayers, 2000), and the lack of a sense of having no personal control over negative events experienced (Sandler, Kim-Bae, & MacKinnon, 2000). Yet a more internal locus of control (greater attribution of experienced outcomes to personal efforts or characteristics than to external agents) did not distinguish college students from families of divorce from those whose parental marriages remained intact (Proulx & Koulack, 1987); nor was greater self-efficacy found in a nonstudent population of late adolescents and adults from families of divorce than among those from intact parental marriages after participants' level of education was controlled (Amato, 1988). It is necessary to go outside the literature on offspring of parental divorce to explore possible personality characteristics that may explain the resilience of college students from families of divorce. Weiss's (1979) clinical studies comparing offspring of single-parent and two-parent families, however, offered some hints about personality characteristics relevant to an explanation of resilience. Although not all children who experience parental divorce have extensive subsequent experience in single-parent families, some parallels exist between characteristics he noted as facilitated by single parenting and characteristics that have been observed among children of divorce. Weiss concluded, for instance, that single parenting—at least for more competent children—can facilitate faster psychological maturation in children. He noted that single parents often promote earlier development of responsibility for the welfare of others and responsibility for one's own behavior. His observations also suggest that by later adolescence, single-parented offspring more often show a sense of personal determination and of freedom from embeddedness in the peer group. Weiss also noted, however, that such children sometimes experience worries not typically shared by their peers in two parent families. There are very few formal empirical studies that directly test Weiss's (1979) observations from his clinical interviews. Some findings with children of divorce, however, are consistent with his observations and conclusions. In particular, children and early adolescents of divorce have been found to be less embedded in peer relationships but more embedded in family issues (Breen & Crosbie-Burnett, 1993). These findings were informative in the search for resilience characteristics because enmeshment seeking has been found to be related to poorer adjustment among college students (Holmbeck & Wandrei, 1993). Freedom from being embedded in the peer group, then, might be a resilience characteristic for college students. Findings from cross-cultural studies are relevant to Weiss's (1979) observations that single parents' child-rearing practices place more responsibility on children and thus promote the children's development of a greater sense of social responsibility. Nadler, Romek, and Sapira-Friedman (1979) compared the social responsibility of Israeli kibbutz-reared and city-reared children. Those from the kibbutzim, where all community members have responsibilities toward the welfare of the group, manifested greater social responsibility. Weisner and Gallimore (1977) reviewed cross-cultural variations in child rearing and found that cultures placing responsibility on children for the care of even younger children are those that place a greater emphasis on social responsibility among adults. These studies support the idea that being reared in circumstances that require responsibility for the welfare of others in the group promotes the development of a sense of social responsibility. A relationship between social responsibility and either parental marital status or adjustment among college students, however, has not been reported. The frequent demands for responsibility to social and work groups encountered in college life, nonetheless, suggested that social responsibility might be a resilience characteristic for college students. Weiss's (1979) observations suggested that social responsibility might be greater among college students from divorced than from intact marriages. Recently Laumann-Billings and Emery (2000) hypothesized that challenges to college students' adjustment posed by parental divorce may result in distress rather than in the more explicit adjustment problems, or symptoms, upon which previous investigators had focused (e.g., Grant et al., 1993, Nelson et al., 1993 and Weiner et al., 1995). They argued that distress is a more subtle response to stressors than is symptom development—and more in keeping with many of the negative sequelae to parental divorce noted by clinicians. With a distress questionnaire focused primarily on family phenomena and a symptom checklist, these investigators found that although college students from divorce families reported no more severe symptoms than their peers from intact-marriage families, they did report more distress. Laumann-Billings and Emery interpreted these findings as evidence that parental divorce does eventuate in more problematic adjustment for college students, but in the subtle forms that had not previously been considered by researchers. These recent findings echo Weiss's (1979) observation that offspring reared in single-parent families, despite their greater responsibility and freedom from enmeshment in the peer group, manifested worries not shared by peers from two-parent families, worries about family issues. Because Laumann-Billings and Emery assessed distress with reference to family phenomena they may have documented what Weiss called “worries.” Laumann-Billings and Emery's fundamental hypothesis, that challenges to college students' adjustment posed by parental divorce may be expressed in distress rather than in symptom development, has yet to be tested independently of phenomena specific to family circumstances. Freedom from peer enmeshment and social responsibility emerged, then, as personality characteristics that may serve as assets for adjustment (considered in terms of freedom from symptoms) to college life and also might be developed in the course of experiences associated with parental divorce. Both are amenable to measurement with late adolescents. Berkowitz and Daniels (1964) have developed a measure of social responsibility that is suitable for use with college students. Collins, Martin, Ashmore, and Ross (1973), based on the thinking of Reisman, Glazer, and Denney (1950), developed a measure called other-direction that assesses an important aspect of peer group enmeshment. It refers to reliance on others' (particularly peers') perceived standards and values as a guide to personal behavior. Other-direction is distinguishable from ostensibly similar constructs such as external locus of control or (lack of) self-efficacy, both of which are relevant to children's adjustment to parental divorce (e.g., Emery & Forehand, 1994 and Sandler et al., 2000). Other-direction is explicitly social and refers to precursive guides to personal behavior rather than to attributions of control over outcomes already experienced or about one's ability to cope with events. To begin the search for the sources of resilience to adjustment problems among college students who are children of divorced parents, then, we tested four hypotheses. We hypothesized first that more social responsibility and less other-direction are associated with fewer adjustment problems among college students, and then, that college students from families of divorce manifest more social responsibility and less other-direction than college students from families with intact parental marriages. Our third hypothesis was that adjustment problems among college students associated with being from a divorced family become apparent after consideration of effects of social responsibility and other-direction. Finally, we hypothesized that college students from families of divorce manifest more distress than those from families with intact marriages, independent of their levels of social responsibility and freedom from other-direction. This last hypothesis recognizes Laumann-Billings and Emery's (2000) distinction between distress and the adjustment problems for which the resilience of college students of divorce is pertinent as well as Weiss's (1979) observation of worries among offspring from single-parent families.

3. Results Hypothesis 1. Greater social responsibility and other-direction are associated with better adjustment among college students. Zero-order correlations between each of the two personality variables (social responsibility and other-direction) and adjustment problems provide a preliminary test of this hypothesis. Greater social responsibility and lower other-direction were associated with fewer adjustment problems [social responsibility, r(191)=−.27, p<.01; other-direction, r(191)=.38, p<.01]. Social responsibility and other-direction, however, were also mutually correlated, r(191)=.21, p<.01. To determine whether social responsibility and other-direction made unique contributions as predictors of adjustment problems, a multiple regression (MR) analysis simultaneously entering both variables as predictors of adjustment problems was used. Both social responsibility and other-direction were predictive of adjustment problems. The partial correlations are shown in Table 1. Because male students scored lower than female students on social responsibility, r(195)=.19, p<.01, and tended to score higher for adjustment problems, r(191)=−.12, p=.10, this analysis was conducted with the addition of gender as a predictor variable. Gender did not make a unique contribution to prediction of adjustment problems. The unique contributions of (higher) social responsibility and (lower) other-direction to adjustment problems were sustained in this analysis that included consideration of gender, however. Table 1. Partial correlations between social responsibility and other-direction with adjustment problems and distress Variable Adjustment problems Distress pr β pr β Gender −.082 −.076 −.137 .054 Social responsibility −.193 −.185** −.162 −.163* Other-direction .344 .338** .173 .171* ** p=.01. * p=.05. Table options Relationships of social responsibility and other-direction with distress were not hypothesized; nonetheless, we explored them. Zero-order correlations showed that both aspects of personality were related to scores for distress, with higher social responsibility associated with less distress, r(191)=−.22, p<.01, and higher other-direction to more distress, r(191)=.21, p<.01. To determine whether social responsibility and other-direction made unique contributions as predictors of distress, an MR simultaneously entering both personality variables was used. Both social responsibility and other-direction were predictive of distress (see Table 1). Because male students scored higher on the measure of distress, r(196)=−.17, p<.05, and lower on social responsibility, r(195)=.19, p<.01, than female students did, this analysis was conducted with the addition of gender as a predictor variable. In this model gender did not make a significant unique contribution to the prediction of distress (β=−.137, p=.054). However, the unique contributions of (higher) social responsibility (β=−.163, p<.05) and (lower) other-direction (β=.171, p<.05) to distress were sustained in this analysis. Subsequent examination of regression analysis data showed that the relationships of social responsibility and other-direction with adjustment problems (or distress) did not differ for undergraduates from divorce and intact-marriage families. That is, these personality characteristics were similarly related to adjustment (and distress) in both family status groups. Hypothesis 2. College students from families of divorce manifest more social responsibility and less other-direction than college students from families with intact parental marriages. Gender was not equally distributed across family status groups and was significantly associated with social responsibility. Hierarchical MR analysis was used to determine whether family marital status (divorce or intact) was related to social responsibility and other-direction, with control for effects of gender. Gender was entered in Step 1, and family status was added as a predictor of the personality variable (social responsibility or other-direction) in Step 2. These two analyses showed that college students from divorce families tended to have somewhat higher social responsibility scores (ß=.129, p=.07) and lower other-direction scores (ß=.114, p=.12) than those from intact-marriage families. Although the relationships between family marital-status groups and these personality variables of social responsibility and other-direction were in the expected directions, they only approached the .05 level of significance. These trends become important because with a mutual zero-order correlation of only −.20, social responsibility and other-direction are largely independent measures. Thus, high social responsibility and low other-direction scores in combination might show different patterns for the students from the divorce family and the intact-marriage group. To determine whether undergraduates from families of divorce showed the combination of high social responsibility and low other-direction more often than their peers from intact-marriage families, we first determined a criterion for this combination. For this purpose, we used the top quartile of scores on social responsibility and bottom quartile of scores on other-direction. The quartiles on social responsibility were determined separately for male and female students because of the observed gender difference for social responsibility. Students with scores in these two quartiles were considered to manifest the combination of personality characteristics associated individually with fewer adjustment problems. The two family status groups were compared on proportions that reached or failed to reach criterion with χ2 to determine whether the combination of high social responsibility and low other-direction was found more often among students from families of divorce than intact-marriage families. More students from divorce families (15.2%) reached this criterion than did those from intact-marriage families (4.3%), χ2(1, N=197)=6.4, p<.05. Although students from families of divorce did not show higher social responsibility or lower other-direction scores than their peers from intact-marriage families when these measures were considered individually, they did more often show the expected combination of characteristics associated with fewer adjustment problems. Weiss (1979) had noted greater responsibility for the welfare of others and less embeddedness in the peer social group among the offspring of single-parent families than those of two-parent families. To explore whether the experience of single parenting was related to social responsibility and other-direction scores in the group of students from families of divorce, we considered each of these measures two ways. First, we explored the correlations of these variables with the number of postdivorce years these students had experienced in a single-parent household; next we conducted t tests on social responsibility and other-direction to compare those who had lived all their teen years in a single-parent family with those who had lived those years in a family with a parent and a stepparent (dual-parent family). Neither social responsibility nor other-direction was significantly associated with either index of single parenting. We found no evidence, then, that the differences between undergraduates from divorce and intact-marriage families could be attributed to circumstances that might be associated with single parenting. Hypothesis 3. Adjustment problems among college students associated with being from a divorced family become apparent after consideration of effects of social responsibility and other-direction. A series of hierarchical MR analyses was conducted on adjustment problems to test this hypothesis. The first analysis entered gender as a predictor in Step 1 to control for the tendency of male students to score higher on adjustment problems although also being more frequent in the intact-marriage group than the divorced-family group. Family status was entered in Step 2. As expected, there were no differences in adjustment problems attributable to family status in this analysis where effects of social responsibility and other-direction are not considered. The next hierarchical MR controlled for the effects of both gender and the personality characteristics on adjustment problems before considering any unique effect of family status. Gender, social responsibility, and other-direction were entered in Step 1 and family status was added as a predictor in Step 2. There was a significant amount of variance on adjustment problems accounted for by family status after variance associated with the full set of control variables was taken into account; divorce family status was associated with more adjustment problems. College students from families of divorce did, indeed, report more adjustment problems than their peers from intact-marriage families but only after the effects of social responsibility and other-direction on adjustment were taken into account. The results of this analysis are shown in Table 2. Table 2. Summary of hierarchical MR analysis of demographic and personality factors predicting adjustment problems (N=197) Variable Summary for adjustment problems B SE B β Step 1 Gender −1.510 1.337 −.076 Social responsibility −.248 .092 −.185** Other-direction .697 .138 .338*** Step 2 Gender −1.930 1.344 −.097 Social responsibility −.269 .092 −.200** Other-direction .717 .138 .348*** Family status 2.631 1.328 .133* R2=.186 (p<.001) for Step 1; ΔR2=.208 (p<.05) for Step 2. ** p=.01. *** p=.001. * p=.05. Table options Because we observed that students from divorce families tended to differ from those from intact-marriage families on both social responsibility and other-direction, we explored further to determine whether adjustment-problem differences would be found when either one of these personality variables were controlled. In these hierarchical MR analyses, Step 1 included gender and one personality variable (either social responsibility or other-direction), whereas in Step 2 family status was added as a predictor variable. No significant family status difference on adjustment problems was found when only social responsibility (family status β=.108, p=.13) was used as a control variable or when only other-direction (family status β=.111, p=.10) was used as a control variable with gender. That is, neither of these personality characteristics alone accounted for sufficient variance to allow family status to emerge as a predictor of adjustment problems. As hypothesized, it was only with prior consideration of both personality variables—the combination of variables on which the family groups had significantly differed—that adjustment problems associated with being a student from a family of divorce emerged. Hypothesis 4. College students from families of divorce manifest more distress than those from families with intact marriages independent of their levels of social responsibility and freedom from other-direction. To control for gender differences in the two family status groups on distress, hierarchical MR was again employed to test for distress differences between students from divorce and from intact-marriage families. Here, Step 1 entered gender as a predictor, and Step 2 added family marital status as a predictor. A third step was used in this analysis to determine whether social responsibility and other-direction contributed to the prediction of distress after effects of gender and family status were considered. Social responsibility and other-direction, then, were added as predictors of distress in Step 3. The results of this analysis are shown in Table 3. Family status added a significant increase in the amount of variance accounted for on distress in Step 2 and uniquely accounted for a significant amount of distress variance in the full model (β=.216, p=.002). Students from families of divorce reported more distress than their peers from intact-marriage families despite their more frequent manifestation of the combination of more social responsibility and freedom from other-direction. When this combination of personality characteristics was taken into account in Step 3, the partial correlation for family status (.221) was greater than before those predictors were considered (.174). That is, the greater distress of college students from families of divorce was even more apparent when personality characteristics associated with less distress were controlled. Table 3. Summary of hierarchical MR analysis of demographic and personality factors predicting distress (N=197) Variable Summary for distress score B SE B β Step 1 Gender −3.274 1.355 −.172* Step 2 Gender −3.877 1.360 −.204** Family marital status 3.278 1.347 .174* Step 3 Gender −3.251 1.328 −.171* Family marital status 4.071 1.308 .216** Social responsibility −.240 .091 −.187** Other-direction .366 .136 −.187** R2=.030 (p<.05) for Step 1; ΔR2=.029 (p<.05) for Step 2; ΔR2=.081 (p<.001) for Step 3. * p=.05. ** p=.01. Table options 3.1. Summary These findings indicate that the college students from divorce families did more often show the set of characteristics (social responsibility and freedom from other-direction) associated with both better adjustment and less distress. Moreover, when effects of that set of characteristics on adjustment were taken into account, then the students from divorce families showed significantly more adjustment problems as compared to students from families with intact marriages. Those students from families of divorce also showed more distress even without control for social responsibility and other-direction although those personality characteristics seem to mask the degree of this difference.