| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||
RESEARCH ARTICLE |
1 Department of Child Development and Family Studies, Purdue University, West Lafayette, Indiana.
2 Department of Human Development and Family Studies, Pennsylvania State University, University Park.
3 Institute for Social Research, University of Michigan, Ann Arbor.
Address correspondence to Karen L. Fingerman, PhD, Department of Child Development & Family Studies, Purdue University, 1200 West State Street, West Lafayette, IN 47907. E-mail: karenf{at}purdue.edu
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionReferences |
|---|
Key Words: Intergenerational relationships • Relationship quality • Ambivalence • Family • Parents • Children
Scholars recognize the importance of the intergenerational ambivalence model for understanding complex emotional qualities of ties between adults and their parents (Luescher & Pillemer, 1998
). This model encompasses both sociological and psychological perspectives. Sociological ambivalence considers incompatible normative expectations in a status or role (e.g., Connidis & McMullin, 2002). Psychological ambivalence occurs at the subjective individual level and involves contradictory cognitions, emotions, and motivations toward the same object (Weigert, 1991). The two perspectives share a focus, however, by considering simultaneous positive and negative experiences in the parent–offspring relationship (Willson, Shuey, Elder, & Wickrama, 2006).
Coexisting positive and negative feelings in a relationship may have detrimental effects on well-being. Uchino and colleagues (Uchino, Holt-Lunstad, Smith, & Bloor, 2004) argued that ambivalent feelings lead to poor outcomes because these relationships are unpredictable and cause stress. There has been little research examining the implications of ambivalent feelings between parents and their adult children, however. This study investigated how parents' and offspring's ambivalent feelings (defined as simultaneous positive and negative sentiments) are associated with each party's well-being.
Ambivalence and Well-Being
The premise that ambivalent feelings may be linked to deleterious outcomes is derived from research regarding social networks more broadly. Beginning in the 1970s, epidemiological studies established that positive qualities of social ties enhance physical and mental health; theorists argued that social support and increased motivation to care for oneself may underlie these associations (see Berkman, Glass, Brissette, & Seeman, 2000 for a review). In addition, theorists have noted that close relationships generate positive emotional states that enhance psychological well-being directly and enhance physiological well-being indirectly (Charles & Mavandadi, 2004).
Similarly, distressful relationships appear to have deleterious effects. Evidence from a longitudinal study of gifted children suggested that negative relationships served as a risk factor for mortality (Friedman et al., 1995). Problematic qualities of relationships also were associated with poorer mental health in a large national sample (Newsom, Rook, Nishishiba, Sorkin, & Mahan, 2005). Scholars suggest that such effects may be due to negative emotional experiences in important ties (Rook, 2001).
In this study we focused on intergenerational ambivalence and considered whether the simultaneous experience of positive and negative sentiments detracts from individuals' well-being. The measurement of intergenerational ambivalence warrants comment. Researchers measure relationship ambivalence either indirectly, as a mixture of positive and negative sentiments toward the same person, or directly, as the subjective feeling of being torn (Luescher & Pillemer, 1998). Psychologists have typically assessed ambivalence indirectly, by asking participants to rate contradictory feelings or attitudes toward an object, and then have combined the ratings to create an ambivalence score. Many researchers have favored such indirect approaches because individuals may not be aware of their conflicted feelings (Priester & Petty, 1996).
Researchers also have found that indirect assessments of ambivalence are associated with direct assessments and with well-being. In a study of social networks among college students, Uchino and colleagues (2004) measured ambivalence in both manners and found a moderate correlation. Further, they found that ambivalent ties measured as a mixture of sentiments were more highly associated with depressive symptoms than were solely negative ties. In addition, people who report having more ambivalent network members demonstrated heightened physiological responses (e.g., elevated blood pressure and cardiovascular reactivity) to laboratory stressors (Holt-Lunstad, Uchino, Smith, Olsen-Cerny, & Nealey-Moore, 2003; Uchino, Holt-Lunstad, Uno, & Flinders, 2001).
Several possible mechanisms may underlie a link between ambivalence and poor well-being. First, the negative feelings in a relationship may detract from the positive effects of support. Second, experiencing a mixture of emotions may be more detrimental than experiencing negative emotions alone, because individuals may adjust to a generally negative tone and no longer react strongly or they may avoid negative social partners. Finally, although ambivalent ties include desirable positive elements, such ties may be unpredictable and thus generate stress.
Extending studies linking ambivalence and well-being in the general social network, we focused on ambivalence between adults and their parents. Prior research has documented associations between parent–offspring relationship quality and well-being. Umberson (1992) found that offspring who indicated greater support and less strain with parents reported lower psychological distress. Shaw and colleagues (Shaw, Krause, Chatters, Connell, & Ingersoll-Dayton, 2004) linked adults' memories of emotional support from parents to psychological well-being and physical conditions. Likewise, Silverstein and Bengtson (1991) examined parents' ratings of positive qualities of ties with a randomly selected child; parents who felt closer to this child had a greater likelihood of survival after widowhood.
To the best of our knowledge, only one study has examined intergenerational ambivalence as a predictor of well-being. Lowenstein's (2007) cross-national research found that parental ambivalence toward grown children was associated with parental quality of life, albeit only weakly when personal resources were controlled for in analysis. Lowenstein measured ambivalence as the subjective feeling of being torn, however, rather than as a combination of positive and negative feelings,. Further, Lowenstein focused only on parents. Our first purpose in this study was to examine whether self-reported ambivalence, measured as a mixture of positive and negative feelings, is associated with well-being among parents and their adult children.
Parent–Offspring Relationships and Well-Being
In this study we also provided a within-family approach to intergenerational ambivalence that goes beyond single-respondent data. Several theoretical perspectives emphasize the importance of individuals' beliefs about relationships for their well-being. Dykstra and Mandemakers (2007) referred to the "social construction perspective" of relationships, arguing that an individual's subjective view of the tie is what matters for his or her well-being. Likewise, attachment theory suggests that individuals develop internalized working models of their relationships as they move from childhood to adulthood (Bowlby, 1982). Nevertheless, studies that link self-reports of relationship quality with well-being do not tell us whether the relationship itself contributes to well-being. Therefore, in this study we also examined associations between well-being and the social partner's beliefs.
Interdependence theory suggests that a partner's feelings about the relationship may play a role in the other party's well-being (Kelley & Thibaut, 1978; Rusbult & Van Lange, 2003). Parties who are interdependent react to one another; their well-being is mutually influenced. Interdependence theory has been applied primarily to romantic ties, but it may apply to other relationships as well. In childhood, parents socialize progeny, and a child's well-being is vulnerable to parents' feelings about that child (Grusec, Goodnow, & Kuczynski, 2000). In adulthood, the role of each party's views of the relationship on the other party's well-being is unclear; the day-to-day lives of parents and offspring are distinct, and interdependence in the relationship may be low.
Nonetheless, when a parent or adult child feels mixed emotions, the other party also may experience positive and negative feelings. Two studies of ambivalence that included both generations found that parents' and offspring's ambivalent sentiments were associated (Fingerman, Chen, Hay, Cichy, & Lefkowitz, 2006; Willson et al., 2006). Thus, it is important to consider both parents' and offspring's feelings about the relationship because those feelings may be interdependent. Moreover, it is important to consider each party's feelings about the relationship, even when interest lies in self-reports of relationship qualities. Kenny and Cook (1999) argued, "Even if one wished to adopt an individualist perspective, it still would be valuable to estimate partner effects to show that they are zero" (p. 435).
Here, we examined how each party's ambivalent feelings are associated with his or her own well-being as well as that of the other party by using the actor–partner interaction model (APIM). APIM is a statistical technique developed to test interdependency in relationships (Campbell & Kashy, 2002). This method treats the dyad as the unit of analysis and considers both self-beliefs and partner ratings of the tie. To the best of our knowledge, this is the first study to examine the parent–adult child relationship by using APIM.
Social Structures, Predispositions, Ambivalence, and Well-Being
Relationship ambivalence is unlikely to affect well-being uniformly across individuals. We considered three factors that might explain associations between intergenerational ambivalence and well-being: generation, gender, and neuroticism. The first two variables are features of family and social structure that may shape individuals' experiences in the parent–offspring tie (Rossi & Rossi, 1990). The latter variable pertains to individual predispositions.
Theories of sociological ambivalence suggest that positions within social hierarchies can engender ambivalence (Connidis & McMullin, 2002). The premise of this view of ambivalence is that social positions may be characterized by unclear or contradictory norms, attitudes, and beliefs. In turn, incompatible expectations generate contradictory feelings or ambivalence (Luescher & Pillemer, 1998). Scholars have given particular attention to gender and generational position (i.e., parent vs. offspring) in intergenerational ties (Rossi & Rossi, 1990), and these positions may also be associated with differences in the experience of ambivalence.
With regard to gender, Connidis and McMullin (2002) argued that gendered aspects of intergenerational relationships evoke greater ambivalence for women than men, because women face competing demands between investment in family, unpaid care, work obligations, and other societal demands. These gender differences may extend into differences in patterns of association between ambivalence and well-being. Structured social relations involving gender lead mothers to be more involved than fathers with their offspring and to feel greater responsibility for how children turn out (Ryff, Lee, Essex, & Schmutte, 1994). Thus, mothers may be particularly susceptible to qualities of ties to offspring. Adult children also tend to report feeling closer to mothers than to fathers (Fingerman, 2001; Rossi & Rossi, 1990). Umberson (1992) found that qualities of an adult child's relationship with his or her mother were more strongly associated with well-being than were the qualities of relationships with fathers. The role of offspring gender is less clear. Although a study of ambivalence in a rural population revealed differences between sons and daughters (Willson, Shuey, & Elder, 2003), other studies report no difference in ambivalence for sons and daughters (Fingerman, et al., 2006; Pillemer & Suitor, 2002). Therefore, we examined offspring gender without predictions.
With regard to generational position, Peeters, Hooker, and Zvonkovic (2006) applied life course theory to understand how differing historical circumstances and changes in societal values and roles can contribute to intergenerational ambivalence, particularly for parents. Moreover, the relative investments of parents and children in their tie (with parents being more invested) may account for differences in the impact of ambivalence on well-being (Shapiro, 2004). Greater parental investment in the tie may lead to unclear norms for parental roles with grown offspring. Parents may view their offspring's achievements as a validation of their success as parents (Ryff et al., 1994), but they no longer exert control over grown offspring. Further, individuals tend to experience stronger emotions in situations in which they harbor a personal investment (Lazarus, 1991), and the effects of interdependence are most evident in highly valued relationships (Berscheid & Reis, 1998). Thus, both sociological aspects of a changing society and psychological aspects of parental investment in the tie suggest that associations between ambivalence and well-being will be particularly evident for parents.
Psychological ambivalence focuses on the simultaneity of positive and negative emotional experiences. Such consideration of emotional experience suggests that individuals' traits and predispositions may contribute to differences in intergenerational ambivalence. However, family science has paid scant attention to how an individual's personality contributes to family ties. Even so, a prior study found that parents reported greater ambivalence when they or their child scored higher on neuroticism (Fingerman et al., 2006). Further, theorists argue that individuals who have tendencies toward negativity and emotionality may be particularly sensitive to relationship experiences (Rusbult & Van Lange, 2003). This sensitivity may extend to intergenerational ambivalence. Indeed, the personality trait of neuroticism may moderate associations between ambivalence and well-being.
In sum, in this study of intergenerational ambivalence and well-being we encompassed a sociological ambivalence framework by considering structured social and family relations that shape the relationship (i.e., generation, gender), and a psychological ambivalence framework by considering the personality trait of neuroticism and by assessing ambivalence as a combination of positive and negative feelings. Our research questions were as follows. First, is intergenerational ambivalence associated with poorer psychological and physical well-being for both parties? Second, does each party's ambivalence show an interdependent association with his or her own well-being and the partner's well-being? Third, do these patterns vary on the basis of structural positions within the relationship (i.e., generation, gender)? Finally, do individuals' predispositions (i.e., neuroticism) moderate these associations?
|
METHODS |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
for details).
The study was limited to adult children and parents who resided within 50 miles of each other. Nearly half of adults in the United States live within 50 miles of their parents (Booth, Johnson, White, & Edwards, 1991), and 80% of parents reside within 50 miles of one of their adult children (Lin & Rogerson, 1995). _Parents and offspring who reside near one another tend to have more frequent contact than distant offspring (Lawton, Silverstein, & Bengtson, 1994), and thus, they may have greater opportunity to affect one another's well-being.
Of the participants, 68% were European American and 32% were African American. Although marital status was not a criterion of the study, most parents (89%) and offspring (64%) were married. A majority of parents (54%) and offspring (83%) worked for pay.
Procedure
We recruited most of the participants (85%) by using purchased lists of telephone numbers in the Philadelphia area, including all listed residential phone numbers in the sampled counties and randomly generated numbers with the appropriate area codes. We enhanced recruitment by using convenience methods (e.g., advertisements, church bulletins, and snowball recruitment; 15%), because mixed recruitment approaches may increase minority participation and may generate a sample with more variable relationship quality (Karney et al., 1995). We stratified by age, gender, and race across recruitment techniques.
Our initial phone screening focused on adults aged 22 to 49 years who had two living parents. If we reached a household in which all adults were older than 50 years of age, we screened the household to determine if these adults had adult children. If they had more than one eligible child, we asked the adult child who had the most recent birthday to participate.
Each family member (adult child, mother, father) completed a telephone interview lasting approximately 1 hour. The adult children then participated in face-to-face interviews separately with their mother and their father. During these sessions, parties completed self-report questionnaires in locations where the other party could not observe them. Offspring responded to questions concerning their mothers and fathers, and each parent responded to questions about the target adult child throughout the study. Questions pertaining to mother and to father were presented in random order across offspring.
Measures
Individual and relationship characteristics
Participants provided background information (e.g., marital status, work status, and family relationships). Ninety percent of parents had more than one son or daughter (M = 2.60 children, SD = 1.81). Participants indicated the frequency of contact with the relationship partner on a scale of 1 (never) to 8 (every day). Parents' and offspring's reports of contact were correlated; on average, participants reported speaking by phone at least once a week (mother, M = 6.85, SD = 1.19; father, M = 6.66, SD = 1.16). Most (68%) participants also reported frequent face-to-face contact (once a week).
Participants rated the importance of the parent or adult child, using six categories: 1 (most important person in your life), 2 (among the 3 most important), 3 (among the 6 most important), 4 (among the 10 most important), 5 (among the 20 most important), and 6 (less important than that; see Fingerman, 2001; Fingerman et al., 2006). We reverse coded this item, so higher numbers equal greater importance of the tie. On average, participants rated the other party as among the 3 to 6 most important people in their lives (M = 4.34, SD = 0.93 for mothers; M = 4.48, SD = 0.79 for fathers; M = 4.50, SD = 0.93 offspring rating of mother; M = 4.28, SD = 0.86 offspring rating of father). An analysis of variance comparing the reports of mothers, fathers, and offspring revealed no significant differences in ratings of importance of the tie.
Neuroticism
We used the 12-item Revised Eysenck Personality Questionnaire (Eysenck, Eysenck, & Barrett, 1985) to assess neuroticism; 
=.73. Example yes–no items included "Are you often fed up?" and "Are your feelings easily hurt?"
Ambivalence
We assessed ambivalence by using four items from the American Changing Lives Study (Umberson, 1992): "How much does he or she make you feel loved and cared for?" and "How much does he or she understand you?" for positive qualities, =.69, and "How much does he or she criticize you?" and "How much does he or she make demands on you?" for negative qualities, =.69. These were rated from 1 (not at all) to 5 (a great deal). These items present advantages for assessing ambivalence. The items are balanced across the positive and negative dimensions. They refer to behaviors rather than global impressions of the relationship. These items have also been used consistently in studies of intergenerational ambivalence (Fingerman et al., 2006; Willson et al., 2003).
Social psychologists use a variety of formulas to combine two polarized scales and estimate ambivalence (see Thompson, Zanna, & Griffin, 1995 for a review). It is important to use a formula that distinguishes between intense positive and negative feelings (ambivalence), and indifference or absence of feelings toward an object. As in other studies of intergenerational ambivalence (Fingerman et al., 2006; Willson, et al., 2003, 2006), we applied Griffin's formula to calculate ambivalence scores:
|
|
Well-being
Because our predictions pertained to global physical and psychological well-being, we created composite measures of these constructs. Consistent with other studies (Antonucci, Ajrouch, & Janevic, 2003), to create a physical health composite, we averaged standardized self-assessments (z scores) of health rated from 1 (poor) to 5 (excellent) and chronic medical conditions. We recoded this index such that a higher score indicates better health. Because of the low frequency of disease among offspring, the physical health score for adult children included only self-rated health.
For psychological well-being, we standardized and averaged a one-item assessment of life satisfaction (Sweet, Bumpass, & Call, 1988), rated on a scale of 1 (very dissatisfied) to 10 (completely satisfied), and the 11 Center for Epidemiological Studies–Depression scales (reverse coded; Radloff, 1977), =.83. Higher scores on this composite represent better psychological well-being. This approach is consistent with the World Health Organization recommendation that positive and negative feelings are key components of psychological functioning (World Health Organization Quality of Life Group, 1998).
|
RESULTS |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
; Kenny, Kashy, & Cook, 2006). APIM allowed us to consider self-report (i.e., actor data) and partner reported data in the same equation predicting each party's well-being. We estimated APIM by using Multilevel Models in SAS PROC Mixed. The composite scales for psychological well-being and physical health served as dependent variables. Independent variables included parents' and offspring's ambivalence scores as self (actor; his or her own ratings) and other (partner; the other party's) ratings. In other words, for each person's well-being, we considered associations with his or her own ambivalence scores as well as with the partner's (the parent's or offspring's) ambivalence scores. APIM accounts for correlations or interdependency in these reports.
We also examined whether patterns of association between ambivalence and well-being differed by (a) generation, (b) gender, and (c) neuroticism as follows. With regard to generation, parents and adult children fit criteria for distinguishable dyads in APIM (Kenney et al., 2006). That is, the parent is always the parent and the offspring is always the offspring. An indistinguishable dyad might include two friends or two roommates. Following the advice of Kenny and colleagues (2006), we included interaction terms between generation and ambivalence to handle distinguishable partners in APIM. Thus, we estimated the models with generation (1 = parent, 0 = offspring), each person's self-reported ambivalence, each partner's reported ambivalence, and the two interaction terms between those reports and generation. We centered ambivalence scores on the grand mean prior to estimating interactions.
With regard to gender, we considered whether associations between ambivalence scores and well-being varied by each party's gender in preliminary analyses by including three-way interactions for Parent Gender x Generation x Ambivalence (with constituent two-way interactions and main effects). The three-way interactions were significant for physical health, but not for psychological well-being. For ease of presentation, we present final models separately for dyads involving mothers and fathers.
We also considered offspring gender in three-way interactions, but patterns did not differ for sons and daughters. Thus, we simply controlled for offspring gender in the final models.
For neuroticism, we estimated models including neuroticism in a three-way interaction with Neuroticism x Generation x Ambivalence and constituent interaction terms and main effects. We centered ambivalence and neuroticism on their grand means prior to calculating interaction terms.
Finally, in preliminary analyses, we considered family contextual variables. We did not find significant associations between well-being and number of siblings or marital status; nor did we find moderating effects of these variables on associations between ambivalence and well-being. We also expected differences in importance of the tie to explain generational differences in associations between ambivalence and well-being. As we mentioned previously, we did not find generational differences in ratings of importance of the tie. Therefore, for parsimony in presentation, we did not include family context variables or importance in the final models.
Demographic characteristics, such as ethnicity and education, may be associated with salience of the parent–offspring tie, and with well-being (Hill & Sprague, 1999; Hayward, Miles, Crimmins, & Yang, 2000). Thus, we included the following control variables in the models: age, ethnicity, education, frequency of contact, and offspring gender. Table 1 includes descriptive information regarding variables in the models.
|
|
). As one can see in Figure 1, mothers were in poorer physical health than offspring overall, but when adult children experienced greater ambivalence toward their mothers, their mothers were in poorer health. Mothers' ambivalence toward their offspring was not strongly associated with offspring health.
|
|
|
Ambivalence, Neuroticism, and Well-Being
We also estimated a model including neuroticism in a three-way interaction, Neuroticism x Generation x Ambivalence, the constituent interactions, and main effects. These models addressed self-reports, and also addressed how a partner's reported ambivalence (e.g., the father) is associated with the other party's well-being (e.g., the offspring), considering that other party's (e.g., the offspring's) level of neuroticism.
Interactions involving neuroticism were not significant for mother–offspring dyads. Therefore, we do not present these findings.
Table 4 presents findings for father–offspring dyads. For psychological well-being, the three-way interaction for Neuroticism x Generation x Self-Reported Ambivalence was significant. For physical health, partner-reported ambivalence mattered; the three-way interaction for Neuroticism x Generation x Partner-Reported Ambivalence was significant. In both models, significant main effects for neuroticism indicated that individuals who scored higher on neuroticism reported poorer psychological and physical well-being.
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
; Umberson, 1992). Second, we examined mothers, fathers, and offspring within the same family, allowing us to consider self-reports and partner reports of ambivalence. Third, we contributed to the sociological ambivalence literature by examining how associations between ambivalence and well-being varied as a function of structural position in the family and in society (i.e., generation and parental gender). Finally, we contributed to psychological ambivalence literature by showing that individual personality traits (i.e., neuroticism) moderate associations between ambivalence and well-being in some instances.
Ambivalence and Well-Being
As we expected, for all three parties—mothers, fathers, and offspring—greater self-reported intergenerational ambivalence was associated with poorer psychological adjustment. Thus, like negative qualities of relationships (Umberson, 1992), mixed emotions between adults and their parents are also associated with more depressive symptoms. However, the findings extend knowledge about relationships between adults and their parents by suggesting that mixed emotions also are associated with poorer mental well-being. Research on social networks provides similar findings. Uchino and colleagues (2004) classified relationships as primarily positive, primarily aversive, or emotionally mixed; the emotionally mixed relationships showed stronger associations with poor psychological well-being than aversive ties did. The current study suggests that intergenerational ambivalence also is associated with psychological distress. When adults feel both positive and negative feelings toward a parent or offspring, they may experience lower psychological well-being because they care about the other party's feelings and desire a positive connection. Not knowing what to expect during each encounter (e.g., conflict or a positive interaction) also may engender stress. In addition, ambivalence may involve love for a person who does not reciprocate that affection or who is disappointing in some other respect. Of course, because of the cross-sectional nature of this study, it is possible that people with poorer psychological well-being experience greater ambivalence. People who are more depressed tend to elicit negativity in relationships and also may interpret interactions more negatively (Gotlib & Beatty, 1985). A key feature of the present study also involved inclusion of partner reports of ambivalence; controlling for the partner's views of the tie, we found that self-reports of ambivalence were associated with well-being. Future research should attempt to disentangle how such mixed feelings are associated with poorer emotional well-being.
This within-family study also revealed complex associations between a partner's intergenerational ambivalence and physical health. Associations between partner ambivalence and well-being were not consistent across mother and father relationships. A mother's physical health was associated with her adult child's feelings of ambivalence, whereas an adult child's health was associated with his or her father's feelings of ambivalence. The sociological ambivalence framework suggests that gender plays a key role in intergenerational ambivalence (Connidis & McMullin, 2002), but that framework primarily pertains to the competing demands regarding women's roles in relationships and other societal expectations. In this study, we did not find a consistent pattern of stronger associations for women than men. Instead, parental gender appeared to play a different role in different relationships.
Interpreting parental gender differences is difficult because of the use of snapshot measurement. Scholars have consistently linked qualities of a parent's and adult child's relationship to well-being by using cross-sectional data, and they presumed that the relationship qualities predict well-being (e.g., Lowenstein, 2007; Shaw et al, 2004; Umberson, 1992). Nonetheless, researchers interested in intergenerational ambivalence speculate that ambivalence arises when a parent's health declines (Fingerman, Hay, Kamp-Dush, Cichy, & Hosterman, 2007; Spitze & Gallant, 2004). Findings in the mother–offspring dyad seem to support this premise regarding ambivalence. When maternal health declines, the mother's illness may evoke ambivalent feelings in her children. Mothers are more involved in caring for children and children expect mothers to serve nurturing roles, even after they are grown. Adult children may help their mothers more when the mothers are in poor health, but they may experience conflicted feelings in doing so (Fingerman et al.). Their mothers also may appear to be more critical and demanding as a consequence of poor health. Indeed, our prior research revealed that offspring experience ambivalence even when they simply worry about maternal health declines (Hay, Fingerman, & Lefkowitz, 2007). Thus, the partner effect may be in reverse; the situation may be that offspring experience greater ambivalence as the mother's health declines, rather than that offspring ambivalence generates those declines.
The pattern with regard to fathers is more perplexing. Although adult children were generally in good health, their health showed a negative association with paternal ambivalence. Elsewhere, Umberson (1992) reported that offspring were more sensitive to relationships with their mothers than to relationships with their fathers. In this study, adult children's health was worse when fathers were more ambivalent. Unfortunately, we suffer a dearth of research on relationships between adults and fathers. Nonetheless, we might speculate that offspring's poor health generates ambivalence for fathers (rather than the reverse). Fathers in this generation were not highly involved in caring for their children. An adult child's illness may lead to ambivalent feelings for fathers who expect their offspring to be independent and who are not accustomed to caring for them.
The findings regarding partner effects also introduce questions about communication of ambivalence in this tie. Observational data reveal that adults communicate in distinct ways with mothers and fathers (Cichy, Lefkowitz, & Fingerman, 2008). Nevertheless, partner effects do not necessarily indicate that partners use direct and open communication; people can communicate feelings unconsciously or in subtle ways that are difficult to measure (Kenny & Cook, 1999). Moreover, as we mentioned previously, the partner effects regarding fathers may reflect cross-sectional associations between health and ambivalence.
In addition, we note the lack of findings regarding the gender of the adult children. The paucity of gender differences here may reflect a bias in the sample. Men who agree to participate in studies with their parents may be more similar to women than men in the general population. Nonetheless, other studies in urban areas in recent cohorts have reported few differences in sons' and daughters' relationships with parents (e.g., Fingerman et al., 2007; Logan & Spitze, 1996), and a national study examining an array of issues (not limited to parents and offspring) also failed to report gender of offspring differences (Shaw et al., 2004). Thus, recent cohorts of sons and daughters may experience fewer differences in relationships with their parents than prior cohorts did.
Finally, the findings present a surprising perspective on how personality traits may moderate family ties. The basic associations between neuroticism and well-being were consistent with our expectations; offspring and fathers who scored higher on neuroticism showed poorer well-being. However, associations between ambivalence and well-being were more evident among offspring or fathers with lower neuroticism. The direct effect of neuroticism on well-being may be so strong that it is only in the absence of high neuroticism that associations between intergenerational ambivalence and well-being are evident. The effects of a particular emotion-laden family relationship may be greater for individuals who do not typically experience relationships as emotion laden. Future research on intergenerational ambivalence should further consider within-family associations between personality and ambivalence, with particular attention as to why these associations may appear only in some (e.g., father–offspring) relationships and not in other (e.g., mother–offspring) relationships.
In sum, this study extended prior research on relationships between adults and their parents by considering associations between both party's experience of ambivalence and their well-being. Self-reports of ambivalence were associated with emotional well-being, whereas partner's reports were associated with physical health. Further, these patterns differed for mothers and fathers, suggesting that adult children have distinct relationships with each parent that warrant further investigation. Likewise, neuroticism played a role in the pattern of associations, but primarily when neuroticism was low. Thus, it is important for scholars to consider an individual's structural positions within intergenerational relationships and his or her personality traits to understand when and why qualities of intergenerational ambivalence may be associated with well-being.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received for publication May 8, 2007. Accepted for publication May 23, 2008.
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
| ||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
