Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2007 Aug 17.
Published in final edited form as: J Marriage Fam. 2006 May;68(2):442–454. doi: 10.1111/j.1741-3737.2006.00263.x

Linked Lives: Adult Children's Problems and Their Parents' Psychological and Relational Well-Being

Emily A Greenfield 1, Nadine F Marks 1
PMCID: PMC1950122  NIHMSID: NIHMS10754  PMID: 17710218

Abstract

This study examined associations between adult children's cumulative problems and their parents' psychological and relational well-being, as well as whether such associations are similar for married and single parents. Regression models were estimated using data from 1,188 parents in the 1995 National Survey of Midlife in the United States whose youngest child was at least 19 years old. Participants reporting children with more problems indicated moderately poorer levels of well-being across all outcomes examined. Single parents reporting more problems indicated less positive affect than a comparable group of married parents, but married parents reporting more problems indicated poorer parent-child relationship quality. Findings are congruent with the family life course perspective, conceptualizing parents and children as occupying mutually influential developmental trajectories.

Keywords: family relationship quality, family stress, intergenerational family relations, life course perspective, marital status, mental health

Distressing circumstances—such as substance abuse, divorce, and long-term unemployment—are not uncommon among adults in the United States (Allegreto & Stettner, 2004; Kreider & Fields, 2002; Pham-Kanter, 2001). Empirical studies have indicated that these problems not only jeopardize the well-being of adults who directly experience them but also endanger the well-being of distressed individuals' family members, including their children and partners (e.g., Bigatti & Cronan, 2002; Werner & Smith, 2001). There has been little empirical investigation, however, of the extent to which adults' problems affect other family members, in particular individuals' parents.

Guided by a family life course perspective, this study investigated linkages between adult children's cumulative number of problems and their parents' well-being. More specifically, we used data from the 1995 National Survey of Midlife in the United States (MIDUS) to test the hypothesis that parents of adult children experiencing more types of problems would report poorer psychological well-being (more negative affect, less positive affect, and less self-acceptance) and poorer relational well-being (poorer parent-child relationship quality and more family relationship strain) than parents of adult children experiencing fewer types of problems. We also examined whether being married (vs. single) moderates these associations, hypothesizing that being married would help to buffer parents from the negative implications of their adult children's greater number of problems.

Theoretical and Empirical Background

This study draws on the family life course perspective as a theoretical framework that orients scholarly attention to the possible linkages between adult children's problems and their parents' well-being. One of the central propositions of the life course perspective is that of linked lives—that is, that people in salient relationships with each other, such as parents and children, occupy mutually influential interlocking developmental trajectories that extend throughout their lives (Elder, Johnson, & Crosnoe, 2003). This theorizing suggests that circumstances in adult children's lives would have implications for the lives of their parents and, accordingly, that adult children's problems would affect their parents' psychological and relational well-being.

Theoretical work and empirical studies on adult parent-child relationships, as well as on stressor events and social networks, implicate several ways through which adult children's problems might jeopardize multiple aspects of their parents' well-being. First, research on distressing circumstances in individuals' social networks suggests that adult children's problems might jeopardize their parents' well-being directly by heightening feelings of stigma, loss, fear, and worry (Maurin & Boyd, 1990), or indirectly by introducing care demands that can create psychological and relational distress (Jendrek, 1993). Second, systems-oriented frameworks, such as family therapy theories (Doherty & Baptiste, 1993), suggest that problems occurring within one family relationship can place strain on other family relationships, which can lead to individuals' poorer psychological and relational well-being. Third, because many believe that parents have a critical influence on how their children develop, adult children's problems might cause parents to think that they have, in part, failed as parents (Ryff, Lee, Essex, & Schmutte, 1994); parents of adult children with problems, therefore, might experience less well-being, particularly in terms of their self-evaluations. Finally, adult children's greater number of problems might jeopardize their parents' well-being by violating parents' normative expectations for their children (Collins & Russell, 1991) and by hindering parents' and children's achievement of mutuality—or interdependence and egalitarianism—in their relationship (Nydegger, 1991). These final points indicate how adult children's problems might be particularly threatening to parents' well-being in terms of the quality of their parent-child relationships.

Although the current literature, as a whole, provides a strong basis from which to hypothesize that adult children's problems can threaten their parents' psychological and relational well-being, few empirical studies have directly investigated these linkages. Pillemer and Suitor (1991) provide one example of an empirical study that has tested for such associations. Using data from parents aged 65 – 100 years in a Canadian national probability sample, the authors compared the well-being of parents of children with at least one type of problem (mental or emotional problems, serious physical health problems, alcohol problems, and/or having a child who had undergone “some serious stress” over the past year) to parents of children without any of those problems. Regression analyses demonstrated that across parents' gender and parent-child living arrangements and controlling for parents' self-rated health, marital status, educational attainment, and access to a confidante, parents of children with at least one problem reported more depressive symptoms than parents of children without problems. More recently, Pillemer and Suitor (2003) investigated linkages between adult children's problems and their mothers' relational well-being. They found that participants reported feeling more ambivalence toward adult children with problems than toward adult children without problems. These findings suggest that adult children's problems are associated with both their parents' poorer psychological and relational well-being.

Despite these findings, knowledge regarding the associations between adult children's problems and parents' well-being is far from complete. First, drawing on a cumulative risk model that suggests that stressors have a multiplicative effect on well-being that is not captured by additive models (Sameroff & Fiese, 2000), as well as considering stress proliferation processes whereby a single stressor in an individual's life can create other stressors (Pearlin, Aneshensel, & Leblanc, 1997), research is necessary to investigate the extent to which adult children's cumulative number of problems is associated with their parents' poorer well-being. Second, previous studies have indicated different patterns of predictors for various dimensions of psychological well-being (Bradburn, 1969; Ryff & Keyes, 1995), as well as the multidimensional nature of intimate relationship quality (Rook, 1998). As such, research is needed to investigate the associations between adult children's problems and diverse aspects of their parents' well-being. Third, although previous studies have investigated gender and coresidence as possible moderators of the associations between adult children's problems and their parents' well-being (Pillemer & Suitor, 1991), other factors might make certain subgroups of parents less vulnerable than others to their adult children's problems. Given previous empirical work suggesting that being married can protect individuals from the negative effects of stressors (Dooley, Prause, & Ham-Rowbottom, 2000; Reid & Hardy, 1999; Rodrigue & Park, 1996), we posited that the well-being of married parents would be less vulnerable than that of single parents to adult children's greater number of problems.

In people's lived experiences, linkages between adult children's problems and parents' well-being are likely complex. In addition to adult children's problems affecting their parents' well-being through a variety of mechanisms, it is also possible that parents' poorer psychological and relational well-being can heighten their children's susceptibility to experiencing problems and that shared characteristics between parents and their children might account for both adult children's problems as well as their parents' poorer well-being. Consider, for example, the personality trait of neuroticism, or the propensity to experience negative emotional instability (McCrae & John, 1992). Previous research has indicated that personality traits such as neuroticism are, in part, genetically based (Johnson & Krueger, 2004). An individual with strong neurotic tendencies, therefore, might be more likely to have children who also have such tendencies. Given previous findings indicating that neurotic individuals are at heightened risk for experiencing a greater number of stressful life events (see Ormel, Rosmalen, & Farmer, 2004, for a review), and that neurotic individuals, on average, experience poorer psychological well-being (Keyes, Shmotkin, & Ryff, 2002), neuroticism might account for both adult children's greater number of problems as well as parents' poorer well-being. This logic indicates the importance of controlling for parental characteristics that are related to both parents' well-being and their reports of adult children's problems, such as parents' neuroticism, educational attainment, employment status, race/ethnicity, income, financial strain, physical health, age, gender, total number of children, and intergenerational coresidence (Aquilino & Supple, 1991; Marks & Choi, 2002; Mroczek & Kolarz, 1998; Ryff & Keyes, 1995). Associations between adult children's problems and parents' well-being that remain after controlling for such factors provide stronger evidence that adult children's problems, at least in part, jeopardize parents' well-being.

In sum, guided by a family life course perspective and previous research, the purpose of this study was to address gaps in understanding of the linkages between adult children's problems and their parents' well-being by (a) employing a “cumulative” risk approach to operationalizing adult children's problems (i.e., developing a cumulative index of problems), (b) considering multiple dimensions of parents' psychological and relational well-being, (c) considering marital status as a potential moderator of associations between adult children's problems and their parents' well-being, and (d) controlling for key factors so as to provide more compelling evidence that adult children's problems might, in part, cause their parents' poorer well-being. More specifically, this study tested two hypotheses:

  1. Parents who report that their adult children have more types of problems will indicate poorer psychological well-being (more negative affect, less positive affect, and less self-acceptance) and poorer relational well-being (poorer parent-child relationship quality and more family relationship strain) than parents who report that their adult children have fewer types of problems.

  2. A greater number of adult children's problems will more strongly jeopardize the well-being of parents who are single rather than married.

Method

Data

This study used data from the 1995 MIDUS. The MIDUS national probability sample that answered both a telephone and mailback survey included 3,032 English-speaking, noninstitutionalized adults, who were between the ages of 25 and 74 when interviewed in 1995. Approximately 82% (N = 2,496) of the total sample reported having biological, adopted, and/or stepchildren. Because survey items assessing this study's main independent variable did not distinguish between problems of younger children and problems of adult children, we limited our analytic sample to the 1,188 respondents whose youngest child was at least 19 years of age. This restriction ensured that our measure of children's problems was exclusively assessing adult children's problems.

The MIDUS sample was obtained through random-digit dialing, with oversamples of older adults and men to ensure an adequate distribution on the cross-classification of age and gender. Sampling weights correcting for selection probabilities and nonresponse allow the sample to match the composition of the U.S. population on age, gender, race, and education. For this study, multiple regression analyses were conducted with both weighted and unweighted data. Because no major differences in results were found, results from the unweighted analyses are reported because these analyses provide estimates with more reliable standard errors (Winship & Radbill, 1994).

The response rate for the MIDUS telephone interview was 70%, and the response rate for the mailback questionnaire was 86.8% of the telephone respondents. Therefore, the overall response rate for the sample that completed both the telephone interview and mailback questionnaire was 60.8%. (For a detailed technical report regarding field procedures, response rates, and weighting, see http://midmac.med.harvard.edu/research.html#tchrpt.)

Dependent Variables

Negative affect and positive affect

Two 6-item scales new to the MIDUS were used to measure positive and negative affect (Mroczek & Kolarz, 1998). To assess negative affect, participants were asked how frequently in the past 30 days they felt (a) so sad nothing could cheer them up, (b) nervous, (c) restless or fidgety, (d) hopeless, (e) that everything was an effort, and (f) worthless. To assess positive affect, participants were asked how frequently in the past 30 days they felt (a) cheerful, (b) in good spirits, (c) extremely happy, (d) calm and peaceful, (e) satisfied, and (f) full of life. Respondents answered each of the 12 affect items on a 5-point scale (1 = all of the time, 5 = none of the time). Items were reverse coded and summed such that higher scores indicated more negative or more positive affect. Cronbach's αs were .86 and .91 for the negative affect and positive affect scales, respectively.

To ease comparison across models estimated for different well-being outcomes (refer to the data analytic sequence described below), we standardized participants' scores on the dependent variables before estimating statistical models. Standardizing participants' scores on the outcome variables allows for more readily discerning the relative size of associations between adult children's problems and different aspects of their parents' psychological and relational well-being. Table 1 displays descriptive statistics and correlations for all analytic variables.

Table 1.

Correlations and Descriptive Statistics for All Analytic Variables

graphic file with name nihms-10754t1.jpg

Open in a new tab

Note: Data from the 1995 National Survey of Midlife in the United States, respondents with at least one biological, adopted, or stepchild who is 19 years of age or older (N = 1,188).

a

Means for dichotomous variables are reported as proportions.

Self-acceptance

A 3-item version of Ryff's self-acceptance index was used to assess participants' self-acceptance (Ryff & Keyes, 1995). Ryff created this version for large survey use as an additive measure designed to represent the conceptual breadth of self-acceptance, which she found in factor analyzing her 20-item scale. One subfactor, related to having a positive attitude toward one's self, is indicated with the item, “I like most parts of my personality.” A second factor, related to feeling positive about one's past life, is indicated with the items, “When I look at the story of my life, I am pleased with how things have turned out so far” and “In many ways I feel disappointed about my achievements in life.” For all 3 items, respondents were asked to indicate the degree to which they agree or disagree with the statements on a 6-point continuum (1 = strongly disagree; 6 = strongly agree). The item regarding participants' feeling disappointed about their achievements in life was reverse coded, and all items were summed such that higher scores indicated more personal growth. This additive index is correlated highly (r < .70) with its parent 20-item, highly reliable scale (Ryff & Keyes).

Parent-child relationship quality

Participants were asked a series of questions about their children in general. One item asked, “Using a scale from 0 to 10 where 0 means the worst possible relationship and 10 means the best possible relationship, how would you rate your overall relationship with your children these days?” Preliminary analyses detected a negative skew in the distribution of participants' responses to this item. Forty-one percent of the participants responded with a 10, and 26% responded with a 9. In an effort to partially correct for the skewed distribution while preserving variation on this index, responses were recoded (0 – 6 = 0; 7 = 1; 8 = 2; 9 = 3; 10 = 4) with higher scores indicating better parent-child relationship quality.

Family relationship strain

Two 4-item scales adapted from Schuster, Kessler, and Aseltine (1990) were used to assess participants' family relationship strain. Participants were asked to indicate the degree to which their family members, not including their spouse or partner, (a) make too many demands on them, (b) criticize them, (c) let them down, and (d) get on their nerves. Participants responded to each item on a 4-point scale (1 = often; 4 = never); scores were reverse coded and averaged such that higher scores indicated more family relationship strain. Cronbach's α was .78.

Independent Variables

Adult children's problems

Participants were asked to indicate whether each of 10 types of problems applied to any of their children in the past 12 months: (a) chronic disease or disability, (b) frequent minor illnesses, (c) emotional problems (e.g., sadness, depression), (d) alcohol or substance problems, (e) financial problems (e.g., low income, heavy debts), (f) problems at school or work (e.g., failing grades, poor job performance), (g) difficulty finding or keeping a job, (h) marital or partner relationship problems, (i) legal problems (e.g., involved in law suits, police charges, traffic violations), and (j) difficulty getting along with people. For each item, respondents were coded as 1 if they indicated that any of their children had experienced that problem or 0 if they indicated that none of their children had experienced that problem or that the problem did not apply to any of their children. Preliminary analyses revealed that bivariate correlations between these 10 types of problems ranged from .04 (chronic disease or disability and difficult getting along with people) to .42 (problems at school or work and difficulty getting along with people). The prevalence of each type of problem ranged from 8% of respondents (alcohol or substance problems) to 38% of respondents (financial problems). All bivariate correlations of specific problems with each outcome were in the expected direction and ranged from .05 (chronic disease or disability and parent-child relationship quality) to .25 (financial problems and family relationship strain).

To create an index of adult children's cumulative problems, we summed participants' responses to the 10 items. Cronbach's α for this cumulative index was .74. Given the small number of participants who reported their children as having 8, 9, or 10 types of problems (15, 4, and 11 respondents, respectively), participants whose children had 8 or more of the 10 problems were combined into a single group, giving this measure a range of 0 – 8. In preliminary analyses, we regressed the measures of parents' well-being on a series of dichotomous variables for 1 through ≥8 problems (0 problems, reference group) and found a consistent monotonic relationship between adult children's greater number of problems and parents' poorer ratings on outcomes. These analyses provided empirical justification for the use of a cumulative index in our statistical models.

Respondents' marital status

Participants indicated whether they were (a) married (66.84%), (b) separated (1.68%), (c) divorced (19.78%), (d) widowed (12.71%), or (e) never married (.67%). Participants who reported that they were married were coded as 1 on marital status; all other participants were coded as 0.

Interaction of adult children's problems by respondents' marital status

An interaction score was assigned to each participant by multiplying the total number of adult children's problems by whether the respondent was married.

Control Variables

Dichotomous variables were created for respondents' gender (1 = female) and employment status (1 = currently employed). A categorical variable was created to indicate respondents' educational attainment, including the categories of <12 years, 12 years (reference group), 13 – 15 years, and ≥16 years. Similarly, a categorical variable was constructed for race/ethnicity, including the categories of non-Hispanic White (reference group), Black, Latina/o, and other. Age was calculated as years since birth at the time of the telephone interview. Income was computed by combining participants' personal annual income with that of their spouse (if applicable). Self-rated health was measured by a global self-assessed health question, which asked participants, “In general, would you say your health is…?” (1 = very poor to 5 = excellent). A 4-item scale was used to assess neuroticism. Participants were asked to indicate the extent to which the following adjectives describe them on a 4-point scale: (a) moody, (b) worrying, (c) nervous, and (d) calm. The item indicating calm was reverse coded, and responses were averaged such that higher scores indicated more neuroticism. Cronbach's α for this scale was .78. Participants' financial strain was measured by a single item that asked participants, “How difficult it is for you (and your family) to pay your monthly bills?” (1 = very difficult; 4 = not at all difficult). Responses were reverse coded such that higher scores indicated more financial strain. Participants' total number of adult children was coded as the number of biological, adopted, and/or stepchildren that they reported. Finally, participants' intergenerational coresidence status was measured by an item that asked, “During the past 12 months, have you had [one or more your adult children and/or grandchildren] living in your home as their place of residence? Visiting overnight, even for an extended period, does not count as living with you according to this definition.” Participants who responded yes to this question were coded as 1, and participants who responded no were coded as 0.

Missing Data

Listwise deletion was employed to handle missing data when estimating multivariate models and constructing measures comprising multiple items (with the exception of the measure of race, for which respondents who did not provide a valid response to the categories of non-Hispanic White, Black, or Latina/o were coded as other). No more than 5% of the analytic subsample was missing on any one of the measures. Bivariate correlations between being missing on the main variables of substantive interest (i.e., the five dependent variables and the measure of adult children's problems) and the sociodemographic control variables indicated that participants with missing data were overall more likely to be non-White, older, in poorer physical health, and to have lower levels of education and household income.

Data Analytic Sequence

We used the ordinary least-squares method to estimate multiple regression models to test the proposed linkages between parents' well-being and adult children's problems. We first tested for gender differences in the associations between adult children's cumulative number of problems and parents' well-being. Findings from previous studies suggest that because women are more likely to be emotionally involved in others' lives than are men (Gilligan, 1982; Kessler & McLeod, 1984), negative events in network members' lives might jeopardize women's well-being more strongly than men's. Additionally, because our analytic sample includes participants who were parents of young children during a historical period in which mothers were expected to be primarily responsible for childrearing (Cancian & Oliker, 2000), we speculated that women might feel particularly responsible for their adult children's problems and therefore might experience poorer well-being because of these problems than men. Accordingly, we estimated five preliminary models (one for each outcome) that included an interaction variable for respondents' gender by their adult children's cumulative number of problems. None of the coefficients for the interaction terms reached statistical significance (p ≤ .05, two-tailed). Therefore, we proceeded by analyzing men and women together.

To test whether adult children's problems are associated with parents' poorer well-being (Hypothesis 1), models were estimated by regressing each dimension of well-being separately on the control variables, respondents' marital status, and parents' reports of their adult children's number of problems. To test whether marital status moderates the associations between adult children's problems and their parents' well-being (Hypothesis 2), the interaction term (adult children's problems by respondents' marital status) was added to the models.

Results

Adult Children's Problems and Parents' Well-Being

Tables 2 and 3 provide a summary of results regarding linkages between adult children's problems and their parents' psychological well-being (positive affect, negative affect, and self-acceptance) and relational well-being (parent-child relationship quality and family relationship strain), respectively. To examine initial evidence for the first hypothesis regarding associations between adult children's problems and parents' well-being, we estimated models that regressed each dimension of well-being on number of adult children's problems (Tables 2 and 3, Model 1). Results indicated that parental reports of adult children with more types of problems were associated with parents' poorer well-being across all five outcomes examined. Parents reporting adult children with a greater number of problems indicated less positive affect (b = −.05, p ≤ .001), more negative affect (b = .05, p ≤ .001), less self-acceptance (b = −.03, p ≤ .05), poorer parent-child relationship quality (b = −.10, p ≤ .001), and more family relationship strain (b = .13, p ≤ .001).

Table 2.

Summary of Regression Analysis for the Effects of Adult Children's Problems on Their Parents' Psychological Well-Being

graphic file with name nihms-10754t2.jpg

Open in a new tab

Note: Data from the 1995 National Survey of Midlife in the United States, respondents with at least one biological, adopted, or stepchild who is 19 years of age or older. Analyses given unweighted data with listwise deletion. Unstandardized coefficients reported in terms of standardized dependent variables.

*

p ≤ .05.

**

p ≤ .01.

***

p ≤ .001 (two-tailed).

Table 3.

Summary of Regression Analysis for the Effects of Adult Children's Problems on Their Parents' Relational Well-Being

Parent-Child Relationship Quality
 Family Relationship Strain
Model 1
 Model 2
 Model 1
 Model 2
B SE β B SE β B SE β B SE β
Control variables
 Female  .39 .06  .19***  .39 .06  .20***  .15 .06  .08**  .15 .06  .08**
 Age  .02 .00  .18***  .02 .00  .18*** −.02 .00 −.16*** −.02 .00 −.16***
 Income −.00 .00 −.04 −.00 .00 −.04  .00 .00  .09**  .00 .00  .09**
 Financial strain −.08 .04 −.06 −.08 .04 −.07*  .15 .04  .13***  .15 .04  .13***
 Employed  .08 .07  .04  .08 .07  .04 −.01 .07 −.01 −.01 .07 −.01
 Education (years)
  <12 −.05 .10 −.02 −.05 .10 −.02 −.14 .10 −.04 −.13 .10 −.04
  13 – 15 −.12 .08 −.06 −.13 .08 −.06  .02 .07  .01  .02 .07  .01
  16+ −.14 .09 −.06 −.15 .09 −.07  .09 .08  .04  .09 .08  .04
 Race/ethnicity
  Black  .14 .15  .03  .13 .15  .03  .05 .14  .01  .05 .14  .01
  Latina/o  .27 .18  .06  .27 .18  .05  .19 .16  .03  .20 .16  .03
  Other  .30 .21  .04  .26 .22  .04  .14 .20  .02  .15 .20  .02
 Coresidence −.09 .07 −.04  .26 .22  .04  .16 .06  .07*  .16 .06  .07**
 Total children  .00 .02  .01  .01 .02  .02  .02 .02  .04  .02 .02  .04
 Self-rated health  .04 .03  .04  .04 .03  .04  .02 .02  .04  .02 .02  .04
 Neuroticism −.14 .05 −.09** −.13 .05 −.09**  .27 .05  .17***  .27 .05  .17***
Main effects
 Married  .22 .07  .10***  .33 .09  .15*** −.03 .07 −.01 −.08 .08 −.04
 Children's problems −.10 .02 −.19*** −.06 .02 −.12*  .13 .02  .26***  .11 .02  .23***
Interaction
 Adult Children's Problems × Married — — — −.06 .03 −.11* — — —  .03 .03  .04
Constant −1.15 .36 −1.21 .36 −.17 .33 −.14 .33
Adjusted R2 .12 .12 .24 .24
Valid N 1,036 1,036 1,026 1,026
Open in a new tab

Note: Data from the 1995 National Survey of Midlife in the United States, respondents with at least one biological, adopted, or stepchild who is 19 years of age or older. Analyses given unweighted data with listwise deletion. Unstandardized coefficients reported in terms of standardized outcomes.

*

p ≤ .05.

**

p ≤ .01.

***

p ≤ .001 (two-tailed).

Being Married as a Possible Protective Factor for Parents' Well-Being

To examine evidence for the second hypothesis regarding whether parents' marital status moderates the associations between adult children's problems and their parents' well-being, we added the interaction term between adult children's cumulative number of problems and parents' marital status to the previous set of models (Tables 2 and 3, Model 2). Results demonstrated a statistically significant association between the interaction term and positive affect (b =.05, p ≤.05) and between the interaction term and parent-child relationship quality (b = −.06, p ≤.05), but no statistically significant associations were found between the interaction term and negative affect, self-acceptance, or family relationship strain. To further interpret the statistically significant interactions, we calculated predicted scores for parents' positive affect and parent-child relationship quality across two different levels of adult children's problems by using estimates from Table 2, Model 2. The baseline model for predicted scores included respondents at the mean level on continuous variables and in the 0 categories for dichotomous variables.

As Figure 1A demonstrates, married parents, in contrast to single parents, experienced a smaller decline in positive affect as their reports of adult children's problems increased from 0 to 8 or more. More specifically, single parents' positive affect was reduced by more than one half of a standard deviation, whereas married parents'positive affect was reduced by one fifth of a standard deviation. These results are consistent with Hypothesis 2, which predicted that parents' reports of adult children's greater number of problems would be more problematic for single parents' well-being.

Figure 1.

Figure 1

Open in a new tab

(A). Predicted Scores of Parents' Positive Affect Across Two Categories of Adult Children's Problems for Married and Single Parents. (B). Predicted Scores of Parents' Parent-Child Relationship Quality Across Two Categories of Adult Children's Problems for Married and Single Parents

An opposite pattern was found with respect to parent-child relationship quality. As Figure 1B demonstrates, the problematic association between parents' reported number of adult children's problems and ratings of parent-child relationship quality was greater among married parents. Married parents who reported that their adult children had 8 or more problems indicated levels of parent-child relationship quality nearly one standard deviation below that of married parents who reported that their adult children had none of the 10 problems assessed. Among comparable groups of single parents, however, the difference in parent-child relationship quality was less than one half of a standard deviation.

Discussion

This study provides consistent evidence that parents who report having adult children with more types of problems demonstrate poorer well-being than parents who report having adult children with fewer types of problems. In addition to replicating the finding of Pillemer and Suitor (1991) that adult children's problems are associated with parents' greater negative affect, results further indicate that adult children's greater number of problems is also problematically associated with parents' positive emotionality (positive affect), self-evaluative well-being (lower self-acceptance), and relational well-being (poorer parent-child relationship quality and more family relationship strain). Although the associations we found were modest in size, they remained statistically significant above and beyond many other factors that are associated with both adult children's problems and their parents' well-being, including psychological factors (e.g., neuroticism), sociodemographic factors (e.g., educational attainment, income), and family structural factors (e.g., intergenerational coresidence). Such findings are consistent with the family life course perspective, which conceptualizes parents and children as occupying mutually influential interlocking developmental trajectories throughout their lives (Elder et al., 2003).

Findings also suggest that the associations between parents' reports of adult children's greater number of problems and parents' poorer well-being are not uniform across all subgroups of parents. Although gender did not moderate the associations between adult children's cumulative problems and parents' psychological or relational well-being, parents' marital status (i.e., married or single) differentiated between the sizes of the associations in terms of two outcomes: positive affect and parent-child relationship quality. Given previous literature suggesting that having a marital partner can protect adults from the negative effects of distressing circumstances (Dooley et al., 2000; Reid & Hardy, 1999; Rodrigue & Park, 1996), we hypothesized that adult children's greater number of problems would more strongly jeopardize the well-being of single parents than married parents. Results provided support for this hypothesis in terms of parents' positive affect; findings indicated, however, that the moderating being married was in the opposite direction regarding parents' parent-child relationship quality. Parents' reports of adult children's greater number of problems were more problematic for married parents' than for single parents' parent-child relationship quality. Supplemental analyses (not shown) in which single parents were separated into distinct marital status categories (separated, divorced, widowed) indicated that this pattern of findings remained across each of the marital status subgroups.

Although results regarding the moderating effect of marriage on the associations between parents' reports of adult children's number of problems and parents' well-being appear contradictory, we speculate that this pattern of results is suggestive of meaningful substantive differences between the outcome domains in which they occur. Marital partners might protect individuals from the negative psychological implications of their children's problems through the various ways in which social support has been theorized to promote individuals' well-being, such as by enhancing individuals' sense of mastery or by alleviating strain resulting from stressors (Pearlin, Lieberman, Menaghan, & Mullan, 1981). Single parents, however, because they are not involved in a marital relationship, might have a greater “generational stake,” or investment in their relationship with their children (Bengston & Cutler, 1976), than married parents. It is precisely because single individuals lack a marital partner that they might be relatively more committed than married parents to maintaining high quality relationships with their children. Accordingly, perhaps single parents more actively engage in strategies to maintain positive relationships, or at least perceptions of positive relationships, with their adult children when adult children's problems place those relationships at risk. Additional theoretical and empirical work is necessary to better understand single and married parents' diverse experiences of their adult children's problems and of their linked lives more broadly.

Several features of this study's design and measures limit full interpretation of its results. First, because of this study's cross-sectional design, inferences regarding the extent to which adult children's greater number of problems causes their parents' poorer well-being are tenuous, even given the many factors that we controlled for in our analyses. Additional studies using longitudinal data are necessary to better understand the extent to which adult children's problems cause their parents' well-being to decrease over time or parents' poorer well-being causes their adult children's problems to increase over time. Moreover, because parents are asked to report on “their children” in general, it is difficult to detect the specific parent-child dynamics within any participant's family. For example, it is unknown whether parents who report their children as having more types of problems are reporting on multiple children with different types of problems or on a single child with multiple types of problems. Also, the current study's measure of relationship strain does not allow for discerning whether adult children's greater number of problems is associated with parents' heightened feelings of strain in terms of their relationships with their children or in terms of their relationships with other nonspousal family members. Additional studies with source-specific measures of family relationship strain would allow for a more precise interpretation of the associations. Finally, because missing data on the measures of substantive interest were associated with respondents' sociodemographic characteristics, our parameter estimates might be biased (Acock, 2005).

Despite these limitations, our study—which draws on data from a U.S. national probability sample and builds on previous empirical and theoretical work—provides consistent evidence that parents' reports of adult children's greater number of problems are associated with diverse aspects of parents' poorer psychological and relational well-being. Results also indicate that being married can serve as both a protective and a vulnerability factor for parents who report having adult children with a greater number of problems, depending on the domain of the outcome of interest. Future studies employing more specified measures and longitudinal designs are necessary to further examine the nature of the associations documented in this study, thereby advancing social scientific understanding of how family factors contribute to individuals' mental health and social relational well-being across adulthood.

Footnotes

Note

An earlier version of this paper was presented at the annual meeting of the National Council on Family Relations, November 17 – 20, 2004, Orlando, Florida. Support for this research was provided by grants from the National Institute on Mental Health (MH61083) and the National Institute on Aging (AG20166 & AG206983).

References

  1. Acock AC. Working with missing values. Journal of Marriage and Family. 2005;67:1012–1028. [Google Scholar]
  2. Allegreto S, Stettner A. Educated, experienced, and out of work: Long-term joblessness continues to plague the unemployed [Electronic version] 2004 Retrieved October 8, 2004 http://www.epinet.org/content.cfm/issuebriefs_ib198.
  3. Aquilino WS, Supple KR. Parent-child relations and parent's satisfaction with living arrangements when adult children live at home. Journal of Marriage and the Family. 1991;53:13–28. [Google Scholar]
  4. Bengston VL, Cutler NE. Generations and intergenerational relations. In: Binstock RH, Shanas E, editors. Handbook of aging and the social sciences. Van Nostrand Reinhold; New York: 1976. pp. 130–159. [Google Scholar]
  5. Bigatti SM, Cronan TA. An examination of the physical health, health care use, and psychological well-being of spouses of people with fibromyalgia syndrome. Health Psychology. 2002;21:157–166. [PubMed] [Google Scholar]
  6. Bradburn N. The structure of psychological well-being. Aldine; Chicago: 1969. [Google Scholar]
  7. Cancian FM, Oliker SJ. Caring and gender. Pine Forge Press; Thousand Oaks, CA: 2000. [Google Scholar]
  8. Collins WA, Russell G. Mother-child and father-child relationships in middle childhood and adolescence: A developmental analysis. Developmental Review. 1991;11:99–136. [Google Scholar]
  9. Doherty WJ, Baptiste DA., Jr. Theories emerging from family therapy. In: Boss PG, Doherty WJ, LaRossa R, Schumm WR, Steinmetz SK, editors. Sourcebook of family theories and methods: A contextual approach. Plenum Press; New York: 1993. pp. 505–524. [Google Scholar]
  10. Dooley D, Prause J, Ham-Rowbottom KA. Underemployment and depression: Longitudinal relationship. Journal of Health and Social Behavior. 2000;41:421–436. [PubMed] [Google Scholar]
  11. Elder GH, Johnson MK, Crosnoe R. The emergence and development of life course theory. In: Mortimer JT, Shanahan MJ, editors. Handbook of the life course. Kluwer Academic/Plenum; New York: 2003. pp. 3–19. [Google Scholar]
  12. Gilligan C. In a different voice: Psychological theory and women's development. Harvard University Press; Cambridge, MA: 1982. [Google Scholar]
  13. Jendrek M. Grandparents who parent their grandchildren: Effects on lifestyle. Journal of Marriage and the Family. 1993;55:609–621. [Google Scholar]
  14. Johnson W, Krueger RF. Genetic and environmental structure of adjectives describing the domains of the Big Five Model of personality: A nationwide U.S. twin study. Journal of Research in Personality. 2004;38:448–472. [Google Scholar]
  15. Kessler RC, McLeod JD. Sex-differences in vulnerability to undesirable life events. American Sociological Review. 1984;49:620–631. [Google Scholar]
  16. Keyes CLM, Shmotkin D, Ryff CD. Optimizing well-being: The empirical encounter of two traditions. Journal of Personality and Social Psychology. 2002;82:1007–1022. [PubMed] [Google Scholar]
  17. Kreider RM, Fields JM. Number, timing, and duration of marriages and divorce: Fall 1996. U.S. Census Bureau; Washington, DC: 2002. pp. 70–80. Current Population Reports. [Google Scholar]
  18. Marks NF, Choi H. Social inequalities, psychological well-being, and health: Longitudinal evidence from a U.S. national study. Research on the Sociology of Health Care. 2002;20:79–106. [Google Scholar]
  19. Maurin JT, Boyd CB. Burden of mental illness on the family: A critical review. Archives of Psychiatric Nursing. 1990;4:96–107. doi: 10.1016/0883-9417(90)90016-e. [DOI] [PubMed] [Google Scholar]
  20. McCrae RR, John OP. An introduction to the five-factor model and its applications. Journal of Personality. 1992;60:175–227. doi: 10.1111/j.1467-6494.1992.tb00970.x. [DOI] [PubMed] [Google Scholar]
  21. Mroczek DK, Kolarz CM. The effect of age on positive and negative affect: A developmental perspective on happiness. Journal of Personality and Social Psychology. 1998;75:1333–1349. doi: 10.1037//0022-3514.75.5.1333. [DOI] [PubMed] [Google Scholar]
  22. Nydegger C. The development of paternal and filial maturity. In: Pillemer K, McCartney K, editors. Parent-child relations throughout life. Erlbaum; Hillsdale, NJ: 1991. pp. 93–112. [Google Scholar]
  23. Ormel J, Rosmalen J, Farmer A. Neuroticism: A non-informative marker of vulnerability to psychopathology. Social Psychiatry and Psychiatric Epidemiology. 2004;39:906–912. doi: 10.1007/s00127-004-0873-y. [DOI] [PubMed] [Google Scholar]
  24. Pearlin LI, Aneshensel CS, Leblanc AJ. The forms and mechanisms of stress proliferation: The case of AIDS caregivers. Journal of Health and Social Behavior. 1997;38:223–236. [PubMed] [Google Scholar]
  25. Pearlin LI, Lieberman MS, Menaghan EG, Mullan JT. The stress process. Journal of Health and Social Behavior. 1981;22:337–356. [PubMed] [Google Scholar]
  26. Pham-Kanter G. Substance abuse and dependence. In: Longe JL, editor. The Gale encyclopedia of medicine. 2nd ed. Gale Group; Farmington Hills, MI: 2001. [Google Scholar]
  27. Pillemer K, Suitor JJ. Will I ever escape my children's problems: Effects of adult children's problems on elderly parents. Journal of Marriage and the Family. 1991;53:585–594. [Google Scholar]
  28. Pillemer K, Suitor JJ. Within-family differences; Paper presented at the annual meeting of the Gerontological Society of America; San Diego, CA. 2003. [Google Scholar]
  29. Reid J, Hardy M. Multiple roles and well-being among midlife women: Testing role strain and role enhancement theories. Journal of Gerontology: Social Sciences. 1999;54:S329–S338. doi: 10.1093/geronb/54b.6.s329. [DOI] [PubMed] [Google Scholar]
  30. Rodrigue JR, Park TL. General and illness-specific adjustment to cancer: Relationship to marital status and marital quality. Journal of Psychosomatic Research. 1996;40:29–36. doi: 10.1016/0022-3999(95)00540-4. [DOI] [PubMed] [Google Scholar]
  31. Rook KS. Investigating the positive and negative sides of personal relationships: Through a lens darkly? In: Spitzberg BH, Cupach WR, editors. The dark side of close relationships. Erlbaum; Mahwah, NJ: 1998. pp. 369–393. [Google Scholar]
  32. Ryff CD, Keyes CLM. The structure of psychological well-being revisited. Journal of Personality and Social Psychology. 1995;69:719–727. doi: 10.1037//0022-3514.69.4.719. [DOI] [PubMed] [Google Scholar]
  33. Ryff CD, Lee YH, Essex MJ, Schmutte PS. My children and me: Mid-life evaluations of grown children and of self. Psychology and Aging. 1994;9:195–205. doi: 10.1037//0882-7974.9.2.195. [DOI] [PubMed] [Google Scholar]
  34. Sameroff AJ, Fiese BH. Transactional regulation: The developmental ecology of early intervention. In: Shonkoff JP, Meisels SJ, editors. Handbook of early childhood intervention. 2nd ed. Cambridge University Press; New York: 2000. pp. 135–159. [Google Scholar]
  35. Schuster TL, Kessler RC, Aseltine RH. Supportive interactions, negative interactions, and depressed mood. American Journal of Community Psychology. 1990;18:423–438. doi: 10.1007/BF00938116. [DOI] [PubMed] [Google Scholar]
  36. Werner E, Smith RS. Journeys from childhood to midlife: Risk, resilience, and recovery. Cornell University Press; Ithaca, New York: 2001. [DOI] [PubMed] [Google Scholar]
  37. Winship C, Radbill L. Sampling weights and regression-analysis. Sociological Methods and Research. 1994;23:230–257. [Google Scholar]

RESOURCES