Home
HOME ARCHIVE SEARCH TABLE OF CONTENTS

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Services
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
PubMed
Right arrow PubMed Citation
The Journals of Gerontology Series B: Psychological Sciences and Social Sciences 59:S2-S8 (2004)
© 2004 The Gerontological Society of America


RESEARCH ARTICLE

Marital Implications of Parent–Adult Child Coresidence: A Longitudinal View

Russell A. Ward and Glenna D. Spitze

Department of Sociology, University at Albany, State University of New York.

Address correspondence to Russell A. Ward, Department of Sociology, University at Albany, SUNY, Albany, NY 12222. E-mail: r.ward{at}albany.edu


    Abstract
 TOP
 Abstract
 Methods
 Results
 Discussion
 References
 
Objectives. This study assesses implications of changes in coresidence with adult children for parents' marital relations, hypothesizing that transitions into coresidence lower marital quality and transitions out of coresidence increase marital quality.

Methods. Panel data from Waves 1 and 2 of the National Survey of Families and Households are used to analyze whether change in three measures of marital quality—time together, happiness, and disagreements—is related to adult child coresidence.

Results. When adult children move out, parent couples increase their time together; there is a tendency for reduced time together when the nest "refills." However, there are no effects of moves in or out on the marital happiness of parents or the number of marital disagreements they have. There is also no effect on time together or on marital quality when one adult child moves out but another moves in during the same period. The presence of younger children has more consistent associations with marital quality.

Discussion. Coresidence with adult children does not appear to be an experience that disrupts the quality of marital relations. It may be that qualitative dimensions of coresidence experiences with adult children matter more than coresidence per se.

PARENTHOOD is stressful, particularly when children are young, and particularly for women. Research has shown that the transition to parenthood leads to lower psychological well-being and marital quality (Belsky & Kelly, 1994Go; McLanahan & Adams, 1987Go), with a rise in marital happiness after the nest empties (Van Laningham, Johnson, & Amato, 2001Go; White & Edwards, 1990Go). To what extent, and under what circumstances, do the effects of parenthood continue when children move into adulthood?

Relations with adult children later in life have inconsistent effects on well-being (Ward & Spitze, 1998Go), reflecting a mix of positive and negative dimensions of such ties (Luescher & Pillemer, 1998Go). Children's problems, however, can be a source of distress for parents (Pillemer & Suitor, 1991Go), as are the common disagreements and conflicts in parent–adult child relations (Clark, Preston, Raskin, & Bengtson, 1999Go; Umberson, 1992Go). Thus, adult children continue to affect parents, but those effects appear to be weaker than those for young children.

What about circumstances in which adult children live (coreside) with parents? One would expect differences between having young children at home and having coresident adult children. The latter, although still receiving some assistance from parents (Spitze & Ward, 1995Go; Ward, Logan, & Spitze, 1992Go), require much less care and attention than young children. They may also be positively selected, as those who stay or return to the nest may get along well with parents and help out more (Aquilino, 1991Go; Aquilino & Supple, 1991Go; Ward & Spitze, 1992Go). However, coresidence with parents in adulthood may imply failure to attain adult status or a crisis such as unemployment, and thus it may create parental stress. Further, coresidence intensifies the nature and effects of parent–child relations, as interaction and exchange are heightened (Rossi & Rossi, 1990Go; Ward et al., 1992Go; Ward & Spitze, 1992Go; White, 1994Go; White & Rogers, 1997Go).

Our research focuses on the implications of adult child coresidence for parents' marital relations in midlife. Longitudinal data allow us to assess change in parents' time together and marital quality in response to moves into and out of the nest by adult children. Prior research has addressed needs and circumstances that lead to adult child coresidence and the nature of parent–child relations in such households. We begin with an overview of those patterns and then discuss how adult child coresidence may affect parents' marital quality.

Patterns and Consequences of Coresidence
Coresidence by adult children and parents is not uncommon. Approximately 30% of parents from their 40s to their 60s coreside with an adult child, and approximately 15% of parents aged 70 or older do so (Aquilino, 1990Go; Ward et al., 1992Go); approximately 25% of children aged 22–25 coreside with a parent (Ward et al., 1992Go). Indeed, adult child coresidence has increased recently, as children are leaving later and are more likely to return to the nest (Goldscheider, Goldscheider, St. Clair, & Hodges, 1999Go; White, 1994Go). Coresidence situations primarily tend to reflect children's needs and circumstances, with adult children using the parental nest as a safety net or home base (DaVanzo & Goldscheider, 1990Go; Ward et al., 1992Go). The alternative scenario of coresidence by frail elderly parents and caregiving children represents a small proportion, even of coresiding older parents and adult children (Aquilino, 1990Go; Ward et al., 1992Go).

Research on the quality of relations between parents and coresident adult children and the effects of coresidence on well-being has relied mostly on cross-sectional data, yielding a positive but incomplete picture. Coresident parents and adult children generally report satisfaction with coresidence and positive relations (Suitor & Pillemer, 1988Go; Ward & Spitze, 1996bGo; White & Rogers, 1997Go). Research on consequences for parental well-being is more mixed. Umberson and Gove (1989)Go reported that both mothers and fathers with coresident children (of any age) have lower well-being, but Menaghan (1983)Go found no significant effect of the presence of coresident children over the age of 22 on parents' well-being. Using panel data, White and Edwards (1990)Go found that nestleaving increases parental life satisfaction if there is continued contact with the child. More generally, parents' distress has been found to be increased by adult children's problems, with no significant difference by whether the child coresides or not (Pillemer & Suitor, 1991Go).

Effects of Children on Parents' Marriages
Although much is known about how children influence their parents' well-being, less research has focused on parents' marriages. Further, that research has tended to focus on minor rather than adult children. It has been grounded mainly in two theoretical traditions. First, on the basis of exchange theory and the derived assumption that spouses receive benefits and experience costs of marital relations (Suitor & Pillemer, 1987Go), it has been suggested that the presence of children modifies the balance between costs and benefits, at least temporarily (Bradbury, Fincham, & Bench, 2000Go). Parents experience greater financial stress (White, Booth, & Edwards, 1986Go), have less time alone together, and tend toward more gender-stereotyped roles (Walzer, 1998Go). The latter is seen as pushing women into more household services and men into greater breadwinning responsibility than they might prefer (Umberson & Gove, 1989Go).

Second, researchers have discussed the impact of children on parents' well-being and marital quality from a role theory perspective (Menaghan, 1983Go; Suitor & Pillemer, 1987Go; White & Edwards, 1990Go). Children potentially increase role strain or role overload because the heavy demands of caring for them are likely to conflict with other roles or create an overwhelming total burden. This is likely to be true more for mothers, who tend to bear more of the child-care burden. The empty nest can also be viewed in terms of role change or loss.

The presence of adult children may have some of the consequences suggested by these theoretical perspectives, but to a lesser degree. Adult children may contribute to family finances or at least support their own living expenses. They may contribute to household labor, although perhaps also adding to the total workload, particularly to the mother's share (South & Spitze, 1994Go; Spitze & Ward, 1995Go). They also may limit parents' time alone but, again, not to the extent that a younger child would (Suitor & Pillemer, 1987Go; Ward & Spitze, 1992Go).

Adult children's presence, and particularly their return, may also have particular consequences that are distinctive from those of younger children. Adult child coresidence may disrupt the normative empty nest of midlife marriages. Continuing coresidence by older children may violate parental expectations and generate conflict. Umberson and Gove (1989)Go suggested that "the benefits of parenting later in the life course partly depend on one's children achieving and maintaining independence from parents" (p. 455). Concerns about children's success may influence the perceived success of the marriage and benefits derived from it. Transitions, as when children return to the nest, can also disrupt routines and relationships. Role theory suggests that any role change, including role loss, can have negative consequences (White & Edwards, 1990Go), and the empty nest is a vulnerable time in marriages (Heidemann, Suhomlinova, & O'Rand, 1998Go).

There is limited but mixed evidence on how parents' marital quality relates to adult child coresidence. Some findings are based on cross-sectional data. For example, Suitor and Pillemer (1987)Go reported no effect of adult child coresidence on marital quality for parents over the age of 65, suggesting that any effects may apply more to coresidence with young adult children. However, they do find an association between the quality of the parent–child relations with coresident adult children and parents' marital conflict. Ward and Spitze (1998)Go found that intergenerational exchanges between "sandwiched" parents in midlife and their adult children had little association with parents' marital quality, as generally high satisfaction in midlife marriages was sustained despite occasional intergenerational burdens.

Some studies have used longitudinal data to examine changes in marital quality with the transition to an empty nest. White and Edwards (1990)Go found that marital happiness increased after the nest emptied, particularly when the last child to leave was younger (indicating recency of launch and perhaps the relief from the launch of a teenager rather than an older child). Menaghan (1983)Go found increased feelings of marital equity, though there was no parallel effect on affection or fulfillment. Van Laningham and colleagues (2001)Go indicated that emptying the nest slows down an otherwise steeper decline in marital happiness over time.

Thus, research on changes in coresidence has produced mixed results on the effects of coresident adult children on parents' marriages. These studies have tended to focus on the initial period of nestleaving by very young adults, with less attention to transitions by older adult children; no study has explicitly examined the effect of a return to coresidence on parents' marital happiness. Research has also focused on subjective dimensions of marital quality, with no attention to more objective and behavioral indicators, such as time together.

Aims and Hypotheses
Our research investigates the effects of transitions in and out of adult child coresidence on parents' time together and two dimensions of their marital quality. Possible selectivity in the characteristics of parents and children who coreside complicates efforts to assess the effects of coresidence (Aquilino, 1991Go; Ward & Spitze, 1992Go). Propensity to coreside may be related to family structure, history, and, in particular, factors associated with the quality of parent–child relations that may partly account for the apparent outcomes of coresidence. In their analysis of panel data, however, White and Rogers (1997)Go concluded that there is no evidence of selection into coresidence of parents and adult children with particularly harmonious relations.

The theoretical perspectives we reviewed imply negative consequences of adult child coresidence for parents' marriages. Exchanges are likely on balance to favor the adult child rather than the parent, and the resulting imbalance could affect marital quality. Similarly, although the "role loss" implied by an emptied nest may be stressful for parents and their marriage, the role perspective also predicts negative consequences from role strain or overload that may result from adult child coresidence. Thus, although we do not view our analysis as explicitly testing the applicability of either theory, we hypothesize positive consequences of transitions out of coresidence and negative consequences of transitions back into living with adult children; the implications of stable coresidence over time are less clear.

First, we hypothesize that transitions out of coresidence with an adult child will increase parents' time together, whereas having an adult child move back into an emptied nest will have the opposite effect. As indicated earlier, the presence of young children in a household is likely to mean that parents have less time for each other, as well as increasingly differentiated roles (Walzer, 1998Go). Although coresidence with an adult child may have different and smaller effects, it is nonetheless likely to reduce parents' time alone with each other (Heidemann et al., 1998Go; South & Spitze, 1994Go; Spitze & Ward, 1995Go; Suitor & Pillemer, 1987Go).

Second, we hypothesize that transitions out of coresidence with an adult child will increase parents' marital happiness and reduce disagreements, whereas a refilled nest will have opposite consequences. Although problems that may lead to a particular child's coresidence are likely to influence parents' well-being (Pillemer & Suitor, 1991Go), the processes we are focusing on in this article relate more generally to the presence of any adult child. However, we also compare the consequences of continued coresidence with the same adult child with those of living at a later point with a different adult child. We also examine the effects of continued coresidence with the same or different adult child, in comparison with those of emptying or refilling the nest.

Third, although there is only weak evidence of gender differences in coresidence experiences, it can be hypothesized that effects of coresidence may be greater for women because of gender-based differences in the nature and impact of family roles and relations. Mothers are likely to be providing more household services to children, and, although these may decrease with the child's age, a gap between mothers and fathers is likely to remain (Spitze & Ward, 1995Go; Ward et al., 1992Go). Menaghan (1983)Go found effects of the empty nest on feelings of marital equity, and, although she did not report gender differences, it seems likely that such concerns would be more salient for wives, given related past research (e.g., Robinson & Spitze, 1992Go).


    METHODS
 TOP
 Abstract
 Methods
 Results
 Discussion
 References
 
Sample
Data are from the first and second waves of the National Survey of Families and Households, collected in 1987–1988 and 1992–1993 (Sweet & Bumpass, 1996Go). The first wave was a national probability sample of 13,017 persons aged 19 and older representing the noninstitutionalized U.S. population, with oversampling of several demographic groups (including Blacks and Hispanics) and household types; 77% of primary respondents were resurveyed (personal interviews and self-administered questionnaires) at Wave 2.

Our analyses use a subsample at Wave 2 who had an adult child (aged 19+) at Wave 1, were married to the same person at both waves, were householders (the owner or renter, to focus on the typical form of coresidence in which adult children live in their parents' household), and had data for all variables used in the analyses (as described in the paragraphs that follow). This yielded a sample of 1,365 persons, with the following characteristics: 43% men, 85% non-Hispanic White, and mean age at Wave 2 of 60.3 years (16% younger than 50, 31% in their 50s, 30% in their 60s, and 23% aged 70+).

Regarding coresidence with adult children, 28% of respondents (n = 384) coresided with an adult child at Wave 1; this number declined to 18% at Wave 2. Put in other terms, 141 (10%) coresided at both waves, 243 (18%) coresided at Wave 1 but not at Wave 2, 109 (8%) did not coreside at Wave 1 but did at Wave 2, and 872 (64%) did not coreside with an adult child at either wave. A substantial majority of coresident households (76–80%) had only one coresident adult child. In addition, 12% of respondents had a child younger than 19 in the household, another 6% had a grandchild in the household, and 92% had an adult child living outside of the household.

Variables
Marital quality
The dependent variables reflect three aspects of marital quality measured at both waves: time spent with each other, marital happiness, and marital disagreements. Time together is self-reported and therefore has subjective elements, but it refers to a more objective aspect of the relationship; we view happiness and disagreements as more subjective indicators of marital quality. To assess time together, respondents were asked, "During the past month, about how often did you and your husband or wife spend time alone with each other, talking, or sharing an activity?" (from 1 = never to 6 = almost every day). Respondents assessed overall marital happiness with a standard item: "Taking all things together, how would you describe your marriage (T1)/relationship (T2)?" (from 1 = very unhappy to 7 = very happy). Frequency of disagreements (from 1 = never to 6 = almost every day) were reported in six areas—household tasks, money, time spent together, sex, in-laws, and children—combined into a scale (range = 6–36, alpha =.76). For missing cases to be reduced, missing items for the scale were recoded to the item mean for respondents with missing data for only one or two of the six items. Because of skewness in distributions of the measures of marital quality, extreme scores were recoded to truncate the tails; comparisons with analyses using the original coding and with analyses using log transformations indicated that empirical patterns are similar across these options. We use the three measures to provide a more comprehensive assessment of the quality of marital relations that reflects both positive and negative dimensions of marital relations, as well as both objective and subjective elements (Johnson, White, Edwards, & Booth, 1986Go). The quality of these midlife marriages was generally quite positive (see Table 1).


View this table:
[in this window]
[in a new window]
 
Table 1. Descriptive Data for Variables Used in the Analyses.

 
Coresidence
We focus on patterns of stability or change in coresidence between parents and adult children. Coresidence status entails four possibilities: coresidence at both waves (with the same or different children), coresidence at Wave 1 but not at Wave 2, coresidence at Wave 2 but not at Wave 1, and not coresiding at either wave. To more clearly assess patterns of change in coresident situations, our analysis divides the sample into two subgroups: first, parents coresiding with an adult child at Wave 1, to compare those who coreside with the same adult child at Wave 2 (n = 91), coreside with a different adult child at Wave 2 (50), and do not coreside with any adult child at Wave 2 (243); and second, parents not coresiding with an adult child at Wave 1, to compare those who still do not coreside (872) with those who do coreside at Wave 2 (109). In combination, these comparisons capture the diverse patterns of stability and transitions in or out of coresidence with an adult child between the two waves.

For the most part, these situations extend past the traditional nestleaving by very young adult children, but they also reflect a variety of ages. Coresident adult children at Wave 1 had a mean age of 23.6, increasing to 26.3 at Wave 2. The mean age was 28.9 if the same adult child coresided at both waves, but it was only 22.8 if a different adult child was coresident at Wave 2. The latter appear to be primarily younger children replacing or remaining as their older siblings leave. Among parents whose households refilled (i.e., who did not coreside with an adult child at Wave 1 but did at Wave 2), the mean age of the coresident adult child at Wave 2 was 25.8, reflecting a mix of younger children aging into adulthood (40% of these coresident adult children had been adolescents living with their parents at Wave 1) and older adult children returning to the nest. In combination, these patterns allow us to look at a more diverse set of coresident households than in previous research: different transitions in and out, as well as stable coresidence by an adult child over approximately 5 years.

Control variables
Analyses included control variables likely to be related to marital quality: age, gender, and race or ethnicity (dummy variables for Blacks and for Hispanics); education (as an indicator of socioeconomic status); and self-rated health (from 1 = very poor to 5 = excellent). Missing data for health were recoded to the mean to conserve sample size. We also included characteristics of coresident children: age, gender (presence of a son), and family status (whether the household included a child or partner of the coresident adult child; of the 250 households with a coresident child, only 10 included that child's partner and only 35 included a child). Presence in the household of a young child (younger than 19) of the respondent (parent) was also included.

Analytic Strategy
We hypothesized that coresidence with an adult child would negatively affect the quality of parents' marital relations, both in terms of time together and marital happiness and disagreements. More specifically, transitions into coresidence would be related to lower marital quality, whereas transitions out would have opposite associations. It is less clear whether continuing coresidence, or a distinction between same versus different coresident adult child, would be related to change in marital patterns.

To address these issues, we focus on change in two respects. First, measures of the dependent variables at Wave 1 are controlled in regression analyses to focus on change in marital quality between Waves 1 and 2. Analyses using change scores for the dependent variables yielded equivalent results for the coresidence variables. Second, we divide the sample according to coresidence or noncoresidence with an adult child at Wave 1, using dummy variables to reflect stability or change in coresidence arrangements at Wave 2. For those coresiding with an adult child at Wave 1, coresidence with the same child at Wave 2 is the omitted category for comparison with coresidence with a different adult child and coresidence with no adult child. For those not coresiding with an adult child at Wave 1, not coresiding at Wave 2 is the omitted category for comparison with having an adult child move into coresidence at Wave 2.

We also hypothesized that coresidence with an adult child would be more strongly related to marital quality for women than men. Following the guidelines of Jaccard, Turrisi, and Wan (1990)Go, we assessed this by testing the statistical significance of change in R2 when multiplicative interaction terms for gender with coresidence status were added to the regression analyses.

Table 1 summarizes descriptive data for variables for the total sample and by gender. Patterns of adult child coresidence and characteristics of coresident children do not differ for men and women. Women report slightly lower marital happiness, but time together and disagreements do not differ significantly by gender.


    RESULTS
 TOP
 Abstract
 Methods
 Results
 Discussion
 References
 
Table 2 summarizes results of regression analyses of time together as a couple, marital happiness, and marital disagreements. Looking first at patterns of coresidence with adult children, we find that increased time together as a couple is associated with having a coresident adult child leave (and there is no statistically significant difference between continuing coresidence with same vs. different adult child). In addition, decreased time together is related to having an adult child move into the household, though this association is marginal (p <.10). These patterns support our first hypothesis. However, coresidence status is not related to changes in either marital happiness or disagreements. In particular, neither a transition out of coresidence with an adult child nor a transition into coresidence is related to these indicators of marital quality. There is also no difference between coresiding with the same adult child or a different adult child. Thus, our second hypothesis is not supported.


View this table:
[in this window]
[in a new window]
 
Table 2. Regression Analyses by Coresidence With Adult Child at T1.

 
Gender exhibits little association with the measures of marital quality. In particular, none of the gender interaction terms add significantly to the variance explained, indicating that, contrary to our third hypothesis, implications of coresidence with adult children do not differ for men and women. However, among parents coresiding with an adult child at Wave 1, women compared with men exhibit a decrease in marital happiness and an increase in marital disagreements (though the latter is marginal, p <.10); in contrast, women compared with men exhibit decreased marital disagreements among parents who did not coreside with an adult child at Wave 1. These patterns suggest that coresidence may be more problematic for women, but the associations are relatively weak and, as we noted, there are no significant gender interactions in the associations between marital quality and coresidence transitions.

Although the control variables are not the focus here, we can note some variations related to age, race or ethnicity, and health. Older respondent age tends to be associated with increases in time together and marital happiness, and with decreased marital disagreements. Better health also tends to be associated with increases in reported marital quality. Black respondents exhibit decreased time together and increased disagreements. The presence of a young child (younger than 19) in the household tends to be associated with reduced time together, lower marital happiness, and more frequent marital disagreements. However, the association between presence of a young child and marital happiness and disagreements is significant only for those without a coresident adult child at Wave 1. This suggests that the presence of an adult child may reduce the effects of a young child. The characteristics of coresident adult children are generally unrelated to marital quality, except that increased time together is associated with older child age (if coresident at Wave 1) and presence of a child or partner.


    DISCUSSION
 TOP
 Abstract
 Methods
 Results
 Discussion
 References
 
We began this article by noting that the burdens of parenthood have received particular attention for parents of young children; we asked whether those burdens continue to be evident after children move into adulthood. Coresidence by parents and adult children offers an opportunity to look at this question; if there are effects, they might be expected to be particularly evident in such circumstances, paralleling the usual situation for parents of younger children.

Panel data were used to assess the implications of coresidence with adult children, in particular transitions into and out of coresidence, for changes in marital quality. We were able to assess diverse circumstances of coresidence with adult children, moving beyond nestleaving by very young adult children to include coresidence stability and transitions by older adult children. On the basis of arguments derived from exchange theory and role theory, we hypothesized that coresidence with adult children would negatively affect both more objective and more subjective dimensions of marital quality, reducing parents' time together and leading to lower marital happiness and more frequent disagreements. We also expected stronger effects for mothers than fathers, on the basis of earlier literature that found such gender differences for younger children and because we expected mothers to provide more household services to coresident children than do fathers.

We find evidence in favor of only our first hypothesis. Changes in adult child coresidence do appear to affect parents' time together as a couple. Transitions out of coresidence are related to increased time together, though transitions into coresidence are only marginallyrelated to less time. Thus, this important aspect of marital interaction appears to be elastic with respect to the presence of coresident adult children, as could be expected when there is another member of the household. But what does this mean about the reported quality of those marriages?

In contrast to these effects on time together, we find that coresidence transitions are unrelated to marital happiness and disagreements. Thus, although these middle-aged and older parents' marriages are affected in the more objective sense of time together, the more subjective assessments of their marriages do not change. It would appear, thus, that any effects on these parents' psychological well-being that can be traced to emptying or refilling the nest do not extend to the quality of the marriage itself. This may be due to the long durations of most of the marriages represented here; perhaps they are less vulnerable than the shorter term marriages that are generally found to be affected by the addition or presence of young children. It may mean that any effects of coresidence attributable to changed exchange relations or modifications in the mix of roles are modest enough not to extend to marital quality.More specifically, it appears that the implications of an adult child in the household are less pronounced than those of a younger child. There is little evidence here or in other research that parental and spousal roles affect each other through role conflict or strain after children have reached adulthood. Parenting a young child is a more encompassing role, whereas the parental role has less depth or breadth with adult children.

We also expected that wives or mothers would be affected more than fathers by coresidence with an adult child, but this hypothesis is not supported by our analyses. We did find some main effects of gender, but these differences were small, and the overall patterns for men and women are very similar. It is important that there are no interactions between gender and changes in coresidence status in our equations for marital quality. Gender differences in parent–adult child relations have been assumed to be significant, with a special focus on the mother–daughter relation (e.g., Rossi & Rossi, 1990Go), but other recent analyses have found fewer differences than expected (e.g., Logan & Spitze, 1996Go; Umberson, Chen, House, Hopkins, & Slaten, 1996Go). The similarity in adult child coresidence experiences for men and women may reflect similar family circumstances for middle-aged mothers or fathers and young adult daughters or sons who are in coresident households (Logan & Spitze, 1996Go; Ward & Spitze, 1996aGo). Furthermore, even if there are some differences in parent–child relations in coresident households, they do not appear to extend to the experience of marital relations for either women or men.

As a unique addition to earlier literature on marital quality, we examined not only potential changes leading to an empty nest but also the effects of a return to the nest by adult children. However, despite our expectation that returns to the nest might signal adult children's problems that would affect parents' evaluation of the success of their marriage, as well as disrupt marital roles and routines, we found no effects of return coresidence beyond a marginal decrease in parents' time together. Neither transitions into nor out of coresidence are associated with marital happiness and disagreements.

Similarly, we had wondered whether it mattered more "which" adult child leaves or returns, rather than "whether" that occurs. We separated out the effects of the same child remaining or a different child entering the nest. We found no difference between these two situations. In addition, the characteristics of coresident children that we were able to incorporate into these analyses exhibited little association with marital quality. However, these are admittedly limited tests of this question. Umberson (1989)Go concluded, as have other researchers who focus on the effects of role incumbencies, that the context is crucial, and that parenting can be a source of both benefits and stress, depending on other factors. For coresidence such factors may include the reasons for coresidence, presumably reflecting problems in the child's life, the adult child's characteristics, and the quality of both past and present parent–child relations. It seems likely that, under some circumstances, returns by "problem" children would indeed be a strain on marriages or lead to more disagreements. However, this situation may be rarer than the more routine returns, which have a mix of mildly positive and negative consequences. Although we could not address this directly in the context of these analyses, we would argue that this is a promising avenue for further research.

Consistent with other research (e.g., that by Van Laningham et al., 2001Go), we find that the presence of younger (minor) children has more consistent associations with the quality of parents' marital relations than the presence of adult children. As we have noted, younger children in the household likely entail more intensive and day-to-day parenting, with greater parental effort and deflection from the spousal role. Our analyses also suggest, however, an intriguing interaction between the presence of younger and older children: We find that the presence of a coresident adult child may buffer the effects of a young child on marital quality, perhaps because the adult child assists the parents in various ways (e.g., babysitting).

An alternative explanation of the null effects for adult child coresidence and marital quality is that coresidence is selective of those with positive relationships. We noted earlier, however, that White and Rogers (1997)Go found that parents and adult children who coreside are not characterized by more positive prior parent–child relations. This is supported by other analyses using a different subset of the data used here, a subset that enabled a focus on family relations over time with particular children as they moved from adolescence into young adulthood (Ward & Spitze, 2002Go). Those analyses found that the quality of family relations during adolescence does not predict whether children subsequently coreside when they reach young adulthood.

Thus, we would argue instead that coresidence and its implications should be understood within the broader context of parent–child relations. Parents and adult children exhibit strong intergenerational ties along a number of dimensions of solidarity, including contact and proximity, affection, feelings of normative obligation, and exchanges of support (Logan & Spitze, 1996Go; Lye, 1996Go; Umberson, 1992Go). Exchanges of assistance between parents and adult children, and especially from parents to adult children, are commonplace and expected. Though only a minority of parents actively assist their adult children at any particular time, and an even smaller minority coreside, helping children is an expected part of the parental role. These normal, and normative, intergenerational exchanges appear to be managed by parents with generally little effect on parent marital quality (Ward & Spitze, 1998Go). This extends to coresidence, which most commonly occurs when children are in young adulthood, a period when support from parents to children is relatively high (Logan & Spitze, 1996Go). Rather than viewing coresidence by adult children as "counternormative" and disruptive of the lives and relationships of parents, it can be seen as a dimension of family solidarity and expected parent–child exchange patterns.


    Acknowledgments
 
Data in this paper are from the National Survey of Families and Households, funded by grants from the Center for Population Research of the National Institute of Child Health and Human Development (Grant HD21009) and the National Institute on Aging (Grant AD10266). The survey was designed and carried out by James Sweet and Larry Bumpass, Center for Demography and Ecology, University of Wisconsin–Madison. An earlier version of this article was presented at the 2001 meeting of the Gerontological Society of America.


    Footnotes
 
Decision Editor: Charles F. Longino, Jr., PhD

Received for publication February 6, 2003. Accepted for publication August 4, 2003.


    References
 TOP
 Abstract
 Methods
 Results
 Discussion
 References
 




This article has been cited by other articles:


Home page
Journals of Gerontology Series B: Psychological Sciences and Social ScienceHome page
C. M. Musil, C. B. Warner, J. A. Zauszniewski, A. B. Jeanblanc, and K. Kercher
Grandmothers, caregiving, and family functioning.
J. Gerontol. B. Psychol. Sci. Soc. Sci., March 1, 2006; 61(2): S89 - S98.
[Abstract] [Full Text] [PDF]


Home page
Research on AgingHome page
J.-H. Ha and D. Carr
The Effect of Parent-Child Geographic Proximity on Widowed Parents' Psychological Adjustment and Social Integration
Research on Aging, September 1, 2005; 27(5): 578 - 610.
[Abstract] [PDF]


This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Services
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
PubMed
Right arrow PubMed Citation


HOME ARCHIVE SEARCH TABLE OF CONTENTS