Abstract
This study examined the moderating effect of parental income on the association between parent–child coresidence and parental affect. Secondary analysis was conducted with data from the ORANJ BOWL panel, a representative sample of adults in New Jersey, aged 50 to 74 years (N = 5,688). Results indicated that income had a significant moderating effect on the association between the adult child’s residential status and parents’ positive and negative affect. Among parents with coresident adult children, an observed decline in positive affect and rise in negative affect were amplified as parental income level increased, suggesting differential strains on parental well-being across income levels.
Introduction
Over the past century, changes in employment opportunities, educational attainment, and income in the United States have enabled young adults to move out of and farther away from the parental home earlier than ever before (Ruggles, 2007). However, for several decades, the number of young adults living with their parents has steadily increased. Today, more than 30% of adults aged 18 to 34 years in the United States live in their parents’ homes, a rate last seen over 65 years ago (Fry, 2016). As a result of their coresidence with parents, young adults are now more likely to live in a multigenerational household than any other age-group, surpassing adults aged 85+ years in this category for the first time in recent history (Cohn & Passel, 2016).
Although explanations for this increase in young adults’ coresidence vary, difficulty finding or maintaining employment, as well as other economic factors, and completion of advanced degrees serve as common themes. One explanation is that increases in attainment of higher education delays full-time employment. Alternatively, economic recession may result in underemployment or unemployment for young adults, driving more young adults back to the parental home or delaying their departure from it (Sandberg-Thoma, Snyder, & Jang, 2015). Living in the same household impacts relationships between parents and their coresident adult children, especially in regard to exchanges of support from parent to child (Kim, Zarit, Birditt, & Fingerman, 2014) and emotional and mental health outcomes for the child and parent (Fingerman, Cheng, Birditt, & Zarit, 2012; Milkie, Bierman, & Schieman, 2008). Relationships between parents and adult children have previously been characterized as ambivalent—possessing the co-occurrence of positive and negative attributes (Birditt, Fingerman, & Zarit, 2010; Gilligan, Suitor, Feld, & Pillemer, 2015). Ambivalence may increase in the coresident context due to the frequency and intensity of relationship exchanges that often accompany multigenerational cohabitation (Huo, Graham, Kim, Birditt, & Fingerman, 2016; Van Gaalen & Dykstra, 2006). However, less frequently addressed in prior work is the impact of coresidence on levels of positive and negative affect of the parent and the role of moderators on the association of coresidence with ambivalent feelings. In particular, the association of a parent’s income with positive and negative affect, in the context of having a coresident adult child is not yet understood.
Previous work demonstrates that parental income is especially relevant in determining the amount and type of support given by parents to their coresident adult children, with parents who have higher incomes being more likely to provide financial as well as emotional support to their adult children than parents with lower incomes (Fingerman et al., 2015). As a result, it is likely that parental income level impacts the association between coresidence and emotional well-being of the parent. This study utilizes data from a large representative sample of older adults in New Jersey to address whether income levels of the parent moderate the impact of coresidence on parental affect.
Coresidence of Adult Children With Parents
Over more than a century, urbanization, declines in agricultural labor, the rise of industry, and changes in sociocultural attitudes and norms have all contributed to a significant decline in coresidence between parents and adult children (Ruggles, 2007). Nonetheless, economic changes, in particular, have significantly impacted more recent rises in coresidence between these two cohorts. Among less educated young adults, diminishing pay and employment opportunities have led to increases in coresidence with parents. Barriers to independent living, such as prohibitive housing costs and debt, have similarly influenced this rise in coresidence among more highly educated young adults (Fry, 2017). The current rise in coresidence between parents and their adult children runs parallel to an overall rise in multigenerational living in the United States that spans across ethnicity and age-groups (Cohn & Passel, 2016).
The Impact of Coresidence on Adult Children and Parents
While coresidence among young adults and parents is common, and may be characterized as a normative response to changing economic and social factors (Cohn & Passel, 2016; Fry, 2016, 2017), representations of coresidence in popular media have characterized the rise of young adults in the parental home as a negative experience for both parents and their children (Steketee, 2011). Such portrayals emphasize the detrimental financial consequences for parents (Fingerman et al., 2012) and the potential for parental conflict with adult children, who are often stereotyped as unmotivated, unemployed, and needy (Steketee, 2011). However, the experience of coresidence can vary widely (Burn & Szoeke, 2016). While some studies note significant drawbacks, including frequent conflicts with children, financial burdens, and negative impacts upon social and marital relationships (Burn & Szoeke, 2016), most coresident parents report satisfaction with their living arrangements (Aquilino & Supple, 1991) and coresident adult children report more daily positive enocounters with parents than negative ones (Fingerman et al., 2017). Some studies describe benefits of coresidence for parents, such as decreased loneliness and greater reciprocation of emotional and financial support, reinforcing the meanginful bonds between parents and adult children (Burn & Szoeke, 2016; Leopold, 2012). These differential findings may in part be explained by the way relationships function between coresiding adult children and parents more generally.
Relationships between adult children and their parents vary in their nature and quality, depending on factors such as the genders of the parent and child, socioeconomic status, cultural background, frequency and means of contact, and residential proximity (Fingerman, Cheng, Tighe, Birditt, & Zarit, 2012). In addition, relationship history impacts the way that parents and their adult children relate to each other, influencing the quality of interactions between generations (Fingerman, Sechrist, & Birditt, 2012). Most parents and young adult offspring view their relationships as relatively healthy (Fingerman, Cheng, Tighe, et al., 2012). Yet, in an intergenerational context, ambivalence exists in relationships between parents and adult children, meaning that both positive and negative sentiments are experienced within a given relationship (Fingerman, Cheng, Tighe, et al., 2012). Coresidence between adult children and their parents creates an opportunity for increased contact, resulting in encounters that may engender ambivalence for parents and adult children alike (Burn & Szoeke, 2016). In the coresident context, this can in part be explained as a consequence of balancing personal bonds with social pressures that may be perceived to exist upon oneself or one another (Van Gaalen & Dykstra, 2006), such as those that dictate when an adult child should achieve financial independence, or how much support is acceptable to receive from a parent.
Of particular interest, the presence of ambivalent relationship sentiments may impact the experience of both positive and negative mood of the parties coresiding. Coresidence has been linked to emotional outcomes for both parents and adult children (Fingerman, Cheng, Birditt, et al., 2012; Milkie et al., 2008), and ambivalent sentiments within a given relationship can pervade other dimensions of daily life through their impact on well-being (Fingerman, Pitzer, Lefkowitz, Birditt, & Mroczek, 2008). Limited work, however, has examined the impact of coresidence on mood specifically, to determine whether parents who live with an adult child experience different levels of positive and negative affect than parents who do not live with an adult child.
The Impact of Income on Coresidence and Affect of Adult Children and Parents
Among the most significant exchanges of support from parent to adult child is that of financial resources. However, the degree to which a parent may provide financial assistance to an adult child is limited by parents’ available financial resources, which may be spread out to assist a number of other children in need. Families with higher numbers of children are more likely to come from lower socioeconomic backgrounds, indicating that across socioeconomic groups, the financial assistance parents provide to children is not impacted by the parents’ available financial resources alone (Fingerman, Miller, Birditt, & Zarit, 2009). For parents with coresident adult children, this may be a particularly significant burden, as few children appear to contribute to household expenses when living in the parental home (Sassler, Ciambrone, & Benway, 2008).
Research by Fingerman et al. (2015) indicates that patterns of support differ among parents of higher and lower socioeconomic backgrounds. Parents of higher socioeconomic status report providing more financial support to their children, in contrast to parents of lower socioeconomic status, who report providing more total support (financial and other forms of support combined) to all their children, with each child receiving less support on the whole. In turn, socioeconomic factors may impact the likelihood of an adult child’s coresidence with parents, with more highly educated young adults being less likely to live with their parents than those with lower levels of education (Fry, 2017). The combination of higher rates of coresidence among adult children with parents of lower socioeconomic status, and the exertion of greater efforts in providing more total support to children among these parents suggests a significant potential for impacts upon the well-being for both parties.
While research is limited regarding the specific impact of income upon the well-being of parents with coresident adult children, the ambivalence of parents toward their adult children increases when they must provide support that contributes toward the independence of their children (Fingerman et al., 2012). When parents allow an adult child to reside with them, many of their financial contributions (such as paying for a child’s personal expenses) or emotional contributions (such as encouraging a child’s efforts to find employment or additional education) could be considered meaningful to a child’s independence. As such, the experience of having a coresident adult child is likely to be associated with greater feelings of ambivalence (Fingerman et al., 2015), which means levels of income may impact the effect of coresidence on the emotional experience of parents as well.
Current Study
Using a large, representative sample of adults aged 50 to 74 years in New Jersey, this study addresses a gap in the literature regarding the impact of adult child coresidence on parental levels of affect. It further examines the moderating effect of income on the association of coresidence with affect. More specifically, the aim of this paper is to address two primary research questions: (a) How does having an adult child in the home impact parental positive affect and negative affect? (b) Is the association of coresidence and parental positive and/negative affect moderated by level of parental income?
Method
Participants
Data from 5,688 people participating in the ORANJ BOWL panel (Ongoing Research on Aging in New Jersey: Bettering Opportunities for Wellness in Life) were collected between November 2006 and April 2008. Study eligibility included being between the ages of 50 and 74 years, living in New Jersey, and having the ability to participate in a 1-hour English-language telephone interview. Panel members were recruited by telephone cold calling using list-assisted random-digit-dialing (RDD) procedures. Details about sampling may be found in Pruchno, Wilson-Genderson, and Cartwright (2010). Using standard American Association for Public Opinion Research calculations, ORANJ BOWL achieved a response rate (RR5) of 58.73% and a Cooperation Rate (COOP3) of 72.88%, consistent with or better than average RDD response rates.
Comparison of characteristics of ORANJ BOWL respondents with those of all persons aged 50 to 74 years living in New Jersey reveals that they have similar racial composition, rates of being born in the state, and marital status distributions. The ORANJ BOWL sample has a slightly higher proportion of females (63.7% to 53.3%) and a slightly higher percentage of individuals with advanced secondary degrees (18.5% to 14.8%).
The analyses that follow utilized data from 4,673 ORANJ BOWL participants who reported having at least one child. Of this subsample, 1,651 were men and 3,022 were women; mean age was 60.9 years (SD = 7.2). The average participant had a 2-year college degree, with education levels ranging from those who had not completed high school (5.6%) to those with a doctoral degree (4.0%). A majority of respondents were White (82.7%); 11.7% were African American. A majority of respondents were currently married (60.9%). The mean household income was between $30,000 and $80,000 (38.9%), with 18.3% reporting income less than $30,000 and 42.8% reporting more than $80,000. Of the 4,673 participants, 1,650 reported having an adult child living in their home at the time of data collection. See Table 1 for sample demographics by coresident status.
Sample Demographic Characteristics by Coresidence Status.
Note. Numbers do not total to full sample size due to missing data.
Measures
Dependent variables
Positive and negative affect
Positive and negative affect were assessed using the Philadelphia Geriatric Center (PGC) Positive and Negative Affect Scale comprising two 5-item subscales, each assessing positive or negative affect, respectively (Lawton, Kleban, Dean, Rajagopal & Parmelee, 1992). Respondents were asked to rate the frequency of experiencing positive and negative emotions in the past week, on a 5-point Likert-type scale ranging from 1 (never) to 5 (very frequently). Items contained in the PGC Positive Affect Scale include happy, warm-hearted, interested, content, and energetic (range: 0–20; M = 14.62, SD = 3.66; α = .76). Items contained in the PGC Negative Affect Scale include irritated, sad, annoyed, worried, and depressed (range: 0–20; M = 6.57, SD = 3.57; α = .78).
Independent variables
Demographic controls
Analyses controlled for demographic characteristics with known associations to parental coresidence with adult children, including age, gender, race (Caucasian or Other), number of children, and education.
Income
Participants reported on annual income, with responses ranging from 1 (less than $15,000) to 6 (greater than $150,000). The mean income level was 3.98 (SD = 1.47).
Coresidence with adult child
Participants reported on the number of children aged 18+ years who lived in their home. Parents of coresident adult children were those participants identified as having at least one child over the age of 18 years living in their home at the time of the interview (N = 1,650). All other participants (N = 3,023) were treated as having no adult child living in their home at the time of the interview.
Analytic Plan
Descriptive statistics were used to describe the sample in terms of demographic characteristics, for the full sample as well as the subsets identified as having or not having a coresident adult child. The continuous variable of income was mean centered prior to analysis to avoid nonessential collinearity and multiplied by coresidence status to create an interaction term (Variance Inflation Factor (VIF) = 1.55, Tolerance = 0.65). Next, two parallel hierarchicial linear regressions were used to test the impact of coresidence on (a) positive and negative affect and (b) the moderating association of income on positive and negative affect. In each analysis, Step 1 controlled for demographic characteristics (Model A), Step 2 examined the main effect of coresidence and income on affect (Model B), and Step 3 included the interaction effect of income and coresidence on affect (Model C). The aim of this analytic technique was to first determine the impact of coresidence on parental affect and then to determine whether income level acted as a moderator upon the association between having a coresident adult child and levels of positive and negative affect.
Results
Table 1 describes the sample on characteristics of gender, race, age, total number of children, income, and affect (positive and negative) by coresidence status. Compared to the full sample, the participants who identified having a coresident adult child were, on average, younger (t(3,580.33) = 18.67, p < .001), had a greater total number of children (t(3,823.42) = 3.90, p < .001), reported higher income (t(3,229.61) = −12.44, p < .001), reported less positive affect (t(4,652) = 4.02, p < .001), and reported more negative affect (t(4,661) = −3.68, p < .001).
Research Question 1—Main Effect of Coresidence on Parent Affect
In addressing the first research question, which asked whether adult child coresidence had an effect upon parent positive and negative affect, we found significant associations of coresidence on affect. For positive affect, Model A first accounted for individual level characteristics of gender, race, age, and total number of children (see Table 2, Model A). Women, Caucasians, older individuals, and individuals with a greater number of children reported significantly higher levels of positive affect. Controlling for demographic associations found in Model A, Model B added the main effects for residential status of the adult child and income of the parent; both coresidence of the adult child (B = −0.47, p <.001) and parental income (B = .58, p < .001) were significantly associated with positive affect. Coresidence was associated with less positive affect while greater income was associated with greater positive affect (see Table 2, Model B).
Hierachical Regression Predicting the Moderating Effect of Income on the Association of Having an Adult Child in the Home With Positive Affect.
Note. N = 4,673 baseline.
*p < .05. **p < .01. ***p < .001.
For negative affect, we again, first accounted for individual level characteristics (see Table 3, Model A). Women, Caucasians, younger individuals, and individuals with fewer children reported significantly higher levels of negative affect. Controlling for demographic associations found in Model A, Model B tested the main effects for coresidence and income. We found significant main effects for both coresidence of an adult child (B = .41, p < .01) and income (B = −.48, p < .001). Coresidence was associated with more negative affect, while greater income was associated with less negative affect (see Table 3, Model B).
Hierachical Regression Predicting the Moderating Effect of Income on the Association of Having an Adult Child in the Home With Negative Affect.
Note. N = 4,673 baseline.
*p < .05. **p < .01. ***p < .001.
Research Question 2—Moderating Effect of Income
In addressing the second research question, which asked whether parental income had a moderating effect upon coresidence of an adult child and parental affect, we found that income had a significant moderating effect for both positive and negative affect. The final models (Tables 2 and 3, Model C) added the interaction effect for income and coresidence. A significant negative interaction for positive affect was evident (B = −0.20, p <.05) indicating a moderating association of income on the association of coresidence and positive affect. As income increased, the impact of coresidence on positive affect was amplified, such that the decline in positive affect in the context of coresidence was greater at higher levels of income (see Table 2, Model C and Figure 1). For negative affect, a significant positive interaction was evident (B = .17, p < .05), indicating a significant moderating association of income on the association of coresidence and negative affect. As income increased, the impact of coresidence on negative affect was amplified, such that the rise in negative affect in the context of coresidence was greater at higher levels of income (see Table 3, Model C and Figure 2).

Moderating effect of income on the association of coresidence and parental positive affect.

Moderating effect of income on the association of coresidence and parental negative affect.
Discussion
Our analyses found that coresidence had a significant positive impact on parents’ negative affect and a significant negative effect on parents’ positive affect. Moreover, our analyses found that parental income level was a significant moderator of the association between coresidence and parental affect. These results carry important implications for future research and furthering our understanding of the impact of coresidence of adult children on parental affective well-being.
The significant association between coresidence and affect indicates that, among parents, coresidence with at least one adult child was related to lower positive affect and higher negative affect. This finding suggests that the presence of an adult child in the home is potentially detrimental to the well-being of the parent. The higher levels of negative affect and lower positive affect may be a result of stress brought on by increased financial strain or parent–child conflict for those with a coresident adult child (Fingerman, Cheng, Wesselmann et al., 2012; Steketee, 2011). As previously noted, the experience of coresidence is likely to evoke intergenerational ambivalence for aging parents and their adult children (Fingerman, Cheng, Tighe et al., 2012; Fingerman et al., 2008; Fingerman et al., 2016). Positive and negative affect may represent indirect measures of relational ambivalence (Suitor, Gilligan, & Pillemer, 2011). Coresidence of an adult child significantly impacted both the positive and negative affect of parents in this sample, and it may also have an influence on other positive and negative parental outcomes not measured here. Our work extends prior findings by demonstrating that coresidence may be an experience that is particularly impactful to affective well-being for parents. Parents living with an adult child may need higher levels of affectitve support from others. Additional work is needed to see if affective responses are directly associated with other measures of relational ambivalence in the context of coresidence and to explore the unique impact of parental affect on other indicators of well-being and health. It may be that higher levels of negative affect, regardless of levels of positive affect experienced, carry greater impacts on other areas of functioning for parents. Additional work should also explore the dyadic nature of affective responses. For example, it may be that the affective experience of coresident adult children influences parental affective responses, as the interplay between dyad members in the context of coresidence may impact each individual’s level of affective well-being as well as other individual and relational outcomes not measured here.
Across participants both with and without coresident adult children, as income level decreased, levels of positive affect were lower and levels of negative affect were higher; however, there was also a significant moderating effect of income on the association of coresidence and parental affect. The impact of coresidence on affective well-being was amplified for parents at higher levels of income. This observation may be explained by differences in the way that parents of various levels of income ascribe meaning to their children’s coresidence, as well as the differential reasons for why adult children reside in the parental home. Prior research indicates that adult children from lower socioeconomic backgrounds, who tend to attain lower levels of education, are more likely than those from higher socioeconomic backgrounds to coreside with parents because of financial need (Fry, 2017), which may mean that the experience of having a coresident adult child is seen as more typical and acceptable for parents at lower levels of income. Parents of lower socioeconomic backgrounds have been found to provide comparable portions of their income toward the financial support of adult children as those with higher incomes, despite the fact that financial resources are more limited among those from lower socioeconomic backgrounds and that these resources are often spread across a greater number of children (Fingerman et al., 2015). With less to give financially to each child, parents of lower socioeconomic status often exhaust more effort to provide their children with other forms of support, though each child typically ends up receiving less support than among higher income families (Fingerman et al., 2015). As a result, it may be that the exertion associated with supporting both coresident and non-coresident children among parents of lower incomes results in a similar emotional burden whether one has an adult child living in the home or not. By contrast, as parental income level increases, the emotional impact of having a coresident adult child may be greater, with fewer peers in these parents’ socioeconomic group undergoing the same experience; in turn, these parents with coresident adult children may experience a greater degree of stress or frustration with their child. In addition, the stereotype that a coresident child is lazy or unmotivated may be more pervasive when parental income levels are higher (Steketee, 2011). Additional work is needed to further explore these associations and the potential mechanisms through which income impacts emotional experience in the context of coresidence.
Overall, these results are strengthened by the use of a large, representative sample of older adults reporting on income, affect, and coresidence; however, this work is not without limitation. Data were limited to information regarding parents, with no means of controlling for or identifying the age, gender, or other information of possible importance regarding adult children. It may be that specific characteristics of the adult children result in differential levels of affect for the parent within the context of coresidence. Future work should examine these characteristics of children further to determine explanatory factors for the impact of adult child coresidence on affect for aging parents. In addition, data did not provide any insight to the timeline of coresidence for adult children in the sample, which could have allowed for analysis regarding the impact of the duration of an adult child’s coresidence with a parent at the point of data collection. Additional work should explore how parental finances are impacted by the duration of a child’s coresidence, as well as how parental income may act as a determinant of the length of a child’s stay, and the reverberating effects on parental well-being. Furthermore, participants were not asked questions regarding the impact of coresidence with an adult child in particular, or regarding their relationships with coresident children, which would have been a desired area for further analysis. Future work should explore the nature of the parent–child relationship of coresident adult children and parents as it relates to household income levels. Finally, the effects found here are statistically significant yet arguably small (R2 = 0.06); additional work should explore what other characteristics that are unaccounted for explain the variance in parental positive and negative affect.
These limitations aside, this research provides insight to the experience of coresidence of an adult child with an aging parent on positive and negative affect and the impact of income level. By focusing upon positive and negative affect, this research adds new insights to the current understanding of intergenerational coresidence between parents and adult children and provides some foundational basis for future work to build upon.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Thanks to the University of Medicine and Dentistry of New Jersey-School of Osteopathic Medicine (UMDNJ-SOM) for funding that enabled establishment of the Ongoing Research on Aging in New Jersey: Bettering Opportunities for Wellness in Life research panel and collection of baseline data.
