Coresident and Noncoresident Emerging Adults’ Daily Experiences With Parents

Abstract

Coresidence between emerging adults and parents is now common in the United States, but we know little about how coresidence influences daily experiences in these ties. Coresident (n = 62) and noncoresident (n = 97) emerging adults (aged 18–30) reported daily experiences with parents and mood for 7 days. During the study week, compared to offspring who lived apart from parents, coresident offspring were more likely to experience positive encounters, receive more support, wish parents would change, feel irritated, and report that their parents got on their nerves. Coresident offspring did not differ from noncoresident offspring with regard to stressful thoughts. Stressful thoughts about parents were associated with more negative daily mood; this effect did not differ for coresident and noncoresident offspring. Findings are discussed with regard to intergenerational ambivalence. In sum, coresident emerging adults were more involved with parents but not more affected by daily experiences with parents.

In the middle of the 20th century, leaving the parental home was considered a mark of adulthood (Arnett, 2007; Mitchell, 2011). Since 2008, however, rates of coresidence between emerging adults and parents in the United States have risen dramatically, with over a third of emerging adults aged 18–30 residing with their parents (Fry, 2013, 2015). Research has focused on reasons for this cohabitation including financial strains of the Great Recession, costs of housing, prolonged education, delayed age of marriage, and having children without a partner (Furstenberg, 2010; Mitchell, 2011; Newman, 2013; South & Lei, 2015). Yet, we know almost nothing about day-to-day experiences when emerging adults live with parents compared to when they live elsewhere. It is not clear whether coresidence generates conflicts and dissent between emerging adults and their parents, whether coresidence is associated with closer bonds between the grown child and parents, or whether coresidence is simply a neutral condition with regard to this tie.

In recent years, regardless of residence, global surveys have found that most emerging adults report contact with parents several times a week or more often (Arnett & Schwab, 2012), and such global surveys may undercount incidental encounters and mundane support (Fingerman, Kim, Tennant, Birditt, & Zarit, 2015; Schwarz, 2012). Nevertheless, most certainly, coresident offspring have more frequent in-person contact with the parents. For offspring who do not reside with parents, phone may be the most frequent mode of contact; 98% of emerging adults and approximately 90% of midlife adults use cell phones (Pew Research Center, 2014). Emerging adults prefer electronic technologies (e.g., text, social media) to interact with peers (Coyne, Padilla-Walker, & Howard, 2015; Lefkowitz, Vukman, & Loken, 2012). Similarly, coresident offspring might text parents regarding scheduling or household items, whereas noncoresident offspring may use electronic means to stay in touch or report daily experiences. Thus, we expected coresident offspring to have more frequent in-person contact, noncoresident offspring to have more frequent telephone contact, but we did not specify differences in coresident and noncoresident offspring’s use of electronic communications with parents.

More importantly, we expected in-person contact and phone contact to generate distinct types of experiences with parents. Social partners who coreside and interact in person throughout the day may experience more frequent positive and negative emotional experiences, more frequent support, and may be more reactive to those interactions than social partners who do not coreside (Akiyama, Antonucci, Takaha-shi, & Langfahl, 2003).

Intergenerational Ambivalence, Coresidence, and Daily Experiences

In this study, the ‘‘intergenerational ambivalence’’ model guided research hypotheses regarding coresident offspring’s daily experiences with parents (Birditt, Miller, Fingerman, & Lefkowitz, 2009; Fingerman, Pitzer, Lefkowitz, Birditt, & Mroczek, 2008; Pillemer & Suitor, 2005; Suitor, Gilligan, & Pillemer, 2011). According to this model, both structural factors and normative beliefs contribute to ambivalent (i.e., mixture of positive and negative) emotional experiences between adults and parents (Lüscher & Pillemer, 1998).

With regard to coresident offspring, increased face-to-face contact (structural factor) is likely to generate opportunities for emotional experiences that do not occur by phone or text. Further, beliefs about the meaning of coresidence also may shape these ties. Theorists argue that situations fraught with unclear norms evoke ambivalence (Pillemer & Suitor, 2005). In the United States, parents and grown children today may hold conflicting beliefs about whether the young person should reside with the parent. In the mid-20th century, leaving the parental home was a marker of adulthood (Arnett, 2000). By the late 20th century, definitions of adulthood shifted toward individual subjective experiences, such as taking responsibility for oneself (Arnett, 2000). Nevertheless, in the 21st century, Americans still endorse preferences for young adults’ autonomy from parents (Fingerman, 2016). This persistence in normative beliefs may reflect what family life-course theorists refer to as a “countertransition’’ (Elder, 1987). A countertransition is one that occurs after a normative transition, such as when a grown child has left home and returns to the home (i.e., ‘‘boomerang’’ child; Mitchell, 2011). In the context of coresidence, regardless of the reasons underlying that coresidence, the expectation that grown children should reside apart may be sufficient to generate a mix of positive and negative feelings. As such, we examined positive and negative daily experiences here with regard to structural factors and beliefs.

Positive and Negative Daily Experiences and Support

Positive encounters.

Structural differences involving in-person contact may foster positive experiences in ways that telephone contact cannot. Indeed, in Spain and Italy where intergenerational coresidence is common, parents and grown children report enjoying one another’s company when they live together (Newman, 2013) and the same may be true in the United States. Coresident offspring may experience simple daily pleasures with parents such as enjoyable everyday activities or sharing a joke (Fingerman, Kim, Zarit, & Birditt, 2016).

Stressful encounters and thoughts.

It is unclear whether coresidence generates stressful encounters. Ambivalence theory suggests that parents and grown children may be particularly sensitive to behaviors that could evoke negative or mixed feelings (Fingerman, Hay, & Birditt, 2004; Pillemer & Suitor, 2005). Via coresiding, parents may get on the offspring’s nerves or serve as a source of irritation. Structural factors involved in sharing space may generate such negative feelings. Indeed, cross-cultural research has suggested that tensions arise more easily when adults and parents live in the same household (Akiyama et al., 2003; Becker, Beyene, Newsom, & Mayen, 2003) and frequent face-to-face contact generates negative feelings (van Gaalen & Dykstra, 2010). Yet, given the role of norms in ambivalence, when coresidence is normative, there may be acceptance of the situation, and tensions may be no more likely to arise than when parties do not coreside. For example, research examining coresidence in the United States before and after the Great Recession found that deleterious effects of coresiding with offspring for parental marital quality were dampened after the Recession (when coresidence became more normative; Davis, Kim, & Fingerman, 2016). Nevertheless, people who live together are still likely to leave dirty socks on the floor or pester the other party about those socks or engage in other annoying habits; even norms accepting coresidence may not override small irritations that arise when sharing daily life.

Regarding stressful thoughts, people may worry or ruminate about problems when they lack in-person contact. Here, beliefs about coresidence may be less important than structural factors regarding in-person versus telephone contact. Limited prior research with midlife and older adults suggests that in the absence of visual facial cues, contact (i.e., by phone or text) may generate worries or concerns about problems (Teo et al., 2015). Thus, noncoresident children may experience more worries about parents on a daily basis.

Daily support from parents.

We also considered daily support. Global surveys find that emerging adults turn to their parents for advice, emotional support, and practical help (Arnett & Schwab, 2012; Bucx, van Wel, & Knijn, 2012; Fingerman, Cheng, Tighe, Birditt, & Zarit, 2012; Fingerman, Cheng, Wesselmann, et al., 2012; Swartz, Kim, Uno, Mortimer, & O’Brien, 2011). Parental financial support is pivotal during the transition to adulthood (Johnson, 2013; Remle, 2011) but may not occur on a daily basis. Thus, we focused on daily parental emotional support, advice, and practical help.

Again, structural factors and beliefs may play a role in daily support from parents. Practical support typically requires inperson contact, so that one person can assist the other with a chore (e.g., laundry, an errand, a meal); coresidence introduces opportunities for parents to provide daily practical support (Becker et al., 2003). Given increases in parental support to young adults over the past decades, however, emotional support and advice may be viewed as normative and both coresident and noncoresident offspring may seek such support.

Implications of Emerging Adults’ Daily Life Experiences With Parents

We also considered the implications of daily experiences with parents for emerging adults’ daily well-being. Almeida (2005) has argued that diary methods are particularly suited for assessing everyday experiences because these methods improve ecological validity and are not subject to memory biases. Moreover, daily stressors play an important role in overall well-being. Indeed, stressors that occur in the context of emotionally salient situations have a greater effect on daily mood than other stressors; daily tensions with close social partners in particular affect mood (Birditt, 2014; Birditt, Finger-man, & Almeida, 2005; Cichy, Stawski, & Almeida, 2014). Moreover, daily stressors are not unique in affecting daily well-being; positive interactions also have been associated with increased positive daily well-being (Ong, Bergeman, Bisconti, & Wallace, 2006). As such, a young adult’s reactions to a passerby are not likely to matter after a brief period of time, whereas a shared laugh or criticism from a parent may affect daily and even next day positive or negative mood.

Further, daily experiences with parents on mood may particularly affect grown children who reside with parents. Frequent contact may heighten the emotional impact of ties to parents (van Gaalen & Dykstra, 2010). Because coresident offspring share a living space, they may also share emotional experiences with parents, and the emotional impact may be heightened.

Other Factors Associated With Coresidence and Experiences With Parents

We considered other factors associated with coresidence or emerging adults’ daily experiences with parents such as offspring gender. Daughters typically have more frequent contact and more emotional experiences with parents than do sons (Suitor, Pillemer, & Sechrist, 2006). Gender also is associated with coresidence (men are more likely to coreside with parents; Fry, 2013). We considered offspring age because younger adults typically have greater involvement with their parents (Hartnett, Furstenberg, Fingerman, & Birditt, 2013; South & Lei, 2015). Further, student status may shape the nature of parent-child daily interactions. Students typically are in frequent contact with parents and receive more frequent support from their parents than emerging adults not enrolled in school (Fingerman, Cheng, Tighe, et al., 2012; Swartz et al., 2011).

Regarding parental factors, mothers are typically more involved with grown children than fathers (Fingerman, 2001; Rossi & Rossi, 1990). We gave particular attention to parental gender by examining daily experiences separately with mothers and with fathers in post hoc tests.

Parental socioeconomic background shapes ties, with better educated parents providing more to the average offspring than lower socioeconomic status (SES) parents (Fingerman et al., 2015; Henretta, Grundy, & Harris, 2002). Finally, a proportion of the current study identified as African American; it is not clear whether African American parents are more involved with grown children than non-Hispanic White parents (Sarkisian & Gerstel, 2004; Suitor, Sechrist, & Pillemer, 2007), but we controlled for minority status (e.g., African American in this sample).

The Current Study

Here, we examined daily encounters with parents, parental support, and mood among emerging adults aged 18–30. Coresident offspring are likely to have more frequent in-person contact with parents than noncoresident offspring, and as a result, we expected to find:

• Hypothesis 1: With regard to daily positive and negative experiences, we expected offspring who coresided with a parent to be more likely to report positive encounters (laughing, enjoying time together) and stressful encounters (irritations, get on their nerves) with that parent. Noncoresident offspring would have more stressful thoughts (worries, wish parent would change, think about problems).

• Hypothesis 2: Regarding support, we expected offspring who coresided with a parent to receive more practical support from that parent. We did not expect differences for emotional support or advice.

• Hypothesis 3: We expected daily positive or negative experiences with parents to have implications for positive and negative mood, but these effects would be greater for offspring who coreside with parents than for offspring who do not.

Method

Sample

The sample included 159 emerging adults aged 18–30 (50% female) from the Family Exchanges Study Wave 2. In 2008, the Family Exchanges Study began with middle-aged adults recruited via listed samples and random digit dialing in the Philadelphia Metropolitan Statistical Area (at that time, 93% of households with middle-aged adults had landlines). The middle-aged adults were screened to have children over the age of 18, and up to three of their children were invited to participate. We oversampled in lower SES neighborhoods and in predominantly African American neighborhoods to generate a diverse sample. A second wave of data collection occurred in 2013, and this wave of data included a daily diary study as a burst of data collection. Unfortunately, Family Exchanges Study did not include a diary study in Wave 1. In Wave 2, returning offspring were invited to participate, and in addition, we invited offspring who had turned 18 since 2008. Participants first completed a survey by telephone or the web (i.e., the ‘‘main survey’’). They then completed brief telephone interviews each day for 7 days (i.e., the ‘‘diary survey’’). Invitation to participate in the diary study was determined by random ID number.

Of 255 offspring invited to the diary study, 230 completed the diary study (90%) before enrollment ceased. The original diary sample ranged in age from 18 to 46. To focus on emerging adults’ relationships with their parents, we excluded offspring over age 30. Among the final sample (N = 159), 85% identified as White and 14% identified as African American, a few participants identified with other ethnic or racial minority groups (2 American Indian, 1 as Asian, and 1 as Hispanic). Regarding families, 96 participants were the only offspring in their family to participate, 27 families had two offspring, and 3 families had three offspring.

A majority of participants (75%) completed all 7 days (total = 1,022 days, M = 6.43 days per participant). Respondents received US$7 for each daily survey with an additional US$1 for completing all 7 days (total US$50).

We compared background characteristics of participants in this sample regarding: (a) offspring who participated in the diary (n = 230) versus those who participated in the main survey only (n = 510) and (b) among those who participated, we compared emerging adult offspring (aged 18–30, n = 159) to older offspring who were excluded (31 or older, n = 71). With regard to the first comparisons, participants in the diary were better educated (13.97 vs. 13.38 years of education, t = 3.97, P < .001) and less likely to have children of their own (26% vs. 36%, χ² = 6.77, p < .01) than participants who only completed the main survey. Comparing the older and younger diary participants, the younger participants were more likely to be students (34% vs. 9%, χ² = 16.57, p < .001) and less likely to be married (16% vs. 59%, χ² = 44.85, p < .001) or have children (9% vs. 65%, χ² = 79.78, p < .001). The younger participants were also more likely to coreside with parents (39% vs. 9%, χ² = 21.99, p < .001). Table 1 includes descriptive information regarding the sample in this study.

Table 1.

Offspring Background Characteristics.

	Coresident Offspring (n = 62)			Noncoresident Offspring (n = 97)
Characteristics	Mean	SD		Mean	SD
Age	23.15	2.92		25.72	2.71
Years of education	13.29	2.08		14.38	1.78
Household incomea	5.51	3.32		4.94	2.70
			Proportions
Female	.42			.55
Work status
Full-time	.45			.72
Part-time	.24			.13
Student status	.39			.31
Racial/ethnic minority	.16			.20
Marital status
Married	.02			.25
Cohabitating	.02			.12
Single/never married	.97			.63
Has children	.03			.12

Note. N = 159.

^a1 = less than US$10,000; 2 = US$I0,00I-US$25,000; 3 = US$25,001-US$40,000; 4 = US$40,001-US$50,000; 5 = US$50,000-US$60,000; 6 = US$60,001-US$75,000; 7 = US$75,00I-US$100,000; 8 = US$100,00I-US$I25,000; 9 = US$I25,00I-US$150,000; 10 = US$150,001-US$200,000; 11 = US$200,001-US$250,000; 12 = US$250,001 or more.

Measures

The study drew on variables from the main and diary surveys. Variables from the main survey included demographics, coresidence with parents, and relationship qualities and life problems in the past 2 years (used for sample comparisons). Variables in the diary study included contact with parents, daily emotional experiences with parents, support from parents, and mood each day. Participants answered questions about each of their parents separately. For parsimony in testing hypotheses, we used findings from both parents; we also estimated analyses separately for mothers and fathers in post hoc tests.

Background characteristics, problems, and relationship qualities.

In the main study, offspring reported their gender, age, student status, and whether they resided with parents (n = 62 residing with parents, n = 97 residing elsewhere; see Table 1). Offspring reported each parent’s gender (156 mothers, 154 fathers) and years of education. Offspring indicated whether they had experienced 10 life problems (e.g., divorce, victim of a crime) and 5 financial problems recently. Finally, offspring rated 2 items regarding positive relationship qualities with each parent (e.g., feeling loved and cared for) and negative relationship quality (e.g., parent makes demands) on a scale from 1 = not at all to 5 = a great deal. We used these variables in initial comparisons of resident and noncoresident offspring but did not include these variables in the subsequent analyses.

Mode of contact and amount of time with parents.

Each day, grown children indicated whether they had contact with the parent, and if so, whether that contact occurred in person, by telephone, or via electronic means (e.g., text, e-mail, social media). As with other questions, participants responded for their mother and father separately. Participants also indicated how much time they spent interacting with each parent: 1 = 0–5 min, 2 = 6–15 min, 3 = 16–30 min to 6 = 4 hr or more; we recoded this variable as 1 = more than 30 min contact or 0 = less than 30 min contact.

Positive and stressful experiences with parents.

Each day, participants reported whether they (a) had an enjoyable interaction or (b) shared a laugh with each parent. Participants also indicated stressful encounters, whether each parent (a) did something irritating or annoying or (b) got on their nerves. Finally, participants indicated stressful thoughts, whether they (a) worried about each parent, (b) thought about problems in their relationship, or (c) wished each parent would change their behaviors. Responses were coded 1 = yes, 0 = no (see Table 2).

Table 2.

Proportion of Offspring Reporting Daily Experiences With Parents During the Study Week.

	Proportion of Offspring Having Experience With Parents That Week		Proportion of Days Offspring Had Experience With Parents
Type of Daily Experience	Coresiding (n = 62)	Noncoresiding (n = 97)	Coresiding (n = 62)	Noncoresiding (n = 97)
Any contact	1.00	.98	.92	.69
In-person	.98	.59	.86	.27
Telephone	.74	.85	.38	.39
Text or e-mail	.61	.74	.36	.33
Interacted more than 30 min a day	.90	.64	.62	.25
Pleasant encounters	.98	.97	.80	.58
Enjoyable interaction	.98	.95	.75	.54
Share a laugh	.94	.89	.67	.39
Stressful encounters	.77	.41	.28	.13
Parent gets on nerves	.71	.32	.23	.11
Irritating or annoying interaction	.68	.37	.20	.11
Stressful thoughts	.82	.70	.51	.30
Wish parent would act differently	.68	.47	.34	.19
Worry about parent	.55	.46	.27	.15
Think about problems with parent	.44	.42	.16	.13
Parental support	1.00	.92	.70	.47
Advice	.87	.85	.52	.37
Emotional support	.86	.68	.38	.25
Practical support	.84	.55	.44	.20

Note. Offspring N = 159; Day N = 1,022.

Support from parents.

Each day, offspring reported whether each parent had provided (a) practical support (fixing something around the house, running an errand), (b) emotional support (listening to concerns or being available if they were upset), or (c) advice (information, help with a decision or suggestions about things they could do) coded 1 = yes, 0 = no.

Daily mood.

Each day, participants rated how much they experienced six positive emotions (e.g., happy, determined, calm; α = .74) and nine negative emotions (e.g., sad, lonely, nervous; α = .82) on a scale from 1 = none of the day to 5 = all of the day. These emotions were drawn from the Positive and Negative Affect Scale and from assessments of daily emotions (Birditt, 2014; Piazza, Charles, Stawski, & Almeida, 2012; Watson, Clark, & Tellegen, 1988).

Analytic Strategy

Descriptive information regarding daily contact is found in Table 2. We reported the proportion of offspring who had each type of experience with a parent during the study week as well as the proportion of days on which offspring reported those experiences. We estimated simple McNemar’s tests to examine whether certain types of experiences (e.g., pleasant encounters) were more likely to occur than other types of experiences (e.g., support) within participants each day. For example, we compared whether—on any given day—a grown child was more likely to experience a stressful thought than a stressful encounter with a parent.

Analyses also examined mode of contact as the outcome (e.g., 1 = had contact, 0 = did not have contact that day) in logistic multilevel models (Proc Glimmix in SAS), with coresidence as a predictor. We asked whether coresident grown children were more likely to have in-person contact than noncoresident offspring each day, and whether noncoresident offspring were more likely to report telephoning or texting (in each case the presence of the contact behavior was coded as 1 and not engaging in that form of contact as 0). Multilevel models take into account the nested structure of responses; offspring reported on each parent for up to 7 days. Thus, models included two parents (mother/father) nested within days and days nested within offspring. We considered an additional nesting level for offspring nested within families. Many participants were the only offspring in their family to participate (n = 96); thus, preliminary analyses revealed that the random effect for family was not significant. Therefore, for parsimony in analyses, we used a three-level model.

We then focused on whether coresidence with parents predicted daily experiences with those parents. Example hypotheses included the expectation that coresident offspring would be more likely to report positive and stressful encounters, but noncoresident offspring would be more likely to report stressful thoughts about each parent. We also expected coresident offspring to be more likely to report receiving practical support. In these logistic multilevel models, the positive and negative experiences or support served as outcomes and coresidence was again a predictor.

Below is an equation to explain this type of model. Level k represents participant offspring, level j day, and level i parent. Thus, parent (i.e., mother or father) is nested within day_j· which is nested within participant_k. Because the dependent variables are binary variables (1 or 0; e.g., whether participant_k on day_j experienced an enjoyable visit with parent_i), we transformed the binary-dependent variables into the probability of the response, using a logit link function (Guo & Zhao, 2000). Based on the logit link function (p_ijk is the probability of the response, η_ijk is the log odds of the response), it is specified as follows:

$$\begin{array}{ll}{\eta }_{ijk}=\hfill & \text{log}\left[p_{ijk}/\left(1-p_{ijk}\right)\right]={\gamma }_{000}+{\gamma }_{100}\left(coresidenc{e}_{ik}\right)\hfill \\ \hfill & +{\gamma }_{200}\left({\text{parent gender}}_{ik}\right)+{\gamma }_{300}\left({\text{parent eduction}}_{ik}\right)\hfill \\ \hfill & +{\gamma }_{001}\left({\text{offspring gender}}_k\right)+{\gamma }_{002}\left({\text{offspring age}}_k\right)\hfill \\ \hfill & +{\gamma }_{003}\left({\text{offspring student}}_k\right)+{\gamma }_{004}\left({\text{offspring minority}}_k\right)\hfill \\ \hfill & +{\nu }_{00}{}_k+u_{0jk}+e_{ijk},\hfill \end{array}$$

where γ₀₀₀ is the intercept (the expected probability of an enjoyable visit when all variables are 0). The slope γ₁₀₀ (core-sidence_ik) represents the likelihood of offspring having an enjoyable visit with coresident parent. γ₂₀₀ (parent gender_ik) and γ₃₀₀ (parent education_ik) represent parent-level covariates for the association with the likelihood of an enjoyable visit. γ₀₀₁ (offspring gender_k) to γ₀₀₄ (offspring minority_k) are offspring-level control variables. ν_00k is the error term for offspring k, u_0jk is the error term accounting for multiple days nested within participants, and e_ijk is the error term for participant k who is responding about two parents on multiple days.

Finally, we looked at whether positive and negative experiences with parents had implications for daily positive and negative mood. Because the outcome was at the participant level, we grouped experiences with parents as to whether the offspring had any (a) positive experience (laughter, enjoyable interaction), (b) stressful encounter (e.g., got irritated, parents got on nerves), (c) stressful thoughts (e.g., worry, wish parent would change, think about problems), or (d) any type of support (e.g., practical help, emotional support, advice). That is, we coded a 1 for pleasant experience if the offspring had shared a laugh or had an enjoyable interaction with either the mother or the father that day. Similarly, if the offspring experienced a worry, thought about problems, or wished a parent would change, they were coded as having a stressful thought that day. Likewise, the offspring experienced worries, thoughts about problems, and wishing parent would change was coded as having a stressful thought that day. In post hoc tests, we reran these analyses to examine associations between mood and each type of experience (laughter, irritation), and we also looked the total numbers of each type of experience. The pattern of findings was generally the same, and for parsimony, we present the category of experience (e.g., any type of support).

Because the variables for mood were continuous, we used standard multilevel models (proc mixed in SAS 9.3). We also included prior day positive and negative experiences to test for possible lagged effects of prior day experiences carrying over into next day mood. Further, we included the interaction terms for Coresidence × Positive Experiences and Coresidence × Negative Experiences to examine whether the associations between positive and negative experiences and daily mood were stronger for offspring who coresided with parents. These models included only two levels: days nested within offspring.

Given the number of analyses involved with regard to the research questions concerning contact, daily experiences, and mood (e.g., 24 models), we set a more conservative significance level of p < .05/24 = p < .002.

In all analyses, control variables included the young adult’s gender (1 = male, 0 = female), age, student status (1 = student, 0 = not a student), and minority status (1 = ethnic/racial minority, 0 = non-Hispanic White). We included parent gender (1 = father, 0 = mother) and years of education in predicting the experiences each day (but not in predicting daily mood). Parents’ ages and minority status were too highly correlated with offspring’s to consider in the same analyses. Post hoc tests were estimated to assure stability of findings for mothers and fathers separately and for different types of daily experiences and daily mood.

Results

Coresident Versus Noncoresident Offspring

Among the 62 coresident offspring, most resided with married parents, but 7 lived with only one parent: 2 resided with a divorced father and 2 with a divorced mother, 1 with a never married father, and 2 with widowed mothers. We first compared coresident offspring to noncoresident offspring in the sample using t-tests and χ² tests. Coresident offspring were younger (M = 23.15 years) than noncoresident offspring (M = 25.72 years; t = 5.67, p < .001) and were more likely to report difficulties finding a job (49%) compared to noncoresident offspring (31%; χ ² = 5.08,p = .029) but did not differ with regard to gender, student status, life problems (e.g., divorce, victim of crime), losing a job, loss of money, cutting back on expenses due to financial problems, or on positive or negative relationship quality with the mother or the father.

Daily Experiences With Parents

Table 2 includes the frequency of daily experiences with parents throughout the study week. All but two participants had contact with a parent during the study week. These two noncoresident offspring were dropped from analyses pertaining to contact (e.g., phone, pleasant encounters; n = 2 offspring, n = 13 days) but were included in analyses pertaining to thinking about parents (e.g., worrying, wishing they would change) and daily mood.

Regarding daily experiences with parents, most participants reported an enjoyable interaction or sharing a laugh during the study week. Approximately two thirds of coresident offspring and a third of noncoresident offspring had an irritating encounter with parents or the parents got on their nerves, but these stressful encounters occurred on only 1 or 2 days of the study on average. Approximately half of participants had stressful thoughts about parents during the study week (e.g., worrying about parent, their problems, or wishing the parent would change). Most coresident and noncoresident offspring received advice and emotional support during the study week. Most coresident participants and over half of noncoresident offspring also received practical support. Yet, support generally occurred on fewer than half the study days.

We also asked whether there were differences in the likelihood of experiencing each type of daily experience. To examine this matter descriptively, we estimated nonparametric McNemar’s tests separately for the offspring who coresided with parents and for offspring who did not coreside with parents. McNemar’s test allows comparisons of 2 × 2 tables (yes/no vs. yes/no responses). For each day of the survey, we compared yes/no reports of pleasant experiences, parental support, stressful thoughts, and stressful encounters in rank order; thus, if pleasant experiences differed from parental support, pleasant experiences also differed from stressful thoughts and encounters. Both groups—coresident offspring and noncoresident offspring—were more likely to report a pleasant encounter than support (McNemar’s exact p < .001), more likely to report receiving some type of parental support than having a stressful thought (McNemar’s exact p < .001), and more likely to have a stressful thought about a parent than a stressful encounter with a parent (McNemar’s exact p < .001) on any given day of the study. We also estimated the McNemar’s tests for daily experiences separately for experiences with mothers and fathers, and the same pattern was evident. In other words, positive experiences with parents (i.e., pleasant and supportive encounters) were more likely to occur on a given day than negative experiences (i.e., stressful thoughts or encounters).

Coresidence and Daily Experiences

Coresidence and contact.

We expected coresident offspring to have more frequent in-person contact and noncoresident offspring to have more frequent phone contact, with no difference for texting. We estimated logistic multilevel models to examine whether coresidence was associated with daily contact in general and with specific modes of contact (e.g., in-person, telephone). As can be seen in Table 3, as expected, coresident offspring were more likely to have in-person contact (odds ratio = 25.94) and offspring who resided with parents were nearly 12 times as likely to spend at least 30 min in contact on a given day than offspring who did not coreside with parents (who had either not had contact or spend less than 30 min in contact on a given day). Coresidence status was not significantly associated with telephone contact or texting (not shown in tables).

Table 3.

Logistic Multilevel Models Predicting Offspring’s Daily Contact With Parents From Coresidence.

	In-Person Contact			More Than 30 Minutes Contact
Variables	B	SE	OR	B	SE	OR
Fixed effects
Intercept	−0.65	1.54		−0.22	1.54
Coresidinga	3.26***	0.26	25.94	2.46***	0.26	11.74
Covariates
Parent variables
Genderb	−0.53***	0.13	0.59	−0.75***	0.13	0.47
Education	0.08	0.05	1.08	0.04	0.05	1.04
Offspring variables
Genderb	−0.42	0.25	0.66	−0.95***	0.26	0.39
Age	−0.08	0.05	0.92	−0.07	0.05	0.93
Student statusc	0.45	0.29	1.57	0.80**	0.29	2.23
Minority statusd	0.33	0.33	1.40	0.34	0.33	1.40
Random effects
Intercept VAR (Level 2: Day)	1.85***	0.36		1.89***	0.37
Intercept VAR (Level 3: Offspring)	1.20***	0.21		1.28***	0.22
−2 (pseudo) log likelihood		9,603.4			9,602.5

Note. Offspring N = 159; Day N = 1,022 for any contact. Offspring n = 157; Day n = 1,009 for in-person contact and more than 30 min; two participants dropped for no contact with parents during study week. Outcomes were coded: 1 = in-person contact, 0 = no in-person contact; 1 = more than 30 min contact, 0 = 30 or fewer minutes contact. OR = odds ratio; VAR = variance.

^a0 = not coresiding with parent, 1 = coresiding with parent.

^b0 = female, 1 = male.

^c0 = nonstudent, 1 = student.

^d0 = non-Hispanic White, 1 = racial minority.

^*p < .05.

^**p <.01.

^***p <.001.

Regarding control variables, using the more conservative p < .002 significance level, participants were more likely to have contact with mothers than with fathers. Daughters were more likely to spend more than 30 min with parents on a given day.

Positive and negative experiences, support from parents.

As expected, coresident offspring were more likely to report each type of positive encounter (laughter, enjoyable interaction; Table 4). With regard to stressful experiences, we expected coresident offspring to be more likely to report stressful encounters with parents (irritations, getting on nerves). But we expected noncoresident offspring to experience more frequent stressful thoughts (worry, thinking about problems, wish parent would change). Findings partially supported hypotheses (see Table 5 for significant findings). Coresident offspring were more likely to report parents got on their nerves and irritated them each day, and coresident offspring also were more likely to wish parents would change. Coresidence was not significantly associated with reports of worries or thinking about problems with parents, however (not shown in tables).

Table 4.

Logistic Multilevel Models Predicting Offspring’s Daily Positive Encounters With Parents From Coresidence.

	Enjoyable Interactions			Shared Laugh
Variables	B	SE	OR	B	SE	OR
Fixed effects
Intercept	− 1.06	1.44		−0.34	1.41
Coresidinga	1.37***	0.25	3.93	1.31***	0.24	3.72
Covariates
Parent variables
Genderb	−1.05***	0.12	0.35	−0.90***	0.12	0.41
Years of education	0.13**	0.05	1.14	0.09	0.05	1.09
Offspring variables
Genderb	−0.70**	0.24	0.50	−0.74**	0.23	0.48
Age	−0.03	0.05	0.97	−0.07	0.05	0.94
Student statusc	0.59*	0.27	1.80	0.40	0.26	1.50
Minority statusd	−0.56	0.31	0.57	0.36	0.30	1.44
Random effects
Intercept VAR (Level 2: Day)	2.45***	0.53		2.26***	0.48
Intercept VAR (Level 3: Offspring)	1.14***	0.24		1.04***	0.23
−2 (pseudo) log likelihood		9,339.9			9,348.1

Note. Offspring n = 157; Day n = 1,009; two participants dropped for no contact with parents during study week. Outcomes were coded: 1 = any enjoyable visit, 0 = no enjoyable visit; 1 = shared laugh, 0 = did not share laugh. OR = odds ratio; VAR = variance.

^a0 = not coresiding with parent, 1 = coresiding with parent.

^b0 = female, 1 = male.

^c0 = nonstudent, 1 = student.

^d0 = non-Hispanic White, 1 = racial minority.

^*p < .05.

^**p < .01.

^***p <.001.

Table 5.

Logistic Multilevel Models Predicting Offspring’s Daily Stressful Experiences With Parents From Coresidence.

	Irritating Interactions			Got On Nerves			Wished Parent Would Change
Variables	B	SE	OR	B	SE	OR	B	SE	OR
Fixed effects
Intercept	−3.00*	1.30		−2.28	1.33		−2.76	1.40
Coresidinga	0.88***	0.22	2.41	1.06***	0.23	2.88	0.86***	0.25	2.36
Covariates
Parent variables
Genderb	−0 77***	0.18	0.46	−0.68***	0.17	0.51	0.29*	0.14	1.34
Years of education	−0.04	0.05	0.96	−0.02	0.05	0.98	−0.14**	0.05	0.87
Offspring variables
Genderb	−0.30	0.21	0.74	−0.14	0.21	0.87	−0.14	0.23	0.87
Age	0.04	0.04	1.04	−0.01	0.04	0.99	0.09	0.05	1.09
Student statusc	0.41	0.23	1.51	0.58*	0.24	1.79	0.41	0.26	1.50
Minority statusd	0.46	0.25	1.59	0.64*	0.26	1.91	0.77**	0.29	2.16
Random effects
Intercept VAR (Level 2: Day)	0.73	0.39		0.83*	0.36		1.07*	0.42
Intercept VAR (Level 3: Offspring)	0.22*	0.11		0.32**	0.11		0.85***	0.16
−2 (pseudo) log likelihood		10,368.5			10,271.9			9,850.7

Note. Offspring n = 157; Day n = 1,009 for stressful encounters (irritating interactions and got on nerves). Offspring N = 159; Day N = 1,022 for wished parent would change. Outcomes were coded: 1 = any irritating interaction, 0 = no irritating interactions; 1 = got on nerves, 0 = did not get on nerves; 1 = wished parent would change, 0 = did not endorse wishing parent would change. OR = odds ratio; VAR = variance.

^a0 = not coresiding with parent, 1 = coresiding with parent.

^b0 = female, 1 = male.

^c0 = nonstudent, 1 = student.

^d0 = non-Hispanic White, 1 = racial minority.

^*p < .05.

^**p <.01.

^***p <.001.

Furthermore, coresident offspring were significantly more likely to experience all types of support: emotional and practical support p < .001 and advice p = .002 (Table 6), though we had initially only predicted such a difference for practical support.

Table 6.

Logistic Multilevel Models Predicting Parental Daily Support From Coresidence.

Variables	Emotional support			Practical support			Advice
	B	SE	OR	B	SE	OR	B	SE	OR
Fixed effects
Intercept	−2.27	1.31		−2.56	1.37		−1.62	1.33
Coresidinga	0.94***	0.23	2.57	1.34***	0.24	3.82	0.72**	0.24	2.06
Covariates
Parent variables
Genderb	0.97***	0.14	0.38	−0.38**	0.13	0.69	−0.83***	0.12	0.44
Years of education	0.09	0.05	1.09	0.07	0.05	1.07	0.10*	0.05	1.11
Offspring variables
Genderb	−0 79***	0.21	0.45	−0.32	0.23	0.73	−0.35	0.22	0.70
Age	−0.02	0.04	0.98	−0.03	0.04	0.98	−0.03	0.04	0.97
Student statusc	0.60*	0.24	1.83	0.54*	0.26	1.72	0.27	0.25	1.31
Minority statusd	0.17	0.28	1.19	0.03	0.30	1.03	−0.24	0.30	0.79
Random effects
Intercept VAR (Level 2: Day)	1 91***	0.36		1.56***	0.37		2 22***	0.47
Intercept VAR (Level 3: Offspring)	0.61***	0.16		0.80***	0.17		0.86***	0.21
−2 (pseudo) log likelihood	9,590.2		9,625.8		9,350.5

Note. Offspring n = 157; Day n = 1,009; two participants dropped for no contact with parents during study week. Outcomes were coded: 1 = any emotional support, 0 = no emotional support; 1 = any practical support, 0 = no practical support; 1 = any advice, 0 = no advice. OR = odds ratio; VAR = variance.

^a0 = not coresiding with parent, 1 = coresiding with parent.

^b0 = female, 1 = male.

^c0 = nonstudent, 1 = student.

^d0 = non-Hispanic White, 1 = racial minority.

^*p < .05.

^**p <.01.

^***p <.001.

Experiences With Parents and Daily Mood

We expected positive and negative experiences with parents to be associated with daily mood; and moreover, we expected these associations to be stronger for coresident than noncoresident offspring. Initial analyses examined whether coresident and noncoresident offspring differed with regard to positive or negative mood; they did not. We estimated separate models for daily positive and negative mood including the interaction term for Coresidence × Each Type of Daily Experience. The interactions terms were not significant in any models; therefore, we present only main effects of coresidence and daily experiences here.

With regard to negative mood, as can be seen in Table 7, stressful thoughts about parents were associated with more negative mood (controlling for prior day’s negative mood). No other significant associations were found for daily positive or negative mood. That is, pleasant encounters with parents, stressful encounters, and receiving daily support from parents were not significantly associated with positive or negative mood at p < .002 (not shown in tables).

Table 7.

Multilevel Models Predicting Offspring Negative Mood From Stressful Thoughts About Parents.

Variables	B	SE
Fixed effects
Intercept	1.00***	0.19
Stressful thoughtsa	0.13***	0.03
Coresidingb	−0.06	0.04
Covariates
Offspring genderc	−0.00	0.03
Offspring age	−0.01	0.01
Offspring student statusd	−0.03	0.04
Offspring minority statuse	−0.06	0.04
Prior day negative mood	0.39***	0.03
Random effects
Intercept VAR	0.03**	0.01
Residual VAR	0.09***	0.01
−2 log-likelihood	582.8

Note. Offspring N = 159; Day N = 1,022. VAR = variance.

^a0 = no worrying, thinking about relationship problems or wishing parents would change, 1 = any worrying, thinking about relationship problems or wishing parents would change.

^b0 = not coresiding with any parents, 1 = coresiding with either parent.

^c0 = female, 1 = male.

^d0 = nonstudent, 1 = student.

^e0 = non-Hispanic White, 1 = racial minority.

^*p < .05.

^**p <.01.

^***p <.001.

Post Hoc Tests

We conducted analyses to assess stability of findings. First, we asked whether findings were the same regarding mothers and fathers. We estimated logistic multilevel models with regard to each type of contact with parent gender coded 1 = father, 0 = mother and including the control variables (but not coresidence). Mothers were more likely to report telephone contact (B = −1.00, p < .001) and any type of contact (B = −1.44, P < .001) with offspring than were fathers. There were no parent gender differences for in-person contact.

We also ran the models for coresidence separately for mothers and fathers using the conservative p < .002 for significance. For both mothers and fathers, the pattern of findings regarding contact was consistent with Table 3. For the models in Tables 4 through 6, however, coresidence did not show an effect for mothers for enjoyable interactions, emotional Parents From Coresidence. support or advice, or with regard to getting on the child’s nerves or irritating the child. In fact, coresidence was only significant for mothers at p < .002 with regard to laughter and practical support. Effects of daily experiences on daily mood did not differ with regard to mothers or fathers.

We also considered whether in-person contact accounts for the observed differences between coresident and noncoresident offspring. We reran models for daily experiences (e.g., Tables 4–6) including only offspring who had in-person contact; on average, 72 offspring had in-person contact with a parent each day. The only significant effect for coresidence in these models involved sharing a laugh.

We reran analyses for daily mood looking at the contribution of each specific type of experience. In these models,thinking about relationship problems with a parent and wishing a parent would change remained significantly associated with more negative mood at p < .002, worrying about a parent was also associated with more negative mood, but only at p < .01.

Discussion

Coresidence between emerging adults and parents in the United States has increased over the past decade. As of 2016, 18- to 34-year-olds in the United States were more likely to live with their parents than with a romantic partner (Fry, 2016). Yet, norms regarding coresidence have not kept pace with this shift. Many media accounts disparage coresidence between generations as fraught with perils and tensions (Ascher, 2015; Collegecandy, 2014; Hoover, 2016). In actuality, coresidence per se does not fundamentally undermine grown children’s ties to parents. Coresidence involves more positives (pleasant encounters, support) than negatives (stressful thoughts or encounters) on a daily basis—but both types of experiences increase with more frequent in-person contact. The probability of experiencing a positive encounter on a given day exceeded the probability of experiencing an irritation or annoyance among coresident and noncoresident offspring alike, suggesting coresidence certainly does not undermine the relationship.

It is unlikely that characteristics of the offspring (such as personality) or the parent-child relationship (such as positive relationship quality) fully account for differences between coresident and noncoresident offspring. Coresidence is not a stable attribute of the child or the relationship—offspring move in and out of the parental home (Davis & Fingerman, 2016; Mitchell, 2011; South & Lei, 2015). Thus, at any given point in time, a particular offspring might coreside or not.

Moreover, although a grown child living at home is not the same as a grown child living in another city in her own apartment, coresidence does not appear to harm relationships with parents. Rather, grown children who coreside simply have more opportunities for positive interactions, parental support, and stressful encounters.

We expected coresidence to amplify effects of daily experiences with parents; that is, experiences with parents would influence well-being more when parties coresided. Yet, for the most part, experiences with parents were not associated with daily mood. Only stressful thoughts about parents were associated with negative mood, but coresidence did not magnify this association. Thus, intergenerational coresidence is associated with daily positive, stressful, and supportive encounters with parents, but these experiences do not disproportionately influence daily mood.

Coresidence, Ambivalent Daily Experiences, and Support

We used the intergenerational ambivalence model to examine daily experiences between young adults and their parents. We had expected grown children who reside with their parents to report positive and negative daily experiences with their parents, and to some extent, findings supported this expectation.

Clearly, people who live together share a variety of experiences throughout the day; this situation is not different for roommates, romantic partners, spouses, or young children and parents. In this study, coresidence increased the likelihood of all types of daily experiences for young adults with their parents: positive encounters, stressful encounters, and support.

Yet, it is not clear these experiences fully reflect interge-nerational ambivalence. Rather, positive experiences outweighed the negative ones. Coresident offspring were more likely to have enjoyable encounters with parents and to laugh and to receive all types of support than noncoresident offspring. Findings are consistent with research in Europe; in countries where coresidence is normative (e.g., Italy and Spain), grown children report that they enjoy living with their parents (Newman, 2013).

Nevertheless, although negative encounters were infrequent, when grown children coresided with parents, they were more likely to report that their parents got on their nerves and generated irritations than were offspring who did not coreside with their parents. Ambivalence theory still helps explain why coresident offspring were more likely to have negative encounters with parents. The theory argues that ambivalence arises due to structural factors and unclear norms (Lüscher & Pillemer, 1998). When parties live together, structural factors such as the need to regulate a shared space may contribute to interpersonal tension, regardless of relationship type (e.g., romantic partner, roommate; Akiyama et al., 2003). Ambivalence theory also predicts that unclear norms may play a role in negative encounters. With regard to adult offspring living at home, parents may interfere or pester based on their prior normative behavior in their role as parents. Indeed, Padilla-Walker, Nelson, and Knapp (2014) observed differences in parents’ and college students’ perceptions of parental authority, and similar differences may feed into day-to-day stressful interactions when emerging adults coreside with parents.

Although coresidence may be viewed as a type of support from parents, it was also associated with a variety of other forms of support. It was not surprising that practical support occurred more often when parties coreside, but coresident offspring also received more emotional support and advice, forms of help that could be provided at a distance. As mentioned, ambivalence theory suggests that adult offspring and parents experience discomfort when these offspring are dependent on the parent (Pillemer & Suitor, 2005). When offspring coreside with parents, however, feelings of dependence may become more normative. Offspring may be more comfortable turning to parents, and parents may find providing such support routine and easy.

In-Person Contact and Parental Gender

To fully understand associations between coresidence and daily experiences, we need to consider other factors such as in-person contact and the gender of the parent. In-person contact accounted for most of the observed differences in positive and negative daily experiences associated with coresidence, such that noncoresident offspring who see their parents in person on a given day were just as likely to experience positive and stressful encounters, emotional support, and advice. Thus, it may not be the coresident status per se that generates ambivalence, but rather that in-person contact offers opportunities for both positive and stressful encounters with parents.

The one area where coresidence matters (regardless of inperson contact)—emerging adults who resided with parents were more likely to share laughter. Perhaps living together provides fodder for teasing or jokes. Researchers also have found strong emotional closeness between grown children and parents who coreside (Mitchell, 2011; South & Lei, 2015). Perhaps people laugh more when they have closer relationships. Regardless, it is important to note that inperson contact can account for all the features of coresidence, except the humor.

Notably, as well, effects of coresidence are particularly important in ties to fathers rather than ties to mothers. That is, when we analyzed the data separately for mothers and for fathers, the associations between coresidence and experiences were only significant for fathers. On a daily basis, mothers communicated with grown children via telephone and text regardless of where the offspring lived. For fathers, coresidence may provide opportunities for increased contact and thus, fodder for a variety of daily experiences that do not occur when fathers and grown children reside in separate households. Nevertheless, associations of coresidence and daily experiences for mothers were in the same direction as those observed for fathers, and the sample size was relatively small; in a larger sample, coresidence might have reached significance for mothers as well.

Offspring Daily Mood and Suggestions for Future Research

In prior research, stressful daily experiences have been associated with increased negative mood at the end of the day (Birditt, 2014; Birditt et al., 2005; Cichy et al., 2014). This study also revealed associations between stressful thoughts about parents and end-of-day negative mood. These effects did not differ for coresident and noncoresident offspring. Thus, the general pattern for emerging adults with their parents is similar to that observed in romantic partnerships and other relationships assessed on a daily basis.

This study has several limitations. The sample was relatively small and did not permit differentiation of distinct patterns of parenting observed in prior studies of emerging adults (Nelson, Padilla-Walker, Christensen, Evans, & Carroll, 2011). For the offspring who were coresiding with parents, we do not know whether these offspring have remained in the nest since childhood (i.e., failure to launch) or returned home after living away from parents (i.e., boomerang child). Mitchell (2011) noted that reasons underlying returning to the nest may differ from reasons for remaining in the home, and qualities of ties to the parents also may differ.

This diary study only surveyed participants across 7 days, whereas the modal period of time for diary studies is 2 weeks (Gunthert & Wenze, 2012). Scholars note a trade-off, however, in the burden of a diary study versus diversity of the sample. Moreover, longer periods of diary data collection are required to capture events that happen rarely (Gunthert & Wenz, 2012), but in this study, contact between emerging adults and parents occurred a majority of study days.

We also did not fully capture technology-mediated contact. Young adults continually adopt new media technologies (Coyne et al., 2015). For offspring who are not proximate to parents, Skype, FaceTime, or videoconferencing technologies might compensate for in-person contact.

Finally, future research might seek to better understand how coresidence affects offspring’s long-term well-being. Although coresidence was not associated with daily mood, it may have repercussions for longer term adjustment, psychological or financial well-being.

In sum, today emerging adults in the United States are more likely to coreside with parents than with a romantic partner. This coresidence is still viewed as nonnormative, however, and grown children may be uncomfortable about living with parents. Yet, intergenerational coresidence does not undermine the grown children’s ties to parents or their daily mood. Of course, emerging adults who live with parents have more frequent in-person contact with them. Thus, living with parents provides opportunities for positive experiences like laughter and support. Coresidence also provides opportunities for negative experiences with parents (such as getting on the young person’s nerves), but these experiences are less frequent than the positive ones. In sum, daily experiences of coresidence involve greater likelihood of positive and negative experiences, but overall, these relationships are not necessarily better or worse than when young people and parents live in separate households.

Author Biographies

Karen L. Fingerman is a professor of human development and family sciences and the director of the Aging Network at University of Texas at Austin. Her research addresses social and emotional aspects of adult development. She is the principal investigator on the Family Exchanges Study which has focused on young adults, their midlife parents, and their aging grandparents.

Meng Huo is a PhD student in human development and family sciences at the University of Texas at Austin. Her research interests include social relationships and daily social interactions across adulthood, and how these relationships and interactions may influence adults’ well-being. She is especially interested in how empathy plays a role in younger and older adults’ social lives.

Kyungmin Kim is an assistant professor of gerontology at University of Massachusetts-Boston. Her research focuses on dynamic aspects of adult child-parent ties across various group levels (dyad, family, and culture/society) and time levels (daily and short-/long-term interval).

Kira S. Birditt is a research associate professor in the Life Course Development Program at the University of Michigan. Her research examines the circumstances under which positive and negative aspects of relationships are associated with physical and psychological well-being.