|Home | About | Journals | Submit | Contact Us | Français|
Daily pleasant or stressful experiences with grown children may contribute to parental well-being. This diary study focused on midlife parents’ (N = 247) reports regarding grown children for 7 days. Nearly all parents (96%) had contact with a child that week via phone, text, or in person. Nearly all parents shared laughter or enjoyable interactions with grown children during the study week. More than half of parents experienced stressful encounters (e.g., child got on nerves) or stressful thoughts about grown children (e.g., worrying, fretting about a problem). Pleasant and stressful experiences with grown children were associated with parents’ positive and negative daily moods. A pleasant experience with a grown child the same day as a stressful experience mitigated effects of those stressful experiences on negative mood, however. The findings have implications for understanding intergenerational ambivalence and stress buffering in this tie.
Parents are often in touch with their grown children; more than half of midlife parents report contact with a grown child every day, and 75% to 90% report contact at least once a week (Arnett & Schwab, 2013; Fingerman, Cheng, Tighe, Birditt, & Zarit, 2012). Indeed, rates of contact between adults and parents have increased over the past 25 years, in part because of cellular phone, e-mail and other technologies that render communication convenient and low cost (Cotten, McCullough, & Adams, 2012; Fingerman, Cheng, Tighe, et al., 2012). The implication of such frequent contact warrants consideration. In childhood, daily interactions with children affect parents’ mood and well-being (Larson & Richards, 1994; Repetti, Wang, & Saxbe, 2011), and the same may be true in adulthood. Despite frequent contact and the potential impact of grown children on parental well-being, however, few studies have examined daily experiences in this tie.
In this study we examined the modalities parents use to be in touch with grown children on a daily basis (e.g., phone, text, in person). We also examined the emotional valence of daily experiences (i.e., pleasant, stressful). We then asked whether positive and negative relationship qualities with each grown child are associated with different modes of contact and daily emotional experiences. Finally, we considered parental daily mood as a function of daily experiences with grown children.
Although extensive research has documented frequent contact between adults and their parents, little is known about how this contact occurs. In this study we examined contact in person, by phone, and by text messaging or e-mail. On the basis of pervasiveness of use (Cotten et al., 2012), we expected parents to be most likely to report telephone contact with their grown children. Texting is equally convenient, but it may be less familiar to some parents ages 45 to 65. We were interested in what modes of contact parents and grown children use and how those modes of contact may be associated with the emotional valence of parents’ daily experiences with grown children.
With regard to emotional valence, minor stressful or pleasant experiences in everyday life may accumulate to influence well-being. Indeed, daily stressors and supportive exchanges in marriage have been associated with daily mood and general well-being (Almeida, Wethington, & Kessler, 2002; Birditt, 2014; Gleason, Iida, Shrout, & Bolger, 2008); the same may be true in intergenerational ties. We considered pleasant encounters, stressful encounters, and stressful thoughts involving grown children each day.
In studies relying on global reports, parents have described frequent pleasant encounters with grown children (Fingerman, 2000). Thus, we expected parents to report pleasant encounters often when asked on a daily basis. Pleasant daily encounters may occur even when the parties do not see each other in person. For example, parents might exchange jokes with a grown child via e-mail or text. The family science literature is surprisingly silent on laughter, yet popular psychology suggests that ties to grown children may be punctuated with humor (Covey, 1997), and we examined laughter in this study.
We expected parents also to report stressful experiences with grown children throughout the study week, though less often than pleasant experiences. Prior research has found that parents report their grown children sometimes irritate them, that grown children get on their nerves, or that parents may regret how they behaved toward a grown child (Birditt, Miller, Fingerman, & Lefkowitz, 2009; Fingerman, 2001; Levitzki, 2009).
Cognitive behavioral theories suggest that individuals’ thoughts about events and other people also influence their mood. Throughout childhood, parents invest energy thinking about their children’s problems, and old habits may persist after children are grown. We expected parents to report stressful thoughts about grown children—that they worry about their children or fret about problems with a child (Hay, Fingerman, & Lefkowitz, 2007; Levitzki, 2009).
We also considered a mixture of experiences: pleasant encounters and stressful encounters with, or stressful thoughts about, grown children on the same day. A vast literature has documented ambivalence in intergenerational ties, suggesting parents frequently experience mixed positive and negative sentiments toward their grown children (Birditt, Fingerman, & Zarit , 2010; Fingerman, Pitzer, Lefkowiz, Birditt, & Mroczek, 2008; Lowenstein, 2007; Pillemer et al., 2007). The ambivalence literature has relied primarily on global ratings, however. In this study, we looked at whether parents reported both positive and negative experiences with their children on the same day, that is, whether such ambivalent experiences co-occur in a short time frame.
We also were interested in whether daily experiences with grown children are associated with global relationship qualities in the tie. We considered relationship qualities and (a) mode of contact and (b) daily emotional experiences.
According to intergenerational solidarity theory, positive features of intergenerational relationships co-occur; that is, ties characterized by frequent contact also involve affection and positive regard (Fingerman, Sechrist, & Birditt, 2013; Silverstein & Bengtson, 1997; Suitor, Pillemer, & Sechrist, 2006). We expected parents to interact often with grown children with whom they had positive relationships across all modes of contact.
The conflict perspective suggests that parents also are involved with grown children with whom they do not get along well (Birditt & Fingerman, 2012). Parents may have daily contact in the context of difficult relationships with grown children for many reasons. Disagreements or arguments arise via interactions; during a conversation, the parties may feel criticized or irritated (Birditt, Rott, & Fingerman, 2009; Fingerman, 2001). Likewise, parents may reach out to children suffering problems (Byers, Levy, Allore, Bruce, & Kasl, 2008; Fingerman, Miller, Birditt, & Zarit, 2009; Pillemer & Suitor, 1991), and parents typically report more negative quality relationships with such children (Birditt et al., 2010).
In sum, we expected parents to report contact with children with whom they have particularly positive relationships as well as children with whom they have conflict. It is not clear whether parents engage in different modes of contact (e.g., in person, by phone) based on relationship qualities with children. Thus, we examined this issue without hypotheses.
We also considered relationship qualities of children who generate pleasant and stressful daily experiences for parents. Persistent qualities of parents’ relationships may be associated with the types of daily experiences they have. Parents can develop strong bonds with children who also generate distress and worry. For example, one study found that parents experienced similar levels of affection for children who were suffering problems as for children who were successful. Yet parents experienced more conflict with the children suffering problems (Birditt et al., 2010). Parallel to these patterns, parents may report pleasant daily experiences with children for whom they had both high positive or negative relationship quality but report daily stressors primarily with children with whom they have more difficult relationships.
Daily events have the greatest effect when people care about those events. As such, daily experiences with grown children may affect parents. The intergenerational investment model suggests that parents are sensitive to relationships with grown children (Fingerman, Cheng, Tighe, et al., 2012). Parental investment in grown children reflects past efforts raising those children and future expectations that grown children serve as a legacy for the parent (Giarrusso, Feng, & Bengtson, 2005; Sayer, Bianchi, & Robinson, 2004). Indeed, recent studies have found that parents’ well-being is closely associated with events in their grown children’s lives, though this studies have not looked at daily experiences (Byers et al., 2008; Fingerman et al., 2008; Greenfield & Marks, 2006). Daily emotional experiences in this tie also may be associated with daily well-being.
Parents’ daily mood in particular may be sensitive to daily experiences with grown children. Research has shown that daily stressors play a key role in dampening daily mood (Birditt, 2014; Birditt, Fingerman, & Almeida, 2005; Bolger, Davis, & Rafaeli, 2003; Cichy, Stawski, & Almeida, 2014; Diehl & Hay, 2010). Similarly, pleasant encounters with social partners may enhance positive mood. Thus, we expected parents to report more positive mood when they had pleasant interactions with grown children and more negative mood when they experienced daily stressors involving their children.
Having both pleasant and stressful experiences with grown children on the same day (i.e., ambivalent experiences) also may be associated with mood. Social psychologists and family scientists have linked ambivalent experiences with social partners, in particular with grown children, to poorer psychological health (Fingerman et al., 2008; Lowenstein, 2007; Uchino, Holt-Lunstad, Uno, & Flinders, 2001). These findings reflect disappointment or a sense of unpredictability in an ambivalent tie. Ambivalent ties hold an expectation of positive experience, but the actual experience is interspersed with negatives that generate poorer mood (Uchino et al., 2001). Likewise, if parents experience pleasant and stressful experiences with a child on the same day, we expected them to report poorer mood.
By contrast, a stress-buffering model suggests positive events may “cancel” effects of a negative event (Cohen & Wills, 1985; Okun, Melichar, & Hill, 1990); that is, pleasant encounters may mitigate the effects of a stressful experience with a child that day. For example, a grown child may call with a problem, upsetting the parent. Later that day, the same child or a different child may call and share a joke at work or a funny story about how his or her toddler sings the ABCs to fall asleep. The amusing story may alleviate the distress over the problem. We examined the stress-buffering hypothesis as well.
We controlled for parent factors that might be associated with daily interactions, including gender. Studies indicate that mothers have more frequent contact with grown children than fathers (Rossi & Rossi, 1990). Likewise, African American families report stronger intergenerational ties than European American families, and stronger reactions to daily experiences with family members (Chatters, Taylor, Lincoln, & Schroepfer, 2002; Cichy, Stawski, & Almeida, 2014). Family size also may determine parents’ activities with each grown child because parents with more children are less likely to interact with any given child (Fingerman, Cheng, Birditt, & Zarit, 2012; Fingerman, Cheng, Tighe, et al., 2012). Similarly, grown children may seek contact with younger siblings still at home and in doing so may encounter their parents. Thus, we controlled for younger children in the home.
We controlled for offspring characteristics that might be associated with parental daily involvement with grown children as well. We controlled for gender because daughters typically have closer ties to parents than sons do (Suitor & Pillemer, 2006), though we also note that recent research suggests that daughters’ and sons’ roles may be converging (Fingerman, Hay, Kamp Dush, Cichy, & Hosterman, 2007). Parents typically are more involved with younger adult offspring than with older adult offspring (Fingerman, Cheng, Tighe, et al., 2012; Swartz, 2009), and we controlled for offspring age. We also controlled for child’s marital status because grown children who are married tend to be less involved with their parents than unmarried grown children (Sarkisian & Gerstel, 2008).
Research has linked qualities of ties with grown children to parental well-being (Fingerman et al., 2008; Fingerman, Cheng, Birditt, & Zarit, 2012; Kiecolt, Blieszner, & Savla, 2011; Umberson, Pudrovska, & Reczek, 2010). Daily diary studies assess everyday experiences not easily captured in global ratings (Gunthert & Wenze, 2012; Laurenceau & Bolger, 2005). This study is the first to ask about mode of contact parents use to stay in touch with grown children (i.e., in person, by telephone, or by text). The study also illuminates everyday life in parentchild ties by examining emotional experiences with grown children each day and their associations with parental daily well-being. Furthermore, in the study we asked how ongoing relationship qualities (positive and negative) with grown children are associated with these daily experiences.
We examined different modes of contact through which parents and grown children maintain everyday contact and addressed three research questions:
We did not make predictions regarding whether associations with mood would endure into the following day, but we examined this issue via lagged effects.
The literature does not provide a basis for hypotheses regarding mode of contact, but we expected parents to report phone contact most often based on ease and familiarity with this form of communication. We also explored whether mode of contact varied by relationship quality; we did not have specific hypotheses regarding the type of contact parents use to be in touch with grown children with whom they have positive or conflicted relationships. We also asked whether parents have particular daily emotional experiences via different modes of contact. Parents may enjoy a grown child via any mode of contact; they may laugh in person, on the phone, or in reaction to a cute message or joke the child e-mails or texts. It is less clear whether stressful encounters occur across all modes of contact. Intuitively, conversations may yield conflict, arguments, or tensions; parents may be less likely to experience stressful encounters by text or e-mail. Yet any form of contact might instigate stressful thoughts by evoking worries or by reminding the parent of the child’s problems. We examined these issues here.
The sample included 247 adults ages 4565 (55% female) from the Family Exchanges Study Wave 2 (http://sites.utexas.edu/adultfamilyproject/family-exchanges-study/) who had at least one child over the age of 18. The initial Family Exchanges Study was recruited via listed samples and random digit dialing in the Philadelphia Metropolitan Statistical Area in 2008, at a time when 93% of households with adults in this age range had landlines. A second wave of data collection occurred in 2013. At Wave 2, participants first completed a 1-hour telephone or web-based survey (called the main survey here). They then completed brief telephone interviews each day for 7 days.
Participants who completed the main survey were invited to participate in the daily diary study until the quota was fulfilled (i.e., 240 individuals). Invitation to participate in the diary study was determined by random ID number (i.e., order of participation in the main study). Of 311 possible participants invited to the diary study, 270 agreed to participate (87%) and, of those, 248 (80%) did participate before enrollment ceased. We excluded from the current analysis one participant whose offspring died between Wave 1 and 2. A description of the sample is given in Table 1. Most participants (n = 198) completed all 7 days, but 49 participants completed fewer than 7 days of the study (1,631 days, M = 6.6 days per participant). Respondents received $7 for each daily survey with an additional $1 for completing all 7 days (total $50).
We compared diary participants to remaining participants from the main survey (n = 242) on background characteristics (e.g., age, gender, education) with t tests (for continuous variables) and chi-square tests (for categorical variables). Participants in the diary study were older (M = 56.42) than those who did not participate (M = 55.24), t(488) = 2.66, p < .001. Also, fewer participants in the diary study were ethnic minority (28%) compared to the main sample (38%; χ2 = 5.60, p = .02). There were no differences on other characteristics (e.g., education, marital status).
The study drew on variables from the main and diary surveys. Variables from the main survey included demographic characteristics (e.g., gender, age) and relationship qualities. Parents provided background information for each grown child but rated relationship qualities for only up to four grown children in this study (n = 578). Nearly all parents (93%) had four or fewer children. Parents with more than four children over age 18 reported on the child who received the most support, the child who received the least support, a randomly selected other child, and their youngest child over age 18 remaining.
Parents reported their own age, gender, marital status, years of education, and 2012 household income in 12 categories that ranged from 1 (less than $10,000) to 12 ($250,000 or more). Participants indicated their ethnicity and race, which were coded 1 = ethnic or racial minority and 0 = non-Hispanic White. Nearly all racial/ethnic minority participants were African American. Participants rated their health on a scale that ranged from 1 (poor) to 5 (excellent; Idler & Kasl, 1991). Participants reported the number of children they had who were age 18 or older, and the number of children under age 18, in their household. Because these numbers were correlated, we controlled for the number of children aged 18 and older and for presence of children under 18 in the household (1 = has children under age 18, 0 = no children under age 18) in analyses. Participants also reported background characteristics for their grown children (see Table 1).
In the main interview, parents reported geographic distance from child in miles; because of skew in this variable we used the log transformation in analyses. Parents also indicated whether the grown child resided in the same household, coded 1 = yes and 0 = no.
In addition, parents rated two items that assessed positive relationship quality: how much each child (a) makes them feel loved and cared for and (b) understands them. Items were rated on a scale that ranged from 1 (not at all) to 5 (a great deal) and were averaged to create a positive quality score (α = .75). Parents also rated two commonly used items regarding negative quality, how much each child (a) criticizes them and (b) makes demands. Items were averaged to create a negative quality score (α = .61; Birditt et al., 2010; Fingerman et al., 2008; Umberson, 1992). Positive and negative relationship quality scores showed a small but significant association (r = −.12, p = .006).
Each day, parents indicated whether they had contact with each child via an unfolding brackets approach to obtain in-depth information only when relevant (Coleman & Ganong, 2008; Danziger, Burgard, & Seefeldt, 2012). First, parents indicated whether they had contact with any child that day. If yes, they indicated which child they had contact with via three modes of contact: (a) in person, (b) by phone, or (c) via text/e-mail. We used these bracketing techniques throughout the survey to gather information about each child.
Each day, we asked whether the parent had a particularly enjoyable interaction with each grown child (yes/no). We also asked whether the parent shared a laugh with each grown child (yes/no) using bracketing techniques. Stressful encounters with grown children. Each day, parents indicated (yes/no) whether (a) they had interactions with the child that were irritating or annoying, (b) the child got on their nerves, and (c) they did not become upset in a potentially upsetting situation, or (d) the parent him- or herself behaved in a way they regretted. Bracketing techniques provided information about each child.
Parents also reported on two stressful thoughts that could occur in the absence of contact: (a) thinking about problems and (b) worrying about each grown child. Similar items have been used in other studies (Birditt, 2014; Hay et al., 2007). Daily relationship quality. To assure relationship quality was stable, we asked about positive relationship quality with each child each day. Each day, parent rated a single item: the extent to which they felt loved and cared for (1 = not at all to 5 = a great deal) by each child (Fingerman et al., 2008; Umberson, 1992).
Participants indicated, on a scale that ranged from 1 (none of the day) to 5 (all of the day), how much they had experienced six positive emotions (e.g., happy, determined, calm; α = .70) and nine negative emotions (e.g., distressed, lonely, nervous; α = .88) each day. These emotions were drawn from assessments of daily emotions (Birditt, 2014; Piazza, Charles, Stawski, & Almeida, 2012; Watson, Clark, & Tellegen, 1988).
This study included three levels of data: (a) participants’ children (nested within each day), (b) days of the study, and (c) participants. We refer to participant level (N = 247), day level (N = 1,631 across 7 days), and offspring-day level (N = 4,146; questions pertained to a maximum of four children for each participant on 7 days). This last unit pertains to the possible responses participants could provide each day; parents could report experiences with one to four children (n = 578 total offspring) each day for up to 7 days.
Analyses relied on multilevel models with two or three levels: (a) offspring nested within (b) days nested within, and (c) parent. Multilevel models account for different numbers of children nested within days, as well as different numbers of days nested within parents (81% of participants completed all 7 days).
We compared relationship qualities of grown children with whom parents had each mode of contact (i.e., in person, by phone, by text) to the children with whom they did not have that mode of contact each day. The main interview included a roster of all grown children in the family. Thus, for each day each grown child received a score of 1 = parent interacted with the child that day or 0 = parent did not interact with the child that day.
We considered three modes of contact (e.g., by text, by phone, in person) as outcomes, relying on logistic multilevel models with the GLIMMIX procedure in SAS. The global ratings of relationship qualities (i.e., positive and negative relationship quality) from the main interview served as independent variables. These models involved three levels: (a) offspring nested within (b) days nested within, and (c) parent. We controlled for parent gender, parental education, health, and racial/ethnic minority status (nearly all parents and grown children had the same racial/ethnic minority status), parents’ number of adult offspring, and whether the parent had younger children under age 18 in the household. We also controlled for offspring: gender, age, and marital status. Geographic distance between the parties was skewed,. and we used the log transformation of geographic distance in analyses.
Many parents (37%) resided with at least one grown child. Therefore, we repeated analyses substituting coresidence for distance. The pattern of findings was nearly the same and, for parsimony, we present only the models using distance (transformed).
Preliminary analyses also established that positive relationship quality ratings were stable across days. Using a 5-point scale, parents rated relationship quality for each child each day (M = 3.71, SD = 1.21). We estimated an intraclass correlation using a null model for ratings of relationship quality for each child, for each day, nested within parent. In this three-level model, 47% of the variance was between participants (parent), 32% of the variance was between children, and only 21% was between days. In other words, participants rated their feelings for each child fairly consistently across days, with daily variability accounting for only one-fifth of the variability.
The research questions involved parents’ daily emotional experiences with grown children: (a) pleasant encounters (e.g., any enjoyable interaction, laughter), (b) stressful encounters (e.g., irritation, got on nerves, potential irritation), and (c) stressful thoughts (e.g., worry, thinking about a problem).
As with the prior analyses, these models relied on logistic multilevel models using the GLIMMIX procedure in SAS. The independent variables were child’s relationship qualities. Again, we controlled for background characteristics and geographic distance. We repeated the analyses with offspring coresidence.
We took a parsimonious approach to presenting models. First, we estimated models that examined each type of pleasant or stressful experience separately. Then we examined a combined index for each of the types of daily experiences. For example, we first estimated separate models for (a) irritating interaction (yes/no), (b) child got on nerves (yes/no), and (c) child was irritating but parent avoided getting upset (yes/no). We ascertained that the pattern of findings was the same for each experience and generated a single index for stressful encounters; that is, if the parent had engaged in any of those experiences with a given child (or multiple activities; e.g., child got on nerves and also parent avoided getting annoyed), we coded 1 = stressful encounter (otherwise 0 = no stressful encounter) for the child that day. Similarly, if the parent had either an enjoyable visit or laughed with that child, that child was coded as 1 = pleasant encounter and otherwise 0 = no pleasant encounter that day. Thus, we report the three types of experiences (pleasant, stressful encounters, stressful thoughts) as the outcomes, but provide supplementary tables on the Journal of Marriage and Family website (http://onlinelibrary.wiley.com/journal/10.1111/(ISSN)1741-3737) for each type of daily experience.
We also considered mixed experiences of pleasant encounters and stressful encounters or stressful thoughts each day. Parents may incur a mix of daily pleasant and stressful experiences in two ways: (a) both types of experiences with the same child on the same day (ambivalence with one child) or (b) both types of experiences with different children on the same day.
We examined parents’ daily positive and daily negative mood associated with daily experiences with grown children. These models involved two levels: (a) days nested within (b) parent. The outcomes were continuous variables, and thus we used standard multilevel models. We considered whether the parent had a pleasant encounter, stressful encounter, or stressful thoughts with any child that day as the independent variables. We controlled for participant age, gender, years of education, health, racial/ethnic minority status, number of grown children, and presence of children under age 18.
In initial models we also included interaction terms for: (a) pleasant encounters × stressful encounters and (b) pleasant encounters × stressful thoughts. In the final models, we included only interaction terms with significant associations with the outcome variables, positive or negative mood (Tabachnick & Fidell, 2013).
Lagged models ascertained whether consequences of experiences from one day persist the next day (beyond experiences that day; Hamaker, 2012). We considered (a) same-day experiences predicting same-day mood and (b) lagged models with a prior day’s experiences predicting the next day’s mood. In lagged models, we controlled for same-day experiences.
We expected parents to report contact with at least one grown child during the study week. On the basis of ease of communication (Cotten et al., 2012), we expected parents to be most likely to report telephone contact with their grown children
As can be seen in the first column of Table 2, nearly all parents (96%) had contact with at least one grown child during the study week. Most parents spoke on the phone with one or more grown children (88%). Moreover, three quarters had seen a grown child in person and two thirds had texted with a grown child. The second column of Table 2 shows the proportion of days on which each type of contact occurred. On average, parents had some sort of contact with a grown child 5 to 6 days that week, with phone calls and in-person contact occurring 3 to 4 days that week.
Table 2 provides descriptive information regarding parents’ daily emotional experiences with grown children. We expected parents to report frequent pleasant encounters with their grown children. Most parents (90%) reported an encounter with a child that they enjoyed, and nearly as many (89%) had laughed with a grown child at least once during the study week.
We also expected parents to report stressful experiences with grown children, though less often than pleasant experiences. Over half of parents reported at least one stressful encounter during the study week (e.g., irritating interaction, child got on nerves, child was irritating but parent avoided getting upset). Only a few parents acted in ways they regretted (10%); thus, we did not consider parents’ regrettable actions further.
More than three fourths of parents reported stressful thoughts that week. Parents worried (65%) or thought about a problem (54%) with a grown child that week.
Parents reported having an enjoyable and a stressful encounter with the same child on the same day relatively rarely (i.e., unit offspring-day). Parents experienced pleasant encounters and stressful thoughts on 10% of offspring-days and pleasant encounters and stressful encounters on 7% of offspring-days.
Therefore, in analyses and in the data in Table 2 we considered whether a parent had both pleasant experiences and stressful experiences with grown children that day, whether with the same child or different children. We present combined indices for pleasant experiences (e.g., laughter or enjoyable encounter) with (a) stressful encounters (e.g., irritating encounter, child got on nerves, parent was irritated but tried not to get upset) or with (b) stressful thoughts (e.g., worry or thought about problem), either with the same child or different children in the family each day.
As can be seen in Table 2, more than three fourths of parents experienced a mix of both pleasant and stressful experiences with grown children at least one day during the study week. Pleasant encounters and stressful thoughts co-occurred most often (68% of parents had both experiences on at least one study day). Finally, few parents (n = 4) reported neither a stressful nor a pleasant experience with grown children that week.
We then tested the research questions regarding relationship qualities and daily contact. We had predicted that parents would be more likely to report daily contact when they got along better or when they had more conflict with a grown child. We estimated logistic multilevel models with daily contact as the outcome, 1 = any contact and 0 = no contact that day with that child, for each mode of contact. Positive and negative relationship qualities served as the independent variables.
As can be seen in Table 3, the findings partially supported our expectations. Parents were more likely to report daily contact across all three modalities with children with whom they had more positive relationship qualities. They were 1.45 times as likely to have in-person contact for each point higher that they rated their positive relationship quality with a child. Similarly, they were 2.44 times as likely to talk on the phone with a grown child each day and 1.82 times as likely to text a child for each additional point of positive relationship quality.
Parents also reported more negative relationship quality with grown children with whom they talked via phone (odds ratio [OR] = 1.59) or text (OR = 1.26). In-person contact was not significantly associated with negative relationship quality.
Next, we examined emotional daily experiences and relationship qualities. We expected parents to be more likely to report pleasant daily encounters with children with whom they had either more positive or more negative relationships. We expected parents to be more likely to report stressful encounters and thoughts when they had more negative-quality relationships with the child. Multilevel models treated pleasant encounters, stressful encounters, and stressful thoughts as the dependent variables. Positive and negative relationship qualities served as the independent variables.
As expected, parents reported pleasant daily encounters with grown children with whom they had either higher positive or higher negative relationship quality (see Table 4). Supplementary Table 1 on the Journal of Marriage and Family website shows models for each type of pleasant encounter (e.g., laughter, pleasant interaction); the pattern of findings is the same as that shown in Table 4.
As expected, parents had greater likelihood of stressful encounters with grown children when they had more negative relationships (see Table 4). Positive relationship quality was not associated with stressful daily encounters. We also examined each stressful situation as a distinct outcome, and the pattern of findings was similar (see online Supplementary Table 2).
The findings also supported expectations regarding greater likelihood of stressful thoughts with grown children with whom parents had more negative relationship qualities. The model also revealed an unexpected finding regarding positive relationship quality, however: Stressful thoughts were more likely to occur regarding offspring with whom parents had less positive relationship qualities. For each point lower on the positive relationship quality scales, parents were 1.25 times as likely to report stressful thoughts with a child each day (OR = 0.75; see Table 4). The pattern of findings was similar for worrying and thinking about problems separately (see online Supplementary Table 3).
Because the mixed experiences could occur via pleasant experiences with one child and stressful experiences with another child, we did not examine associations with relationship quality.
We also examined associations between mode of contact and pleasant and stressful encounters. The literature provides little information regarding mode of contact, but we speculated that parents may enjoy a pleasant experience with grown child via any mode of contact. We intuitively expected that conversations serve as a conduit for arguments and conflict and thus that stressful encounters may occur by phone or in person. We expected stressful thoughts to be associated with all three modes of contact. In these models, pleasant encounters, stressful encounters, or stressful thoughts served as the dependent variables, and the three modes of contact (e.g., in person, by phone, by text) served as independent variables.
With the exception of in-person contact and stressful thoughts, all modes of contact were associated with pleasant encounters and stressful encounters and thoughts. As expected, all three modes of contact were associated with pleasant encounters. We had not expected text messages to be associated with stressful encounters, but all three modes of contact were associated with stressful encounters (see Table 5). Texting and telephone contact were associated with stressful thoughts, but in-person contact was not associated with stressful thoughts about a given child on a given day.
These findings suggest that contact is necessary for a pleasant or a stressful encounter to occur, but the mode of contact does not generate a specific emotional experience. Because the modes of contact did not differentiate pleasant and stressful experiences each day, we did not consider mode of contact with regard to parental daily mood.
We examined daily positive and negative mood as a function of pleasant and stressful experiences (e.g., pleasant encounters, stressful encounters, stressful thoughts) with grown children. Because the outcome was mood at the daily level, we considered whether the parent had each type of experience with any grown child each day. We did not consider relationship qualities with grown children and daily mood because (a) relationship qualities were relatively stable and did not vary by day and (b) the daily emotional experiences were associated with relationship qualities. As such, we focused on how daily experiences with grown children were associated with parents’ daily mood.
We also considered interaction terms for (a) pleasant encounters × stressful encounters and (b) pleasant encounters × stressful thoughts that same day. Only the interaction term for pleasant encounters × stressful encounters the same day was significantly associated with negative mood. For parsimony, we did not include nonsignificant interaction terms in the models (see Table 6).
We expected pleasant experiences with a grown child each day to be associated with more positive mood. We expected stressful encounters, stressful thoughts, and mixed pleasant and stressful experiences to be associated with more negative mood each day. As can be seen in Table 6, pleasant encounters with a grown child were associated with more positive mood on any given day, after controlling for the prior day’s mood and experiences. Stressful encounters and stressful thoughts were not associated with diminished positive mood.
As expected, negative mood was associated with stressful encounters and stressful thoughts that day (controlling for prior day’s mood and experiences). Moreover, the interaction term for pleasant encounters × stressful encounters showed a negative association with negative mood; that is, when parents had a pleasant encounter with a grown child on the same day as a stressful encounter, the pleasant encounter partially mitigated the effect of the stressful encounter on negative mood; parents reported less negative mood on those days. Figure 1 is a diagram of the difference in negative mood when parents experienced both a pleasant and a stressful encounter and when they did not. As can be seen in the figure, having a stressful encounter was associated with more negative mood that day. Yet when parents also experienced a pleasant encounter with a grown child that day, their negative mood was diminished in comparison to when they did not have such a pleasant encounter.
We estimated additional models to understand the findings. We considered the number of children with whom parents had different experiences and mood. To do so, we reestimated the multilevel models for mood with control variables (including number of children), comparing daily pleasant and stressful experiences with (a) no children, (b) one child, and (c) two or more children (vs. no children) that day. The pattern of findings was similar to those reported in Table 6. Pleasant and stressful experiences with one or more children were associated with daily mood.
Frequency of contact between parents and grown children has increased over the past 25 years (Fingerman, Cheng, Tighe, et al., 2012); nearly every parent reported contact with a grown child during the study week, and parents reported such contact on 5 or 6 days a week, on average. Indeed, rates of contact in this daily study were higher than in surveys that have used global estimates of contact (Arnett & Schwab, 2013; Fingerman, Cheng, Tighe, et al., 2012). This discrepancy may reflect daily contact that is incidental and minor—a son phones from the store to ask about the brand of soup for a recipe; a daughter texts her parents with a joke. These brief encounters may not register when parents report frequency of contact globally, but they pepper daily life.
In addition, this study broke new ground regarding (a) mode of contact (i.e., in person, by telephone, by text or e-mail), (b) pleasant and stressful daily experiences, and (c) daily mood. In the following section we consider the implications of findings in each of these areas.
Cultural, economic, and technological factors contribute to recent increased contact between generations. In this study, parents and their grown children had contact via phone, by text, or in person on 56 days that week. This frequent contact likely reflects the widespread use of cell phones that make such contact cheap and easy (Cotten et al., 2012; Fingerman, Cheng, Tighe, et al., 2012). In the past, modes of communication were limited (Cotten et al., 2012). Telephone calls occurred via one shared landline in a household, and long-distance calls were expensive (Fingerman, Cheng, Tighe, et al., 2012). Thus, a grown child might relate a conflict from work during a Sunday visit or phone call, but the problem had either already been resolved or the sense of immediacy was lost. The current high frequency of contact may result in greater emotional contagion between generations. Frequent contact provides increased opportunity for parents to empathize with their children and may make their own emotional lives more susceptible to events in their children’s lives. In the current era, grown children relay their experiences in real time (often accompanied by photos). As such, parents may experience their grown children’s ups and downs in a raw, unprocessed form, influencing their own mood more readily.
Future research also might focus on the role of e-mail, text messages, and social media platforms (e.g., Instagram, Twitter) in daily contact as these communications become more pervasive in this tie. Much of the contact may be mundane or trivial, but parents also reported pleasant and stressful experiences via phone and text, suggesting that these forms of real-time contact affect their emotional state.
Parents report happiness when their daily interactions with young children are enjoyable (Nelson, Kushley, English, Dunn, & Lyubormirsky, 2013; Nelson, Kushley, & Lyubomirsky, 2014). Likewise, in this study nearly all parents reported at least one enjoyable encounter and had shared a laugh with a grown child that week. Thus, parenting grown children has an “up” side in daily life.
More than half of parents, however, reported stressful encounters with grown children during the study week. Moreover, the findings were consistent with prior research indicating that parents report worrying about grown children (Hay et al., 2007); most parents reported stressful thoughts throughout the week. We did not inquire about pleasant thoughts because of concerns about potential social desirability and reactivity involved in such questions. In pilot studies nearly all parents, when prompted, claimed to have pleasant thoughts about their grown children. Future research should focus on other emotional qualities of parents’ daily thoughts about their grown children.
Our findings fit expectations that parents had contact with grown children with whom they had both positive and negative relationship qualities across all three modes of contact. Yet parents reported more negative relationship qualities with grown children with whom they had contact via telephone and text, and not with grown children whom they saw in person.
As such, conflict and intergenerational ambivalence may be technologically mediated. Solidarity theory suggests that parents may choose to get together in person with offspring whom they view more favorably (Fingerman et al., 2013; Silverstein & Bengtson, 1997). Alternately, in-person contact may mitigate negative relationship qualities. These findings are also consistent with the fact that in-person contact was not associated with worrying or fretting about problems with grown children. Future research should ask whether in-person contact dispels negative relationship qualities in some way.
As we had expected, all three modes of contact were associated with reports of pleasant experiences with grown children. We had expected conversations to serve as a conduit for arguments and conflict and thus stressful encounters to occur by phone or in person, yet stressful encounters also occurred via text. Thus, pleasant and stressful encounters may occur through any mode of contact.
Stressful thoughts were limited to text and telephone. In-person contact was not associated with worrying or fretting about problems with grown children. If in-person contact somehow dispels worries about grown children, this finding may also partially explain the lack of significant findings regarding in-person contact and negative relationship qualities. Perhaps in-person contact allows participants to avoid or to more easily resolve negative thoughts and this process, in turn, mitigates negative relationship qualities.
As we had expected, parents engaged in laughter and pleasant interactions with children with whom they had difficult relationships as well as with children with whom they get along well. This pattern is consistent with research showing that parents experience positive feelings toward their difficult children as well as their more favored children (Birditt et al., 2010).
Also, as expected, stressful encounters and stressful thoughts were restricted to offspring with whom parents reported negative relationships. Indeed, parents reported diminished positive relationship quality when they worried about their grown children or thought about their problems. This pattern extends an understanding of the family conflict perspective (Birditt & Fingerman, 2012) according to which grown children with whom parents report more difficult relationships offer some rewards but also are a source of stressful encounters and stressful thoughts on a daily basis.
The present findings also lend complexity to the intergenerational ambivalence perspective (Fingerman et al., 2008; Lowenstein, 2007; Pillemer et al., 2007). Prior research has not examined parents’ mixed emotional experiences with grown children on a daily basis. We did not collect data regarding ambivalence per se during a particular interaction with grown children but instead pleasant and stressful experiences occurring at any point that day.
It is surprising that parents did not report having both a pleasant and a stressful experience with the same child on the same day very often. Such experiences represented only a small fraction of daily experiences in this tie. Of course, other assessments of ambivalence, such as feeling conflicted or torn regarding a child (Pillemer et al., 2007), might have yielded different findings.
Nonetheless, parents in this study often reported some type of pleasant experience with a grown child and a stressful experience with another child on the same day. This suggests that global reports of ambivalence or mixed feelings toward grown children may reflect a gestalt of experiences. Throughout the day, parents may experience sentiments that leave them with conflicted emotions. The study adds to a burgeoning literature regarding distinctions between momentary experiences and global cognitive representations of those experiences (Gunthert & Wenze, 2012). In prior studies, parents’ global reports have indicated mixed emotions toward a given child (e.g., Fingerman et al., 2008; Lowenstein, 2007), but their daily experiences do not indicate that such mixed emotions occur in the moment with a specific child.
The findings in this study support expectations that parents would report more positive mood on days when they had pleasant encounters with their grown children and more negative mood on days when they had stressful encounters or stressful thoughts about children.
Researchers have documented that parents’ daily interactions with children in childhood are associated with parental mood (Repetti et al., 2011), and the same was true in adulthood. Studies have shown that parental well-being is susceptible to grown children’s problems in a global way (Fingerman, Cheng, Birditt, & Zarit, 2012; Pillemer & Suitor, 1991; Umberson et al., 2010). This study suggests that parents may suffer vicarious distress on a daily basis via worries and thinking about their children’s problems.
As in other diary studies (Gunthert & Wenze, 2012), in this study we examined events and mood each day, but nonetheless causality should be interpreted with caution. For example, the association between stressful encounters and negative mood may occur because parental poor mood generates conflicts with a grown child on a given day, rather than the reverse. The analyses that included prior-day experiences indicated some stability in moods across days, but also two carryover effects. Previous days’ stressful encounters seemed to depress next days’ positive mood, and previous days’ stressful thoughts seemed to contribute to next days’ negative mood. Further investigation of the characteristics and predictors of unresolved stressful thoughts and encounters may provide helpful information about reasons for ongoing distress.
The most intriguing findings regarding parental mood pertain to days when parents experienced both pleasant encounters and stressful experiences with grown children. Although ambivalence theory suggests that such mixed experiences generate poorer mood (Uchino et al., 2001), this was not the case; instead, our findings support a stress-buffering model (Cohen & Wills, 1985; Okun et al., 1990): The pleasant encounter with a grown child seemed to diminish the repercussions of the stressful experiences. The experiences may have occurred with different children, and thus we did not tap ambivalence per se. Moreover, unfortunately, the temporal order of these events is unclear, and the pleasant event may precede the stressful on on any given day.
Nonetheless, these findings warrant follow-up. A study using global reports of relationships with grown children did not find that an affectionate relationship with a grown child mitigated the deleterious effects of a problematic child (Fingerman, Cheng, Birditt, & Zarit, 2012). Therefore, the translation of daily stress buffering to parental general well-being warrants future research.
This study has other limitations. Rates of contact in this study were higher than in studies that have used global ratings (Arnett & Schwab, 2013). Perhaps this study captured mundane contact that is not well remembered in global ratings (Gunthert & Wenze, 2012). Alternately, although research suggests that participants do not alter behaviors in a daily diary paradigm (Gunthert & Wenze, 2012), parents may have sought contact with grown children because of participation in the study.
Moreover, this study focused on 1 week and did not capture vicissitudes over multiple weeks or months. Moreover, the ability to understand whether events co-occurred was limited by assessments at end of the day. Future research might use momentary experience sampling throughout the day (Reis, 2012). The Family Exchanges Study included adults from a wide cross-section of socioeconomic status backgrounds, however, and adults with little education may be unfamiliar with the technologies or websites used in studies that attempt to sample experience throughout the day. .
Parents remain involved in the daily lives of their grown offspring, with frequent contact across a variety of modalities. Findings from this study suggest that daily experiences are a contributing source of enduring ties between parents and their adult offspring. Even when parents report the relationship is conflicted, they at times enjoy their children’s company. Yet offspring can generate distress for parents through their daily interactions or parents’ thoughts. A complete portrait of family life may include a mix of both good times and bad, with the respective amount of these experiences contributing to parents’ well-being. In sum, this study provides fodder for examining daily experiences in this tie: Daily experiences matter.
This study was supported by grants from the National Institute on Aging under award number R01AG027769 the Family Exchanges Study II, the Psychology of Intergenerational Transfers (Karen L. Fingerman, Principal Investigator). The MacArthur Network on an Aging Society (John W. Rowe, Network Director) also provided funds. This research also was supported by Grant 5 R24 HD042849 awarded to the Population Research Center at The University of Texas at Austin by the Eunice Kennedy Shriver National Institute of Child Health and Human Development. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Karen L. Fingerman,
Kyungmin Kim, *
Kira S. Birditt,
Steven H. Zarit,