Social support and well-being among older adult married couples: A dyadic perspective

Abstract

The present study takes a dyadic approach to examine how social connectedness (i.e., neighborhood support, family and friend support, and social network characteristics) is associated with husbands’ and wives’ well-being during older adulthood. Participants included married couples (N = 832) from the National Social Life, Health and Aging Project, a national probability sample of older adults aged 57 and older. Using a series of actor–partner interdependence models, we found that wives’ and husbands’ closeness with their own social network, family support, and neighborhood social ties were all significantly associated with emotional well-being, and closeness to one’s social network was associated with self-rated health. Friend support was associated with emotional well-being only for women, while family support was associated with self-rated health for men only. Four partner effects emerged, with husbands’ friendship support being positively associated with wives’ self-rated health, husbands’ greater talk frequency with his social network being associated with wives’ lower emotional well-being, wives’ greater talk frequency with her social network being associated with husbands’ lower self-rated health, and wives’ closeness to their social network being positively associated with husbands’ self-rated health. The present study has implications for programming and the role of social connectedness beyond marriage.

Keywords

Aging couple relationships marriage older adulthood social networks well-being

Spouses are often viewed as central to each other’s social networks, especially in later life, when overall network size decreases (e.g., Lang & Carstensen, 1994). Married individuals are afforded health benefits including increased levels of self-rated health and decreased levels of functional limitations (Hughes & Waite, 2009; Williams & Umberson, 2004). Moreover, the well-being of husbands and wives is often associated with their spouses’ reports of well-being (e.g., Beach, Katz, Kim, & Brody, 2003). Higher levels of physical and mental health in married individuals can be attributed to many factors internal to the marriage, but also to factors external to it, including social support outside of a marriage (e.g., Golden, Conroy, & Lawlor, 2009; Pienta, Hayward, & Jenkins, 2000).

The role of social networks is often overlooked in the context of marriage (e.g., Felmlee & Sprecher, 2000), especially in later life. Much research on marriage focuses on marital interactions without taking into consideration how social networks external to the dyad may be linked to both individual and spousal well-being. Research that does examine social networks in the context of marriage tends to examine young, newlywed couples (e.g., Jackson, Kennedy, Bradbury, & Karney, 2014). The present study aims to address this gap by taking a dyadic approach to examining how social connectedness (i.e., social network characteristics, family and friend support, and neighborhood support) is associated with husbands’ and wives’ well-being during older adulthood. Throughout the article, well-being will refer to both emotional well-being and self-rated health, the two primary dependent variables of interest, based on how others have conceptualized well-being (e.g., Helliwell & Putnam, 2004).

Social networks and individuals

Feeling connected with others is associated with higher levels of life satisfaction and lower levels of loneliness and social isolation (Bishop, Martin, & Poon, 2007; Cornwell & Waite, 2009). When an individual does not have a supportive social network, physical health may suffer; lower levels of social and emotional support are associated with poorer overall physical health, mental health, and health behaviors, even for individuals who are married (Strine, Chapman, Balluz, & Mokdad, 2008). Social support is also inversely associated with mortality, with a greater risk of mortality among women who report lower levels of social support (Lyyra & Heikkinen, 2006). Therefore, examining how individuals are socially connected to others is necessary.

Specific types and sources of social support, such as support from family, friends, or neighbors, may be associated differently with well-being. Social contact with family members tends to be more obligatory, whereas social interactions with friends and neighbors are more voluntary in nature. The more obligatory nature of family relationships may contribute to interactions being more ambivalent, or simultaneously emotionally positive and negative (Fingerman, Hays, & Birditt, 2004). The ambivalent nature of family relationships might mean that those relationships are more emotionally fraught than friendships or relationships with neighbors, and this may have implications for well-being. Indeed, Huxhold, Miche, and Schuz (2014) found that informal activities with friends were associated with increases in life satisfaction and decreases in negative affect, while informal activities with family members were only associated with increases in negative affect. These findings from Huxhold and colleagues (2014) further stress the importance of delineating between sources of social connectedness.

The associations between friendship support, neighborhood social ties, and well-being factors may change over the life course. Compared to younger adults, older adults’ friendship support is associated with fewer depressive symptoms (Thomas, 2016). During older adulthood, social activities with friends are associated with greater life satisfaction, whereas during middle adulthood, both family and friends are associated with subjective well-being (Huxhold, Miche, & Schuz, 2014). Neighbors also play a role in the social interactions of older adults and are a potential resource that older adults can utilize. Older adults tend to have more contact with other acquaintances, including neighbors, than middle aged adults (Wrzus, Hanel, Wagner, & Neyer, 2013), making the study of neighborhood ties during older adulthood especially relevant.

Social networks in the context of marriage

Marriage may provide individuals with the presence of additional supportive social network members (Curran, McLanahan, & Knab, 2003). Kalmijn (2003, p. 232) coined the term “marital capital” to discuss how marriage brings two social networks together, potentially providing additional social resources for spouses. Relationships with other supportive individuals may transfer to a spouse, which can largely be explained by interdependence theory. Interdependence theory posits that the behaviors of one individual in a relationship influence the other individual (e.g., Kelley, 1983). For example, an individual’s experience of high levels of friendship support may benefit both his/her own well-being and that of his/her spouse/partner. However, individuals who have low levels of friendship support may find themselves relying primarily on spouses for support. This could potentially place an additional strain on the spouse and, in turn, be associated with poorer well-being.

A spouse’s social network may, however, provide an individual with much needed social support outside of the marriage and “spillover” may occur. Prior research has often examined spillover in the context of work–family balance (e.g., Heller & Watson, 2005; Stevens, Minnotte, Mannon, & Kiger, 2007). Yet, this research can be applied to marriages and social networks. Positive well-being outcomes may occur for an individual in cases where a spouse is experiencing positive relationships with family and friends. However, it could be that negative aspects of relationships outside of a marriage could also “spillover” into a spouse’s life. It is also necessary to consider that spouses do not always share their entire social network with one another. Using a sample of adults under age 65, Kalmijn (2003) found that 43% of friends are not shared with a partner. Since friends are not always shared, spouses’ friendship networks may be associated with varying outcomes for each spouse.

Gender, social networks, and marriage

Social network experiences and well-being also differ by the gender of the respondent. Some gender differences in social network experiences might be due to socialization. Women are often socialized to be expressive (i.e., adept in social and emotional skills), whereas these traits are not often emphasized for men, especially in Western cultures (Löckenhoff et al., 2014). Women also tend to retain a greater number of “separate” friends throughout life and may potentially benefit from friend support more than men as a result (Kalmijn, 2003; Walen & Lachman, 2000). When examining friendship, Antonucci, Lansford, and Akiyama (2001) found that women reported greater depressive symptoms when they did not have a friend as confidante; this was not true for men. Mixed findings exist with regard to family support, gender, and well-being. Men have been found to benefit more from family support as compared to women (Cable, Bartley, Chandola, & Sacker, 2013). A study that addressed the role of social support after a new diagnosis of heart disease among older adults found that women reported better self-rated health from family contact to a greater degree than men, yet women and men benefitted similarly when heart disease was chronic (Donato, Leon-Perez, Wallston, & Kripalani, 2018). Therefore, dyadically examining married couples’ social networks can further understand how the different types of social networks of both partners may play a role in one another’s well-being and the potential gendered nature of these associations.

Spouses’ gender may also play a role in how the dyad navigates their social relationships. Men may be especially susceptible to their wives’ friendship experiences; husbands may find themselves relying on a spouse for access to social networks (e.g., Waite & Harrison, 1992). In terms of gender, women often find themselves in the role of kin-keeper, maintaining social ties with other family members (e.g., Hagestad, 1986). This role of kin-keeper may extend to other relationships including sustaining relationships with neighbors and friends. Indeed, Davidson (2004) found in a qualitative study that older married men tended to have couple-oriented social networks. However, at the same time, age may alleviate the gendered expectation for men to refrain from expressing emotions, enabling men to cultivate more emotionally intimate relationships than earlier in life. Examining social network characteristics relevant to both partners is necessary to have a full understanding of how marriages and partnerships are embedded in larger social networks.

Present study

The present study examines the associations between one’s own and a partner’s social network characteristics, family, friend, and neighborhood social support; and emotional well-being and self-rated health. The main effects hypothesis posits that social connectedness factors can be directly attributed to well-being (Cohen & Wills, 1985), providing an explanation for the potential link between social connectedness factors and well-being. Because of the enmeshment experienced by married older adults, interdependence theory (Kelley, 1983) serves as a theoretical basis. We are especially interested in examining how individuals’ social connectedness may be associated with their spouses’ outcomes. The socioemotional selectivity theory (SST) and the social convoy model (SCM) also serve as theoretical frameworks. The SST posits that as we age, we become more focused on the quality of relationships, pruning relationships that we may not view as important or positive, and investing time and energy into those that are important (Lang & Carstensen, 1994). Thus, the relationships that remain in later life might be particularly salient for older men’s and women’s well-being. The SCM functions as a theoretical framework due to its emphasis on the multidimensionality of social networks and the inclusion of multiple domains of social connectedness in the present study (e.g., Antonucci & Akiyama, 1987). The SCM emphasizes the role of varying avenues of social support (e.g., friends, neighbors, family members) while simultaneously considering quality and quantity of social relationships. For example, family relationships may have stronger associations with well-being than other relationships as family tends to be deemed as a more central and obligatory tie. Together, the SCM, SST, and interdependence theory surmise that certain social connectedness factors may be more strongly associated with well-being, especially later in life, and that these well-being factors among spouses are interrelated.

Based on the literature reviewed above, the following three research questions guided the present study: (1) Are individuals’ neighborhood social ties, family and friend support, and social network characteristics associated with their own (a) emotional well-being and (b) self-rated health? (2) Are individuals’ neighborhood social ties, family and friend support, and social network characteristics associated with their spouses’ (a) emotional well-being and (b) self-rated health? (3) What role does the gender of a spouse play in the potential associations between neighborhood social ties, family and friend social support, and social network characteristics and (a) emotional well-being and (b) self-rated health?

Method

Participants

The present study includes 832 married couples from the second wave of the National Social Life, Health and Aging Project (NSHAP), a national probability sample of adults over age 57 (Waite et al., 2014). Participants were included in the sample if both an individual and their spouse participated in the second wave of the NSHAP. The mean age of married respondents at the second wave was 70.91 (SD = 7.71), 11.5% identified as Hispanic, 11.3% identified as Black, 7.5% identified as others, and 50% identified as female (see Table 1 for descriptives and spousal comparisons). Only heterosexual married couples are included in the present study as the vast majority of men (96.4%) and women (95.0%) included in the NSHAP identified as heterosexual (Lindau & Gavrilova, 2010).

Table 1.

Descriptive statistics and mean-level differences.

	M	SD	df	F	Range
Age			1, 1662	27.14**	36–99¹
Male	71.72	6.80
Female	69.69	8.66
ADLs			1, 1662	9.56**	0–15
Male	0.54	1.32
Female	0.43	1.21
Marital happiness			1, 1658	9.07	1–7
Male	6.35	1.33
Female	6.44	1.12
Talk frequency (inverse hyperbolic sine)^a			1, 1660	3.88*	1.19–8.2
Male	6.61	0.93
Female	6.71	0.86
Closeness			1, 1659	0.67	0.5–3
Male	1.91	0.51
Female	2.01	0.51
SN size			1, 1661	8.48**	1–5
Male	3.61	1.91
Female	3.71	2.01
Family support			1, 1643	1.45	0–4
Male	2.83	1.17
Female	2.93	1.15
Friend support			1, 1630	2.34	0–4
Male	2.16	1.31
Female	2.32	1.33
NST			1, 1487	0.01	0–9
Male	4.54	2.10
Female	4.47	2.10
Self-rated health			1, 1661	1.13	1–5
Male	3.26	1.07
Female	3.24	1.05
Emotional well-being
Overall mental health			1, 1660	0.15	1–5
Male	3.67	0.97
Female	3.62	0.96
CESD^b			1, 1649	2.35	14–44
Male	15.54	4.59
Female	15.79	4.80
Happiness			1, 1659	1.077	1–5
Male	3.71	0.84
Female	3.74	0.87
Loneliness^b			1, 1415	0.40	0–9
Male	2.79	2.15
Female	2.82	2.21
	%	%	df	χ²	Range
Race—Black^c	Black	Not Black	1	0.03	0–1
Male	10.3	89.7
Female	10.1	89.9
Ethnicity—Hispanic^c	Hispanic	Not Hispanic	1	0.14	0–1
Male	11.0	89.0
Female	11.6	88.4
Education^c	≤ HS degree	> HS degree	1	0.31	0–1
Male	38.9	61.1
Female	37.6	62.4

Note. NST = neighborhood social ties, M = mean, SD = standard deviation, NSHAP = National Social Life, Health and Aging Project, SN = social network, ADL = activity of daily living, HS = high school, CESD = Centers for Epidemiologic Studies Depression Scale.

^a An inverse hyperbolic sine was applied due to skewed responses.

^b Reverse scored in analyses.

^c Percentage and χ² were included for categorical responses as a χ² analysis was conducted for dichotomous variable.

¹Participants were included in the NSHAP if at least one spouse was aged 57 and over. The vast majority of spouses were over age 57, as can be seen by the M and SD.

*p < .05; **p < .01.

Instruments

Self-rated health

The following question was used to assess self-rated health: “Would you say your health is excellent, very good, good, fair, or poor?” This item was measured on a 1 (excellent) to 5 (poor) scale and was reverse scored, such that lower scores indicated worse self-rated health.

Emotional well-being

Emotional well-being was constructed as a latent variable, with loneliness, depressive symptoms, general happiness, and overall mental health serving as manifest indicators. The 3-item NSHAP Felt Loneliness Measure was used to assess loneliness. Questions include “How often do you feel isolated from others?” “How often do you feel that you lack companionship?” and “How often do you feel left out?” Responses range from 0 (never or hardly ever) to 2 (often). All responses were reverse coded such that lower levels equated to more felt loneliness. Prior research indicated acceptable levels of internal consistency (α = .72; Hughes, Waite, Hawkley, & Cacioppo, 2004), with the present sample having a good internal consistency (α = .81). Depressive symptoms were assessed with an 11-item version of the Center for Epidemiological Studies–Depression scale. Responses were added together and each item was scored on a 1 (rarely or none of the time) to 4 (most of the time) scale. All items were reverse coded with the exception of life enjoyment and happiness. Lower scores indicate greater depressive symptoms. Prior research has indicated a high level of internal consistency for this measure (α = .85) in older adult populations (Callahan & Wollinsky, 1994), with the present sample having an acceptable level of internal consistency (α = .78). Happiness was assessed with 1 item that ranged from 1 (unhappy usually) to 5 (extremely happy), “If you were to consider your life in general these days, how happy or unhappy would you say you are, on the whole….” Overall mental health was assessed by the following question that was assessed on a 1 (poor) to 5 (excellent) scale, “What about your emotional or mental health? Is it excellent, very good, good, fair, or poor?”

Social ties with neighbors

Three items were used to assess social ties with neighbors, including “How often do you and people in this area visit in each other’s homes or when you meet on the street?,” “How often do you and other people in this area do favors for each other?,” and “How often do you and other people in this area ask each other for advice about personal things?,” and were assessed on a 1 = never to 4 = often (scale). These 3 items were added together. Higher levels of neighborhood social ties were indicated by higher values. The present sample had an acceptable level of internal consistency (α = .75).

Support from family

Two questions addressed support from family: “How often can you open up to members of your family if you need to talk about your worries?” and “How often can you rely on them for help if you have a problem?” Participants could answer the questions on a 0 (never) to 4 (often) scale, with higher scores indicating greater family support. The responses from the two questions were summed together, with higher responses indicating more family support. Because this measure is a 2-item measure, a Spearman–Brown coefficient was calculated to check reliability. A positive correlation emerged between opening up to family members and relying on family members, r_s = .421, p < .001.

Support from friends

The same two questions as support from family were used to examine support from friends, with the referent changed from family to friends as a whole. A Spearman–Brown coefficient found a positive correlation between the 2 items, r_s = .561, p < .001.

Social network characteristics

Information about social network characteristics was taken from the social network roster portion of the in-person questionnaire. Participants could identify up to five individuals to whom they confide and were then asked questions regarding their relationship with that individual, which could include a spouse or partner (Cornwell, Schumm, Laumann, Kim, & Kim, 2014). Information about interactions with a spouse or partner were not included in the present study, which focused on talk frequency, social network size, and closeness to social network outside the marital relationship. The decision to use these items from the social network roster largely stemmed from suggestions by Cornwell and Waite (2009) and Cornwell, Laumann, and Schumm (2008) to include a variety of questions that examine social network characteristics. Talk frequency was assessed by asking participants “How often do you talk to this person?” Responses ranged from every day, several times a week, once a week, once every 2 weeks, once a month, a couple times a year, once a year, and less than once a year. These responses were then placed on a year metric so that we could assess how frequently a participant talked to the individual per year, and this transformation was done for each non-partner/spouse confidant (Cornwell, Laumann, & Schumm, 2008). Total contacts per year for each confidant were then summed, with higher responses indicating greater talk frequency with the identified network members. An inverse hyperbolic sine transformation was then applied to account for skewed responses. Social network size was measured by counting the number of non-partner/spouse individuals noted on the social network roster. Responses ranged from 0 to 5. Closeness to social network members was measured with one question that asked participants, “How close do you feel is your relationship with x?” Responses ranged from 1 (not very close) to 4 (extremely close), with higher responses indicating closer relationships. Using Cornwell et al.’s (2008) recommendations, average closeness of network members was created by adding together the closeness responses and dividing by the number of non-spousal network members mentioned.

Covariates

Covariates included education, age, race, ethnicity, marital happiness, and activities of daily living (ADLs) limitations. Age was measured as a continuous variable. Race was coded as 0 = White/Caucasian and 1 = Black/African American. Ethnicity was coded as 1 = Hispanic or Latino and 0 = Not Hispanic or Latino. Education was coded as 0 = high school degree/equivalent or less and 1 = associates degree or higher. Marital happiness was assessed by one question on a 1 (very unhappy) to 7 (very happy) scale. Five items were used to assess ADLs (Katz, 1983). The scale ranged from 0 (no difficulty) to 3 (unable to do), with higher scores indicating greater difficulties with ADLs. These items were added for a measure of ADLs and the scale indicated good internal consistency (α = .82).

Analytic plan

Actor–partner interdependence models (APIMs) in Mplus were used in the present study due to the dyadic nature of our primary research question (see Campbell & Kashy, 2002; Garcia, Kenny, & Ledermann, 2015; see Figure 1 for an example). A total of four APIMs were conducted. The four models included the following 3 items as independent variables in their respective model: (1) neighborhood social ties (see Figure 1), (2) support from family, (3) support from friends, and (4) social network characteristics. Associations between an individual’s reports of the previous three constructs and their own emotional well-being and SRH were examined (i.e., actor effects) along with how one’s partner’s reports of the previous 3 items were associated with an individual’s emotional well-being and SRH (i.e., partner effects). The maximum likelihood estimation with robust standard errors estimator was used in Mplus to account for non-normality among variables of interest and for missing data (Muthén & Muthén, 1998–2017).

Figure 1.

Sample APIM. APIM = actor–partner interdependence model.

Results

Before conducting the main analyses, we checked the measurement model to assess the latent variable of emotional well-being. All of the items (i.e., happiness, depressive symptoms, overall mental health, and loneliness) were significantly loaded on emotional well-being and a good model fit ensued (comparative fit index [CFI] = .96, standardized root mean square residual [SRMR] = .04; see Table 2 for the overall measurement model and the measurement model within each analysis). Race, ethnicity, level of ADLs, age, and education served as covariates. The non-significant results and covariates can be found along with the results in Table 3.

Table 2.

Standardized coefficients for the emotional well-being measurement model within each model.

	Female	Male
Overall measurement model
Happiness	.666***	.689***
Overall mental health	.641***	.640***
Depressive symptoms	.685***	.697***
Loneliness	.518***	.585***
Model 1: NST
Happiness	.586***	.552***
Overall mental health	.766***	.767***
Depressive symptoms	.609***	.563***
Loneliness	.344***	.282***
Model 2: Family support
Happiness	.558***	.544***
Overall mental health	.773***	.782***
Depressive symptoms	.562***	.610***
Loneliness	.279***	.336***
Model 3: Friend support
Happiness	.548***	.551***
Overall mental health	.790***	.796***
Depressive symptoms	.570***	.612***
Loneliness	.323***	.361***
Model 4: SN characteristics
Happiness	.560***	.546***
Overall mental health	.775***	.792***
Depressive symptoms	.562***	.609***
Loneliness	.316***	.339***

Note. NST = neighborhood social ties, SN = social network.

***p < .001.

Table 3.

Standardized coefficients for APIMs with NST, family and friend emotional support, and SN characteristics.

	Emotional well-being				Self-rated health
	Male		Female		Male		Female
	Actor effects	Partner effects¹	Actor effects	Partner effects²	Actor effects	Partner effects¹	Actor effects	Partner effects²
Model 1: NST
Age	−.044		−.041		−.065		−.089**
Black	−.045		−.055		−.020		−.112**
Hispanic	.008		−.104*		−.034		−.170**
ADLs	−.286**		−.326**		−.328**		−.284**
Education	.139**		.112*		.118**		.177**
Marital happiness	.206**		.177**		.058		.011
NST	.148**	.006	.122*	−.006	.069	−.020	.043	.056
Model 2: Family support
Age	−.050		−.045		−.041		−.084
Black	−.115		−.068		−.056		−.106
Hispanic	−.006		−.105		−.018		−.172
ADLs	−.305**		−.287**		−.312**		−.259**
Education	.112**		.108**		.127**		.166**
Marital happiness	.193**		.180**		.052		.023
Family support	.192**	.023	.169**	−.018	.077*	.016	.056	−.030
Model 3: Friend support
Age	−.047		.000		−.041		−.067*
Black	−.104*		−.056		−.043		−.095**
Hispanic	.002		−.079		−.020		−.164**
ADLs	−.296**		−.293**		−.317**		−.275**
Education	.099*		.124**		.112**		.170**
Marital happiness	.205**		.154**		.055		−.005
Friend support	.078	.043	.118**	.075	.042	.049	.045	.111**
Model 4: SN characteristics
Age	−.042		−.036		−.059		−.079*
Black	−.138**		−.082*		−.051		−.102**
Hispanic	−.013		−.106**		−.003		−.160**
ADLs	−.046		−.289**		.039		−.250**
Education	.099*		.113**		.112**		.163**
Marital happiness	.207**		.168**		.062		.000
Network size	.029	−.019	−.008	.074	.039	.049	.051	.016
Closeness to SN	.175**	.034	.090*	.067	.077*	.068*	.077*	.041
Talk frequency	.002	.016	.040	−.124*	−.049	−.080*	−.034	−.017

Note. APIM = actor–partner interdependence model, NST = neighborhood social ties, SN = social network, ADL = activity of daily living, CFI = Comparative Fit Index, RMSEA = root mean square error of approximation, SRMR = standardized root mean square residual. Model 1—CFI = .89, RMSEA = .05, SRMR = .06; Model 2—CFI = .90, RMSEA = .05, SRMR = .05; Model 3—CFI = .90, RMSEA = .05, SRMR = .05; Model 4—CFI = .89, RMSEA = .05, SRMR = .05.

¹The association between a female spouse’s social connectedness factor and her male spouse’s well-being.

²The association between a male spouse’s social connectedness factor and his female spouse’s well-being.

*p < .05; **p < .01.

Neighborhood social ties

An APIM was conducted examining both partners’ neighborhood social ties and well-being outcomes (see Table 3). The model had acceptable fit (CFI = .89, root mean square error of approximation (RMSEA) = .05, SRMR = .06). Actor effects for the positive association between neighborhood social ties and emotional well-being emerged for both men (β = .148, p = .006) and women (β = .122, p = .010). No partner effects emerged for emotional well-being and no actor or partner effects emerged for self-rated health.

Support from family and friends

Next, we conducted two APIMs to examine the association between (1) both partners’ family support and well-being outcomes and (2) both partners’ friend support and well-being outcomes (see Table 3). The models for family support (CFI = .90, RMSEA = .05, SRMR = .05) and friend support (CFI = .90, RMSEA = .05, SRMR = .05) had acceptable fit.

For the model that included family support, husbands’ family support was positively associated with their own emotional well-being (β = .192, p < .001) and self-rated health (β = .077, p = .019). Wives’ greater family support was associated with their own greater emotional well-being (β = .169, p < .001). No partner effects emerged for family emotional support.

For friend support, wives’ greater friend support was associated with their own higher levels of emotional well-being (β = .118, p = .005). Husbands’ greater levels of friendship support was associated with wives’ higher levels of self-rated health (β = .111, p = .001); this was the only partner effect that was found.

Social network characteristics

The final APIM examined the association between each partner’s social network characteristics (i.e., talk frequency, closeness, and social network size) and well-being (i.e., self-rated health, emotional well-being; see Table 3). The model had acceptable fit (CFI = .89, RMSEA = .05, SRMR = .05). Husbands’ and wives’ greater social network closeness was associated with their own higher levels of emotional well-being (β = .175, p < .001 for husbands and β = .090, p = .034 for wives). Husbands’ greater talk frequency was associated with wives’ lower levels of emotional well-being (β = −.124, p = .021). No other actor or partner effects emerged for wives or husbands for emotional well-being as the outcome. Husbands’ greater closeness to their own social network was positively associated with their own self-rated health (β = .077, p = .027). Wives’ social network closeness was positively associated with their own self-rated health (β = .077, p = .019) and also was positively associated with husbands’ self-rated health (β = .068, p = .041). Wives’ greater talk frequency was associated with husbands’ poorer self-rated health (β = −.080, p = .045).

Discussion

The present study assessed whether one’s own and one’s spouse’s social network variables are associated with emotional well-being and self-rated health among older adult married couples. Maintaining social ties outside of marriage is positively associated with individual well-being, although the findings for spousal well-being are more mixed. The present study builds on prior literature by assessing how multiple aspects of social support, above and beyond the spousal relationship, are associated with well-being at both the individual and the dyadic level.

General findings for actor effects

Our first research question explored how an individual’s social connectedness is associated with their self-rated health and emotional well-being. Neighborhood social ties, family support, and closeness to one’s social network were associated with higher levels of emotional well-being for both husbands and wives via actor effects, providing some support for the main effects hypothesis (Cohen & Wills, 1985). Even though social support received by neighbors has generally declined over time (e.g., McPherson, Smith-Lovin, & Brashears, 2006), neighbors still appear to play a role in older adults’ lives, which aligns with prior studies (e.g., Wrzus et al., 2013). Since married couples living together share neighbors, they may be experiencing similar levels of benefits from neighbors. The SST lends support for the reasons why we did not find any individual associations between social network size for either women or men; the number of social network members may not be as relevant later in life (Lang & Carstensen, 1994). Further, the SCM would posit that those who are closest to us, such as family members and those we see on a daily basis (i.e., neighbors), may exert greater effects on our overall well-being than more distant ties (Antonucci & Akiyama, 1987).

General findings for partner effects

The second research question addressed partner effects. These findings show how spouses’ social connectedness factors are, and are not, associated with a partner’s well-being later in life. Husbands’ greater talk frequency with their social network was negatively associated with their wives’ emotional well-being; wives’ greater talk frequency with their social network was negatively associated with their husbands’ self-rated health. Interestingly, talk frequency only had associations with well-being in terms of partner effects—no actor effects were found—and in both cases the associations with spouses’ outcomes were negative. Interdependence would suggest that individuals may be particularly susceptible to the frequency of social contact engaged in by a spouse. Thus, the type of contact individuals are engaging in with social network members should be explored further. For example, husbands might engage with their social networks to a greater degree when their wives’ emotional well-being is compromised because their social engagement needs are not being met within the marriage. This may be the case as wives are typically viewed as “kin-keepers” within a marriage (e.g., Hagestad, 1986). Because these associations are cross-sectional, it is equally likely that spouses’ poorer well-being is prompting husbands and wives to engage in greater levels of social contact outside the marriage. For example, wives could have greater contact with social network members regarding caregiving when a husband is in poorer health, especially since women are more likely to provide caregiving than men (e.g., National Alliance for Caregiving, 2015).

Wives’ greater closeness to their social network was positively associated with their husbands’ self-rated health, and husbands’ greater friend support was positively associated with wives’ self-rated health. Again, the cross-sectional nature of the study design does not allow for causal ordering between these variables, but this pattern of findings suggests that individuals’ health is related to certain types of social connectedness on the part of spouses. Having a wife who is close to her social network or a husband who receives adequate levels of support from friends could allow or encourage spouses to engage in better self-care practices because some of their spouses’ needs are being met outside the marriage. Alternatively, having a healthy spouse might create an environment where one is free to engage in and receive benefits from social relationships. This largely aligns with the concept of spillover that is often applied to work–family balance (e.g., Heller & Watson, 2005) and exemplifies how social interactions outside of a marriage may “spillover” into the marriage.

The role of gender

The third research question pertained to gender similarities and differences, and our findings generally show that there are more similarities than differences among men and women. The present study found that men and women both reported associations between their own family ties and emotional well-being; however, only men reported associations between their own family ties and self-rated health. These mixed findings for family and well-being later in life align with prior studies which show that men benefit more from family support than do women (e.g., Cable et al., 2013). Family is considered an obligatory relationship that married men may feel more obliged to maintain compared to other relationships. Wives may help to maintain these family ties for men on the basis that family relationships are more obligatory in nature and due to marriage creating shared family ties (e.g., Hagestad, 1986). More work should be done to fully disentangle gender differences in the associations between family and well-being.

Of the differences, one notable difference is that friendship support is associated with emotional well-being for women, whereas this is not the case for men. This may stem from gendered norms surrounding friendship, as women are more likely to report wanting to have intimacy in their friendships as compared to men (e.g., Fehr, 2004) and the present findings show that these gendered expectations may carry into late adulthood. Intimacy is associated with well-being (Reis, Sheldon, Gable, Roscoe & Ryan, 2000), which may be why women experience ties between support from friends and emotional well-being. The finding that husbands’ friendship is associated with wives’ self-rated health could signal that a wife is in well enough health to help a husband to maintain his friendship network. Among those in opposite sex relationships, married men’s social networks are more likely to involve wives, which is why wives’ health might be important (e.g., Davidson, 2004; Hagestad, 1986; Umberson, Thomeer, & Lodge, 2015).

Considerations and implications

Programming for older adults could promote avenues of social interaction for older adults as interaction is associated positively with self-rated health and the holistic measure of well-being used here. The integration of older adults into the social environmental of neighborhoods is a promising avenue for future interventions based on the present study. Interventions may consider how neighborhood gatherings promote interactions between neighbors, which may be especially beneficial for older adults. Moreover, neighborhood structure should also be considered. Those who live in more walkable areas with shared spaces are more likely to know and converse with their neighbors, which translates into increased social capital (Leyden, 2011; Richard, Gauvin, Gosselin, & Laforest, 2009). Neighbors may be associated with emotional well-being due to looking out for each other, providing a sense of belongingness and community, and they may decrease loneliness by increasing support. This support may not only help to deter loneliness, but may be beneficial in cases where an individual needs help with daily tasks (e.g., Cramm, van Dijk, & Nieboer, 2013). Examining neighborhood factors at the dyadic level is a fruitful area for future research.

Friendship is another area to consider for programming, especially for women, due to the associations between friendship support, emotional well-being, and self-rated health. Interventions, including the Friendship Enrichment Program, have previously demonstrated increases in the creation of new friendships and improvements in current friendships (Stevens, Martina, & Westerhof, 2006). Delivering such programs, perhaps by community-based organizations, may prove helpful for older adults.

Strengths and limitations

Strengths of the study include using a nationally representative data set and examining a variety of social connectedness measures. The use of dyadic data enabled us to examine how spouses are interdependent with one another, garnering additional support for interdependence theory (Kelley, 1983). This allowed us to gain a better understanding of how specific aspects of individuals’ social support are, and are not, interrelated with spouses’ well-being. Additionally, the study examined couples’ social connectedness factors outside of marriage all while controlling for marital happiness and demographic variables.

Measures pertaining to dimensions of social networks were limited to a maximum of five people outside of the romantic relationship, which may have created an artificial ceiling on the individuals about whom social network characteristic items were asked. Questions on friend and family support asked about general relationships with family and friends, rather than about a specific relationship. Caution should be used in interpreting the findings since the data are cross-sectional; we cannot tell how relationships unfold over time or how relationships may be associated with well-being at a later time point. In the future, longitudinal studies could shed light on how health and social ties evolve over time within couples. Future research may also want to consider why an individual is receiving such support as different types of support may differentially impact well-being factors, especially considering how individuals’ greater talk frequency was associated with poorer well-being outcomes for spouses.

Conclusion

This series of APIMs showed that aspects of both one’s own, and a partner’s, social network characteristics are associated with well-being. Individuals should particularly be cognizant of fostering close relationships and supportive interactions with both family members and neighbors, considering all were positively associated with emotional well-being for men and women. Family relationships and neighborhood relationships are relationships that are often fostered over extended periods of time, highlighting how emphasis should be placed on these two relationships throughout the life course. For wives, friendship is a particularly salient relationship and has positive associations with well-being. This set of analyses was guided by interdependence theory and lent some support for the notion that an individual’s well-being may be associated with their own and a spouses’ social connectedness. The present study provides evidence for how social support above and beyond the spousal dyad is necessary to consider and should be examined in a time period where emphasis is often placed on the spousal relationship.

Footnotes

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Ashley E. Ermer

Open research statement

As part of IARR’s encouragement of open research practices, the author(s) have provided the following information: The data used in the research are available. The data can be obtained at: https://www.icpsr.umich.edu/icpsrweb/NACDA/studies/34921/datadocumentation. The materials used in the research are available. The materials can be obtained at:

References

Antonucci

T. C.

Akiyama

(1987). Social networks in adult life and a preliminary examination of the convoy model. Journal of Gerontology, 42, 519–527. doi:10.1093/geronj/42.5.519

Antonucci

T. C.

Lansford

J. E.

Akiyama

. (2001). Impact of positive and negative aspects of marital relationships and friendships on well-being of older adults. Applied Developmental Science, 5, 68–75. doi:10.1207/S1532480XADS0502_2

Beach

S. R. H.

Katz

Kim

Brody

G. H.

(2003). Prospective effects of marital satisfaction on depressive symptoms in established marriages: A dyadic model. Journal of Social and Personal Relationships, 20, 355–371. doi:10.1177/0265407503020003005

Bishop

A. J.

Martin

Poon

(2007). Happiness and congruence in older adulthood: A structural model of life satisfaction. Aging and Mental Health, 10, 445–453. doi:10.1080/13607860600638388

Cable

Bartley

Chandola

Sacker

(2013). Friends are equally important to men and women, but family matters more for men’s well-being. Journal of Epidemiology and Community Health, 67, 166–171. doi:10.1136/jech-2012-201113

Callahan

C. M.

Wolinsky

F. D.

(1994). The effect of gender and race on the measurement properties of the CES-D in older adults. Medical Care, 32, 341–356. doi:10.1097/00005650-199404000-00003

Campbell

Kashy

D. A.

(2002). Estimating actor, partner, and interaction effects for dyadic data using PROC MIXED and HLM: A user-friendly guide. Personal Relationships, 9, 327–342. doi:10.1111/1475-6811.00023

Cohen

Wills

T. A.

(1985). Stress, social support, and the buffering hypothesis. Psychological Bulletin, 98, 310. doi:10.1037/0033-2909.98.2.310

Cornwell

Laumann

E. O.

Schumm

L. P.

(2008). The social connectedness of older adults: A national profile. American Sociological Review, 73, 185–203. doi:10.1177/000312240807300201

10.

Cornwell

Schumm

L. P.

Laumann

E. O.

Kim

(2014). Assessment of social network change in a longitudinal survey. Journal of Gerontology: Psychological and Social Sciences, 69, S75–S82. doi:10.1093/geronb/gbu037

11.

Cornwell

E. Y.

Waite

L. J.

(2009). Social disconnectedness, perceived isolation, and health among older adults. Journal of Health and Social Behavior, 50, 31–48. doi:10.1177/002214650905000103

12.

Cramm

J. M.

van Dijk

H. M.

Nieboer

A. P.

(2013). The importance of neighborhood social cohesion and social capital for the well-being of older adults in the community. The Gerontologist, 53. doi:10.1093/geront/gns052

13.

Curran

S. R.

McLanahan

Knab

(2003). Does remarriage expand perceptions of kinship support among the elderly? Social Science Research, 32, 171–190. doi:10.1016/S0049-089X(02)00046-7

14.

Davidson

(2004). “Why can’t a man be more like a woman?” Marital status and social networking of older men. The Journal of Men’s Studies, 13, 25–43. doi:10.3149/jms.1301.25

15.

Donato

K. M.

León-Pérez

Wallston

K. A.

Kripalani

(2018). Something old, something new: When gender matters in the relationship between social support and health. Journal of Health and Social Behavior, 59, 352–370.

16.

Fehr

(2004). Intimacy expectations in same-sex friendships: A prototype interaction-pattern model. Journal of Personality and Social Psychology, 86, 265–284. doi:10.1037/0022-3514.86.2.265

17.

Felmlee

Sprecher

(2000). Close relationships and social psychology: Intersections and future paths. Social Psychology Quarterly, 63, 365–376. doi:10.2307/2695846

18.

Fingerman

K. L.

Hay

H. L.

Birditt

K. S.

(2004). The best of ties, the worst of ties: Close, problematic, and ambivalent social relationships. Journal of Marriage and Family, 66, 792–808. doi:10.1111/j.0022-2445.2004.00053.x

19.

Garcia

R. L.

Kenny

D. A.

Ledermann

(2015). Moderation in the actor-partner interdependence model. Personal Relationships, 22, 8–29. doi:10.1111/pere.12060

20.

Golden

Conroy

R. M.

Lawlor

B. A.

(2009). Social support network structure in older people: Underlying dimensions and association with psychological and physical health. Psychology, Health and Medicine, 14, 280–290. doi:10.1080/13548500902730135

21.

Hagestad

G. O.

(1986). The family: Women and grandparents as kin-keepers. In Pifer

A. J.

Bronte

(Eds.), Our aging society: Paradox and promise (pp. 141–160). New York, NY: W.W. Norton & Company.

22.

Heller

Watson

(2005). The dynamic spillover of satisfaction between work and marriage: The role of time and mood. Journal of Applied Psychology, 90, 1273–1290. doi:10.1037/0021-9010.90.6.1273

23.

Helliwell

J. F.

Putnam

R. D.

(2004). The social context of well–being. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences, 359, 1435–1446. doi:10.1098/rstb.2004.1522

24.

Hughes

M. E.

Waite

L. J.

(2009). Marital biography and health at mid-life. Journal of Health and Social Behavior, 50, 344–358. doi:10.1177/002214650905000307

25.

Hughes

M. E.

Waite

L. J.

Hawkley

L. C.

Caccioppo

J. T.

(2004). A short scale for measuring loneliness in large surveys: Results from two population-based studies. Research on Aging, 26, 655–672. doi:10.1177/0164027504268574

26.

Huxhold

Miche

Schuz

(2014). Benefits of having friends in older ages: Differential effects of informal social activities on well-being in middle-aged and older adults. Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 69, 366–375. doi:10.1093/geronb/gbt029

27.

Jackson

G. L.

Kennedy

Bradbury

T. N.

Karney

B. R.

(2014). A social network comparison of low-income Black and White newlywed couples. Journal of Marriage and Family, 76, 967–982. doi:10.1111/jomf.12137

28.

Kalmijn

(2003). Shared friendship networks and the life course: An analysis of survey data on married and cohabiting couples. Social Networks, 25, 231–249. doi:10.1016/S0378-8733(03)00010-8

29.

Katz

(1983). Assessing self-maintenance: Activities of daily living, mobility, and instrumental activities of daily living. Journal of the American Geriatrics Society, 31, 721–727. doi:10.1111/j.1532-5415.1983.tb03391.x

30.

Kelley

H. H.

(1983). Love and commitment. In Jones

W. H.

Perlman

D. W.

(Eds.), Advances in personal relationships (pp. 117–143). New York: Freeman.

31.

Lang

F. R.

Carstensen

L. L.

(1994). Close emotional relationships in late life: Further support for proactive aging in the social domain. Psychology and Aging, 9, 315–324.

32.

Leyden

K. M.

(2011). Social capital and the built environment: The importance of walkable neighborhoods. American Journal of Public Health, 93, 1546–1551. doi:10.1037//0882-7974.9.2.315

33.

Lindau

S. T.

Gavrilova

(2010). Sex, health, and years of sexually active life gained due to good health: Evidence from two US population based cross sectional surveys of ageing. BMJ, 340, c810. doi:10.1136/bmj.c810

34.

Löckenhoff

C. E.

Chan

McCrae

R. R.

De Fruyt

Jussim

De Bolle

… Nakazato

(2014). Gender stereotypes of personality: Universal and accurate? Journal of Cross-Cultural Psychology, 45, 675–694. doi:10.1177/0022022113520075

35.

Lyyra

Heikkinen

(2006). Perceived social support and mortality in older adults. The Journals of Gerontology: Series B, 61, S147–S152. doi:10.1093/geronb/61.3.S147

36.

McPherson

Smith-Lovin

Brashears

M. E.

(2006). Social isolation in America: Changes in core discussion networks over two decades. American Sociological Review, 71, 353–375. doi:10.1177/000312240607100301

37.

Muthén

L. K.

Muthén

B. O

. (1998–2017). Mplus user’s guide (8th ed). Los Angeles, CA: Muthén & Muthén.

38.

National Alliance for Caregiving and AARP. (2015). Caregiving in the US. Retrieved from https://www.caregiving.org/wp-content/uploads/2015/05/ 2015_Caregiving intheUS_Final-Report-June-4_WEB.pdf

39.

Pienta

A. M.

Hayward

M. D.

Jenkins

K. R.

(2000). Health consequences of marriage for the retirement years. Journal of Family Issues, 21, 559–586. doi:10.1177/019251300021005003

40.

Reis

H. T.

Sheldon

K. M.

Gable

S. L.

Roscoe

Ryan

R. M.

(2000). Daily well-being: The role of autonomy, competence, and relatedness. Personality and Social Psychology Bulletin, 26, 419–435. doi:10.1177/0146167200266002

41.

Richard

Gauvin

Gosselin

Laforest

(2009). Staying connected: Neighborhood correlates of social participation among older adults living in an urban environment in Montreal, Quebec. Health Promotion International, 24, 46–57. doi:10.1093/heapro/dan039

42.

Stevens

N. L.

Martina

Westerhof

G. J.

(2006). Meeting the need to belong: Predicting effects of a friendship enrichment program for older women. The Gerontologist, 46, 495–502. doi:10.1093/geront/46.4.495

43.

Stevens

D. P.

Minnotte

K. L.

Mannon

S. E.

Kiger

(2007). Examining the “neglected side of the work-family interface”: Antecedents of positive and negative family-to-work spillover. Journal of Family Issues, 28, 242–262. doi:10.1177/0192513X06294548

44.

Strine

T. W.

Chapman

D. P.

Balluz

L. S.

Mokdad

A. H.

(2008). Health-related quality of life and health behaviors by social and emotional support. Social Psychiatry and Psychiatric Epidemiology, 43, 151–159. doi:10.1007/s00127-007-0277-x

45.

Thomas

P. A.

(2016). The impact of relationship-specific support and strain on depressive symptoms across the life course. Journal of Aging and Health, 28, 363–382. doi:10.1177/0898264315591004

46.

Umberson

Thomeer

M. B.

Lodge

A. C.

(2015). Intimacy and emotion work in lesbian, gay, and heterosexual relationships. Journal of Marriage and Family, 77, 542–556. doi:10.1111/jomf.12178

47.

Waite

L. J.

Cagney

Dale

Huang

Laumann

E. O

McClintock

M. K.

… Cornwell

(2014). National social life, health, and aging project (NSHAP): Wave 2 and partner data collection.

48.

Waite

L. J.

Harrison

S. C.

(1992). How women in mid-life allocate social contacts among kith and kin. Social Forces, 70, 637–654. doi:10.1093/sf/70.3.637

49.

Walen

H. R.

Lachman

M. E.

(2000). Social support and strain from partner, family, and friends: Costs and benefits for men and women in adulthood. Journal of Social and Personal Relationships, 17, 5–30. doi:10.1177/0265407500171001

50.

Williams

Umberson

(2004). Marital status, marital transitions, and health: A gendered life course perspective. Journal of Health and Social Behavior, 45, 81–98. doi:10.1177/002214650404500106

51.

Wrzus

Hanel

Wagner

Neyer

F. J.

(2013). Social network changes and life events across the life span: A meta-analysis. Psychological Bulletin, 139, 53–80. doi:10.1037/a0028601