Social support, social strain, loneliness, and well-being among older adults

Abstract

This study proposed that, among older adults, higher support and lower strain received from each of the four relational sources (spouse/partner, children, family, and friends) were associated with reduced loneliness and improved well-being and that loneliness might mediate the relationship between support/strain and well-being. Structural equation modeling was conducted using a national sample of adults aged 50 years and older (N = 7,367) from the Health and Retirement Study. Findings indicated that support from spouse/partner and friends alleviated loneliness, while strain from all the four sources intensified loneliness; higher support and lower strain from various sources directly and indirectly improved well-being, with indirect effects mediated through reduced loneliness. It was concluded that, in later life, various sources of support/strain engender distinct effects on loneliness and well-being, and loneliness serves as one of the psychological pathways linking support/strain to well-being.

Full text

http://spr.sagepub.com/
Relationships
Journal of Social and Personal
http://spr.sagepub.com/content/31/2/141
The online version of this article can be found at:
DOI: 10.1177/0265407513488728
May 2013
2014 31: 141 originally published online 31Journal of Social and Personal Relationships
Yixin Chen and Thomas Hugh Feeley
An analysis of the Health and Retirement Study*
Social support, social strain, loneliness, and well-being among older adults:
Published by:
http://www.sagepublications.com
On behalf of:
International Association for Relationship Research
can be found at:Journal of Social and Personal RelationshipsAdditional services and information for
http://spr.sagepub.com/cgi/alertsEmail Alerts:
http://spr.sagepub.com/subscriptionsSubscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
What is This?
- May 31, 2013OnlineFirst Version of Record
- Feb 11, 2014Version of Record >>
at University at Buffalo Libraries on February 11, 2014spr.sagepub.comDownloaded from at University at Buffalo Libraries on February 11, 2014spr.sagepub.comDownloaded from

Article
Social support, social
strain, loneliness, and
well-being among older
adults: An analysis
of the Health and
Retirement Study*
Yixin Chen
Thomas Hugh Feeley
University at Buffalo, The State University of New York, USA
Abstract
This study proposed that, among older adults, higher support and lower strain received
from each of the four relational sources (spouse/partner, children, family, and friends)
were associated with reduced loneliness and improved well-being and that loneliness
might mediate the relationship between support/strain and well-being. Structural equa-
tion modeling was conducted using a national sample of adults aged 50 years and older
(N¼7,367) from the Health and Retirement Study. Findings indicated that support from
spouse/partner and friends alleviated loneliness, while strain from all the four sources
intensified loneliness; higher support and lower strain from various sources directly and
indirectly improved well-being, with indirect effects mediated through reduced loneli-
ness. It was concluded that, in later life, various sources of support/strain engender dis-
tinct effects on loneliness and well-being, and loneliness serves as one of the
psychological pathways linking support/strain to well-being.
Keywords
Interpersonal communication, loneliness, older adults, social contact, social interactions,
social strain, social support, well-being
Corresponding author:
Yixin Chen, Department of Communication, University at Buffalo, Buffalo, NY 14260, USA.
Email: yixinche@buffalo.edu
*An earlier version of this article received Top Paper Award from the Health Communication Division of the
2012 Conference of the International Communication Association.
Journal of Social and
Personal Relationships
2014, Vol. 31(2) 141–161
ªThe Author(s) 2013
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0265407513488728
spr.sagepub.com
J S P R
at University at Buffalo Libraries on February 11, 2014spr.sagepub.comDownloaded from

The United States is facing an aging population, with individuals aged 50 and older
numbering over 99 million (i.e., 32% of the total U.S. population) (U.S. Census Bureau,
2012). Considering that a large number of Americans have entered older adulthood, it is
essential to explore psychosocial determinants that may impact this population’s well-
being, an important health outcome involving optimal experience and functioning (Ryan
& Deci, 2001). One unique risk factor relating to older adults’ well-being is loneliness—
one of the most painful of all human experiences, and a pervasive one among the elderly,
with about 17%of Americans aged 50 years and older reporting feelings of loneliness
(Cacioppo, Hawkley, & Thisted, 2010; Cacioppo, Hughes, Waite, Hawkley, & Thisted,
2006; Sullivan, 1953; Theeke, 2010).
A number of factors have been found to influence loneliness and well-being in later
life, such as levels of social support (i.e., positive social interactions) and social strain
(i.e., negative social interactions) in social relationships (Newsom, Nishishiba, Morgan,
& Rook, 2003; Newsom, Rook, Nishishiba, Sorkin, & Mahan, 2005; Shiovitz-Ezra &
Leitsch, 2010). The aging process is often accompanied by a decline in physical and cog-
nitive functions and a loss of social network members. As a result, this may increase
older adults’ need for social support and vulnerability to social strain, and it may also
trigger or contribute to feelings of loneliness. Previous studies have examined either the
unique impact of loneliness (e.g., Cacioppo et al., 2006, 2010) or the joint influence of
social support and social strain on health outcomes among older adults (e.g., Newsom
et al., 2003, 2005). However, the mechanisms by which these three factors function
together to affect well-being in later life are still being tested, and the various sources
of support and strain of older adults have not been examined simultaneously. The present
study has two aims: the first is to examine the unique effects of social support and social
strain received from various sources on loneliness and well-being among older adults;
the second is to explore whether loneliness could account for the association between
social support/social strain and well-being in later life.
Loneliness and well-being
Loneliness is a state of emotional distress accompanying perceived deficiencies in the
quantity and/or quality of one’s social relationships (Peplau & Perlman, 1982). Recently,
Cacioppo, Hawkley, and colleagues articulated a theory of loneliness and health, arguing
that loneliness hasa unique and detrimentaleffect on physicaland psychological health (e.g.,
Hawkley & Cacioppo, 2010). Studieson the aging population have generally supported this
theory. For example,greater loneliness has been linked to increased systolic blood pressure
(Hawkley, Thisted, Masi, & Cacioppo,2010), increased depression (Cacioppo et al., 2010),
and poorer physical health (Cornwell & Waite, 2009) among older adults.
Much research on the impact of loneliness in later life has been devoted to negative
health outcomes, but overlooked positive psychological health outcomes, such as well-
being (e.g., Cacioppo et al., 2006, 2010). There are many conceptualizations of well-
being in the extant literature. The current study focuses on subjective well-being,
defined as individuals’ self-evaluation of their lives, as subjective well-being is consid-
ered an essential element of positive psychological health (Diener, Sapyta, & Suh,
142 Journal of Social and Personal Relationships 31(2)
at University at Buffalo Libraries on February 11, 2014spr.sagepub.comDownloaded from

1998). Studies have reported loneliness to be negatively associated with both health-
related quality of life (Liu & Guo, 2007) and subjective well-being among older adults
(Windle & Woods, 2004).
Social contact and loneliness
Loneliness may vary as a function of the amount of social contact one has (Gross, Juvonen,
& Gable, 2002). Social contact is defined as individuals’ daily social interactions (Jones,
1981). Major sources of social contact are marriage, partnership, family, and friendships
(Berkman & Syme, 1994). Studies in the 1990s found that a low frequency of social
contact with family and friends was associated with greater loneliness among the elderly
(e.g., Bondevik & Skogstad, 1998; Holme´n, Ericsson, Andersson, & Winblad, 1992;
Mullins & Dugan, 1990). More recent research found that the frequency of contact with
children and friends was not significantly related to loneliness among older adults (e.g.,
Routasalo, Savikko, Tilvis, Strandberg, & Pitka¨la¨, 2006). These contradictory findings
suggest that it may be premature to infer a causal relationship between social contact and
loneliness in the aging population. It is possible that loneliness experienced in later life is
not due to a low frequency of social contact but due to perceived lack of interpersonal
intimacy or low quality of social relationships. The current study treats social contact as a
control variable and operationalizes it as older adults’ frequency of contact with children,
family members, and friends.
Social support and social strain
The lack of intimate social contacts is more likely to induce loneliness than the lack of
regular social contacts for older adults (Green, Richardson, Lago, & Shatten-Jones,
2001). Such intimate social contacts are important sources of social support in later life.
Social support from an interpersonal communication perspective is understood as sup-
portive behavior performed for an individual by others and is often assessed by an indi-
vidual’s perception of received support (Burleson & MacGeorge, 2002; Goldsmith,
2004). Received support can be further categorized into emotional support, informational
support, and instrumental support (House, 1981). Emotional support has been reported to
be especially consequential, with significant physical and psychological health outcomes
(Burleson, 2003). Thus, the current study operationalizes social support as older adults’
perception of emotional support (i.e., positive social interactions) received from their
social network members (Newsom et al., 2003, 2005).
Research on social support in later life examines positive social interactions, which
could be considered as one dimension of social relationships (Bengtson, Giarrusso,
Mabry, & Silverstein, 2002). It should be noted that not all social relationships are
beneficial and pleasant and that frequent social contact may actually increase the
chances of conflicts, disputes, or strained relations (i.e., negative social interactions).
Such negative social interactions among social relationships are referred to as social
strain (Shiovitz-Ezra & Leitsch, 2010), social negativity (Bertera, 2005), or negative
social exchanges (Newsom et al., 2003, 2005). Studies on the aging population have
recommended that the independent effects of positive and negative social interactions
Chen and Feeley 143
at University at Buffalo Libraries on February 11, 2014spr.sagepub.comDownloaded from

be examined simultaneously (Newsom et al., 2003; Shiovitz-Ezra & Leitsch, 2010).
Thus, the current study treats social strain as an important dimension of social relation-
ships and operationalizes it as older adults’ perception of negative social interactions
received from their social network members (Newsom et al., 2003, 2005).
Social support and social strain represent unique dimensions of social relationships
(Bengtson et al., 2002). One would expect that social support and social strain are
complementary, so that high levels of one imply low levels of the other; however,
relationships can paradoxically embrace high support and high strain or low support and
low strain. In any given relationship, an individual can feel loved, understood, or cared
for, and also feel rejected, criticized, or ignored. This love–hate dynamic is often evident
within families and has been captured by the family solidarity–conflict model proposed
by Bengtson, Rosenthal, and Burton (1995).
The solidarity–conflict model
The solidarity–conflict model contends that intergenerational relationships consist of
seven dimensions—association, affect, consensus, function, norms, family structure, and
conflict—with the first six classified as solidarity (Bengtson et al., 1995). This frame-
work emphasizes that intergenerational relationships are multidimensional (Silverstein
& Bengtson, 1997) and that each of the seven dimensions is distinct (Bengtson et al.,
2002). The model was later challenged by Luescher and Pillemer (1998), who argued
that the concept of ambivalence—mixed and contradictory feelings toward a relation-
ship—is an alternative and a more useful perspective for studying intergenerational
relations. The ambivalence concept was viewed by Bengtson et al. (2002) as com-
plementing rather than competing with the solidarity–conflict model.
Although the solidarity–conflict model with the addition of the ambivalence concept
has mostly been used for studying relationships between generations, such as older
parents’ relations with their adult children (e.g., Lowenstein, 2007), it is also applicable
for the wider social network of older adults, such as the relationship with a spouse, a family
member, or a friend. Similar to intergenerational relationships, any social relationship of
older adults can potentially incorporate solidarity, conflict, and ambivalence. There is
empirical evidence showing that solidarity, conflict, and ambivalence coexist among older
adults’ family and nonfamily relationships (Fingerman, Hay, & Birditt, 2004).
It is beyond the scope of the present study to incorporate all dimensions delineated by
the paradigms of solidarity–conflict and ambivalence. However, three important
dimensions from these paradigms—association (frequency of social contact), affect
(feelings of emotional intimacy), and conflict (feelings of tension or criticism)—cor-
responding to social contact (treated as a control variable), social support, and social
strain separately are included in the present study. Ambivalence is not considered in the
current study, as the concept of this construct needs to be refined and it is difficult to
measure (Luescher, 2004; Rappoport & Lowenstein, 2007).
Effects of social support and social strain on loneliness
Previous studies involving social support and loneliness in the aging population suggest
that social support is generally associated with lower loneliness, if social support is
144 Journal of Social and Personal Relationships 31(2)
at University at Buffalo Libraries on February 11, 2014spr.sagepub.comDownloaded from

assessed by a global measurement without differentiating sources of support (e.g.,
Cacioppo et al., 2010). However, findings are mixed when different sources of support
are delineated. For example, Shiovitz-Ezra and Leitsch (2010) reported that social sup-
port from family was a negative predictor of loneliness, while social support from friends
was not a predictor of loneliness, in a sample of adults aged 57–85 years. Stevens and
Westerhof (2006) found that only social support from partner and friends was signifi-
cantly related to lower levels of loneliness, whereas social support from family members
was unrelated, among a sample of adults aged 40–85 years. These findings suggest sup-
port from different sources may exhibit different effects on loneliness, and it appears
necessary to differentiate sources of support in research on older adults.
While research has established that social support protects older adults from lone-
liness (e.g., Cacioppo et al., 2010), only two studies have examined the link between
social strain and loneliness. A study by Stevens and Westerhof (2006) found that neg-
ative interactions with one’s partner and non-kin were related to higher levels of lone-
liness. A later study by Shiovitz-Ezra and Leitsch (2010) reported that social strain
from family was positively related to loneliness, but social strain from friends was unre-
lated. Taken together, these findings indicate that social strain from different sources
may yield different effects on loneliness, and higher social strain from any specific
source is likely to create a greater sense of loneliness for older adults. The current study
sets out to address this possibility.
Effects of social support and social strain on well-being
It is proposed that the effects of different sources of social support/social strain on well-
being be examined independently, as each type of relationship may exhibit distinct influ-
ences on critical outcomes. For example, the spousal relationship generally has centrality
status among all possible social relationships. In the support network of older adults, the
spouse is usually the preferred source of social support, if available (Cantor, 1979).
Among older adults, the spousal relationship was also reported as the most positive and
negative, and its quality was more strongly related to well-being than the quality of rela-
tionships with family or friends (Antonucci, Lansford, & Akiyama, 2001). In addition,
the nature of marriage and kinship connections gives relationships with spouse and chil-
dren an involuntary character, which can create contradictory sentiments (e.g., affection
vs. conflict) in these relationships (Hogerbrugge & Komter, 2012). A national longitu-
dinal study has reported that low social support and high social strain from spouse or
child were associated with increased mortality among middle-aged and older adults
(Birditt & Antonucci, 2008).
In contrast to family relationships, a defining feature of friendships is their voluntary
character, such that individuals have the option to withdraw from an unsatisfying
friendship (Lawton, Silverstein, & Bengtson, 1994). In fact, some extended family
relationships may also exhibit voluntary character: individuals may discontinue a rela-
tionship with an extended family member if the relationship is not perceived as bene-
ficial. Thus, friendships and extended family relationships are less likely to exhibit
conflicts or tensions (examples of social strain), compared to family relationships
with spouse or children. There is evidence that voluntary support from friends such as
Chen and Feeley 145
at University at Buffalo Libraries on February 11, 2014spr.sagepub.comDownloaded from

companionship can improve daily well-being in later life (Sherman, de Vries, & Lans-
ford, 2000). Taken together, there is compelling reason to differentiate and examine the
various sources of social support/social strain when studying the well-being of older
adults.
Loneliness as a mediating factor
The exact mechanisms by which social support, social strain, and loneliness jointly
influence well-being in older adulthood remain unclear. Berkman, Glass, Brissette, and
Seeman (2000) presented a conceptual model of how social networks impact health,
arguing that it is ‘‘a cascading causal process beginning with the macro-social to psycho-
biological processes that are dynamically linked together’’ (p. 846). Their conceptual
model identifies psychological states and traits (e.g., self-esteem and self-efficacy) as
one of the proximate pathways through which social support influences health status.
Berkman et al.’s (2000) model does not list loneliness as a possible psychological
pathway, but there is empirical evidence that social support influences health through
loneliness. For example, Stroebe, Stroebe, Abakoumkin, and Schut’s (1996) study
indicated that the impact of social support on lower psychological symptoms (depression
and somatic complaints) was partially mediated by reduced loneliness. Two studies by
Segrin and Domschke (2011) and Segrin and Passalacqua (2010) found the relationship
between social support and improved health was fully mediated through decreased
loneliness. These findings thus complemented Berkman et al.’s (2000) model, indicating
that loneliness may be one of the psychological pathways through which social support
affects physical or mental health.
Although Berkman et al.’s (2000) conceptual model does not incorporate social
strain, the authors remarked that not all relationships are positive and that social rela-
tionships may have powerful impacts on health through acts of abuse, violence, and
trauma (examples of negative social interactions). A recent study by Fiori and Consedine
(2013) on first-year college students reported that effects of positive (or negative) social
exchanges on positive (or negative) emotional well-being were partially mediated by
loneliness. This finding suggests loneliness serves as an important mediating factor in
the relationships between social support/social strain and critical health outcomes. Rely-
ing on Berkman et al.’s (2000) conceptual model and empirical findings, we propose a
filtration model in which distal social relationship factors (social support and social
strain) operate through proximal psychological factors (e.g., loneliness) to influence
well-being. The underlying assumption is that distal social relationship factors will have
an impact on well-being to the extent that they filter down to affect psychological states
such as loneliness. In light of this reasoning, lower support and higher strain may each
have an indirect detrimental effect on well-being through increased loneliness, which
itself is deleterious to well-being in later life.
Research hypotheses and questions
The initial aim in this investigation is to test the independent effects of social support and
social strain received from four sources (i.e., spouse/partner, children, family members, and
146 Journal of Social and Personal Relationships 31(2)
at University at Buffalo Libraries on February 11, 2014spr.sagepub.comDownloaded from

friends) on loneliness and well-being among older adults, as no single existing study in this
inquiry has included different sources of support and strain. Two hypotheses are posed:
H1: Higher social support from each of the four sources is associated with (a)
lower loneliness and (b) higher well-being.
H2: Higher social strain from each of the four sources is associated with (a) higher
loneliness and (b) lower well-being.
The second aim is to explore the possible mediating role of loneliness in the relationship
between each source of support/strain and well-being in later life. Given that the present
study is the first attempt in this inquiry and that loneliness may serve as one of the psy-
chological pathways linking support/strain to well-being, two research questions (RQs)
are posed:
RQ1: Does loneliness mediate the relationship between each of the four sources of
social support and well-being?
RQ2: Does loneliness mediate the relationship between each of the four sources of
social strain and well-being?
Figure 1 shows the hypothesized theoretical model with direct and indirect paths linking
four sources of support/strain to loneliness and well-being.
Method
Participants
Data for this study came from the 2008 Health and Retirement Study (HRS) conducted
from February 2008 through February 2009 (Health and Retirement Study, 2010; visit
Loneliness
Support
(Spouse)
Strain
(Spouse)
Support
(Children)
Support
(Friends)
Support
(Family)
Well-
being
Strain
(Children)
Strain
(Family)
Strain
(Friends)
Figure 1. Hypothesized theoretical model showing the relationship between each construct.
Chen and Feeley 147
at University at Buffalo Libraries on February 11, 2014spr.sagepub.comDownloaded from

http://hrsonline.isr.umich.edu). The HRS is a national longitudinal study of the eco-
nomic, health, marital, and family status, as well as public and private support systems,
of Americans aged 50 years and older. The HRS was sponsored by the National Institute
on Aging (grant number NIA U01AG009740) and was conducted by the University of
Michigan, USA.
Participants were 7,367 older adults who completed phone interviews regarding
demographics and health conditions and the Leave-Behind Questionnaire in the 2008
HRS. The Leave-Behind Questionnaire collected additional information from partici-
pants without adding to the interview length, and included questions on level of partic-
ipation in general activities, relationships with others, and views on both general and
specific aspects of life. Participants’ ages ranged from 50 to 108 years (M¼69.40,
SD ¼10.43), and of the total, 3,022 (40.3%) were male. Participants’ years of education
ranged from 0 to 17 years (M¼12.47, SD ¼3.22). A majority of participants were mar-
ried (60.3%) and most of them were born in the United States (89.7%).
Measures
Demographic variables and self-reported health status. Age, gender, education, marital
status, and self-reported health status were included in the analysis as control variables.
Self-reported health status was assessed by one item: ‘‘Would you say your health
is ...?’’ The response options ranged from 1 (excellent), 2 (very good), 3 (good), 4 (fair),
to 5 (poor). This item was re-coded so that a higher value indicates a better health status.
Social contact. The current study operationalizes social contact using three indices: (1)
social contact with children, (2) social contact with family members, and (3) social
contact with friends. Each of the three indices was measured by three items, which took
the general form: ‘‘On average, how often do you do each of the following with any of
your children/family members/friends, not counting any who live with you?’’ ‘‘Each of
the following’’ ranged from ‘‘meet up (include both arranged and chance meetings),’’
‘‘speak on the phone,’’ to ‘‘write or e-mail.’’ The response options ranged from 1 (three
or more times a week), 2 (once or twice a week), 3 (once or twice a month), 4 (every few
months), 5 (once or twice a year), to 6 (less than once a year or never). Items were
re-coded so that a higher value indicates a higher frequency of social contact. To reduce
the number of predictors and facilitate the analysis, social contact with each source was
calculated by adding up the frequency of communicating with each source using three
communication channels (meeting up, speaking on the phone, and writing or e-mailing).
This procedure generated three indices of social contact mentioned above.
Social support. Social support was measured by three items of a social support scale
developed by Walen and Lachman (2000). Similar measurements were used by previous
studies and were found to be reliable (e.g., Bertera, 2005). Three items assessing social
support include: ‘‘How much do they really understand the way you feel about things?’’
‘‘How much can you rely on them if you have a serious problem?’’ and ‘‘How much can
you open up to them if you need to talk about your worries?’’ Items were asked in four
loops in reference to participants’ spouse/partner, children, family members, and friends.
148 Journal of Social and Personal Relationships 31(2)
at University at Buffalo Libraries on February 11, 2014spr.sagepub.comDownloaded from

The response options ranged from 1 (alot), 2 (some), 3 (a little), to 4 (not at all). Items
were re-coded so that a higher value indicates a higher level of social support. Social sup-
port from each of the four sources was calculated separately by the average of the above
three items measuring support from the corresponding source. This procedure generated
four scales of social support and their reliabilities were: a¼.81 for support from spouse/
partner; a¼.82 for support from children; a¼.86 for support from family members;
and a¼.83 for support from friends.
Social strain. Social strain was measured by four items of a social strain scale developed
by Walen and Lachman (2000). Similar measurements were used by previous studies
and were recommended as reliable scales (e.g., Bertera, 2005). Four items measuring
social strain were: ‘‘How often do they make too many demands on you?’’ ‘‘How much
do they criticize you?’’ ‘‘How much do they let you down when you are counting on
them?’’ and ‘‘How much do they get on your nerves?’’ Items were asked in four loops in
reference to participants’ spouse/partner, children, family members, and friends. The
response options ranged from 1 (alot), 2 (some), 3 (a little),to4(not at all). Items were re-
coded so that a higher value indicates a higher level of social strain. Social strain from each
of the four sources was calculated separately by the average of the above four items mea-
suring strain from the corresponding source. This procedure generated four scales of social
strain and their reliabilities were: a¼.79 for strain from spouse/partner; a¼.78 for strain
from children; a¼.79 for strain from family members; and a¼.76 for strain from friends.
Loneliness. Loneliness was assessed by a shortened version of the UCLA Loneliness Scale
(Version 3) (Russell, 1996), which has been well established and has been found to have
excellent psychometric properties in previous studies (e.g., Segrin & Domschke, 2011).
This measurement took the general form: ‘‘How much of the time do you feel ...?’’ The
four items for completing this question were: ‘‘you lack companionship,’’ ‘‘left out,’’
‘‘isolated from others,’’ and ‘‘alone.’’ The response options ranged from 1 (often), 2
(some of the time), to 3 (hardly ever or never). Items were re-coded so that a higher value
indicates a higher level of loneliness. The reliability of this measurement was a¼.85.
Well-being. Well-being was assessed by the Satisfaction with Life scale (Diener,
Emmons, Larsen, & Griffin, 1985), which has been widely used by previous studies
as a measure of well-being with favorable psychometric properties (e.g., Steinfield, Elli-
son, & Lampe, 2008). This scale contains the following five items: ‘‘In most ways my
life is close to ideal,’’ ‘‘The conditions of my life are excellent,’’ ‘‘I am satisfied with
my life,’’ ‘‘So far, I have gotten the important things I want in life,’’ and ‘‘If I could live
my life again, I would change almost nothing.’’ The response options ranged from 1
(strongly disagree)to7(strongly agree). A higher value indicates a higher level of
well-being. The reliability of this scale was a¼.88.
Analysis plan
The hypothesized model shown in Figure 1 was tested using structural equation mod-
eling (SEM). The SEM analysis was conducted using AMOS software (version 20.0) and
Chen and Feeley 149
at University at Buffalo Libraries on February 11, 2014spr.sagepub.comDownloaded from

maximum likelihood estimation. Figure 1 depicts 17 causal paths without showing the
measurement portion of the model. In addition to the causal paths shown in Figure 1, the
model also specifies covariances between each pair of the eight exogenous variables
(i.e., four sources of support and four sources of strain). The covariances are not shown in
Figure 1 to maintain a better overview of the model. The direct and indirect effects of
four sources of support and four sources of strain on well-being (the outcome variable)
were estimated, along with the direct effects of these eight exogenous variables on lone-
liness (the mediator) and the direct effect of loneliness on well-being.
As w
2
is usually significant with large samples (Kenny, 2012), several alternative fit
indices were examined to assess model fit. These fit indices included the comparative fit
index (CFI), the Tucker–Lewis index (TLI, also known as the non-normed fit index), the
root mean square error of approximation (RMSEA). The CFI and the TLI values larger
than .90 and .95 are considered acceptable and excellent fit (Kline, 1998), and the
RMSEA values smaller than .05 and .08 are considered close fit and reasonable fit
(McDonald & Ho, 2002).
Results
Descriptive statistics of study factors and a zero-order correlation matrix are provided in
Table 1. Results of the SEM analysis indicated that the model provided a good fit to the
data, CFI ¼.92, TLI ¼.90, and RMSEA ¼.043 (90%confidence interval (CI) ¼.042–
.044), although the w
2
was significant at w
2
(584, N¼7,367) ¼8,425.08, p< .001, and
w
2
/df ¼14.43. Overall, four sources of support and four sources of strain accounted for
approximately 41%of the variance in loneliness (R
2
¼.408); four sources of support,
four sources of strain along with loneliness explained approximately 28%of the variance
in well-being (R
2
¼.284).
The standardized path coefficients including direct effects and indirect effects are
presented in Table 2. As shown in Table 2, support from spouse (b¼.405, p< .001)
and support from friends (b¼.08, p< .001) were both significantly related to lone-
liness, but support from children and support from family were not. Strain from spouse
(b¼.118, p< .001), strain from children (b¼.092, p< .01), strain from family
(b¼.062, p< .05), and strain from friends (b¼.117, p< .001) were all significantly
related to loneliness.
Table 2 also shows that the direct effects of support from spouse (b¼.126, p< .001),
support from children (b¼.117, p< .001), strain from spouse (b¼.125, p< .001), and
loneliness (b¼.307, p< .001) on well-being were significant; other direct paths were
not significant. In addition, Table 2 shows that the indirect effects of support from spouse
(b¼.124, p< .001) and support from friends (b¼.024, p< .001) on well-being through
the pathway of loneliness were both significant, but the indirect effects of support from
children and support from family were not; the indirect effects of strain from spouse
(b¼.036, p< .001), strain from children (b¼.028, p< .01), strain from family
(b¼.019, p< .05), and strain from friends (b¼.036, p< .001) on well-being
through the pathway of loneliness were all significant.
In sum, support from spouse/partner and friends reduced loneliness, while support
from children and family failed to do so; strain from all four sources increased
150 Journal of Social and Personal Relationships 31(2)
at University at Buffalo Libraries on February 11, 2014spr.sagepub.comDownloaded from

Table 1. Descriptive statistics and zero-order correlation matrix of study variables.
Variable 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
1. Age — .03* .14** .19** .10** .06** .05** .09** .05** .18** .06** .04** .07** .22** .20** .15** .02 .08**
2. Gender — .04** .24** .01 .18** .14** .15** .18** .12** .11** .21** .09** .04** .06** .02 .07** .02
3. Education — .11** .31** .18** .07** .26** .14** .01 .06** .08** .07** .05** .06** .06** .13** .11**
4. Marital status — .12** .05** .03* .05** .18** .03* .09** .11** .01 .04** .03** .05** .25** .18**
5. Health status — .14** .07** .20** .15** .10** .03* .09** .11** .13** .13** .12** .25** .35**
6. Contact with children — .43** .38** .05** .39** .13** .14** .01 .07** .03* .05** .13** .13**
7. Contact with family — .30** .02 .18** .49** .15** .02 .02 .02 .03* .08** .06**
8. Contact with friends — .07** .08** .07** .39** .03 .03 .00 .04** .14** .13**
9. Support from spouse — .20** .10** .11** .49** .18** .17** .16** .45** .34**
10. Support from children — .39** .25** .12** .39** .19** .14** .22** .24**
11. Support from family — .29** .08** .15** .19** .06** .14** .12**
12. Support from friends —.08** .05** .01 .07** .14** .10**
13. Strain from Spouse — .36** .30** .31** .39** .30**
14. Strain from children — .52** .42** .29** .24**
15. Strain from family — .48** .28** .20**
16. Strain from friends — .27** .16**
17. Loneliness —.41**
18. Well-being —
Mean 69.84 1.59 12.47 .60 3.08 11.79 9.95 11.17 3.44 3.27 2.90 3.06 1.97 1.69 1.57 1.41 1.50 4.95
SD 9.95 .49 3.23 .49 1.10 3.05 3.22 3.18 .68 .73 .88 .75 .69 .64 .63 .49 .54 1.52
*p< .05, **p< .01.
151
at University at Buffalo Libraries on February 11, 2014spr.sagepub.comDownloaded from

loneliness; higher support and lower strain from spouse/partner directly and indirectly
improved well-being, with indirect effects mediated through reduced loneliness; higher
support from children was directly associated with higher well-being; lower loneliness
fully mediated the effects of higher support from friends and of lower strain from chil-
dren, family, and friends on higher well-being. Thus, H1a, H1b, and H2b were partially
supported, and H2a was supported. Loneliness served as a partial mediator linking sup-
port/strain to well-being.
Alternative models
Demographics and three sources (children, family members, and friends) of social
contact both serve as potential alternative explanations, variables that could potentially
account for the effects of social support, social strain, and loneliness on well-being. Two
alternative models using the proposed model (Figure 1) as the baseline model were
tested. The first alternative model added demographics and self-reported health status.
Fit statistics indicated that this alternative model did not have a better fit to the data than
the baseline model, CFI ¼.91, TLI ¼.88, and RMSEA ¼.042 (90%CI ¼.041–.043);
the w
2
was significant at w
2
(719, N¼7,367) ¼10,043.26, p< .001, and w
2
/df ¼13.97.
The second alternative model added three sources (children, family members, and
friends) of social contact. Fit statistics indicated that this second alternative model
did not provide a better fit to the data than the baseline model, CFI ¼.82, TLI ¼.79,
and RMSEA ¼.055 (90%CI ¼.055–.056); the w
2
was significant at w
2
(911,
N¼7,367) ¼21,498.74, p< .001, and w
2
/df ¼23.56. Thus, the proposed baseline model
provided a more parsimonious fit for the data.
Discussion
This study examined the unique effects of various sources of social support/social strain
on loneliness and well-being in a national sample of older adults. Results indicated that
Table 2. Standardized path coefficients from structural equation modeling analysis.
Standardized direct
effects on loneliness
Standardized direct and
indirect effects on well-being
Predictors Direct Indirect
Support from spouse .405*** .126*** .124***
Support from children .038 .117*** .012
Support from family .013 .027 .004
Support from friends .08*** .015 .024***
Strain from spouse .118*** .125*** .036***
Strain from children .092** .023 .028**
Strain from family .062* .046 .019*
Strain from friends .117*** .049 .036***
Loneliness .307***
*p< .05, **p< .01, ***p< .001.
152 Journal of Social and Personal Relationships 31(2)
at University at Buffalo Libraries on February 11, 2014spr.sagepub.comDownloaded from

loneliness is a unique negative predictor of well-being and exerted the strongest effect on
well-being. The findings with respect to social support and social strain were mixed.
Specifically, social support from one’s spouse/partner and friends alleviated loneliness,
with spousal support exhibiting a much stronger effect than friend support; social strain
from all four sources intensified loneliness. Social support had a stronger effect on
well-being than did social strain, and support from one’s spouse/partner was the most
important source for improved well-being, followed by support from children, and then
support from friends. Furthermore, social strain from one’s spouse/partner showed the
strongest negative effect on well-being among all sources of strain. Finally, higher sup-
port and lower strain directly and indirectly improved well-being, with indirect effects
mediated through reduced loneliness.
Study contributions
Three contributions of this study warrant mention. First, social support and social strain
were examined both simultaneously and separately as antecedents of loneliness and
well-being. It is widely known that social support serves as a coping resource to protect
individuals’ physical and mental health (Cohen & Wills, 1985). However, the possible
independent effect on positive psychological health (e.g., well-being) resulting from
social strain is less understood. The present findings indicated that both social support
and social strain exhibited direct effects on well-being and that social support had a
stronger association with well-being than did social strain. These findings are inconsis-
tent with previous studies, which reported that only social strain was positively related to
depression, while social support was unrelated (e.g., Mavandadi, Sorkin, Rook, &
Newsom, 2007), or that social strain had a larger impact on mental health indicators
(i.e., anxiety and mood disorders) than did social support (e.g., Bertera, 2005).
A likely explanation for this discrepancy is the sampling difference. For example,
Bertera’s (2005) study was based on a national sample of young and middle-aged adults
with a high prevalence of psychiatric disorders. Individuals with psychiatric disorders
may be more vulnerable to the impact of social strain, as it may entail more stress and
anxiety, which intensify preexisting psychological symptoms. The current investiga-
tion was based on a national sample of older adults, and thus these findings are general-
izable to the population of older adults in the US. A second explanation may be the
differences of the measurements on social support and social strain. For instance,
Mavandadi et al.’s (2007) study used a global measure of social support consisting
of four types of support: companionship, emotional, instrumental, and informational
support. In contrast, the social support measure in the present study focused on emo-
tional support. We consider examining each type of support separately a better prac-
tice, as existing research has documented that different types of support may exert
distinct effects on well-being (Reinhardt, Boerner, & Horowitz, 2006). Still a third rea-
son for these varied findings may be the different choices of outcome variables. For
example, the outcome variables in Bertera’s (2005) study were anxiety and mood dis-
orders and the one in Mavandadi et al.’s (2007) study was depression. These negative
psychological health outcomes may be more sensitive to the impact of social strain
than social support. In contrast, the well-being outcome in our study represents positive
Chen and Feeley 153
at University at Buffalo Libraries on February 11, 2014spr.sagepub.comDownloaded from

psychological health, which may be subjected to the influences of both social support
and social strain.
Although the current finding that social support yielded a stronger effect on well-
being than social strain is inconsistent with the previous findings mentioned above, this
finding is in line with socioemotional selectivity theory (SST). SST argues that indi-
viduals’ selection and pursuit of social goals are determined by their perception of time
(Carstensen, Isaacowitz, & Charles, 1999). According to SST, when individuals get
older, they may perceive time as limited and are consequently motivated to reduce
their range of social contacts and focus on relationships that are emotionally rewarding
(Carstensen et al., 1999). As the current participants were in later life, negative social
interactions might become less salient for them, and thus have comparatively less
impact on well-being than positive social interactions (Birditt, Jackey, & Antonucci,
2009).
A second contribution of this study is the method of categorizing and accounting for
the various sources of support and strain (i.e., spouse/partner, children, family members,
and friends). This practice is important for two reasons. The first reason is that it revealed
that only social support from spouse/partner and friends reduced loneliness, while social
support from children and family failed to do so. This finding suggests that loneliness
may be a two-dimensional construct and is in line with Weiss’ (1973) typology of emo-
tional and social loneliness, which argues that emotional loneliness results from an
absence/deficiency of intimate relationships, while social loneliness stems from a lack
of general social relationships. It is possible that among older adults, the intimacy with
a spouse/partner alleviates emotional loneliness, while a social network of friends
diminishes social loneliness. The finding that lower levels of support received from
spouse/partner had a greater influence on loneliness than did lower levels of support
received from friends indicates that emotional loneliness is a more pervasive problem
than is social loneliness in later life.
The second reason is that the extant research on the impact of social relationships on
older adults’ health or well-being failed to differentiate sources/providers of support and
strain (e.g., Mavandadi et al., 2007; Newsom et al., 2005). Although some studies exam-
ined different providers of support, they did not include social strain in their investiga-
tions (e.g., Merz & Huxhold, 2010). The decision to include these four sources generated
richer results and revealed varied strengths of association between social support/social
strain and well-being in later life.
One important finding is that support from friends is beneficial for older adults’ well-
being, while support from family is not significant. This finding is inconsistent with
Merz and Huxhold’s (2010) study that reported emotional support from kin was signif-
icantly and positively related to well-being, while emotional support from non-kin was
insignificant, among older adults. One possible reason is that our study differentiated
children and family members among kin members, while Merz and Huxhold (2010) did
not. As indicated by our findings, support from children significantly improved older
adults’ well-being. It appears that older adults rely more on children for social support,
as children may be perceived as much closer than other family members, with whom
relationships may be more distant. Also, Merz and Huxhold (2010) examined support
from non-kin, which may include friends and acquaintances, while we only examined
154 Journal of Social and Personal Relationships 31(2)
at University at Buffalo Libraries on February 11, 2014spr.sagepub.comDownloaded from

support from friends. Perhaps for older adults, only support from non-kin who are con-
sidered friends is beneficial for well-being. Our findings suggest that a more nuanced
approach may be needed to tease out the complex relationships between different
sources of support and well-being in later life.
In addition, our findings indicated that spouse/partner is the most important source of
both social support and social strain. This finding is in line with the solidarity–conflict
model, which argues that affection and conflict can coexist among close family rela-
tionships (Bengtson et al., 1995). This finding is also consistent with Birditt et al.’s (2009)
study, which suggested that the relationship with spouse/partner most likely exhibits
negative patterns compared to relationships with children or friends among older adults.
A third contribution of this study is that it tested whether loneliness mediates the
relationship between social support/social strain and well-being. There are likely numer-
ous mechanisms through which social support promotes well-being. Our findings clearly
indicated that part of the effect of social support on well-being is mediated by reduced
loneliness. Perhaps social support provided to older adults reassures them that they are
loved, needed, and cared for, and such supportive messages prevent or reduce loneliness,
which itself has a deleterious effect on well-being. On the other hand, our findings jus-
tified that there is a possible pathway linking social strain to well-being through
increased loneliness. In contrast to the caring and supportive messages sent by social
support, social strain experienced by older adults likely sends the message that they are
unloved, unwanted, or neglected, and such hurtful messages actually increase the likeli-
hood of feeling lonely, which in turn decreases their well-being. In addition, the finding
that loneliness serves as a mediator in the relationship between social support/social strain
and well-being supports Berkman et al.’s (2000) conceptual model of how social net-
works impact health. Based on our findings, it appears that social support and social strain
are distal factors and that part of their effects on health outcomes are mediated through the
pathway of more proximal psychological factors (e.g., loneliness).
Theoretical implications
Findings from the present study have three implications for the development and testing
of theories in this area of inquiry. First, the present study found that social support
alleviated loneliness, while social strain intensified loneliness; thus, it appears necessary
for theories of loneliness (e.g., Hawkley & Cacioppo, 2010) to take into account the
deleterious impact of negative social interactions. Incorporating social strain into the-
ories of loneliness promises to provide a full-range understanding as to the independent
impacts of positive and negative aspects of social relationships on the occurrence of
loneliness. Second, as social support and social strain received from different sources
exhibit distinct impacts on loneliness and well-being, this may suggest that future
research on social relationships should differentiate support/strain from various sources,
rather than using a global measure consisting of support/strain from all possible provi-
ders. Clearly, a more fine-grained distinction among support/strain providers would cre-
ate a more nuanced picture of the complex relationships among social support, social
strain, loneliness, and well-being. Third, as the current study found that social support
and social strain both exhibited direct and indirect effects on well-being, a partial
Chen and Feeley 155
at University at Buffalo Libraries on February 11, 2014spr.sagepub.comDownloaded from

mediational model may be more appropriate to understand the mechanism by which
social support and social strain act on well-being, in addition to the well-established
main-effect and buffering models (Cohen & Wills, 1985). This finding may also prompt
researchers to explore other potential mediators between social support/social strain and
well-being.
Practical implications
In practice, the current findings have important implications for developing intervention
strategies toward the improvement of well-being among older adults. As psychosocial
factors (social support, social strain, and loneliness) were found to have a significant
impact on well-being in older adulthood, health interventions targeting older adults
should focus on enhancing support and minimizing strain among social relationships
in an effort to alleviate feelings of loneliness. Specifically, these interventions might
be implemented at both interpersonal and group levels.
At the interpersonal level, intervention programs should focus on maintaining older
adults’ positive interactions with their existing close social contacts. Specifically,
intervention approaches can be directed at encouraging close contacts of older adults to
provide companionship, make home visits, or send caring and supportive messages
through phone calls/e-mails. Such intervention programs should first target older adults’
spouse and children for two reasons. One reason is that spouse and children are the most
important sources of social support, as shown by the present findings. Another reason is
that individuals in later life may constrain their social contacts to those with whom they
feel closely connected, as suggested by SST (Carstensen et al., 1999). Spouse and chil-
dren may become especially important in fulfilling the roles of intimate social contacts in
later life; thus, support from these sources may be more effective in improving the well-
being of older adults than from other sources (e.g., non-kin).
At the group level, intervention programs could help older adults to form support
groups among peers. A friendship enrichment program in the Netherlands has been
successful in helping older women to cope with loneliness (Stevens, 2001). Beneficent
friendships built with other older adults in these support groups may compensate for the
absence of close social contacts and thus buffer feelings of loneliness. Because of per-
ceived similarities, support from peers may be more effective in alleviating loneliness
and improving well-being among support recipients. In addition, such self-help groups
could serve as a monitoring system for any negative interactions, such as domestic
violence, abuse, or neglect, in the social relationships of older adults, with the aim
of eliminating or reducing social strain and preventing its recurrence in the future.
Limitations
Several limitations of this study should be considered when interpreting findings. First,
this study operationalized social support as emotional support, but did not examine other
types of support, such as instrumental support and informational support, as measure-
ments on the latter two types of support were not available in the 2008 HRS. Although
existing literature has suggested that emotional support is most consequential (Burleson,
156 Journal of Social and Personal Relationships 31(2)
at University at Buffalo Libraries on February 11, 2014spr.sagepub.comDownloaded from

2003), including all types of support in the analysis may provide an optimal view of the
independent effect of each type of support, as different types of support may have dis-
tinct effects on well-being (Reinhardt et al., 2006). Similarly, social strain in the present
study was operationalized as a measure opposite to emotional support, while other
unsupportive behaviors, such as failure to offer instrumental support and bad/unwanted
advice and information, were not examined (Newsom et al., 2005).
Second, this study considered social contact, social support, and social strain, which
represent three important dimensions of social relationships. However, there are likely
other dimensions (e.g., consensus and ambivalence), according to the solidarity–conflict
model (Bengtson et al., 1995) and the ambivalence paradigm (Luescher & Pillemer,
1998). Third, loneliness in this study was measured by a shortened version of the UCLA
loneliness scale, which treated loneliness as a unidimensional construct (Russell, 1996).
It is possible that loneliness is a multidimensional construct, consisting of emotional
loneliness and social loneliness (Weiss, 1973). Finally, although loneliness was found to
partially mediate the relationship between social support/social strain and well-being,
other factors (e.g., perceived control) may also act as potential mediators in such rela-
tionships (Windle & Woods, 2004) and need to be considered in conjunction with lone-
liness. It should also be noted that the causal relationships among social support/social
strain, loneliness, and well-being cannot be inferred due to the correlational and cross
sectional nature of this study.
Conclusion
This study contributes to the gerontology literature by highlighting the distinct
associations of positive and negative social interactions with loneliness or well-
being. Findings are also important in terms of the identification of the independent
association of each relational source of social support/social strain with loneliness or
well-being, and the justification of the role of spouse and children as the most
important sources of support in later life. Additionally, this study evinces that enhanced
support and diminished strain directly and indirectly relate to improved well-being, and that
the indirect relationships are mediated through reduced loneliness. Health interventions are
recommended at the interpersonal and group levels, with the aims of alleviating loneliness
and improving the well-being of older adults. Future research may want to examine other
types of received support/strain as well as other dimensions of social relationships (e.g.,
ambivalence), differentiate between emotional loneliness and social loneliness, explore
other potential mediators, and conduct experimental studies to further investigate the rela-
tionships among social support, social strain, loneliness, and well-being in the aging
population.
Acknowledgments
The authors thank Dr. Kristopher J. Preacher of Vanderbilt University for his advice on the struc-
tural-equation-modeling analysis of this study. The authors are also grateful to the two anonymous
reviewers and Associate Editor Konstantinos Kafetsios for their insightful comments and
suggestions.
Chen and Feeley 157
at University at Buffalo Libraries on February 11, 2014spr.sagepub.comDownloaded from

References
Antonucci, T. C., Lansford, J. E., & Akiyama, H. (2001). Impact of positive and negative aspects
of marital relationships and friendships on well-being of older adults. Applied Developmental
Science,5, 68–75.
Bengtson, V., Giarrusso, R., Mabry, J. B., & Silverstein, M. (2002). Solidarity, conflict, and
ambivalence: Complementary or competing perspectives on intergenerational relationships?
Journal of Marriage and Family,64, 568–576. doi:10.1111/j.1741-3737.2002.00568.x
Bengtson, V. L., Rosenthal, C. J., & Burton, L. M. (1995). Paradoxes of families and aging.
In R. H. Binstock & L. K. George (Eds.), Handbook of aging and the social sciences (4th
ed., pp. 253–282). San Diego, CA: Academic Press.
Berkman, L. F., Glass, T., Brissette, I., & Seeman, T. E. (2000). From social integration to health:
Durkheim in the new millennium. Social Science and Medicine,51, 843–857.
Berkman, L. F., & Syme, S. L. (1994). Social networks, host resistance, and mortality: A nine year
follow-up study of Alameda County residents. In A. Steptoe & J. Wardle (Eds.), Psychosocial
processes and health: A reader. (pp. 43–67). New York, NY: Cambridge University Press.
Bertera, E. M. (2005). Mental health in U.S. adults: The role of positive social support and social
negativity in personal relationships. Journal of Social and Personal Relationships,22, 33–48.
doi:10.1177/0265407505049320.
Birditt, K., & Antonucci, T. C. (2008). Life sustaining irritations? Relationship quality and
mortality in the context of chronic illness. Social Science and Medicine,67, 1291–1299.
doi:10.1016/j.socscimed.2008.06.029
Birditt, K. S., Jackey, L. M. H., & Antonucci, T. C. (2009). Longitudinal patterns of negative
relationship quality across adulthood. Journal of Gerontology: Psychological Sciences,
64B, 55–64. doi:10.1093/geronb/gbn031.
Bondevik, M., & Skogstad, A. (1998). The oldest old, ADL, social network, and loneliness. West-
ern Journal of Nursing Research,20, 325–343. doi:10.1177/019394599802000305
Burleson, B. R. (2003). Emotional support skills. In J. O. Greene & B. R. Burleson (Eds.), Hand-
book of communication and social interaction skills (pp. 551–594). Mahwah, NJ: Erlbaum.
Burleson, B. R., & MacGeorge, E. L. (2002). Supportive communication. In M. L. Knapp & J. A.
Daly (Eds.), Handbook of interpersonal communication (3rd ed., pp. 374–424). Thousand
Oaks, CA: Sage.
Cacioppo, J. T., Hawkley, L. C., & Thisted, R. A. (2010). Perceived social isolation makes me sad:
5-year cross-lagged analyses of loneliness and depressive symptomatology in the Chicago
health, aging, and social relations study. Psychology and Aging,25, 453–463. doi:10.1037/
a0017216
Cacioppo, J. T., Hughes, M. E., Waite, L. J., Hawkley, L. C., & Thisted, R. A. (2006). Loneliness
as a specific risk factor for depressive symptoms: Cross-sectional and longitudinal analyses.
Psychology and Aging,21, 140–151. doi:10.1037/0882-7974.21.1.140
Cantor, M. H. (1979). Neighbors and friends: An overlooked resource in the informal support sys-
tem. Research on Aging,1, 434–463.
Carstensen, L. L., Isaacowitz, D. M., & Charles, S. T. (1999). Takingtime seriously: A theory of socio-
emotional selectivity. American Psychologist,54, 165–181. doi:10.1037/0003-066x.54.3.165
Cohen, S., & Wills, T. A. (1985). Stress, social support, and the buffering hypothesis. Psycholo-
gical Bulletin,98, 310–357. doi:10.1037/0033-2909.98.2.310
158 Journal of Social and Personal Relationships 31(2)
at University at Buffalo Libraries on February 11, 2014spr.sagepub.comDownloaded from

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.
Diener, E., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The satisfaction with life scale.
Journal of Personality Assessment,49, 71–75.
Diener, E., Sapyta, J. J., & Suh, E. (1998). Subjective well-being is essential to well-being. Psy-
chological Inquiry,9, 33–37. doi:10.1207/s15327965pli0901_3
Fingerman, K. L., Hay, E. 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
Fiori, K. L., & Consedine, N. S. (2013). Positive and negative social exchanges and mental
health across the transition to college: Loneliness as a mediator. Journal of Social and Per-
sonal Relationships. Epub ahead of print 27 January 2013. doi:10.1177/0265407512473863
Goldsmith,D.J.(2004).Communicating social support. Cambridge, UK: Cambridge University
Press.
Green, L. R., Richardson, D. S., Lago, T., & Schatten-Jones, E. C. (2001). Network correlates of
social and emotional loneliness in young and older adults. Personality and Social Psychology
Bulletin,27, 281–288. doi:10.1177/0146167201273002
Gross, E. F., Juvonen, J., & Gable, S. L. (2002). Internet use and well-being in adolescence. Jour-
nal of Social Issues,58, 75–90. doi:10.1111/1540-4560.00249
Hawkley, L. C., & Cacioppo, J. T. (2010). Loneliness matters: A theoretical and empirical review
of consequences and mechanisms. Annals of Behavioral Medicine,40, 218–227. doi:10.1007/
s12160-010-9210-8
Hawkley,L.C.,Thisted,R.A.,Masi,C.M.,&Cacioppo,J.T.(2010).Lonelinesspredicts
increased blood pressure: 5-year cross-lagged analyses in middle-aged and older adults. Psy-
chology and Aging,25, 132–141. doi:10.1037/a0017805
Health and Retirement Study (2010). (2008 Core, Final, Version 1.0) public use dataset. Produced
and distributed by the University of Michigan with funding from the National Institute on
Aging (grant number NIA U01AG009740). Ann Arbor, MI.
Hogerbrugge, M. J. A., & Komter, A. E. (2012). Solidarity and ambivalence: Comparing two per-
spectives on intergenerational relations using longitudinal panel data. The Journals of Geron-
tology, Series B: Psychological Sciences and Social Sciences,67B, 372–383. doi:10.1093/
geronb/gbr157
Holme´n, K., Ericsson, K., Andersson, L., & Winblad, B. (1992). Loneliness among elderly people
living in Stockholm: a population study. Journal of Advanced Nursing,17, 43–51.
House, J. S. (1981). Work stress and social support. Reading, MA: Addison-Wesley.
Jones, W. H. (1981). Loneliness and social contact. Journal of Social Psychology,113, 295–296.
Kenny, D. A. (2012). Measuring model fit. Retrieved from http://davidakenny.net/cm/fit.htm
Kline, R. B. (1998). Principles and practice of structural equation modeling. New York, NY: The
Guilford Press.
Lawton, L., Silverstein, M., & Bengtson, V. (1994). Affection, social contact, and geographic
distance between adult children and their parents. Journal of Marriage and the Family,56,
57–68. doi:10.2307/352701
Liu, L., & Guo, Q. (2007). Loneliness and health-related quality of life for the empty nest elderly in
the rural area of a mountainous county in China. Quality of Life Research,16, 1275–1280. doi:
10.1007/s11136-007-9250-0
Chen and Feeley 159
at University at Buffalo Libraries on February 11, 2014spr.sagepub.comDownloaded from

Lowenstein, A. (2007). Solidarity–conflict and ambivalence: Testing two conceptual frameworks
and their impact on quality of life for older family members. Journal of Gerontology: Social
Sciences,62B, S100–S107. doi:10.1093/geronb/62.2.S100
Luescher, K. (2004). Conceptualizing and uncovering intergenerational ambivalence. In K. Pille-
mer & K. Luescher (Eds.), Intergenerational ambivalences: New perspectives on parent–child
relations in later life. (pp. 23–62). Oxford, UK: Elsevier Science.
Luescher, K., & Pillemer, K. (1998). Intergenerational ambivalence: A new approach to the study
of parent–child relations in later life. Journal of Marriage and the Family,60, 413–425. doi:10.
2307/353858
Mavandadi, S., Sorkin, D. H., Rook, K. S., & Newsom, J. T. (2007). Pain, positive and negative
social exchanges, and depressive symptomatology in later life. Journal of Aging and Health,
19, 813–830. doi:10.1177/0898264307305179
McDonald, R. P., & Ho, M. R. (2002). Principles and practice in reporting structural equation anal-
yses. Psychological Methods,7, 64–82.
Merz, E.-M., & Huxhold, O. (2010). Wellbeing depends on social relationship characteristics:
Comparing different types and providers of support to older adults. Ageing and Society,30,
843–857. doi:10.1017/s0144686x10000061
Mullins, L. C., & Dugan, E. (1990). The influence of depression, and family and friendship
relations, on residents’ loneliness in congregate housing. Gerontologist,30,377–384.
Newsom, J. T., Nishishiba, M., Morgan, D. L., & Rook, K. S. (2003). The relative importance of
three domains of positive and negative social exchanges: A longitudinal model with compara-
ble measures. Psychology and Aging,18, 746–754. doi:10.1037/0882-7974.18.4.746
Newsom, J. T., Rook, K. S., Nishishiba, M., Sorkin, D. H., & Mahan, T. L. (2005). Understanding
the relative importance of positive and negative social exchanges: Examining specific domains
and appraisals. Journal of Gerontology: Psychological Sciences,60B, P304–P312. doi:10.
1093/geronb/60.6.P304
Peplau, L. A., & Perlman, D. (Eds.). (1982). Loneliness: A sourcebook of current theory, research,
and therapy. New York, NY: Wiley-Interscience.
Rappoport, A., & Lowenstein, A. (2007). A possible innovative association between the concept of
inter-generational ambivalence and the emotions of guilt and shame in care-giving. European
Journal of Ageing,4, 13–21. doi:10.1007/s10433-007-0046-4
Reinhardt, J. P., Boerner, K., & Horowitz, A. (2006). Good to have but not to use: Differential
impact of perceived and received support on well-being. Journal of Social and Personal
Relationships,23, 117–129. doi:10.1177/0265407506060182
Routasalo, P. E., Savikko, N., Tilvis, R. S., Strandberg, T. E., & Pitka¨la¨, K. H. (2006). Social
contacts and their relationship to loneliness among aged people: A population-based study.
Gerontology,52, 181–187. doi:10.1159/000091828
Russell, D. W. (1996). UCLA loneliness scale (Version 3): Reliability, validity, and factor struc-
ture. Journal of Personality Assessment,66, 20–40.
Ryan, R. M., & Deci, E. L. (2001). On happiness and human potentials: A review of research on
hedonic and eudaimonic well-being. Annual Review of Psychology,52, 141–166. doi:10.1146/
annurev.psych.52.1.141
Segrin, C., & Domschke, T. (2011). Social support, loneliness, recuperative processes, and their
direct and indirect effects on health. Health Communication,26, 221–232. doi:10.1080/
10410236.2010.546771
160 Journal of Social and Personal Relationships 31(2)
at University at Buffalo Libraries on February 11, 2014spr.sagepub.comDownloaded from

Segrin, C., & Passalacqua, S. A. (2010). Functions of loneliness, social support, health behaviors,
and stress in association with poor health. Health Communication,25, 312–322. doi:10.1080/
10410231003773334
Sherman, A. M., de Vries, B., & Lansford, J. E. (2000). Friendship in childhood and adulthood: Les-
sons across the life span. International Journal of Aging and Human Development,51, 31–51.
Shiovitz-Ezra, S., & Leitsch, S. A. (2010). The role of social relationships in predicting loneliness:
The national social life, health, and aging project. Social Work Research,34, 157–167.
Silverstein, M., & Bengtson, V. L. (1997). Intergenerational solidarity and the structure of adult
child–parent relationships in American families. American Journal of Sociology,103,
429–460. doi:10.1086/231213
Steinfield, C., Ellison, N. B., & Lampe, C. (2008). Social capital, self-esteem, and use of online
social network sites: A longitudinal analysis. Journal of Applied Developmental Psychology,
29, 434–445. doi:10.1016/j.appdev.2008.07.002
Stevens, N. (2001). Combating loneliness: A friendship enrichment programme for older women.
Ageing and Society,21, 183–202. doi:10.1017/s0144686x01008108
Stevens, N., & Westerhof, G. J. (2006). Partners and others: Social provisions and loneliness
among married Dutch men and women in the second half of life. Journal of Social and Per-
sonal Relationships,23, 921–941. doi:10.1177/0265407506070474
Stroebe, W., Stroebe, M., Abakoumkin, G., & Schut, H. (1996). The role of loneliness and social
support in adjustment to loss: A test of attachment versus stress theory. Journal of Personality
and Social Psychology,70, 1241–1249. doi:10.1037/0022-3514.70.6.1241
Sullivan, H. S. (1953). The interpersonal theory of psychiatry. New York, NY: Norton.
Theeke, L. A. (2010). Sociodemographic and health-related risks for loneliness and outcome dif-
ferences by loneliness status in a sample of U.S. older adults. Research in Gerontological Nur-
sing,3, 113–125. doi:10.3928/19404921-20091103-99
U.S. Census Bureau. (2012). Resident population by sex and age: 1980 to 2010. Retrieved from
http://www.census.gov/compendia/statab/2012/tables/12s0007.pdf
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
Weiss, R. S. (1973). Loneliness: The experience of emotional and social isolation. Cambridge,
MA: MIT Press.
Windle, G., & Woods, R. T. (2004). Variations in subjective wellbeing: The mediating role of a
psychological resource. Ageing and Society,24, 583–602. doi:10.1017/s0144686x04002107
Chen and Feeley 161
at University at Buffalo Libraries on February 11, 2014spr.sagepub.comDownloaded from