Social capital and the care networks of frail seniors.
Keating, Norah ; Dosman, Donna
SOCIAL CAPITAL HAS BEEN A KEY framework in conceptualizing the
place of social ties in quality of life. The foundational work of
scholars like Putnam (1995), Bourdieu (1986), and Coleman (1990) has
influenced our understanding of the broad structure of the relationships
between social networks, norms, and social trust; the processes through
which they work; and the positive outcomes which include various types
of mutual benefit. Indeed, social capital has had such widespread appeal
that identifying the nature of social relationships and their impact
"has become a veritable cottage industry across the social
sciences" (Szreter and Woolcock 2004:650).
Such proliferation of work on social capital has lead to criticism
that too often it is used as a "multi-purpose descriptor[s] for all
types and levels of connections" and thus, lacks the specificity
that would add to its theoretical and practical value (Muntaner, Lynch,
and Smith 2000:108). In this paper, we endeavor to increase the
specificity of social capital as a useful framework through addressing a
particular social group (families of frail seniors), and a particular
benefit (care to an older family member). We draw on social capital
literature to frame our understanding of the social capital inherent in
families of frail older adults, and hypothesize their abilities to
benefit their family members.
SOCIAL CAPITAL AND CLOSE TIES
Social networks are viewed as the building blocks of social capital
(Lauder et al. 2006), the resources or "stock" developed over
time through trust and norms of reciprocity (Rankin 2002; Reimer et al.
2008), which "facilitate coordination and cooperation for mutual
benefit" (Putnam 1995:67). Structural features of such networks
provide different types of connections among members and across
networks, resulting in varying potential to take collective action. From
this perspective, social capital is an asset grounded in social networks
(Reimer et al. 2008).
Two structural features of networks viewed as having differential
social capital potential are those that connect people within networks
and those that link people to other networks or link networks
themselves. Bonding social capital is viewed as reflective of relations
within homogenous groups that have strong ties among members,
characterized by intimate relationships in which the others' needs
are known (Kavanaugh et al. 2005; Putnam 2000). Such groups are well
suited to provide the social and psychological supports its members need
for managing their day-to-day activities (Policy Research Initiative
2003). Families fit the description of groups with strong ties: high
amounts of time spent together, emotional intensity, mutual confiding
and reciprocal services (Kavanaugh et al. 2005), and a strong sense of
shared identity (Reimer et al. 2008).
Bridging social capital is more heterogeneous and is based on weak
ties among network members (Granovetter 1973), better suited to
providing instrumental resources than emotional support (Kavanaugh et
al. 2005). Such ties are useful in connecting people to external assets
and to information (Policy Research Initiative 2003) that help them
obtain access to community resources (Scharf, Phillipson, and Smith
2005). Families seem less likely to have these network assets given
their homogeneous membership and tight bonds may exclude others and
create barriers to information exchanges (Zacharakis and Flora 2005).
However, nonkin, such as friends and neighbors who comprise an element
of social networks, may have more potential to provide links to
community support given their structural position as nonkin whose ties
are more discretionary and individualistic than those of close kin
(Litwin 2001). These relationships could be important to older adults
because numerous direct ties to people who have alternative routes to
valuable resources increases a person's chances of receiving needed
support (Barrett and Lynch 1999; Cornwell, Laumann, and Schumm 2008).
Structural features reflective of bonding and bridging social
capital may be differentially useful. In a statement about the relative
merits of these features of social networks, Putnam (2000:23) argues
that "bonding is good for getting by, but bridging is crucial for
getting ahead." In the context of families providing care to a
frail older adult, one might hypothesize that families would be in a
good position to provide care themselves because they have strong
normative obligations to care and long histories together (Furstenberg
2005). In contrast, family solidarity, which is essential to the
provision of intense levels of care over time, may mitigate against weak
ties to people who can move between groups (Fukuyama 1995), leaving
families without adequate resources with which to broker connections to
formal community services to assist their frail older members.
Families have not been among groups of interest in social capital
research (Donati and Prandini 2007). Yet within the context of research
and public policy on aging, the contemporary discourse on families is
congruent with social capital assumptions. There is a belief that
families are the basis of the system of support to frail seniors,
providing the most responsive, knowledgeable, and nuanced care (Havens,
Donovan, and Hollander 2001). Yet there are also concerns about their
fragility as services are increasingly rationed (Chappell and Penning
2003) and the welfare state maximizes individual risk and responsibility
(Meyer and Wilmoth 2006). Families' abilities to both sustain high
levels of care and gain access to formal services that might assist them
in caring are increasingly important in supporting an aging population
(Wiles 2003; Winslow 2003) and may well exceed their resources.
Little research has focused on deconstructing these assumptions
about families and care. Yet contemporary social capital theorists
caution that it is important to consider the purpose and outcomes of
different ties within their individual contexts (Leonard and Onyx 2003),
avoiding overly simplistic interpretations of the pattern of connections
among people that may mask determinants of positive outcomes (Muntaner
et al. 2000). In this paper, we address two assumptions about families
and social capital: that families are the basis of support to frail
older adults, and that they have limited resources to link to formal
services.
CARE NETWORKS OF FRAIL OLDER ADULTS
What then might be the key characteristics of the social networks
of older adults that would increase the likelihood that they would
receive care if needed? Research on social networks of older adults
shows that in general they have strong, close, and stable social ties
including long-standing kin and friend relationships (Peek and Lin 1999;
van Tilburg 1998). Yet one of the truisms of network analysis is that
"mere presence of a tie between two people does not equate with the
provision of support" (Walker, Wasserman, and Wellman 1993:72).
While structural characteristics of social networks of older adults
provide the potential for supportive actions, access to social resources
may vary considerably (Langford et al. 1997). "Cashing in"
social capital requires actualizing this support potential (Jennings
1999; Tijhuis et al. 1998).
Support networks are groups of family members and friends that are
responsive to the everyday needs of older adults. These support
relationships provide the basis for care, an intense form of support
that goes beyond what is required in everyday life to encompass tasks
and services needed to support frail older adults (Barrett and Lynch
1999; Hanson et al. 1997; Walker, Pratt, and Eddy 1995).
Four main types of support networks have been found: family-based,
friend-based, diverse, and limited (Fiori, Smith, and Antonucci 2007).
Close-family networks include mostly close and distant kin. Many of
these networks provide extensive day-to-day support, although some are
less intimate with few family members actively involved (Aartsen et al.
2004; Litwin and Landau 2000; Melkas and Jylha 1996). Similarly, some
friend-based networks have active exchanges of practical help (Melkas
and Jylha 1996), while in others relationships are cordial but without
active provision of resources (Litwin and Landau 2000). Diverse networks
have a mix of kin and nonkin and are typified by frequent contact among
them (Aartsen et al. 2004). Those with limited network ties often have
no spouse, few if any children, and little contact with friends. Their
networks are small and socially isolated with little supportive
interaction (Fiori, Antonucci, and Cortina 2006; Litwin 2001; Stone and
Rosenthal 1996).
Clearly support network resources in old age are not universally
available (Aartsen et al. 2004) suggesting that some older adults may
not receive care if needed. The limited empirical evidence of the
structure of care networks shows that they may be less diverse than
support networks. Researchers studying the evolution of support to care
networks have found two main trends. The first is that there is an
increasing concentration of family members in care networks of frail
older adults (Aartsen et al. 2004; van Tilburg 1998). Family-focused
care networks have small numbers of close kin (Wenger and Keating 2008;
Wolff and Kasper 2006) who are predominantly female and living with or
close to the person for whom they are caring (Fast et al. 2004). Spouses
and adult children are the most common family caregivers (see e.g.,
Chappell 2003; Walker et al. 1995). They provide the highest amount of
care and most intimate care tasks (Li 2004), but are least likely to
receive formal services (Lyons, Zarit, and Townsend 2000; Winslow 2003).
Some researchers have found that friendship-oriented support
networks are unlikely to provide care (Lyons et al. 2000; van Groenou
and van Tilburg 1997), although others have found that friends remain
part of care networks (Cranswick 2003; Reinhardt, Boerner, and Benn
2003). Where present, friends and neighbors in care networks increase
the likelihood that formal services will be received (Barrett and Lynch
1999), evidence that they are good mediators between caregiving families
and service providers (Li 2004).
Less is known about how other structural characteristics of care
networks may influence their caring capacity. Care networks have
characteristics that may reflect high levels of bonding social capital:
higher proportions of women, long-standing relationships, people living
nearby, and small numbers including those caring alone (Tijhuis et al.
1998; Wolff and Kasper 2006). However, increasing proportions are also
in the labor force. Care networks in which all members are employed
provide the fewest weekly hours of care (Keating et al. 2008), though
their employment status may afford opportunities to link with community
services. Social capital assumptions allow us to further our
understanding of resources that families might draw upon to care for
their older family members.
We address two main research questions toward better understanding
the structure and actualized social capital of care networks of frail
older Canadians:
1. What are the types of care networks of older adults and how do
they differ in their structural characteristics? Findings related to
this question will allow us to develop a more nuanced picture of the
differential potential of care networks to provide care themselves and
to broker care from the formal care sector.
2. How is care-potential actualized? Findings related to this
question will allow us to consider diversity in the amount of care
provided and in the use of formal care services.
METHODS
Data for this study were drawn from the nationally representative
Statistics Canada 2002 General Social Survey on Aging and Social
Support. The sample was drawn using random digit dialing. Interviews
were conducted by telephone. Those living in the northern territories
(Yukon, Northwest Territories, Nunavut), or in institutions, and those
without telephones were excluded. Data were collected throughout the
year. The response rate was 83.8 percent. From the total sample of
24,870 Canadians age 45 years and older, a sample was drawn of adults
aged 65 and over who indicated that they had received assistance from
family/friends (n = 2,407). Care was operationalized as having received
assistance with one or more tasks in the previous 12 months because of a
long-term health problem or disability. Care tasks include meal
preparation or cleanup; housekeeping; household maintenance or repair;
grocery shopping; transportation; help with banking or bill paying; and
personal care. Care receivers indicated who provided assistance in which
tasks. They were also asked how frequent these tasks were performed and
for how much time was spent performing the tasks. Care networks were
operationalized as all family members/friends that the respondent said
had provided them with one or more tasks in the previous 12 months
because of their long-term health problem or disability. Additionally,
formal care is operationalized as all care provided to the respondent by
individuals from nongovernmental organizations, paid employees of the
respondent, or government employees.
Analyses
Care network was the unit of analysis. The first set of analyses
addressed the question of what is the structure of care networks of
older adults. The sample was differentiated into care network types by
utilizing structural network characteristics in a cluster analysis
procedure. Structural characteristics were: gender composition
(proportion of the care network comprised of women); age composition
(proportion of the care network between 45 and 64 and proportion of the
care network over 65); relationship composition (proportion of the care
network comprised of kin); proximity (proportion of the care network
residing with the senior recipient and proportion of the care network
more than 1/2 day's travel away from the senior recipient);
employment (proportion of the care network employed full or part time);
and network size (number of members of the care network). These network
characteristics were chosen based on key characteristics of support and
care networks (see e.g., Fast et al. 2004; Fiori et al. 2007; van
Groenou and Tillburg 1997).
K-means cluster analysis method was used. This method divides the
sample into subgroups based on their similarity across a group of
variables. The clusters are derived by determining their similarity on
the set of structural network characteristics described above.
Dissimilarity between clusters is measured by squared Euclidean
distance, which involves summing and squaring across all the variables.
Variables with larger variances are weighted more heavily in the cluster
solution. All the variables in these analyses with the exception of
network size are expressed as proportions ranging from 0 to 1,
consequently the network size variable was standardized to ensure that
it contributed equally to the cluster formation. Solutions are sought
that minimize the within-group differences and maximize the
between-group differences on the set of structural network
characteristics.
In the second set of analyses, we examined the question of how the
structure of care networks may influence actualized social capital: the
amount of care provided by the networks, and by formal care services.
Three sets of analyses were conducted:
1. The average number of hours per week of care provided by family
and friends for each of the six network types was calculated.
2. The proportion of frail seniors who received formal care in
addition to family and friend care was calculated for each of the six
care network types.
3. Patterns of distribution of hours of care provided were
examined. The sample was divided into two groups: older adults who
received family/friend care only and older adults who received
family/friend plus formal care. Distribution of hours of care within
each of these two groups was examined by calculating the average number
of hours of care per week provided by family/friends and by formal
services for each of the six care network types. Difference of means
tests was conducted using an ANOVA estimation controlling for the
differing level of disabilities.
RESULTS
The final result of the cluster analysis was a six-cluster solution
of typologies of frail seniors' care networks. The six-cluster
solution resulted in distinctly different clusters which offered the
most information with the fewest number of clusters (Table 1).
Three of the six care network types were almost entirely based in
close-family relationships. All had more than 80 percent close kin
though they differed on age distribution, and on proximity and
employment composition. Children at home (5.3 percent of all networks)
included adult children who were middle-aged, employed, and living with
the older adult; Lone spouse (13.4 percent of networks) were older
husbands or wives, living with their frail partner. Spouse and children
(11.2 percent of networks) were close kin, female, diverse in age,
proximity, and employment status. These three close-family care networks
comprised 29 percent of all care network types. This proportion is lower
than expected given previous findings that most members of care networks
are close kin.
Three care network types were based in a broader set of members
that included both friends and family. The friend-and-family networks
ranged from 17 percent to 42 percent friends. All lived at a distance
from the older adult but differed on other characteristics. The largest
proportion of Older diverse (9.0 percent of networks) networks was
friends, but with smaller proportions of close kin and distant kin. Most
were over age 65, highlighting the presence of same generation
caregivers. Close kin and friends (46.7 percent of networks) were mostly
close kin, with a smaller proportion of friends, were middle aged and
employed. Younger diverse networks (14.4 percent) were a mix of
(distant) kin and friends, middle aged and younger, and employed. This
network type had the lowest proportion of women. Together these three
network types made up the majority (71 percent) of all networks.
These six care network types are more diverse than previously
found. Close-family networks reflect most closely the tightly knit
groups hypothesized to be high on bonding social capital.
Friend-and-family networks have a mix of characteristics, such as age
and employment status, in addition to kin and friends that may enhance
their ability to link to community resources.
Hours of care provided by the close-family and friend-and-family
networks provide evidence of some of the expected differences in
actualized social capital based on bonding and bridging potential.
Networks differ considerably in the hours of care they provide (Table
2). The three close-family networks provide highest hours of care,
between 10 and 18 hours per week. These networks are tightly knit; most
are adult children and spouses with high normative obligations to care.
All have coresident caregivers. In contrast, friend-and-family networks
provide fewer hours than any of the close-family networks, between 2.9
and 6.9 hours per week. Among them, most care is provided by networks
with the highest proportion of close kin: Close kin and friends have 76
percent of members who are close relatives.
Overall, these findings suggest a positive relationship between the
proportions of close kin in a network and the hours of care provided,
evidence that it is not just family ties, but ties to close kin, such as
spouse and children, that are key resources in "getting by"
under the difficult circumstances of frailty of an older person.
Bridging social capital was estimated in part by calculating the
proportion of each of the six care network types that received formal
services (Table 3). As expected, the highest proportion of networks
caring without formal services is among the close-family networks. Among
them, two network types least likely to bridge to formal help are those
who live with the frail older adult. Proximity may be an indicator of
highest bonding among networks whose lives are focused on the day-to-day
support of the older adult. Only 28 percent of Children at home networks
and 23 percent of Lone spouse networks received formal help.
In contrast, networks that were most heterogeneous were more likely
to bridge to formal services. Half of Close kin and friends and 43
percent of Older diverse networks linked to formal services. These links
can be explained only partially by the presence of ties to friends.
Older diverse networks indeed have the highest proportion of friends and
thus theoretically have the highest linking potential. Yet Close kin and
friends, a family-based network type, are the most likely network to
receive formal services. It may be that for these networks, a
combination of characteristics increases access to formal services. High
labor force participation may increase knowledge of community resources
while being in mid life and older may be indicative of time to develop
skills in gaining access to them.
It is important to note that the majority of each of the network
types received no formal help. It may be that overall membership in care
networks is not sufficiently diverse to allow for linking to formal
services needed to support a frail older adult. It is also possible that
given the constraints on formal health-care resources across the
country, even strong linking skills may not be adequate to garner scarce
care resources.
Finally, patterns of distribution of hours of care provide further
information on the relative importance of bonding and bridging resources
in care networks. Table 4 has two sections. The first column shows mean
weekly hours of care provided by care networks when no formal care was
being received. The second set of columns show the mean weekly hours of
care provided by care networks by formal services, and the total of
these two.
These findings reveal patterns of care across networks and sources
of care. There are similarities across networks both in hours of care
provided in the absence of formal care, and of hours of formal care
received. When networks are caring alone (i.e., without formal
services), mean hours of care are similar across four of the care
network types between six and eight hours per week. (Exceptions are that
Younger diverse networks provide less care, while Spouse and children
networks provide substantially more.) Among networks in receipt of
formal services, mean hours of formal services also show little
variation between four and seven hours per week across all six network
types.
These similarities in care by networks caring alone and in formal
service hours across networks are in sharp contrast to the substantial
differences in care hours when networks are caring along with formal
services. When formal care is also provided, close-family networks
provide between 18.8 and 23.3 hours of care per week; friend-and-family
networks provide between 3.6 and 7.3 weekly hours. It may be that
close-family networks are reluctant to link to formal services and do so
only when they have exhausted their personal caring resources. In
comparison, lower care obligations of friend-and-family networks may
lead them to limit their care inputs after linking with formal services.
DISCUSSION
Results from these analyses provide information to address two
assumptions about families and social capital: that families are the
basis of support to frail older adults; and that they have limited
resources to link to formal services.
A key finding in this study is the centrality not just of families,
but of close kin, in caring for frail older adults. The three care
networks based on close, nuclear, family relationships most closely
resemble the family group believed to epitomize bonding social
capital-devoted spouses and adult children who have the highest
normative obligations to provide care (Szinovacz and Davey 2008). Care
provision by these networks is evidence of their strong commitment. Even
in the face of limited formal help, they provided relatively high hours
of care.
All six network types had some close-kin members, suggesting that
the high bonding inherent in close-kin family connections is a necessary
prerequisite to the emergence of care networks. Nonetheless, the
presence of friends and more distant kin indicates structural capacity
to bridge to formal community resources, particularly among the three
friends-and-family network types. These networks had diverse memberships
typical of groups that can link to other resources and were more likely
to receive formal services than were close-family networks. Importantly
though, all of these networks had some family members. While families do
not provide all of the care without their commitment, care networks are
unlikely to emerge. It appears that while care work has not entirely
been transferred to families (Armstrong and Armstrong 2005; Chappell and
Penning 2003; Hales 2007; Penning 2002), responsibility for care does
lie with them.
In intense caregiving situations, bonding has its costs.
Close-family networks are least likely to receive formal help,
especially if network members live with the cared-for person. Networks
that receive formal help get few hours of assistance despite providing
high levels of care. There is a longstanding concern of the isolation of
such proximate caregivers whose high care demands may lead to poor
health or social isolation (Michelson and Tepperman 2003; Seltzer and
Wailing 2000). Recent findings show that among adult child caregiver
groups, tensions over equity in caregiving can lead to poor family
relationships, feelings of exclusion, and even bitter court battles
(Lashewicz et al. 2007).
Questions about sustainability of care emerge from the analyses of
patterns of care when formal services are received. Overall,
friend-and-family networks provide less care than do close-family
networks. These differences are accounted for by the much lower levels
of family-and-friend care when formal services are provided. Differences
may be a result of when community services are sought. Networks with
predominantly weak ties may link to formal services earlier because they
have bridging capacity and because they lack the bonding social capital
required to assume primary responsibility for the ongoing work of care.
From the data reported here, we are not able to estimate the proportion
of support networks based on friend and neighbor relationships that do
not make the transition into caregiving, though prior research would
suggest that these proportions will be small (Lyons et al. 2000; van
Groenou and van Tilburg 1997).
In contrast, the high bonding of close-family networks may result
in delaying the search for formal help. Close kin may be reluctant to
ask for assistance for tasks that they believe they should do themselves
(Rosenthal, Martin-Matthews, and Keefe 2007). Having poor bridging
resources or being reluctant to use them may mean that services are
received only when initiated by community service providers who become
aware of high needs of isolated caregivers as a result of health crisis.
Indeed as Szreter and Woolcock (2004) have argued, these close-family
networks can function in both a socially inclusive and exclusive way, at
times to the detriment of their own quality of life and ability to care
for their frail relative. Longitudinal research on care networks will be
required to tease out these possible differences in existence of
bridging resources, willingness to use them, and efficacy of these
strategies.
Other characteristics of care networks in addition to relationship
may enhance or constrain caring resources. Networks living with the
cared-for person reflect structural characteristics and caring behaviors
indicative of high bonding. Examination of these networks over time may
provide insights into how such bonding social capital acts as a resource
to frail older adults. Little is known for example, about whether such
older adults remain longer in the community than those cared for by less
homogeneous networks or whether high emotional and physical costs to
caregivers result in earlier depletion of their abilities to sustain
care.
Employment might be expected to be a bridging resource because it
provides access to community connections. However, we found no clear
patterns of relationship between employment composition of networks and
amount of formal care received. In fact, in close-family networks, high
levels of employment are associated with lowest hours of formal care
received. Yet these high-employment close-family networks differ
considerably in the amount of care they provide themselves. The high
bonding of Children at home networks may trump competing demands of
employment in terms of time devoted to caregiving. In contrast, lower
levels of normative obligation to care in friends-and-family networks
may mean that Younger diverse networks with their high rates of
employment are limited both in their willingness to broker formal care
and to provide high levels of personal caregiving.
Findings of previous research on gender of individual caregivers
that women provide higher hours of care (MacDonald, Phipps, and
Lethbridge 2005; Stobert and Cranswick 2004) are replicated at the
network level. Spouse and children networks that have the highest
proportion of women caregivers provide the highest hours of care while
Younger diverse networks with lowest proportion of women provide fewest
hours of care. Nonetheless highly bonded proximate networks, such as
Children at home and Lone spouse, with balanced gender composition
provide high levels of care when formal services are provided.
Longitudinal research would help tease out the relative importance of
gender, family obligations, and caregiving demands in provision of care.
In conclusion, social capital theory has been useful in further
understanding how structural ties among care networks relate to their
capacity to provide care and to link to formal resources that might
assist them. Through increasing the specificity of social capital as a
framework, we hope that we also have addressed some of the concerns of
social capital scholars that arguments suffer from "logical
circularity" (Portes 1998:19). Portes argues that scholars must
separate social capital from its outcomes; demonstrate that social
capital is present before the outcomes that it is expected to produce;
and recognize that there are downsides to social connections.
The rich literature on families and caregiving has provided
empirical evidence of the nature of bonding and bridging social capital
in caregiving families and the potential for heterogeneity in these two
types of social capital. To our knowledge, the outcome of amount of
formal services received is the first test of this heterogeneity.
Further empirical tests with different outcomes, such as levels of
family conflict, would strengthen this theoretical specificity. However,
there is evidence from these analyses that bonding social capital in
close-kin relationships provides the basis of support to frail older
adults and the potential for their exclusion from linkages to community
resources, while bridging social capital in more heterogeneous
family-and-friend networks increased the likelihood of linking to formal
resources. Severe limitations in the availability of formal care
resources may account for lack of variation in the number of formal care
hours received across network types. There are hints of important
influences of other structural characteristics of care networks on
amount of care provided by networks in light of limited access to formal
resources.
Because the focus here is on families, which are ongoing social
groups, we believe that there is a strong case for arguing that in this
situation, the presence of social capital is demonstrably before the
outcomes. Ultimately, directionality can best be determined by empirical
work that is longitudinal. There is need of more research in which
structural changes in care networks are traced along with a set of
outcomes, such as receipt of formal services and levels of family
conflict.
Finally, much has been learned about how patterns of connections
among members of networks of older adults may determine positive
outcomes for them (Muntaner et al. 2000). Yet there also is powerful
evidence from prior research that "bonding has its costs."
Examples cited here of poor family relationships, feelings of exclusion,
court battles may well escalate in the contemporary economic context of
scarce formal care resources and increasing demands of an aging
population. It is time to move beyond assumptions that close-family ties
necessarily result in the best care outcomes for frail older adults or
their family members.
References
Aartsen, M.J., T. van Tilburg, C.H.M. Smits and K.C.P.M.
Knipscheer. 2004. "A Longitudinal Study of the Impact of Physical
and Cognitive Decline on the Personal Network in Old Age." Journal
of Social and Personal Relationships 21:249-66.
Armstrong, P. and H. Armstrong. 2005. "Public and Private:
Implications for Care Work." Sociological Review 53(Suppl
2):169-87.
Barrett, A.E. and S.M. Lynch. 1999. "Caregiving Networks of
Elderly Persons: Variation by Marital Status." Gerontologist
39:695-704.
Bourdieu, P. 1986. "The Forms of Capital." Pp. 241-58 in
Handbook of Theory and Research for the Sociology of Education, edited
by J. Richardson. New York: Greenwood Press.
Chappell, N.L. 2003. "Correcting Cross-Cultural Stereotypes:
Aging in Shanghai and Canada." Journal of Cross-Cultural
Gerontology 18:127-47.
Chappell, N.L. and M.J. Penning. 2003. "Family Caregivers:
Increasing Demands in the Context of 21st-Century Globalization?"
Pp. 455-62 in The Need for Theory: Critical Approaches to Social
Gerontology, edited by S. Biggs, A. Lowenstein and J. Hendricks.
Amityville, NY: Baywood.
Coleman, J. 1990. Foundations of Social Theory. Cambridge, MA: The
Belknap Press of Harvard University Press.
Cornwall, B., E.O. Laumann and L.P. Schumm. 2008. "The Social
Connectedness of Older Adults: A National Profile." American
Sociological Review 73:185-203.
Cranswick, K. 2003. General Social Survey Cycle 16: Caring for an
Aging Society 2002. Ottawa, ON: Statistics Canada.
Danati, P. and R. Prandini. 2007. "The Family in the Light of
a New Relational Theory of Primary, Secondary and Generalized Social
Capital." International Review of Sociology 17:209-23.
Fast, J., N. Keating, P. Otfinowski and L. Derksen. 2004.
"Characteristics of Family/Friend Care Networks of Frail
Seniors." Canadian Journal on Aging 23:5-19.
Fiori, K., T. Antonucci and K. Cortina. 2006. "Social Network
Typologies and Mental Health among Older Adults." Journals of
Gerontology Series B: Psychological Sciences and Social Sciences
61:25-32.
Fiori, K., J. Smith and T. Antonucci. 2007. "Social Network
Types among Older Adults: A Multidimensional Approach." Journals of
Gerontology Series B: Psychological Sciences 62 B:322-30.
Fukuyama, F. 1995. Trust: The Social Virtues and the Creation of
Prosperity. New York: Free Press.
Furstenberg, F.F. 2005. "Banking on Families: How Families
Generate and Distribute Social Capital." Journal of Marriage and
Family 67:809-21.
Granovetter, M.S. 1973. "The Strength of Weak Ties."
American Journal of Sociology 78:136080.
Hales, C. 2007. "Crisis or Commotion? An Objective Look at
Evidence on Caregiving in Families." Family Matters 76:18-23.
Hanson, B.S., P.-O. Ostergren, S. Emstahl, S.-O. Isacsson and J.
Ranstam. 1997. "Reliability and Validity Assessments of Measures of
Social Networks, Social Support and Control: Results from the Maim6
Shoulder and Neck Study." Scandinavian Journal of Social Medicine
25:249-57.
Havens, B., C. Donovan and M. Hollander. 2001. "Policies That
Have Positive or Negative Impacts on Informal Care in Canada."
Paper presented at the Congress of the International Association on
Gerontology, Vancouver, British Columbia, July.
Jennings, A. 1999. "The Use of Available Social Support
Networks by Older Blacks." Journal of National Black Nurses
Association 17:519-25.
Kavanaugh, A., D. Reese, J. Carroll and M. Rosson. 2005. "Weak
Ties in Networked Communities." Information Society 21:119-31.
Keating, N., D. Dosman, J. Fast and J. Swindle. 2008. "Sharing
the Work: Care Networks of Frail Seniors in Canada." Pp. 165-83 in
Blurring the Boundaries: Ageing at the Intersection of Work and Home
Life, edited by A. Martin-Matthews and J. Phillips. London, UK:
Routledge.
Langford, C.P.H., J. Bowsher, J.P. Maloney and P.P. Lillis. 1997.
"Social Support: A Conceptual Analysis." Journal of Advanced
Nursing 25:95-100.
Lashewicz, B., G. Manning, M. Hail and N. Keating. 2007.
"Equity Matters: Doing Fairness in the Context of Family
Caregiving." Canadian Journal on Aging 26(Suppl 1):91-102.
Lauder, W., S. Reel, J. Farmer and H. Griggs. 2006. "Social
Capital, Rural Nursing and Rural Nursing Theory." Nursing Inquiry
13:73-79.
Leonard, R. and J. Onyx. 2003. "Networking through Loose and
Strong Ties: An Australian Qualitative Study." Voluntas 14:189-204.
Li, L.W. 2004. "Caregiving Network Compositions and Use of
Supportive Services by Community-Dwelling Dependent Elders."
Journal of Gerontological Social Work 43:147-64.
Litwin, H. 2001. "Social Network Type and Morale in Old
Age." Gerontologist 41:516-24.
Litwin, H. and R. Landau. 2000. "Social Network Type and
Social Support among the Old-Old." Journal of Aging Studies
14:213-28.
Lyons, K.S., S.H. Zarit and A.L. Townsend. 2000. "Families and
Formal Service Usage: Stability and Change in Patterns of
Interface." Aging and Mental Health 4:234-43.
MacDonald, M., S. Phipps and L. Lethbridge. 2005. "Taking Its
Toll: The Influence of Paid and Unpaid Work on Women's
Well-Being." Feminist Economics 11:63-94.
Melkas, T. and M. Jylha. 1996. "Social Network Characteristics
and Social Network Types among Elderly People in Finland." Pp.
99-116 in The Social Networks of Older People: A Cross-National
Analysis, edited by H. Litwin. Westport, CN: Praeger.
Meyer, M.H. and J.M. Wilmoth. 2006. "Changing Demographics,
Stagnant Social Policies." Research on Aging 28:265-68.
Michelson, W. and L. Tepperman. 2003. "Focus on Home: What
Time-Use Data Can Tell about Caregiving to Adults." Journal of
Social Issues 59:591-610.
Muntaner, C., J. Lynch and G.D. Smith. 2000. "Social Capital
and the Third Way in Public Health." Critical Public Health
10:107-24.
Peek, M.K. and N. Lin. 1999. "Age Differences in the Effects
of Network Composition on Psychological Distress." Social Science
and Medicine 49:621-36.
Penning, M.J. 2002. "Hydra Revisited: Substituting Formal for
Self- and Informal In-Home Care among Older Adults with
Disabilities." Gerontologist 42:4-16.
Policy Research Initiative. 2003. Social Capital: Building on a
Network-Based Approach. Ottawa, ON: Policy Research Initiative.
Portes, A. 1998. "Social Capital: Its Origins and Applications
in Modern Sociology." American Review of Sociology 24:1-24.
Putnam, R. 1995. "Bowling Alone: America's Declining
Social Capital." Journal of Democracy 6:65-78.
Putnam, R. 2000. Bowling Alone: The Collapse and Revival of
American Community. New York: Simon and Schuster.
Rankin, K. 2002. "Social Capital, Microfinance, and the
Politics of Development." Feminist Economics 8:1-24.
Reimer, B., T. Lyons, N. Ferguson and G. Polanco. 2008.
"Social Capital as Social Relations: The Contribution of Normative
Structures." The Sociological Review 56:256-74.
Reinhardt, J.P., K. Boerner and D. Benn. 2003. "Predicting
Individual Change in Support Over Time among Chronically Impaired Older
Adults." Psychology and Aging 18: 770-79.
Rosenthal, C.J., A. Martin-Matthews and J.M. Keefe. 2007.
"Care Management and Care Provision for Older Relatives amongst
Employed Informal Care-Givers." Ageing and Society 27:755-78.
Scharf, T., C. Phillipson and A.E. Smith. 2005. "Aging in a
Difficult Place: Assessing the Impact of Urban Deprivation on Older
People." Pp. 153-74 in New Dynamics in Old Age: Individual,
Environmental and Societal Perspectives, edited by H.-W. Wahl, C.
Tesch-Romer and A. Hoff. Amityville, NY: Baywood Publishing.
Seltzer, M.M. and L.L. Wailing. 2000. "The Dynamics of
Caregiving: Transitions during a Three-Year Prospective Study."
Gerontologist 40:165-78.
Stobert, S. and K. Cranswick. 2004. "Looking After Seniors:
Who Does What for Whom?" Canadian Social Trends 74:2-6.
Stone, L. and C. Rosenthal. 1996. "Profiles of the Social
Networks of Canada's Elderly: An Analysis of 1990 General Social
Survey Data." Pp. 77-97 in The Social Networks of Older People: A
Cross-National Analysis, edited by H. Litwin. Westport, CN: Praeger.
Szinovacz, M.E. and A. Davey. 2008. "The Division of Parent
Care between Spouses." Ageing and Society 28:571-97.
Szreter, S. and M. Woolcock. 2004. "Health by Association?
Social Capital,
Social Theory, and the Political Economy of Public Health."
International Journal of Epidemiology 33:650-67.
Tijhuis, M.A.R., H.D. Flap, M. Foets and P.P. Groenewegen. 1998.
"Selection in the Social Network: Effects of Chronic
Diseases." European Journal of Public Health 8:286-93.
van Groenou, M.B. and T. van Tilburg. 1997. "Changes in the
Support Networks of Older Adults in the Netherlands." Journal of
Cross-Cultural Gerontology 12:23-44.
van Tilburg, T. 1998. "Losing and Gaining in Old Age: Changes
in Personal Network Size and Social Support in a Four-Year Longitudinal
Study." Journal of Gerontology: Social Sciences 53(B):S313-23.
Walker, A.J., C.C. Pratt and L. Eddy. 1995. "Informal
Caregiving to Aging Family Members: A Critical Review." Family
Relations 44:402-11.
Walker, M.E., S. Wasserman and B. Wellman. 1993. "Statistical
Models for Social Support Networks." Sociological Methods and
Research 22:71-98.
Wenger, G.C. and N. Keating. 2008. "The Evolution of Networks
of Rural Older Adults." Pp. 3342 in Rural Ageing: A Good Place to
Grow Old?, edited by N. Keating. London, UK: Policy Press.
Wiles, J. 2003. "Informal Caregivers' Experiences of
Formal Support in a Changing Context." Health and Social Care in
the Community 11:189-207.
Winslow, B.W. 2003. "Family Caregivers' Experiences with
Community Services: A Qualitative Analysis." Public Health Nursing
20:341-49.
Wolff, J.L. and J.D. Kasper. 2006. "Caregivers of Frail
Elders: Updating a National Profile." Gerontologist 46:344-56.
Zacharakis, J. and J. Flora. 2005. "Riverside: A Case Study of
Social Capital and Cultural Reproduction and Their Relationship to
Leadership Development." Adult Education Quarterly 55:288-307.
NORAH KEATING
University of Alberta
DONNA DOSMAN
University of Alberta
The authors wish to acknowledge the Social Sciences and Humanities
Research Council (SSHRC) Major Collaborative Research Initiative for
support of the Hidden Costs, Invisible Contributions Research Program
(2003-2008) of which this research is a part.
Dr. Norah Keating, Professor and Co-director Research on Aging,
Policies and Practice, Department of Human Ecology, University of
Alberta, Edmonton, AB, T6G 2N1, Canada. E-mail:
norah.keating@ualberta.ca
Table 1
Care Network Types and Characteristics
Network composition characteristics
Network Proportion of
characteristics network members
Size
Relationship % close kin
% distant kin
% friends
Gender % a women
Age % < 45 years
% 45-64 years
% > 65 years
Proximity % same household
Employment % employed
Network type
Lone Children Spouse and Close kin Older Younger
spouse at home children and friends diverse diverse
1.0 1.6 1.4 1.9 1.4 1.3
94 84 80 76 31 17
3 10 7 7 27 58
3 7 13 17 42 25
49 43 61 54 42 38
0 67 22 29 20 73
14 22 47 62 7 20
86 11 31 9 74 7
99 82 53 25 22 4
4 72 44 66 14 62
Table 2
Mean Weekly Hours of Family/Friend Care by Network Type
Mean weekly hours of care
Care network type from family/friend networks
Lone spouse 10.9
Children at home 10.2
Spouse and children 18.1
Close kin and friends 6.9
Older diverse 6.2
Younger diverse 2.9
Table 3
Proportion of Family/Friend Care Networks in
Receipt of Formal Care
Proportion
receiving
Care network type formal care
Lone spouse 23
Children at home 28
Spouse and children 40
Close kin and friends 50
Older diverse 43
Younger diverse 37
Table 4
Hours of Care Provided by Family/Friend and Formal Care
Providers by Care Network Type
Mean Mean weekly hours
weekly hours Family/friend plus formal care
Family/friend
care only Total
family/
Total Family/ friend+
family/friend friend Formal formal
Care network type care only care care care
Lone spouse 8.0 * 18.8 4.1 22.9 *
Children at home 6.0 * 23.3 4.0 27.7 *
Spouse and children 16.8 ** 19.1 7.1 26.2 **
Close kin and friends 6.0 * 7.3 4.8 12.1 *
Older diverse 6.9 5.8 6.2 11.9
Younger diverse 2.4 3.6 5.9 9.5 *
* p < .001.
** p < .01.
