Age differences in the effects of network composition on psychological distress

Abstract

The main goal of this research is to better understand age differences in the effects of social networks on mental health. Using a social network approach to revise the convoy of social support model (Kahn and Antonucci, 1980), we examine specifically how two aspects of social support networks (kin composition and convoy dimensions) influence psychological distress for older and younger samples (18–59 and 60+). We hypothesize that kin composition will influence distress in general but especially for the younger sample, while two competing hypotheses for the convoy of social support model are tested. Using data from a three-wave panel health study, structural equation models (LISREL 8.20) indicate that a greater proportion of kin in the perceived support network and the presence of family members in the inner circle of the convoy significantly reduce distress, primarily for the younger sample. Implications for the convoy model are discussed.

Introduction

Though the importance of networks has long been acknowledged, researchers have become increasingly aware of the contribution of interpersonal ties to the health and well-being of individuals (e.g., House et al., 1988 for reviews). The social networks of individuals, or the direct and indirect ties linking people together through such relationships as kinship or friendship, provide many important services (Lin et al., 1981. Indeed, in the past twenty years, the importance of social networks has been established in research on numerous outcomes, including social support (Kadushin, 1982), occupational status and employment (Granovetter, 1973, 1982; Lin, 1982), power and influence in organizations (i.e., Cook and Emerson, 1978), physical and mental health (Berkman and Syme, 1979; Berkman, 1984; George, 1989; Haines and Hurlbert, 1992), and health care utilization and help seeking behaviors (Horwitz, 1977; Pescosolido, 1991, 1992, 1996). Network ties have often been described as social resources that provide critical social support and information (Campbell et al., 1986; Lin, 1982; Pescosolido, 1996).

Interest in social networks has gained particular momentum in the field of mental health. Research on the effects of social support and social networks on depression, life satisfaction, and other aspects of well-being has increased substantially (e.g., George, 1989; House et al., 1988). In particular, social networks have become especially relevant to research on older individuals. The importance of informal networks in, for example, the provision of care to the aging population has been well-documented (Stoller and Pugliesi, 1991). However, the specific ways in which social networks influence the mental health of older individuals and possible explanations for these effects deserves more attention than received thus far in the research literature.

The term ‘social network’ tends to be used interchangeably with ‘social support’ (House et al., 1988) with little attempt to truly differentiate these two diverse concepts. In general, the distinction between social networks and social support is that networks are the structures through which social support, the functional behavior, is provided (Antonucci, 1985). Social support can be conceived as being embedded within the social network and as being one function of the social network (Lin and Ensel, 1989). One important research issue is how to capture both the social network and social support, or the structure and the function, simultaneously. This research focuses on one way in which both network and support features are addressed. We apply the convoy of social support model (Kahn and Antonucci, 1980) to a social network perspective to examine age differences in the effects of network composition on psychological distress.

The convoy of social support model (Antonucci, 1985; Antonucci and Akiyama, 1987; Kahn and Antonucci, 1980) provides one approach through which to integrate social support and social networks. Originating from both social role and social support theories, Kahn and Antonucci combine these two perspectives to better examine the varying effects of social support across the life course. The term ‘convoy’, which was borrowed from anthropologist David Plath (1975), refers to a protective layer of family and friends who surround an individual throughout the life course and help ‘negotiate life's challenges’ (Antonucci and Akiyama, 1987). By its nature, a convoy implies support provided by a set of others who are related to a focal person or ego. In other words, an individual moves throughout life encompassed by groups of people to whom that person is related through the exchange of social support. Theoretically, these layers of support are dynamic and, at the same time, lasting throughout life. That is, convoys should change in some ways but remain stable in others. For instance, parents and children typically remain in each other's support networks. On the other hand, friends, co-workers, and neighbors may be continue to be part of the support convoys or may drop out, depending on the role that the central person occupies.

Kahn and Antonucci (1980) present a metaphor of a convoy as a set of three concentric circles that surround a person throughout his/her life. Each circle represents a different level of closeness to the individual, with the inner circle symbolizing those toward whom the focal person feels very close. These relationships in the innermost convoy circle are the most stable and, hypothetically, the least bound by role requirements, while those in the second and third convoy circles are somewhat dependent on role relationships and are more likely to change over time.

In applying the convoy model to age differences with respect to network effects on psychological distress, it is important to develop two concepts that integrate social networks and social support: kin composition and the inner circle of the convoy. Kin composition refers to whether support providers are related to the focal person. This is one of the important features of a support network because, typically, immediate living kin are important parts of individuals’ support networks (e.g., Horwitz, 1977; Wellman and Wortley, 1989). Network research indicates that most, though clearly not all, of the networks with higher proportions of kin are supportive (Horwitz, 1977; Litwak, 1989; Wellman, 1990; Wellman and Wortley, 1989). Almost all parents and children provide emotional support (though more from mothers and daughters). Immediate kin also tend to be the primary service providers, which is not necessarily true for extended kin (Wellman and Wortley, 1989). For instance, research indicates that family is the primary source of support during a health crisis (Wellman, 1990). Typically, it is only when family members are not available that friends and/or formal services are turned to for help (Litwak, 1985, 1989).

Kinship ties are also better able to remain active and intimate over greater distances than friendship ties (Wellman and Wortley, 1989). In other words, friends must reaffirm their ties continually, while kinship ties are relatively reliable and stable without needing reciprocity (Wellman, 1990; Wellman and Wortley, 1989). Thus, network research suggests that the kin versus non-kin distinction within the support network is important when an aspect of mental health is the outcome.

The second concept through which to integrate social support and social networks is the inner circle of the convoy. The convoy model suggests that it is the placement or presence of people in the support network rather than their role relationship to the respondent that influences mental health. The particular aspect of convoy composition that is a primary interest in this research is the inner circle of support. For instance, if a spouse is not included in the inner circle, then it makes no difference that he or she is technically ‘kin’ in the support network; the spouse relationship is only important to the extent that it is considered very close and ‘supportive’ by the respondent. The expectation is that the people in the inner circle make the largest difference in mental health because these people are assumed to be the most important support providers (Antonucci and Akiyama, 1987). Following this supposition, the argument of the convoy model, as far as mental health is concerned, is that the relationship of the support provider to the focal person is not as crucial to psychological distress as where that support provider is placed in the circles of the convoy.

Both kin composition and placement in the inner circle measure the strengths of relationships: one with role relationships and the second in terms of position or location of the relationship in the convoy. Both concepts, precisely because they address strengths of relationships, should influence mental health, which in this study is measured in terms of psychological distress. Thus, the following hypotheses will be evaluated in this research:

Hypothesis 1

Kin composition of the perceived network will have a negative effect on psychological distress.

Hypothesis 2

Relationships present in the inner circle of the convoy will have a negative effect on psychological distress.

The general hypotheses regarding the effects of kin composition and inner circle placement on distress do not necessarily apply to all ages. One reason why it is important to examine the varying effects by age of network dimensions is that previous research has indicated age differences in distress or depression. The relationship between age and depression is complex, but, very briefly, older adults report more depressive symptoms than younger adults (e.g., George, 1989). However, older adults are less likely to actually experience a depressive disorder (see George, 1989 for reviews). The relationship between age and depression is also curvilinear in nature in that younger and older adults experience more depressive symptoms while depression declines in middle-age adults (Lin et al., 1989; Mirowsky and Ross, 1992). The decrease of depression in middle adulthood and increase in late life may reflect life-cycle gains and losses in employment, marriage, and socioeconomic status (Mirowsky and Ross, 1992).

Another reason why it is important to examine age differences in the effects of kin composition on psychological distress is that expectations for help from kin may vary depending on age of the respondent. For instance, reciprocity, may be of increased importance to older individuals. Research has indicated that older individuals typically do not wish to receive support unless they can reciprocate (Krause, 1990; Lee, 1985). The receipt of care from family members and/or the inability to return support may ultimately lead to dependence and influence mental health (Krause, 1987). Furthermore, other researchers argue that the effects of support differ depending on whether the support is provided by family or friends (Litwak, 1985; Litwak, 1989), particularly for the older population. Support from family tends to be non-voluntary, while support from friends typically is voluntary and based on mutual gratification (Norris et al., 1994). Thus, the provision of support by family may be evaluated as fulfillment of an obligation and have little effect on well-being for older individuals, while friend support may be an affirmation of the importance of the recipient to the provider and thereby be evaluated positively and have positive effects on well-being (Norris et al., 1994). Therefore, due to depression varying by age, as well as help from kin potentially varying by age group, it is likely that the effect of kin dimensions of the support network on depression will also vary by age. Specifically, we hypothesize that:

Hypothesis 3

There are age differences in the effects of kin composition on distress such that higher kin composition will lower distress only for younger individuals.

There are two competing age-specific hypotheses for the effects of the inner circle placement on distress:

Hypothesis 4

There are no age differences in the effects of inner circle placement on distress.

Hypothesis 5

There are age differences in the effects of inner circle placement on distress such that having kin relationships in the inner circle of support will decrease psychological distress only for younger individuals.

There may be age differences in the composition of the inner circle (in that there may no longer be spouses included, and there may be more friends and other family members), but if the argument is that those who are very close help with problems and having people who are very close lowers distress, then age differences should be not be significant. Furthermore, the relationships in the inner circle should not be relevant. However, if the logic of kin composition of the support network is accurate, then having kin relationships present in the inner circle of support will decrease distress only for younger individuals.