|
COPYRIGHT 2000 JAI Press, Inc.
INTRODUCTION
The social networks in which older people are embedded constitute a major source of personal well-being and a principal resource for personal care in later life (Gallo 1984; Wenger 1984). One of the more important aspects of the social network phenomenon addressed in gerontological research in the recent past has been the effect of different forms of network structure or social network type (Litwin 1995; Wenger 1991). Indeed, studies of elderly cohorts in a number of countries have found a significant relationship between network type and a range of support and well-being outcomes (Litwin 1996; Scharf and Wenger 1995).
The study reported upon in this article broadens the current state of inquiry regarding the network types of older people in a number of ways. First, it focuses upon the networks of the old-old--persons age 75 and older. Despite the fact that network support is perceived as particularly important in very old age, the social networks of the old-old have not been widely addressed in the research literature (Bowling and Browne 1991).
Second, the article examines the application of a social network inventory not previously employed in the study of elderly samples (Cochran et al. 1990). The current analysis considers both logistic and methodological implications of this network inventory in the context of research on the old-old. Moreover, specific attention is paid to the capacity of the instrument to accurately tap the phenomenon of social support among elderly respondents.
Third, the inquiry considers a series of personality factors not taken into account in previous research in this specific area; that is, factors that may potentially confound the observed influence of network type. The personal coping resources and the coping styles of older adults may stand behind the formation of their social network as well as behind the extent of social support they currently receive.
The present investigation is based on data drawn from a study of European-born elderly persons residing in Tel Aviv and the surrounding vicinity. The inquiry examines whether different social network types may, indeed, be discerned among persons age 75 and older, and if so, whether differing network types are related to different levels of social support. In addition, the analysis explores whether the network type construct makes an independent contribution to explaining social support, even when taking into account the effect of older people's coping resources and coping styles.
LITERATURE REVIEW
Network Type
The importance of a social network for the well-being of older people has been widely documented (Auslander and Litwin 1991; Ell 1984; Gallo 1984; Gottlieb 1981; Levitt et al. 1985-1986). On the other hand, an interesting aspect of the network phenomenon that has received less attention in the literature is the notion of network type. This construct reflects the particular structure or "constellation" of the interpersonal groupings in which older persons are embedded. Initial inquiry indicated that differing network types tend to have differing effects on their focal older members (Wenger, Davies, and Shahtahmasebi 1995).
Network types among elderly cohorts have been derived on the basis of varying criteria. These have included network size and tie multiplexity (Mugford and Kendig 1986), availability of close kin, level of involvement of family, friends and neighbors, and degree of interaction with community groups (Wenger 1991), and role relationship categories (Auslander and Litwin 1990). A recently derived network typology among elderly immigrants utilized six structural criteria for delineation: size, composition, percentage of intimate ties, frequency of contact, duration of ties, and geographic proximity (Litwin 1995).
Despite the different bases for the delineation of network types, a degree of commonality may be observed among several network typologies. In this regard, Wenger (1996) cited the presence of both narrow family-based and wide family-based networks, friend- or neighbor-based groupings, wider community-focused networks, and private restricted networks. Correspondingly, a recent secondary analysis of a probability sample of persons age 60 and over in Israel derived a somewhat similar set of network types (Litwin 1997).
The Measurement of Social Support
The notion of network type is significant, in that it implies differential availability of social support according to one's particular network constellation. Social support is defined as the range of interpersonal aids that people require for daily functioning such as augmentation of self-concept, sense of belonging, cognitive guidance, concrete assistance in fulfilling tasks, and feeling loved and admired (Litwin 1996). Such supports have been classified into three main categories of assistance: affective, affirmative, and instrumental (House and Kahn 1985).
A major issue in the study of the relationship of network type and social support is how each of these phenomena is measured. Social network inventories generally tap both the network's structure and its content; that is, provision of support (Norbeck, Lindsey, and Carrieri 1981). One particular inventory incorporates structure and content within a single measure of network connectedness (Lubben 1988). Another addresses structural and interactional aspects of the network only (Wenger 1991). Each type of network inventory has both strengths and limitations.
The Norbeck Social Support Questionnaire (NSSQ), for example, asks respondents to name the persons important to them, and then collects the characteristics of each. The instrument also operationalizes the three support categories--affective, affirmative, and instrumental--asking respondents the extent to which they would turn to each person listed on a series of needs. The NSSQ has served as the basis for both network type delineation and study of the supportiveness of different network types (Litwin 1995; Auslander 1996).
Two problems arise from this approach to the measurement of social support. Insofar as all persons listed receive a score on each support task, the instrument confounds support with network size. Although the proffered solution to this dilemma is to average support scores by network size, this calculation may distort the support outcome in a different manner. The second problem also arises from seeking a support score for each person on each task. When faced with having to rank the supportiveness of their children, for example, older respondents may feel compelled to give a high rating, even if the children are not, in fact, supportive.
A network inventory that overcomes both these problems has been applied in a study of young families (Cochran et al. 1990). Like the NSSQ, this instrument generates a name set for each respondent...
Read the full article for free courtesy of your local library.
|
Bookmark this article
Print this article
Link to this article
Email this article
Digg It!
Add to del.icio.us
RSS
