Cohort Differences in Social Network Types of Older Adults: a Latent Class Approach
Year: 2018
Type: article
Abstract: Epidemiologic research, using unidimensional measures, demonstrates protective effects of social relationships for late-life health but fails to capture the complexity of older adults’ social networks. Our aim is twofold: (1) to develop a typology of older adults’ social networks using a multidimensional approach, simultaneously considering the structure, function, and relationship quality; and (2) to determine whether social network types of older adults differ by age cohort. We obtained data from the National Social Life, Health, and Aging Project, a longitudinal population-based study of older adults’ health and social lives. Latent class analysis was used to classify 3,005 adults (57–85 years) into distinct network types using nine observed social network characteristics. Stratified analysis was used to compare network types of the recent (75-85years-wave 3) and preceding cohorts (75-85years-wave 1). Five network types emerged in the full sample: diverse, supportive partner network (31%), average partner network (27%), large, supportive network without partner (16%), partner-centered network (15%), and restricted, family-centered network without partner (11%). In the stratified analysis, a four-class solution adequately captured the social networks of both cohorts. However, differences were observed in network type prevalences and the distribution of responses to the observed network characteristics. The recent cohort had fewer members in the large, supportive network without partner (14%vs.29%) and more members in the partner-centered network (34%vs.19%) compared to the preceding cohort. They also reported greater support from friends compared to the preceding cohort. Constructing network types can ultimately allow us to inform development of relevant risk assessment tools and interventions.
Source: Europe PMC (PubMed Central)
Institutions: University of Michigan, Institute for Social Research
Language: English
Subfield: Health
Field: Social Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W2907486654