Social Networks and Social Inequality in Later Life

Abstract

The projected growth of the older adult population in the United States coincides with deepening social inequalities. This dissertation is motivated by the observation that social network ties profoundly shape older adults’ quality of life, but that some of the most consequential features of these networks have yet to be incorporated into frameworks for understanding social stratification across the life course. In this dissertation, I examine how properties of older adults’ personal social networks intersect with dimensions of social inequality that are among the key social determinants of later life well-being. Each chapter uses data from the National Social Life, Health, and Aging Project (NSHAP), a population-based, longitudinal study of community-residing older adults in the United States, in a series of descriptive and multivariable analyses designed to assess the focal research questions. The first chapter explores how older adults’ personal networks are shaped by childhood circumstances. I find that higher parental socioeconomic status during childhood is associated larger, less kin-based, and more expansive personal networks, while higher levels of childhood family happiness are associated with denser and more kin-centric social networks in later life. The second chapter proposes that personal networks are a social context that carries implications for older adults’ perceptions of discrimination. I find that more kin-based personal networks predict less frequent experiences of discrimination; however, among black older adults who experience more frequent discrimination, more kin-based personal networks are associated with race-based discrimination. The third chapter argues that the stability of older adults’ personal networks is a function of residential neighborhood conditions, including concentrated disadvantage, residential instability, and social ties. I suggest that there may be nuances in how the stability of kin and non-kin ties are shaped by these conditions. I find that neighborhood concentrated disadvantage is associated with the loss of kin network ties over time, but that higher levels of neighborhood social ties predict higher rates of non-kin turnover – that is, the addition and loss of non-kin network members over time. I conclude the dissertation with a discussion of common themes and directions for future research that emerged across the three chapters.