|
|
Matt A. Nelson, University of Minnesota
Coresidential kin in the United States have been studied extensively with long-run data availability with census and survey microdata. Data limitations hindered studying non-coresidential kinship networks, especially prior to 1950 in the United States. This paper leverages three linked census datasets to estimate the extent, structure, and prevalence of parents, children, siblings, grandparents, aunts and uncles, and first cousins geographic proximity for nearly 170 million unique individuals between 1850 and 1950. Individuals lived proximately to non-coresidential kin over the life course reflecting kin availability. Over the period, kin proximity declined from 34% in 1880 to 13% in 1950 in the same enumeration district (40-30% in the same place). Decomposition of the decline suggest that 95% of the decline between 1900 and 1950 can be described by declining preferences for living near kin (especially among farmers, non-migrants, whites, the Midwest and South, and women aged 20-49) with little explanation due to demographic shifts. Given the decline coincides with decreasing fertility and transformations in American family structure, the declining preference likely reflects larger attitudinal and economic changes in the United States during this time period. This paper will further discuss the limitations of these kinship networks for other analyses.
No extended abstract or paper available
Presented in Session 152. Understanding Kinship and Family Dynamics