This article compares the determinants of living alone in later life in Spain and Sweden, two countries with relatively similar levels of economic development from a global view point but different family systems and institutional contexts. With microdata coming from census (Spain) and linked administrative registers (Sweden), logistic regression techniques, including a nonlinear regression–based decomposition of differences between, are used to estimate the weight of different factors behind the residential choices of elderly women. Theoretical expectations are validated. Levels of living alone are associated with age, childlessness, marital status, and education in both populations. Population characteristics (compositions effects) explain only a small part of the differences in living alone between both countries, while behaviors (rate effects) account for the larger part of the variation. Therefore, among elderly women proximate determinants of living arrangements produce different outcomes in different sociocultural environments largely determined by existing family systems.