The effects of sociodemographic, familial, and residential variables in explaining mortality differentials: A comparison of the Dahrendorf and Weberian social class models

This study utilized the National Longitudinal Mortality Study (NLMS) to examine the effects of social class on the probability of dying using modified Dahrendorf and Weber social class models. More specifically family income and occupational prestige were used as the central variables in analyzing these respective models. The sample size for this study is comprised of 209,959 respondents who are in the age range 25--64. The two social class models were analyzed using multiple logistic regression and was analyzed separately for males and females. In addition, the additive and interactional effects of these variables were examined. It was hypothesized that the social class model of Weber (using occupational prestige) would be a better predictor of the probability of dying than the social class model of Dahrendorf (using family income). However, results from this study show that the Dahrendorf model proved to be the better statistical model for predicting the probability of dying. Although adjusting for the effects of race, family size, and urban residence reduced disparities between Blacks and Whites, these differences were not totally eliminated.