Content area
Abstract
This thesis grows out of a long-standing research interest in the association of health status and survival outcomes with marital status. In particular, we have been troubled by the lack of adequate research on the underlying processes that produce the observed marital status differentials in health. In this analysis, we focus on marriage selection on the basis of health as a logical starting point to identify the relative importance of marriage selection and marriage protection. Our goal is two-fold: first, we extend the conventional argument to recognize that marriage selection may operate on the basis of more broadly defined health-related characteristics and behaviors, rather than simply on severe physical and mental handicaps; and second, we broaden the existing marriage choice model by taking into account health components as either direct selection criteria or as mediating factors through which individual and family socioeconomic characteristics affect marriage behaviors.
We address these two issues by examining the effects of health-related characteristics and behaviors on first marriage rates among young American adults based on data from the National Longitudinal Survey of Youth 1979-1991. Using event history analysis, we investigate the overall association between each of the health-related variables and marriage rates, and the effects of these health-related variables net of the influence of other health variables and of various socioeconomic factors. We also examine possible gender differentials and age effects of health-related characteristics and behaviors on marriage rates.
The findings suggest that first marriage is selective on the basis of health among young adults. Specifically, marriage selection results in lower marriage rates for (1) persons with certain physical characteristics, namely obesity and short stature; and (2) for persons with unhealthy behaviors, such as heavy alcohol consumption and use of hard drugs. In contrast, the association between the presence of health limitations and first marriage rates is modest and statistically insignificant.
Our results also provide evidence that the specific pathways through which marriage selection operates differ somewhat by gender and that the effects of certain health-related behaviors on marriage rates vary by age.
