Rainwater, Elizabeth
Working paper no. 2005-05
Abstract
There is a persistent relationship between early socioeconomic status and morbidity and mortality that is well documented in the literature. In this thesis, I explore two of the theories proposed in the literature to account for this relationship—Iaccess to resources and effects of general intelligence. I ask three main questions. First, does the inclusion of measures of early SES beyond family income add to a model of mortality? Second, do other measures of family context, such as number of siblings and intact family, which affect available resources, add to our understanding of the relationship between childhood context and adult mortality? And finally, which theoretical explanation, resources or intelligence, best explains the risk of mortality? To answer these questions, I analyze survival models of data from the Wisconsin Longitudinal Study, a cohort sample of adult, Wisconsin high school graduates that has been followed since 1957. I find that parental income is the only significant early SES variable in the model, and that the only other significant measure of family context is number of siblings. Both of these findings fit well with the resources argument, as parental income is the SES measure most closely associated with resources, and number of siblings indicates over how many individuals family resources must be spread. The effects of both indicators of general ability—IQ and grades—are negative for women, and neither is significant. Among men, the effects of IQ and grades run in opposite directions, with higher IQ increasing the risk of mortality net of grades, and higher grades decreasing the risk of mortality net of IQ. I believe that these results lend support to the access to resources hypothesis, and they do not tend to support the intelligence hypothesis.