The educational gradient in health is a well-established phenomenon. Despite receiving a great deal of scholarly attention, questions about the universality of the relationship remain. Specifically, there is relatively little research investigating whether the relationship extends to racial/ethnic minority populations. Also, the mechanisms through which education improves health are not well understood. This dissertation adds to our understanding of the educational gradient in health by focusing on two questions. First, does education affect health equally across racial/ethnic groups? And, second, do the intervening variables that connect education to health operate equivalently across groups?
Using data from the National Comorbidity Survey Replication, the National Survey of American Life and the National Latino and Asian American Study, I analyze the health returns of education across racial/ethnic groups in the United States. I conduct analyses at both the pan-ethnic level which includes non-Hispanic whites, non-Hispanic blacks, Hispanics and Asians and at the more specific race/national-origin groups. I employ various methodologies, including logistic regression models and structural equation modeling, to test hypotheses about how and why educational attainment affects health across groups.
Taken together, I find that in general, educational attainment is positively related to better self-rated health. However, interesting racial/ethnic differences emerge, both at the pan-ethnic and at the national-origin level. I also find that while the mechanisms linking education and health are similar across racial/ethnic group, the magnitude of effects varies significantly across groups. I therefore conclude that education does not equally affect health across groups.