Erik Hernandez , Pennsylvania State University
The paper examines the relationship between education and type 2 diabetes prevalence among four cohorts at ages 60-64 (born 1935-39, 1940-44, 1945-49, and 1950-54). The analysis involves logistic regressions and Fairlie decomposition analysis to measure the effect changing educational composition across these cohorts. First, results indicate a stable effect of education on diabetes prevalence across cohorts, that is, the education gradient did not change. Persons with higher levels of education (a bachelor degree) were less likely to be diagnosed with diabetes than persons with lower levels of education, for each cohort. Second, a changing educational composition, that is, increases from one cohort to the next, affected the level of diabetes prevalence. Later cohorts experienced greater levels of educational attainment, and these higher attainments resulted in lower diabetes prevalence than would have been the case if there had been no educational expansion.
Presented in Session 5. Health & Mortality 1