The object of this research is to examine the effect of education on mortality using age-race-sex-education specific death rates. From 1989, educational attainment has been added to standard death certificate. Combining death counts from death certificates and population estimates derived from the Census Public Use Microdata Sample (PUMS), I computed death rates by age group, by race, by sex, and by educational attainment. Two time periods are examined: the 1990 and the 2000.

By fitting Poisson distributed death rates into a Log-Rate model, I tested the main effect and the mortality differentials. Empirical analysis confirms that mortality risk is higher for less educated population; empirical analysis also shows that men face higher mortality risk than women and blacks face higher mortality risk than whites. There exists race difference between blacks and whites on the impact of education in reducing morality risk. This study confirms the importance of education and may contribute to similar mortality studies using Poisson regression models.