The authors examined the independent associations of educational attainment and ethnicity with behavioral risk factors for cardiovascular disease using data from the 1989 baseline survey for the New York State Healthy Heart Program. This telephone survey used the Centers for Disease Control Behavioral Risk Factor Survey interview instrument and was conducted in eight communities (total population, approximately 1.24 million people) in New York State. The response rate was 65.5° (n = 4,179); 3,606 subjects aged 20–64 years with self-described ethnicity of white (n = 1,935), black {n = 1,035), or Hispanic (n = 636) and of known educational status were retained in the analysis. After adjustment for age, sex, and ethnicity, significant associations were found beween educational attainment and smoking, lack of regular exercise, overweight, diet atherogenicity, and knowledge about blood pressure and cholesterol. After adjustment for age, sex, and educational attainment, associations were found between ethnicity and most of these same variables. Blacks and Hispanics generally had less favorable risk factor profiles. These data indicate that the differences in cardiovascular disease risk profiles between whites and blacks or Hispanics cannot be fully explained by underlying differences in educational attainment. The differing patterns of risk factor distribution by educational attainment within ethnic groups have implications for the segmentation of risk reduction programs. Am J Epidemiol 1991; 134:567–82.