To examine baseline fat-related dietary behaviors of white, Hispanic, and black participants in Minimal Contact Education for Cholesterol Change, a National Institutes for Health-funded cholesterol screening and education project conducted in New England. A sample of 9,803 participants who joined the study at baseline (n=7,817 white; n=1,425 Hispanic; and n=561 black). Participants completed baseline questionnaires that included demographic and psychosocial items as well as the Food Habits Questionnaire, a dietary assessment tool measuring fat-related dietary behaviors. They also had their blood cholesterol level and height and weight measured. Analysis of variance (ANOVA) was used to compare racial/ethnic groups on continuous demographic variables, and the chi(2) test of association was used to compare groups on demographic categorical variables. Multivariate analysis of variance (MANOVA) was used to compare mean differences between racial/ethnic groups on six behavioral subscales (Fat Factors) differentiating domains of behavior related to fat intake and to compare 27 individual fat-related dietary behaviors. After adjusting for sex, age, marital status, education, employment status, and percent time lived in the United States, white participants had the lowest Food Habits Questionnaire summary score (2.44) (indicating a lower fat diet), followed by Hispanic (2.61) and black (2.68) participants. The three ethnic groups also differed on the prevalence of Fat Factors and specific fat-related dietary behaviors. White participants were more likely to use lower-fat alternatives, to avoid frying, to replace meat, and to modify meat to make it lower in fat. However, they were least likely to eat fruits and vegetables for snacks and desserts. Hispanic participants were more likely to engage in fat-avoidance behaviors and to eat fruits and vegetables for snacks and desserts. Black participants were less likely to eat meatless meals and modify meats to make them lower in fat. Black and Hispanic participants were more likely than white participants to fry foods. Hispanics were less likely to read food labels for nutrition information. The most and least prevalent fat-related behaviors also differed by each ethnic group, showing that different behaviors were more and less easily implemented by each ethnic group. The results of this study suggest that there is a need for improvement in dietary behaviors related to fat intake, especially for blacks and Hispanics, and that the specific dietary behavior issues differ widely by ethnicity. These results can be used by nutrition educators and researchers to help them decide what messages to emphasize in dietary counseling, nutrition education programs, and materials. The results can also be used to help design better dietary assessment tools and more effective interventions for culturally diverse populations.