Objectives: To assess the nutrition knowledge levels and dietary intake pattern of schoolchildren belonging to two groups of different socio-economic status (SES; high income/high SES and low income/low SES). Design: A purposive sampling method was employed. A validated food-frequency questionnaire was administered to assess the dietary intake of schoolchildren in four schools from two different socio-economic strata in the month of January 2001. The children were divided into two groups, one serving as the experimental group and the other as the control group. Subjects: Two hundred and seventy-two children aged between 12 and 14 years. Results: There was a significant improvement (P , 0.001) in the knowledge levels of high-SES schoolchildren as compared with low-SES schoolchildren. A significant difference was observed in the intake of protective foods like milk and milk products, green leafy vegetables and fruits between the two income groups. However, children from the high SES background preferred fast foods such as noodles and corn flakes to traditional foods. Irrespective of income group, most of the children consumed carbonated beverages. Conclusions: There was a significant difference in the intakes of protective foods and fast foods between the different income groups. However, the increased intake of fast foods and carbonated beverages by the children irrespective of SES needs to be discouraged as a part of nutrition education. The study indicated the need for repeated interventions for improvement of nutrition knowledge levels in low-SES children. KeywordsQualitative assessment Food-frequency questionnaire Socio-economic factors Traditional foods Nutrition educationIt is important to study the existing pattern of dietary intake in a given population to determine the food preferences, beliefs and food fads of the community, so as to plan an appropriate nutrition education package. Estimation of the community's dietary intake and precision in the collection of data depend on the resources available and the characteristics of the study population. Data on national food availability and food consumption patterns of communities provide information on dietary intake for large populations. Diet records, dietary recall methods and food-frequency questionnaires (FFQs) are common tools employed to measure the food consumption patterns of individual subjects in the epidemiology of diet-related diseases 1 . The food frequency method involves the collection of information about a list of food items for which average frequency of consumption is determined with reference to a specified period in the past. It requires neither specialised training for respondents nor a lengthy interview, and hence places a lighter burden on respondents, which encourages their participation.Compared with diet history and 24-hour recall methods, the FFQ method requires less specialised interviewer training. It can be undertaken by face-to-face personal interview, by telephone or by self-administration including mail surveys 2,3 ...