Food frequency questionnaire (FFQ) is a practical tool for the measurement of usual food intake in large surveys because it gives a quick approximation to 'true' dietary intake. This study was carried out to compare the semiquantitative FFQ with three day 24-hour diet recalls (24-hr DR) in assessing intake of energy, total fat, fatty acids and vitamin A, C and E among Malaysian women. This semi-quantitative FFQ which was developed specifically for the Malay and Indian ethnicities has 200 food items and categorized according to three mealtimes namely breakfast, lunch or dinner and morning or afternoon snacks. A total of 51 Malay and 28 Indian women aged between 30 to 60 years were selected as study subjects. The result of the study shows that majority of study subjects were within the normal EI/BMR ratio when their energy intake was assessed by semi-quantitative FFQ (70%) and 24-hr DR (74%). However, 10% of study subjects became over-reporters when their intakes were assessed using the semi-quantitative FFQ. Analysis of t-test shows there is no significant difference (p > 0.05) on the mean intake of energy, total fats, saturated, monounsaturated and polyunsaturated fatty acids and vitamin A, C and E between semi-quantitative FFQ and 24-hr DR. Percent mean differences were also less than 10% for all nutrients included in this study. This indicates that the semi-quantitative FFQ can produce comparable results with 24-hr DR. Energy adjusted correlation coefficient values for all studied nutrients were total fat (r = 0.64, p = 0.02), saturated fatty acids (r = 0.59, p = 0.01), monounsaturated fatty acids (r = 0.52, p = 0.03), polyunsaturated fatty acids (r = 0.57, p = 0.02), vitamin A (r = 0.69, p = 0.01), retinol (r = 0.55, p = 0.01), beta carotene (r = 0.74, p = 0.01), vitamin C (r = 0.64, p = 0.02) and vitamin E (r = 0.69, p = 0.01). Cross-classification for both methods into quartiles of intake resulted in correct classification into the same or adjacent quartile from 82% to 96% of the study subjects. Only 3% of the subjects were grossly misclassified. As a conclusion, this semi-quantitative FFQ gives estimation as good as 24-hr DR for intakes of energy, total fat, fatty acids and vitamin A, C and E among Malaysian women specifically for the Malay and Indian ethnicities. This semi-quantitative FFQ is a useful tool in dietary intake assessment for research use especially for epidemiological study on diet and disease relationship such as cardiovascular, cancer and diabetes.