The purpose of this study was, using compositional data analysis, to examine dietary composition patterns of obese women. Based on their percent body fat (BF%), two groups of women, 50 each, were selected from an urban women health study (N=480). The obese group included 50 women with the highest percent body fat (M±SD: Age=54.06 ± 7.57 yrs; BF%=48.55 ± 2.97) and the "control" group included 50 with the lowest body fat (Age=52.36 ± 8.02 yrs; BF%=31.48 ± 5.59). The food-frequency data were collected using the Block Dietary Data Systems for individual diet analysis and body fat percentages were measured using the BodPod body composition system. The diets were comprised of three categories, calories from fat (CFat), calories from protein (CPro), and calories from carbohydrates (CCarb). Individuals' total caloric intakes were computed by summing the categories. CFat, CPro, and CCarb were then converted into the percentages of the total caloric intake. The percentages were analyzed using the compositional data analysis to determine whether there was a group (obese vs. control) difference in dietary composition. Ternary diagram was employed to examine the compositional pattern of the data. The results of the logistic normal distribution tests (i.e., Anderson-Darling, Cramer-von Mises, and Wetson tests) showed that the data met the assumption of logistic normal distribution. Ternary diagrams indicate that there was no major difference in the composition of diet between the obese and control groups. Both groups consumed a greater percentage of carbohydrates (M±SD: obese=50.4±6.7%, control=51.1±7.2%) than protein and fat. Both groups consumed similar percentages of protein (M±SD: obese=17.4±3.7%, control=17.6±3.7%). The percentage of fat was also comparable (M±SD: obese=32.1±7.5%, control=31.3±7.6%). The results of the partial least square and maximum likelihood estimation tests also confirmed that there was no statistically significant difference between the groups (loglikelihood statistic=468.37, p>.05) in the dietary composition. In conclusion, the reported dietary consumptions of obese women in this study were very similar to those, who had lower body fat in the control group. It indicates that other variables in the diet (e.g., the percentage of the types of fat intake), or the amount of exercise, may be more important factors in determining body fat percentages.