There is growing concern regarding childhood obesity and its impact on children's health. Many states and school districts are mandating assessments because of obesity related concerns. Body Mass Index (BMI) has been used in schools as an educational or screening tool to communicate with parents about their children's risk status. There is concern, however, regarding the accuracy of BMI as a screening tool since it does not take into account an individual's muscle mass or bone density, which could lead to a bias toward over identification of African American children for risk status. Percent body fat, as estimated by skinfold measurements, is the recommended measure of body composition on the FitnessGram, while BMI is recognized as an acceptable alternative assessment. In light of the limitations associated with the use of BMI, the purpose of this study was to examine the consistency of risk identification when using skinfolds vs. BMI. Participants were 471 middle school physical education students (216 males, 255 females; 185 African American, 286 European American) enrolled in a public school in the southeastern United States. Height, weight, triceps and calf skinfolds were taken by qualified research assistants. BMI was calculated, and percent body fat was estimated using a prediction equation (Lohman, 1992). A 2 (gender) x 2 (race) MANOVA with BMI and percent body fat as the dependent variables yielded significant gender and race effects. Univariate follow-ups revealed that females had higher BMI's and percent body fat than males. For race, European Americans had lower BMI's than African Americans, but estimated percent body fat did not differ by race. Using the age and gender appropriate standards from FitnessGram, participants were classified as either in the healthy zone, or at risk for both body composition measures. A 2X2 contingency table was constructed to determine the consistency of the classification across the two measures. The risk classification for BMI and skinfold measures was the same for 88.9% (419) children, but the classification was inconsistent for 11.1% (52). Chi-square analyses indicated that a disproportionate number of African American children were identified as at risk using BMI. Although the consistency of risk classification using the two measures of body composition was high, these results suggest that BMI should be interpreted cautiously, especially for African Americans. This is an important consideration for physical education teachers as they strive to maximize the effectiveness of body composition measures as educational tools.Keyword(s): measurement/evaluation, multiculturalism/cultural diversity, obesity issues