Cross-validation of the NCAA method to predict body fat for minimum weight in collegiate wrestlers.

Abstract

In 1998, the National Collegiate Athletic Association (NCAA) adopted a new rule that required minimum weight testing for collegiate wrestlers. The objective of the study was to cross-validate the method used by the NCAA to estimate minimum weight in collegiate wrestlers. The NCAA skinfold equation was cross-validated against a criterion value from hydrostatic weighing (HW). SETTING The subjects were tested at the Universities of Wisconsin and Iowa. A sample of 93 college wrestlers from the Universities of Wisconsin and Iowa (mean +/- SD; age = 20.20 +/- 1.67 years, height = 171.98 +/- 6.63 cm, weight = 74.44 +/- 11.48 kg) were studied. Cross-validation included analysis of the standard error of estimate (SEE), total error (TE), and residual plots. The mean body fat from the NCAA prediction (10.61 +/- 3.58%) was not significantly different than HW (9.70 +/- 3.95%). The SEE was low (2.32%), and the TE was low (2.49%). The difference in methods was related to the size of the HW value. The residual plot (y = -0.26x + 3.45, R(2) = 0.198) suggests that fat is overestimated in the leaner wrestlers and underestimated in fatter wrestlers. The authors found the NCAA method to be a valid predictor of body fat in this sample of 93 collegiate wrestlers under the conditions of the study. Although some bias was seen across the range of fatness, these data support the NCAA method to estimate body fat in college wrestlers for establishment of minimum weight.