Four methods of assessing body composition and percent body fat, hydrodensitometry (HD), bottle buoyancy (BB), ultrasound (US), and anthropometry, were administered to 64 Minnesota Olympic development female ice hockey players. Anthropometry included three previously developed regression equations; two sport-specific (Going et al. & Meleski et al.) and one generalized equation (Jackson, Pollack, & Ward). Mean age was 16.0 years, mean height 165.5 cm, and mean weight 60.8 kg. Subjects included 8 goalies, 22 defensive players, and 34 offensive players. No significant differences were found among positions for height or weight. Defensive players' skinfolds were significantly larger than offensive players (p .0007).; Using hydrodensitometry as the criterion, repeated measures analysis of variance and Bonferroni/Dunn post hoc test revealed ultrasound (p = .0013), Going et al. formula (p = .0001), and Jackson et al. formula (p = .0001) to be significantly different from hydrodensitometry and each other. No significant differences were found between hydrodensitometry and bottle buoyancy (p = .8920), hydrodensitometry and Meleski et al. formula (p = .0347), and bottle buoyancy and Meleski et al. formula (p = .048). Means (M), standard deviations (SD), correlation coefficients (R), standard errors of estimate (SEE), and total errors (TE) for percent body fat were as follows: HD (M = 17.54%, SD = 6.58%); BB (M = 17.41%, SD 7.04%, R = .946, SEE = 2.28% TE = 1.24%); US (M = 15.24%, SD = 2.98%, R = .544, SEE = 2.50%, TE = 20.33%; Jackson et al. (M = 20.16%, SD = 5.22%, R = .670, SEE = 3.88%, TE = 20.99%); Meleski et al. (M = 18.55%, SD = 4.19%, R = .724, SEE = 2.89%, TE = 8.09%); Going et al. (M = 23.2%, SD = 5.85%, R = .723, SEE = 4.04%, TE = 45.27%).; Stepwise regression analysis performed on anthropometric data resulted in the following formula: Body density = 1.145 - .003 (bicep skinfold) + .001 (waist circumference). R and SEE for the formula were .783 and .008 respectively.
展开▼
