EVALUATION OF SKINFOLDS MEASUREMENT IN ESTIMATING BODY FAT OF CHINESE CHILDREN

Abstract

The purpose of the study was to evaluate skinfolds prediction models in estimating body fat (%fat) of Chinese children age 8 to 13 years old. A total of 141 Chinese children (66 girls, mean age = 10.39+/−1.11 yrs.; 75 boys, mean age = 10.60+/−1.29 yrs.) recruited from local primary schools participated in this study. Triceps, subscapular, and calf skinfolds measurements were taken from children following standard protocols. Percent fat was then computed from published equations developed by previous researchers. The %fat criterion was measured by hydrostatic weighing, with residual volume was measured by oxygen dilution method. Regression analysis found that %fat estimated from the sum of triceps and calf equations (Slaughter et al., 1988) shared small variances with the criterion with large standard error of estimates (SEE) for both boys (R2 = .25, SEE = 8.02%) and girls (R2 = .21, SEE = 6.71%). Subsequent stepwise regression analyses indicated that the best prediction model for boys was a linear combination of triceps and subscuplar (R2 = .71, SEE = 4.99%), and for girls was a linear combination of calf and subscuplar (R2 = .64, SEE = 4.49%). The SEEs were 37.78% and 33.08% lower than the Slaughter equations for boys and girls, respectively. When only triceps was used for boys and only calf was used for girls, the second best models were found (R2 = .68, SEE = 5.18% for boys; R2 = .53, SEE = 5.07% for girls). In conclusion, when estimating %fat of Chinese children from skinfolds, the suggested models were triceps and subscapular for boys (%fat = 1.145+0.952*triceps+0.545*subscapular), and calf and subscapular for girls (%fat = 3.794+0.494*calf+0.933*subscapular). When subscapular is eliminated, only triceps for boys (%fat = 0.31+1.481*triceps) and only calf for girls (%fat = 2.577+1.277*calf) would be sufficient to produce satisfactory body fat estimation.