This study was designed to determine prediction equations of body density (BD) for athletes using anthropometric variables and to examine validity of the prediction equations. The subjects were 211 male and 198 female athletes aged 18 to 22 years. The subjects were measured for standing height, body weight, skinfold thickness, girth as well as body composition. Body composition was estimated from densitometry using underwater weighing method and pulmonary residual volume measurement. Skinfold thickness was measured at 9 sites on the right side of the body with an Eiken-type ski nfold caliper and 7 measures of girth were taken using a cloth tape. All measurements were done three times from April to October : pre-, mid- and post-competitive season, in order to find out a suitable site reflecting body composition change. Using multiple regression analysis, equations to estimate BD were obtained from standing height, body weight, skinfold thicknesses and girths. The effective prediction equations for BD were as follows : For Males
ED=1.11104-0.00053 (sum of chest, abdomen and quadriceps skinfold thicknesses in mm) -0.00027 (waist girth in cm) .
R=0.851, SEE=0.0051.
For Females
BD=1.11861-0.00054 (sum of abdomen, triceps and subscapula skinfold thicknesses in mm) -0.00054 (waist girth in cm) .
R=0.826, SEE=0.0062.
A cross-validation analysis of these prediction equations for BD correlated highly with hydrodensitometrically determined BD (r=0.832, SEE=0.0053 for males and r=0.812, SEE= 0.0062 for females) . Thus the prediction equations developed in the present study will be applicable to athletes.