Purpose: The purpose of our study is to provide useful information for me-asurement methods of childhood obesity and to provide clinical significance of bioelectrical impedance analysis. Methods: We measured height, weight, waist to hip ratio, skinfold thickness and bioelectrical impedance from 206 elementary students in Kwangju from May to July 1997. Results: In males; 1) There were statistically high correlations between tri-ceps skinfold thickness and body mass index (BMI), percentage of body fat (%BF), weight, Rohler index (RI), total fat weight (TFW), waist to hip (W/H) ratio, height, lean body mass (LBM) (all, p<0.001). 2) There were statistically high correlations between subscapular skinfold thickness and BMI, RI, weig-ht, %BF, TBW (all, p<0.001). 3) There were statistically high correlations between %BF and BMI, triceps skinfold thickness, RI, TFW, subscapular skinfold thickness, weight (all, p<0.001). In females; 4) There were statistically high correlations between triceps skinfold thickness and BMI, weight, LBM, height, %BF (all, p<0.001). 5) There were statistically significant correlations between subscapular skinfold thickness and BMI, weight, LBM, RI, height. 6) There were statistically high correlations between percentage of body fat and BMI, RI, triceps skinfold thickness. The percentage of body fat according to the bioelectrical impedance analysis highly correlated with that calculated by skinfold thickness in males (r=0.57, p<0.01) and in females (r=0.30, p<0.01). Conclusion: The results of our study suggests that bioelectrical impedance analysis is a useful method of estimating body fat in the childhood obesity.