The purpose of this study was to determine the validity and reliability of four methods of estimating body composition in 331 participants (177 boys, 154 girls) between 12-17 years of age. Percent body fat (%BF) was assessed once on one day using DEXA and twice using the sum of two skinfold (SF) and three bio-electrical impedance analysis (BIA) devices: OMRON hand-to-hand BIA, TANITA 521 foot-to-foot BIA, and TANITA 300A foot-to-foot BIA. The same assessments were repeated on 79 of the participants on a second day. DEXA was used as the criterion method of estimating %BF. The agreement between the estimates of %BF from the sum of two SF and the three BIA devices and DEXA was evaluated using linear regression and Bland-Altman analyses. Although the two analyses generally led to similar conclusions about each of the four prediction methods, the specific interpretations of each analysis varied because of the inherent differences in the analyses. In an attempt to determine if any of the four prediction methods were interchangeable with DEXA, the 95% confidence interval (CI) and prediction interval (PI) around the line-of-best-fit through the data are reported. The results of this study indicate that (a) all of the methods used in this study to estimate %BF were reliable within and between days, (b) the TANITA 300 BIA device performed poorly in both boys and girls and should not be used to assess body composition in children and adolescents, (c) none of the four prediction methods performed well in both boys and girls across the entire range of %BF values of the subjects in this study, (d) the sum of two SF, OMRON and TANITA 521 are acceptable for use in large population-based studies but are not recommended when the accurate assessment of body composition of an individual is critical, in which case (e) criterion methods of assessing body composition should be used.



College and Department

Life Sciences; Exercise Sciences



Date Submitted


Document Type





Body composition, DEXA, adolescent, electrical impedance, child, skinfold