Objective: We compared adiposity from triceps and subscapular skinfold thickness (SF) with total body fat from dual energy X-ray absorptiometry (DXA) in their associations with HOMA insulin resistance for a large sample of US adolescents. Methods: We analyzed cross-sectional data from 1496 adolescents, ages 12.0-17.99 year examined as part of the US national health and nutrition examination survey (NHANES) cycles 2001-04. Data collected included skinfold thicknesses, DXA-based total body fat (DXF), serum insulin and fasting glucose for homeostasis model assessment of insulin resistance (HOMA-IR). Regression models adjusting for complex survey design effects and correspondence analyses were used to study associations with HOMA-IR and concordance between the two measures of adiposity. Results: SF and DXA models had identical precisions for estimating continuous HOMA-IR, based on bootstrap estimations. For boys, RMSEs = 0.53 (95% CI: 0.49-0.58) and 0.53 (95% CI: 0.48-0.57) for SF and DXA models respectively; and for girls RMSEs were SF = 0.45 (95% CI: 0.41-0.48) and DXA = 0.46 (95% CI: 0.42-0.48). Both adiposity models identified high proportions of the same individuals (positive percent agreement: 80-81%) with exact agreements exceeding 92% for identifying adolescents at elevated risk of IR in both sexes. Conclusions: Triceps and subscapular skinfold thicknesses estimate HOMA-IR and identify those at highest risk for insulin resistance as well as total body fat from DXA. Skinfold thickness provides an inexpensive and widely applicable measure of fatness that is appropriate for studies of insulin resistance and perhaps other metabolic variables in adolescents.