This study was done to examine the relationships among adiponectin and multiple risk factors for CVD and T2D in a large, racially diverse sample of youth, and to determine if relationships were moderated by gender, race and overweight status. The sample consisted of 1215 Black and White children and adolescents, aged 7-18 years, selected from 2211 subjects who participated in the Cardiovascular Health in Children III Study. Subjects who were selected had frozen serum samples available for analysis, and had complete data on the following variables: race, age, gender, pubertal stage, cardiovascular fitness, high density lipoprotein cholesterol (HDL-C), low density lipoprotein cholesterol (LDL-C), triglyceride, insulin, glucose, systolic and diastolic blood pressure, body mass index (BMI), waist circumference, and sum of skinfolds. SAS 9.1.3 was used for all statistical analyses. Generalized Estimating Equations were used to account for the increased correlation among subjects within school clusters. Adiponectin means were lower in Black subjects and in male subjects overall, and specifically lower in Black males than in other race/gender groups. Adiponectin means were also lower in subjects with a family history of diabetes, but there were no differences in adiponectin means by Tanner stage or fitness level. BMI z-score, waist circumference and sum of skinfolds were each inversely related to adiponectin, but BMI z-score and waist circumference provided the best models for prediction of adiponectin when compared by the Quasi-Likelihood in Independence Model Criterion (QIC). Adiponectin was positively associated with HDL-C and inversely associated with insulin and systolic blood pressure in multivariate regression with other risk factors, but the relationships with insulin and systolic blood pressure were dependent on one or more measures of adiposity. There was an interaction between HDL-C and gender, in that the relationship between adiponectin and HDL-C was stronger in female than in males. Insulin was only related to adiponectin in female subjects. In summary, BMI z-score and waist circumference are strongly related to adiponectin. Race, gender and HDL-C are consistently related to adiponectin, independent of adiposity measures, but insulin and family history of diabetes are also important variables to consider in the study of adiponectin.