Background The relationships between environmental correlates of adolescent obesity are complex and not yet well defined by current research, especially when considering age and gender. Objective The purpose of this study was to test a model of proximal (home) and distal (neighbourhood) environmental correlates of obesity for adolescent age and gender groups. Methods This was a descriptive, cross-sectional study, using the 2007 National Survey of Children's Health of 39 542 children ages 11-17 years. Results The model fit the data well for early adolescents (ages 11-14 years) (root mean square standard error of approximation [ RMSEA] 0.040, 90% confidence interval [ CI]: 0.039-0.041; comparative fit index [ CFI] 0.947; Tucker- Lewis index [ TLI] 0.929) and middle adolescents (ages 15-17 years) ( RMSEA 0.037, 90% CI: 0.036-0.038; CFI 0.052; TLI 0.937). The model also fit the data well for boy adolescents ( RMSEA 0.038, 90% CI: 0.037-0.039; CFI 0.951; TLI 0.935) and girl adolescents ( RMSEA 0.038, 90% CI: 0.037-0.040; CFI 0.949; TLI 0.932). Conclusions All models provide loadings of the environmental correlates of adolescent obesity for specific age and gender groups that can be used for early identification of risks and targeted interventions. [ABSTRACT FROM AUTHOR]