Obesity prevalence among inmates in the United States is unknown. Since incarceration disproportionately affects minorities, excluding inmates from surveys may bias national obesity estimates. Including inmates may also help explain racial obesity disparities among men. This descriptive study summarizes obesity prevalence among US male inmates and analyzes the effect of incarceration on national prevalence estimates. Data for male inmates came from the 2002 Survey of Inmates in Local Jails and the 2004 Survey of Inmates in State and Federal Correctional Facilities. Data for the non-incarcerated US adult male population came from the 2004 National Health Interview Survey. Self-reported weight and height data were analyzed from men aged 25-59 years for all surveys (obesity equaled BMI ≥ 30.0 kg/m(2)). Pooled inmate obesity prevalence was less than non-incarcerated estimates across all race/ethnic-education subgroups. However, unlike non-incarcerated estimates, inmates had obesity disparities between Hispanics and non-Hispanic Whites. Merging inmate and non-incarcerated estimates lowered obesity prevalence among men aged 25-39 with lower education levels. Merged estimates showed a positive obesity gradient within Whites by education. This study indicates that the exclusion of inmates from national obesity estimates leads to overestimates in obesity prevalence, particularly for low SES White and Black men.