Abstract Background Tuberculosis (TB) remains one of the most important infectious diseases worldwide. A comprehensive approach towards disease control that addresses social factors including stigma is now advocated. Patients with TB report fears of isolation and rejection that may lead to delays in seeking care and could affect treatment adherence. Qualitative studies have identified socio-demographic, TB knowledge, and clinical determinants of TB stigma, but only one prior study has quantified these associations using formally developed and validated stigma scales. The purpose of this study was to measure TB stigma and identify factors associated with TB stigma among patients and healthy community members. Methods A cross-sectional study was performed in southern Thailand among two different groups of participants: 480 patients with TB and 300 healthy community members. Data were collected on socio-demographic characteristics, TB knowledge, and clinical factors. Scales measuring perceived TB stigma, experienced/felt TB stigma, and perceived AIDS stigma were administered to patients with TB. Community members responded to a community TB stigma and community AIDS stigma scale, which contained the same items as the perceived stigma scales given to patients. Stigma scores could range from zero to 30, 33, or 36 depending on the scale. Three separate multivariable linear regressions were performed among patients with TB (perceived and experience/felt stigma) and community members (community stigma) to determine which factors were associated with higher mean TB stigma scores. Results Only low level of education, belief that TB increases the chance of getting AIDS, and AIDS stigma were associated with higher TB stigma scores in all three analyses. Co-infection with HIV was associated with higher TB stigma among patients. All differences in mean stigma scores between index and referent levels of each factor were less than two points, except for incorrectly believing that TB increases the chance of getting AIDS (mean difference of 2.16; 95% CI: 1.38, 2.94) and knowing someone who died from TB (mean difference of 2.59; 95% CI: 0.96, 4.22). Conclusion These results suggest that approaches addressing the dual TB/HIV epidemic may be needed to combat TB stigma and that simply correcting misconceptions about TB may have limited effects.