This study examined whether the distribution of tobacco use and related psychosocial risk factors among youth in urban India vary by socioeconomic status (SES). Data were derived from a cross-sectional survey of students enrolled in the 6th and 8th grades in 32 schools in Delhi and Chennai (N=11,642). The survey was conducted in 2004, before the implementation of a program designed to prevent and reduce tobacco use (MYTRI). Mixed-effect regression models were used (a) to determine the prevalence of tobacco use among private (higher SES) and government (lower SES) school students, (b) to investigate whether certain psychosocial factors were associated with increased tobacco use, and (c) to determine how these factors varied by school type. Ever-use of multiple forms of tobacco (e.g., gutkha, bidis, and cigarettes) was more prevalent among government school students than private school students. After adjusting for city, gender, grade, and age, we found the prevalence rate for ever-use of any tobacco product to be 18.9% for government school students, compared with 12.2% for private school students (p<.01). Students in government schools scored lower than private school students on most psychosocial risk factors for tobacco use studied here, indicating higher risk. Government school students scored the lowest for refusal skills, self-efficacy, and reasons not to use tobacco. Social susceptibility to chewing tobacco and social susceptibility to smoking were strong correlates of current tobacco use among government school students. Exposure to tobacco advertising was also a strong correlate of current tobacco use for government school students but not private school students. In two large cities of India, students attending government schools are using many forms of tobacco at higher rates than private school students. The psychosocial risk profile of government school students suggests they are more vulnerable to initiation and use and to outside influences that encourage use.