Higher infant mortality among tribal populations in India is well-documented. However, it is rare to compare factors associated with infant mortality in tribal populations with those in non-tribal populations. In the present paper, Cox proportional hazards models were employed to examine factors influencing infant mortality in tribal and non-tribal populations in the Central and Eastern Indian states using data from the District Level Household Survey-III in [2007][2008]. Characteristics of mothers, infants, and households/communities plus a program variable reflecting the place of pregnancy registration were included in the analyses. We found that the gap in infant mortality between tribal and non-tribal populations was substantial in the early months after birth, narrowed between the fourth and eighth months, and enlarged mildly afterwards. Cox regression models show that while some factors were similarly associated with infant mortality in tribes and non-tribes, distinctive differences between tribal and non-tribal populations were striking. Sex of infants, breastfeeding with colostrum, and age of mother at birth acted similarly between tribes and non-tribes, yet factors such as state of residence, wealth, religion, place of residence, mother's education, and birth order behaved differently. The program factor was non-significant in both tribal and non-tribal populations.