Recent cross-national studies have found only moderate support for the idea that population pressure and resource scarcities may lead to political violence, contrary to much of the case study literature in the field. This article suggests that the level of analysis may be at heart of this discrepancy. In a time-series study of 27 Indian states for the period of 1956-2002, propositions that violent conflict may be associated with high population pressure on renewable natural resources, with youth bulges and with differential growth rates between ethnic and religious groups are tested. The study analyzes the relationship between population pressure and political violence, using three different and independently collected datasets measuring armed conflict, political violence events and Hindu-Muslim riots. The great availability of disaggregated demographic, socioeconomic and environmental data for India makes possible a more detailed assessment of the population pressure hypotheses than can be tested for a global sample of states. The results generally provide more support for the resource scarcity and conflict scenario than recent global studies, supporting claims that the scarcity-violence nexus may be better studied at a sub-national level. Scarcity of productive land is associated with higher risks of political violence, particularly when interacting with high rural population growth and low agricultural yield. Other central aspects of the resource scarcity scenario are not supported; structural scarcity, measured by rural inequality, as well as high urbanization rates are not associated with higher levels of political violence. The study further suggests that youth bulges increase the risk of all three forms of political violence, especially in states with great male surpluses. Youth bulges, when coinciding with high levels of urban inequality, is the only form of demographic pressure to statistically increase the risk of Hindu-Muslim rioting. Finally, there is some indication that relatively high Hindu growth rates are positively related to armed conflict, while religious heterogeneity in itself appears to be unrelated to political violence in India.