a district level analysis
This paper deals with political violence in Bangladesh. In this paper, we show that the degree of political violence varies widely in districts and argue that this can be explained by the difference in political competitiveness. We use election competition as a proxy of political competitiveness and attempt to study whether it can explain variation in political violence in districts. This paper tests the hypothesis using a multiple regression model and finds substantial support for the hypothesis. It is also found that allocation of development expenditure, population density and urban centricity, along with location of regional headquarters are strong determinants of political violence.