Although many have argued that election violence depresses voter turnout, previous research has found inconclusive empirical support for this claim. Nonetheless, its frequent strategic use suggests that politicians believe violence will have some sort of a significant effect on voter behavior. Using survey data from 32 African countries and geo-tagged event data from a smaller subset of these countries, I examine the impact of exposure to election violence on intention to vote and candidate support. Mirroring findings from the literature on negative campaigning in the United States, I find that election violence is associated with increased support for opposition parties and decreased support for ruling parties. Based on this analysis, election violence appears to be, at best, an ineffective campaign strategy, and under many circumstances, actually reduces a candidate’s electoral support
