Violence and voting behavior :|evidence from 2015 Turkish elections

Abstract

How does violence affect voting behavior? Using the Turkish elections in June and November 2015 as a case study, this thesis tries to answer this question. To do that, first, I conduct a variety of quantitative analyses based on district-level data I have gathered on the location of deadly clashes and attacks as well as the funeral locations of security forces and civilians in between two elections. Second, in order to unpack possible mechanisms behind statistical correlations, I use qualitative evidence gathered from a variety of primary and secondary sources. In my analysis, I emphasize the importance of context and argue that the literature on voting behavior and violence/terrorism should pay greater attention to the contextual factors such as the nature of electoral competition, electoral strategy and counter-terrorism policies of political parties, the targets (civilian or security force) and the level of violence, the nature of warfare, and identities and political orientations of the civilians. Moreover, I argue that under high levels of violence, macro-dynamics rather than the micro-dynamics of the conflict influence voting behavior.