Scholars of social movements recognize the threat that opposition protests pose to the stability of an authoritarian regime. Extensive research has been done on how mass protests can induce regime change, and in turn, democratization. We know less about the micro-level processes by which protests can influence the voter behaviour of the ordinary citizen. This thesis aims to fill this gap, by using evidences from the 2019 social unrest in Hong Kong. By combining protest data from ACLED and the Anti-ELAB Research Archive, with a post-election survey from Hong Kong Election Study, I examine how citizens who did not vote for opposition parties in the previous election respond to anti-regime protests and police brutality in their neighbourhood. The findings imply that the effects of protests and police brutality are dependent on whether the respondents identify themselves as “Hongkongers” or as “Chinese”. By using logistic regression, I find that opposition protests and police brutality can mobilize “Hongkongers”, who previously abstained from voting, to vote for opposition parties. However, it also refrains those who identify themselves as “Chinese” from doing the same. I also find that nonviolent protests increase the likelihood that previous regime supporters, who identify themselves as “Hongkongers”, will defect to opposition parties.
Nilsen, Amalie (2021) Can Pre-Election Protests Influence Voter Behaviour? The Case of Hong Kong's Anti-ELAB Movement. MA thesis, Department of Political Science, University of Oslo, Oslo, Norway.