Social Media Bots, Trolls, and the Democratic Mandates in Sub-Saharan Africa

Abstract

This exploratory study investigates the role that bots and trolls played on social media in widening the gaps of political partisanship in sub-Saharan Africa. Taking the case of the 2018 Zambian by-elections, the authors examined the relationship between online social media content that propagates hate and organized trolling efforts in Zambia. The study used machine learning tools to identify the origin of the bots on Facebook and Twitter accounts (trolls) of the two major political parties in Zambia (PF and UPND). Online posts that accounted for the election campaigns and the aftermath in the year 2018 were considered for the study. Findings suggest that social-mediated conversations were divided along political lines and that the examined trolling accounts systematically took advantage of the existing echo chambers to create hate messages on Zambian social networks. In other words, the findings indicated that the online hate messages that accounted for violence were neither created by the PF or UPND political parties as earlier studies suggest but by bots and trolls.