Some of the most significant changes to have occurred in Sub-Saharan Africa in the last two decades can be found in many countries' mass media environments. Political liberalization, technological advances, and market reforms have made television, FM radio, newspapers, and the Internet available over increasingly large swaths of territory in Africa. As a result, millions are finding it easier and cheaper to access entertainment, information from diaspora communities, and, perhaps most importantly, news about politics in their country. Social scientists have only begun studying how these new developments in communication and information exchange are affecting citizens' knowledge about politics and, consequently, political competition in the region. This research project seeks to understand, more specifically, how recent developments in African mass media have affected citizens' electoral decision-making processes. Dominant theories of voting behavior in Africa focus heavily on factors such as ethnic identities and candidates' distribution of relatively cheap goods during campaign periods. However, does the increasing availability of information, such as journalistic accounts of official corruption, government distributional practices, and policy outcomes, affect the salience of these factors in voters' decision-making calculi? This project develops a theory of information-collection strategies and electoral decision-making in Sub-Saharan Africa that attributes some of the importance of factors such as ethnicity and goods distribution to the political information that they provide to voters. The theory posits that, as other, possibly confounding facts about politics become more widely available, the importance of these other factors, such as ethnicity and goods distribution, in voters' decision-making processes will decline. In order to examine the validity of these hypotheses, a series of laboratory experiments will be conducted in a number of locations in Uganda. Subjects will be exposed to various types of information about hypothetical two-candidate races for local government positions. Following these informational treatments, which will include cues such as candidates' educational levels, past performance as elected officials, and positions on controversial local issues, participants' levels of support for and evaluations of co-ethnics and goods distributors will be measured. These experiments will result in the creation of a dataset on which theories on voting behavior in Sub-Saharan Africa can be tested. The findings of this research should improve our understanding of how variations in access to political information affect outcomes such as inter-ethnic relations, clientelistic exchange, and electoral competition in Sub-Saharan Africa, and other developing democracies as well.
