This dissertation examines the differences in ethnic favoritism across African countries. In the first chapter, using survey data from Afrobarometer we consider to what extent ethnic favoritism arises in sub-Saharan African countries by examining whether a person’s perception of government fairness depends on whether he shares the same ethnicity as the country’s leader. I allow associations to differ not only across time but across types of countries such as democracies versus nondemocracies. Such an approach not only can determine whether ethnic favoritism arises but in what cases it might be more pronounced. Findings indicate that the associations with perceived ethnic favoritism have held steady over time. Differences between low- and middle-income countries are also small. Greater differences arise across political regimes as democracies show less ethnic bias than do nondemocracies. More diverse countries also show more ethnic bias. In the second chapter, with the information from two rounds of the Afrobarometer survey, we consider to what extent gender and ethnicity matter for perceived access and quality of education and health facilities. We allow associations to differ not only across time but across types of countries such as democracies versus nondemocracies. Such an approach not only can determine whether ethnic and gender inequalities arise but in what circumstances they might dissipate. Our findings indicate no differences in the perceived access and quality of education and health facilities nor do differences arise in various subgroups of countries such as democracies, resource-rich countries, ethnically diverse countries, or middle-income countries. The third chapter examines how the Nigerian president’s changes in ethnicity and religion affect certain outcomes. Since 2000, Nigerian presidents have come from different ethnic groups and practiced different religions. Using several rounds of the Demographic and Health Surveys, we exploit this variation in leadership to examine if households see greater material wealth or greater access to public services when the president comes from their ethnic group or when he practices the same religion. The analysis indicates that ethnic groups do not benefit (at least, contemporaneously) from having a co-ethnic as president. In fact, co-ethnics are less likely to have access to electricity, protected water, improved sanitation, and more likely to be in the poorest quantile.