The purpose of this study is to answer the following overarching question: how does ethnic diversity among immigrant and native populations impact xenophobia? Many studies answer this question by examining the effects of relative immigrant group size. Instead, I argue that group size increases xenophobia when immigrants are ethnically visible, crossing salient linguistic, religious, or racial boundaries. In three investigations I look at the effects of the following factors on xenophobia: ethnic diversity in the immigrant population, ethnic diversity in the broader society, and being cultural marginal. Analyzing multilevel models using cross-national data from the European Social Survey (ESS), I examine the effects of regional and national contexts of immigrant visibility on xenophobia. I define xenophobia as the perception of immigrant threat. I also test the hypothesis that average xenophobia is higher among individuals living in more ethnically diverse countries. In the second investigation, I reexamine immigrant visibility, this time using Swiss ESS data to compare across municipalities. I also consider the effects of living adjacent to rather than in an immigrant-rich community. In a final investigation, I again analyze cross-national ESS data to determine the effects of being different from the cultural majority on xenophobia. I find that xenophobia is higher among individuals living in more religiously diverse countries. Also, for those living in communities with few to no immigrants, the presence of immigrants in surrounding areas amplifies xenophobia. Comparing across countries and regions within those countries, I find that the size of the ethnically visible population does not affect xenophobia. However, immigrant visibility does increase xenophobia in the Swiss context. All other things equal, cultural minority and majority members do not appear to differ in their levels of xenophobia. However, individuals who perceive marginality tend to be more xenophobic than those who do not. The interesting exception is religious minority members, who are least xenophobic, but only when they perceive marginality. The findings cast doubt on the size argument of group threat theory, which predicts that xenophobia is higher where there are more immigrants. Even when measured in terms of the most ethnically visible and potentially most culturally threatening, immigrant group size does not explain cross-national differences in individual xenophobia. It seems to explain attitudes in some national contexts, but not others. Group size may only evoke perceptions of immigrant threat under certain necessary conditions as an interaction effect. Living adjacent to immigrant-rich communities amplifies xenophobia, but in a way that is partly attributable to contact. Counter to the predictions of cultural marginality theory, being culturally different does not universally lead one to espouse more tolerant views toward immigrants. The religious exception may stem from increased contact with immigrants, rather than simply the increased sympathy for other marginalized peoples implied by cultural marginality theory.