Background: Previous studies have consistently identified an association between higher levels of education and lower levels of anti-immigrant sentiment, but the underlying reasons for this relationship remain unclear. Therefore, this research aims to help explain why education matters for attitudes toward immigrants. This thesis consists of two studies, where in I examine the role of two features of education, specifically the aquistition of knowledge and social relationships. In the first study, I analyze how two aspects of teaching in schools are associated with anti-immigrant sentiment. I examine education as a means to knowledge by investigating whether the content of education, such as critical thinking and multicultural education, is inversely related to students' prejudice. I also look into the certification of teachers and whether this has a similar effect. In the second study, I investigate the school as an arena for social interaction and examine the relationship between outgroup contact and prejudiced attitudes. Previous research has found that outgroup contact, especially contact in the form of outgroup friendship, is effective in reducing prejudice. Based on these previous findings, I study how the Secondary Transfer Effect (STE) of friendship with a specific outgroup member is associated with attitudes towards other ethnic or racial groups. Method: The two studies rely on a cross-sectional survey of Swedish high school students (aged 16-18) administered by 'Forum för levande historia' (Forum for living history) and Statistics Sweden during the 2009-2010 academic year. Additionally, survey participants' responses are matched to registry data. This thesis also uses multilevel (MLM) and ordinary least squares (OLS) regression models. Results: Results show that both aspects of education are correlated with lower levels of anti-immigrant sentiment. The first study, in which education is a means to knowledge, demonstrates that education focused on critical thinking and multiculturalism are negatively correlated with anti-immigrant sentiment. Furthermore, results show that more exposure to teachings about xenophobia and racism is associated with less prejudice. However, when controlling for exposure to critical thinking, as well as learning about religions and cultures, the results show that exposure to learning about xenophobia and racism are no longer significant. Moreover, there is a negative association between exposure to teachers with a teaching certification or teaching degree and prejudice. Results from the second study, in which the school functions as an arena for social interaction, shows that positive attitudes associated with intergroup friendship not only generalize to the ethnic outgroup of that friend but, more importantly, also to other secondary outgroups. STEs are most frequently found where boundaries between ingroups and outgroups are perceived to be the thickest. Thus, the presence of STEs appears to be group-specific. Previous research on perceived social distance in Sweden and other countries help shed light on these findings. In summary, the results of this thesis provide evidence of two different complimentary accounts of the negative relationship between education and prejudice.