selective Colombian university that led to large differences in mean classmate ability for students in the same programs. In a … regression discontinuity design, we find that students in higher-ability classes were more likely to fail courses and drop out …

university. We focus on two core variables: the share of non-English-speaking students and the diversity within the group of non …-English-speaking students with respect to their linguistic background. English-speaking students are largely unaffected by the ethno …-linguistic classroom composition.Non-English-speaking students benefit from a larger diversity in their performance and increase their …

Italy and exploiting the random assignment of first year students to college accommodation, we find that roommate peer …

local students in Shanghai, where local students all go to public schools, but migrant students are endogenously selected … all student groups. We conduct policy experiments to examine the effect of transferring migrant students from private …

educational outcomes of low-income students and has negative effects on high-income students. We further document a range of …

disruptive students away to disciplinary schools, removed students are exposed to highly disruptive peers. Using rich … administrative data on Texas high school students, we leverage within school-year variation in peer composition at disciplinary … schools to estimate the effects. We show that for students placed in disciplinary schools, exposure to more disruptive peer …