Three essays on student peer effects

I analyze immigrant and gender peer effects among junior and high school students. The samples are drawn from a large, nationally representative study of adolescents, with detailed individual and school information. Intra-school grade-level variation provides the identification, while school selection is controlled with fixed effects. The availability of short and long-term academic indicators, as well as extensive data on attitudes, behaviors and friendship networks permits new insights into the ways in which adolescents affect each other.;In Chapter 1, I analyze the impact of first and second generation immigrants from Mexico on both native schoolchildren and the Mexican-Americans themselves. I find little evidence of between or within-group peer effects. My analysis also shows that classmates are not the relevant peer group for immigrant students, since a significant portion of them form strongly assortative friendship networks that are not constrained by grade levels.;Building upon this analysis, Chapter 2 looks at how Hispanic and Asian immigrants affect native non-Hispanic white and black students. I find that Hispanic immigrants overall appear to have little effect on native whites and blacks, but white and black natives appear to respond differently to Asian classmates. The proportion of first or second generation Asian students in the grade is positively correlated with white students' academic outcomes. The proportion of first generation Asians is also positively correlated with the outcomes of black students; but having more second generation Asian classmates is correlated with lower grades, more delinquent behaviors, and lower self-reported desire and likelihood of college attendance among blacks.;Chapter 3 presents evidence on gender peer effects. I find that high school girls who have more female classmates are more likely to graduate, but there is no significant correlation with other academic outcomes of girls or boys. I do find significant behavioral effects: having more female classmates is positively correlated with sexual activity among girls and sexual awareness among younger boys; it is also correlated with more positive attitudes towards school but, at the same time, greater likelihood of delinquent, and potentially dangerous, behaviors among boys.