Peer Effects from Students with Limited English Proficiency: How Does Sharing a Classroom with LEP Students Affect Native English Speakers?

Scott A. Imberman

In the 1990’s, enrollment of students with limited English proficiency (LEP) grew by 105% compared to overall enrollment growth of 24%. As the LEP population in US schools grows, these students will increasingly interact with native English speakers. However, while there is some research about the effectiveness of immersing LEP students in regular classrooms on the LEP population, we know little about the impact they have on their English speaking classmates. Indeed, a proper assessment of the costs and benefits of programs for LEP students such as Bilingual education needs to take into account the impacts on non-LEP students. In this project we address this question by estimating the impact of LEP classmates on non-LEP students’ educational outcomes. Using student level data for all students in the state of Texas from 1998 through 2009, we exploit quasi-random assignment of LEP students to mainstream classes generated by a policy rule in Texas. This rule requires school districts to provide bilingual education in a particular language only if there are at least 20 LEP students in the same grade and language. When a bilingual education class is formed, exposure to LEP classmates jumps down. This enables us to employ a regression discontinuity approach to identify the causal impact of LEP classmates. We supplement this approach with a strategy that uses natural variation in immigrant share within schools across time and grade-levels. Through these strategies we will establish the first causal estimates of peer effects from students with limited English proficiency.

About Scott A. Imberman

N/A