Effects of Dual Language Immersion on English Learner Reclassification: A Discrete-Time Survival Analysis

• In Event: 47.055 - Curricular Policy and Language-Minority Student Inclusion

Abstract

This paper uses data from a lottery-based study in which students were randomly assigned before kindergarten to dual-language immersion programs or to monolingual instruction within an urban U.S. district. We use discrete-time survival analysis to estimate the causal effect of random assignment to immersion on time to reclassification for ELs in binding lotteries. To examine the role of partner-home language match in immersion efficacy, we disaggregate our estimates in the binding lottery sample. We find lower median lifetimes for immersion students regardless of whether their home languages match the partner language. However, we find a more consistent advantage when the languages do match.

Authors

• Jennifer L. Steele, American University

• Graziello Covello, American University

• Robert Slater, American Councils for International Education

• Gema Zamarro, University of Arkansas

• Trey Miller, RAND Corporation

• Jennifer Li, RAND Corporation

• Michael Bacon, Portland Public Schools