Axes of Variation in Bilingual Program Implementation: Learning From School and District Leaders

• In Event: 30.046 - Can the "Blunt Instruments" of Education Policy Affect Classroom Practice for English Learners?

Abstract

After several decades of experimental, quasi-experimental, and correlational research, there is an emerging consensus that bilingual education, on average, results in a small to moderate linguistic and academic advantage for students acquiring English (August & Shanahan, 2006; Cheung & Slavin, 2012). However, these studies, by and large, do not interrogate the details of bilingual program implementation. We know less, therefore, about what elements of bilingual program design result in the greatest gains for students, and what explains variation in the impact of bilingual programs. This presentation addresses those gaps in the literature, drawing on findings from in-depth interviews with 22 current and former bilingual school principals and district administrators for bilingual education. In the presentation, we have two primary objectives: 1) establish key axes of variation in bilingual program implementation and 2) discuss veteran practitioners’ perceptions of the relative effectiveness of each.

Theory suggests several ways in which bilingual education supports students’ learning. First, developing full literacy and oracy in one’s native language provides a foundation from which to acquire a second language (Cummins, 1991). Second, providing academic content instruction in students’ dominant language allows greater access and understanding (Krashen, 1981). Finally, bilingual instruction can increase students’ academic and personal self-efficacy by creating a context in which their home language, history and culture are placed on equal footing to those from English-speaking households (Gándara & Orfield, 2010). There is a clear typology of bilingual education programs which identifies programs by duration (early exit versus late exit) and by student composition (bilingual versus dual immersion) (Baker, 2011). However, there is less clarity on variation in the implementational features of bilingual programs and their relative effectiveness, the two primary questions addressed in this study.

Data include 22 in-depth interviews with current and former principals and district administrators for bilingual programs in a large, urban school district in California. Each structured interview lasted between one and two hours and covered issues of bilingual program goals, structure, composition, curriculum and instruction. Interviews were transcribed and coded using Dedoose software. Codes were developed both from theory and from the data, and coding was triangulated across multiple coders. Analysis of the coded interviews is underway.

Preliminary results suggest that there is wide variation across programs within the same program type. For example, two programs both categorized as early exit bilingual programs may look very different in practice. Key axes of variation include, among other features, the degree of district oversight and program articulation, the structure of English language instruction, the relative degree of integration with English-speaking students, the presence of parental and family participation, and teachers’ academic expectations of students acquiring English.

This presentation pushes analyses of bilingual education beyond impact analyses and basic typologies. Instead, it lays out key features of bilingual programs and identifies, from a leadership perspective, how features interact to support or hinder program effectiveness. This contributes to conversations about ways in which policies about instructional program models for English learners can be designed in more meaningful ways.

Authors

• Ilana Marice Umansky, University of Oregon

• Misael Flores, University of Oregon

• Nikki Cox

• Sean F. Reardon, Stanford University

• Rachel Valentino, Stanford University