Effects of Transitional Bilingual Education, Two-Way Bilingual, and Structured English Immersion Programs on the Literacy and Oracy of Spanish-Dominant Children*

Grantee Johns Hopkins University

Key Staff: Robert Slavin, Margarita Calderon, Nancy Madden

Research Questions:

• What are the relative effects of teaching reading, writing, and oracy to Spanish-dominant students in structured English immersion versus transitional bilingual education or two-way bilingual education, from kindergarten to grade 3, on English literacy, oracy, vocabulary, comprehension, writing, and other variables?
• How do the effects of structured English immersion versus the other two instructional strategies vary according to child and home factors such as race, gender, initial skills in Spanish and English, initial cognitive skills, home language use, family structure, and family resources?
• How do the effects of structured English immersion versus the other two instructional strategies vary according to school and teacher factors such as quality of program implementation, classroom oral language environment, teacher language proficiency, use of Spanish or English in parts of the day other than literacy periods, and district and state policy environments?

Design:

Approximately 545 Kindergarten children who are Spanish-dominant English language learners were randomly assigned to conditions within schools. Three conditions were studied: structured English immersion (SEI); transitional bilingual education (TBE); and, dual language education. Two sets of schools were recruited. In one set, a total of 387 students in 10 schools were randomly assigned to receive either the structured English immersion or the transitional bilingual program. In the second set, approximately 158 students in 5 schools were randomly assigned to either structured English immersion or dual language instruction. The study followed three cohorts of Kindergarten students. The study of SEI versus dual language instruction suffered from high levels of school and student attrition, making the findings difficult to interpret.

Cost/Duration: $5,395,652 over 6 years (October 1, 2003–September 30, 2009)

Current Status: (February 2011)

This project is complete. The final project progress report has been received.

Key Findings:

• At the end of kindergarten and first grade, students in SEI had statistically significantly better English receptive vocabulary, letter word identification, work attack, and passage comprehension (not assessed in kindergarten) outcomes than students in TBE.
• By the end of second and third grades, when many students in TBE had switched to instruction in English, the positive effects of SEI on English skills were statistically significant for only one of the four outcomes in each grade. At the end of second grade, there was a statistically significant positive effect of SEI on word attack. At the end of third grade, there was a statistically significant positive effect of SEI on receptive vocabulary.
• When the students reached fourth grade, they only received English instruction. At the end of fourth grade, there were no statistically significant differences between the two groups on English outcomes, although the effect sizes ranged from 0.21 to 0.31 (all in favor of SEI).
• The findings for Spanish language outcomes are the reverse of the results for English outcomes. At the end of kindergarten and first grade, students in TBE had statistically significantly better Spanish outcomes than students in SEI. By the end of second and third grades, there were statistically significant differences in Spanish outcomes favoring TBE for three of the four outcomes (letter word identification, word attack, and passage comprehension). At the end of fourth grade, the only remaining statistically significant difference in Spanish outcomes was for passage comprehension, although the effect sizes for the other three Spanish outcomes ranged from 0.04 to 0.33.

* This grant was awarded under the English Language Acquisition Evaluation Program, CFDA 84.305P