New Ways of Studying Fluency in English
Harold Smithhsmith [at] su.edu
Shenandoah University, Winchester, Virginia, USA
The Santa Ana school district has four major kinds of ELD: Transitional Bilingual Education, which gives native language instruction along with ESL; Immersion, which gives sheltered instruction in English; combined Transitional Bilingual Education and Immersion; and Mainstream Limited English Proficiency.
Differences observed in achievement levels for reading and math in Santa Ana's various ELD appeared to be due primarily to students' placement in a particular program, not relative effectiveness of the programs. That is, students were not randomly assigned to different programs. Students who entered Santa Ana schools above kindergarten level, and those moving between schools, were more likely to be placed in mainstream or mixed Transitional Bilingual Education or Immersion. And students in Immersion tended to enter (and exit) at higher levels of fluency.
Several factors appeared to predict reading and math achievement for nonnative speakers: special education, movement between schools, test language, students' English language development levels, primary language development levels, and English Language Development programs.
Each program was more effective at some fluency level(s) than others. Transitional Bilingual Education seemed most effective from Pre-Production to Early Production levels, Immersion worked best in helping students progress from Early Production to Speech Emergence. Mainstreaming was the least effective program below the Intermediate level; it did best from Intermediate to Advanced Fluency levels. Transitional Bilingual Education best helped students progress from Advanced Fluency to Fully English Proficient levels.
If students, families, and schools know that it may take seven or more years to reach full fluency, they might have more reasonable expectations of student progress in developing English fluency. Kindergarten students, for example, who were assigned to Transitional Bilingual Education programs began nearly a full level below others, and by grade 5 they were still about one third of a level below.
While Mitchell and team saw that as "closing the gap by nearly half," one may wonder if there is some better program that would develop fluency more quickly. Teachers, researchers and administrators should continue looking for more effective and efficient ways to build English language fluency in nonnative students. Though it was not the focus of their study, Santa Ana's results indicate that how students are taught has a strong influence on how long it takes to attain fluency, and that seems to be even more important than who is studied.
Reference: Mitchell, D.; Destino, D.; and Karam, R. (1997) Evaluation of English language development programs in the Santa Ana Unified School District. California Educational Research Cooperative, School of Education, University of California, Riverside.
The Internet TESL Journal, Vol. V, No. 2, February 1999
http://iteslj.org/
http://iteslj.org/Articles/Smith-Fluency.html