The Acquisition of Academic English by Children in Two-Way Programs: What does the Research Say?
Stephen Krashen
Presented at NABE, 2004, Albuquerque, NM
Two-way bilingual programs are those in which majority language children and minority language children attend the same classes at the same time, the idea being that majority language children will acquire the minority language and minority language children will acquire the majority language. There has been a great deal of excitement about these programs; claims have been made that two- way has been shown to be the most effective form of bilingual education, and the best possible program for language minority children in general. It is important that we take a careful look at the research to see if these claims are supported.
The purpose of this paper is to examine one aspect of the research on two-way programs, specifically to examine the progress of language minority students in acquiring academic English. The focus here is on performance on tests of reading comprehension, a measure considered to be the gold standard for academic English. The questions to be discussed are these: (1) Do English learners in two-way programs show evidence of significant acquisition of English? (2) Do they outperform children in non-bilingual (all-English) options? (3) Do they outperform children participating in other forms of bilingual education? Some data is also available comparing two-way children with native speakers of English.
I will review the studies in ascending order of their ability to answer these three questions. First, I briefly discuss studies that had to be excluded from this analysis. Then, we look at studies that lacked comparison groups. These studies are able to provide us with some information regarding the first question. The third group consists of studies in which two-way students are compared to similar students in mainstream or submersion classes, an attempt to answer question (2). The fourth group consists of studies comparing two-way students to those in other kinds of bilingual education programs, an attempt to answer question (3). Finally, I include a section on studies in which two-way children are compared to native speakers of English.
Group One: Excluded Studies
I begin this review by excluding a number of studies from the analysis, reports in which few details are reported (Thomas and Collier, 1997), reports in which it is not possible to determine how well English learners did because their scores are combined with those of native speakers of English (Christian, Montone, Lindholm, and Carraza, 1997; Rhodes, Crandell, and Christian, 1990, Lindholm and Fairchild, 1990, Sensec, 2002) .and reports in which two-way students were not tested in English (Coy and Litherland, 2000). In addition, in several of these studies, a substantial percentage of minority language speakers appear to have already acquired a considerable amount of English before beginning the program. These studies do not shed light on any of the three issues to be discussed.1 Two studies labeled “dual language” comparisons were excluded because comparisons were not monolingual speakers of English but were bilingual but English-dominant (Perez and Flores, 2002; Oller & Eilers, 2002).
1