Noonan's Wrong on Bilingual Education

By STEPHEN KRASHEN

It is unfortunate that Ken Noonan's appointment to the California State Board of Education has resulted in a repetition of the myth that dropping bilingual education is a good thing.

Superintendent Noonan has been given credit for the increase in test scores of English learners in Oceanside after he eliminated bilingual education in the city's school district following the passage of Proposition 227. But scientific analyses strongly suggest that Noonan's policies had nothing to do with the increase.

In California, a new test, the SAT9, was introduced at the same time Proposition 227 passed. Research done by Robert Linn and others shows that when a new standardized test is introduced, scores at first appear to be low, and then increase each year as teachers and students become more familiar with the format and content of the test. Thus, when we see gains on a new test, it is not clear if the gains represent real learning or simply better test preparation.

Predictably, SAT9 scores increased throughout California for the next few years after it was introduced. California's "test score inflation" was particularly severe because of the intense pressure to increase scores.

What is crucial is that the increase occurred for all students, fluent English speakers and English learners, a rising tide that lifted all boats. In fact, Stanford researcher Kenji Hakuta and his colleagues have shown that gains for Oceanside's English learners were similar to gains made in many California districts that retained bilingual education.

Uncontrolled standardized-test scores are, however, a poor way to judge whether a teaching method works. A much better way is the use of controlled studies that compare groups of children similar in language proficiency and background, with one group receiving bilingual education and the other "English-only."

Nearly all scholars who have reviewed this research have concluded that students in bilingual programs generally acquire more English than children in all-English programs; at worst, the programs produce similar results. Researcher Jay Greene, for example, concluded that "efforts to eliminate the use of the native language in instruction ... harm children by denying them access to beneficial approaches."

Bilingual programs use the first language in ways that accelerate English language development. When students learn subject matter in their first language, subject matter classes taught in English become more comprehensible, resulting in more acquisition of English. An English learner who knows science well, because of instruction in the first language, will be more successful in a science class taught in English than an English learner without this background.

Also, learning to read in the primary language is a shortcut to learning to read in English. Scientific evidence shows that it is easier to learn to read in a language you already understand, and once you can read in one language, this ability transfers to the second language, even if the writing systems are different.

These days schools are told to base teaching practice on "scientific" evidence. It is strange that the overwhelming scientific evidence supporting bilingual education is ignored.

Stephen Krashen is professor emeritus, Rossier School of Education, University of Southern California. He is the author of "Condemned Without a Trial: Bogus Arguments Against Bilingual Education" (Heinemann, 1999).