WHEN THE state issued its test scores last month showing that immigrant students with little or no English performed better than last year, supporters of Proposition 227 seized on the results as proof that bilingual education doesn’t work.
But it is far too soon to say whether bilingual education should be banished forever. At issue is whether two year’s worth of test scores can be used to settle the pedagogical challenge of how best to educate California’s 1.4 million students classified as “limited English proficient.”
The simple answer is: they can’t. “Some people are claiming, based on no systematic study and on a few rough numbers, that the program (mandated by Prop. 227) is a success,“ said Delaine Eastin, California’s superintendent of public instruction. “But we are not nearly at the point where we can say that.”
In 1998 California voters decreed that, beginning in the fall, immigrant students had to be placed in classes taught in English, unless parents applied in writing for a waiver to keep their children in bilingual classes. Much of the cheerleading for Prop. 227’s success has been based on comparing test scores in the Oceanside School District near San Diego, where all bilingual programs were phased out, with those in the neighboring Vista School District, where half of the students are still in bilingual programs.
The test scores showed that English learners in Oceanside scored higher than a similar population of students at Vista, and improved at a faster rate, especially in the earlier grades.
Oceanside has become a magnet for reporters, mainly because Ken Noonan, its superintendent, was for decades a leading crusader for bilingual education. Today, he says he was wrong, and has emerged as a prominent advocate for the “English immersion” approach. Reporters for the New York Times, the Wall Street Journal and other outlets have written enthusiastic articles or editorials on the same theme: Oceanside proves that bilingual education doesn’t work.
But a closer analysis shows the danger of selecting one or two districts at random from among the state’s 1,000 districts to prove that fundamental point. David Cowles, Vista’s superintendent told The Chronicle his district had inaccurately coded the test scores for its most advanced English learners, and tallied them along with the scores of regular students. As a result the district’s English learners appear to be performing worse than they actually are. Cowles says once the numbers have been recalculated, they will show his students are doing about as well as those at Oceanside.
The scores also can’t indicate whether school districts like Oceanside spend more time teaching students to take the test than others. It is also common sense that students who are taught in English will be better prepared to take a test administered in English. How they will do in later grades — or learn academic subjects taught in a language they don’t know very well — is still unproven.
The media experts have also contrasted scores of last year’s students to this year’s students, which means they are comparing two completely different groups of children. A better way would be to evaluate the same children from one year to the next to see how they are progressing. It is also a questionable practice to draw firm conclusions based on only the two sets of scores — this year’s and last’s. “Most experts will tell you it takes measurements at three points in time to make accurate measurements,” said Joan Herman, associate director UCLA’s Center for Research on Evaluation, Standards and Student Testing.
It’s also possible that much or all of the recent increases in test scores can be explained by changes other than ending bilingual education. In addition to the Prop. 227 reforms, the state has introduced many others, including reducing class size in K-3 grades, improving teacher training, and implementing new curriculum standards. The fact that all California’s students showed improvements in test scores shows that something other than ending bilingual education is at work here. It is no coincidence that students in elementary schools, where many of the reforms were concentrated, showed the biggest increase.
Complicating the discussion is confusion over basic terminology. It will come as a surprise to many that Prop. 227 did not require students to be taught in a purely “English-only” classroom, but in one where English is “overwhelmingly” used. Three quarters of the state’s English learners are enrolled in “structured English immersion” classes, where they get substantial help from teachers or aides in their native language. Nearly 20 percent are still in traditional bilingual classes, where students are taught most (but not necessarily all) subjects in their native language. Only 5 percent are in “total English immersion” classes where instruction is completely in English. Without careful study, these variations make it almost impossible to assess what approach works best.
Prop. 227 could turn out to be the answer. But for now, limited English students still lag far behind their English speaking peers — even in acclaimed districts like Oceanside that have ended bilingual education as we know it. That is an argument for continuing to search for better approaches. There is a deeper challenge: how to prepare all students to participate in a multicultural world in which mastery of more than just English will be a prized asset. Speaking English will be crucial. So will speaking Spanish, Chinese or Korean. On that issue, the language of Prop. 227 is completely silent.