For a man with training in theoretical physics, Ron Unz makes a great spinmeister. Last week, as the yawning gap in test scores between California’s limited-English-speaking (LEP) students and its fluent English-speakers shrank by an inch or two, Unz, author of Proposition 227, proclaimed victory.
In the year since voters passed the measure against bilingual education, the performance of the average California LEP student, by Unz’s calculations, rose by 18 percent in reading and 21 percent in math. “There are probably few educational reforms,” he declared, “which have had such rapid and substantial impact.”
But to get those numbers, Unz, a Silicon Valley software millionaire, mushed together all grades and, worse, used percentile rank rather than student scores to measure gains. Thus, his 18 percent gain in reading is derived from the fact that last year the state’s LEP students ranked at the 15.6 percentile when measured against all students; this year they reached the 18.4 percentile. (The average student scores 50 percent.)
But there’s less here than appears. To make such a gain, a fourth-grade student would have had to get roughly 37 answers right (out of 80) rather than 34 or 35, which makes Unz’s mushed-together gain in reading closer to 7 percent than to 18 percent. And since much of the gain can be attributed to better test preparation, its significance shrinks still more.
And yet, Unz deserves a great deal of credit for addressing what is surely one of the most serious problems in California education — and until Proposition 227, one of the most neglected.
One-fourth of California’s students are designated as LEP. In fourth-grade reading, for example, where the average score among the state’s English-proficient students is slightly above the national average (at the 53rd percentile), the average LEP student comes in at the 17th percentile.
In effect, California now appears to have two segregated school systems: one predominantly middle class, suburban and more or less average; the other, with some shining exceptions, beset with problems. And as LEP enrollment expands, as seems likely, the problems will grow with it.
More important, the numbers call attention to many things we don’t know. What percentage of our LEP students, for example, are recent arrivals in this country, and thus could be expected to register dismal scores on a test that’s given in English? How many have been here for five years or more or, indeed, were born here and are thus clearly the products of inadequate schools?
It would be useful if we had at least a sampling of test scores for each of these different groups of students. Such a sampling would yield a far better idea of how well the schools are doing. For a man with Unz’s interests and resources, it might be a productive investment.
Nor do we know how they’re being taught. In his campaign for Proposition 227, Unz assumed that as long as LEP students were being taught in English, much of the problem would be solved.
But there’s at least anecdotal evidence that in the years before the passage of his initiative, a great many LEP students, only a third of whom were in bilingual classes, weren’t really being taught to read in any language. And while Proposition 227 strongly discourages reliance on students’ native language in the classroom, no one in California is collecting systematic data on its implementation.
Eight months ago, Unz was talking about suing school districts for what he believed was widespread evasion of his measure. Now, based on a different set of anecdotes, he proclaims it a success.
And it may indeed be a success — not because of the inconclusive scores, but because it has started forcing school districts to treat immigrant students like all others, and not as inmates of a sheltered workshop.
But it’s only a start. Even after the passage of Proposition 227, CABE, the California Association for Bilingual Education, and other groups (including some within the state Department of Education) were pressing hard, though so far without success, to exempt LEP students from some of the state’s curricular standards. They sought to maintain what was, in effect, a separate curricular track.
At the same time, the Legislature is moving a bill, AB 144, that would exempt from the Standardized Testing and Reporting program most students who have been in U.S. schools for two years or less and who have not had at least six months’ instruction in English. The bill, by Assemblywoman Carole Migden, has passed the Assembly, and is likely to pass the Senate as well.
Gov. Gray Davis’ office says he has “concerns about it.” One hopes he will veto it. If he doesn’t, we’ll be back to the sheltered workshop.
The backers of Migden’s bill argue that children who know little English can’t be expected to understand even the directions of a test that’s given in English. But testing all students does establish a baseline, a starting point against which to measure progress. And if those measurements are keyed to the length a student has been in U.S. schools, we may also be able to make some judgments about the effectiveness of the teaching — and the relative quality of the teachers — that those students get.
The small gains on this year’s test have been attributed to all sorts of things: class-size reduction; greater emphasis on phonics and other basic instruction; the phaseout of bilingual education; and the high stakes that the state’s school accountability policies are putting on success.
In that context, Unz’s initiative was a necessary, though not a sufficient, instrument of school improvement for the 1.4 million California kids officially labeled as LEP. If Unz were now to follow up, rather than celebrate, the victory could be greater still.