YOU ARE HERE: LAT HomeCollections

A Special Report / THE 'STANFORD 9' TESTS | REPORT
CARD: How L.A. County Schools Fared on Statewide Exams
: What the Scores Tell Us

Pockets of Progress, but Question Marks, Too

Bilingual education: Experts say small increases make it difficult to tell if Prop. 227 or other reforms are responsible.


A year after voters approved the anti-bilingual education law Proposition 227, tens of thousands of California schoolchildren with limited English skills showed only modest gains in Stanford 9 test scores released Thursday.

Statewide, third-grade reading scores for English learners, who make up 21% of the test takers in California, rose only 4 percentile points.

In the Los Angeles Unified School District, where about 313,400 students, or 46%, are classified as limited English-proficient, third-grade reading scores for English learners crept up from the 13th percentile to the 15th. In math, they rose from the 24th percentile to the 29th.

Third grade is a benchmark year for students, after which it becomes difficult for them to learn to read; the vocabulary expands so dramatically in the fourth grade that few poor readers catch up.

Testing experts believe the increases are so small that it may be hard to separate the impact of Proposition 227 from that of other reforms implemented over the past year.

"It's going to be hard to infer a whole lot from these scores," said Kenji Hakuta, an education professor at Stanford University and an expert in bilingual education. "By blaming the language of instruction as the bad guy, Proposition 227 may have only distracted us from all the other things we need to do to significantly improve schools."

The initiative dismantled most bilingual education programs in public schools statewide and required that students with limited English skills be placed in structured English immersion classes. The aim is to have them transferred into all-English classes within one year.

Reflecting overall trends statewide, English learners posted their best scores in earlier grades, where class-size reduction, new textbooks, teacher training and renewed emphasis on basic skills and phonics are apparently starting to pay off--particularly in math.

In later grades, overall scores rose 1 or 2 percentile points, which was predictable given that students were more familiar with the exam process in this, the second year of administering the Stanford 9 test.

Nonetheless, stalwarts on each side of the bilingual education debate tried to interpret the modest improvements along political lines separating fierce critics and stubborn loyalists.

"These numbers really are remarkable," said Silicon Valley millionaire Ron Unz, who co-wrote the law that essentially erased a method of teaching used for three decades in California.

"The overall test scores of immigrant kids in California have gone up by a margin of 20% in less than a year, which is unprecedented among educational reforms," he said. "These percentile increases represent huge gains that dwarf other reforms such as class-size reduction, which at a cost of $4 billion created improvements of less than 1 percentile point."

Wrong, all wrong, argued Thomas Saenz, regional counsel for the Mexican American Legal Defense and Educational Fund.

"Ron should know better than that; there's no indication that Proposition 227 is the cause of any increase," he said. "In fact, a lot of other reforms have been concentrated in the earlier grades where educators believe the greatest gains are to be made.

"If he was right, we would expect to see gains in kids in 227 programs and no gains or smaller gains for kids in bilingual education," he said. "Instead, we see similar gains across the board. That means we have to look someplace else for an explanation."

Surprisingly, however, California's English learners bucked a national trend by avoiding the mysterious drop in math test scores chalked up by high school students fluent in English. Instead, English learners in grades two through 11 showed consistent gains in math. In second and third grades, they saw their scores rise by as much as 7 percentile points.

Pending a formal study of that finding, Hakuta said the higher math scores may be linked to the fact that English learners perform better on the mechanical math questions than on problem-solving that involves complex language.

At Walnut Park Elementary School in Huntington Park, Principal Kenneth Urbina was not all that surprised that his Spanish-speaking students' test scores edged upward, because "we were already in a fighting mode and rose to the occasion."

Reading scores among limited-English third-graders at Walnut Park climbed from the 10th percentile to the 14th. Their math scores rose from the 16th percentile to the 24th.

"We pored over previous test score data and we provided instruction with an eye toward making improvements," said Urbina, whose working-class school has about 900 English-language learners out of a total of 1,100 students. "And Proposition 227 forced our teachers to take a close, hard look at how children learn language."

Los Angeles Times Articles