It's Hard To Get Straight Answers On Standard Tests

Education: Difficulty gauging impact of limited-English students on schools' overall scores brings hard feelings, court intervention.

Roughly three in four Santa Ana students struggled through a standardized test they didn’t understand. In Garden Grove, that was the case for about half the students.

And that, in a nutshell, is why the statewide test results are mired in a court battle just two days before they were supposed to be released to the public via the Internet.

A San Francisco court judge Thursday ordered the state Department of Education to publicize only scores that do not include students who speak limited English.

The decision favored Bay Area school administrators who filed a complaint, arguing that scores of students who do not speak English are not meaningful because the exam is not designed for such students.

The scores in Orange County, where 25 of the 27 school districts already have released at least partial results, vividly show why the scores of the limited-English students have become a sore point among educators, sometimes pulling down a school’s overall scores to the point where it is difficult to tell how the English-speaking students fared.

Limited-English students consistently have scored in the bottom quarter or so of students nationwide, the Orange County results show. In schools with a sizable population of those students, the overall scores end up looking mediocre to dismal, even when most of the English-speaking students tested at or above average.

Consider the Santa Ana Unified School District.

There, second-graders who speak English ranked at the 49th percentile in reading, meaning they did better than 49% of their peers in a national sample. Limited-English students, who represented 67% of the second-graders tested, scored at the 17th percentile, pulling down the district’s overall second-grade score to the 21st percentile rank.

“If you look at Santa Ana’s English-only scores, they are hovering around the middle. All of a sudden Santa Ana looks normal,” said Howard, whose district overall is 71% limited-English speaking. “But when we bring in the whole limited-English population, the numbers look low.”

Because those students’ scores often are so much lower than their English-speaking counterparts’, even a smaller number of tests taken by limited-English students can skew the scores.

In the Newport-Mesa Unified School District, for example, 29% of the fourth-graders are not fluent in English and tested at the 18th percentile in the standardized test. Native English speakers scored at the 68th percentile, with the overall ranking at the 51st percentile.

With numbers like that, many Orange County educators were cheering the court decision, which will result in a showing of overall higher scores when the state publishes its scores. That publication, originally scheduled for Tuesday, has been delayed indefinitely by the court decision.

“When you give a test to a kid who doesn’t know the language it’s written in , you don’t get valid results,” said Al Sims, testing administrator at Garden Grove Unified School District where 50% of the 34,000 students tested do not speak fluent English.

Although the court ruling does not apply to local school districts, Sims said his district chose to withhold scores of limited-English students as a safeguard.

Santa Ana Unified director of bilingual education Howard Bryan called the ruling a “very fair decision” that ensures scores will be presented equally.

“The whole issue of including limited-English students is skewing the scores,” Bryan said.

Concern over testing students who are not fluent in English erupted in March when the San Francisco Unified School District refused to give the state test to its 6,000 non-fluent students. District officials took the state to court and won two rulings in its favor.

Meanwhile, most of the state’s other public schools tested fluent and non-fluent students alike.

In Irvine Unified, where 11% of its students speak limited English, overall test figures were generally at the 70th percentile, well above the national averages. But scores of limited-English students lowered districtwide results in some grade levels by four points.

Irvine Unified testing coordinator Beverly Huff said news of the judge’s decision was bittersweet.

“It’s a good step at the wrong time,” Huff said.

The advantages of the recent court order is that districts won’t be unfairly judged, she said. But there is value to having the test data on limited-English students, Huff added.

“If we want to talk about all students then we should test them all and be able to pull out specialized populations to do analyses,” she said. “There’s nothing wrong with that.”

The downside, Huff added, is that the ruling comes days before the state was supposed to post all California test scores broken down by individual schools, she said. Further complicating matters, various districts have been publicly releasing information over the past two weeks. The remaining few districts are left in a quandary of what they are permitted to do.

“It’s difficult being midstream,” Huff added. “Everyone’s wrestling with their data.”

Two Orange County school districts, Orange Unified School District and Savannah School District, have not released test scores.

Grading the Schools

Four more Orange County school districts released scores in the state’s new standardized test of basic skills, the Stanford 9. The test was given to students in grades two to 11 from mid-March to mid-May.

Editor’s note: The percentile figures show how student scores ranked, on average, against a nationally selected group in spelling, reading, language and math. High school students were tested in reading, language, math, science and social studies. Those at

the 50th percentile, for example, scored higher than one half of the pool and lower than the other half.

“LEP” designates students who are limited English proficient. “R-FEP” refers to students who were formerly LEP but have been reclassified as fluent English proficient, and “FEP” refers in general to students who have a native language other than English but are also fluent in English.

Comments are closed.