Scores mirror demographics

EDUCATION: Schools with poorer and immigrant populations generally posted poorer scores.

It’s not hard to predict where to find schools with high test scores. Look for neighborhoods with expensive homes, well-educated parents and low numbers of non-English speaking pupils.

Schools with low scores are just as easy to spot ahead of time. Look for neighborhoods with high rates of poverty, immigrants and transient apartment dwellers.

In most cases, scores on the Stanford 9 tests are a product of the background of kids who were tested — not what students learned in class.

“I believe the scores reflect the challenges we face in a community like Santa Ana,” said Al Mijares, superintendent of Santa Ana Unified School District, which has among the lowest scores, highest rates of poverty and most limited-English students in Orange County. “But I don’t want that to become an excuse that poor people can’t learn.”

In addition to raw scores for each school, the state Department of Education also released scores for different groups of students: limited-English, males and females, economically disadvantaged.

Orange County and California have more than 10 times as many students who speak limited English as the sample group of students that was used to establish averages on the Stanford 9 tests.

Remove the scores of limited-English students, and state and local students outperform the national average — the 50th percentile — in the vast majority of categories. Among fluent speakers of English across Orange County, 10th-grade reading — in the 49th percentile — was the only category in which students scored below the national average.

Even in low-scoring Santa Ana, students fluent in English beat the national average in 26 of 43 test categories.

“This shows that California does do well on basic skills test, especially kids who speak fluent English,” state Superintendent Delaine Eastin said.

But questions remain about the accuracy of the test results. Scores were delayed for three weeks because of computer programming errors by test publisher Harcourt Educational Measurement that miscalculated the scores for 144,000 students in year-round schools and accidentally grouped scores of more than 200,000 fluent English speakers with those who speak limited English.

Language proficiency designation for 419,550 students statewide — almost 10 percent of all test takers — was still missing Thursday.

And the stakes on the tests are rising.

This fall, 430 schools statewide that averaged below the 50th percentile will be chosen for extra support and eventual closure or takeover if they fail to improve their Stanford 9 results. For students, low scores will lead to mandatory summer school and repeating grades if they fail to make progress.

“All of us are so sensitive to the matter of test scores that we have to be able to rely on them,” Mijares said.

Mijares said he is already using the test scores to evaluate teachers, principals and administrators. Those who fail to show improvement over the next three to five years should look for a different job, he said.

“In the case of a teacher who’s underperforming, we have a responsibility to remove that person,” he said. “If there’s a pervasive problem at a school, we have to look at the leadership. The price tag is too high. We’re talking about students’ lives.” Register staff writer John Gittelsohn can be reached at [email protected]

Comments are closed.