State Test Finds Students Lagging : Education: For second straight year, few meet standards for reading, writing and math. Schools superintendent Eastin says reform is vital.

For the second straight year, California students have failed to measure up to demanding standards in reading, writing and mathematics, according to the final results released Tuesday in the state’s pioneering effort to boost public school performance.

Scores on the California Learning Assessment System (CLAS) tests show that only a tiny fraction of the state’s fourth, eighth and 10th graders are advanced enough to meet the state’s expectations for top students--the ability to write clearly and confidently, to grasp complexities when reading or to apply a sophisticated mathematical sensibility to real-world problems.

Moreover, a solid majority of students in all three grades failed to meet even minimally acceptable performance levels in all three subjects, with the showing in math by far the weakest of the three.

In response to the new scores--as well as soon-to-be released national data that is expected to contain similar findings--new state Superintendent of Public Instruction Delaine Eastin announced she will convene task forces to consider a wide variety of explanations for the low scores and recommend corrective actions within four months.

“The numbers are just abysmal,” said Eastin, at a news conference at a Sacramento elementary school, where she was joined by representatives of major education, business and parent groups.

Although she cited a list of factors--such as overcrowded classes, large numbers of students who are poor or do not speak English, and a shortage of time and money for teacher training--she said California must reform its schools so that by the end of third grade, every student is able to read and achieve basic competency in math.

Gov. Pete Wilson also decried the results--calling them “absolutely unacceptable”--and using their release to argue for his education agenda, which would make it easier to fire incompetent teachers, free districts from the state’s lengthy education code and establish uniform performance standards.

*

The 2-year-old CLAS system was designed to measure student achievement against objective standards, rather than comparing the performance of each school to a statewide average, as past exams had done.

Last year, results were deemed dismal across the state, especially in math, and some educators complained the new high students were beyond the reach of most students.

This year’s results were little better.

More than 40% of the 10th graders tested in math, for example, achieved only the lowest of six possible levels, meaning that they showed “little or no mathematical thinking and understanding of mathematical ideas.” More than three-quarters of the students in all three grades scored at one of the three lowest levels, indicating less than basic understanding.

Reading was only slightly better, with more than 60% of the students in each grade failing to achieve at a level four or above--meaning those students were not even able to muster “a partial mastery of knowledge and skills fundamental for proficient work.”

The scores in writing, which has been part of the state’s assessment system since 1987, were slightly stronger; about 54% of the students in the eighth grade scored at the lowest three levels, 61% in the 10th grade and 68% in grade 4.

Asian American students performed substantially better than their counterparts in all three subjects and all three grades. Even so, only 30% of the Asian American 10th graders scored at a minimally acceptable or better level in math. Among white students, 20% were at that basic level or above; among African Americans, only 3% ranked above the basic level, and among Latinos the figure was 5%.

The results were based on tests students took last spring. State testing officials warned that the scores cannot be compared to those from 1993, because of changes in the test questions and in scoring procedures aimed at ensuring greater accuracy in response to problems that bedeviled the system last year.

The tests themselves--hailed at their introduction two years ago for making California a national leader in improving student assessment--have generated almost as much controversy as the low scores.

Last year, the release of CLAS scores fueled distrust in some quarters when it became clear that technical errors had resulted in the release of inaccurate scores, or no scores at all, for about 250 schools.

And the tests already had become the target of conservative and religious parents and groups, who zeroed in on reading passages they said were too provocative and writing tasks they claimed violated students’ privacy by probing their beliefs.

Legal battles over whether districts could be forced to administer the tests and mounting criticism of the state education department’s response to the uproar led to the demise of the test, which many educators had embraced as an important tool for guiding the reform and improvement of school curriculum.

In September, Wilson vetoed a $26-million bill that would have kept the testing program alive. Wilson’s office is working with the Legislature on a bill to develop a new assessment system.

*

With the release of the final set of scores Tuesday, some educators and business leaders repeated their concerns that the loss of the CLAS test threatens gains California schools have made in making curriculum more challenging and more in line with job market demands.

“Its demise may send a chilling signal through the educational system nationally,” said Eva Baker, a researcher who heads the Center for the Study of Evaluation at UCLA. CLAS had made the state a leader by requiring students to demonstrate their knowledge in essays, open-ended math problems and other so-called “authentic” activities, she said.

“What worries me is that the aftertaste will linger in the wrong way,” Baker said. “There is no reason to believe that tests that ask people to write on complex issues need to raise the specter of invasion of privacy.”

But that approach, requiring students to write essays in all three subject areas, introduced a variety of problems. Their answers were costly to score, provoked controversy among parents, and led to a variety of technical missteps.

Last year, in an effort to save money, the state tried to project results for schools and districts by grading only some of the students’ answers, resulting in the release of results later deemed unreliable.

In an effort to guard against that, every test was graded this year and state testing officials said they would not report scores for any schools or grades in which more than 25% of the students were not allowed by their parents to take the test. The state also did not report scores for small schools or schools where the demographic makeup changed substantially from last year.

About 500 schools and dozens of districts, including five in Los Angeles County, did not receive any test results because of those procedures. Others, however, had a low participation rate and still received scores, once again raising questions about what the scores mean.

In West Covina, only 10% of the district’s sophomores, 40% of the eighth graders and 45% of the fourth graders chose to take the test after the district gave parents the option of having their children exempted.

Still, the state reported scores for the district and each of its schools, which for the 10th grade pushed the working class community into the ranks of such perennially high-achieving districts as those serving San Marino, Beverly Hills and La Canada Flintridge.

Testing officials said the scores were released because the district must have failed to report to the state the number of parents who chose to have their children excluded.

But John F. Costello, West Covina’s superintendent, said the state erred again. “What you’ve got is a continuation of the mishandling of this particular test,” Costello said. “The first year you can always understand the mistake, but it’s more difficult the second time.”

Other educators were concerned about the nature of the test.

Janet Salem, the assistant principal at Jordan High School, said she believed several portions of the CLAS tests were too subjective. “Some of the questions were personal,” Salem said. “How do you correct the subjective components of the test?”

Absenteeism during the several days that the test was administered also might have skewed the results at Jordan, she said. “It’s dependent on the whole group being present the whole three or four days.”

Of Jordan’s 10th graders, 99% scored at the lowest three levels in math, 84% at those levels in writing, and 88% in reading.

Countywide, Los Angeles lagged behind many in the state, with the usual range of high scores in affluent districts and dismal scores in districts with high numbers of poor children, such as Compton and Bassett Unified in La Puente.

*

One bright note was that seven districts in the county cracked the list of the top 20 in the state at the 10th grade level, led by the San Marino district, which posted California’s highest scores. The rankings were based on the combined percentages of students scoring at the top three levels in all three subjects.

Gretchen Whitney High School in the ABC District, a highly selective school that admits students based on high test scores, was the top high school in the state, followed by California Academy in Long Beach and San Marino High School.

At the elementary and middle school levels, however, San Marino did not receive any scores because too many parents declined to have their children participate.

In Ventura County, school officials were pleased that fourth- and eighth-grade students scored above those statewide in all three subjects. But some educators were concerned about the results for 10th-graders, whose scores dipped below those across the state in writing.

The high school students’ mediocre results could have resulted from the large number of students who opted not to take the test, said Charles Weis, who heads the Ventura County office of education.

“On the surface, we can take some solace in the fact that we are doing better than the kids statewide,” said Weis. “But I’m not sure what these scores really tell us.”

Orange County students exceeded the statewide performance in every subject and grade. The county ranked second behind Marin County in fourth grade, third behind Nevada and Marin counties in 10th grade, and placed sixth in the eighth grade.

The strongest performance came on the 10th grade reading exam, where Orange County ranked best in the state.

“Overall, I’m elated that Orange County did so well compared to the statewide average, but I’m also disappointed that we didn’t do better,” said John F. Dean, superintendent of Orange County schools. “We’d like to be 100% in everything. We keep striving for this.”

Individual county schools excelled. The brightest star was Roy Andersen Elementary in Newport, which posted the highest combination of reading, writing and math scores at the fourth-grade level.

Among school districts, the Laguna Beach Unified School District topped the fourth grade results. Superintendent Paul M. Possemato attributed the high scores to his district’s emphasis on language skills.

“I’m a strong advocate for the CLAS-type testing,” he said. “I believe that one has to have test programs that force the examination of, the measurement of, and the use of language and thinking in all subject areas.”

Times staff writers Jodi Wilgoren and Beth Shuster and correspondent Kay Saillant contributed to this story.