Revised SAT Test Yields Large Gains

By RICHARD LEE COLVIN

Aug. 24, 1995 12 AM PT

TIMES STAFF WRITER

College-bound seniors who took a renamed and rewritten SAT in 1995 chalked up the biggest gains in more than a decade, reversing a steady slide in verbal scores and adding momentum to an upward trend in math scores that began in 1980.

The improvement was particularly dramatic among students near the top of their class and women, although men also registered strong scores. Verbal scores for African American and Latino students surged, but their math scores did not.

Experts in education reform cautioned against drawing broad conclusions from the one-year jump, especially because it comes from students taking a new version of the venerable examination that has been used by colleges since 1927 in deciding who to let in.

But Wednesday, U.S. Secretary of Education Richard W. Riley joined officials of the College Board, the private group that administers the test, in attributing the higher scores to more than a decade of reform. Students today take far more academic “solids” such as chemistry and trigonometry, advanced literature courses and honors history than they did in 1983, when the modern school improvement movement was launched.

“You have to be willing to take the difficult courses, and those kinds of things are paying off,” Riley said.

The scores to be released today show a five-point increase, to 428, in the average verbal score, and a three-point increase in math, to 482, on a scale of 200 to 800 points.

The verbal score continues to lag 32 points behind what was recorded in 1970 while the math score is the best since 1971. More than 1 million students, 41% of this year’s graduates, took the 2 1/2-hour test.

Riley said the progress is slower than he would like but that the test score increase “certainly indicates to me that what we are doing in terms of standards . . . happens to be working, and it is no time to get off the sustained drive for improvement.”

He said he feared that a proposal by Republicans in the House of Representatives to cut $36 billion from federal spending on education over seven years could derail the progress; the Clinton Administration is proposing a $40-billion increase in education spending during that period.

California students contributed to the national upswing, boosting their average verbal score four points and their math score three points. The state’s math score of 485 was three points above that of the nation, but its verbal score of 417 lagged 11 points behind.

State Supt. of Public Instruction Delaine Eastin said the 128,000 California students who took the test were far more likely than those nationally to be minorities and to come from low-income and non-English-speaking families.

She said California students whose first language is English outpaced their peers nationwide by seven points in math and tied them on the verbal segment of the test.

Students in the Los Angeles Unified School District, the nation’s second largest, fell further behind the state and the nation, dropping two points to 351 in verbal and six points to 418 in math.

“The scores for too many students in too many schools remain too low,” said Board of Education President Mark Slavkin. “We need to move quickly with our reforms and other efforts to show results. This is their only shot. There is a sense of urgency.”

The College Board said scores in large cities and rural areas, where students were less likely than suburban students to take a full complement of academic courses, were below those of better-prepared students in suburban communities. Even when they took the same courses as their suburban counterparts, urban students’ scores lagged.

The annual release of scores on the SAT, which is now known as the Scholastic Assessment Test instead of the “aptitude” test, has been a trying event for educators since the 1960s, when scores began plunging, touching off annual waves of criticism of American schools and convulsive movements to seek solutions.

The College Board began updating the test in 1987 in an effort to reflect and support trends toward more problem solving in mathematics and more emphasis on reading for meaning and understanding. But this is the first year that the new version of the test has been widely used.

Robert L. Linn, an education professor at the University of Colorado at Boulder and co-director of UCLA’s Center for Research on Evaluation, Standards and Student Testing, reviewed the College Board’s methodology, and said the score increases this year probably were not caused by changes in the test.

But, he said, the explanation is probably more complicated than simply that students are taking more academic courses.

The new math test includes, for the first time, 10 problems that require students to work out a solution, rather than choose among a list of possible answers. Students also were encouraged to use calculators, which have been banned in the past, although most of the questions do not lend themselves easily to calculator use.

Eliminated from the verbal part of the test was the much-despised antonyms section, which bedeviled generations of students by forcing them to memorize the meaning of relatively obscure words such as “mutability,” which means subject to change, or “salubrious,” which means healthful, as well as their opposites.

The new test has more and lengthier reading passages that the College Board said were supposed to be more accessible to students. Samples of the passages from the test covered relatively arcane topics such as the difference between English “garden cities” and modern cities.

College Board President Donald M. Stewart said that despite the changes the test still measures students’ developed math and verbal skills. “It’s how we’re measuring it that’s changed,” he said.

The College Board has been conducting a national campaign to acquaint students with the new test format for five years. The campaign caused far more students to prepare for the exam by enrolling in commercial testing courses as well as seminars offered by schools and universities, which might in and of itself explain the increase.

Kaplan Education Centers, which says it is the nation’s largest standardized test preparation service, said it experienced a 50% increase in SAT-related business as a result of students’ concerns about the new exam.

Jonathan Grayer, Kaplan’s president and CEO, said the changes in the test eliminated some of the most difficult questions--such as the antonyms portion--and made it easier for top students.

“I’d hate to conclude that our nation’s educational problems are solved based on one year of data with the new test,” Grayer said.

But Michael K. Smith, an assistant professor of education at the University of Tennessee at Knoxville, wrote in an article in the journal Phi Delta Kappan last year that the changes were not that dramatic and still had little significance for high school educators.

He and others said the relatively small changes in the test mean that the College Board has wasted an opportunity to add to the momentum for educational reform.

The higher scores this year represent progress, but “certainly not huge progress,” said David D. Marsh, a USC education professor. “The SAT is still not very well aligned with all the things that we want kids to learn in high school.”

Stanford University Prof. Michael Kirst said the changes are welcome, but he characterized them as more of a tuneup than an overhaul.

He said he doubted that the higher scores could be explained by the fact that students are taking harder, academic courses. That trend, he said, had been going on for more than a decade without producing a corresponding increase in SAT scores.

“If these trends would continue for a while, then it would be positive, but I just wish they would have occurred earlier,” Kirst said.

Barbara Lerner, a Princeton education consultant who has designed and studied tests, agreed. Unfortunately, she said, it will not be easy to follow that trend because the College Board plans a radical overhaul of how scores are reported next year.

“Once you get the new figures next year, it will look like our verbal score went way up . . . but it won’t necessarily have changed at all,” she said.

Stewart was more upbeat, saying that he was “cautiously optimistic” that the new scores signal a turnaround in American schools that has resulted in part from a more serious attitude among students.

“This is the best prepared class in recent history.” Stewart said.

Despite the overall increases, the gap persists between the academic achievement of white and Asian American students on the one hand and African American and Latino students on the other. African American students, for example, still score on average 150 points lower than Asian American students and 110 points lower than white students in math.

All ethnic groups registered gains on the verbal portion of the test, ranging from a two-point increase for Asian Americans to seven points for American Indians. African American and Mexican American students each raised their average by four points and white students were up by five points.

The average score for white and Asian American students on the math part of the test rose by three points while the average for African American students remained unchanged and dropped by one point for Mexican American students.

Results for Latino students whose families originated in countries other than Mexico are reported separately by the College Board.

L.A. RESULTS: L.A. schools’ students fared poorly on SAT test. B1

(BEGIN TEXT OF INFOBOX / INFOGRAPHIC)

Changes in the SAT

College-bound seniors in the Class of 1995 took a new version of the Scholastic Assessment Test. The College Board, which administers the exam, maintains that the overhauled version has the same level of difficulty and reliability as the old test. The 30-minute standard written English test has been dropped. Some other major changes:

Verbal Section

* 78 questions, down from 85

* 51% of questions require reading, up from 29%.

* Reading passages are 400 to 850 words, up from 200 to 450.

* Questions focus on critical reading skills.

* Students analyze and compare a pair of passages with two points of view.

* Antonym questions dropped.

* 75 minutes to answer questions, up from 60.

****

Math Section

* More emphasis on application of concepts and interpretation of data.

* 10 questions in a new form that ask students to produce their own answers and enter them on grids.