SAT Math Scores Are Up

This fall's college freshmen were the last to take the old SAT -- and they did well on the mathematics portion, posting a 2-point gain, to an average of 520. Over the last 10 years, the average math score increased by 14 points, a gain that College Board officials said was significant and attributed to increases in the number of students taking rigorous math courses in high school.

But the statistics released by the College Board on Tuesday also had plenty of sobering news: Verbal scores were flat. And over 10 years, verbal scores increased by only 4 points, to an average of 508. In addition, over the last 10 years gaps in performance levels among members of ethnic and racial groups have grown. Over the last decade, for example, the average score for Asian Americans rose by 25 points on the SAT math test, while the score for black increased by an average of 9 points. That leaves the average for African American students, 431, at 149 points behind the Asian American average of 580.

The following table shows the breakdowns on scores and gains by racial and ethnic groups.

SAT Average Scores and Gains, by Race and Ethnicity, 2005

College Board officials attributed both the good news about rising math scores and the bad news about racial gaps to preparation in high school. Over the last 10 years, increasing number of students, across racial and ethnic groups, have been taking precalculus during high school, with the national average increasing from 37 to 48 percent of students.

But the gaps are striking in course preparation by racial and ethnic group. In high school, 44 percent of Asian students take calculus, compared to 28  percent of white students, 19 percent of Mexican-American students, and 14 percent of black students. Gaston Caperton, president of the College Board, said that the single most important thing to improve educational performance would be to offer tougher courses in all schools, and to encourage more students to take those courses.

The College Board, as it typically does, invited a group of students to appear with the board's officials. And while the students also talked about the importance of high school preparation, several also mentioned their sense that the students who can afford to hire tutors or attend coaching sessions do better on the SAT than those who can't afford to do so.

Students who can afford SAT tutoring aren't necessarily smarter than other students, but "they learn the formula," said Deborah Francois, a high school senior from Brooklyn. "It's unfair."

During the press briefing, College Board officials didn't respond to her comment or similar remarks from another student the board had invited to the briefing. Historically, the College Board has played down the potential impact of coaching on SAT scores. And reports about the more sophisticated and expensive test-prep services achieving large gains have embarrassed the board, as they raise questions about the fairness of colleges' use of the SAT.

In an interview after the briefing, Caperton said that coaching typically helps students increase by about 40 points, so he acknowledged some gains. But he said that most of those gains were not from the actual tutoring sessions, but from practicing. Tutoring programs frequently involve taking numerous practice tests, he said, and students who can't afford tutoring can gain the same benefits from taking practice tests provided by schools or books. "It's preparation," he said.

The New Writing Test

The big news about the SAT in the last year has been the addition of a writing test. While three administrations of the SAT this spring have featured the new test, those SAT scores are not included in the overall averages released as most of the students who took the SAT this spring are juniors, not seniors.

Given the enormous interest in the new writing test, the College Board broke with normal practice and released average scores on the spring administrations of the new SAT, along with comparison averages for the spring administrations in previous years. Board officials stressed that these averages would probably decline by next year when total figures for the year are released, because spring test-takers are usually the more ambitious and better prepared students.

But of those students who did take the SAT in the spring, the math average was 537, up 4 points from the average for last spring; the verbal average was 519, down 2 points from a year ago; and the new writing average was 516.

The College Board did not release ethnic or racial breakdowns for the spring scores, but Wayne Camara, vice president for research and psychometrics at the board, said that early indications are that "subgroup gaps are similar or smaller" than they are for other portions of the test.

The writing test was added amid complaints from colleges that the old SAT was flawed to ignore such an important skill. But data released Tuesday of a survey of admissions officers at competitive colleges found that many are being cautious about how they will consider scores on the writing test.

The survey, by Kaplan Test Prep and Admissions, found that 47 percent of colleges are adopting an interim policy of not counting the score on the new SAT and 15 percent are giving the new writing score equal weight with the math and verbal tests. Many other colleges will be looking at the writing scores, but not necessarily counting them equally with other scores. While the Kaplan survey found some skepticism about the new writing test, it also found that the small group of colleges giving it full weight include some of the most elite institutions around, such as Harvard and Stanford Universities.

At the College Board briefing, Caperton said that the new writing test sent an important message to high school students: "Writing is a requirement for doing well in the college and in life."