SAT II Better Predictor Of College Success, UC Says
Its name strikes fear in the hearts of high school seniors. College-admissions officers use it to gauge their applicant pools. And a whole cottage industry of test-preparation companies has grown up to help students master it.
But the SAT, the nation's most widely used college-entrance exam, may not be as good at predicting success in college as the more subject-oriented SAT II, a study by researchers at the University of California concludes.
The researchers found that the SAT II achievement tests were better at predicting freshman-year grades for students entering the University of California system, the nation's largest four-year university system, with 170,000 students on 10 campuses. The authors of the study also say that SAT II scores were less likely to be affected by students' socioeconomic backgrounds.
The results are sure to add fuel to the national debate about the importance of the SAT, which is taken by more than 1.3 million high school seniors each year. Some educators have argued that colleges place too much emphasis on SAT results.
UC President Richard C. Atkinson proposed last February that the SAT be eliminated as a requirement in UC's admissions process. He recommended using the SAT II or a statewide, standardized achievement test instead. ("UC President Pitches Plan To End Use of SAT in Admissions," Feb. 28, 2001.)
The request is currently under review by UC's faculty senate. It also must gain the approval of the university system's board of regents.
The study appears to contradict an earlier review commissioned by the College Board, the sponsor of the SAT and the SAT II, which found that the SAT was a solid measure of students' academic performance in the first year of college. ("SAT Said to Be Reliable Predictor of College Success," May 9, 2001.)
Doubts at College Board
For the new study, the UC researchers examined the records of nearly 78,000 freshmen who entered the university system from 1996 to 1999. UC has required applicants to take both the SAT, which stresses verbal and mathematical aptitude, and the SAT II, which tests for mastery of specific subjects, since 1968.
"We find that rather than looking for aptitude, looking for what they know now and can master is a better predictor" of success in college coursework, said Saul Geiser, a researcher and co-author of the study. "That was our number-one finding."
Asked why the analysis found that the SAT II was a better predictor than the SAT I, Mr. Geiser declined to offer an explanation. "I have my theories, but I don't have all of the data to back them up yet," he said. "We're looking into causes. All we know is the results are there."
Amy Schmidt, the director of higher education research for the New York City-based College Board, noted that University of California students are a "very select group" and said the findings might not hold true elsewhere.
"The results could be different from the rest of the nation," Ms. Schmidt said.
Mr. Geiser doesn't dispute that. "I have no way to know if these results are generalizable to other colleges and universities," he said. "But it's true for us."
Ms. Schmidt also noted that the UC study found little difference in the ability to predict college success by combining the SAT, the SAT II, and high school grade point average as compared to using just the SAT II and high school GPA.
The 'Right Message'
Mr. Geiser said that none of the factors was very good at predicting success. "Our ability to predict in general is not very good," he said. But Mr. Geiser defended the report's findings, saying that "achievement is a better predictor and fairer."
Aptitude tests seek to measure innate intelligence, Mr. Geiser said, while achievement tests seek to measure the mastery of certain subjects. "With achievement [tests], if you study hard and work, you can do better," he said. "The achievement message sends the right message."
Vol. 21, Issue 10, Page 14