High-school grades are far more reliable than SAT I scores in predicting how well minority students will do in college, according to a new study by researchers at the University of California, Davis.

Merna Villarejo, a professor emerita of microbiology at UC Davis, will present the findings Saturday, Nov. 17, at a conference, "Rethinking the SAT in University Admissions," sponsored by the Academic Senate Center for Faculty Outreach at UC Santa Barbara.

The researchers -- examining the college performance of 1,274 minority students who entered UC Davis as freshmen from 1988 through 1994 -- found that high-school preparation was associated with persistence and performance from basic science classes all the way through to graduation.

"In fact, high-school GPA is the single most important predictor of all positive academic outcomes we measured," said Villarejo, who directs the National Institutes of Health Initiative for Minority Student Development at UC Davis.

Villarejo and postgraduate researcher Amy Barlow compared the success of 397 students who enrolled in UC Davis' Biology Undergraduate Scholars Program with 877 students who chose not to participate in the intensive enrichment program.

The Biology Undergraduate Scholars Program (BUSP), established in 1988, offers disadvantaged students supplemental courses and seminars, academic counseling and research opportunities to help them succeed in the biological sciences. The program is supported with grants from the Howard Hughes Medical Institute, National Institutes of Health-Initiative for Minority Student Development, California Alliance for Minority Participation (NSF), and the Genentech Foundation.

In a study with implications for university admission procedures, Villarejo and Barlow found that higher SAT scores did correlate with students' success in basic math and chemistry, as well as their chances of graduating with at least a B-average -- the minimum required for most graduate and professional programs.

However, high-school GPA far outweighed SAT scores in predicting the students' chances of success in college. Moreover, the study found that a program of academic enrichment and personal support, such as BUSP, can largely compensate for poor high-school preparation.

"For this population of minority and disadvantaged students at UC Davis, SAT I scores are of no predictive value if we are interested in merely getting students through to graduation, with any GPA in any major," Villarejo said. "Rather, high-school GPA provides the best indicator of whether a student will graduate."

"If we are interested in selecting students who are likely to excel academically in any major, both verbal and math SAT I scores appear to be of some value, with 100-point increases in SAT I score yielding a 50 percent and 33 percent increase, respectively, in this outcome.

"Comparing these modest SAT effects with the estimated 14.5-fold impact of a one-point increase in high-school GPA, however, calls into question the validity of the weighting SAT I scores have generally received in college admissions. At the very least, a quantitative admissions formula heavily weighted in favor of high school GPA is warranted for this population."