High School Grades Outperform SAT

University Testing


FairTest Examiner - July 2007


Researchers have long known that high school grades forecast first year college performance more accurately than any standardized test does. Even leading test-makers such as the College Board, Educational Testing Service, and ACT, Inc., admit that fact, though the evidence is often buried in obscure technical reports.


Now a major study based on 80,000 University of California (UC) students concludes that high school grade point average (GPA) is also the strongest predictor of four year undergraduate performance. Moreover, GPA was found to be less closely tied to applicants' ethnicity and family income.

"Validity of High School Grades in Predicting Student Success Beyond the Freshman Year," by Saul Geiser and Maria Veronica Santelices from the Center for Studies in Higher Education at UC Berkeley, analyzes the correlations among high school grades from college prep courses, SAT scores, undergraduate grades and graduation rates, to reach three profound conclusions:

1) high school GPA is the best predictor of four year college outcomes for all academic disciplines, campuses and freshman cohorts in the UC systems;

2) the predictive strength of high school grades increases after the freshman year of college, and actually explains a greater proportion of four year undergraduate performance than of first year grades; and

3) when used in admissions, GPA has less adverse impact on low-income and underrepresented minority students than standardized tests.

In addition, Geiser and Santelices found that no combination of factors could account for more than 30 percent of the variance in college grades, suggesting that institutions that rely heavily on small difference in predicted outcomes to make admissions decisions are misguided.

These findings directly contradict claims of test proponents that "objective" scores from largely multiple-choice exams, such as the ACT and SAT, are needed to offset the variability among high school marking systems and the impact of grade inflation. In fact, GPA proved superior to the SAT at ultra-selective schools such as UC Berkeley and UCLA, as well as at less competitive campuses such as UC Riverside.

Based on these results, the authors call for deemphasizing standardized tests in college admissions, a recommendation that is certain to spur the test-optional movement. Once again, a careful review of the data demonstrates that test scores do not measure merit.