The editorial "Testing the SAT -- for Fairness" {July 13} not only ignored the main reason behind the case against the SAT but presented arguments that fall flat in the face of statistical research.

The Educational Testing Service, maker of the SAT, states the sole purpose of the exam is to serve as a predictor of first-year college grades. Yet, females have higher grades than males in all subjects in high school and college, but averaged 61 points lower than males on the SAT last year.

Against the claims that the SAT is biased against women and ethnic minorities, The Post's editorial argued that scoring disparities between women and men, and between blacks and whites, are merely a reflection of the discrepancies in these individuals' educational backgrounds and socioeconomic status.

These factors, however, offer at best only a partial explanation. For example, when the performance of black men and women of equal income levels is compared, the average SAT verbal score differential is only reduced by half. Even ETS admits that it does not know why women score so much lower than men do.

With regard to course selection, ETS' own study concludes that on tests of mathematical skill, "controlling for years of study does not appear to explain completely the differ-ences." At a hearing held recently by the House subcommittee on civil and constitutional rights, the admissions director of MIT testified that its own research showed that women with lower SAT scores than men earned grade point averages equal to those of these men. The MIT study involved identical course work, thus dispelling the argument that women take easier classes than men do and therefore get better grades.

This is why consumer groups, state legislatures and Congress are looking into these exams and questioning their use. One has to question why we rely so heavily on these tests when admissions officers and statisticians agree that high school grades, despite their simplicity, are the best predictors of college performance. More important, behind this discussion lies an even larger one over the use of standardized tests in tracking children at a young age.

Women and minorities may be getting an unfair shake at the college level, but even more insidious is the use of such tests as a measure of our children's abilities and as the deciding factor in their educational fate.


U.S. Representative (D-Calif.)

Chairman, Subcommittee on

Civil and Constitutional Rights