Study Calls SAT ‘Fraud,’ Biased Against Women

The Scholastic Aptitude Test, a key tool used for college admissions, discriminates against women by including questions that favor men and is a “consumer fraud,” according to a report released Thursday.

The 121-page report, published by the Center for Women Policy Studies, a women’s advocacy group, said that the test does not accurately predict the college performance of 780,000 women each year. These women have diminished chances of entering the 1,500 colleges that rely on the SAT and decreased access to scholarship money, the report said.

The report, “The SAT Gender Gap: Identifying the Causes,” is based on a federally funded two-year analysis of more than 100,000 students who took the test in June, 1986, and November, 1987.

The SAT’s main purpose is to predict the first-year college grades of those taking the test. In 1988, average female scores were 13 points below males on the verbal section and 43 points lower on the math section, the study said, even though women’s grades are higher than men’s in college classes. Fifty-two percent of the 1.5 million students taking the test each year are female.

“A test that under-predicts the performance of more than half the people who take it should be considered a consumer fraud,” said Phyllis Rosser, an independent researcher who conducted the study. Rosser is director of the Equality in Testing Project, a private firm that studies the causes and impact of sex bias in standardized testing. She is also a senior research associate at the Center for Women Policy Studies.

The study found that 23 questions of the 145 on the test in November, 1987, were discriminatory because they contained a large sex difference. Men outscored women on 21 of those questions and women did better on only two of the questions. Also, the June, 1986, test contained 17 questions with a sex difference, with men doing better on 13 of them. The study considered a question discriminatory if it had more than a 10% difference in correct answers between men and women.

The study also found that women scored lower than men from similar backgrounds, including race and income.

“These questions are tipping the balance,” Rosser said. “As a result, four years of high school achievement go right down the drain. They’re wiped out by a three-hour test.”

The Educational Testing Service, which publishes the SAT, defended the validity of the test. “Virtually all the different scores are not due to bias in the test, but due to real educational differences in those taking the test,” said Nancy Burton, director of the ETS admissions testing program.

However, ETS has been adjusting the SAT to throw out questions that are either too easy or too difficult for any group, including women and minorities, Burton said.