Ten Myths About the SAT

2. The SAT is only the messenger: score gaps merely reflect differences in students' academic backgrounds.

Even when students are matched for academic preparation, there are still large gender and racial gaps in their SAT scores. A federal court reviewed the test-makers' best arguments, which took into account variables such as ethnicity, parental education, high school classes, and proposed college major, and concluded that "...[U]nder the most conservative studies presented in evidence, even after removing the effect of these factors, at least a 30 point combined differential [between males and females] remains unexplained."

3. SAT scores are precise, like inches on a yardstick.

The College Board's ATP Guide long stated, "Users learn to understand and appreciate the meaning of a score of 430 in the same way that they have learned to understand and appreciate the meaning of, say, 14 inches..." But it notes in other publications that the SAT's margin of error is approximately 60 points. Even more incredible, the test-makers admit that two students' scores must differ by about 120 points before anyone can be sure that the differences are meaningful. Adding to the confusion, ETS "recentered" test scores in 1995. This changed the formula used to convert raw scores into the SAT's 200-800 point scale, resulting in an average score increase of nearly 100 points. If the SAT is a "common yardstick," as the test-makers say, it must be made of elastic.

4. At least the SAT is more accurate than high school grades.

On the contrary, despite all the differences between courses and grading standards, high school grade point average (GPA) is still the best predictor of first year college grades -- which is all the SAT claims to predict. As a student moves through college, SAT scores become even less accurate predictors, with high school GPA and rigor of courses trumping the SAT in forecasting bachelor's degree attainment. This shows just how inaccurate the SAT really is.

5. But colleges still need test scores to make admissions decisions.

The nearly 400 colleges and universities that admit a substantial number of applicants without regard to test scores show that you can have a rigorous admissions process without the SAT. Highly selective institutions can follow the example of Bowdoin College, which has found that the diversity and quality of its students improved after it made the SAT optional 25 years ago. Fewer than 150 colleges in the country reject more than half of their applicants. Admissions officers at these schools have many other ways to deal with differences in high school curriculum and quality.

6. When girls and boys are matched by ability, the score gap on the SAT-Verbal section disappears.

The College Board's way of "matching by ability" is to match test-takers' SAT scores! So if the test is flawed as an overwhelming body of research and legal cases indicates, then matching by scores is also skewed. The real cause of the gender gap is bias in the test: despite the fact that they receive better grades in high school and college in comparable classes, females receive lower scores on the SAT. This remains true even when course taking patterns and course difficulty are accounted for. The test-maker has been unable to adequately account for this discrepancy, but independent research shows that the timed, multiple-choice format of the exams, the roles of females in test questions, the penalty for guessing, and "stereotype bias" may all play roles in artificially depressing females' test scores.

7. Test coaching doesn't work.

Studies collected by FairTest show that good coaching programs can raise a student's scores by 100 points or more. Many of these courses are very expensive ($800 and up), and teach little more than test-taking strategies specific to the SAT. The fact that short-term coaching works undermines the test-makers' claim that the SAT measures skills and knowledge learned over a long period of time. It also adds another income-related bias to the test, since students who come from families that can afford an expensive coaching class are already more likely to score higher on the test. Moreover, why do both ETS and the College Board sell test preparation products if coaching doesn't work? 

8. The test-makers' "exhaustive bias reduction procedures" guarantee that the test is fair.

Whatever else their procedures may be doing, they are not eliminating bias from the test. ETS' own research indicates that the overall format of the test itself is to blame. The SAT is a fast-paced, multiple-choice test which rewards strategic guessing -- a non-academic skill at which males tend to excel. The methods used to screen individual test items for bias, such as "differential item functioning," fail to eliminate large discrepancies between students of different racial groups. One analysis of the October 1998 SAT showed that out of 78 Verbal and 60 Math questions, there were no items on which African Americans or Chicanos outperformed Whites.

9. The SAT is needed to counteract grade inflation.

No definitive proof exists showing that grade inflation is running rampant throughout U.S. high school, as many in the testing industry claim. Even if one assumes that grades are going up at a much faster rate than test scores, this trend would be a rising tide that lifts all boats. Applicants can still be accurately compared with one another since everyone's grades would be increasing and rank-in-class does not change.

10. The SAT measures what you need to know in college.

The SAT is a mind game that has nothing to do with skills necessary for higher education: it tests a tiny range of techniques, mainly how quickly you can choose among four or five answers without thinking deeply about any of them. For example, research shows that over 40% of reading comprehension items can be answered correctly without reading the passage. Some of the many qualities you need in college that the SAT cannot measure are writing ability, strategic reasoning, higher order thinking skills, experience, persistence and creativity.