There are two major categories of tests: norm-referenced tests and criterion-referenced tests. The SAT I is a norm-referenced test. According to Bond (1996), "The major reason for using a norm-referenced tests (NRT) is to classify students. . . . Although no test can measure everything of importance, the content selected for the CRT is selected on the basis of its significance in the curriculum while that of the NRT is chosen by how well it discriminates among students." (Bond, L. A. (1996). Norm- and criterion-referenced testing. ERIC/AE Digest.) On a norm-referenced test, scores are determined by referencing the raw scores to a norming group (a relative interpretation of the raw score), while criterion-referenced tests are supposed to measure what students can or cannot do with particular subject matter (an absolute interpretation of the raw score).

The SAT is constructed the way that it is and timed because, as a norm-referenced test, it must discriminate among students; some students must do well, some must be average, and some have to do poorly relative to the rest of the test takers. Putting a time limit on the SAT helps assure that some test takers will do better and some will do worse. Timing the test also assumes that in "verbal and mathematical reasoning," working faster equals working better, something that is often not true in high school, college, or the world after college. (For a good discussion of this, see Ungerleider, D. & Maslow, P. (2001). Association of Educational Therapists: Position paper on the SAT. Journal of Learning Disabilities, 34(4), 311-314.) The fact that qualified students can take the test with extended time would seem to indicate that some students' reasoning skills are just as good as others with similar scores, even though they aren't as fast.

Does extra time on the SAT help?

The difference between a 720 and an 800 Verbal score can be as few as five raw score points (based on the actual number of right, wrong, and unanswered items) on the 78-item Verbal section. And, the difference of between a combined score of 1430 and a combined score of 1600 can be as few as 10 raw score points on the full 138-item test.

Extra time on a speeded test will help most, but not all students. The ones who will be helped most will be those who are coached on test-specific techniques and testwiseness skills. They will be able to use the pacing strategies, guessing strategies, etc. that are taught in coaching courses and get extra points, and avoid mistakes that will lose points. When the writing sample/essay is added to the SAT I, there is no question that students with extra time will be advantaged.

Test prep experts (as opposed to the testing specialists referenced above) will tell you that most of the students who are scoring below 1000, and particularly those below 800 combined, unless they have a legitimately diagnosed learning disability that necessitates extra time on speeded tests, may not benefit significantly from extra time. However, the opposite is true for many students scoring above 1000, and it is even truer for those above the 1200 to 1300 range. For these students, extra time will often mean a higher score, and often, but not always, a significantly higher score. And, well-coached students across the score spectrum will have more time to practice what they have learned in the coaching courses. (Why believe the test prep experts and not testing specialists? As David Owen wrote in “None of the above,” if you want to know how a safe works, ask the safe cracker, not the manufacturer.)

Giving students who need it extra time is fair and necessary. And removing the asterisk is fair and necessary. But we are fooling ourselves, or letting the powers that be fool us, if we believe that this system will not be abused by some, who are already advantaged, to the detriment of those who are already disadvantaged. Abuse of extra time, even with the asterisk/flagging system in place, has been demonstrated (Weiss, K.R. (2000). Audit confirms disparities in SAT testing: White, affluent students are granted disproportionate share of time extensions based on learning disabilities, state report finds. Los Angeles Times, December 1, A-3). The SAT is not, and will continue to not be, a true “standardized” test.

The only fair solution to the SAT timing problem is to give students the time that they need to show what they know, not how fast they can spew it out. (See New York Times, July 18, 2002, “Testing for Aptitude, Not for Speed” by Howard Gardner.)

Of course, no one the high school side will look forward to spending even more of our Saturdays at our schools (i.e., test centers) or trying to find proctors willing to do so.

Bradford R MacGowan, Ed.D

09/02