The SAT Score Debate
by Daniel Luzer
Room for Debate, the New York Times section in which different people discuss issues of national importance, this week tackles the SAT. “Critics claim the exams can be gamed. If that’s the case, why are they still so important in college admissions?” is the prompting question. But this might be the wrong concern.
Stop worrying about the gaming, writes Paul Siemens of Advantage Testing of Los Angeles. The examination is “not an I.Q. test”:
Both the SAT and ACT assess knowledge of vocabulary, grammar and math, while exercising a student’s reading and problem-solving abilities. Such skills are not merely helpful in high school academics but critical for success in college and in life. There are no “tricks” or “shortcuts” a tutor can impart that will circumvent the difficulty of the questions. The only way to prepare successfully for these tests is to apply the hard work that is at root in any successful academic endeavor.
University-level entrance exams are not intended as intelligence tests, and for good reason: to the extent it can be measured, intelligence is a limited predictor of academic success at that level. Colleges aren’t simply looking to enroll the smartest students; rather they are seeking mature, talented, well-rounded, motivated, service-oriented and accomplished students of every background. Standardized tests provide a reasonably reliable barometer of the extent to which a student has been able to master the general high school curriculum in a way that will prepare her well for college.
He’s right that there are, at least on some level, important things the SAT can tell admissions officers about the potential for a high school student to perform effectively in college. He’s also essentially right about the gaming. Students who aren’t prepared for the intellectual rigors of higher education can’t “game” the SATs to get scores that indicate that they are. While there’s some preparation and skills someone can learn to do better on the examinations, there’s only so much that test prep can do.
But he’s just wrong when he asserts that the SATs aren’t an I.Q. test. They are, and we’ve known this for a while. Meredith Frey and Douglas Detterman demonstrated back in 2003, that “SAT is mainly a test of g [general intelligence].” The authors also provided “equations for converting SAT scores to estimated IQs,” because those examinations measure essentially the same things.
We shouldn’t be surprised by that. Granted, the SAT does not, as originally intended, actually predict success in college. But it does measure intelligence, and that certainly matters for success in a academic environment.
The real problem with the SAT, in fact, appears to be something very different. Part of the reason there’s so much worry about students “gaming” the test is that while the SAT measures intelligence, a difference of 100 points or so on the exam doesn’t really indicate much about cognitive ability.
But those sorts of differences matter a great deal in terms of college admissions, and that 100 points or so is precisely what students can fix by prepping for the SATs.