Institutionalized Voodoo: The Use of SAT Scores in College Admissions
by Dan Mess
In my research project, I will be looking at the biases inherent in the SAT, the history of its enshrinement as a valid - and often critical - criterion for college admissions and scholarships, and the ways that the SAT is used at Miami.
The SAT is one of the latest attempts for people to categorize others by intelligence, while assuming that intelligence is a monolithic entity that can be simply measured upon a linear scale. There has been a long history of attempting to use various results from this ranking of intelligence to “prove” that there is a hieararchy of ethnic groups in regards to intelligence, that men are more intelligent than women, and that the upper classes are innately, genetically superior to the working class.
This kind of thinking goes back to the days of craniometry in the 19th century, when researchers attempted to prove white superiority by asserting that intelligence could be measured by brainpan capacity and whites had bigger capacities than Native Americans and especially bigger capacities than people of African descent. It continued through the atempts of people to measure the size of the different protions of the brain and support claims of white superiority, and through the attempt of scientists to examine the bodies of criminals to prove that criminality was innate in people’s physiology.
Alfred Binet, a French scientist, was the first to pose the idea of testing people for mental capability instead of measuring various parts of their bodies. He developed a test in 1904 for the French minister of public education that was a diagnostic test so that teachers could identify students who might need remedial help early on. Binet himself was very careful to never state that there was an absolute scale of intelligence, or that his test was foolproof. Nor was the test meant to separate people into “smart” and “stupid” categories; it was simply a diagnostic tool for teachers to be able to help students early so that they could stay on track with their peers. His test did result in a single number, but the number was only to be used for the purpose of identifying which students might need some extra help.
Unfortunately, American psychologists got their hands on the test and strayed far, very far from Binet’s original intentions. They insisted on reifying the IQ number derived from the Binet test so that it was supposed to be a test of the true intelligence of all students. Binet’s vision of a test that would be used to help students who needed extra attention was perverted into a test that sorted out student into varying gradations of intelligence and did end up. The psychologists assumed that all of the varying mental activities of the brian could be measured by a single test with a single score, and they assumed that intelligence was inherited, not environmental or cultural.
R.M. Yerkes, an American psychologist, convinced the army to test 1.75 million enlisted men in order to “prove” that IQ was a legitmate measure of intelligence. Carl Campbell Brigham, another American psychologist, then went on to use those scores to “prove” the superiority of Western European ethnic groups over everyone else in a book entitled A study of American Intellgence. (This book was part of the evidence produced to usher through the Immigration Restriction Act of 1924, which “imposed harsh quotas against nations of inferior stock.” (quote from The Mismeasure of Man) He then went on to write the first SAT test. This is not a very good pedigree for the test, obviously.
SAT scores are currently used just as Binet’s IQ scores were: as a single linear scale used to rank people and find the “best and the brightest.” ETS (Educational Testing Services, Inc.), the corporation that administers the SAT, does caution people that SAT scores are not the end-all and be-all on a students academic aptitude and that many other factors need to be looked at in the college admission process. This still doesn’t keep SAT scores for being directly equated with academic fitness in popular culture (“Billy’s pretty dumb: he got under a thousand on the SAT!”) or from being used to make cutoff decisions for colleges or scholarship programs.
What are the main problems with the SAT? As I see it, there are three. The first is that SAT scores for 88% of incoming college first year students are as bad a predictor for first year grades as a roll of dice. As the students progress through school, the SAT becomes even worse at predicting their later grades. It’s outrageous that anything that obviously is such a flawed predictant of college success is used in admissions.
The second problem with the SAT is that it is a biased test. What are the biases of the SAT? They are fourfold: it is racist, classist, sexist, and ageist.
It’s no secret the minorities tend to do worse than whites on the SAT. Studies have shown that SAT scores consistantly underpredict minority students’ grades and overpredict white students’ grades.
There is a high correspondence between family income and SAT score, but a very low correspondence between family income and college grades.
Women on average do poorer on the SAT then men do, yet women have on average higher grades in both high school and college.
The longer that a person has been out of high school, the poorer he tends to do on the SAT. Yet studies have shown that reasoning and problem-solving skills increase throughout one’s thirties and forties.
The third problem with the SAT is that it doesn’t measure much of anything. It’s supposed to measure a student’s aptitude, her overall fitness for college. It is not meant to draw upon cultural or specific academic knowledge (beyond that of the most basic); it is only supposed to measure verbal and mathematic reasoning skills. ETS has claimed that the SAT is impervious to coaching except for some disclaimers that a review of basic math concepts would be helpful as well as some vocabulary-building skills and test-taking skills. (SATs now come with helpful hints such as, “If you don’t know an answer to a question, but can eliminate at least one possibility, guess.”)
However, the fact is that the SAT is eminently coachable, with many students exiting prep courses from Kaplan or The Princeton Review with hundred point gains on their scores or more. ETS took umbrage at the claims of these coaching courses. After all, if the test can be coached for, then it can’t be really measuring any innate aptitude for academics that’s been learned over years and years, can it? The ETS then had the FTC (Federal Trade Commission) investigate the claims of the coaching schools. Unfortunately for ETS, the coaching courses’ claims turned out to be accurate. Therefore, the SAT does not live up to its propaganda and does not measure pure academic aptitude.
With all of the above in mind, I’m going to interview Diane weber, a Senior Director of Admissions about Miami’s uses of the SAT in the admission process and our reasoning for doing so. (Our meeting is scheduled for March 27th.) I will also attempt to get statistics from a source in Roudebush that will allow me to calculate the level of correspondence between SAT score and grades at Miami, as well as the graduation rate. My working hypothesis is that the results for the statistics for Miami will follow the trend of the nationwide statistics and show the four biases mentioned above.
Return to the Topic Menu
IMPORTANT: For each Research Response, make sure the title of the response is different than previous titles shown above!