Downgrading SATs Makes Sense

Many conservatives are groaning over a major new report from a commission of higher education luminaries calling on colleges to de-emphasize the SAT for college admissions.

The catcalls from the right erupted after the National Association of College Admission Counseling suggested that colleges should rethink their reliance on the SAT for admissions. Wrongheaded, de-evolutionary, politically correct in the extreme, and void of common sense, the critics said the NACAC report is a frontal attack on academic standards and will lead to the ruin of American higher education.

We’ve heard the dire warnings before, countless times. And countless times the cries that the sky is falling have been wrong.

The defense of the SAT as the linchpin of the college admissions process contains at least two major propositions, both of questionable merit.

First, there’s the “necessary evil” argument. The SAT is the only objective, common yardstick colleges have to make fair-minded comparisons of students’ academic qualifications for admission to college. Grade inflation, declining standards in American schools, the wide diversity in academic rigor among high schools, etc. , all make this “common yardstick” even more necessary now than ever. The SAT may not be perfect, but it’s the best tool colleges have to make reasoned and fair assessments of a student’s fitness for college-level work.

That’s the meritocratic argument for the SAT, and it’s simply not supported by either the scientific evidence nor by the actual experience of many colleges and universities that have de-emphasized the SAT in recent years.

We’ve known for many years that the SAT is a relatively weak predictor of academic success in college, adding little value to what admissions offices can glean from high school grades, class rank, writing samples and other assessment tools that predict performance in the real world, including the classroom.

All those alternatives to the SAT — the supposedly mushy and unscientific measures of student achievement that — in fact are more accurate assessments of skills and achievements than any multiple choice test could ever be. Time and again, the various defenses of the SAT over years have failed to adequately address the vast social scientific literature that decimates the case for the SAT.

Probably the single important study in recent years affirming this conclusion was published in 2001 by the University of California, when UC researchers Saul Geiser and Roger Studley examined some 80,000 student records between 1996 and 1999. Geiser and Studley found that a student’s high school grade point average was the single best predictor of one’s grade point average at the university, followed by the SAT II subject tests. The SAT I, the so-called reasoning test of verbal and math ability, came in a distant third. Indeed, while the SAT II and high school GPA combined accounted for about 22 percent of the variance in students’ college grades, adding the SAT I to those assessments improved the prediction by a meager 0.1 percent.

The SAT defense also likes to trot out the medical analogies: ignoring the SAT evidence on student ability or college quality would be like physicians ignoring medical evidence from tests like colonoscopies or mammograms.

On the surface, the analogy to medical care that the SAT adherents like to invoke seems irrefutable. More information is better than less information, in education or medicine, right? In fact, medicine is chock full of examples in which certain medical tests have proven to be counterproductive and costly exercises that don’t lead to better health.

Physicians don’t perform mammograms on 20-year-old women nor colonoscopies on 30-year-old men, because the information gleaned from such testing isn’t useful. More and more patients are now demanding body CT scans for early detection of cancer. The tests are costly, and indicate little more than the normal wear and tear on the body. Patients get more information, but it’s of marginal utility that doesn’t save lives.

The medical analogy to the SAT, then, is deeply flawed. Just because we have the technology to gather more information about patients or students does not mean that the tests are worth doing.

And then, there’s the “backdoor affirmative action” claim. According to this reasoning, politically correct educators, hoping to create more racial and ethnic diversity, want to get around laws and policies against racial preferences by lowering testing standards and admitting more unqualified minorities.

The argument is spurious. First, the presumption that “unqualified” students are admitted because testing standards are lowered is not consistent with the experience of the many colleges and universities that have reduced the importance of the SAT, including Bates College, the University of Texas at Austin, Oregon State University, and scores of others.

Second, it defies the economic logic of the higher education industry to claim that colleges want to quit the SAT in order to commit “backdoor” affirmative action.

Despite rhetoric suggesting that they are democratic enterprises extolling the public good and rooted in egalitarian values, American colleges and universities are businesses. They don’t maximize profits or supply 10K’s to the Securities and Exchange Commission, but they are in the business of maximizing institutional prestige and endowments. College “quality” in the higher education market is determined by an institution’s selectivity in admissions, which in turn is determined by the median SAT score of entering freshmen.

American higher education’s growing reliance on SAT scores has never been about upholding the principles of meritocracy. Rather, an institution’s median freshman SAT score serves as an important market signal that establishes and reinforces the institution’s brand identity as an elite institution for the select few or a mediocre institution that serves ordinary students. Selective colleges compete fiercely for prestige in this market, and the median SAT statistic is the most important tool for establishing that most intangible of qualities.

The industry’s prevailing measure of quality, dubious as it is, creates various subsidiary markets in higher education, chief of which is the market for prestige. Competing for prestige, colleges employ an array of sophisticated marketing tools to identify, recruit and enroll students with the highest SAT scores possible, and shun those with scores that fail to meet institutional objectives. Those objective are driven by the institution’s need to satisfy its various constituencies, including alumni, parents, and big financial donors. It’s the rare selective college or university that sees its institutional interest in placing less emphasis on the SAT, the very tool that ensures its place in the higher education pecking order.

The college rankings game, orchestrated by U.S. News & World Report, entrenches the dominance of SAT scores as a measure of a college’s quality. It’s a phony measure, at best. Instead of measuring what colleges actually do for students once they arrive on campus — the equivalent of the actual performance criteria that most American business enterprises are judged by — the current college rankings paradigm says, in essence, that you’re a good college because you’ve admitted students with high SAT scores and that you’ve turned away those with modest SAT scores. That’s like saying Sony is a good electronics company, not because of its production quality, but because the company purchased the best copper wires from its raw materials suppliers.

The “backdoor” affirmative action claim is dubious on another count as well. As endowment-maximizing enterprises, colleges and universities don’t really care about affirmative action as a matter of morals, justice or ethics. Rather, affirmative action is politically useful. Colleges have embraced race-based affirmative action because it has allowed them to maintain admissions and financial aid policies that largely serve the mostly white and affluent demographic groups that help them compete for prestige and grow institutional endowments. Affirmative action allows universities to espouse the values of “diversity” while actually engaging in a program of self-interested elitism.

No conspiracy theories are needed to account for these outcomes. The economic logic of the higher education marketplace plus the historical accidents that over the years positioned the SAT at the center of selective college admissions fully account for the admissions and financial aid trends in higher education that have greatly benefited the most affluent families. Colleges and universities are not nefarious organizations; they are self-interested institutions trying to survive in an intensely competitive market with other self-interested institutions.

So what does the NACAC commission’s report really mean? The educators on the widely representative commission who produced this report include William Fitzsimmons, the dean of admission and financial aid at Harvard; Nicholas Lemann, the dean of Journalism at Columbia; and a dozen other educators from public and private schools, colleges and universities across the country.

I take them at their word when they say their goal is to “stimulate a rich and sustained process of introspection” about how colleges and universities pick and choose their students. The commission’s goal is to “assume control of the conversation about the use of standardized tests in admission.”

What does that mean? It doesn’t mean sacrificing standards on the altar of diversity. It doesn’t mean that meritocracy is dead. Although they don’t put it in these terms, the commission members have, in essence, called for the restructuring of the higher education market so that the SAT would no longer presumptively occupy the position of cultural icon for college quality and individual accomplishment.

That’s a position no test ever deserved, and the time is right to bury that notion for good.


Another view of the SATs


One thought on “Downgrading SATs Makes Sense”

  1. As an employer of undergraduates and graduates of professional schools, and as a teacher of the latter, I can identify a critical flaw in your analysis. That flaw is this: Administrators use college grades as a measure of “success in the classroom” and then find little correlation of this “success” with SAT scores. This is so because, with the exception of objectively graded programs such as engineering, college grades are offer little as a true measure of merit. The grading scales have become so inflated and compressed that it is difficult to separate the morons from the achievers. Grades are handed out by faculty who either don’t believe in grades at all or are fatigued by the complaints from students and others as to their supposed irrelevance. To me as an employer, I want to see results measured on objective tests. If you haven’t studied and been graded in college math, physics, or engineering, show me the SAT. That “gender studies” grade is meaningless.
    By the way, you dispense of the “backdoor affirmative action” issue much too quickly. Perhaps you forget that the SAT itself was revamped last decade (by adding a writing component) for the singular purpose of increasing scores among female test takers. Since that did not have the effect of boosting scores by other politically targeted groups, the next logical step was to discount the value of the test altogether. It seems you have joined that club quite easily.

Leave a Reply

Your email address will not be published. Required fields are marked *