SAT--A Failing Test
Educators have long known the rap sheet on the SAT, the college entrance exam that millions of young people have taken as a rite of passage for some seventy-five years. Since its inception, the SAT has become among the most scrutinized and controversial of standardized tests. And yet, the exam--and the mental testing culture that has sustained it in the United States--has been remarkably impervious to the attacks on it over the years.
Recently, however, the SAT suffered a body blow when the president of the University of California system proposed dumping the exam. Don't expect colleges and universities to defect from the SAT en masse--it's too deeply entrenched for that. But in announcing his far-reaching proposal in February, UC president Richard Atkinson legitimized open discussion of a heretofore taboo subject for large and selective universities: whether they (and society) would be better off without the test.
Atkinson, an eminent cognitive psychologist, knows well the list of particulars against the exam in question, the so-called SAT I "reasoning test." As the progeny of the first intelligence test commercialized in the United States, the SAT has proven to be a weak predictor of a student's actual performance in the first year of college; after that, its usefulness vanishes completely. Moreover, the SAT has proven to be a vicious sorter of young people by class and race, and even gender--and has served to sustain the very upper-middle-class privilege that many of the exam's supporters claim to oppose. The latest figures from the College Board, the SAT's sponsor, show that a test taker can expect an extra shot of fifteen to fifty points on his or her total SAT I score for every $10,000 that Mom and Dad bring home. Call it the Volvo Effect: a boost that peaks out at the highest levels of family income. Being white, on average, confers an extra 200-point advantage over a black test-taker. Atkinson hopes that replacing the SAT I with the SAT II subject tests will lessen such disparities and more accurately reflect what students study in high school. In fact, scores on both exams are powerfully correlated with each other, and UC's own data show that the SAT II also sorts harshly by class, race and gender. More helpful, Atkinson intends to revamp the entire UC admissions process by requiring campuses to evaluate applicants more comprehensively than under the old numerical formulas, judging a high school student's achievements in light of his or her social and economic circumstances.
The SAT's shortcomings have become especially vivid in recent years, as courts, voters and policy-makers in several states, including the UC Board of Regents in 1995, have ordered public universities to dismantle their affirmative action programs. Post-affirmative action, UC's most selective campuses have seen freshman acceptance rates wane for blacks and Hispanics. Meanwhile, the state's Hispanic population is forecast to skyrocket from about 11 million in 2000 to 18 million over the next two decades. Hispanic high school graduates will surge 74 percent over the next decade, while numbers of white graduates are expected to grow just 2 percent.
In light of these trends, the usual justifications for the SAT's continued dominance as a gatekeeper to UC would no longer wash. Yes, since 1968 the admissions test has been a bureaucratically convenient way to sort and weed large numbers of college aspirants. Yes, UC's relatively high SAT scores made it look good in the test-score fashion show put out by US News & World Report. Yes, the test was a common yardstick. But it was also a crooked one, inflicting enormous social costs.
Of course, there will be complaints that Atkinson's tossing the SAT will lead to the ruination of a great university: As UC opens the floodgates to hordes of the academically unfit, standards will plummet. We've heard it before, as when the University of Texas system enacted its "top 10 percent" law after the 1996 federal appeals court ruling in the Hopwood case, which ordered the state's universities to end their affirmative action programs. Beginning in 1997, any Texas high school senior graduating in the top 10 percent of her class earned automatic admission to Texas public universities--regardless of SAT scores. Did this produce the collapse of a great university? Hardly. At the flagship University of Texas, at Austin, SAT scores of students admitted under the top 10 percent law, as expected, fell markedly compared with their peers from pre-Hopwood days. And yet, their classroom performance actually bettered their pre-Hopwood counterparts (that is, those in the top 10 percent who did meet the SAT threshold), holding steady even in engineering, business and science. To top it off, by 2000, enrollments of Hispanics and African-Americans had been restored to their pre-Hopwood levels.
Ultimately, UC's faculty senate and the Regents could dash Atkinson's hopes for a new era in the university's approach to college admissions. Nevertheless, he has accomplished something of unquantifiable benefit by helping to pry open a badly needed debate about the meaning of merit in American higher education. Will we be a nation that judges young people based on what they have accomplished and what they've overcome to do so, or by how well they fill in bubbles on a standardized test that is itself of questionable merit?