What's Holding American Students Back? The SAT

Conceived as a way to identify talent, standardized testing is leaving too many students behind. The U.S. needs new answers

What's Holding American Students Back? The SAT
Conceived as a way to identify talent, standardized testing is leaving too many students behind. The U.S. needs new answers (Photograph by Getty Images)
Photograph by Getty Images

University of Wyoming President Robert Sternberg was stupid in elementary school. IQ tests said so. Knowing his scores, his teachers in the 1950s expected him to perform badly, and he agreeably lived down to their expectations. In fourth grade a teacher named Virginia Alexa saw something special in him and conveyed her high expectations. Almost overnight he became an A student. He went on to earn a bachelor’s degree from Yale University and a doctorate in psychology from Stanford, and later served as president of the American Psychological Association. Not so stupid after all. “My entire future trajectory changed as a result of just one teacher,” Sternberg writes in a 2010 book, College Admissions for the 21st Century.

He worries about “stupid” students who don’t have a Virginia Alexa looking out for them. It’s not only IQ tests that defeat students, he says. It’s also the SAT and ACT, the college-admissions tests that he says are—contrary to their developers’ assertions—“basically IQ tests in disguise.” Sternberg says he thinks college applicants should also be asked to demonstrate their creativity, practical intelligence, and even wisdom, qualities which are in shorter supply than cleverness. “If you look at why this country is so screwed up,” he says, “it’s not because the people running it have low SATs.”

The U.S. rode to economic supremacy with the world’s highest share of young college grads, but now its percentage of graduates at the typical age of graduation is behind those of Australia, Denmark, Finland, Iceland, Ireland, Japan, the Netherlands, New Zealand, Norway, Poland, Portugal, the Slovak Republic, Sweden, and the U.K., the Organisation for Economic Co-operation and Development says.

Meanwhile, Americans who don’t go to college lack the skills they need for middle-class jobs as plumbers, welders, electricians, and health workers. “The skills gap in America has nearly reached a crisis point,” Jorge Ramirez, president of the Chicago Federation of Labor, told Bloomberg Businessweek earlier this year.

The SAT and its rival, the ACT, are part of the problem. Designed to ferret out hidden talent, the tests have become, for some students at least, barriers to higher education. Scores are highly correlated with family income; Harvard law professor Lani Guinier calls the SAT a “wealth test.” Type “SAT” into Amazon.com, and you’ll have to scroll past more than 200 test-prep volumes before you get to one book that’s a history or critique of the test. Because the SAT and ACT are now thought of as yardsticks of ability, students who do poorly on them are marked—or mark themselves—as failures. Overreliance on the SAT and ACT threatens to make America’s institutions of higher education even more elitist, adding to income inequality and harming U.S. competitiveness. The irony is that these were the very ills the tests were designed to combat.

Since the earliest days of the republic, there have been two schools of thought about the merits of sorting students, as recounted in Nicholas Lemann’s 1999 book The Big Test: The Secret History of the American Meritocracy. Thomas Jefferson, who believed in a “natural aristocracy,” said that in Virginia all white boys and girls should get a free public education from ages 6 to 8, after which “twenty of the best geniuses”—boys only—“will be raked from the rubbish annually and be instructed, at the public expense.”

New Englander Henry Adams was less disdainful of the rubbish. He said Jefferson’s natural aristocracy was no better than regular old aristocracy: “I would trust one as soon as the other with unlimited power.”

Jefferson’s side, the sorters, won. The SAT was launched in 1926 as a variant of an intelligence test used in World War I to place soldiers and sailors. Harvard adopted it in 1934. The University of California long resisted using standardized tests but in 1968—swamped by more qualified applications than it could handle—began requiring applicants to submit SAT scores as a way to screen out lower achievers. By this past academic year almost 1.7 million students took the SAT, and about 1.8 million took the faster-growing ACT.

Lately the influence of the tests has generated a backlash. Admissions officers at about 850 four-year colleges now make standardized tests optional for some or all of their applicants, according to FairTest, a nonprofit watchdog. A certain amount of self-interest is at work: If weak students don’t submit scores, then average reported scores go up and their schools ascend in the annual U.S. News college ranking. To be less cynical, the tests do stigmatize low scorers and distract people “from what they really need to do, which is mastering academic subjects in their high school,” says Wake Forest University sociologist Joseph Soares, whose school went SAT-optional in 2008.

Jay Rosner, executive director of the Princeton Review Foundation, once analyzed rarely disclosed “item-level” data from old SATs and found a troubling pattern. The College Board drops questions if they tend to be answered incorrectly by students who otherwise do well on the test—or if they tend to be answered correctly by students who otherwise do poorly. That seems like an admirable attempt to control quality, but it reinforces the status quo: Questions that white and Asian males don’t do particularly well on are systematically shorn from the tests.

The College Board—the nonprofit consortium of colleges, high schools, and other organizations that creates the SAT—has repeatedly jiggered the test to respond to critics, most obviously in 2005, when it added a writing section that boosted the highest possible score to 2400 from 1600. Huge disparities remain. Asians score the highest on the test, and their average rose this past academic year even as the scores of all other ethnic groups fell.

The College Board’s president, David Coleman, is a member of the educational elite with a strong do-gooder streak. As a student at Yale he started a program for students to tutor low-income pupils at New Haven’s Hillhouse High School. Coleman and his team are completing a major revision of the SAT to be unveiled in January 2014 and launched in the spring of 2015. He wants the test to “propel” students toward deeper learning of real things. The test will be based on what students study in school and not “shrouded in mystery,” he says. That means fewer abstruse vocabulary words (like “abstruse”) and essays that are based on documents so human graders can evaluate the correctness of their writers’ arguments, not just their style. “It is not different in a flashy, strobe-light way,” Coleman says. “I hope it will be greeted almost with a sense of relief.”

Coleman is taking a step in the right direction, but the SAT and ACT are still fundamentally about sorting by smarts. Imagine if hospitals evaluated incoming patients the way colleges evaluate applicants: Only the healthiest cases would be admitted. Thanks in part to the pernicious influence of published college rankings, schools have an incentive to entice more students to apply simply in order to reject them.

For the good of a country that’s losing its lead in the global race for knowledge, it would be more productive to expand opportunities for learning than to monkey with the tests that parcel out existing slots. Increased government funding of postsecondary education is one way to open the bottleneck and reduce the importance of standardized tests. Massive open online courses—MOOCs—are a more exciting answer. They’re cheap and highly democratic, and anyone can enroll at any time. A MOOC is all about the knowledge, not the credential. Which is the way it should be, right?

Sternberg, the formerly stupid first-grader, wound up running the University of Wyoming this fall after academic postings at Yale, Tufts, and Oklahoma State. At all three schools his research showed that measuring students’ creativity and practicality could predict their college success better than plain SAT scores could. The message: Real life is messy. You’re not given five answers to choose from. And America shouldn’t depend on something resembling an IQ test to rake geniuses from the rubbish.

Before it's here, it's on the Bloomberg Terminal. LEARN MORE