The Problem With the GRE

The exam “is a proxy for asking ‘Are you rich?’ ‘Are you white?’ ‘Are you male?’”

Team Static / fstop / Corbis

Updated on March 2, 2016

Early one Saturday morning last year, Christian Vazquez hopped on his bicycle and pedaled from his Highland Park home to the campus of Cal State, Los Angeles, one of many designated testing facilities for the Graduate Record Examination. The GRE, as it’s better known, is a test required for admission to what may amount to thousands of Master’s- and doctoral-degree programs—from astronomy and English to journalism and zoology—in the United States.

Vazquez, 24, who studied for the GRE over the course of about four months using a free study guide, was already an academic success story. Raised in east L.A. by a mother who worked full-time as a bank-loan processor, he was the first person in his family to attend college. During his undergraduate education, Vazquez lived at home and commuted via bike or bus to California State University Northridge four days a week, a two-hour trip each way, and paid his way through school by working as a cashier at a Kohl’s department store. Vazquez graduated in 2013 with a B.A. in English, earning mostly As. When he went job hunting, he had one thing in mind: “I knew I wanted a job where I made an impact on other people’s lives,” he says. He ended up with two part-time positions, as a teaching assistant at an elementary school helping mainly at-risk Latino children and as a tutor at Pasadena City College, where he also enrolled in literature and writing classes for fun. On his commutes to work, Vazquez dreamed of a grander future: becoming a college English professor.

Not able to afford the thousands of dollars often required for a prep course, he studied for the GRE on his own. But his test results weren’t what he hoped: “I didn’t feel like they were good enough to be sent in to any of the programs I was considering.”

Stories like Vazquez’s highlight the limitations of standardized admissions tests like the GRE—or at the high-school level, the SAT and ACT—and the obstacles they can pose to otherwise talented students, many of whom are disadvantaged minorities. Most graduate programs accept or require GREs as part of a prospective student’s admissions-application package and the stakes are well-known. Grad-school admissions is a competitive venture; researchers call grad school a “high-status opportunity.” For Fall 2014 admission, 2.15 million students applied to graduate school in the U.S. Only 22 percent of the doctoral applicants and about 48 percent of the Master’s-degree applicants were accepted, however. Score well on the GRE and your odds of entry to this high-status opportunity look better. Don’t score well, and—at least traditionally—you may have to consider something else to do with your life. Critics say an over-reliance on these tests leaves people like Vazquez—somebody who has demonstrated grit and passion for his field and a dedication to learning and teaching—out in the cold without just cause.

“In our society we put a huge premium on the kinds of analytical problems the GRE measures. So if you’re a good abstract analytical thinker, you’ll do well on these tests,” says Robert J. Sternberg, a cognitive psychologist and professor of human development at Cornell University. “The GRE is like taking a cancer test that was invented in the 1940s, though. Most of us wouldn’t have confidence in the results from a cancer test developed then. We have more knowledge and a far better understanding of intelligence and ability now.” Some students simply don’t achieve stellar GRE results although they may be intelligent and exceptionally capable, says Sternberg, who has studied both intelligence and college admissions for more than three decades.

In 2014, professors from Vanderbilt and the University of South Florida published a column in the science journal Nature denouncing the GRE as a test that fails because it takes a toll on student diversity—mainly the numbers of women, minorities, and economically disadvantaged students with high academic potential but relatively low GRE scores. Last December, the president of the American Astronomical Society posted an open letter asking chairs of university departments that grant degrees in the astronomical sciences to reconsider the use of the GRE.

One of the problems critics cite is that as a predictive test the GRE is something of a flop, only managing to (weakly) predict those who will do well in their first year of grad school. That’s it. GRE scores say little about whether a student has the perseverance, creativity, and intellect required to finish a graduate program or, more importantly, to add something to their professional world afterward.

Even the nonprofit creator of the GRE, Educational Testing Service (ETS), warns that there’s only a tenuous connection between test scores and success in graduate school. According to the ETS report, “Toward a Description of Successful Graduate Students,” “The limitations of graduate school admissions tests in the face of the complexity of the graduate education process have long been recognized.” The report acknowledges that critical skills associated with scholarly and professional competence aren’t measured by the GRE. Despite this report, David Payne, the vice president of ETS and chief operating officer for the Global Education division, told me in an email that he believes the GRE “measures the skills that are important for success at the graduate level: verbal reasoning, quantitative reasoning, critical thinking, and analytical writing.”

Perhaps more alarmingly, as with the SAT, high GRE test scores time and again tend to correlate with a student’s socioeconomic status, race, and gender. Research dating back decades from the University of Florida, Stanford, New York University, the University of Missouri, and  ETS has shown that the GRE underpredicts the success of minority students. The University of Missouri study looked at GRE scores and first-year graduate school GPAs of 160 minority students who earned graduate degrees from distinguished universities focused on research. The average first-year graduate-school GPA of these students was decent at 3.51, and they all ultimately finished their programs. But, according to the researchers, “in observing the range of GRE scores of these graduate students, it becomes clear that according to some graduate-school admissions policies, some students should not have been admitted to graduate study let alone actually graduate. This strongly suggests the necessity of focusing on other factors besides the usual criteria and to look beyond the talented top 20% of applicants.”

ETS studies have also concluded the GRE particularly underpredicts for women over 25, who represent more than half of female test-takers. Research from as far back as the 1960s leads experts to believe that the inconsistencies in GRE performance trace to a combination of factors including access to coaching, a disparity in educational opportunities that better prepare some students for the test, the content of the test, the way students are tested, and even the student’s own insecurities regarding race and gender. Sternberg puts it bluntly: “The GRE is a proxy for asking ‘Are you rich?’ ‘Are you white?’ ‘Are you male?’”

Payne argued that the score discrepancies are largely attributable to forces outside of the ETS's control and rejected the notion that they imply bias. "There are a number of factors that contribute to observed differences in scores, such as variation in course-taking patterns, interests, knowledge and skills, or differential educational, economic, and social systems in which everyone does not receive equal opportunity," he said, stressing that the tests' “fairness and validity” are one of ETS's top priorities. He cited a number of safeguards on which the ETS relies to ensure those goals are met, including a rigorous auditing process and a “meticulous system of checks and balances.”

Still, the differences in performance are what most trouble Julie R. Posselt, an assistant professor of higher education at the University of Michigan and the author of Inside Graduate Admissions: Merit, Diversity, and Faculty Gatekeeping. “The GRE adds information, but it is a noisy signal that says little about a student’s ability to be successful as a scholar. Yet in many programs it’s treated as a very significant piece of information. And unfortunately, requiring very high GRE scores for admissions undermines diversity,” says Posselt, who recently studied the admissions process at 10 top-ranked doctoral graduate programs.

In her research, Posselt found that mediocre GRE scores were often used as reason to immediately eliminate students in a highly competitive admissions process. And Posselt also found that more than half of the faculty members who sat on the admissions panels she studied erroneously equated GRE scores with a student’s native or raw intelligence. Google “how to raise GRE test scores,” however, and you’ll find the most common recommendation is to hire a private tutor or attend one of the many pricey GRE-prep classes offered by companies such as Princeton Review or Kaplan. It’s widely acknowledged that test-takers can be coached to do well on the GRE if they’re able to fork over the hours and thousands of dollars for prep classes or tutoring. For many students, though, the mere cost of taking the GRE—about $205, roughly 20 hours of minimum wage work in California—is prohibitive enough.

Recently, some programs have become “test-optional,” meaning students choose whether to submit their GRE scores. However, test-optional isn’t a quick fix for increasing diversity, warns Posselt. “When liberal-arts colleges go test-optional with the SAT, it doesn’t necessarily increase diversity the way you’d like it to. What it almost always does is increase the college’s image regarding selectivity because more people apply and those who do report scores tend to have higher ones.”

In an ideal world, Posselt would like a better test—say, something that consistently shows people who score well are more successful as graduate-school candidates. For now, though, she’d be thrilled if ETS simply provided better information to the people charged with making admissions decisions. “ETS score reports could provide the percentile ranking based on the test-taker’s national origin, field of study, and maybe parent education, race, and gender. That at least would allow faculty on the admissions panels to compare students with similar test-takers versus all test-takers.”

Too many graduate schools, says Posselt, are overly focused on snapping up already successful “winners”—falsely accepting high GRE scores as part of the evidence that these students are indeed winners—and avoiding the perceived risk of imperfect test-takers. But she warns against deciding not to apply to graduate school just because of your GRE scores, like Vazquez did.

“So much about [graduate] admissions is idiosyncratic,” says Posselt. “Depending on where you’re applying, who the other applicants are in a given year and who happens to be sitting on the admission committee, your odds of admission or rejection could be wildly different.”  Students shouldn’t view any one factor, says Posselt, as a certain deal breaker.

Robert Sternberg’s work on intelligence concludes that practical intelligence, the ability to experience setbacks and changes and still figure out how to have a fulfilling and purposeful life, is one of the key indicators of high overall intelligence.Vazquez has replaced his dream of grad school and professorship with something he feels even better about, and is now gearing up to take the CBEST, the standardized test required to become a public-school teacher in California. “I’ve decided I want to be an elementary-school teacher and be that person who inspires kids to go on, do something that matters in this world,” he says. In the meantime, he’s helping kids in El Sereno learn to read and write. “The kids I work with helped me change my perspective. Maybe I’ll try the GRE again sometime, but right now it’s not worth it.”