For years, the GRE has faced criticism over its role in the admission of graduate students. Various studies have suggested that departments rely too heavily on the GRE and as a result end up minimizing the chances that they will admit female, black and Latino applicants. And failing to admit more of such applicants may well doom efforts to diversify the faculties of many colleges.
Now, a new campaign is about to begin to encourage graduate departments to stop focusing as much as they have been on GRE scores. The campaign is going to be led by the Educational Testing Service, which produces the GRE, among other tests.
In an interview Friday, David G. Payne, vice president of ETS, said that the organization believes in the GRE as much as ever, and that graduate departments should require it. But he said that two recent developments have convinced him misuse of the GRE in admissions may be widespread, and failing to end that misuse could make it more likely that graduate programs eventually drop the GRE as an admissions requirement.
- The publication of Inside Graduate Admissions: Merit, Diversity and Faculty Gatekeeping (Harvard University Press). Julie R. Posselt, the author and an assistant professor of higher education at the University of Michigan, observed doctoral admissions committees of top programs in different disciplines, granting them anonymity if they would let her see everything. The book documents that many leading departments, despite saying otherwise, are reluctant to admit anyone who does not have extremely high GRE scores. Payne said he found the book persuasive and feared this was true not only for the top programs Posselt examined, but for many others as well.
- The American Astronomical Society adopted a policy to encourage departments to make the GRE optional or to stop using it. The move was notable in that astronomy is a science field, and many such disciplines have been particularly reliant on the GRE.
Payne said ETS would soon be offering graduate departments assistance in showing why they can admit applicants from a wider band of ETS scores than many currently do. Payne said that if departments provide data on those who are admitted with slightly lower GRE scores than is the norm for their program, he believed research would show that a broader range of GRE scores can lead to comparable levels of academic success in doctoral programs.
Likewise, he said ETS would start working with departments on some of the other measures they currently rely on, to see if validity research shows that they are as effective as graduate admissions committees believe.
These research projects, Payne said, could produce evidence that could lead to an increase in diversity in graduate programs. He noted that one of Posselt's key findings was that graduate admissions committees in elite programs focus on the prestige of applicants' undergraduate colleges and are likely to overlook talented prospects from less-well-known institutions. Payne said he thought validity research would show that success in graduate school does not in fact depend on having attended an elite undergraduate institution. By conducting such research and then by sharing it widely, Payne said, ETS may encourage graduate departments to cast wider nets for talent.
He also noted that the various biases Posselt uncovered (an obsession with both elite undergraduate colleges and a high narrow band of acceptable GRE scores) both hinder the enrollment of minority and female candidates. And he said he viewed the GRE as a tool to help those from nonelite undergraduate institutions demonstrate their potential.
Payne also acknowledged the challenges of bringing about change. Doctoral admissions committees are generally small groups of faculty members, and panel membership changes from year to year.
Posselt, in an interview, said that since her book was published, she has had a number of discussions with ETS, and she praised the organization for working on these issues. She said that she wasn't sure how quickly doctoral admissions committees would respond to research, but that it was important to try.
"Clearly there is a need to consider students out of the very narrow bands of scores that get privileged," she said. "Some of ETS's own research has found that extremely high scores don't always translate into high grades" in doctoral programs.
Shift in Astronomy
The shift by the astronomy association represents what ETS is hoping to avoid in other fields.
The statement urging departments to stop requiring the GRE noted that the norm in astronomy has been to require both the GRE and its physics subject test, the PGRE. Evidence "suggests that GRE and PGRE scores are poor predictors of success in graduate study in the astronomical sciences," the statement says, citing studies at Harvard University of graduate students in physics and of larger pools of students. "Research by the Education Testing Service … demonstrates that GRE scores correlate with demographic characteristics unrelated to potential for graduate study, such as gender, race and socioeconomic status. These correlations persist even in the GRE's recently revised general test."
Other studies have shown, the society statement says, "that misusing GRE scores, particularly by establishing score thresholds, fuels the underrepresentation of white women and minorities in graduate programs. ETS itself states, 'A cutoff score [on the GRE] should never be used as the only criterion for denial of admission or awarding of a fellowship.'"
Payne said he believes that departments would be limiting their knowledge by dropping the GRE or subject tests as requirements. But he said the astronomy statement raised legitimate questions about misuse of the GRE, which is part of why he wants to change the way departments use the test.
Posselt said that she believed that astronomy organization's move was significant. She said that, in graduate admissions, change is more likely to come in disciplines than through a particular university.
A Failed Effort at Broadening Admissions Criteria
ETS has tried various efforts in the past to try to promote the idea that test scores and grades aren't everything in admissions. Payne said Friday that one of those efforts -- the Personal Potential Index, or PPI -- would be phased out. The idea behind the index was to make faculty recommendations more comparable by asking questions designed to identify applicants' noncognitive skills.
Questions were designed to draw out faculty members on applicants' creativity, resilience, ethics and more. ETS had hoped that its system -- which in pilots seemed to identify many of those skills in a more diverse subset of graduate applicants than those who typically excel on the GRE -- would help colleges identify a diverse pool of talent.
Payne said that while results were encouraging, too few applicants and graduate programs used the PPI. As a result, when ETS planned large-scale validity studies, it didn't have enough results to be statistically valid, and decided to phase out the program.
The problem, he said, is one that ETS will face in its new campaign with graduate admissions committees as well. "Asking them to do something different is an uphill battle," he said.