To Diversify STEM Graduate Programs, Eliminate the GRE

By: Aníbal Tornés Blanco, Biology PhD student at the University of Michigan 

If you have ever applied to a graduate program, you very likely had to take the standardized Graduate Record Examination (GRE). While historically this exam has been a central part of the application package for most masters and PhD programs, in recent years its utility as an objective predictor of graduate student performance has been thrown into question. A growing body of research has emerged showing “a great divide” in GRE scores. In 2014, a study published in Nature described how women score on average 80 points lower than men in the physical sciences and how underrepresented minorities (URM) such as African Americans score 200 points lower than white students. Moreover, the exam is prohibitively expensive for many URM students, further impeding efforts to get into graduate programs. Instead of predicting success in graduate school, the exam serves primarily as a mechanism to filter URM students out of graduate school applicant pools. 

I have experienced this injustice directly: coming from a low-income background and with hopes to pursue my academic goals, it took around 65 hours of minimum wage work in Puerto Rico for me to pay the $205 GRE test fee and for score reports ($27 each). I sold popsicles by day while juggling my academic responsibilities by night in hopes of getting into my dream school. After paying $500 to the ETS, I now had the “opportunity” to apply to graduate school–though of course, I had no guarantee of getting accepted. My experience is not unique: this financial hurdle presents a serious impediment to all low-income and working-class students applying to graduate school. 

The COVID-19 pandemic has further exaggerated these challenges, and added new ones. Eight million people have fallen into poverty since March, and even more have been laid off or have seen their work significantly reduced. Now, a $500 investment in an exam is even more unrealistic for many students. And the test is now exclusively offered remotely, meaning that test takers now must have access to a stable internet connection and an operating system that meets certain requirements, both of which may present significant hurdles for students from low-income backgrounds. Under pandemic conditions, admissions committees must now take into account that even test-taking conditions are not equitable across the applicant pool. 

Recently, however, graduate programs–most noticeably those in STEM–have begun to recognize that these standardized tests are weak predictors of success in graduate school, and many have dropped the GRE as an admissions requirement. Dr. Joshua Hall from UNC-Chapel Hill has been tracking this trend, curating a public spreadsheet (BioGRE.info) that lists all of the biological and biomedical programs that have dropped the test. In 2019, he published these findings in an article in Science, demonstrating that at that point, 74 biological and biomedical graduate programs had dropped it as a requirement. At the time of writing, over 380 STEM programs have followed suit. Today, there are over 380 graduate STEM programs that have completely dropped the GRE requirement. Furthermore, both the National Institute of Health (NIH) and the National Science Foundation (NSF) have also dropped the GRE requirement for Graduate Research Fellowship Program (GRFP) grants, and the ETS has dropped the discipline-specific GRE subject tests entirely because they cannot be offered remotely.

The decision that many institutions have made to eliminate the exam will force admission committees to look at applicants in a more holistic manner. However, dropping the exam on a somewhat ad-hoc basis is not sufficient–universities need to step back and systematically reexamine its usefulness.

Luckily, many graduate admission committees from top-ranked institutions have begun doing exactly this. These programs have adopted one or more of the following three techniques to initiate conversations within their own faculty members and raise awareness of how poor a predictor the GRE is in the context of biomedical research:

1.     Have the graduate admission committee members take a GRE practice test. The University of Minnesota director of the Biomedical Graduate Research, Education, and Training, Dr. Jon Gottesman told me that prior to taking this practice test, many of the faculty had never seen the GRE. This strategy yielded important results: faculty realized that the quantitative section questions were not correlated with the math required for graduate studies in biology. These conclusions were shared by the president of the American Astronomical Society who posted an open letter in 2015 asking chairs in astronomical sciences graduate programs to rethink the use of the GRE General Test.
2.     Have the graduate admission committee members make decisions without looking at the GRE. The Northeast Alliance for Graduate Education and the Professoriate (NEAGEP) has suggested that programs trying to remove the GRE from their admissions process could conduct a preliminary study where faculty rank prospective candidates while blinded from GRE scores. After gaining access to these scores, the faculty members could determine whether or not the student rank would change. This model was built in 2017 by 60 faculty and administrators from 26 institutions around the country that got together in UMass Amherst to discuss alternatives for GRE scores in admission processes.
3.     Have the graduate admission committee members write a report outlining the importance they attribute to the GRE when considering prospective students. In 2017, the Program in Biomedical Sciences (PIBS) at the University of Michigan and its director, Dr. Scott Barolo, obtained feedback from faculty members. The PIBS voluntary faculty members wrote a report with the strongest arguments for and against the GRE requirement. Both the “pro” and “con” positions were then shared in a town hall which concluded that if faculty cannot agree on a test’s usefulness and predictive power, requiring students to invest time and money in the test is unethical.

While these are important steps, universities must go further: higher education institutions must make coordinated, unilateral decisions to eliminate the exam entirely, as it merely measures sex and skin color rather than success in graduate school. If graduate schools are serious about their purported commitment to increasing representation in STEM, removing these arbitrary hurdles that disproportionately affect URMs must be a top priority. The GRE has made the glass ceiling metaphor feel very real today, especially for URMs, and it must be shattered if we truly want to make science for everyone. This historic admission year requires historic measures to be taken in order to increase diversity and representation in science.

 

Aníbal is a 1st year PhD Biological Chemistry student who was born in La Habana, Cuba and escaped shortly after to Puerto Rico where he pursued his B.S. in Integrative Biology at the University of Puerto Rico – Rio Piedras. His early work with a non-profit science museum in Puerto Rico inspired him to become an educator for his community and to play an active role in the Científico Latino team to streamline the PhD pathway for diverse generations of scientists to come.