Director’s Corner

Theodore Hodapp

GRE – What can it tell you?

A substantial portion of the physics community relies on the Physics Graduate Record Exam (P-GRE) to provide an “unbiased” assessment of a student’s preparation for completing a doctoral degree in physics. Unfortunately, the P-GRE is neither unbiased, nor an accurate assessment of a student’s chances of completing a degree.¹

Consider the student who grew up in an inner-city neighborhood, where the high school math teacher had to teach to the lowest common denominator, and where physics wasn't offered because the principal didn’t think his students could “handle it.”² This same student attends a small, public, underfunded 4-year college, where the physics program graduates few students, and many of the upper-division courses are either not offered, or given as “reading courses,” where the student has to essentially learn the topic on his or her own. How well will this student fare on the P-GRE? Will the exam be a good indicator of his or her potential to complete a physics PhD? Should we exclude this student from getting a PhD?

What about students who were not told or did not understand how the test would be used to sort their graduate school application? Or, as we have heard from students in the APS Bridge Program, they took the test, and based on their score, decided that physics graduate school was not for them. In cases like these, we are missing the opportunity to provide equitable opportunities for students who have the drive, but perhaps lack some preparation, to join us as our colleagues. If we approach graduate admissions with the thought that we will select the “best and the brightest,” by selecting students with high P-GRE scores, we will likely miss the “tenacious and capable”³ — students who have the desire but may need a little guidance or perhaps a semester of undergraduate E&M to get them ready for graduate studies.

And then there is the issue that some students don’t take the GRE-style multiple-choice tests very well (last time I checked, being able to take multiple-choice tests is rarely seen as a measure of excellence in research). Or very well documented issues like stereotype threat, where students of a particular group perform more poorly on high-stakes exams (like the GRE) when they fear they will confirm a stereotype of their group (e.g., poor math or science skills).

The P-GRE may be telling you something about a student’s pedigree, but very little about their potential. And just by requesting or requiring the exam, you will be telling many students not to apply to your program. Is this the message you want to transmit? The NSF decided it did not want to say this and dropped the GRE requirement for their prestigious graduate research fellowships a number of years ago. A number of prestigious universities have dropped their GRE requirement, and found an increase in the number and diversity of their applicant pool.

If you feel your department cannot stop using it, there are ways to reduce the bias introduced by the exam: (1) don’t require it; (2) don’t use it as a minimum score (ETS, the makers of the exam, even say this is inappropriate, and yet more than a third of all physics programs still follow this practice⁴); (3) don’t show the score to the admissions committee first (this introduces something called “anchoring bias,” where a reviewer will rate an application more poorly when a lower score is seen first, than if they were unable to see the score to begin with); (4) educate your admissions committee on the biases inherent in such exams that work against underrepresented groups, and (5) design a set of written guidelines for reviewing applications, and discuss these with the admissions committee — a way of minimizing arbitrary or unintended bias.

Finally, stay tuned for a proposed APS statement on the GRE. The APS Committees on Education, Minorities, and the Status of Women in Physics have recently proposed one for consideration to the Panel on Public Affairs. Let’s not waste talent by missing them because of this exam.

(Endnotes)

¹ C. Miller, B. Zwickl, J. Posselt, R. Silvestrini, & T. Hodapp, “Typical Physics PhD Admissions Criteria Limit Access to Underrepresented Groups but Fail to Predict Doctoral Completion” Sci. Adv. (in press).

² A. Kelly, K. Sheppard, “Secondary school physics availability in an urban setting: Issues related to academic achievement and course offerings,” American Journal of Physics 77 p902 (2009)

³ R. E. Scherr, M. Plisch, K. E. Gray, G. Potvin, & T. Hodapp, “Fixed and growth mindsets in physics graduate admissions” Phys. Rev. Phys. Educ. Res. 13, 020133 (2017).

⁴ G. Potvin, D. Chari, & T. Hodapp, “Investigating approaches to diversity in a national survey of physics doctoral degree programs: T`he graduate admissions landscape” Phys. Rev. Phys. Educ. Res. 13, 020142 (2017).

Disclaimer – The articles and opinion pieces found in this issue of the APS Forum on Education Newsletter are not peer refereed and represent solely the views of the authors and not necessarily the views of the APS.