A Challenge and an Opportunity for Graduate Admissions
Addressing equity is imperative. That means revamping more than just GRE requirements.
Opinion By Kirin E. Furst, Rachel C. Scholes, Aidan R. Cecchetti, and Jonathan O. Sharp
As part of our nationwide reckoning with systemic racism, engineers and academics have been called to confront the inequities in our fields. Despite years of university diversity and inclusion initiatives, racial disparities remain stark in graduate engineering. Black students continue to be severely underrepresented; women and people of other marginalized identities remain underrepresented as well. Lack of diversity in our graduate programs translates into lack of diversity in our faculty and the engineering organizations that hire our graduates, limiting our ability to address the needs of the diverse populations we serve.
Engineering faculty control many departmental practices, such as graduate admissions, that present broad opportunities to increase graduate student diversity. As Yvette E. Pearson, associate dean in Rice University’s School of Engineering and Engineering Change podcast host, puts it, if we applied the mindset we use for engineering challenges to develop equitable practices in our research groups and departments, we could make substantial progress in diversifying engineering. In that spirit, we want to share what we learned by taking an engineering-inspired approach to reforming graduate admissions.
The COVID-19 pandemic has already provoked some reexamination of graduate admissions practices, with various institutions dropping GRE requirements. This may be a good first step, because the exam cost and overweighting of GRE scores by admissions committees reinforces racial, gender, and socioeconomic inequities in access to graduate school. However, the GRE is just one of many components prone to bias in graduate admissions. Simply removing GRE requirements may have little impact on equity in graduate admissions and could even make admissions less equitable by shifting weight to other problematic metrics. Graduate admissions is a complex system of many components. Thus, making it more equitable requires an engineering approach.
Over the past several years, our NSF Engineering Research Center has explored practices for reengineering graduate admissions processes. Our goals were to increase diversity while improving how we select for attributes that contribute to graduate student success. Although our existing procedures considered those attributes, we lacked direct, transparent methods of evaluation. In our literature review, we learned that traditional metrics such as GRE scores and undergraduate institution reputation do not necessarily translate into student outcomes such as completion of degree programs or number of publications. Instead, personal attributes such as persistence, intrinsic motivation, and creativity are better predictors of graduate student success. These attributes should be evaluated with targeted essay prompts and a detailed evaluation rubric.
Diversity of the admissions committee is also critical. We experienced firsthand the value of feedback from diverse stakeholders by engaging graduate students in the admissions process alongside faculty and staff. Finally, we learned the importance of acknowledging unconscious bias in letters of recommendation—for example, recognizing that gendered descriptions can disadvantage applicants.
To support our departments and others in reengineering graduate admissions, we compiled a guide for implementing evidence-based practices that promote equity. This guide consists of an extensive literature review, surveys from our universities, and documentation of practices implemented in our departments. We focused on five practices that, when implemented as part of a systematic, holistic review, can make admissions decisions more equitable and better aligned with programmatic goals: (1) diversifying admissions committees, (2) using rubrics, (3) de-emphasizing GRE scores, (4) soliciting personal history or diversity statements, and (5) controlling for potential bias in recommendation letters.
When implemented as part of holistic review, several of these practices have measurably increased the diversity of admitted and enrolled graduate student cohorts. Studies of graduate professional programs with large cohorts offer the clearest demonstration of the potential of holistic review; for example, one medical school found holistic review increased the number of underrepresented minority applicants selected for interviews by over 200 compared to review using traditional academic metrics—a more than tenfold increase.
We acknowledge that reengineering graduate admissions does not directly address low diversity in the applicant pool for graduate engineering programs (the “pipeline”), which can be a barrier. However, we must not wait until the pipeline is full to make our admissions systems more equitable. The time to reengineer graduate admissions is now, because it’s the right thing to do for our students and our disciplines—and because it’s good engineering!
Kirin E. Furst, Rachel C. Scholes, Aidan R. Cecchetti, and Jonathan O. Sharp are current and former members of the NSF Engineering Research Center for Re-Inventing the Nation’s Urban Water Infrastructure (ReNUWIt) and contributors to the referenced guide, “Evidence-Based Practices for Systematic Holistic Review in Graduate Engineering Admissions,” available at http://renuwit.org/diversity/resources.
See Databytes for more on GRE requirements in doctoral admissions.