What is the Relationship between Demographic Background and Test Score Submission to GRE-Optional Graduate Programs? It’s Complicated

Faculty members are committed to increasing the diversity of their students attending graduate school. One strategy that has been implemented in service of this goal is making the submission of GRE test scores optional. However, almost no research has examined the relationship between student demographics and propensity to submit GRE scores. This is the first study to do so using a relatively large sample (N > 15,000). We studied two demographic categories that are underrepresented in many academic disciplines: Females and racial/ethnic minorities (Black/African American, Hispanic/Latino, Multiracial, Native American/Pacific Islander applicants). Given the prominence of international students in U.S. graduate programs we also examined their presence in pools of GRE scores and nonsubmitters. Analyses were conducted at the overall sample level for applied, admitted, and enrolled students, and also master’s and doctoral programs in five academic disciplines. International students comprised larger portions of GRE score submitters than nonsubmitters among applied, admitted, and enrolled students. Members of racial/ethnic minorities comprised a larger percentage of GRE score nonsubmitters than submitters among applied but not admitted or enrolled students. Additionally, the 95% confidence intervals for female score submitters and non-submitters overlapped. More fine-grained analyses revealed additional differences: Racial/ethnic minority applicants were less likely to submit scores to humanities master’s, and STEM doctoral, programs. International applicants were more likely to submit scores to social sciences, nursing, and STEM master’s, and humanities and STEM doctoral, programs. Confidence intervals overlapped for all other comparisons. Descriptive trends varied across demographic groups, disciplines, and degree levels. Our results suggest that GRE-optional policies are neither a panacea nor a poison for graduate program diversity and that the effects of these policies need to be considered at fine-grained levels of analysis (e.g., academic discipline, degree level).