UNIVERSITY PARK, Pa. — A new study led by a researcher at Penn State offers evidence that more comprehensive, or holistic, admissions practices can serve to provide opportunities for more students in graduate education — especially in science, technology, engineering and mathematics (STEM) fields.
The study, co-authored by Kelly Rosinger, associate professor of education and public policy at Penn State, and published in The Journal of Higher Education, quantified the effects of a comprehensive holistic admissions intervention developed by the research team on actual enrollment outcomes in 26 STEM doctoral programs at five research universities in California over three years. The researchers found that participating departments saw sustained increases in applications, admissions and enrollment of students from historically underrepresented backgrounds, according to Rosinger.
The intervention centered on a series of in-depth workshops, developed by the research team, that provided faculty with training and tools to evaluate an applicant more comprehensively, assessing their academic achievement and potential alongside their other accomplishments. The workshops guided departments through critical self-assessment of their existing admissions practices. A major focus was reducing overreliance on Graduate Record Examination (GRE) scores and elite academic pedigrees, which often benefit students from historically advantaged backgrounds. The GRE is a standardized test that is part of the admissions process for many graduate schools in the U.S. and Canada, along with a few other countries.
“Graduate and professional education has long been an engine of opportunity, but it’s also a space where inequities — particularly for students of color and women in STEM — are deeply entrenched,” Rosinger said. “This study helps us understand what happens when departments make intentional, structural changes to the way they assess candidates. The impact is real and measurable.”
The research not only captured how faculty implemented new admissions practices but also measured shifts in the racial and gender composition of applicants and admitted and enrolled students.
“Instead of asking, ‘What has this applicant already achieved?’ faculty were trained to ask, ‘What is this applicant’s potential given their context?’” Rosinger explained. “That shift —toward contextualized evaluation — opens the door for more students with high potential who may not come from traditional academic pipelines.”
Faculty participants became part of a consortium, meeting regularly for discussions, troubleshooting and long-term collaboration around admissions reform.
To measure the impact of the intervention, Rosinger and her team employed a quantitative research design. They collected longitudinal data from participating departments, comparing practices and enrollment patterns before and after the intervention. Surveys captured how departments were shifting their recruitment, review and mentoring approaches. Meanwhile, administrative data tracked changes in the demographic makeup of applicant pools and enrolled cohorts over time.
“We’re making a dent in a very large, very rigid system,” she said. “This isn’t a one-year fix — it’s about shifting the way institutions think about talent over the long haul.”
The study is rooted in an organizational learning framework, which focuses on lasting behavioral and cultural change rather than quick policy fixes. The collaborative community built during the workshops proved essential to sustaining momentum after the formal intervention ended, Rosinger said.
“These weren’t isolated policy tweaks,” Rosinger explained. “These were programs that took a hard look at their admissions systems, made intentional changes, and kept coming back to reflect and refine those changes together.”
The findings are particularly relevant, she said, at a time when institutions across the country are reexamining their admissions policies.
“Holistic review is already being used effectively in states like California where affirmative action has been banned for decades. This study shows it’s not just theoretically promising — it can work,” she said. “While not every program can replicate the full community-based model, key elements — like rubrics, training and faculty reflection — can be widely shared and adopted. That’s how real, lasting change happens.”
Rosinger’s co-authors on the paper are Julie Posselt, professor of higher education in the USC Rossier School of Education; and Casey W. Miller, associate dean for research and faculty affairs and professor in the College of Science at the Rochester Institute of Technology.
This study was supported by the U.S. National Science Foundation through an Innovations in Graduate Education grant.