Researchers at the Oregon Health & Science University-Portland State University School of Public Health are leveraging data science to support students from diverse and underserved backgrounds pursuing science, technology, engineering and mathematics, or STEM.
A new study, published this week in the Journal of STEM Outreach, examines effective approaches for measuring key demographic variables and offers recommendations for inclusive demographic data practices across student-serving programs. This study aligned researchers and STEM practitioners across OHSU, Portland State University and the Portland Metro STEM Partnership.
Recognizing students beginning their STEM training may not have had equal access, accommodations or preparation, the study describes considerations for programs to welcome and retain trainees using demographic data. This work builds on federal efforts to reach underrepresented populations in STEM and biomedical science.
Leveraging inclusive data science
Demographic data allow programs to identify groups who are facing barriers or being actively excluded or marginalized from participation in STEM programs. Demographics like age, gender, race and ethnicity are often used in STEM training programs, but additional demographics are needed to capture the complexity of student identities as well as the contextual factors that influence STEM training.
Examples include using open-ended prompts for students to self-describe, expanding answer options beyond the gender binary, representation of LGBTQIA+ populations, inclusion of socioeconomic considerations like houselessness, and sharing data practices with trainees regarding the privacy and safety of their shared information.
Additionally, because everyone is more than a single demographic, this project explores intersectionality across demographic groups and aims to provide insights, best practices and recommendations for STEM programs to improve demographic data practices. This allows programs to more effectively address inequities and support recruitment and retention of underrepresented students.
“Demographics enable the study of intersectionality needed to truly move the needle for equity, diversity, inclusion and accessibility,” said Lisa K. Marriott, Ph.D., associate professor in the OHSU-PSU School of Public Health and principal investigator of OHSU’s NIH-funded Science Education Partnership Award. “It’s so important to make sure students feel welcome and seen in training programs, and on the back end, making sure they are accurately represented within the data.”
The study emerged from previous work by Marriott’s team building the STEM Assessment and Reporting Tracker, or START, which helps schools and STEM programs measure their students' development in science. When expanding demographics for START, other STEM programs shared they also had questions about demographics, presenting an opportunity to study demographics collaboratively.
The study shares lessons learned from STEM practitioners and equity researchers, offering recommendations for responsible, effective practices for demographic data needed to broaden and diversify the biomedical workforce.
Community-engaged partnerships
This work is just one component of the university’s ongoing efforts to build capacity in inclusive data science for programs that support STEM.
Marriott’s lab also earned a Science Education Partnership Award focused on data science, which aims to train middle and high school students in equitable data science while collaboratively studying how STEM programs are incorporating demographics into their work.
“Inclusive demographic data collection is an essential piece of the puzzle, because they’re necessary to understand if all students have equal and consistent access to STEM education,” Marriott said. “We know these students will provide so many valuable contributions to research and scientific discovery, but we need to ensure they’re given that opportunity.”
Research reported in this publication was supported by the National Institutes of Health (NIH) Science Education Partnership Awards (R25GM129840, R25OD010496, and R25GM150166 to LKM) and a NIH award for Excellence in Diversity, Equity, Inclusion, and Accessibility (DEIA) Mentoring (R25GM129840-04S1 to LKM) administered by the National Institute of General Medical Sciences. Informatics for START supported by the OHSU Clinical and Translational Research Institute (UL1TR002369). The work is solely the responsibility of the authors and does not necessarily represent the official view of the National Institutes of Health.