Young-Hee Cho Chi-Ah Chun ^* Hector Ramos Paul Buonora Vasanthy Narayanaswami Kim-Phuong L Vu

• California State University, Long Beach, Long Beach, United States

Very few studies have examined the relationship between student characteristics and their acceptance to research training programs that use holistic selection. The present study addressed this question using institutional and applicant data of three NIH undergraduate training programs at California State University, Long Beach. Its first aim was to examine whether the applicants to the training programs were representative of the broader campus population. Its second aim was to investigate whether applicants who were accepted to the programs using a holistic selection process differed in academic discipline, demographics, and psychosocial characteristics from applicants who were not accepted. Information on students' majors, race/ethnicity, and gender was obtained from the university records or applications submitted by students. Majors were categorized as either biomedical or behavioral disciplines, while URM status was defined as students who self-identified their race and ethnicity as African American/Black, Native American, Native Hawaiian/Pacific Islander, or Hispanic. Applicants' psychosocial characteristics were obtained from a separate online survey. The acceptance status of applicants was obtained from the training programs' records. The applicant and non-applicant groups showed similar distribution of demographic characteristics regarding URM status and gender. Moreover, students' academic discipline and other demographic variables were not associated with application status at either the lower division (LD) or upper division (UD) levels. Although psychosocial characteristics measured with the online survey were not considered in the selection process, post-hoc analyses showed that LD applicants with higher grit and UD applicants with higher science interests were more likely to be accepted to the programs. The equal representation of URM and female students in the applicant and nonapplicant groups suggests that students from these traditionally underrepresented groups in STEM were just as likely to apply to our training programs. Furthermore, while the holistic selection process resulted in comparable acceptance rates across URM status and gender, it appeared to favor LD applicants with higher grit and UD students with higher science interests. These findings imply that research training programs can effectively recruit diverse students from underrepresented populations in STEM by using intentional outreach and recruitment efforts coupled with an objective and holistic selection process.