Researchers at North Carolina A&T State University will conduct a longitudinal study using big data analytics to advance the knowledge and evidence base for successful strategies to encourage underrepresented minorities to pursue STEM pathways. The goals of the project are to characterize and understand pathways to STEM, assess STEM intervention effectiveness at universities across the nation based on institutional system/program metrics for student success, and recommend ways to increase the number of underrepresented minority women pursuing STEM degrees and careers. The study will examine factors that influence transitions to STEM for undecided/non-STEM students and demonstrate specific systematic/programmatic interventions that are effective. The research will be guided by a triangulation of social cognitive career theory, student persistence theory, human capital theory, and vocational anticipatory socialization theory. The project is an innovative approach to better understanding factors that motivate students to elect STEM degree programs, and it is expected to develop a core set of best practices metrics for STEM transitions and pathways.
The primary data sources for the study are the High School Longitudinal Survey and the Beginning Postsecondary Students Survey. The data will be analyzed to examine student pathways into STEM, with a focus on students who are undecided about their majors upon entering community college and four-year degree programs. The specific research questions are (1) what are the academic, socioeconomic, and demographic supports and barriers for underrepresented minorities (URM)and undecided/non-STEM major transitions to STEM?; (2) what effective pathways and higher education institutional systems/programs support URMs, undecided/non-STEM successful transitions to STEM?; (3) which institutions have data-based evidence to support their best practice intervention strategies?; (4) what implementable strategies will support successful transitions and effective pathways into STEM, particularly for URMs and undecided/non-STEM students from minority serving institutions?; and (5) what shared metrics are recommended for institutions to use nationally to assess successful transitions and effective pathways for URMs into STEM? Analytical techniques will include Random Forest, Principal Component Analysis, XGBoost, and Deep Learning to analyze the large datasets and conduct the meta studies on the variables. The study results could impact policy and practice at higher education institutions at large.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
