The art and science of selecting graduate students in the biomedical sciences: Performance in doctoral study of the foundational sciences.

The goal of this study was to investigate associations between admissions criteria and performance in Ph.D. programs at Boston University School of Medicine. The initial phase of this project examined student performance in the classroom component of a newly established curriculum named "Foundations in Biomedical Sciences (FiBS)". Quantitative measures including undergraduate grade point average (GPA), graduate record examination (GRE; a standardized, computer-based test) scores for the verbal (assessment of test takers' ability to analyze, evaluate, and synthesize information and concepts provided in writing) and quantitative (assessment of test takers' problem-solving ability) components of the examination, previous research experience, and competitiveness of previous research institution were used in the study. These criteria were compared with competencies in the program defined as students who pass the curriculum as well as students categorized as High Performers. These data indicated that there is a significant positive correlation between FiBS performance and undergraduate GPA, GRE scores, and competitiveness of undergraduate institution. No significant correlations were found between FiBS performance and research background. By taking a data-driven approach to examine admissions and performance, we hope to refine our admissions criteria to facilitate an unbiased approach to recruitment of students in the life sciences and to share our strategy to support similar goals at other institutions.