In Holistic Admissions, a Combination of Non-Academic Explanatory Variables Has Significant Predictive Value for Applicant Ranking

A medical school with a focus on community engagement has innovated its admissions process to include three different interview formats and one novel task format. Each component is designed to assess specific attributes of applicants, including teamwork skills, cultural competence, and service orientation. Correlations between these components are low, consistent with the original purpose that each should assess different attributes. To understand the use of the data by the members of the committee that ranks applicants, the authors created a model of seven explanatory variables, comprised of the three interview ratings and one task rating, a review of the written applications, and two measures of past academic performance. With regression analysis, the model significantly predicted applicant rankings, with most of the predictive capacity retained after omission of academic metrics. The results display that the school has developed innovations that allow for a reduced dependence upon academic history, and instead uses a truly holistic approach that is tailored to its mission. Most importantly, the work establishes that the admissions committee uses all the diverse forms of data provided to make decisions, which until this point has been an open question in holistic admissions.