Lines in the sand: pre-interview rank and probability of receiving admission to medical school

Abstract

Background: We provide an examination of one medical school’s attempt to determine whether their cut-off point for number of interviews offered is congruent with the probability these applicants’ have for admission post-interview.

Methods: Offer probability was determined by organizing pre-interview rankings from 2013-2017 (n = 2,659) applicant cohorts into bins of 50 applicants and finding the quotient of successful and total applicants in each bin.A linear-by-linear association Chi-square test and adjusted standardized residuals with an applied Bonferroni correction were used to determine if the observed frequencies in each bin were different than expected by chance. A Spearman Correlation analysis between pre- and post-interview ranks was conducted.

Results: All applicants have between a 50.0% and 76.4% chance of admission. Observed frequencies are different than chance (χ(1)=50.835, p<.001), with a significantly greater number of offers seen in the bins between 1 and 100 (p<.001 for both bins). There is a weak positive relationship between pre- and post-rank, r_s(2657)= 0.258, p<.001.

Conclusion: The results indicate the number of interviews conducted does not exceed a threshold wherein individuals with a relatively low chance of admission are interviewed. Findings are interpreted with respect to ethical resource allocation for both programs and applicants.