Investigating the Reliability of the Medical School Admissions Interview
- Cite this article as:
- Kreiter, C.D., Yin, P., Solow, C. et al. Adv Health Sci Educ Theory Pract (2004) 9: 147. doi:10.1023/B:AHSE.0000027464.22411.0f
Purpose: Determining the valid andfair use of the interview for medical schooladmissions is contingent upon a demonstrationof the reproducibility of interview scores. This study seeks to establish thegeneralizability of interview scores, firstassessing the existing research evidence, andthen analyzing data from a non-experimentalindependent replications research design.Methods: Multivariate andunivariate generalizability analyses areconducted using data from a structuredinterview obtained from a population of medicalschool applicants over two years.Results: The existing literaturedoes not provide sufficient evidence regardinginterview reliability. In this study,interview scores derived from a standardizedinterview were found to display low to moderatelevels of reliability. Interview scores do notappear to possess the level of precision foundwith other measures commonly used to facilitateadmissions decisions.Discussion/Conclusion: Given theresults obtained, the fairness of using theinterview as a highly influential component ofthe admission process is called into question. Methods for using interview data in apsychometrically defensible fashion arediscussed. Specifically, attention to decisionreliability provides guidance on how interviewscores can best be integrated into theadmissions process.