Multiple Mini Interviews vs Traditional Interviews: Assessing Reliability and Equity in Medical School Admissions [Letter]

Dear editor

I found the study by Banks et al¹ on racial and socioeconomic factors in medical school interviews insightful, particularly in its comparison of Multiple Mini Interviews (MMIs) with traditional formats. A key strength is its use of a comprehensive dataset spanning multiple admissions cycles, enhancing reliability. Through correlation analyses and effect size calculations, the authors illustrate how structured interviews influence demographic disparities—an important aspect of refining admissions processes. While the study contributes meaningfully to equitable admissions discussions, certain methodological and interpretative aspects warrant further scrutiny.

First, the study uses the Casper test as a benchmark for interpersonal and intrapersonal competencies, raising concerns. While widely used, research suggests Casper exhibits moderate racial and socioeconomic disparities, potentially perpetuating inequities rather than mitigating them.² Though the authors acknowledge this limitation, they interpret MMI-Casper correlations as evidence of reliability without fully addressing whether Casper is an appropriate comparator.

Second, the study suggests that MMIs and traditional interviews perform similarly in reducing demographic differences. However, effect sizes reveal nuanced disparities. While MMIs helped narrow demographic gaps for African American applicants, they slightly widened them for underrepresented in medicine (URM), Hispanic, and disadvantaged applicants. Though relatively small, these differences indicate that MMIs do not provide uniform benefits across disadvantaged groups. Prior research links socioeconomic status to MMI performance, likely due to unequal access to preparation resources.³ Future studies should explore targeted interventions—such as structured coaching—to level the playing field for lower-income applicants.

Additionally, the transition from traditional interviews to MMIs coincided with broader changes in interview structures and faculty training, complicating efforts to isolate MMIs’ specific impact on bias reduction. Since structured traditional interviews improve scoring consistency,⁴ a direct comparison between structured traditional interviews and MMIs—rather than unstructured interviews—would provide clearer insights.

Lastly, the authors propose reducing the weight of academic metrics such as the Medical College Admission Test (MCAT), undergraduate Grade Point Average (GPA), and Casper scores in favor of MMIs to improve equity. While this aligns with holistic admissions principles, it risks weakening predictive validity. Despite limitations, academic metrics correlate with medical school performance and licensing exam outcomes.⁵ A balanced approach would integrate multiple assessment tools while refining interview methods to maximize fairness.

In summary, this study provides valuable insights into interview formats and applicant evaluations. However, future research should reassess Casper’s suitability as a benchmark, investigate MMIs’ differential impact on specific demographic groups, and explore structured traditional interviews as a viable comparator. By refining admissions processes with a data-driven, equity-focused approach, medical schools can advance both fairness and predictive validity in candidate selection.