Skip Navigation LinksHome > February 2013 - Volume 88 - Issue 2 > The Possible Impact of the MMI and GPA on Diversity
Academic Medicine:
doi: 10.1097/ACM.0b013e3182830390
Letters to the Editor

The Possible Impact of the MMI and GPA on Diversity

Albanese, Mark PhD

Free Access
Article Outline
Collapse Box

Author Information

Professor Emeritus, University of Wisconsin-Madison, and director of research, National Conference of Bar Examiners, Madison, Wisconsin;

Back to Top | Article Outline

In Reply to Norman et al:

Jerant et al, in their study of the impact of a Multiple Mini Interview (MMI) on candidates offered admission to their institution, found statistically significant results favoring female candidates who were extraverted, agreeable, slightly older, and did slightly poorer on the MCAT physical sciences subscore.1 The authors’ concern over the narrowing of diversity based upon the extraversion finding prompted Norman and colleagues’ letter. They argue that extraversion could easily be a desirable quality in physicians and that the MMI overcomes some of the known problems with the traditional interview.

From my view, none of the personality results probably much matter. The extraversion mean for the group offered admission was 3.8 versus 3.5 for those not offered admission, a difference of only about 0.38 SDs, a value between a small (0.2) to medium (0.5) effect size by Cohen’s criteria.2 The results for agreeableness were even smaller (0.2 SDs).

There are two statistically significant results, however, that are both meaningful in size and also can have a direct impact on the larger society. The first is that men made up 48.0% of the candidates brought in for the MMI but only 40.7% of those who were offered acceptance. This appears to be part of a larger trend. In the United States, in 2009–2010, men received only 42.8% of the 1,650,014 bachelors’ degrees awarded.3 This percentage has been about the same since 2003. These are probably the results of a host of initiatives to correct historical gender inequities.4

Thus, unless provisions are made for coming to a soft landing, things will over-correct. If the MMI adds a layer of gender correction on top of previous corrections, and if the differences in gender are not found to be related to candidates’ eventually becoming better physicians, it would not be too difficult to apply a statistical adjustment to compensate when selecting future candidates.

A second issue pertains to representation of the 19–21 year-olds (YOs) and the 25–39 YOs. The younger group constituted 21% of those sitting for the MMI and the older group constituted 27%. Among those offered acceptance, the percentages were 17% and 30%, respectively. This outcome is not surprising, since the MMI is a performance-based examination that requires demonstration of skills that often come with maturity. One might say that the outcome is, in fact, fair because we want to select mature individuals who understand what medicine is all about and will truly yield to its life calling.

However, if some 19–21 YOs have fast-tracked to be competitive enough to warrant invitations to the MMI, that is a remarkable accomplishment. The commitment to their studies needed to do that is likely to leave them with less life experience and maturity than even that of their age-mates, let alone individuals who have lived 25% longer. But it will extend their careers another 5 years, and they are likely to accumulate less debt. So, like most things in life there are trade-offs. Do we select for maturity and life experience at the expense of a shorter career and perhaps a less concentrated practice (because of distractions of continued debt and commitments accumulated by life’s challenges), or do we risk youth and immaturity for long practice but potential misfit? With the physician shortage likely to continue to worsen, we may be forced to make decisions like this in a more draconian and less humane manner.

Surprisingly, Jerant and colleagues found no relationship of the MMI to conscientiousness. If they were able to neutralize the MMI as a tool for determining conscientiousness, which they attribute to their screening criteria, how they did this could be very important to report to the medical education community.

My last point is that Jerant’s institution, unlike many medical schools, attracts a very accomplished applicant pool. The MCAT scores of those invited for the MMI ranged from a mean of 10.0 for Verbal Reasoning to 11.4 for Biological Sciences. With students who achieve such scores, admission committees have room to “play” with noncognitive variables with little risk of admitting unqualified students. Medical schools that do not attract such high-performing applicants may need to be more careful. Their desire for noncognitive diversity may lead to admission of students who will struggle and may ultimately fail. The MMI is a great tool, but like any tool, you have to use it right or you can break things.

Mark Albanese, PhD

Professor Emeritus, University of Wisconsin-Madison, and director of research, National Conference of Bar Examiners, Madison, Wisconsin;

Back to Top | Article Outline


1. Jerant A, Griffin E, Rainwater J, et al. Does applicant personality influence Multiple Mini-Interview performance and medical school acceptance offers? Acad Med. 2012;87:1250–1259 Accessed December 5, 2012

2. Cohen J Statistical Power Analysis for the Behavioral Sciences. 19882nd Edition Hillsdale, NJ Lawrence Erlbaum Associates

3. Institute of Educational Sciences, National Center for Education Statistics. Accessed December 5, 2012

4. Shelley J. Ten reasons why men aren’t going to college [unpublished]. Accessed December 5, 2012

© 2013 Association of American Medical Colleges


Article Tools