"Healthy" Medical School Admissions

What is the MCAT? The Medical College Admissions Test (MCAT) serves as the entry screening assessment for most U.S. medical schools. Developed by the Association of American Medical Colleges (AAMC) and administered to over 55,000 students annually, the 6-hour exam consists of 221 multiple-choice items and two essays. Divided among four sections, scores on Biological Sciences, Physical Sciences, and Verbal Reasoning are each reported on a scale from 1 (lowest) to 15 (highest). The Writing Sample receives a score of 1 to 6, converted to an alphabetical scale ranging from J (lowest) to T (highest).
The MCAT does a poor job of predicting success in medical

According to the AAMC, the MCAT is designed to “assess mastery of basic concepts in biology, chemistry, and physics; facility with scientific problem solving and critical thinking; and writing skills.” According to several independent researchers however, the skills the MCAT tests relate most closely to the first two years of medical school classroom work but have little relevance in predicting success during clinical training or actual practice as a physician. William McGaghie writes in the journal Academic Medicine: “No physician answers pages of multiple-choice questions when he or she practices clinical medicine‚ĶGrades predict grades, test scores predict test score, ratings predict ratings, but attempts to demonstrate scientific convergence among such indicators of professional competence have not been successful.”1

While the MCAT’s developer claims the exam should play a central part in forecasting success in medical school, the predictive value of the test is in fact quite small. Several studies confirm that the predictive ability of the MCAT is confined to the basic science portion of the medical school curriculum, explaining only 9-16% of the variance, or difference, in grades during the first two years. MCAT scores prove to be an even weaker indicator of performance during clinical training (the third and fourth years of medical school).

One research study showed that as a student progresses through medical school the power of the pre-admission interview ratings to predict medical school grade point average (GPA) generally increases over time while the power of MCAT scores decreases.2 Another study considered the effects of MCAT scores and “non-cognitive” measures on basic science grades and clinical competence in medical school. While only 4% of the variation in the ratings of clinical competence were related to MCAT scores, 14% could be explained by psycho-social measures.3 While no one factor contributes greatly to predicting success, test scores are clearly only weak predictors whose value decreases as students progress through training.

Strict use of the MCAT disadvantages students of color and harms communities of color

Substantial score differences between Whites and students of color lead to an underepresentation of minorities in medical school when MCAT scores play a large role in admissions decisions (see table). In states where affirmative action programs have been eliminated, the numbers of students from historically underrepresented groups in medical school have plunged dramatically. In 1997 the number of African Americans, Latinos, and Native Americans accepted to publicly-supported medical schools in California, Texas, Mississippi, and Louisiana (all states that eliminated affirmative action for the first time that year) dropped by 27%.

The Journal of Blacks in Higher Education estimated that if the top 10 highest-ranking medical schools in the nation dismantled their affirmative action policies and replaced them with strict MCAT cut-off scores, only 7 Black students would meet entrance requirements.4 Another study revealed that under admissions in which MCAT scores are of primary importance, the percentage of Blacks enrolled at the 25 most selective medical schools would drop from 10% to 1%.5 Derek Bok, former president of Harvard University, cautioned against this phenomenon: “The most dramatic effects in eliminating affirmative action would be on medical and law schools because the more selective the institution, the greater the impact, and nothing is more selective than medical school and law schools.”

The consequences of a shortage of doctors of color would be devastating. More than half of all patients seen by Black doctors are Black; in contrast, White doctors practice in communities in which on average only 5% of the population is Black. Black doctors practice in communities in which nearly half of the population lives in poverty.6 If a state no longer educates significant numbers of Black doctors, it is estimated that communities with large Black populations would be four times as likely to have doctor shortages.

In fact, since test scores substantially fail to predict success, medical school admissions offices can promote equity and excellence simultaneously by deemphasizing test scores. At the University of California, Davis medical school, a twenty-year study measured the effects of affirmative action policies on promoting racial diversity on campus. Students admitted under affirmative action – 20% of whom as part of their “special consideration” status did not need to meet MCAT and GPA minimums – had graduation rates and performance reviews in residencies that were “remarkably similar” to students admitted under standard criteria.7

Coaching strips away the MCAT’s claim as an “objective” measure Like all standardized admissions tests, the MCAT is extremely susceptible to coaching. Despite its insistence that coaching is not effective, the AAMC sells MCAT practice tests for $40 each. Coaching companies offer more intensive experiences, with large score gains guaranteed. Test takers can spend $1299 for one major test prep company’s MCAT preparation classroom course. They can also participate in an on-line course for $499, or can arrange for 15 hours of private tutoring for $1999. The impact of the coaching “steroid” in boosting scores leads to a substantial disadvantage for students who cannot afford expensive test preparation. Yet admissions officers have no way of knowing which applicants have benefited from coaching and which have not, eroding whatever “objective measure” the MCAT promoters claim the test can provide.
Medical school admissions without the MCAT?

Medical school admissions officers can and do make sound decisions about who to admit without considering MCAT scores. One approach is to offer provisional medical school admission to a student as he or she enters the freshman year of college. Approximately 30 medical schools, including the Johns Hopkins, University of Rochester, Northwestern, Brown, and Tulane, pursue such a policy guaranteeing a seat to incoming undergraduates if they maintain a certain GPA.

Another approach to de-emphasizing test scores is offered through the Expanded Minority Admissions Exercise (EMAE). Developed by the AAMC, EMAE trains admissions committees in the use of “non-cognitive” variables for the assessment of applicants of color. Some of the factors admissions committees learn to consider include: leadership, motivation, realistic self-appraisal, family and community support, maturity and coping capability, and communication skills.8 While the program focuses on students of color, the criteria used and the holistic approach to reviewing applicants can certainly be applied more broadly.

The Texas A & M University System offers another admissions route that doesn’t require MCAT scores. Applicants coming from areas of Texas with a shortage of medical services can become part of the Partnership for Primary Care Program. Participation in the program allows for the waiver of all MCAT requirements, provided the individual pledges to work in underserved areas upon graduation.

In a 1991 report, the Southern Regional Education Board highlighted the admissions practices of medical schools with high enrollments of people of color. Among their most significant findings were that these schools used MCAT scores to compare students of color among themselves and to diagnose special needs a student may have, and not as an absolute admissions hurdle. In contrast, medical schools with low enrollments of students of color tended to emphasize MCAT scores and use arbitrary minimum scores.9 A letter to the editor in the Chronicle of Higher Education about the report issued a warning: “The single greatest barrier to admission of minority applicants is an over-reliance by many medical schools on test scores as predictors of success in the curriculum.”

1 McGaghie, W. “Perspectives on Medical School Admissions.” Academic Medicine, Vol. 65 (No. 3), March 1990, pp. 136-139.
2 Elam, C.L.; Johnson, M.M.S. “Prediction of Medical Students’ Academic Performances: Does the Admission Interview Help?” Academic Medicine, Vol. 67, 1992, S-28-S30.
3 Hojat, M., Robenson, M., Damjanov, I., Veloski, J.J., Glaser, K., & Gonnella, J.S. “Students’ Psychosocial Characteristics As Predictors of Academic Performance in Medical School.” Academic Medicine, 68, 1993, pp. 635-637.
4 “White Coats, White Faces: Calculating the Impact on Black Admissions to Medical School If Affirmative Action Were Eliminated.” Journal of Blacks in Higher Education, Autumn 1997, pp. 15-17.
5 “Why a Nationwide Ban on Race-Conscious Admissions Will Sharply Curtail Black Enrollments at the Nation’s Highest-Ranked Medical Schools.” Journal of Blacks in Higher Education, Spring 1999, pp. 22-25.
6 Cited in JBHE, 1997.
7 Davidson, R. and Lewis E. “Affirmative action and other special consideration admissions at the University of California, Davis, School of Medicine.” JAMA, Vol. 278, 1997, pp. 1153-1158.
8 Association of American Medical Colleges, Expanded Minority Admissions Exercise (EMAE), 1999. http://www.aamc.org
9 Denton, D. Recruitment and Retention of Minority Medical Students in SREB States. Atlanta, GA: Southern Regional Education Board, 1991.

2000 MCAT scores by race/ethnicity and sex

African-American/Black 6.0 6.3 6.3 N
American Indian 7.3 6.9 7.2 O
Caucasian American/White 8.3 8.4 8.6 O
Mexican American/Chicano 6.9 7.1 7.3 O
Asian American 7.6 8.9 8.8 O
Puerto Rican-Mainland 7.7 8.0 8.0 M
Puerto Rican-Cmnwlth. 4.6 5.4 5.3 K
Other Hispanic 7.2 7.5 7.8 O
Female 7.7 7.7 8.0 O
Male 7.8 8.7 8.7 O
ALL TEST TAKERS 7.8 8.2 8.3 O

Approximately 54,000 test-takers, 52.3% female
Source: American Association of Medical Colleges, Summary Data on the Combined April/August
2000 MCAT