Undergraduate education signals the start of a physician's educational continuum, with medical school admission committees serving as “gatekeepers” to the profession. Besides weighing undergraduate science and nonscience grade point averages (GPA) and Medical College Admission Test (MCAT) scores, admission committees also consider the breadth and difficulty of undergraduate work.1 Since the Preparation for Medical Education in the Liberal Arts College2 in 1953 to the 1984 Physicians for the Twenty-first Century Report on the General Education of the Physician and College Preparation for Medicine,3 medical educators have advocated a “baccalaureate education that encompasses broad study in the natural and the social sciences and in the humanities.”3
For many, a liberal arts education meets this expectation and provides an excellent preparation for medical school. In fact, the Oberlin Report, using data from 48 private liberal arts colleges, affirmed that leading liberal arts colleges rank at or near the top of all American institutions of higher education in the training of scientists.4 Report findings further suggested that this success was due to the close linkages maintained between teaching and faculty research, fostering interest in science. Liberal arts curricula offer students the opportunity to study a core area or set of required courses that cover a broad spectrum of disciplines - requirements that enable a broad exposure to humanities, social sciences, and the arts while still developing the scientific skills necessary for medicine.5 Small class size, close student-faculty interactions, and use of collaborative learning techniques are also seen as hallmarks of a liberal arts education.
The education that liberal arts colleges provide might best be described as preprofessional, since a disproportionate number of their students enroll in graduate or professional programs upon graduation. However, virtually no undergraduate professional education is offered at a liberal arts college.6 Accordingly, liberal arts institutions are defined by the Carnegie Classification of Institutions of Higher Education as “primarily undergraduate colleges with major emphasis on baccalaureate programs...that awarded at least half of their baccalaureate degrees in liberal arts fields.”7 Using this definition, 228 of 3,941 higher education institutions in the United States are designated as Baccalaureate Colleges-Liberal Arts.7
At a 2002 conference sponsored by the Center of Inquiry in the Liberal Arts at Wabash College, 17 medical school admission deans from across the United States were invited to discuss “Liberal Arts Education and the Medical Profession.” Preliminary data shared among conferees included the numbers of liberal arts applicants and matriculants at schools represented, as well as anecdotal evidence regarding medical school performance of liberal arts students. All but one of the conference attendees reported that the proportion of liberal arts students accepted exceeded that contained in the applicant pool. This anecdotal evidence was borne out by Association of American Medical Colleges (AAMC) data showing graduates of liberal arts colleges having higher acceptance rates than graduates of any other Carnegie class of schools over the past five years.5 However, it was recognized that little empirical research had been conducted in the medical school setting to explore how admission committee members perceived students with liberal arts degrees or how those students subsequently performed in medical school.
The purpose of this preliminary study was twofold: (1) to examine admission committee members' assessments of the utility of a liberal arts education as preparation for medical school and the medical profession; and (2) to examine medical school performance of students completing undergraduate degrees at liberal arts institutions. To that end, the current study addresses three questions regarding graduates of liberal arts programs. First, using survey data, we examined admission committee members' perceptions of applicants completing degrees from liberal arts colleges. Second, we compared preadmission academic characteristics of matriculants from liberal arts colleges with graduates of comprehensive doctorate or master's granting universities. Third, we compared these two groups across various measures of academic, clinical, and extracurricular performance.
An 11-item, open-ended questionnaire containing liberal arts education-related questions was administered to admission committee members at a public medical school in the southeastern United States. The questionnaire, based upon work by McGrath in preparation for the Wabash conference,8 addressed: (1) differences between candidates from liberal arts colleges and comprehensive universities (in general and with regard to science preparation); and (2) unique contributions of liberal arts colleges (e.g., size, instruction, academic or social environment, extracurricular involvement) to educational quality. Committee members were also asked whether they inquired about liberal arts education or quality of science/nonscience education during admission interviews. Completed questionnaires were obtained from 16 of 17 members (response rate = 94%).
Student subjects consisted of the 2001 and 2002 graduates at the study institution. Of the total of 190 graduates, 45 (23.7%) had completed their undergraduate degree at Carnegie-classified liberal arts college. Preadmission academic data included undergraduate science, nonscience, and total GPA, and MCAT scores on four scales- Verbal Reasoning (VR), Physical Science (PS), Writing Sample (WS), and Biological Science (BS). The medical school academic performance measures collected were average class ranking and scores on the Clinical Performance Examination (CPX) – a comprehensive test of basic clinical skills administered to all students following completion of the M3 year.
Additional data were gathered from study subjects who, as applicants, voluntarily completed a confidential, two-page questionnaire along with their secondary application materials. Subjects' participation and responses had no bearing on their admissions status, and the project protocol was approved by the medical center institutional review board. The questionnaire addressed preferred learning methods (lecture, computer, self-study, group discussion), approaches to learning (problem solving, memorizing facts, digging out knowledge, discussing issues, applying information, organizing concepts, spoon-feeding), and medical specialty preferences (surgery, internal medicine (IM), pediatrics, family medicine (FM), pathology, obstetrics/gynecology, psychiatry, neurology, and emergency medicine).
These factors were assessed using a paired comparison scaling methodology.9 In this traditional format, each dichotomous “forced choice” response is contrasted with all other possible combinations (or pairs) – allowing respondents to make comparative rather than absolute judgments. Thus, preferences for the nine medical specialties, for example, resulted in 36 pairs of comparisons systematically listed in arbitrarily-chosen order: The first category (surgery) being contrasted with all others; next, IM being contrasted with all others except surgery, and so forth. Using a series of computations, scores were then calculated separately across each response category (medical specialty). Given responses to all pairs, this “reduced subset cyclic design” results in a complete ranked set of options – ranging, in this case, from 0 (preferred below all alternatives) to 100 (preferred above all alternatives). In this fashion, it is possible to relate the preference for each specialty relative to all others, with measures representing the percentage of time a given response is chosen relative to the alternatives. Applicants' preferences toward instructional methods and approaches to learning were calculated similarly, with an average score computed across three dimensions: (1) comfort; (2) effectiveness; and (3) interest.
Finally, medical school honors and activities such as Alpha Omega Alpha (AOA) membership, the medical honorary, and number of activities self-reported in the Graduation Program (including membership in service and medical specialty clubs, volunteer activities, and elected offices) were also examined.
Among admission committee members, 62.5% perceived no difference between applicants from liberal arts colleges and comprehensive doctorate or master's granting universities. Of those who did note differences in liberal arts students, factors such as their being more broadly educated, well rounded, and having more experience in integrated courses were mentioned. However, all members felt that liberal arts colleges offered an academic and social environment conducive to developing the qualities desired in a future physician—such as social skills and the ability to “think on your feet.”
Fifty percent of committee members did perceive differences between science majors at liberal arts colleges and comprehensive universities—most favoring students from comprehensive universities because of a diversity of advanced courses, more specialized knowledge of science, and enhanced research opportunities. Further, half also thought that sciences are taught differently in liberal arts colleges. Those perceiving a difference commented that students at liberal arts schools tend to work more often in small groups, receive more individual faculty attention (and less instruction by teaching assistants), and be assigned more research-oriented, collaborative projects (as opposed to large group lectures).
Eleven of 16 members (68.7%) saw no difference in extracurricular involvement between graduates of liberal arts colleges and comprehensive universities. Those noting differences attributed them to the size of the college, stating that it was “easier to be a big fish in a small pond.” However, nine of the 16 (56.2%) respondents thought the institutional size had no impact on the quality of premedical advising received by liberal arts students. Forty-three percent of respondents identified influential factors in a liberal arts education relevant to medical school and future physicians, including: (1) encouragement to interact in small groups; (2) a tendency of nonscience majors to demonstrate empathy and compassion; (3) an enhanced ability to relate to others; and (4) a desirable grounding in history, language arts, philosophy, ethics, political science, economics, and anthropology.
Regarding the actual admission interview, most members did not pose direct questions of applicants concerning their liberal arts education. Indirect questions included reason for choice of schools, how professors evaluate and test subject matter, educational expenses, hobbies, volunteer activities, advice given by premedical advisor, and societal and cultural influences on medical practice. Sixty percent of respondents reported asking interview questions related to the applicant's quality of science or nonscience education. Topics included: (1) preparation from undergraduate courses; (2) research activity and involvement; (3) the relevance of research to medical practice; (4) strengths and weaknesses of their educational program; and (5) their most challenging class, and why.
The second and third research questions addressed characteristics of liberal arts students prior to admission (n = 177) and also as program graduates (n = 190). Compared to their counterparts with undergraduate degrees from comprehensive universities, students from liberal arts colleges did not differ significantly in MCAT performance. Nonliberal arts matriculants did have significantly (p ≤.05) higher total undergraduate GPA (3.61 vs. 3.51), although science and nonscience components did not differ. Regarding instructional method preferences, no statistically significant differences were found – although students with liberal arts backgrounds tended to prefer lecture less (42.9% vs. 49.5%) and group discussion more (62.7% vs. 58.8%) than their nonliberal arts counterparts.
Some statistically significant differences were found among learning preferences. Liberal arts matriculants preferred “problem solving” and “memorizing facts” significantly less than students from larger comprehensive universities – instead preferring to learn by “discussing issues” (t = -2.59, df = 75, p = .01). Among the eight medical specialties examined, only one was preferred significantly more by one group: liberal arts graduates – prior to admission to medical school – expressed significantly greater interest in the practice of psychiatry (t = -2.01, df = 175, p = .04).
Supplemented with data collected at graduation – some four years after these prior measures – we examined a variety of performance indicators. Comparison of class rank at medical school graduation indicated no difference between liberal arts and nonliberal arts graduates. In terms of clinical performance, CPX test data were analyzed along nine different skills – including chief complaint, history-taking, physical exam, communication and interpersonal skills. Making Bonferroni corrections for multiple comparisons, no statistically significant differences were found between student groups.
Lastly, using information on various curricular and extracurricular endeavors supplied by graduates at commencement, we examined level of involvement by calculating an unweighted total of years spent across activities. Looking first at membership in AOA – the national medical honor society – we found a larger (but not statistically significant) proportion of liberal arts students elected as members (20.0% vs. 15.9%). This same trend was observed among recipients of 6 college commencement awards, where a greater percentage of students with liberal arts backgrounds received major commendations (11.1% vs. 6.9%). In terms of total activities during medical school (e.g., student groups, academic/professional organizations, discipline-specific clubs etc.,), no statistically significant difference was found.
In his summary of the Wabash conference proceedings, McGrath notes that the medical school admission deans valued the following cognitive and noncognitive qualities in medical school applicants: (1) fluency of the scientific language; (2) intellectual curiosity and passion for studying an area of interest; (3) research experience contributing to creativity, independence of thought, and love of learning; (4) collaborative learning skills and collegiality; (5) cultural sensitivity; (6) personal integrity and the ability to act ethically; (7) lifelong commitment to learning; and (8) self-confidence based on a clear assessment of one's own strengths and weaknesses.5
While liberal arts colleges undoubtedly assist their graduates in developing many of these aforementioned skills and attitudes, they clearly hold no monopoly in this arena. More importantly, these attributes – while relevant to the learning and practice of medicine – are not reliably and validly assessed in most medical programs, either among applicants or matriculants.
Like participants in the Wabash conference, admission committee members surveyed at the study institution noted few differences between candidates from liberal arts colleges and comprehensive doctoral or master's degree granting universities – in part due to definitional problems. Comments on the questionnaire were telling. “Most of these questions are impossible to answer since they lump all liberal arts colleges together and that's untenable.” “Many of these questions are extremely dependent on the particular college, there is a big difference between colleges.” “I tend to look more at individual and personal experiences as opposed to where those experiences are obtained.”
While committee members may not believe they give extra consideration to students from liberal arts colleges, the data do indicate that those students enter medical school with lower total GPAs. However, there were no overall differences in either MCAT scores or general academic performance within medical school. Because admission committee members have access to mean GPA and MCAT scores of applicants and matriculants from colleges and universities across the nation, they have likely taken institutional grading trends into account in their selection decisions.
With regard to instructional methods, students from liberal arts colleges tended to prefer lecture less and group discussion more than their classmates who attended comprehensive universities. Similarly, in consideration of learning preferences, liberal arts graduates favored “discussing issues” over “memorizing facts” and “problem solving.” These points likely reflect students' familiarity and comfort with different pedagogies in their undergraduate settings. Liberal arts graduates had the opportunity to experience and value the faculty interaction and discourse common in smaller classes using collaborative learning techniques.
Liberal arts graduates were more interested in the practice of psychiatry than their classmates from comprehensive universities. Because psychiatrists must understand the biological, psychological, and social components of illness, they treat the “whole person”10 – assisting in the adaptation of individuals coping with stress, crises, and other problems. As it incorporates aspects of the human condition, the breadth of their undergraduate training—combined with a preference for discussing issues—makes psychiatry an understandable early career preference for liberal arts graduates.
Examination of class rank at graduation and performance on the CPX revealed no differences between these groups of students. However, some differences did appear among the more intangible aspects of extracurricular involvement and awards received. Although not statistically significant, liberal arts graduates were somewhat more likely than graduates from comprehensive universities to be members of AOA and to have received awards during commencement. Such recognition may speak to interests in scholarship, leadership, and service that were sparked or nurtured during their liberal arts undergraduate experiences.
Because this study was conducted at a single public institution that attracts liberal arts graduates from a relatively small regional pool, the generalizability of these findings is limited. Further, additional factors that may have contributed to differences between liberal arts and comprehensive graduates were not addressed, including numbers of undergraduate hours taken in science, social science, and humanities courses, or undergraduate extracurricular activity in service, leadership, or research. Additionally, this study did not assess the quality of the undergraduate experience. Moreover, we recognize that many comprehensive universities are striving to incorporate positive aspects of liberal arts education into their undergraduate programs through such initiatives as honors colleges or topical majors.
While the present study failed to find a strong “liberal arts effect,” we believe trends were evidenced that warrant further study. Therefore, we believe it would be useful to expand this research to incorporate a larger sample of medical schools and liberal arts graduates, looking at a range of preadmission and medical school performance variables. Drawing upon work by Blaich (unpublished manuscript, 2003) it may also be helpful to further refine the operational definition of liberal arts education. As society places more demands on physicians and, in turn, on medical educators, further examining the utility of a liberal arts education would be relevant at both at the undergraduate and medical school levels.
1.Mitchell KJ. Traditional predictors of performance in medical school. Acad Med. 1990;65:149–158.
2.Servinghaus AE, Carman HJ, Cadbury WE, Jr. (Eds.) Preparation for Medical Education in the Liberal Arts College: The Report of the Subcommittee on Preprofessional Education of the Survey of Medical Education. New York: McGraw Hill, 1953.
3.Muller S. (Chair). Physicians for the Twenty-First Century, GPEP Report, Washington, D.C: Association of American Medical Colleges, 1984.
4.Davis-Van Atta D, Carrier SC, Frankfort F. Educating American Scientists. Oberlin, Ohio: Office of the Provost, Oberlin College, 1985.
5.McGrath M. Admission Deans' Roundtable: Medicine as a Liberal Art. The Advisor, March, 2003.
6.Breneman DW. Liberal Arts Colleges: Thriving, Surviving, or Endangered? Washington, D.C: The Brookings Institute, 1994.
7.The Carnegie Foundation for the Advancement of Teaching. Category Definitions. (http://www.carnegiefoundation.org/CIHE2000/defNotes/Definitions.htm
). 2000 Edition. Accessed March 13, 2003.
8.McGrath M. Information about the meeting on medicine and the liberal arts at Wabash College on December 14. Personal communication by e-mail, October 30, 2002.
9.Ghiselli EE, Campbell JP, Zedeck S. Measurement Theory for the Behavioral Sciences. San Francisco, CA: W. H. Freeman, 1981.
10.American Board of Medical Specialists. Which Medical Specialist for You? Evanston, IL: American Board of Medical Specialists. 2000.