Douglas Grbic

Medical school mission statements as reflections of institutional identity and educational purpose: a network text analysis.

Purpose A medical school’s mission statement (MS) is an expression of its vision and a reflection of the broader social environment in which it is embedded. The authors examine how the institutional identity of U.S. medical education is projected through the MSs of all U.S. MD-granting medical schools. In addition, the authors examine the extent to which differences between subsets of schools (public versus private; research-intensive versus “social mission”-oriented) are reflected in their respective MSs. Method Whereas existing studies have typically sought to uncover core themes using content-analytic approaches, for this 2011 study the authors combined text analysis with network analysis to elaborate core concepts within larger networks of meaning across MSs. They computed measures of centrality (i.e., density or “connectivity”) for each concept and presented the aggregations of MSs as semantic networks (i.e., representations or maps of the relationships among concepts). Results Analysis across all MSs revealed a dense set of ties among a core set of concepts, indicating a general uniformity among all medical schools; however, the authors identified appreciable differences within subsets of MSs (public versus private and research-intensive versus social-mission-focused). Conclusions MSs of U.S. medical schools are meaningful statements about values and direction. The use of social network text analysis provides an innovative method for understanding the structure of meaning within MSs and, potentially, how these documents might speak to institutional performance outcomes.

Ambiguity tolerance of students matriculating to U.S. medical schools.

Purpose To examine the psychometric adequacy of a tolerance for ambiguity (TFA) scale for use with medical students. Also, to examine the relationship of TFA to a variety of demographic and personal variables in a national sample of entering U.S. medical students. Method The authors used data from the 2013 Association of American Medical Colleges Matriculating Student Questionnaire in which questions on TFA were included for the first time that year. Data from 13,867 entering medical students were analyzed to examine the psychometric properties of the TFA scale. In addition, the relationships of TFA to sex, age, perceived stress, and desire to work in an underserved area were analyzed. Finally, the relationship of TFA to specialty preference was examined. Results The TFA scale was found to be psychometrically adequate for use in a medical student population. TFA was found to be higher in men and in older students. Lower TFA was associated with higher perceived stress levels. Students with higher TFA were more likely to express desire to work in an underserved area. Different levels of TFA may be associated with certain specialty preferences. Conclusions These findings support the assessment of TFA to understand how this personal characteristic may interact with the medical school experience and with specialty choice. Longitudinal work in this area will be critical to increase this understanding.

The Role of Socioeconomic Status in Medical School Admissions: Validation of a Socioeconomic Indicator for Use in Medical School Admissions

Purpose Socioeconomic status (SES) impacts educational opportunities and outcomes which explains, in part, why the majority of medical students come from the upper two quintiles of family income. A two-factor SES indicator based on parental education (E) and occupation (O) has recently been established by the Association of American Medical Colleges (AAMC). This study validates this two-factor indicator as applicable to the national pool of medical school applicants. Method The AAMC SES EO indicator classifies applicants into five ordered groups (EO-1 through EO-5) based on four aggregated categories of parental education and two aggregated categories of occupation. The EO indicator was applied to the 2012 American Medical College Application Service applicant pool. The authors examined the associations that the EO category had with six additional and independent indicators of socioeconomic (dis)advantage, as well as with demographic and educational characteristics and life experiences. Results The EO indicator could be applied to 89% of the 2012 applicants. The lower the EO category, the stronger the association with each of the six indicators of socioeconomic disadvantage. Other notable, but weaker, associations with the EO indicator were differences by age, race/ethnicity, performance on the Medical College Admission Test, community college attendance, and certain self-reported life experiences. Conclusions The EO indicator provides a simple, intuitive, widely applicable, and valid means for identifying applicants from socioeconomically disadvantaged backgrounds. This affords admissions committees an additional factor to consider during the holistic review of applicants in order to further diversify the medical school class.

Dimensions of Diversity and Perception of Having Learned From Individuals From Different Backgrounds: The Particular Importance of Racial Diversity.

Purpose Selective higher education institutions that take race into account in admissions decisions must be able to demonstrate that their policy is justified by a compelling governmental interest, is narrowly tailored, and is the least restrictive means for achieving that interest. The authors thus investigate whether, among medical students, the association between racial diversity (as distinct from other forms of diversity) and learning from individuals from different backgrounds is unique. Method The authors examined six dimensions of diversity, including racial/ethnic diversity, among the 2010, 2011, and 2012 cohorts of fourth-year medical students in the United States. They also examined students’ responses to two Medical Student Graduation Questionnaire items pertaining to learning from individuals from different backgrounds. They modeled the association between each of the school-level dimensions of diversity and the student-level responses to having learned from others with different backgrounds, and they assessed whether associations vary across different groups of students. Results Racial/ethnic diversity is unique in its very strong association with student perceptions of having learned from others who are different. The association between racial/ethnic diversity and student perceptions of having learned from others who are different is especially strong for members of historically underrepresented minority groups. Conclusions Compared with other forms of diversity, racial/ethnic diversity has a unique association with students’ perceptions of learning from others who are different. This association is of particular relevance to admissions and diversity policies in an era of strict scrutiny of these policies.

Non–English-Language Proficiency of Applicants to US Residency Programs

More than 25 million US residents have limited English proficiency, an 80% increase from 1990 to 2010.1 Limited English proficiency (LEP) may impede participation in the English-language–dominant health care system.2 Little is known about the non–English-language skills of physicians in training. This study characterizes the language diversity of all US residency applicants through the Electronic Residency Application Service and contrasts applicant language skills with the predominant languages of the US population with LEP.

Predicting Peer Nominations Among Medical Students: A Social Network Approach.

Purpose Minimal attention has been paid to what factors may predict peer nomination or how peer nominations might exhibit a clustering effect. Focusing on the homophily principle that “birds of a feather flock together,” and using a social network analysis approach, the authors investigated how certain student- and/or school-based factors might predict the likelihood of peer nomination, and the clusters, if any, that occur among those nominations. Method In 2013, the Jefferson Longitudinal Study of Medical Education included a special instrument to evaluate peer nominations. A total of 211 (81%) of 260 graduating medical students from the Sidney Kimmel Medical College responded to the peer nomination question. Data were analyzed using a relational contingency table and an ANOVA density model. Results Although peer nominations did not cluster around gender, age, or class rank, those students within an accelerated program, as well as those entering certain specialties, were more likely to nominate each other. The authors suggest that clerkships in certain specialties, as well as the accelerated program, may provide structured opportunities for students to connect and integrate, and that these opportunities may have an impact on peer nomination. The findings suggest that social network analysis is a useful approach to examine various aspects of peer nomination processes. Conclusions The authors discuss implications regarding harnessing social cohesion within clinical clerkships, the possible development of siloed departmental identity and in-group favoritism, and future research possibilities.

Community College Pathways to Medical School and Family Medicine Residency Training

PURPOSE Medical students who attend community college are more likely to express intention to train in family medicine. This study examined whether community college attendance is associated with family medicine residency training in a national sample of US medical school graduates. METHODS We performed a cross-sectional analysis using the Association of American Medical Colleges matriculant files of residency trainees who graduated from medical school between 2010 to 2012. Residency specialty (family medicine vs other) was modeled using logistic regression analysis; the key independent variable was community college attendance, with categories of non–community college (reference); community college while in high school; community college after graduating from high school, then transfer to 4-year university; and community college after graduating from a 4-year university or as a postbaccalaureate. The logistic model adjusted for age, sex, race/ethnicity, years in medical school, parental education (a marker of socioeconomic status), and high school US region. RESULTS Of the 43,382 medical school graduates studied, 25.9% attended community college and 8.7% trained in family medicine. In unadjusted analysis, graduates attending community college while in high school, after graduating from high school with transfer to 4-year university, or after graduating from a 4-year university or as a postbaccalaureate (12.0%, 12.7%, and 10.8%, respectively) were more likely to train in family medicine compared with their peers who did not attend community college (7.7%). Respective adjusted odds ratios were 1.47 (95% CI, 1.33-1.63; P <.001), 1.27 (95% CI, 1.06-1.52; P = .009), and 1.17 (95% CI, 1.06-2.29; P = .002). Among family medicine residents, 32.7% of those who were white, 35.2% of those Asian, 50.8% of those Latino, and 32.7% of those black or African American attended community college. CONCLUSIONS US medical school graduates who attended community college were more likely to train in family medicine, suggesting community college is an important pathway for increasing the primary care workforce.

First year medical student peer nominations of professionalism: A methodological detective story about making sense of non-sense: Medical Student Peer Nominations

This article explores the assessment of professionalism within a cohort of medical students during a sequential 13‐week medical school histology and anatomy course. Across seven data points, students were asked to identify a professionalism role model from amongst their peers and to score Likert‐structured rationales for their decision. Based on density scores, an initial social network analysis identified six peer‐nomination “stars.” However, analysis of these stars revealed considerable variability and random‐like “noise” in both the nomination and explanation data sets. Subsequent analyses of both data sets explored the possibility of underlying patterns in this noise using tests of reliability, principal components factor analysis, and fixed‐effects regression analysis. These explorations revealed the presence of two dimensions (professional vs. supportive) in how students sought to explain their nomination decisions. Although data variability remained quite high, significantly less variability was present in the professional than in the supportive dimension, suggesting that academic helpfulness rationales are both empirically distinct and more mutable than rationales grounded in professionalism‐related factors. In addition, data showed that the greater the stability in ones choice of a professionalism role model nomination over the T1–T7 data periods, the more stable ones reasons for that nomination—both for professionalism and supportive dimensions. Results indicate that while peer assessment of professionalism by first‐year medical students may not be very reliable, students can differentiate between more personal and professional factors, even at this early stage in their professional development. Formal instruction within the pre‐clinical curriculum should recognize and address this distinction. Anat Sci Educ.