INTRODUCTION Very few studies have reported on the effect of admission tests on medical school dropout.1 Recently Urlings-Strop et al. found the relative risk of dropout to be 2.6 times lower for ‘selected students’ than for ‘lottery admitted controls’.2 The main aim of our study was to evaluate the predictive validity of admission testing versus grade-based admission on dropout. METHOD This prospective cohort study followed 6 cohorts of medical students admitted to the medical school at University of Southern Denmark (USD) in 2002-2007 (N=1544). Half the students were admitted based on highest prior grades (quota 1), while the other half went through a composite non-grade based admission test (quota 2). Educational as well as social predictor variables (doctor parent, origin in the developed world, parenthood, parents live together, parent on benefit, university educated parents) were also examined. The outcome of interest was students’ dropout status 2 years after admission. Multivariate logistic regression analysis was used to model dropout. RESULTS Quota 2 (admission tested) students had a lower relative risk of dropping out of medical school within 2 years of admission compared to quota 1 students (OR=0.56, CI95% [0.39-0.80]). In addition, Danish students without traditional Gymnasium exams had higher relative risks of dropout (OR=2.49, CI95% [1.52-4.07]), as did students who did not assign a first priority to the program on the national admission form (OR= 1.88, CI95% [1.30-2.74]). None of the aforementioned social variables contributed significantly to the dropout model. Interestingly, students composite admission test scores from the quota 2 process did not predict dropout either. DISCUSSION Selection by admission-testing appeared to have an independent, protective effect on dropout in this setting despite the lack of association between admission test scores and dropout. Pre-university grade point averages are merely broad measures of basic academic skills, and not measures of motivation for any particular program. One explanation for our result could be that, there is in fact an independent association between admission-test survival and program retention – a program specific admission test survivability factor - regardless of admission-test content, prior education, and program priority. The generalisability and other important limitations of the results (e.g. missing data, potential misclassifications, omitting potentially relevant explanatory variables, sample size, method of analysis, outcome chosen) will be discussed. CONCLUSION Selection by admission-testing appeared to have an independent, protective effect on dropout in this setting, despite the lack of association between admission test scores and dropout. REFERENCES 1.O’Neill L, Wallstedt B, Eika B, Hartvigsen J. Factors associated with dropout in medical education: a literature review. Med Educ (In press). 2.Urlings-Strop LC, Stijnen T, Themmen APN, Splinter TAW. Selection of medical students: a controlled experiment. Med Educ 2009;43 (2):175–83.
Abstract Book. Amee 2011, 2011, p. 16-17
Main Research Area:
AMEE 2011 - An International Association for Medical Education
Association for Medical Education in Europe (AMEE)