Medical school admission test: advantages for students whose parents are medical doctors?

Abstract

Background: Admission candidates especially in medicine do not represent the socio-demographic proportions of the average population: children of parents with an academic background are highly overrepresented, and those with parents who are medical doctors represent quite a large and special group. At Göttingen University Medicine, a new admission procedure was established with the intention to broaden the base of applicants towards including candidates with previous medical training or lower final school grades. With a view to family background, we wished to know whether candidates differ in the test scores in our admission procedure.

Methods: In February 2014 we asked all admission candidates of Göttingen University Medicine by questionnaire (nine closed, four open questions) about the academic background in their families, specifically, the medical background, school exam grades, and previous medical training as well as about how they prepared for the admission test. We also analysed data from admission scores of this group (semi-structured interview and four multiple mini-interviews). In addition to descriptive statistics, we used a Pearson correlation, means comparisons (t-test, analysis of variance), ANOVA, and a Scheffé test.

Results: In February 2014 nearly half of the applicants (44%) at Göttingen University Medicine had a medical background, most frequently, their parents were physicians. This rate is much higher than reported in the literature. Other socio-demographic baseline data did not differ from the percentages given in the literature. Of all applicants, 20% had previous medical training. The group of applicants with parents who were medical doctors did not show any advantage in either test-scoring (MMI and interview), their individual preparation for the admission test, or in receiving or accepting a place at medical school. Candidates with parents who were medical doctors had scored slightly lower in school exam grades.

Conclusion: Our results suggest that there is a self-selection bias as well as a pre-selection for this particular group of applicants. This effect has to be observed during future admission procedures.