Students coached for an admission test perform less well throughout a medical course

Background: Undertaking commercial coaching to improve one's chance of selection into medical school is widespread. Although its effect on selection test performance appears to be relatively minimal, its impact on the predictive validity of the tests is unknown. Aims: To examine whether commercial coaching for the Undergraduate Medical and Health Sciences Admissions Test (UMAT) changes its ability to predict the subsequent academic performance of medical students. Methods: The first two cohorts to enrol in a new Australian medical school provided information at the time of their selection interview about whether or not they had undertaken a commercial coaching course to help prepare for the UMAT. Final academic grades for each year of the degree and overall grade point average (GPA) of coached students were compared with those of non-coached students. Moderated regression analyses examined differences in the relationship between UMAT scores and examination results while controlling for entry UMAT scores and past academic performance. Results: Coached students had a lower GPA than those who were not coached. In cohort 1, coached students performed more poorly than non-coached students in every year of their degree. This effect, while similar, was not statistically significant in cohort 2. Conclusions: Differences in selection process and learning context between the two cohorts may explain why coaching was only significantly related to the performance of one cohort. Further research is required to ascertain if coached students develop a learning style that hinders ongoing acquisition of knowledge, which might have serious implications for job performance after graduation.