Accuracy of Forecast of Hypothetical Admission Criteria to Medical Studies

Based on a prospective study on the factors of success in medical studies we calculated the effects of a hypothetical access limitation by means of school achievements. We included the marks in Mathematics, English, Physics and Chemistry and the number of class repetitions. By means of discriminant analysis we created three models of a hypothetical numerus clausus: by including all marks into the model a proper and most unerring model could be achieved while the least percentage of false negative students was achieved in a model including the marks in Mathematics and English and the number of class repetitions. But this approach would cause very high costs because it also would include the highest rate of false positive students which – in reality – did not succeed in the decisive end-of-the-year examination. Furthermore additional predictive factors (sex and mothertongue) – which are not suited to serve as access criteria – were tested and their influence on predictability is presented.

25.02.2007 | Lukas MITTERAUER, Gerhard HAIDINGER & Oskar FRISCHENSCHLAGER (Wien)