Predicting aural performance in a tertiary music training programme

Abstract

[Abstract]: Selection of students for entry to tertiary music training programmes is a difficult task traditionally undertaken by painstaking individual auditions, perhaps under highly variable local conditions as the selection team moves from region to region. A number of musical aptitude batteries have been developed but do not appear to have been widely embraced largely because of poor predictive validity. In the present study, a battery of tests designed to assess aural skills, an important component of musical training, was developed by a team of psychologists and musicians at the University of Southern Queensland (USQ). The battery consisted mostly of tests of pitch discrimination and intonation. All students (N=91) currently enrolled in the Music Programme at USQ completed the musical tests. A regression model was then constructed using test scores as independent variables and students' end of semester aural training performance scores as the dependent variable. The resulting model accounted for 36 per cent of the variance in scores on final exam results, a result well beyond chance expectations. Further testing with a revised version of the battery is underway.