摘要：Recent research in medical education suggests that program specific admission testing could have a protective effect against early dropout. Little is known about the effect of program specific admission testing on dropout in other areas of higher education. The aim of this paper was to examine if admission strategy was also independently associated with dropout for sports science students in a university setting. The study design was a prospective cohort study with a 2 year follow-up. The population was 449 sports science students admitted to a university in the years 2002-2007. The analysis used was multivariate logistic regression and the predictors examined were: admission group (grade-based or admission tested) as well as educational and socio-demographic variables. The outcome was dropout within 2 years of study start. Admission testing offered superior protection against dropout compared to grade-based admission. This result may fit with elements of previous dropout theory, student-environment fit theory and perhaps also with self-efficacy theory. Nyere forskning inden for medicinsk uddannelse indikerer at uddannelsesspecifikke optagelsesprøver kan have en beskyttende effekt i forhold til tidligt studiefrafald, men for andre universitetsuddannelser end Medicin synes denne sammenhæng endnu ikke at være blevet grundigt belyst. Formålet med dette studie var derfor at undersøge, om optagelsesprøver også beskyttede mod tidligt frafald blandt idrætsstuderende på universitetet. Studiedesignet var et prospektivt kohortestudie med to års opfølgning. Populationen var 449 idrætsstuderende, som blev optaget på Syddansk Universitet i årene 2002-2007. Data blev analyseret med multivariat logistisk regression, og følgende typer af prædiktorer for frafald blev undersøgt: Optagelseskvote (kvote 1 eller kvote 2 udprøvede), andre uddannelsesrelaterede variable samt udvalgte socio-demografiske variable. Effektmålet var studiestatus (frafaldet/ikke-frafaldet) to år efter studiestart. Vi fandt, at optagelsesprøver (kvote 2 udprøvning) virkede beskyttende i forhold til tidligt studiefrafald på idrætsstudiet sammenlignet med karakterbaseret optagelse (kvote 1 optagelse). Dette resultat kunne være i overensstemmelse med dele af allerede eksisterende frafaldsteorier, student-environment fit teori og måske også med self-efficacy teori.
