Predicting the academic achievement of students bridging to engineering: the role of academic background variables and diagnostic testing

Abstract

Although the number of engineering students is increasing, dropout rates remain high. This problem is also present in the Faculty of Engineering Technology (FET) at KU Leuven, Belgium, which resulted in the need for an in-depth analysis of the academic achievement of the bridging students there. This study examines the contribution of a range of predictors, both cognitive and non-cognitive. The examined predictors are: general characteristics, academic background variables and variables tested in a diagnostic test. A multiple linear regression model for the 2015–2016 chohort accounted for an explained variance of 36% of the students’ academic achievement. After combining three cohorts, we managed to explain 43% of the variance in students’ academic achievement. As expected, the academic background variables are the most important predictors. The diagnostic tests are less predictive but their role is important, since they encourage students to participate in associated interventions.