DETERMINING THE EFFICACY OF BRIDGING PROGRAMMES IN THE FACULTY OF SCIENCE AT THE UNIVERSITY OF KWAZULU-NATAL

The vast majority of mainly black African students enrolling at a higher education institution come from a township school where a lack of resources and teacher training creates an environment of rote learning where students leave with only a superficial understanding of some of the linguistic and numeracy concepts needed to successfully complete a higher education based field of study. To address this problem universities have put in place additional teaching programmes that are designed to help bridge this gap. This paper examines the efficacy of some of these bridging programmes using regression adjustment and propensity score matching methods to control for a possible selection bias that can occur with observational studies. To control for a possible selection bias that may occur when selection into treatment is being determined by a set of confounding variables that may be unobservable, Heckman's switching regression model was also fitted to the dataset that was collected.