Does IEB make the grade? Alternative testing methods and Educational outcomes: The case of the IEB in South Africa

According to the Independent Examinations Board (IEB, 2015), students who write the IEB National Senior Certificate school-leaving exam are at a distinct advantage and seem to be better prepared for the pressures and challenges faced during their university years than are those students who wrote the Department of Basic Education (DBE) exams. Although the underlying curriculum is no different, the IEB exam is thought to be more challenging and to encourage more critical thinking and deeper engagement with the material than the DBE exam. Thus, this research paper aims to provide a rigorous investigation of whether those students who write the IEB exam at the end of their matric year achieve higher university grades in their first year of study, as well as a decomposition of this effect into a teaching effect and a testing effect. This is done by exploiting within-school variation of examination boards. Given that studies investigating independent school impacts on university performance have predominantly been conducted internationally (McNabb et al., 2002; Ogg et al., 2009; Smith & Naylor, 2001; Smith & Naylor, 2005), this paper will add to the literature in the South African context. By using the techniques of OLS, quantile regression, binary choice probit models and ordered probit models, this paper attempts to provide a holistic view of the effect that the IEB school-leaving examination has on a student’s academic performance at a tertiary level. The data used in this study is also unique, in that it is made up of an amalgamation of student record data obtained from the University of Cape Town (UCT), as well as governmental survey data. This paper finds that the IEB examination has a strong positive effect of between 1.6 and 6.5 percentage points on first-year GPA at UCT, particularly in the Medicine and Engineering faculties. Furthermore, this effect is present, but decreasing across the entirety of the performance distribution. Students with an IEB matric are significantly more likely to achieve a 2nd class pass or higher at the end of their first year of study than are comparable students from Former African schools. When decomposing the IEB effect into a teaching effect and a testing effect, it was found that the majority of the impact of the IEB comes simply from the different exam, and that teaching effects are minimal. A further finding of interest is that the IEB effect seems to be independent of resource availability, and that simply the exposure to the alternative testing method is sufficient for students to see significant improvements in their university performance. These results are robust to changes in functional form, and provide a strong and clear picture that perhaps South Africa should be adopting more of the IEB policies towards teaching and learning on a national scale.