Die voorspelling van die akademiese prestasie van alternatiewelik toegelate studente

English: Several considerations in post-apartheid South Africa compel historically white universities in particular, to admit more black students. Because of inequalities in the education system, the majority of black students have not been given an equitable opportunity to prepare for university studies. As academic failure can lead to large-scale financial and personal losses, it is essential to maximise these students' chances for academic success. Universities try to achieve the latter objective by employing two strategies. Firstly, selection methods are used to admit only those educationally disadvantaged applicants with sufficient prospects for academic success. Secondly, bridging and support programmes are implemented in an attempt to eliminate the academic drawback of educationally disadvantaged students. In view of the above, the University of the Orange Free State (UOFS) has created alternative admission channels for educationally disadvantaged students. For example, in some cases students with matriculation exemption, but lacking the minimum prescribed matriculation performance, are admitted conditionally to the university on the basis of a psychometric evaluation and a counselling interview. In the current study these students were referred to as discretionarily admitted students. The support interventions for such discretionary admissions in their first year are less intensive than an extensive bridging programme. Doubt exists about the validity of traditional predictors (Le. matriculation performance and aptitude tests) for predicting academic performance of educationally disadvantaged students. Noncognitive predictors may be more valid in this regard. However, South African research on both the prediction of the academic performance of educationally disadvantaged students and the effectiveness of academic bridging and support programmes is limited. The research results that are available, produce contradictory findings. In the first place this dissertation dealt with the prediction of academic performance of discretionarily admitted students on the basis of their matriculation performance, scholastic aptitude, English reading understanding, personality traits, interests, faculty of registration, gender and age. Secondly, the effectiveness of the interventions to improve their academic performance was studied. The first research question was studied by combining all the predictors simultaneously in a standard multiple regression equation. Apart from the interest field of sociability, none of the predictors (that is, including the traditional predictors) correlated statistically significantly (at the 5% level) with the first-year performance of discretionary admissions. The value of the multiple correlation coefficient (R = 0.58) however did not deviate so much from the (simple) correlations found typically between the matriculation and university performance of educationally non-disadvantaged students. In spite of the shortcomings of this study, it would appear that the use of traditional predictors for selection decisions on discretionarily admitted students are under suspicion. The sociability score of these students apparently do have prediction value. The second research question was studied by comparing the university performance of discretionary admissions with the performance of a group of directly admitted black students as well as two other groups of alternatively admitted black students. This was done by means of co-variance analyses (on the basis of hierarchical multiple regression analyses) of academic performance in the first, second and third years of study in which matriculation performance was controlled as co-variant. In all three years of study, group membership explained significant amounts of the variance in university performance in addition to the variance already explained by matriculation performance. The results of this study therefore suggest that after differences in previous academic performance was taken into account statistically, the support interventions that discretionary admissions are subjected to might be effective, but not for all such students, particularly not in the second and third year