The number of students that apply for studies at South African universities has increased dramatically over the past years, but the resources of the universities have stayed the same. It is therefore imperative that academic performance can be predicted on a short- and long-term basis. The Developing Expertise model might be able to predict academic performance on a short- and long-term basis. This study focussed primarily on academic performance on short-term basis by focussing on three of the model’s elements. The three elements that were investigated were: metacognition, intrinsic motivation and critical thinking. One of the contributions of this study is the refinement of some of the elements of the expertise model. The study helped to determine the validity of the instrument in a South African context. The measurement instrument can also be used as a selection meganism on short-term basis. In order to measure each of the elements of the expertise model, a measuring tool consisting of five questionnaires was compiled. A convenience sample of first year psychology students provided 326 students for the study. A second sample, compiled from the original sample, consisting of 209 first year students was used for the multiple regression analyses. These students are registered for psychology for the first time in 2005. The students were placed in three groups according to the student’s means of registering for the CBT centre. The first group was those students who completed the instrument on computers. The second group consisted of those students who completed a paper and pencil version of the measurement. The third group consisted of the Mamelodi students. These students completed a paper and pencil version of the measurement and brought it back a week later. The statistical analysis included factor analysis, correlation analysis and multiple regression analysis. Based on the literature research and the results of the statistical analysis, the expertise model was expanded to include metamemory as the functional unit for metacognition. Motivation can be divided in three subcomponents, namely: self-efficacy, intrinsic value and test anxiety. The motivational component, knowledge of cognition and external control can be used as predictors of academic performance on a short-term basis. The findings of this study can only be used as a selection meganism of academic performance on a short-term basis and therefore the results must be used with caution. / Dissertation (MA (Psychology))--University of Pretoria, 2007. / Psychology / unrestricted