Historically, difference in performance across dominant and subordinate groups on tests of cognitive ability have been observed. Of particular importance is the underperformance of groups on tests. It is proposed that this results from the underlying assumptions of conventional tests. Contextual models (which argue that cognitive ability is socially determined) and research pertaining to Piagetian theory argue that understanding children's socio-ecological contexts, as well as their underlying cognitive processes, enhances assessment of cognitive ability/ competence. Additionally, the multitude of factors influencing performance in an intra- and intercultural assessment situation should be considered. A new test of cognitive ability based on Piagetian Genetic Epistemology, the Counter test, has been developed by Dr V. Grover. Previous exploratory research has indicated performance differences across designated racial groups attending different education systems. This research focuses on understanding performance (on the Counter Test) across "black" and "coloured" groups within the same education system. A fortuitous sampling technique was used to select a sample consisting of 20 "black" and 20 "coloured" children, aged 8 and 9 years. Demographic data was obtained to understand their socio-ecological context. They were administered the Draw-A-Man and Counter Test (first administration). Results obtained on the Counter Test indicated that designated racial group does not influence performance on the test (1 = 0.203, d.£ = 38, Q > 0.01) or significant underperformance within each group (1 = 6.901 "coloured", 1 = 9.68 "black", d.£= 19, l2. < 0.01). The latter is similar to findings on previous research. Possible explanations are given, but further vigorous investigations are indicated. On second administration, the unstandardized, structured steps based on Feuerstein's Learning Potential Assessment Device were administered to 14 children - criterion underperformance from 1 yr. 6mo to 2 yr. 9mo on the Counter test. The clinical interpretation of 6 children's performance across administrations enhanced understanding of their cognitive processes and the contingencies affecting their use and indicated overall improvement in performance. This indicates that "actual" competence can be assessed through more appropriate assessment procedures. Implications for assessment in a multi-cultural context are considered.