This article examines the suitability of the Raven Standard Progressive Matrices Test (SPM) for groups of white, coloured, Indian and black pupils in Standard 7 in South Africa. The four groups show very little difference in test reliabilities, the rank order of item difficulties, item discrimination values, and the loadings of items on the first principal component. Consequently, from a psychometric point of view, the SPM is not culturally biased. However, the test is not culture or ethnic ‘blind’ either. This is revealed by discriminant analysis which shows that the largest proportion of the black and white testees follow the general pattern of their own group. (This tendency is far less evident in the coloured and Indian groups.) The groups also show large mean test score differences, especially between black and white pupils where the differences is nearly 3 SD units. Regarding the nature of these differences, it was found that items which discriminate best within the groups were also the items that showed the largest differences between the groups.