A preliminary investigation of the suitability of the WAIS-III for Afrikaans-speaking South Africans

Abstract

The standardised version of the Wechsler Adult Intelligence Scale-III (WAIS-III) for English-speaking South Africans is currently widely used. Although an Afrikaans translation of the verbal subtests of the WAIS-III is provided in the manual, this translation has not been standardised. This preliminary investigation of the suitability of the translation for use with Afrikaans-speaking South Africans indicated comparability of the performance of an urbanised sample with good quality education and similar groups tested with the English version. The results also revealed significant differences between the urbanised Afrikaans speakers and a relatively non-urbanised group of Afrikaans speakers with inadequate quality education. There were no significant relationships of performance with gender and age. University graduates/students obtained higher scores than college graduates/ students and subjects without tertiary education. Multivariate analyses of the contribution of demographic variables to variations in scores confirmed the significant effect of quality of education on test performance. The reliabilities for the verbal subtests ranged from acceptable to good, and intersubtest correlations were at the expected level. An analysis of item difficulty revealed the expected sequence of progressively more difficult items, but with some exceptions. While some scoring criteria may need to be re-examined, inconsistencies also appear to be related to the role of schooling and social background. Limitations of the study pertain to the design and the small sample size that do not allow for conclusive evidence in terms of construct validity and in terms of measurement equivalence between the Afrikaans and English versions.