Abstract
Seven hierarchical clustering methods were applied to the WAIS-R scaled scores of 121 medical rehabilitation clients. This process was employed to investigate the possibility of method-dependent results and to determine the stability of the clusters. The WAIS-R number-of-factors issue was addressed by examining the agglom-eration procedure when data were at a three-cluster stage, followed by a two-cluster stage. Results were found to vary with the method employed; a gross difference in cluster structures was obtained from three of the seven methods. The traditional two-factor interpretation best fits the data. However, as each of the one-, two-, and three-factor solutions could have been interpreted with differing degrees of confidence, especially in light of the small sample size, it would appear that the underlying constructs are not well-separated. Caution in the use of isolated measures and in the acceptance of one WAIS-R factor solution over another is advised.
Get full access to this article
View all access options for this article.
