Abstract
Six hierarchical clustering methods were applied to subtest intercorrelations for each of the nine age groups in the WAIS-R standardization sample. The multiple-methods approach was adopted to overcome method-dependent results and to challenge the stability of clusters across methods. The WAIS-R number-of-factors issue was addressed by observing the agglomeration procedure when data were at a three-cluster solution, followed by a two-cluster solution. The 114 clustering results were interpreted by(a) summarizing cluster membership of subtests across age groups and methods and (b) quantitatively comparing obtained results with results expected by the historical two- and three-factor solutions for the WAIS-R. Varying degrees of support were found for "g," two-, and three-factors. Results of this hierarchical examination of the WAIS-R constructs, coupled with the history of evidence for all three structures of the WAIS-R, points to the possibility of three simultaneously, plausible solutions.
Get full access to this article
View all access options for this article.
