Abstract
<p>The latent structure of the Wechsler Adult Intelligence Scale-Fifth Edition (WAIS-5) is the subject of recent psychometric debate. Although the publisher's initial confirmatory factor analysis (CFA) supported a theoretically derived five-factor model, a recent independent evaluation of the US standardisation sample found that complementary exploratory and confirmatory analyses favoured a hierarchical four-factor representation. The present study examined whether these findings could be independently replicated using the published WAIS-5 Australasian standardisation sample (N = 677).Exploratory factor analyses were conducted using principal axis factoring with Promax rotation, followed by Schmid-Leiman transformation and model-based reliability and dimensionality analyses. Confirmatory analyses compared competing oblique, higher-order, and bifactor models evaluated in the US sample.Factor-retention procedures did not support the proposed five-factor structure. Exploratory analyses consistently recovered four interpretable factors corresponding to Verbal Comprehension, Perceptual Reasoning, Working Memory, and Processing Speed, with Visual Spatial and Fluid Reasoning failing to emerge as distinct latent constructs. Confirmatory analyses identified the same higher-order (4h3) and bifactor (4b3) models as the preferred representations of the data. Model-based reliability analyses indicated a dominant general factor with smaller contributions from the broad cognitive abilities. Despite modest differences in model parameters, the central conclusions closely replicated those of the US sample.</p>