Invariance of the Measurement Model Underlying the Wechsler Adult Intelligence Scale-IV in the United States and Canada

Abstract

A measurement model describes both the numerical and theoretical relationship between observed scores and the corresponding latent variables or constructs. Testing a measurement model across groups is required to determine if the tests scores are tapping the same constructs so that the same meaning can be ascribed to the scores. Contemporary tests of intelligence describe a number of closely related cognitive abilities, each ability being sampled by a set of observed scores. This study examined the invariance of the measurement model underlying the Wechsler Adult Intelligence Scale—IV (WAIS-IV) in the U.S. and the Canadian standardization samples. The model satisfied the assumption of invariance across samples with subtest scores reflecting similar construct measurement in both samples. Consistent with previous research with the WAIS-III, slightly higher latent variable means were found in the Canadian WAIS-IV normative sample. The results demonstrate the generality of construct validity in measurement of cognitive abilities across U.S. and Canadian samples and highlight the importance of local norms.