Reliability and factorial validity of the Canadian Wechsler Intelligence Scale for Children–Fifth Edition

The reliability and factorial validity of the Wechsler Intelligence Scale for Children–Fifth Edition: Canadian (WISC-V) was investigated. The higher-order model preferred by Wechsler (2014b) contained five group factors but lacked discriminant validity. An alternative bifactor model with four group factors and one general factor, akin to the traditional Wechsler model, exhibited the best global fit. The general factor accounted for 33.8% of the total variance and 67.6% of the common variance, but none of the group factors accounted for substantial portions of variance. All together, the general and group factors accounted for 50% of the total variance. Omega reliability coefficients demonstrated that reliable variance of WISC-V factor index scores was primarily due to the general factor, not the group factors. It was concluded that the cumulative weight of reliability and validity evidence suggests that psychologists should focus their interpretive efforts at the general factor level and exercise extreme caution when using group factor scores to make decisions about individuals.