Measurement Invariance of the WISC-V across a Clinical Sample of Children and Adolescents with ADHD and a Matched Control Group

Abstract

Measurement invariance of the Wechsler Intelligence Scale for Children, Fifth Edition (WISC-V) 10-primary subtest battery was analyzed across a group of children and adolescents with ADHD (n = 91) and a control group (n = 91) matched by sex, age, migration background, and parental education or type of school. First, confirmatory factor analyses (CFAs) were performed to establish the model fit for the WISC-V second-order five-factor model in each group. A sufficiently good fit of the model was found for the data in both groups. Subsequently, multigroup confirmatory factor analyses (MGCFAs) were conducted to test for measurement invariance across the ADHD and control group. Results of these analyses indicated configural and metric invariance but did not support full scalar invariance. However, after relaxing equality constraints on the Vocabulary (VC), Digit Span (DS), Coding (CD), Symbol Search (SS), and Picture Span (PS) subtest intercepts as well as on the intercepts of the first-order factors Working Memory (WM) and Processing Speed (PS), partial scalar invariance could be obtained. Furthermore, model-based reliability coefficients indicated that the WISC-V provides a more precise measurement of general intelligence (e.g., represented by the Full-Scale IQ, FSIQ) than it does for cognitive subdomains (e.g., represented by the WISC-V indexes). Group comparisons revealed that the ADHD group scored significantly lower than the control group on four primary subtests, thus achieving significantly lower scores on the corresponding primary indexes and the FSIQ. Given that measurement invariance across the ADHD and the control group could not be fully confirmed for the German WISC-V, clinical interpretations based on the WISC-V primary indexes are limited and should only be made with great caution regarding the cognitive profiles of children and adolescents with ADHD.

Keywords: ADHD; Wechsler Intelligence Scale for Children; clinical sample; fifth edition; measurement invariance.

Figure 1

The WISC-V second-order five-factor model for the ADHD and control group including standardized estimations of all regression weights, subtest intercepts, and intercepts of the first-order factors for the 10 WISC-V primary subtests (M₄ in Table 3). Note. Second-order factor: g = general intelligence. First-order factors: VC = Verbal Comprehension, VS = Visual Spatial, FR = Fluid Reasoning, WM = Working Memory, PS = Processing Speed. Subtest indicators: SI = Similarities, VC = Vocabulary, BD = Block Design, VP = Visual Puzzles, MR = Matrix Reasoning, FW = Figure Weights, DS = Digit Span, PS = Picture Span, CD = Coding, SS = Symbol Search. All standardized parameter estimates are significant at p < .001.

Figure 2

The WISC-V second-order five-factor model for the ADHD and control group including standardized estimations of all regression weights, subtest intercepts, and intercepts of the first-order factors for the 10 WISC-V primary subtests (M₅^† in Table 3). Note. Second-order factor: g = General Intelligence. First-order factors: VC = Verbal Comprehension, VS = Visual Spatial, FR = Fluid Reasoning, WM = Working Memory, PS = Processing Speed. Subtest indicators: SI = Similarities, VC = Vocabulary, BD = Block Design, VP = Visual Puzzles, MR = Matrix Reasoning, FW = Figure Weights, DS = Digit Span, PS = Picture Span, CD = Coding, SS = Symbol Search. All standardized parameter estimates are significant at p < .001.

Figure 3

The WISC-V second-order five-factor model including standardized estimations of all regression weights, subtest intercepts, and intercepts of the first-order factors for the 10 WISC-V primary subtests (M₆^†† in Table 3). Note. Second-order factor: g = General Intelligence. First-order factors: VC = Verbal Comprehension, VS = Visual Spatial, FR = Fluid Reasoning, WM = Working Memory, PS = Processing Speed. Subtest indicators: SI = Similarities, VC = Vocabulary, BD = Block Design, VP = Visual Puzzles, MR = Matrix Reasoning, FW = Figure Weights, DS = Digit Span, PS = Picture Span, CD = Coding, SS = Symbol Search. RMSEA = root mean square error of approximation, CFI = comparative fit index. All standardized parameter estimates are significant at p < .001.