Higher-order factor structures for the WISC-IV: implications for neuropsychological test interpretation.

Area of Science:

• Psychometrics
• Child Psychology
• Educational Assessment

Background:

• Conflicting factor models exist for the Wechsler Intelligence Scale for Children-Fourth Edition (WISC-IV).
• Disagreement persists on whether to interpret WISC-IV results using a general intelligence factor or specific sub-factors.
• Previous research often favored hierarchical or bifactor models for WISC-IV interpretation.

Purpose of the Study:

• To investigate the best-fitting factor model for the WISC-IV using confirmatory factor analysis.
• To re-examine how different WISC-IV models should guide clinical test interpretation.
• To compare the fit of hierarchical, bifactor, and correlated factor models for the WISC-IV.

Main Methods:

• Confirmatory factor analyses were conducted on the WISC-IV standardization sample (N=2,200, ages 6-16).
• The study evaluated model fit for various higher-order factor structures, including hierarchical and bifactor models.
• Model comparisons were performed to determine the most appropriate structure for WISC-IV interpretation.

Main Results:

• Bifactor models for the WISC-IV were challenging to identify.
• The correlated factors model, without a general higher-order factor, demonstrated the best data fit.
• The WISC-IV's basic four-factor structure provided a superior fit compared to hierarchical and bifactor models.

Conclusions:

• The findings support the WISC-IV's technical manual structure, emphasizing a four-factor model.
• Test interpretation should prioritize the specific factor level rather than a general intelligence indicator.
• This research provides evidence for a nuanced approach to WISC-IV interpretation in clinical practice.