This work contributes to the investigation of a factorial model that is more appropriate for Brazilian children with learning disabilities (LD). 263 WISC-III test protocols of public school students with academic difficulties referred by their teachers for psychological evaluation were analyzed. Statistical techniques of exploratory factor analysis and confirmatory factor analysis were performed. This study, besides corroborating the factor structure defined in the Brazilian standardization, meets the results of the international research for the definition of four-factor model as the best adjusting for the LD population. Although we have also identified two three-factor models as advantageous as to the fit, parsimony and theoretical interpretability, the four-factor structure is the most suitable for clinical interpretation of the scores that express the LD group cognitive abilities, since it allows us to leverage existing standards of WISC-III for the general population.
Wechsler scales; Factor analysis; Validity; School difficulties
