Key points are not available for this paper at this time.
Abstract This study examined the internal validity of the Wisconsin Card Sorting Test in two samples. In an alcohol-dependent sample (n= 151), exploratory factor analysis of internal validity, using Lisrel 8, showed that the six principal scores derived from the test could be fitted to a model with one factor and correlated errors. In a sample of university students (n= 75), we replicated the one factor internal validity model identified in the alcohol-dependent sample, and we also examined the alternate-form reliability of the Wisconsin. The reliability data, together with a reanalysis of information from the most recent edition of the test manual, suggest that the test should not be used for clinical purposes in its present form.
Bowden et al. (Fri,) studied this question.