Short forms of the Wisconsin Card Sorting Test (WCST) have been developed and studied in adult populations, however studies addressing their use in children are lacking. This study compared the full WCST to two short forms in a sample of 174 school-age children who were referred for neuropsychological evaluation. Multiple regression was used to predict standard scores (SSs) on the full WCST. Percent scores were obtained, and normative data from the WCST manual was then used to obtain SSs. We found that scores from the short forms were significantly correlated with corresponding scores on the full WCST, however a high proportion of children obtained short form SSs, which differed significantly from the SSs obtained on the full WCST. It is recommended that clinicians use the full WCST with children, unless at least four categories are reached in the first deck.