Individuals with autism spectrum disorder (ASD) are hypothesized to exhibit relative strengths in verbal and non-verbal reasoning and weaknesses in working memory and speed of information processing. The purpose of the present investigation was to determine the degree to which this cognitive profile as measured by the Wechsler Intelligence Scale for Children--Fourth Edition (WISC-IV; Wechsler 2003a) cognitive proficiency index (CPI; measure of working memory and processing speed) and general ability index (GAI; measure of verbal and non-verbal reasoning) could accurately distinguish between a referred sample of 79 school-aged students diagnosed with ASD and two non-clinical comparison groups: (a) 2200 children in the WISC-IV standardization sample and (b) 216 school-aged students referred for psychoeducational testing whose school-based evaluations did not result in a diagnosis. Results indicated that the ASD sample exhibited significantly lower mean scores on the CPI when compared to the two control groups. However, diagnostic utility statistics indicated that a randomly selected participant from the ASD subgroup would exhibit a larger difference between the GAI and CPI than a randomly selected participant from the two control groups 51.9-66.0% of the time. Consequently, the GAI > CPI cognitive score profile exhibits low diagnostic accuracy for individuals with ASD. Psychologists who work in applied settings are cautioned against using group trends to guide decision-making for individual clients.