Ability of the Test of Variables of Attention (TOVA) to distinguish between referred children with attention-deficit/hyperactivity disorder (ADHD) and other (OTHER) clinical diagnoses were studied. The ADHD group differed from the OTHER group on TOVA variables and most measures from the Revised Conners Teacher Rating Scale (RCTRS) and ADD-H Comprehensive Teacher's Rating Scale (ACTeRS). The criteria of any one TOVA variable > 1.5 standard deviations from age and sex adjusted means correctly identified 80% of the sample with attention deficit disorders and 72% of the sample without attention deficit disorder. Cases misclassified by teacher ratings were often correctly classified by the TOVA and conversely. The TOVA makes a unique and important contribution to diagnostic evaluations.