The Fear Survey Schedule for Children-Revised (FSSC-R), Revised-Children's Manifest Anxiety Scale (RCMAS), and the Modified State-Trait Anxiety Inventory for Children (STAIC-M) are three widely used self-report measures of childhood anxiety. While previous studies have established the reliability of these measures, their validity in discriminating anxious from non-anxious youngsters remains to be established. The present study examines the discriminant validity of the three measures by comparing clinic referred samples of boys with an anxiety disorder (n = 105) or ADHD (n = 59) with a community sample of never-psychiatrically-ill boys (n = 49). Results indicated that the two patient groups differed significantly from the never-psychiatrically-ill group on the RCMAS and STAIC-M, but the anxious and ADHD groups did not differ from each other. None of the three groups differed on the FSSC-R. The implications of these findings for the assessment of childhood anxiety disorders are discussed.