The purpose of this investigation was to define the psychometric properties of the Anxiety Rating for Children - Revised (ARC-R), a clinician rating scale for the assessment of anxiety symptoms in children and adolescents. The ARC-R is comprised of an Anxiety subscale and a Physiological subscale. In a clinical sample (N = 22), the test-retest reliability and interrater reliability were investigated. In a nonoverlapping clinic sample of school refusers (N = 199), the internal reliability and convergent, divergent, and discriminant validity were evaluated. Test-retest (r = .93) and interrater reliability (r = .95) were excellent. There was good internal reliability of items (Cronbach's alpha = .80). The Anxiety subscale of the ARC-R correlated somewhat higher with self-report anxiety instruments (r = .62) than with self-report depression instruments (r = .54-.56). The Anxiety subscale of the instrument discriminated between children with and without an anxiety disorder. This new instrument was shown to be a reliable measure of clinician-rated anxiety. The validity of the ARC-R Anxiety subscale was demonstrated. The potential uses of this instrument for clinical and research purposes are highlighted.