Résumé : Aim The aim of this study was to examine the reliability and validity of the Dyskinesia Impairment Scale (DIS). The DIS consists of two subscales: dystonia and choreoathetosis. It measures both phenomena in dyskinetic cerebral palsy (CP). Method Twenty-five participants with dyskinetic CP (17 males; eight females; age range 5-22y; mean age 13y 6mo; SD 5y 4mo), recruited from special schools for children with motor disorders, were included. Exclusion criteria were changes in muscle relaxant medication within the previous 3 months, orthopaedic or neurosurgical interventions within the previous year, and spinal fusion. Interrater reliability was verified by two independent raters. For interrater reliability, intraclass correlation coefficients were assessed. Standard error of measurement, the minimal detectable difference, and Cronbach's alpha for internal consistency were determined. For concurrent validity of the DIS dystonia subscale, the Barry-Albright Dystonia Scale was administered. Results The intraclass correlation coefficient for the total DIS score and the two subscales ranged between 0.91 and 0.98 for interrater reliability. The reliability of the choreoathetosis subscale was found to be higher than that of the dystonia subscale. The standard error of the measurement and minimal detectable difference values were adequate. Cronbach's alpha values ranged from 0.89 to 0.93. Pearson's correlation between the dystonia subscale and Barry-Albright Dystonia Scale was 0.84 (p<0.001). Interpretation Good to excellent reliability and validity were found for the DIS. The DIS may be promising for increasing insights into the natural history of dyskinetic CP and evaluating interventions. Future research on the responsiveness of the DIS is warranted. © The Authors. Developmental Medicine & Child Neurology © 2012 Mac Keith Press.