The objective of this study was to test the interobserver and intraobserver reproducibility of videographic gait analysis in paediatric cerebral palsy patients following botulinum toxin type-A injections. Standardized videos according to gait-specific functional criteria were assessed for 1 h. For intraobserver reproducibility, 13 of 17 items showed good/moderate correlation, one mediocre and three poor correlation. For interobserver reproducibility, 12 of 17 items showed excellent, good or moderate agreement, five items showed poor agreement. Overall, 9 of 17 items showed good intraobserver or interobserver reproducibility. Mean interobserver scores for sagittal plane items were slightly lower than mean intraobserver scores (kappa=0.509 and 0.536, respectively). In conclusion, video analysis of children with cerebral palsy based on functional criteria is useful.