Objective:To evaluate the reliability and validity of the Chinese version of the Edinburgh visual gait score (EVGS-CN) for children with cerebral palsy.Methods:The EVGS-CN was established following international guidelines for translation and cross-cultural validation of health status questionnaires. Videos of 30 children with cerebral palsy were assessed independently by six raters (with different levels of experience in gait analysis) using the EVGS-CN. Inter- and intra- observer reliability were evaluated using intraclass correlation coefficients (ICCs). The correlation analysis and group comparison were used to test the technique′s criteria-related validity, convergent validity, and discriminant validity.Results:The ICC values of the 17 items in the EVGS-CN ranged from 0.20 to 0.87 for inter-observer reliability, and from 0.41 to 0.90 for intra-observer reliability. Most items showed good inter- and intra-observer reliability among experienced raters, but only a moderate level when used by inexperienced raters. The EVGS-CN results were strongly correlated with those of physician rating scale (PRS) ( r=0.77, P≤0.001) and observational gait scale (OGS) ( r=-0.85, P≤0.001), moderately correlated with the total gross motor function measure-D/E (GMFM-D/E) score ( r=-0.55, P≤0.01), and strongly correlated with 10MWT times ( r=-0.69, P≤0.001) and timed up and go (TUG) times ( r=0.60, P≤0.001). Moreover, significant differences in average EVGS score were found between different gross motor function classification system (GMFCS) levels and between affected limbs on different sides. Conclusion:The EVGS-CN demonstrates satisfactory reliability and validity in evaluating children with cerebral palsy when it is used by an experienced or inexperienced rater.