Objective:To test the reliability and validity of a multi-dimensional, quantitative video evaluation system (the Multi-dimensional system) for evaluating hand dysfunction among stroke survivors so as to provide an objective basis for its clinical application.Methods:Sixty stroke survivors with single dysfunctional hand were evaluated using the Multi-dimensional system and also using the Fugl-Meyer Upper Extremity Assessment (FMA-UE) and the Action Research Arm Test (ARAT) and in terms of their active range of hand motion (AROM-H). The Multi-dimensional system′s internal consistency, composite reliability, split-half reliability, parallel-forms reliability as well as intra- and inter-observer reliability were quantified. Its validity was tested in terms of content validity, structural validity, convergent validity and criterion validity.Results:The system′s Cronbach′s α was determined to be 0.86. Its one-dimension composite reliability was 0.939, its split-half reliability coefficient was 0.88 and its parallel-forms reliability was 0.922. The inter-evaluator reliability was between 0.965 and 0.998 [95%CI= (0.919, 0.999)], and the single evaluator ICC value was 0.973 to 0.998 [95%CI= (0.937, 0.999)]. The validity test produced a content validity index of 1, and the Pearson correlation coefficients between each sub-item and the total score were all greater than 0.60. The KMO value for structural validity was 0.882, and there was a common factor in the component matrix which reflected 60.9% of the information in the original variables. The system′s convergent validity AVE was 0.609 with Pearson correlation coefficients between the multidimensional system and the FMA-UE and ARAT both > 0.70.Conclusion:The multi-dimensional quantitative video evaluation system has good reliability and validity in the evaluation of stroke survivors with hand dysfunction.