Adaptation and validation of the chinese movement assessment battery for children-second edition

BACKGROUND: This study aims to adapt the MABC-2 to the Chinese context, to establish normative scoring standards for children in mainland China, and to evaluate the tool's applicability within this population. METHODS: Data from 321 children aged 7-10 years were standardized using the range transformation method. The raw scores were then converted into standard scores (1-19) based on the percentile distribution to establish a norm conversion table for the Chinese version of the MABC-2. Subsequently, data from another independent sample of 314 children were analyzed to evaluate the psychometric properties. Reliability was assessed using the intraclass correlation coefficient (ICC) for inter-rater reliability, along with the separation and reliability coefficients from many-facet Rasch analysis, the Pearson correlation coefficient for test-retest reliability, and Cronbach's α coefficient for internal consistency. Validity was examined through the item-level content validity index (I-CVI) for content validity, and confirmatory factor analysis (CFA) using indices such as the RMSEA and CFI for construct validity. All analyses were conducted using SPSS, AMOS, RStudio, and Facets software. RESULTS: According to the Kruskal-Wallis test, significant differences were found in the raw scores of the seven test items for manual dexterity, aiming and catching, and balance across different age groups (p < 0.05). After converting the raw data of children aged 7-10 years into standard scores ranging from 1 to 19, a corresponding standard score conversion table for the MABC-2 Chinese version was established. Following the conversion of standard scores for 314 children, a series of reliability and validity analyses were conducted. Item analysis showed that the Pearson correlation coefficients for the eight test items ranged from 0.19 to 0.51, and test-retest reliability ranged from 0.784 to 0.929, indicating strong correlation and stability among the test items. The Cronbach's α coefficient was 0.706, which increased to 0.715 after removing the drawing trail item, demonstrating good internal consistency for the MABC-2 Chinese version. ICC analysis showed high inter-rater consistency, with ICC(2,1) = 0.971 (95% CI [0.94, 0.99], p < 0.001). Multi-faceted Rasch analysis indicated that raters' ratings were consistent, with a reliability coefficient of 0.90, and the differences were significant and reliable. Content validity was assessed with an I-CVI mean of 0.93, and structural validity analysis showed RMSEA = 0.021, CFI = 0.994, confirming good model fit. Discriminant validity was significant (r > 0.4, p < 0.001). After removing the drawing trail item, the model fit and discriminant validity improved further, with the AVE square root for manual dexterity increasing to 0.54, indicating good discriminant validity. CONCLUSION: The MABC-2 Chinese version, constructed based on data, demonstrates good psychometric properties in terms of reliability and validity. The revised scale is applicable for assessing the motor coordination abilities of Chinese children aged 7-10 years. The revised version is valuable in evaluating children's manual dexterity, aiming and catching abilities, and balance. It can serve as a reference for children's motor interventions and training program designs.