Abstract: Background：Based on consensusbased standards for the selection of health measurement instruments（COSMIN）, we evaluated the psychometric properties of the evaluation tools for topics in different domains of the clinical practice guideline for children and adolescents with cerebral palsy (CANDLE) and established the inclusion and exclusion criteria of evaluation tools in CANDLE. Objective：To systematically evaluate Gross Motor Function Measure88 and 66 (GMFM88 and GMFM66) based on COSMIN methods, clarify the evidence level of each psychometric property, provide evidence for its clinical practice, and explore the value of COSMIN methods in assessing field observation evaluation tools. Design：A systematic review. Methods：MEDLINE, Embase and SinoMed databases were searched for relevant research literature on GMFM88 and GMFM66 measurement properties. The updated COSMIN Risk of Bias checklist was used to evaluate the methodological quality of all measurement properties, and updated criteria for good measurement properties were used to score each measurement property study. The GRADE method of COSMIN revision was adopted to assess the evidence quality. Main outcome measures：The content validity of GMFM88 and GMFM66. Results：GMFM88 had better evidence for content validity than GMFM66, and both of them have highquality reliability research. GMFM88 had more evidence to support its internal consistency and measurement errors than GMFM66, but it lacked structural validity research. GMFM88/66 had highquality evidence in concurrent validity and concurrent responsiveness, and low or very lowquality evidence in discrimination validity and discrimination responsiveness. GMFM66 showed better criterion responsiveness than GMFM88. Conclusion：GMFM88/66 has highquality evidence in reliability, internal consistency, and construct validity. COSMIN methods could be useful for the assessment of field observation evaluation tools.

Key words: Gross motor function measurement, Consensus-based standards for the selection of health measurement instruments, Psychometric properties, Content validity