BACKGROUND: Caregivers' feeding practices significantly influence children's eating behaviours. We previously developed a Caregivers' Complementary Feeding Practice Scale (CCFPS) for children aged 6-23 months in China. We aimed to comprehensively assess the CCFPS's reliability and validity, thus establishing a reliable and robust tool for evaluating caregivers' feeding practices. METHODS: Through convenience sampling, we recruited 928 caregivers from community health centres. We assessed the scale's reliability with Cronbach's alpha and split-half analysis. Additionally, we used the content validity index (CVI) for content validity. Lastly, we examined structural validity with exploratory factor analysis (EFA) and confirmatory factor analysis (CFA). RESULTS: The scale's Cronbach's alpha coefficient was 0.783, and its split-half reliability coefficient reached 0.750. The CVI was 0.910. EFA supported the scale's six-factor structure, which explained 51.43% of the total variance. The CFA results demonstrated that the fit indices of the model were higher than the acceptable level, indicating the scale provides a good level of validity. CONCLUSIONS: The CCFPS is a reliable and valid tool for assessing caregivers' feeding practices with infants and young children aged 6 to 23 months in urban China, suitable for research and clinical applications.