Regular physical activity has been shown to help improve mental health, quality of life, and well-being. Physical inactivity among children and adolescents is becoming a major global public health problem. China's policy emphasizes the need to improve the physical health level of adolescents, and emphasizes the construction of an evaluation system with multi-dimensional content and diverse methods. The development and application of youth sports health literacy (SHL) assessment tools will further promote the in-depth development of physical health education for middle school students.

To develop and verify the reliability and validity of the SHL questionnaire for middle school students in China based on the Rasch analysis of classical test theory (CTT) and item response theory (IRT), and to explore the form and appropriateness of different levels and categories of adolescent SHL measurement items.

Based on the theoretical structure of health literacy, a questionnaire was developed based on the actual situation of Chinese adolescents through policy combing, literature research and expert interviews. In June to July 2022, a total of 1 651 middle school students in urban and suburban counties in four provinces and cities with different levels of economic development were selected for on-site testing. The quality analysis of the SHL questionnaire was carried out by combining CTT and Rasch model analysis, including one-dimensional test, individual and item reliability and discrimination analysis, model fit test, difficulty analysis and item-specific fit index analysis, internal consistency reliability, Spearman correlation analysis and hierarchical exploratory factor analysis.

A total of 1 651 valid samples were included, with an effective recovery rate of 99.88%. The single-dimensional test of motor knowledge and skill dimension of the classification items of the SHL questionnaireⅡwas basically passed, and the reliability of the items and participants was >0.7, and the separation degree was >1.5. The difficulty distribution of the White plot entries follows a normal distribution, but the upper entry does not correspond to it. Three items were poorly fitted, and the overall fit was good. The reliability of the three dimensions of scale items was good, Cronbach's α coefficients were 0.943, 0.878 and 0.921, respectively, and the overall Cronbach's α coefficient was 0.956. Exploratory factor analysis extracted 6, 5 and 3 factors in each dimension, and finally the behavior dimension did not delete the entries, the motivation dimension deleted 5 items, and the influencing factor dimension deleted 3 items to form a questionnaire.

The theoretical framework of the middle school students' SHL questionnaire is scientific and reasonable, and the quality analysis results of the questionnaire based on CTT and Rasch model are good, which has good pertinence and practical significance, and the quality evaluation of the revised questionnaire needs further research.