Objective To translate the Autism Eating Behavior Questionnaire (AEQ) for children with autism spectrum disorder (ASD) into Chinese and evaluate its reliability and validity.

Methods Following the Brislin translation model, the AEQ was translated through forward translation, back translation, cross-cultural adaptation, and pretesting to develop the Chinese version of the AEQ. Using purposive sampling, parents of 273 children with ASD who visited the Third Affiliated Hospital of Sun Yat-sen University from March to May 2026 were recruited for the survey, and the reliability and validity of the questionnaire were evaluated.

Results The Chinese version of the AEQ consisted of 32 items across six common factors. The item-level content validity index ranged from 0.813 to 1.000, and the scale-level content validity index was 0.965. Exploratory factor analysis extracted six common factors, accounting for 65.172% of the cumulative variance. The six factors were named oral motor/texture restriction, food refusal, food selectivity, ritualistic eating, overeating, and emotional eating, respectively. Confirmatory factor analysis showed that all indicators reached acceptable standards (χ²/df = 2.441; the root mean square error of approximation (RMSEA) was 0.073 ＜ 0.1; the comparative fit index (CFI) was 0.872 ＞ 0.8; the goodness-of-fit index (GFI) was 0.802 ＞ 0.8; the incremental fit index (IFI) was 0.873 ＞ 0.8; and the Tucker-Lewis index (TLI) was 0.859 ＞ 0.8). The overall Cronbach’s α coefficient of the questionnaire was 0.922, with a test-retest reliability coefficient of 0.649 and a split-half reliability coefficient of 0.926.

Conclusion The Chinese version of the AEQ demonstrates good reliability and validity and provides a scientifically reliable reference tool for the early identification of eating behavior risks among children with ASD in clinical settings in China.