AIM: To develop the 17-item Asthenopia Survey Questionnaire (ASQ)-17 by Rasch analysis, and to generate a predictiveness score. METHODS: Totally 739 participants were recruited and 680 were involved in the result analysis in this prospective, cross-sectional study. Three rounds of Rasch analysis were used to analyze the psychometric characteristics of items and options. RESULTS: Phase 1 assessed the original ASQ-19, adjusted the item scoring mode to a four-point Likert response rating scale and combined the 18th and 19th items into a new item. Phase 2 deleted the 11th item. Phases 3 and 4 assessed the new ASQ-17. All the evaluation indexes of ASQ-17 were acceptable. The Infit and Outfit MnSq values of items were 0.67-1.48, the variance explained by the principal component and the unexplained variance explained by the first contrast were 53.90%-59.40% and 1.50-1.80 in three dimensions. The curve peaks of scores in each dimension were separated and in the same order. The PSR and PSI values were 2.80 and 0.89, respectively. The mean scores of dimensions A (9.5±4.1 vs 3.5±3.2), B (7.3±3.3 vs 2.5±2.7), C (4.3±2.2 vs 1.4±2.0) and total (21.1±8.1 vs 7.4±7.0) in asthenopia participants were significantly higher than those without asthenopia (all P<0.001). The area under the curve in two groups was 0.899 (P<0.001). Youden’s index was up to the maximum value of 0.784 when the cut-off value was 12.5. CONCLUSION: ASQ-17 has stronger option sorting and suitability than ASQ-19. It is an effective assessment tool for asthenopia with an optimal cut-off threshold value of 12.5, which is suitable for diagnosis and curative effect evaluation.