Objective To develop a self‑efficacy scale for caregivers of elderly patients with dysphagia, and to test its reliability and validity. Methods Based on role theory, self‑efficacy theory, and the biopsychosocial medical model, an initial item pool was constructed through literature review and qualitative interviews. The preliminary version of the self‑efficacy scale for caregivers of elderly patients with dysphagia was formed after 3 rounds of expert consultation. Preliminary investigation inventory was designed based on preliminary version of the scale. Using the convenience sampling method, 30 caregivers of elderly patients with dysphagia from 2 Class⁃Ⅲ level⁃A hospitals in Chengdu were selected for a preliminary survey. Based on the results of the preliminary survey, the scale was adjusted to form the clinical test version. Using the convenience sampling method, 602 caregivers of elderly patients with dysphagia selected from May to November 2024 in 2 hospitals as above were formally surveyed. Item analysis was performed to screen items, and the reliability and validity of the scale were tested. Results The self⁃efficacy scale for caregivers of elderly patients with dysphagia consisted of 20 items across four dimensions: social care beliefs, emotional care beliefs, therapeutic care beliefs, and intake care beliefs. Exploratory factor analysis indicated a cumulative variance contribution rate of 53.891%. Confirmatory factor analysis demonstrated excellent overall model fit; the modified model indicator showed χ²/df =1.891, root mean squared error of approximation=0.054, and goodness⁃of⁃fit index, incremental fit index, Tucker⁃Lewis index, and comparative fit index all >0.9. The composite reliability for the scale ranged from 0.751 to 0.827, and the average variance extracted (AVE) ranged from 0.388 to 0.461, with all inter‑dimension correlation coefficients being smaller than the square roots of the corresponding AVE values. The item‑level content validity index ranged from 0.883 to 1.000; the scale‑level content validity index was 0.800, and the scale‑level average content validity index was 0.967. The criterion‑related validity was 0.419 (P<0.01), indicating a moderate positive correlation. The overall Cronbach's α coefficient of the scale was 0.853, the split‑half reliability was 0.736, and the test‑retest reliability was 0.899. Conclusion The self‑efficacy scale for caregivers of elderly patients with dysphagia demonstrates good reliability and validity and can serve as an effective assessment tool for measuring self‑efficacy in this target population.