<![CDATA[Early detection of autism in children aged 16–30 months is crucial for timely intervention and improved developmental outcomes. This study aims to analyze the effectiveness of the ChatGPT-based M-CHAT educational model in enhancing parental knowledge and assisting with the completion of the Modified Checklist for Autism in Toddlers (M-CHAT). This model is designed as an interactive assistant that provides explanations about autism, the function of the M-CHAT, and guidelines for self-completion and initial interpretation of screening results. The research method employs a quasi-experimental approach with a pre-test and post-test design involving 60 parent respondents. The research instrument consisted of a 1-5 Likert scale questionnaire measuring three domains: knowledge about autism and M-CHAT, self-completion ability, and satisfaction and usability of the model. Validity was analyzed using Pearson correlation, reliability using Cronbach's alpha, and effectiveness was tested using a paired sample t-test. The results showed that all questionnaire items were valid and reliable. The average pre-test scores in the domains of knowledge, completion, and usability significantly increased in the post-test (p < 0.001). Thus, the ChatGPT-based M-CHAT educational model is effective in improving parents' understanding, self-completion ability, and user satisfaction and can be used as a tool for early autism screening.]]>
Copyrights © 2025
