Development of e-modules for science subjects based on artificial intelligence technology to improve concept mastery and self- efficacy of students of grade IV elementary school

The study aimed to test the validity, practicality, and effectiveness of e-modules of science subject content based on Artificial Intelligence (AI) technology to improve concept mastery and self-efficacy of 4th-grade elementary school students. This research and development model uses Thiagarajan's (1974) development model of Define, Design, Development, and Dissemination, abbreviated as 4D. The data analysis technique used is data analysis of validity, practicality, and effectiveness tests of e-modules, which are processed into quantitative and qualitative data. The results showed that the feasibility of e-modules was categorized as very good, as evidenced by the construct validity test, which obtained a feasibility percentage of 94.16% with the category of "very feasible.” The content validity test obtained a feasibility percentage of 80%, classified as "feasible.” The practicality of using e-modules through limited trials on a group of students using student and teacher observation sheets obtained a percentage of 93.75% with the category "very practical." The results of learning observations by teachers obtained a percentage of 96.05% with the category "very practical.” The effectiveness of the e-modules on concept mastery using pretest-posttest instruments obtained an N-gain of 0.762, categorized as "high.” Furthermore, the testing of students' self-efficacy using a questionnaire instrument obtained a percentage of 70.23%, categorized as “effective.” Based on the results, this e-module is declared feasible and effective in improving students' concept mastery and self-efficacy in the science subject of grade IV elementary science.