Enhancing Students' Scientific Argumentation Skills in Islamic Religious Education Through AI-Powered Differentiated Learning

Abstract

This study investigates the effectiveness of AI-assisted learning in enhancing students’ scientific argumentation skills within Islamic Religious Education (PAI) at SMKN Widang. As digital tools become more integrated into education, understanding their impact on higher-order thinking skills is increasingly relevant. A quasi-experimental design was employed, involving two groups: an experimental class (XI-A) that used AI tools such as ChatGPT for constructing arguments and receiving instant feedback, and a control class (XI-B) that received conventional instruction. Data were collected through argumentation tests, classroom observations, and interviews. Statistical analysis was conducted using an independent sample t-test in SPSS 26. The experimental group showed a significantly higher improvement in argumentation skills (Mean = 76.34, SD = 15.53) compared to the control group (Mean = 55.34, SD = 17.99), with a large effect size (Cohen’s d = 1.25, p < 0.05). AI tools facilitated student engagement by enabling real-time feedback, personalized content, and multiple perspectives, particularly in discussing contemporary issues like brawls and substance abuse from an Islamic viewpoint. These findings suggest that AI-based learning can effectively support scientific argumentation in religious education contexts by fostering deeper analysis and critical thinking. Despite limitations—such as being restricted to one institution and a single AI tool—the results highlight the potential of AI integration for contextual and individualized learning. Future studies could explore broader applications across disciplines and educational levels.