The Impact of an AI-Based Educational Platform on Student Teachers’ Self-Regulated Learning

Abstract

This study investigated the effectiveness of an AI-based lesson planning platform in enhancing self-regulated learning (SRL) skills among student teachers at Sultan Qaboos University. To address a critical gap in AI–SRL integration within Middle Eastern teacher education contexts, this research developed a culturally adapted, Arabic-enabled platform aligned with national curriculum standards. Utilizing a quasi-experimental pre-test/post-test design, the SRL skills of an experimental group (n=14) using an AI-based lesson planning platform were compared with those of a control group (n=15) who followed traditional lesson planning methods. During a four-week intervention, the AI-based platform was utilized to streamline the planning process by providing ready-made lesson plans that adapt to teachers’ objectives, strategies, and evaluation methods. The platform enables teachers to customize and modify each section of their plan by offering personalized suggestions tailored to the instructional needs of each teacher. An SRL questionnaire was constructed and validated for data collection. This instrument comprised 20 items across four dimensions: short-term planning, real-time self-monitoring, rapid self-evaluation, and adjustment of strategies. The questionnaire demonstrated excellent psychometric properties, exhibiting good overall internal consistency (Cronbach's α = 0.89), with individual dimension reliabilities ranging from 0.79 to 0.87, and good test–retest reliability (r = 0.84). Content validity was established by a review conducted by six expert faculty members specializing in educational psychology and teacher education, while construct validity was confirmed using an exploratory factor analysis that revealed a four-factor structure explaining 68.5% of total variance. The statistical analysis indicated a significant improvement in the SRL dimensions for the experiential group, rising from a pre-test mean of 75.07 (SD=10.5) to a post-test mean of 102.20 (SD = 12.5), t (13) = -6.93. p = 0.005. This yields an average gain of 26.93. In contrast, the control group exhibited no significant increase, with only a slight mean improvement of 6.87 (pre-test: M = 78.8, SD = 11.5; post-test: M = 85.67, SD = 13.0; t (14) =-1.8, p = 0.438). The positive effect was evident across all four SRL sub-dimensions, highlighting the inclusive impact of the AI-based platform. This research advances the field of AI in education, with substantial implications for curriculum design for teacher education and professional development.