This study aims to empirically analyze the effects of information classes using AI Digital Educational Materials(AIDT) on middle school students' student agency and learning engagement. As the educational paradigm shifts in the era of digital transfor...
This study aims to empirically analyze the effects of information classes using AI Digital Educational Materials(AIDT) on middle school students' student agency and learning engagement. As the educational paradigm shifts in the era of digital transformation, learner-centered education is being emphasized, and AIDT is garnering attention as a core tool for its implementation. However, empirical research on its actual effectiveness in the educational field, especially within the context of the information subject, remains insufficient. For this research, a pre-test/post-test control group design was employed with 92 first-year middle school students (N=92) in Gwangju.
The students were assigned to either an experimental group (n=46) using AIDT or a control group (n=46) using traditional paper-based textbooks for 12 class sessions over six weeks. Student agency was measured using the scale developed by Lee et al. (2023), and learning engagement was measured using the scale by Cha et al. (2010). The collected data were analyzed using SPSS 31.0.0 for independent t-tests (to verify homogeneity), MANOVA (to comprehensively analyze group differences), and subsequent ANOVA. A mixed-methods approach was adopted, incorporating qualitative data from open-ended surveys (n=46) and
in-depth interviews (n=8) with the experimental group to supplement the quantitative findings. The results confirmed that the two groups were homogeneous in all sub-factors of student agency and learning engagement at the pre-test. In the post-test analysis, student agency showed a significant overall multivariate effect (Pillai’s Trace = .182, p < .01), which was primarily attributed to significant improvements in the 'self-regulation' sub-factor at both personal and community levels (ps < .05). Conversely, learning engagement did not show a significant multivariate effect in the MANOVA analysis (p = .088). However, the sub-factor analysis revealed a significant improvement in the 'class activities' factor (p = .014). The qualitative analysis strongly supported these quantitative results,
indicating that AIDT facilitated students' self-regulative learning strategies and active participation in class activities. In conclusion, this study confirmed that AIDT-based instruction shows a positive effect on learners' competencies, particularly centered on fostering self-regulation skills and promoting active engagement in class activities. Although statistical significance was not secured in some factors due to limitations such as the short duration of the study or experimental design, this warrants further verification through future research. Nevertheless, this study provides significant implications that AI digital tools can be utilized as effective tools for sophisticated teaching and learning strategies targeting specific core competencies of learners in the field of information education.
Keywords: AI Digital Education Materials, AI Digital Textbook, AIDT, Information Education, Student Agency, Learning Engagement