<![CDATA[Generative AI (GenAI) offers interactive explanations, feedback, and personalized support in higher education, yet its educational value depends on how such support is structured. Evidence remains limited regarding whether structured GenAI co-tutoring can simultaneously improve independent learning outcomes and learning motivation in concept-oriented Information Systems courses. This study examined the effectiveness of structured GenAI co-tutoring compared with conventional instruction in an undergraduate Information Systems Concepts course in Indonesia. A quasi-experimental nonequivalent control group pretest–posttest design involved 60 students from two intact classes, with 30 students in each condition. Learning outcomes were measured using parallel forms of the Information Systems Concepts Achievement Test, while learning motivation was assessed using the EVC Light scale. Data collection comprised pre-intervention assessment, the instructional intervention, and post-intervention assessment; achievement tests were completed without GenAI access to capture independently demonstrated learning. ANCOVA was used to compare posttest outcomes while controlling for corresponding pretest scores. The experimental group achieved a significantly higher adjusted ISCAT-B score than the control group, with an adjusted mean difference of 7.89 points, F(1, 57) = 27.65, p < .001, partial η² = .327. Learning motivation was also significantly higher in the experimental group, with an adjusted EVC Index difference of 0.053, F(1, 57) = 13.53, p < .001, partial η² = .192. The study extends evidence on structured GenAI support by examining independently assessed conceptual learning alongside expectancy–value–cost motivation in Information Systems education. In practice, lecturers can adopt a sequence of independent attempts, guided AI feedback, verification, and revision, followed by assessment without AI assistance.]]>
Copyrights © 2026
