Indonesia Learning in the Digital Era: How AI-Based Chatbot Satisfaction Shapes Academic Performance among Indonesian Students

Joni Suhartono, I Gusti Made Karmawan, Drajad WiryawanNon-Scopus11 Desember 2025Afiliasi: ID

Abstrak

This study analyses how satisfaction with AI-based chatbots shapes academic performance among Indonesian university students, a topic of growing importance as AI tools become embedded in higher education. Using data from 361 respondents and employing PLS-SEM, the research evaluates the effects of five antecedents – usefulness, ease of cognitive engagement, trustworthiness, quality of results, and emotional support – on overall satisfaction and adoption continuity. The findings reveal that ease of cognitive interaction, the quality of chatbot-generated output, and emotional support significantly enhance satisfaction, underscoring the dual cognitive-affective value of AI systems in learning environments. Conversely, perceived usefulness and trustworthiness do not significantly influence satisfaction, indicating that students value interactional fluency and personal support more than basic functional or credibility attributes. Furthermore, satisfaction strongly predicts adoption continuity, reflecting students' willingness to integrate chatbots into ongoing academic activities. Adoption continuity, in turn, exerts a substantial positive effect on academic performance, improving students' comprehension, assignment quality, engagement, and independent learning. These results demonstrate that AI-based chatbots can function as impactful personalised learning companions that significantly contribute to academic development. The study offers practical insights for educational institutions and AI developers. To maximise their learning impact, chatbots should prioritise adaptive dialogue, contextuallyrelevant content generation, and emotionally responsive features. While the findings provide a robust empirical foundation, future research should broaden the institutional sample, employ mixed-method designs, and explore behavioural differences across academic disciplines to deepen understanding of AI-supported learning

Identifier

Scroll to Top