Journal of Science and Mathematics Education
Vol. 2 No. 2 (2026): Journal of Science and Mathematics Education, June 2026

AI-Powered Personalized Learning in Indonesian Senior High School Chemistry Education: Evidence from a Quasi-Experimental Study

Reza Firmansyah (Department of Chemistry Education, Faculty of Mathematics and Natural Sciences, Universitas Negeri Surabaya, Indonesia)
Novia Kurniawati (Department of Chemistry Education, Faculty of Mathematics and Natural Sciences, Universitas Negeri Surabaya, Indonesia)
Bagas Dwi Prasetyo (Department of Chemistry Education, Faculty of Mathematics and Natural Sciences, Universitas Negeri Surabaya, Indonesia.)



Article Info

Publish Date
23 Jun 2026

Abstract

This study investigated the effectiveness of an AI-powered AI-powered adaptive learning platform in improving students’ academic achievement in chemistry and explored their overall learning experience. The research was conducted at SMA Negeri 3 Surabaya, involving 72 students of class XI MIPA in the 2023/2024 academic year. A quasi-experimental pretest-posttest control group design was employed. The experimental group (XI MIPA 2, n = 36) used an AI-powered adaptive learning platform incorporating artificial intelligence algorithms to personalize learning pathways, whereas the control group (XI MIPA 1, n = 36) received conventional instruction. Academic achievement was measured using a validated chemistry knowledge test covering reaction rates, chemical equilibrium, and thermochemistry, while learning experience was assessed through a structured questionnaire. The results showed that the experimental group achieved significantly higher posttest scores (mean = 83.7) than the control group (mean = 72.4), with a statistically significant difference (t = 6.43, p < 0.001) and a large effect size (d = 1.02). The experimental group also obtained a higher N-Gain score (0.61, moderate) than the control group (0.35, low). In addition, students in the experimental group reported highly positive learning experiences across all measured dimensions. These findings suggest that AI-powered personalized learning has strong potential to enhance chemistry achievement and student engagement in Indonesian senior high school settings.

Copyrights © 2026






Journal Info

Abbrev

josme

Publisher

Subject

Agriculture, Biological Sciences & Forestry Biochemistry, Genetics & Molecular Biology Chemical Engineering, Chemistry & Bioengineering Chemistry Education Materials Science & Nanotechnology Mathematics Physics

Description

Journal of Science and Mathematics Education (JOSME) is a multidisciplinary and a medium of communication for researchers, academicians, and practitioners that provide a means for sustained discussion of relevant issues that fall within the focus and scopes of the journal which examined empirically. ...