The Effect Of Generative Ai-Integrated Problem-Based Learning Model On Students' Mathematical Problem-Solving Ability In Sletv Material

Abstract

Mathematical problem-solving ability is one of the core competencies students must master, particularly in the topic of Systems of Linear Equations in Two Variables (SLETV), which requires stepwise reasoning and the mathematical modeling of contextual situations. This study aims to examine the effect of implementing the Problem-Based Learning (PBL) model integrated with generative AI technology on students' mathematical problem-solving ability in the SLETV topic. This research employed a quantitative approach with a quasi-experimental nonequivalent control group design. The population consisted of eighth-grade students at a junior high school, with samples selected through purposive sampling and divided into an experimental group and a control group. The experimental group received PBL instruction enriched with generative AI used as a virtual tutor and a generator of contextual problem illustrations, while the control group received conventional PBL instruction without generative AI assistance. Data were collected through a mathematical problem-solving test based on Polya's indicators and analyzed using an independent t-test after normality and homogeneity assumptions were satisfied. The results show that the mean problem-solving score of the experimental group was significantly higher than that of the control group. These findings indicate that integrating generative AI into the PBL syntax strengthens the stages of problem orientation, exploration of solution strategies, and student reflection on the SLETV topic, thereby resulting in a more optimal improvement of mathematical problem-solving ability compared to conventional PBL instruction.