The Effectiveness of Mathematics Learning with an Open-Ended Approach Reviewed from Students' Self-Regulated Learning

Abstract

This study investigates the effectiveness of mathematics learning with an open-ended approach in improving students' Self-Regulated Learning (SRL) compared to the conventional scientific approach in Indonesian junior high schools. A quasi-experimental design with pretest-posttest control group was employed at a Muhammadiyah Junior High School in Yogyakarta, involving 48 grade VIII students divided into experimental (open-ended approach) and control (scientific approach) groups. Data were collected using a validated SRL questionnaire measuring three dimensions: cognitive, motivational, and behavioral. The intervention was conducted over four learning sessions on linear function topics, with data analysis including paired t-tests, one-sample t-tests, and Hotelling's T² test at a significance level of 0.05. The results showed that the open-ended approach significantly improved students' SRL with an average increase of 10.16 points (from 65.38 to 75.54, p = 4.613e-06), surpassing the control group's increase of 8.77 points. The motivation dimension demonstrated the largest improvement (5.70 points), followed by behavioral (3.04 points) and cognitive (1.50 points) dimensions. Following the intervention, 92% of experimental class students achieved high or very high SRL categories, compared to 87.5% in the control class. Hotelling's T² test confirmed significant differences between approaches, with the open-ended method proving more effective in developing students' autonomy, responsibility, and self-reflection. These findings suggest that open-ended learning effectively fosters intrinsic motivation and learning independence, making it a recommended strategy for developing SRL in 21st-century mathematics education.