<![CDATA[Physics education within the Kurikulum Merdeka emphasizes the development of students' problem-solving ability as an essential competency for addressing twenty-first-century challenges. However, physics teaching practices in many Indonesian schools continue to rely heavily on teacher-centered instructional approaches, limiting students' opportunities to develop systematic problem-solving ability. This study aimed to examine the effect of the Self-Regulated Learning (SRL) model on senior high school students' physics problem-solving ability. A quasi-experimental study employing a nonequivalent control group pretest–posttest design was conducted with 64 tenth-grade students at SMA Negeri 3 Singaraja, Indonesia. Students' problem-solving ability was measured using a validated essay test, and the data were analyzed using one-way Analysis of Covariance (ANCOVA) with pretest scores as covariates. Prior to the intervention, the SRL and conventional groups demonstrated comparable mean pretest scores of 36.67 (SD = 5.40) and 35.83 (SD = 4.81). After the intervention, students in the SRL group achieved a higher mean posttest score of 73.33 (SD = 5.32) than those in the conventional group, 69.17 (SD = 5.14). ANCOVA revealed a statistically significant effect of the learning model on students' problem-solving ability (F = 17.067, p < 0.001), with a large practical effect (partial η² = 0.219). Furthermore, the estimated marginal mean difference between the two groups was 5.095 points (95% CI [2.498, 7.691]), indicating a consistent advantage of the SRL model over conventional instruction. These findings demonstrate that SRL is substantially more effective in enhancing students' ability to understand problems, formulate solution strategies, implement appropriate procedures, and evaluate solutions. Therefore, the SRL model represents an effective student-centered instructional approach for strengthening physics problem-solving ability and supporting the implementation of the Kurikulum Merdeka.]]>
