Abstract This study is motivated by the low learning outcomes of fifth-grade students at SDN Pegirian II Surabaya in the subject of light properties, where only 23% of students achieved the Minimum Competency Criteria (KKM) of 85. This issue is believed to be related to the teaching methods used, which are insufficiently varied and tend to be passive, leading to difficulties for students in understanding concepts such as reflection, refraction, and the color spectrum. To address this problem, the study aims to evaluate the effectiveness of the experimental method in improving students' learning outcomes. The study employs Classroom Action Research (CAR) conducted in two cycles. In the first cycle, students performed experiments with guidance from the teacher, while in the second cycle, students conducted experiments independently using Student Worksheets (LKPD). The results showed a significant improvement in students' learning outcomes. Before the intervention, the majority of students did not meet the KKM, but after the first cycle, 50% of students achieved the KKM, which increased to 70% in the second cycle. Some students who initially scored very low also showed significant improvement, although there was variation in students' responses to this method. The study concludes that the experimental method is effective in enhancing students' understanding and learning outcomes in the subject of light properties and recommends its implementation in science teaching at the elementary level to create a more interactive and meaningful learning experience.