<![CDATA[This quantitative study aims to determine the improvement and effectiveness of a Problem-Based Learning (PBL) model using a Deep Learning approach on students’ scientific reasoning regarding the topics of temperature and heat. A pre-experimental design with a one-group pretest-posttest approach was applied in this study. The sample consisted of 36 11th-grade MIPA students at a public high school in Bandung, selected using convenience sampling. Data were collected using a scientific reasoning test adapted from the Lawson Classroom Test of Scientific Reasoning (LCTSR), which had been tailored to the subject matter. The results showed a significant improvement in scientific reasoning, with the average pretest score increasing from 28.31 to 58.61 on the posttest, and achieving an average N-gain score of 0.43 (moderate category). The greatest improvement occurred in the correlational aspect (N-gain = 0.54), followed by proportional (0.49), hypothesis-deductive (0.44), probabilistic (0.42), conservation (0.36), and variable control (0.25). Furthermore, the Wilcoxon signed-rank test confirmed that this model is significantly effective (Z = -5.242, p < 0.001). These findings indicate that the Problem-Based Learning model with a Deep Learning approach enhances scientific reasoning by promoting active inquiry and reflective thinking. It was concluded that the implemented learning model effectively improves students’ scientific reasoning, particularly in analyzing relationships between variables in physics concepts.]]>
Copyrights © 2026
