<![CDATA[This study aims to examine the effect of the STEM Hybrid Research-Based Learning (STEM-Hybrid RBL) model on high school students’ scientific reasoning skills in the context of environmental change, compared with the Discovery Learning model. The study employed a quasi-experimental method using a pretest–posttest control group design and involved 84 tenth-grade students from SMA Negeri 1 Kota Sukabumi, Indonesia, who were selected through purposive sampling and assigned to an experimental class (n = 42) and a control class (n = 42). The research instrument was a multiple-choice scientific reasoning test that met the validity criteria and demonstrated a high reliability coefficient (0.94). Data were analyzed using the Mann–Whitney U test because the data did not meet the assumptions of normality and homogeneity. Effect size (r) and N-gain analyses were also conducted to examine the magnitude of the difference and the improvement in each scientific reasoning indicator. The results revealed a statistically significant difference between the experimental and control classes (U = 459.50, Z = −3.860, p < .001), with a medium-to-large effect size (r = .42). The experimental class achieved a higher mean posttest score (91.65) than the control class (77.75). The N-gain analysis for each indicator showed the greatest improvement in proportional reasoning and correlational reasoning, whereas identifying and controlling variables showed the lowest improvement in the experimental class. Student responses to the STEM-Hybrid RBL model were categorized as good to very good, with an average score of 79.6%. These findings indicate that STEM-Hybrid RBL is a promising learning model for improving high school students’ scientific reasoning skills and may be applicable to various science learning contexts at the secondary school level.]]>
