This study aims to determine: (1) the effect of the Discovery Learning model on improving students’ science learning outcomes; (2) the effect of the Inquiry Learning model on improving students’ science learning outcomes; and (3) the significant difference in the gain value of science learning outcomes between students taught using the Discovery Learning model and those taught with the Inquiry model. This research employed a quasi-experimental method with a Pretest-Posttest Non-Equivalent Multiple Group Design. The population consisted of all fifth-grade students at SD Negeri 9 Lawa, and the sample was selected purposively, consisting of 48 students divided equally into two classes: Class VA (Discovery Learning) and Class VB (Inquiry Learning). Data were collected through pretests and posttests using a 15-item multiple-choice test. The analysis included normality tests, homogeneity tests, N-Gain calculations, and hypothesis testing using the Mann-Whitney U test with a significance level of 0.05. The results showed that the Discovery Learning model improved students’ learning outcomes from an average pretest score of 59.44 to a posttest score of 78.11, while the Inquiry Learning model increased scores from 56.39 to 76.11. The average N-Gain for Discovery Learning (0.42) was slightly higher than for Inquiry Learning (0.41), indicating a moderate improvement in both groups. Although both models significantly enhanced learning outcomes, Discovery Learning demonstrated a marginal advantage. The findings suggest that active, student-centered learning models, particularly Discovery Learning, can effectively improve science learning outcomes in elementary school. Teachers are encouraged to integrate these models into instructional practices to foster students’ critical thinking, problem-solving skills, and conceptual understanding.
Copyrights © 2025