<![CDATA[Although previous studies have reported improvements in science learning following the use of experimental methods, limited attention has been given to the pedagogical processes underlying these improvements and their applicability across conceptually different topics within a single classroom action research sequence. This study aimed to improve the learning process and outcomes of fifth-grade students in Science and Social Studies (IPAS) through the experimental method. The study employed Classroom Action Research (CAR) based on the Kemmis and McTaggart model and was conducted in two cycles, each comprising two meetings, involving 23 fifth-grade students with heterogeneous academic abilities. Data were collected through teacher and student activity observation sheets and end-of-cycle achievement tests and analyzed descriptively. The findings demonstrated progressive improvements in student learning outcomes, with the mean score increasing from 69.13 at baseline to 75.22 in Cycle I and 81.52 in Cycle II. Classical mastery increased from 39.13% at baseline to 56.52% in Cycle I and 95.65% in Cycle II. Teacher activity implementation increased from 70.31% to 93.75%, while student activity increased from 64.29% to 95.24%. Improvements were observed across two conceptually distinct topics: mixtures and solutions in Cycle I and heat and its transfer in Cycle II. These findings suggest that experimental activities, when supported by structured procedural guidance and systematic reflection, can facilitate improvements in classroom participation and learning achievement in elementary IPAS. However, given the single-class CAR design without a comparison group, these findings should be interpreted as intervention-associated improvements rather than definitive evidence of a causal effect.]]>
