<![CDATA[Effective implementation of Indonesia's Merdeka Curriculum requires instructional approaches that connect academic content to students' real-life contexts, yet primary classrooms frequently remain dominated by teacher-centered, rote-based instruction. This study examines whether the Contextual Teaching and Learning (CTL) model can meaningfully improve the quality of Integrated Science and Social Studies (IPAS) instruction at the primary level. A Classroom Action Research design following Kemmis and McTaggart's cyclical model was conducted across two cycles with 26 Grade IV students at SDN 3 Sojol, Donggala Regency, Indonesia. Data were collected through validated written tests (KR-20 = 0.78), structured observation sheets (17 teacher indicators; 13 student indicators), and semi-structured interviews, analyzed using both quantitative mastery criteria and Miles and Huberman's qualitative framework. Classical learning mastery improved from 65.4% in Cycle I to 88.5% in Cycle II, surpassing the 85% criterion. Teacher activity scores progressed from 2.86 (Good) to 3.71 (Very Good), and student participation rates increased across all 13 behavioral and cognitive indicators. Unexpectedly, a subset of behaviorally passive students consistently achieved above-average test scores, indicating a discrepancy between observable participation and cognitive engagement. The findings affirm that systematic, iterative CTL implementation transforms IPAS learning toward meaningful, student-centered inquiry. The study contributes process-level evidence demonstrating that CTL effectiveness requires the reflective alignment of all seven components, not merely the introduction of real-world materials. Challenges in equitable group participation and time management remain areas for ongoing pedagogical development.]]>
