Lia Agesti
Master's Program in Elementary Education, Faculty of Teacher Training and Education, Universitas Jambi, Jambi, Indonesia

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

From Local Culture Projects to Deeper Learning: Strengthening Conceptual Understanding and Critical Thinking in Elementary Natural and Social Sciences Lia Agesti; Ahmad Syarif; Eka Sastrawati
ISEJ : Indonesian Science Education Journal Vol. 8 No. 1 (2027): October-January
Publisher : Yayasan Darussalam Bengkulu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62159/isej.v8i1.2975

Abstract

Elementary Natural and Social Sciences learning often remains oriented toward recall, leaving students with limited opportunities to explain concepts, justify decisions, and apply knowledge to everyday situations. This classroom action research investigated whether deep learning principles integrated with Project-Based Learning (PjBL) could strengthen fourth-grade students’ conceptual understanding and critical thinking. The participants were 22 students at SD Negeri 225/IX Kasang Solok, Muaro Jambi, Indonesia. Two action cycles, each comprising planning, action, observation, and reflection across two meetings, used contextual projects: a local-culture mind map, a pop-up book, a “rule tree” poster, and a rules-classification game. Data were collected through observation rubrics, concept-understanding tasks, critical-thinking assessments, field notes, interviews, and documentation, then analyzed descriptively using mean percentage scores and reflective comparison across meetings. Conceptual-understanding scores increased from 65.3 to 74.2 in Cycle I and from 67.6 to 80.0 in Cycle II. Critical-thinking scores increased from 66.85 to 73.3 in Cycle I and from 67.4 to 79.6 in Cycle II. Reflection indicated that contextual questions, structured group roles, scaffolding, project presentation, and written reflection supported the strongest improvement. The findings suggest that culturally situated projects can operationalize deep learning by connecting conceptual representation, reasoning, application, and reflection. Nevertheless, the single-class action design limits causal and broader generalization claims.