THE POTENTIAL OF STEAM-BASED SCIENCE LEARNING WITHIN A DEEP LEARNING FRAMEWORK FOR THE TOPIC OF ENVIRONMENTAL POLLUTION

Abstract

This study aimed to analyze the potential for implementing an integrated STEAM–Deep Learning model on the topic of Environmental Pollution at the junior secondary school level, prior to full-scale implementation. A descriptive quantitative survey approach was employed, involving science teachers and a Grade IX class as research subjects. Readiness was measured across four variables—Learning Tools, Supporting Facilities, Human Resources (Teachers), and Student Habits—using an expert-validated Likert-scale questionnaire, and data were analyzed using descriptive mean scores converted into qualitative categories. The results showed that Supporting Facilities obtained the highest mean score of 3.25 (fairly adequate), Learning Tools scored 2.76 (fairly good/valid), Human Resources (Teachers) scored 2.72 (moderate), and Student Habits obtained the lowest score of 2.24 (rare). Overall, the potential for implementing STEAM–Deep Learning on the topic of environmental pollution is fairly significant, but two main challenges remain: strengthening teachers' pedagogical competence in integrating the Engineering and Art dimensions, and shifting students' learning habits from surface, teacher-directed routines toward mindful, meaningful, and joyful project-based inquiry. These findings imply that successful adoption of STEAM within the Deep Learning framework depends less on physical infrastructure, which is already comparatively ready, than on sustained teacher professional development and the deliberate cultivation of independent, project-oriented learning habits among students.