Application of Humanistic Theory Assisted by Deep Learning in Science Learning Grade 4 SDN 108/1 Sungai Rumbai

Abstract

This study investigates the integration of humanistic learning theory with a deep learning approach in science education for fourth-grade students at SDN 108/1 Sungai Rumbai. We used a descriptive qualitative method and purposive sampling to select 14 students who actively participated in science learning. The research utilized validated instruments, including observation sheets, questionnaires, and student reflection notes, with a validity score ≥ 0.80 and reliability (Cronbach’s Alpha) of 0.85. Data were analyzed thematically for qualitative aspects and using descriptive statistics and the Wilcoxon test for quantitative learning outcomes. Findings indicate significant improvements in student engagement, understanding of concepts, and emotional expression. Learning activities, supported by Canva and Wordwall, fostered interactive and joyful learning environments. Students demonstrated increased focus, confidence, and the ability to relate science material to personal experiences. Before the intervention, only 68% of students met learning targets, whereas after the implementation, 91.7% achieved satisfactory outcomes. Despite limitations in sample size and duration, the study emphasizes the promise of combining psychological theory and digital tools to improve science learning outcomes. This integration offers a promising model for 21st-century learning that is mindful, meaningful, and emotionally engaging.