<![CDATA[This study aims to develop deep learning-based instructional materials on the topic of states of matter and their changes, to determine the feasibility of the developed materials, and to examine their effectiveness in improving students’ cognitive process skills. The study employed a modified 4D Research and Development (R&D) model consisting of the define, design, develop, and limited dissemination stages. The participants were 35 fourth-grade students at SDN Serdang 1, Banten Province, Indonesia. The developed products included learning outcomes, learning objective sequences, teaching modules, student worksheets, Canva-based learning media, and assessment instruments. Data were collected through expert validation, student response questionnaires, and pretest–posttest assessments. The data were analyzed using descriptive percentage analysis and normalized gain (N-gain). The results showed that the developed instructional materials were categorized as highly feasible in terms of content, media, and language aspects. Furthermore, the effectiveness test indicated a moderate improvement in students’ cognitive process skills after the implementation of the instructional materials. The highest improvement was observed in applying skills (C3), followed by analyzing (C4) and evaluating (C5). These findings indicate that deep learning-based instructional materials can effectively support meaningful science learning and improve elementary school students’ cognitive process skills. Therefore, the developed instructional materials may serve as an alternative instructional model aligned with the demands of 21st-century education.]]>
Copyrights © 2026
