Needs Analysis For Developing Deep Learning-Based Teaching Materials To Enhance Elementary Students’ Science Literacy

Abstract

Science literacy is an important competency that needs to be developed since basic education to equip students in understanding, applying, and making science-based decisions in daily life. However, the results of international studies show that the science literacy of Indonesian students is still low, one of which is due to the limitations of contextual and meaningful teaching materials. This research aims to analyze the need for the development of Deep Learning-based science teaching materials to improve the science literacy of elementary school students. The research uses a Researchand Development (R&D) approach with the ADDIE model, which is focused on the Analysis stage. The subjects of the study include the Science Learning Outcomes Phase B document for class V, two elementary school teachers, and 28 grade V students at SDN 05 Bengkulu City. Data collection techniques include documentation studies, questionnaires, and semi-structured interviews. Data were analyzed descriptively, quantitatively, and qualitatively, as well as through gap analysis. The results of the study show that the science book in class V has contained indicators of science literacy in the aspects of competence, procedural and epistemicknowledge, and scientific attitudes, but has not presented a local context and science content that is close to students' lives. Teachers and students need teaching materials that are interactive, contextual, flexible, and utilize visual and digital media. These findings indicate that the development of Deep Learning-based science teaching materials that integrate contextual science literacy and innovative learning media is needed to improve concept understanding, scientific process skills, and the ability to apply science in the daily lives of elementary school students.