Improving students’ physics representation competence with an Android-based representation training model

Abstract

Representational competence is vital for learning and solving problems in physics, yet many students struggle to master it, and teachers encounter challenges in fostering its development. This study addresses the issue by developing an Android-based training model focused on linear motion kinematics, designed using the analysis, design, development, implementation, and evaluation (ADDIE) research and development (R&D) framework and validated by experts. A total of 127 undergraduates participated through questionnaires, interviews, and observations. The model incorporates feedback and scaffolding to guide students’ understanding and practice. Implementation results showed significant improvements in representational competence. N-Gain scores reached 0.35 (medium) in experimental group I and 0.61 (medium) in experimental group II. Statistical analysis using the Wilcoxon signed-rank test confirmed these gains were significant (p<0.05) with large effect sizes (r=0.871; r=0.862). Further, the Kruskal-Wallis’s test revealed significant differences between groups, and Games-Howell post hoc analysis indicated that integrated classroom use was more effective than independent practice. Student responses demonstrated high practicality and positive engagement, reinforcing the model’s usability. These findings highlight the novelty of an expert-validated, scalable Android-based platform as an accessible tool to enhance representational competence in physics education. Future research should investigate its broader application across physics topics and its long-term impact on learning outcomes.