Evaluasi Media Pembelajaran Digital Matematika Menggunakan Integrasi Analisis Korespondensi Berganda dan Klasterisasi K-Medoids

Abstract

Learning linear inequalities still encounters challenges in terms of visualization and interactivity, while the evaluation of digital learning media generally relies only on descriptive analysis, resulting in less comprehensive findings. This study aims to evaluate the effectiveness of RShiny-based digital learning media through the integration of Multiple Correspondence Analysis (MCA) and K-Medoids clustering as a scientific novelty in examining multidimensional patterns of student perceptions. The media was implemented among 101 tenth-grade students from three schools using a five-construct questionnaire that had been previously tested for validity and reliability. The results showed that most students responded positively to the interface, clarity of material, and ease of use; MCA identified dominant indicators including increased motivation, conceptual understanding, and assessment outcomes, while K-Medoids produced two perception clusters, namely highly positive and moderate. Therefore, the RShiny-based media is proven effective in enhancing students’ motivation, conceptual understanding, and academic performance in learning linear inequalities.