Pengembangan Math Book Digital Berbasis Deep Learning Untuk Numerasi Siswa Kelas IV

Abstract

This study aims to develop and evaluate the feasibility and effectiveness of a Deep Learning-based Math Book Digital to improve fourth-grade students’ numeracy skills. The research employed a Research and Development (R&D) method using the ADDIE model, which includes analysis, design, development, implementation, and evaluation stages. The participants consisted of material experts, media experts, and fourth-grade elementary school students. Data were collected through expert validation questionnaires, student and teacher response questionnaires, and pretest–posttest assessments. The results showed that media expert validation obtained an average score of 3.41 (85.25%) and material expert validation reached 3.60 (90%), both categorized as highly valid. Student responses in small and large group trials were 91.74% and 91.69%, while teacher responses were 93.33% and 86.67%, indicating the media was very practical. Statistical analysis revealed a significant improvement in students’ numeracy skills (Sig. = 0.001 < 0.05) with an average N-Gain score of 0.7254 in the high category. Therefore, the Deep Learning-based Math Book Digital is valid, practical, and effective for supporting numeracy learning in elementary schools.