International Journal of
Information and Education Technology

Abstract

This study aims to reconstruct a character education model in elementary schools through the integration of digital reflective applications with the Socio-Emotional Learning (SEL) approach, which combines personal values, school culture, and social-emotional intelligence. Using a mixed quantitative-qualitative method, this study involved 28 students as a single experimental group sample. The instruments were tested for validity and reliability using the Rasch model, while the effectiveness of the model was analyzed using paired-sample t-tests, normality tests, N-Gain, and Partial Least Square Structural Equation Modeling (PLS-SEM) modeling. Qualitative data were analyzed using NVivo to uncover the depth of students’ narrative reflections. The results showed significant improvements in all character indicators (p < 0.001) with an average N-Gain of 0.2955. The Rasch model confirmed the validity of the instruments with proportional logit distributions and no bias between groups. Structural modeling revealed that responsibility and self-awareness play a central role in shaping an interconnected character network. Students’ narrative reflections reinforce evidence of authentic value internalization, marked by the emergence of themes such as empathy, emotional regulation, and honesty. The synthesis of all analyses demonstrates that the developed model is not only statistically effective but also pedagogically transformative. This study recommends the integration of digital reflective applications as an evidence-based character education strategy relevant to addressing the challenges of 21st-century values education.