Implementasi Pengembangan Model Deep Learning Berbasis Komunitas untuk Peningkatan Literasi Teologi Multikultural

Abstract

The discourse on the implementation of Deep Learning in the Merdeka Belajar Curriculum aims to shift education from memorization to critical thinking and in-depth understanding. However, in the field, educators face obstacles due to the lack of technical models, especially in religious education, which prioritizes memorization. This study aims to design a community-based Deep Learning model for multicultural theological education, to bridge understanding. The method used is Research and Development (R&D) with the Borg and Gall model, through ten stages of development, including product revisions and field trials to ensure the model's feasibility. The study subjects comprised 22 elementary schools in Purwakarta, West Java, selected purposively. Data collection techniques included a questionnaire survey with 132 respondents (teachers and principals), interviews with 44 Islamic Religious Education teachers, and classroom observations. Quantitative analysis used descriptive statistics and paired t-tests, combined with qualitative analysis. The results showed that the community-based Deep Learning model effectively improved students' critical thinking skills and in-depth understanding of multicultural theology, with an average score increase of 28% compared to conventional learning. It is recommended that this model be applied more widely in Islamic Religious Education and that integrated supporting media be developed to facilitate its implementation by teachers.