Efektivitas Problem Based Deep Learning dalam Mengembangkan Kemampuan Pemecahan Masalah Matematis Siswa Indonesia

Abstract

This study aims to examine the effectiveness of the Problem Based Deep Learning (PBDL) model in developing Indonesian students' mathematical problem-solving skills through the Systematic Literature Review (SLR) approach. The background of this study is the low ability of students to link mathematical concepts to everyday contextual situations, especially in the aspect of problem solving. Analysis was conducted on accredited articles from databases such as Sinta, Google Scholar, and ResearchGate published in the last five years. The results of the study indicate that the implementation of PBDL is consistently more effective than conventional methods in improving critical, creative, collaborative thinking skills, and deep mathematical understanding. The success of PBDL implementation is influenced by contextual problem design, small group-based learning strategies, and student independent reflection. However, several limitations were also identified, such as teacher readiness in implementing this approach, variations in research designs that make it difficult to generalize results, limited access to technology in certain areas, and the lack of research examining students' affective and metacognitive aspects. Therefore, systematic integration of PBDL into the mathematics curriculum and teacher training based on Higher Order Thinking Skills (HOTS) is highly recommended. Further studies are recommended to explore the long-term effectiveness of PBDL across educational levels and socio-cultural contexts.