RECONSTRUCTION OF A MODEL FOR EVALUATING TRANSFORMATIVE LEARNING BASED ON RECONSTRUCTIVISM IN THE ERA OF DEEP LEARNING

Abstract

The transformation of education in the era of Deep Learning calls for a paradigm shift in learning assessment one that no longer focuses solely on measuring cognitive learning outcomes, but also on students’ ability to understand, reflect on, and transform social reality. This article aims to reconstruct a transformative learning evaluation model based on the philosophy of reconstructivism that is relevant to the implementation of Deep Learning and the Merdeka Curriculum in Indonesia. The research employs a qualitative approach using the Systematic Literature Review (SLR) method and conceptual reconstruction. Data sources were obtained from reputable national and international journal articles published between 2023 and 2026, educational policy documents, and various relevant literature. Data analysis was conducted through thematic content analysis, involving the stages of identification, categorization, interpretation, and conceptual synthesis. The results of the study indicate that the reconstructionalism-based transformative evaluation model consists of four main components: context, input, process, and product (CIPP-T). This model positions social issues as the context for learning, teacher readiness and the digital ecosystem as inputs, deep learning based on social projects as the process, and changes in critical understanding and social action as the products of learning. The reconstructed model aligns strongly with the principles of the Merdeka Curriculum, particularly in strengthening the Pancasila Student Profile, authentic assessment, and project-based learning. These findings provide a conceptual contribution to the development of a more humanistic, reflective, and transformative educational evaluation in the era of deep learning.