Trends and Developments in Educational Program Evaluation Models in the Digital Era: A Systematic Literature Review

Abstract

Digital transformation has fundamentally reshaped how educational program evaluation is designed and implemented. Evaluations that once focused solely on end results have evolved into continuous, data-driven processes supported by learning analytics and artificial intelligence (AI). This study aims to review the trends and developments in educational program evaluation models over the past decade (2015–2025), highlighting the paradigm shift from conventional frameworks toward digital-hybrid approaches. Using a Systematic Literature Review (SLR) guided by the PRISMA protocol, forty-two scientific articles were selected from Scopus, ERIC, SpringerLink, DOAJ, and SINTA 2 databases. The findings reveal that classical models such as CIPP, Logic Model, and Kirkpatrick’s Four Levels remain the dominant frameworks but have been substantially adapted through the integration of digital indicators and learning analytics. The incorporation of learning analytics and AI has enhanced the accuracy, efficiency, and predictive capacity of educational evaluations. However, challenges related to data validity,   privacy,   ethics,   and   evaluator   literacy   persist   in   implementing   digital evaluations. The study concludes that the success of educational evaluation in the digital era depends on the synergy between classical methodology, technological innovation, and ethical data governance. The findings provide a theoretical and practical foundation for developing sustainable, accountable, and contextually relevant digital-based evaluation models for Indonesia’s education system.