PEDAGOGICAL FRAMEWORK FOR INTEGRATING AUTOMATED ASSESSMENT SYSTEMS IN HIGHER EDUCATION

Abstract

This article analyzes the pedagogical foundations, didactic limitations, and practical integration mechanisms for the implementation of automated assessment systems in higher education institutions. The main goal of the study is not to fully technologize assessment, but to develop a pedagogical framework that combines automated tools, teacher expertise, and active student reflection. The article summarizes current scientific research on online assessment, digital formative assessment, learning analytics, automated feedback, generative artificial intelligence-based assessment, and data governance. As a result, a seven- component model for integrating automated assessment systems is proposed: constructive coherence, task architecture, level of automation, human-in-the-loop control, feedback design, analytical monitoring, and ethical-legal management. The model is concretized using the example of mathematics and computer science, and it is based on which assessment elements should be fully automated, which should be semi-automated, and which should retain human expertise. The study shows that the true pedagogical value of automated systems lies not in speed and scalability, but in providing the teacher with evidence-based decision-making and the student with clear and timely feedback. However, technological superiority can become a pedagogical risk if algorithmic bias, transparency, academic integrity, data security, and student feedback literacy are overlooked.