Scenarios of Pedagogical Support to Promote Subject Learning Success Among University Students Based on Academic Risk Analysis

The paper considers a strategy for shifting from predicting academic risks to implementing personalized measures of pedagogical support in the context of the digital transformation of education. This study is aimed at developing and testing scenarios for providing this support to university students based on learning analytics data. It is based on the outcomes of a predictive model implemented in the Pythia early warning system. This system has been designed for weekly monitoring and grouping students at high, medium, and low academic risk, followed by differentiated support scenarios. A pedagogical experiment with 209 students resulted in a statistically significant increase in academic performance in the experimental group compared to the control group, as well as a positive dynamic of students, having transferred from high to lower academic risk groups. The study demonstrates the feasibility and effectiveness of a scenario-based approach with regard to the pedagogical support of student academic success in subject learning, based on academic risks analysis. A key achievement of the work is considered the development of a risk-oriented approach, which makes possible a shift from uniform teaching methods to personalized interaction with students classified into high, medium, and low academic risk groups.