Methodology for identifying academic disciplines in the risk group in terms of academic performance

Abstract

The relevance of the study is due to the lack in the literature of practical models and algorithms for improving the quality of education by identifying disciplines that are risky for academic performance. Such disciplines become a “stumbling block” on the way of successful mastering by students of the educational program. The consequence is a make-up examination, losing scholarships, academic arrears, expulsion. Risky disciplines require the adoption of corrective management actions to reduce the risks of unsatisfactory teaching and/or learning. The purpose of the study is to identify indicators for assessing the risk of academic achievement and developing an evidence-based methodology for identifying disciplines at risk. For the statistical analysis, the academic performance results of 462 second-year students in various fields of study in 92 disciplines were used. Two approaches have been developed to identify disciplines included in the risk group in terms of academic performance: 1) calculation of the total risk score for each discipline; 2) clustering of disciplines. Both approaches are based on: performance indicators; non-risky area for determining the values of these indicators; identification of disciplines that go beyond the non-risky area. Each of the approaches made it possible to single out the disciplines that were risky for academic performance (22 – by the first method, 27 – by the second one). A high correlation of the disciplines identified by the two methods was revealed. For disciplines with the “exam” form of control – r=0.99, with the “credit” form of control – r=0.77. The proposed methodology variants can be applied in universities that operate educational process management systems to support the functions of students’ contingent movement and accounting for academic performance.