The quality of higher education and managing retention rates represent major strategic challenges for Higher Education Institutions (HEIs) globally, with student dropout being a critical issue. Currently, a robust theoretical framework for applying Multi-Criteria Decision-Making (MCDM) methods is lacking, which hinders the development of well-founded decision-making tools to address this problem. The primary objective of this work was to create such a framework by not only listing the determinant factors but also classifying them into clusters to facilitate the robust application of MCDM in the context of HEI student dropout. The methodology involved a rigorous systematic review of the literature in the Web of Science (WoS) database covering the period 2021–2025, which led to the identification and synthesis of 17 distinct factors determining student persistence or dropout. The core idea is that the ranking derived from frequency can support two distinct expert-evaluation strategies: Focusing on high-frequency factors (e.g., top 5) because they are well-anchored and easier for experts to evaluate, or focusing on under-represented factors (e.g., rank 10 or below) to explore gaps and identify novel intervention levers. These factors were subsequently prioritized by frequency and grouped into three hierarchical clusters based on their theoretical nature and confirmed statistical interdependencies. This research provides a solid foundation, offering the necessary theoretical framework for future MCDM studies on HEI dropout to be conducted on a robust, complete, and well-justified basis, moving beyond the random selection of factors.
Nechita et al. (Mon,) studied this question.