Benchmarking higher education excellence: insights from QS rankings

Purpose: This study aims to address a gap in traditional university ranking methodologies by investigating the interrelations among key indicators featured in the QS rankings, within the broader context of benchmarking in higher education. Design/methodology/approach: Utilizing the 2024 QS ranking data and a Bayesian Belief Network (BBN) model, this research explores the interconnected relationships among indicators such as “academic reputation,” “employer reputation,” “faculty-to-student ratio,” “sustainability” and others to predict university rankings. Findings: The developed model achieves 80% predictive accuracy and shows that strong performance in “employment outcomes,” “academic reputation” and “employer reputation” contributes to higher overall scores. In contrast, weaker performance in “academic reputation” and “sustainability” is associated with lower scores. Among these factors, “academic reputation” is the most informative indicator for predicting the overall score. Originality/value: This research contributes to the literature by emphasizing the interconnections among ranking criteria and advocating for network-based models for benchmarking in higher education. Particularly, it underscores the importance of “sustainability” in forecasting rankings, aligning well with the broader theme of predicting university performance and societal impact. This study offers valuable insights for researchers and policymakers, promoting a comprehensive approach that considers the interdependencies among criteria to enhance educational quality and address societal change within the framework of benchmarking in university rankings.