In recent years, the evaluation and ranking of universities have become major concerns for higher education institutions and policymakers. Among various evaluation dimensions, research-related criteria play a central role in shaping a university’s academic reputation and ranking status. However, one of the key challenges in this area is identifying the relative importance of these criteria and accurately modeling their influence on ranking levels. While previous studies have employed either traditional statistical methods or multi-criteria decision-making (MCDM) techniques, most lack a coherent integration of objective weighting and predictive modeling approaches.
This study addresses this gap by applying a hybrid methodology to real-world data on the research performance of Iranian universities. First, the CRITIC method (Criteria Importance Through Intercriteria Correlation)—an objective and data-driven technique that accounts for both variability and intercorrelation among criteria—is used to calculate the relative weights of the research indicators. These calculated weights are then incorporated into an ordinal logistic regression model (OLR) to assess the contribution of each criterion to the probability of a university being placed within specific ranking levels.
The key contribution of this research lies in its structured integration of two complementary methods, enabling both objective weighting and interpretable statistical analysis. Unlike traditional MCDM approaches that often produce aggregate scores with limited diagnostic value, the proposed model provides a deeper understanding of how each criterion affects ranking outcomes—both in magnitude and direction.
The findings reveal that certain frequently used research metrics have limited or even negligible predictive value, while others exert a strong and statistically significant influence on institutional rankings. These insights offer practical guidance for university administrators, research planners, and policymakers who aim to enhance institutional research strategies and improve their positions in national and international ranking systems. Furthermore, the proposed methodological framework is adaptable and may be extended to other educational systems and ranking contexts beyond Iran.