Beyond Rankings: A K-means Approach to Evaluating Research Universities in Emerging Higher Education Systems

This study introduces a novel approach to evaluating research universities in developing countries, using Türkiye as a case study within the broader context of global higher education trends. By combining the national University Ranking by Academic Performance (URAP-TR) metrics with K-means clustering analysis, we address the limitations of international ranking systems in assessing institutions outside the Global North. Our comparative analysis of 23 Turkish research universities, implemented using Python and scikit-learn, resulted in three distinct clusters that reflect diverse patterns of institutional development. This clustering approach allows for a nuanced comparison of university performance within Turkey's higher education landscape, while also connecting to global debates on university rankings and performance metrics. A focused examination of Istanbul University-Cerrahpasa illustrates how this method can inform targeted improvement strategies, offering insights applicable to institutions in similar contexts worldwide. By moving beyond traditional rankings, this approach facilitates data-driven decision-making in higher education policy and institutional strategy.