On The Necessity Of Multiple University Rankings

Nowadays university rankings are ubiquitous commodities; a plethora of them is published every year by private enterprises, state authorities and universities. University rankings are very popular to governments, journalists, university administrations and families as well. At the same time, they are heavily criticized as being very subjective and contradictory to each other. University rankings have been studied with respect to political, educational and data management aspects. In this paper, we focus on a specific research question regarding the alignment of some well-known such rankings, ultimately targeting to investigate the usefulness of the variety of all these rankings. First, we describe in detail the methodology to collect and homogenize the data and, second, we statistically analyze these data to examine the correlation among the different rankings. The results show that despite their statistically significant correlation, there are many cases of high divergence and instability, which can be reduced by ordered categorization. Our conclusion is that if, in principle, someone accepts the reliability of university rankings, the necessity and the usefulness, of all of them is questionable since only few of them could be sufficient representatives of the whole set. The overabundance of university rankings is especially conspicuous for the top universities.