Although some news outlets are reporting that Leiden University has launched a ‘new’ university ranking, in fact this institution has just released an ‘improved’ version of the ranking exercise it first introduced back in 2007/2008, but which has not been very visible in the interim. The new iteration of the Leiden Ranking aims to provide highly accurate measurements of the scientific impact of universities and their involvement in scientific collaboration. In other words, it focuses only on the research performance of universities (to the exclusion of other types of research organisations, but including teaching hospitals). Furthermore, Leiden offers an exclusively discipline-based ranking, and only for the sciences and social sciences. The arts and humanities are explicitly excluded from the exercise, meaning that if your university has a strong profile in these areas, you may stop reading this article now.  

Also purposely excluded are such subjective factors as a reputational survey and self-reporting from universities. These improvements made to the Leiden Ranking are meant to address doubts that plague the better-known Academic Ranking of World Universities, and even the previous versions of the Leiden Ranking itself. A detailed methodology is published on the ranking website as a sign of transparency, explaining with examples: 

• the introduction of an impact indicator based on the proportion of top 10% publications; 
• collaboration indicators based on geographic distances; 
• fractional counting of collaborative publications; 
• the possibility of excluding non-English language publications; and 
• stability intervals.

The new indicators, while expected to improve the accuracy of the ranking itself, are sending contradictory messages with regard to research impact (by favouring established journals over open access journals not ranked highly in the journal hierarchy), collaboration (by counting long-distance collaboration and discounting co-authourship), and language of research (by recognising the English citation bias through exclusion of non-English publications). Meanwhile, its ‘multi-rank’ approach - which gives users the ability to rank institutions according to their own criteria of interest – may enhance the relevance of this particular ranking tool. However, this depends greatly on how well-informed the users are about the indicators.     

Leiden Ranking

University World News