The QS World University Rankings for individual subjects were released today. As reported in the Irish Examiner, University College Dublin did particularly well at bringing subjects into the top 100 (now has 13) and Trinity College Dublin did particularly poorly (now has 2).
The overall ranking of universities is based on six indicators, four are produced from statistical data and two are based on reputational surveys. An academic reputation score is produced from the returns made by 74.6k academics, 3.8% identified themselves as being in the fields of Geography and Area Studies. For each of five main areas (Arts and Humanities, Engineering and Technology, Life Sciences and Medicine, Natural Sciences, Social Sciences and Management), respondents were asked to identify their own area of expertise and then to list up to ten universities from their own country (excluding their own employer) and 30 internationally that they consider excellent for research in this area. They do the same for their specialist field (from a list of 52, results were published for 46). This produces 40% of the overall score for each university and a variable percentage of the score for different subjects within universities. I received a request to complete this survey just recently but Maynooth, under any of its likely names, did not appear in the drop-down list of institutions that I could claim as an affiliation, and thus the questionnaire aborted. On another day it did.
44,000 responses from employers produced an employability measure with the measure again based on the ten domestic and 30 international universities considered excellent for recruiting, and also in which fields they mainly recruit. This gives 10% of the university’s weighted score. So, half the score is based on these subjective assessments. The benefit to older and larger institutions is evident.
These two factors are also included in the subject rankings. They are supplemented by two measures of research output. The first is citations per faculty, using data derived from Scopus (these scores are then normalized for individual disciplines). The other is the so-called H-Index. An author’s papers are ranked in order of citations in on other journals. H is the point before which the number of citations falls below the rank of the paper in the rank-order of papers based on citations for that person. If your fifth most cited paper has 12 citations and your sixth has five, then your H-Index is 5, the rank preceding the point at which citations fall below rank (these numbers are then normalized for individual disciplines).
For Geography 60% of the subject score comes from academic reputation, 10% from employer reputation, 15% from (normalized score of) citations per faculty, and 15% from the (normalized score of) average H-Index of faculty. Fully 70% of the Geography score comes from those measures the benefit the older and larger universities.
Perhaps not surprisingly Maynooth University has few disciplines that do well in the QS rankings. This year, Maynooth’s Geography and English Departments were only subjects placed in the top 200 for their subjects in the world rankings. We might conclude, then, that English and Geography at Maynooth did well to overcome the relatively low visibility of a small and new university both with global employers and global academics. The highest ranked Geography department in the Republic of Ireland is UCD.
The values for the separate elements of the subject score are given for the current rankings.
We can see that Trinity and UCD have very good reputations with employers, and that UCD has a strong reputation among academics. The Trinity score on this last measure is depressed by virtue of missing data due to incorrect data reporting in previous years. On the measure of citations per paper Maynooth, Queen’s Belfast and Trinity do well. The H-Index really seems to pick out Trinity geographers.
I have commented before on the way some of these rankings work and QS is more reliant on subjective evaluation of reputation that are other rankings, but – What is to be made of this? First, it is heartening that five Irish Geography departments are in the top-200. Second, if international visibility is to be so heavily influenced by data derived from Scopus, then, it will hurt journals not included there. Third, for international visibility to rely solely on journals surely presents a confusing picture since, as most of us know, in Geography the citation of books far outstrips the citation of individual articles, so for the measure of research productivity to ignore what is arguably the most direct form of academic influence is bizarre, however convenient for the companies that sell the journals.