A league of their own

Universities are complicated beasts, with multiple products and multiple stakeholders. It is folly to think that any one metric can ever completely or even adequately capture what goes on. That said, there is a large and booming business in metrics. And a large and booming industry in debunking them for good or evil. But, there they are and there they will stay. Rankings are not cardinally useful perhaps : what does 8th best mean?  They are and should be seen as ordinal with 8th better (whatever that means) than 80th, but the degree of ordinality might be one to ponder (what is the difference between 8th and 10th anyhow?)

For what its worth, the QS rankings are a decent attempt to move the issues along, relying on a mix of academic peer review (asking in essence other academics to say what schools and universities they think are good), employer feedback (would you hire someone with a degree from this school) and citations (a flawed but ‘acceptable’ measure of research impact). Not only do they do this for universities as a whole, but perhaps more intriguingly they do so for subject areas. Their methodology is well described and complex.

The rankings are made up with different weightings – some areas have a higher weight on one of the areas above than other, reflecting an attempt to , for example, recognize that citations are more relevant for science than humanities, and for some areas longer histories will give secularly higher ‘peer review’ simply from visibility. Is this perfect? By no means. Is it better than nothing? I suggest so. Here are the weights by subject area: it makes little sense to count citations in Performance art, while it makes a lot more sense in Medicine say.

Approx 2500 institutions were  ranked in some way in the mix, of the c 20,000 universities and third/fourth level institutions estimated to exist, which suggests that appearing in the top 200 is quite an achievement in and of itself.

Where do we find ourselves: below are some overall rankings by subject for 2011 (details of rankings below 50 not given except by broad category e.g. 51-100).

There is good news for irish higher education in the rankings but its not perhaps the news that the government would hear. Smart green nanobots are all well and good but we do not have massive evidenced university strength in these areas. We do have strength in some social and humanities areas, as well as emergent clusters. For what they are worth, see below. I have bolded the areas in which an Irish university (ok, ok, TCD the only one) appears in the top 50 worldwide

• Sociology:  TCD 48; UCD 51-100
• Statistics and OR: no Irish university in top 200, which is worrying as this is a key business analytics function and one that has great job opportunities.
• Politics and International Relations: TCD 50, UCD 51-100
• Law: TCD, QUB, UCD and UCC all in the 51-100 ranking, indicating a massive strength in legal scholarship on the island. We rarely hear of law as an area in which the government is going to invest, but clearly we have strength and it would make sense to build on this.
• Economics and Econometrics: TCD and UCD are both ranked in the 51-100 area, and it is interesting that there is in place a jointly taught PhD programme in this area. Again it shows a cluster, Dublin based, of world ranked excellence in this area, which if one were to take the ESRI into account is even stronger
• Accounting and Finance: TCD 51-100, QUB and UCD 100-151; some potential here it would seem especially when one considers the close linkages with economics. I might also modestly note, being in that area, that there are significant size differences between the three schools – TCD has 4 finance faculty and 3 accounting, the others have…more
• English: TCD 32, UCD 51-100
• Modern Languages TCD 51-100
• History: TCD 39, QUB and UCD 51-100
• Philosophy: TCD and UCD both 51-100
• Geography: TCD and UCD 51-100, QUB 151-200
• Linguistics: no Irish university was ranked in the top 200 here
• Computer Science and Engineering: UCD 51-100, TCD 101-150,
• Civil and Structural Engineering: TCD and UCD 100-151, QUB 151-200
• Chemical Engineering: TCD QUB and UCD 51-100, NUIG 101-150, UCC 151-200
• Electrical/Electronic Engineering: TCD and QUB 51-100, UCC 100-150, DIT (its only entry in a subject in the top 200) 151-200, also NUIG and UCD here
• Mechanical Engineering: TCD and UCC 51-100, NUIG 101-150, UCD and QUB 151-200, again another engineering cluster of excellence
• Medicine: TCD 51-100, UCD 101-150, UCC and QUB 151-200. Interesting that there is no mention of the RCSI here…
• Biological Science: TCD 51-100, UCD and UCC 101-150, NUIG 151-200,
• Psychology: TCD 48, UCD 51-100, QUB 151-200
• Chemistry: TCD 36, QUB 51-100, UCD 101-150, UCC 151-200, which when combined with the chemical engineering rankings suggests that this is an area in which as an island we punch heavily worldwide
• Physics: TCD 49, UCC QUB and UCD 151-200
• Metallurgy and Material Science: UCD 151-200
• Mathematics: TCD 15, a fantastic achievement indeed…UCD 101-150, UCC and QUB 151-200
• Environmental Sciences: TCD, UCD and UCC all 101-150, QUB 151-200
• Earth Science: TCD 101-150, NUIG, UCC and UCD 151-200

Very obviously, on these data anyhow, there is a clear distinction between TCD and UCD on the one hand (and I would suggest that TCD have an edge there as they alone gain rankings in the top 50 worldwide on a number of subjects) and the other universities and institutes on the other hand. Neither NUIM nor UL nor DCU   feature which might suggest that they form a separate pool, with the remaining universities in the middle.

This is of course deeply unfashionable, flies in the face of the rush towards ‘one size fits all’ , has implications for the idea of spreading the loot to all and sundry (which in a resource constrained environment means relative starving of the top at the expense of others) and would suggest that we already have two world class universities across a broad range of fields. If Ireland really wants to thrive in the knowledge economy, the findings here might suggest some clusters, both of areas and of universities, towards which funding and resources might flow disproportionally.

We might also wonder if it is a good way of aligning incentives to have the Prof of an area that is ranked by employers, academics and peers as being of the top class worldwide paid essentially the same as one that is to the same stakeholders all but invisible. Where is the incentive there?

This is not to say that we close or do not fund or downplay the other universities or areas ; it is to say that we reinforce success and hold it out as an example and challenge to the others. And no, we shouldn’t chase the rankings : we should chase excellence in research, in fitting employer needs, and in teaching, and that will cause higher rankings to emerge. Whether the present third level system of funding, resourcing and reward does this is another question.