UK linguistics Research Assessment Exercise results: hard to be humble

« previous post | next post »

Everyone who's anyone in British higher education knows that today at one minute past midnight the results of the latest Research Assessment Exercise (RAE) were released. And as I believe I have occasionally mentioned here, in the small part of my life that is not devoted to Language Log, I moonlight as Head of Linguistics and English Language (known as LEL) at the University of Edinburgh. So you'll naturally want to know how well we did in the RAE. That's why I'm still up after midnight (Greenwich Mean Time).

Well, you can easily check the published details for yourself now, at the relevant RAE results web page, as soon as their server stops crashing (it was a bit over-excited just after midnight). So it would be silly for me to let my natural innate modesty hold me back. The truth is out there: LEL ranks absolute highest in the UK for the proportion of its work falling in the 4* "world-leading" category. And not only that, but its numbers are so strong that if you compute a sort of absolute volume of world-leading-research by multiplying the number of Edinburgh linguists considered in the exercise (36) by the percentage of their work that was considered world-leading (30%), you get a number (10.8) that cannot be matched even by adding together the figures for any two other departments of linguistics in the United Kingdom.

The closest approach would be by adding Lancaster's 6.34 to Newcastle's 4.27, to get 10.6. For any two other departments you get less. Adding three or four departments together generally yields less: for example, adding together Cambridge (10 * 20% = 2) and Oxford (16.5 * 10% = 1.65) and University College London (14.5 * 20% = 2.9) and York (13 * 20% = 2.6) would yield only 9.15 for the four programs combined.

According to the published figures, LEL at Edinburgh actually contributes about 22% of the entire output of 4*-quality research work done in the United Kingdom.

So what my colleagues here have built here (for I am just a relatively recent arrival) is not just a program that is top of the league table, but one that is stunningly good. Really, I'm a naturally shy and retiring guy (you know that), but facts must be faced.

The way the RAE worked involved a national panel which put out a call to all UK universities to pick out for each department or subject area those of their academic staff who were judged to be productive in research, and to submit four publications by each of those, plus various other data. In each subject a national panel worked for a year or so reading and grading everything submitted. They mixed in a score for the research environment and a small percentage for esteem indicators like major prizes and awards, and then presented the results as a vector of five integers: first, a rounded percentage corresponding to work of a quality that leads the world in originality, significance, and rigour (4*); then another corresponding to quality of international excellence but not at the very highest level (3*); another corresponding to internationally recognized quality (2*); another corresponding to nationally but perhaps not internationally recognized quality (1*); and a fifth corresponding to work falling below nationally recognized standards for research work (Unclassified).

LEL at Edinburgh entered everyone on the permanent staff as research productive (departments were allowed to simply leave some people out of the accounting, lowering the number of people submitted but raising the average quality, but we didn't do that), and submitted four selected publications by each person. And the result, after we had spent just over a year on tenterhooks, was this vector:

4* 3* 2* 1* U
30% 30% 30% 10% 0%

LEL at Edinburgh was the only linguistics program to reach 30 in the 4* category. And, with its 36 productive linguists, it is also in absolute terms (by count of academic staff members entered on the return) the largest linguistics program in the country. Perhaps the most amazing thing is that this is only part of the story about Edinburgh. Next door to us is the astonishing School of Informatics, where there are so many computational linguists and other language and speech researchers that they would be very highly ranked in their own right if they decided to call themselves a linguistics department.

So what should be the advice given to an ambitious student who is considering the possibility of studying linguistics in the United Kingdom, either at undergraduate or graduate (what British universities call postgraduate level? It is going to be hard for us to be humble. We will try, of course: my colleagues here come from a variety of different countries around the world, but many are British, and thus very modest and self-effacing despite their brilliance. But with these figures out, even these shy people will have to admit, if pressed, that if you want to study in the biggest language sciences community in the U.K., and the best one as judged by volume of work judged to be of world-leading quality, it looks like you should make plans to head for Edinburgh.

Comments are closed.