Friday, 11 October 2013

Going to Beijing and Fukuoka

On Sunday I’m going to East Asia for three weeks, mainly Beijing (China) but also a short trip to Fukuoka (Japan). Very much looking forward to it: I have been in Fukuoka a few times in recent years, but was only in Beijing once, in 1995: actually, the picture above was taken on that trip (I know I’m not a great photographer). At that time, there were still more bicycles than cars – let’s see how it will be now ...

Update: the chart below (from the very useful air quality website) provides a hint on how things are today...

Sunday, 6 October 2013

The discrepancy between the UK and global rankings of UK universities

Compare two recent rankings, the Times and Sunday Times University League Tables 2014 (one of the UK rankings) and the Times Higher Education (THE) World University Rankings 2013-14 (one of the global rankings):


rank
Times League Tables 2014
THE World Rankings 2013-2014
1
Cambridge
Oxford
2
Oxford
Cambridge
3
LSE
Imperial
4
St. Andrews
UCL
5
Imperial
LSE
6
Durham
KCL
7
Bath
Edinburgh
8
Exeter
Manchester
9
UCL
Bristol
10
Warwick
Durham

Bristol (15)
Edinburgh (22)
Manchester (26)
KCL (27)
St. Andrews (16)
Warwick (20)
Exeter (22)
Bath (38)

So, six universities are in both top 10, yet, all of them with different ranks; the other universities are not ranked very well in the other ranking, some of them not even in the top 20.  How to explain it? Wikipedia tells us that this can be attributed to the different methodologies:
“(i)nternational university rankings primarily use criteria such as academic and employer surveys, the number of citations per faculty, the proportion of international staff and students and faculty and alumni prize winners. The national rankings, on the other hand, give most weighting to the undergraduate student experience, taking account of teaching quality and learning resources, together with the quality of a university's intake, employment prospects, research quality and dropout rates.”
However, such an indicator-based explanation is not really helpful since many of the indicators may actually be proxies for something else: for example, employment prospects may not actually reflect the quality of the teaching but the reputation of universities based on their research; and the subjective opinion polls / surveys of both types of rankings are likely to be shaped by reasons which have nothing to do with the quality of the university in question.
  Thus, to get a better picture of the discrepancy, I would rather suggest to look at the differences between the actual rankings (see the table above): in the THE ranking larger and older universities tend to be higher ranked; universities that specialise on social sciences and humanities (as opposed to natural sciences) tend to be lower ranked; finally, possibly due to the use of international surveys, it may matter if a university is based in a big and internationally well-known city.
  Of course, the factors mentioned in the previous paragraph should not matter for a quality based ranking. Thus, this already makes the value of the international rankings (such as the THE one) fairly doubtful. In addition, the specific UK rankings (such as the Times League Table) are preferable since they can make use of better data available at the UK level (research assessment scores, student surveys etc), whereas international rankings have to rely on more shaky indicators since only those are available for all countries of the world.