While the majority of universities in the QS World University Rankings are comprehensive, specialist institutions have made a big impact in recent years.
Take a look at the leaders in this year's QS World University Rankings. Seven of the top ten are big, general universities, starting with Harvard in top spot. However, the remaining three are specialist technical universities: MIT, Caltech and Imperial College London.
Look further down the list and you will see that this pattern is repeated. Most of the places in our top 200 are taken by universities which are involved in pretty much the full range of subjects. At Harvard, for example, you can study Japanese, become an astronomer, or learn dentistry - and the same applies at most of the universities that we rank highly.
Capturing excellence
But in the top 200, as in the top ten, there is a significant group of universities whose members break this rule. Twenty-two institutions on our list have a name involving science, technology or engineering. But there is exactly one with a name which suggests an emphasis on the social sciences. It is LSE, the London School of Economics and Political Science, which appears in our 2009 rankings in 67th place.
This very rough examination of our rankings is not perfect. It conceals the reality that universities have a wide range of missions and specialisms. But for us as compilers of the rankings, it tells us that we are doing something right. We designed our system to capture excellence in broad-based academic institutions, and it seems that we have succeeded.
Major institutions dominate these rankings for a number of reasons. The first is that the biggest single element of our rankings scores, up to 40% of the possible total, comes from a peer review of the world’s academics, who are asked to list up to 30 universities whose work they esteem in their own field. We do our best to achieve balance among these academics between the five main areas of scholarly life: science, biomedicine, technology, the social sciences, and the arts and humanities.
If a university is not active in most or all of these areas, it is hard to get a big score on this measure.
We omit universities which do not teach undergraduates, and any that do not work in at least two of these five areas of academic life. This means excluding many fine institutions, some of which feature strongly in the Shanghai Jiao Tong University Academic Ranking of World Universities. Most are medical schools, such as the University of California at San Francisco.
A similar argument applies to the employer review that gives universities up to 10% of their score in our rankings. Many of the big employers we poll are looking for clever graduates in any subject, and are less attracted to specialist universities than to general ones.
Research leaders
So why do science and technology specialists defy the odds to do well in our rankings? Part of the reason is that a further 20% of our score is available for citations per staff member. This measure inevitably favors science and medicine.
In addition, the global nature of science research means that these universities bring in both staff and students from around the world, guaranteeing them a good showing in our measure of international attractiveness. Often they are well-funded. Governments recognize their importance as generators of innovation and employment, and they are uniquely good at bringing in industrial funding.
And while specialist arts universities tend not to have a physics department or a medical school, the lower cost of getting involved in the arts and social sciences means that many supposed science and technology universities do have at least a toehold in these subjects.
MIT’s School of Humanities, Arts and Social Sciences is active in music, languages and history. Such a school is especially important for US universities, where students are expected to take minor subjects unrelated to their main field of study.
Undoubtedly the biggest area in which science and technology universities get involved in the social sciences, though, is via business schools. MIT’s Sloan School is the best-known of the many business schools attached to a big-name technology university.
This leaves the question of just why LSE is where it is in our rankings. We know from many conversations that LSE representatives themselves find this a topic of unending interest.
Social science specialists
First, LSE should take heart. Things elsewhere are worse. In the Shanghai ranking, LSE appears in the 201-302 band, far below its position with QS. (Shanghai does not give the precise rank of universities this far down its ranking.)
The reason is that Shanghai’s system is very heavily dependent upon publications and impact in science and medicine, where LSE is not active. LSE appears in position 234 in the Webometrics ranking of universities on the web, despite its significant international visibility in politics, economics and business.
A look at the columns of data we give on LSE explains the problem. LSE has a perfect 100 score on three of our criteria: international staff, international students, and employer review. But its citations per faculty score was only 29.
The maximum score of 100 on this measure was obtained by ten institutions, including Harvard, as well as two science specialists, MIT and Caltech. Although LSE also does creditably on academic peer review for a specialist institution, and has a respectable but not stellar staff/student ratio, the other measure we apply, it is very hard to compensate for this gap.
LSE is not alone in having this problem. Just a kilometre away in London, the School of Oriental and African Studies suffers from the same syndrome, and there are other examples around the world.
Change to come?
Over time, it is possible that the citations issue will ease as social science research becomes more scientific, with more publishing and citation. Equally, we know that data on citations, publications and academic impact are simple to buy and manipulate.
This means that there may be more rankings systems which make use of them and in which institutions such as LSE find it hard to look good. But systems that use academic peer review will never reach a position in which a university with such a specialist remit as LSE can fairly be ranked alongside Harvard or University College London.
Several points arise from this analysis. One is that, as QS has often said, no one ranking system is able to cope with the full range of universities that are out there. Any system intended to capture excellence in the social sciences is certain to place the LSE near the top on a world scale. LSE is fifth in our 2009 ranking of universities for the social sciences, fifteenth in a 2004 citations-based league of politics departments, and eighth in a recent world ranking of economics departments.
The more important lesson is that it is important to read university rankings tables, not just to look at who comes where. It is obvious from LSE’s scores in the columns of data we provide that it is an extraordinary place which is loved by staff, students and employers, but also that it is unusual in its publishing patterns and in other ways.
Rankings prove that it is a world-class institution, but also give readers fair warning that its place in a system designed to rank full-service universities is bound to be only part of the story.