Students, academics and companies placing research contracts all need to know which are the best universities in the world. And the measures used to identify them are crucial, explains John O'Leary.

Higher education has become so international that it is no longer enough for the leading universities to know that they are ahead of the pack in their own country.

Students are prepared to look abroad for the best course, even at undergraduate level; firms scour the world to place research contracts; and academics are more mobile than ever. When the newly merged Manchester University was launched last month, among the goals in its first strategic plan was to become one of the top 25 universities in the world. But who is to say which those are and - crucially - how they should be identified?

Domestic league tables are controversial enough, but there are extra pitfalls associated with international comparisons. The rankings that start on the page opposite represent a first attempt to compare the world's top universities in the round. The process has been kept simple, partly because so few indicators of quality in ÖйúAƬ translate reliably across borders, but also to avoid any suggestion that the data have been manipulated to produce a particular outcome. The five indicators have been chosen to reflect strength in teaching, research and international reputation, with the greatest influence exerted by those in the best position to judge: the academics. University staff from every continent have given their verdict on the top institutions in their field, rather than delivering a more impressionistic judgement of quality across the board. Subsequent features will identify the leaders in different disciplines, but here we examine the aggregated results of the survey.

ÖйúAƬ

Other measures were considered and discarded for a variety of reasons.

Some, such as a survey of graduate recruiters, may be revisited in future but produced too limited a response to be reliable. Others, such as spending on libraries, were too closely linked to national prosperity. Some proved impossible to compile because of a lack of comparable data.

ÖйúAƬ

Alan Gilbert, Manchester University's president - a prime example of the globalisation of ÖйúAƬ, having been headhunted from Melbourne University - identified the recruitment of Nobel laureates as one indication of international excellence for his institution. But the leading academic prizes were another factor omitted from our tables to make them as contemporaneous and consistent as possible.

Nobel prizes and Fields medals account for almost a third of the points in the list of top universities compiled this year by Shanghai Jiao Tong University. But why count only these prizes? And why credit the universities where prizewinners studied, some at the turn of the century before last? Why, indeed, credit universities where winners carried out their research, often at least 20 years previously, rather than the institution that now benefits from their presence?

The Shanghai list also awards a fifth of its points on the basis of articles published in Science and Nature , thereby conferring a big advantage on universities with strengths in the areas covered by these journals. A further 40 per cent rides on two overlapping citation indices, with a final 10 per cent devoted to a complex measure compensating for the advantages enjoyed by big universities. The Times Higher ranking rates universities as they are now, or at least as they were at the time of the most recently published statistics. The use of citations and staffing levels helps institutions dominated by the sciences, but the measures are as neutral as possible. When the next rankings are published in 2005, more improvements will no doubt have been made.

It will take a big change to shift Harvard University from top place, however. Strong performances on all five measures confirm what most observers have long suspected: that Harvard is in the position to which all leading universities aspire. The riches of its endowment will make the university hard to challenge, but its performance is not simply a matter of money. A reputation for being the best in the world acts as a magnet for the most talented students and staff.

ÖйúAƬ

Other positions in the table are less predictable, and no doubt some are the result of quirks in the methodology or the different ways statistics are compiled worldwide. But despite taking seven of the top ten places, US institutions are certainly less dominant than most would have predicted.

The strong showing by the University of California, Berkeley will encourage other public universities but, across the Atlantic, so will the presence of Oxford and Cambridge universities and ETH Zurich in the top ten. The peer review, in particular, demonstrates that there are highly regarded universities in many parts of the world. Japan, Australia, China and Singapore all have representatives in the top 20. And even Australians may be surprised to find six of their universities in the top 50 - more than any country except the US and the UK.

Where scores are close, as they are lower down the table, there is no suggestion that one university is definitively better than another.

However, the ranking offers a snapshot of the leading institutions on a set of criteria that are valued around the world.

ÖйúAƬ

Statistics section  

World university rankings 2004
Back to index page

Download full-colour version
(331K)