The Importance of Stability in World University Rankings

Source: http://s0.geograph.org.uk/geophotos/01/69/80/1698049_fe2c325e.jpg

The QS World University Rankings for 2012/13 appeared last month. How good a guide are they to the changing fortunes of top universities?

We believe that the rankings have achieved a level of stability that makes them a valuable tool for students, and for university managers and other education professionals. The changes made to the Rankings over time, notably the introduction of the employer survey in 2005,have been few and simple. We think that this stability makes it possible to use the Rankings to track genuine change.

As an example of the Rankings’ high level of stability, look at this year’s top 20 institutions. 19 of them were in the top 20 in 2011. The only new entrant is Toronto, up from 23 to 19. It replaces Edinburgh, down from 20 to 21. This is a modest change and we can easily identify the reasons for it. Toronto has improved in several of our quantitative measures of university performance, including its attractiveness to international staff and students and its faculty/student ratio. By contrast, Edinburgh’s figures show little significant change. This reinforces our belief that universities can only stay level in the Rankings by improving steadily.

The stability of our Rankings also allows us to comment on how the Massachusetts Institute of Technology has replaced Cambridge in top spot. The top few institutions in the Rankings all score highly on every measure. Otherwise they would not be there to start with. But MIT has improved its international faculty score tangibly since 2011. By contrast, the only measure on which Cambridge has improved materially is citations per faculty member. But here it still remains well behind MIT, at 29th spot in the world compared to MIT’s 15th.

It is noticeable too that there are few very large jumps up or down in the rankings measures for the top universities. The bigger changes are mostly found among the more modestly ranked institutions. Even with the Z-score system which we use to remove the damaging effect of statistical outliers, the scores for lower-ranked institutions bunch more closely than for higher-placed universities. So big change is more likely lower down. The scores for the top 20 universities run from 100 for MIT to 89.5 for Duke in 20th, while the scores for the 400th and 420th institutions range from 25.9 to 24.7.

This stability allows us to illuminate key questions such as the changing status of Asia’s top universities. We see that Peking and Tsinghua, China’s top institutions, have improved their attraction to employers this year, and have seen modest improvements in their overall ranking. Hong Kong University, down one place to 23 but still the top Asian institution, has had a big fall in its citations count and is 215 in the world on this measure. This indicator shows up here as HKU’s biggest weakness. Japan’s top university, Tokyo, is seventh in the world for academic opinion, but produces little cited research (down eight places to 99) and remains poor at attracting overseas staff and students, a chronic issue with all Japanese universities. And the striking rise of KAIST in Korea, from 90 to 63, is driven by improvements in academic and employer opinion as well as in citations.

We expect the Rankings to provide more insights like these in coming years. We also expect their stability to continue. Partly this is because of their scale. The academic and employer surveys this year involved 46,079 and 25,564 responses respectively, a massive number which makes big shifts in opinion unlikely. Only 14 top-100 institutions moved by 10 or more places in the academic survey between 2011 and 2012, although there were some bigger shifts in the employer survey. The citations, faculty/student ratio and international measures are all based on significant data-gathering which we have been doing on a consistent basis for some years.

We are not resistant to ideas for improving these rankings, but we do intend to continue them in something like their present form. Rather than changing them drastically, we intend to go on adding to them, as we have already done with our Asian and Latin American rankings, the World University Rankings by Subject and the Best Student Cities ranking.

More QS Insights

Sign up for industry insights

Receive the latest insights, expertise and commentary on the topics which matter most in higher education, straight to your inbox.

Sign up