September 20, 2010

Analyzing the World University Rankings

Print More

It is that time of year when the London-based company Quacquarelli Symonds releases its annual update on the popular and controversial World University Rankings. Surely enough, the 2010 rankings were reported in The Sun on Sept. 17. Cambridge overtook Harvard at the top spot this year. Cornell fits in relatively nicely as 16th in the world and 12th in the nation.But a closer look at the ranking raises some questions. A surprising total of four English schools are ranked top-seven in the world, and perhaps more remarkably, the “quality” of these schools has ostensibly improved dramatically in the past six years. When Q.S. first launched its rankings in 2004, Cambridge, University College of London (UCL), Oxford and Imperial College were ranked sixth, 34th, fifth and 14th in the world, respectively. These schools are now purportedly first, fourth, sixth and seventh in the world. The four schools collectively rose by 41 spots — an unbelievable average of 10 spots per school — in just six years.I decided to conduct a pilot study to test the Q.S. claims. Considering that winning the Nobel Prize is a widely recognized indicator of academic excellence, positive contributions to the society and school reputations as a whole, I decided to use the number the number of Nobel Prizes won by each institution to see if they corresponded with the QS list. Now, according to Q.S., the top four U.K. institutions are all at the level of Harvard, Yale, MIT and the University of Chicago, and are superior to Caltech, Princeton, Columbia and Penn. If true, such rivalry should be approximately reflected in their Nobel Prize count as well. In an analysis of a time-series data of universities grouped according to their Q.S. rankings, the results are quite surprising. The U.K. schools were ahead in the early 1900s but began to lag behind after the ’50s. Most notably, the prize performances of the four U.K. schools have dramatically declined in the past three decades, with just three awards won altogether. During the same time period, the U.S. “rivals” have collected 37 prizes, and the “lower ranked” U.S. schools have won 26 prizes altogether. Of course, small differences may be reconciled by arguing the shortcomings of prize as a “complete” measure, but differences of this magnitude cannot easily be justified. It is a little bit like insisting that a tennis player with three Grand Slam titles is better than another player with 26 titles.In fact, the Q.S. methodology has been widely panned by academics worldwide. Here are some of the most critical problems. First, the survey respondants need not have had any direct experience with the universities they rate. This is as ridiculous as asking people to rate the quality of airline services without having been on them. Second, the survey response rate is abysmal at around one percent, which makes it prone to serious sampling problems. Third, since the survey is conducted by an English company, it naturally attracts higher response rates from places like the U.K., Australia and Hong Kong, and thus leads to more generous reviews of institutions in these places. This survey constitutes half of the ranking scores. The other 30 percent are, in the words of an influential British economist and a former Bank of England’s Monetary Policy Committee David Blanchflower, “laughable as they tell us zero about quality … This is an index that penalises the best to help the mediocre.” Even the citation-based index that complete the remaining 20 percent seems to lack quality control, as an obscure Gwangju Institute of Science and Technology received perfect score alongside Caltech and MIT.We must be cautious with rankings like these. Q.S. appeals to the public by showing what the readers worldwide want to see rather than what best reflects the truth. English readers want to believe that four of their universities are the very best in the world. Australian readers want to believe that Australian National University is 20th in the world. As a son of a professor at Chinese University of Hong Kong, I also want to believe that the university I grew up in is on par with New York University. But believing this would only make me an ignorant or dishonest person. As Cornellians let us be astute and ignore these rankings. And to Q.S.: There is no doubt that Cambridge and Oxford are world-class universities, and UCL and Imperial are great schools, but as Blanchflower puts it best, “the U.K. is not home to four of the top ten universities in the world, sorry.”Lawrence Jin is a junior in the College of Arts and Sciences. He may be reached at ljj9@cornell.edu. Guest Room appears periodically this semester.

Original Author: Lawrence Jin