Thursday, August 18, 2005

World university rankings


Update (13 November 2005)

I have a post about the 2005 rankings by the Higher Education Supplement (THES) of the London Times.


Update (26 August 2005)

Check out an earlier ranking (pdf) of world universities by the (London) Times published in November 2004, using a different set of criteria. It features the 'Indian Institute of Technology' (location is not mentioned, so it is probably the entire IIT system) at the 41st position.

How much significance do you attach to these rankings? Particularly, if this exercise is done on an international scale? Even if you, like me, don't really give a damn, do you (again, like me ;-) want to get a sneak peek at who is where in the pecking order? You might be interested in looking at the 'much awaited' Academic Ranking of World Universities - 2005 (the third in the series), published by the Institute of Higher Education, Shanghai Jiao Tong University.

Harvard, Cambridge, Stanford, UC-Berkeley, MIT, Caltech, Columbia, Princeton, University of Chicago and Oxford are the top 10 universities. Overall, the top 25 list features nineteen universities from the US, and two from Asia (Tokyo and Kyoto).

Three Indian universities make it to the top 500 list: IISc (World rank: 301-400, Asia-Pacific rank: 37-65), IIT-Kharagpur (401-500, 66-93) and University of Calcutta (401-500, 66-93).

12 Comments:

  1. Anonymous said...

    Why do they consider articles published in Nature and Science to be very important? Most of the people who do these rankings do.

  2. Anonymous said...

    Well....for what ever it's worth......they needed some methodology to compile the rankings, and Nature and Science are taken to be benchmarks in the sciences for "highest impact" publishing, if you go by citations and "coolness" of research. A majority of pathbreaking work goes there....

    But that ofcourse is not true if you are in engineering (Computer science or something). So, i guess they must have taken other indicators (they have a disclaimer for social sciences and humanities, so they might have something similar for other fields).

    Abi....I dont give a damn either....but couldn't resist taking a peek at the pecking order...have link, must click.

    Aren't I quite relieved to see that i'm a part of the higher pecking order? Now, FINALLY, I can sleep in peace after doing research for all these years in obscurity :-)))))

  3. Abi said...

    Vishnu, Sunil: thanks for your comments. You both nailed one of the main problems with the Shanghai ratings: high weightage (20 %) given to Nature and Science. A related problem -- and IMHO an even more egregious problem -- is the weight (30 %) assigned to Nobel laureates (among both alumni and faculty). There are at least two reasons:

    First, when you give such a large -- 30 % is HUGE! -- weightage to actual Nobels and none at all to all those pieces of work that almost made it to a Nobel prize, what you have is essentially a 'winner takes all' methodology.

    The second problem is the small number (5 to 10) of Nobel prizes given out every year. If Stanford get one this year and Cambridge gets none, their relative ranking could change next year. Similarly, if JNU, say, wins a Nobel this year, I would bet that it would just get right into the top 100 or 200! A good ranking system should not allow one-time events and outliers to influence relative positions so strongly.

    A similar problem exists with the criterion using publications in Nature and Science; however, the effect should be much smaller, because there are a much larger number of papers (50 issues * 50 papers * 2 = 5000 papers as an upper bound) in these journals every year.

  4. Abi said...

    It is getting curiouser and curiouser.

    Take a look at the grades received by the University of Calcutta which (together with IIT-KGP) has a rank of between 401 and 500. It gets a relative grade of 15.4 and 12.9 for alumni Nobel and faculty Nobel, respectively! On the other hand, it has zero for Nature and Science publications, as well as for highly cited researchers.

    Hmmm..., interesting, isn't it?

  5. Anonymous said...

    A good ranking system should not allow one-time events and outliers to influence relative positions so strongly.

    Absolutely! The Nobel factor can change the ranking overnight. I guess there should be other criteria which make the ranking scheme more consistent.

    If I had chosen my grad school by someone else's rankings, I would have been elsewhere.

  6. Anonymous said...

    If I had chosen my grad school by someone else's rankings, I would have been elsewhere.

    Indeed Vishnu.......I chose my grad school only based on the research going on here......which I could see by the record my boss has over the years.

    But....though most of these rankings are very far from perfect....it does give a decent indication of who the "big boys" are......most of the schools in the top 50 list are research powerhouses. Which one is better than the other is completely subjective after that, and bottles down to specific interest.

    For what ever it's worth.....:-)

  7. Anonymous said...

    Hi! Abi:

    Commenting for the first time on your Blog (which I enjoy).

    A couple of quick observations on the World University Rankings...

    1. Look at the over-all scores of Univs. Harvard has 100, the next best (Cambridge) has only 73.6, followed by a bunch which are above 70. While Harvard may be good, I do not think it is SO good, for this huge difference.

    2. I know Rockefeller Univ. (rank 30)is a highly respected univ. But, it is also a highly specialized university, primarily looking into biological sciences. Given this, it is surprising that it is ranked 30. Can explain it also. Since it is into Biology, most of they do have a lot of publications in journals like Nature and Science. Also, they have one or two Nobels on their roster.

    Therefore, giving too much weightage to Nobels and Nature papers skwes the ranking significantly, as you guys have pointed out already. But, it doesn't surprise me, since these rankings are coming out of Shanghai Jiatong, which itself is not a great Univ. by any means. There is a new-found desire in many Asian univs to benchmark themselves against the best in the world and Harvard and Cambridge are readily on the lips of people who enforce this bench-amrking (recall PC's budget speech).

    Ram

  8. Anonymous said...

    THES had given a rank of 4 to the "Indian Institutes of Technology" in Engineering. Now, are the rank-hungry people happy?

  9. Anonymous said...

    I am really against the rankings given by these people. IISc is more into Engineering and it has its own impact worldwide. There are two more universities in Asia,namely, Nanyang Technological University and National University of Signapore,which did not find its place in top 100.
    Moreover its difficult to rank a university on certain parameters.As each university listed in the ranking has their own strengths and weakness. It all depends on our perception to discern these rankings.I personally feel that IISc should be in higher rank.

  10. Anonymous said...

    bogus!!

  11. Anonymous said...

    The subject of ranking universities or colleges is subjective. It really depends upon the pain and patience of survey team members have. It is not possible to evaluate Indian Universities/colleges in one year. In my opinion it is all bogus and done to the advantages of few.

  12. Anonymous said...

    The ranking is very subjective in its own manner but i also feel that the top 100 universities in the thes ranking is worth to consider if you want to get a more than average level of education...

    http://worldbest-universities.blogspot.com