The Student Room Group

The Times Higher Education World Reputation Rankings is released

Gap beyween UK Universities and the UK "Golden Triangle" widens

The University of Cambridge and the University of Oxford have been bumped down a place in the 2014 World Reputation University Rankings. With Harvard and MIT managing to maintain their first and second place, Stanford University knocked Cambridge off the number three spot, pushing Oxford to number five. The US clearly dominates the rankings with seven of the top 10 places being taken by American Institutions. You can see the full table at the Times Higher Education

Times.jpg

http://www.timeshighereducation.co.uk/world-university-rankings/2014/reputation-ranking


With 10 institutions in the rankings the UK still has the most top 100 representatives after the US. However, this has been interpreted by analysts as evidence that the gap between top UK institutions and the "Golden Triangle" Oxbridge and Imperial College London is continuing to widen. London’s Universities continue to perform well with six of them being in the top 100, more than any other city in the world.

04 - University of Cambridge
05 - University of Oxford
13 - Imperial College London
24 - London School of Economics and Political Science
25 - University College London
43 - King's College London
46 - University of Edinburgh
51-60 - University of Manchester
91-100 - London Business School
91-100 London School of Hygiene & Tropical Medicine

There are 20 countries represented in the world top 100 reputation list. Taking a global view there were a number of significant changes. Seoul National University and the Hong Kong University of Science and Technology were identified as rising stars in the world’s biggest academic reputation survey. Seoul rapidly moved through the rankings from 41st to 26th. This was mirrored across Asia with other Asian institutions making progress in the rankings. Singapore moved from 27th in 2011 to 21st in 2014. Although the University of Tokyo missed out on the top 10 for the first time since the rankings were established, Japan continued to remain Asia’s leading nation five 5 institutions in the rankings.

You can find the full TSR report here

Scroll to see replies

These league tables really should come with a massive health warning for anyone who is currently choosing a university - they have big limitations, which are explained rather well here http://www.thestudentroom.co.uk/showthread.php?t=1096372
Reply 2
Original post by She-Ra
Gap beyween UK Universities and the UK "Golden Triangle" widens

The University of Cambridge and the University of Oxford have been bumped down a place in the 2014 World Reputation University Rankings. With Harvard and MIT managing to maintain their first and second place, Stanford University knocked Cambridge off the number three spot, pushing Oxford to number five. The US clearly dominates the rankings with seven of the top 10 places being taken by American Institutions. You can see the full table at the Times Higher Education


It's not "Times Higher Education University Global Rankings" but "The Times Higher Education World Reputation Rankings", based on a survey among scholars worldwide asking "Which universities do scholars think are the best on the planet?".

World Reputation Rankings 2014 methodology

As research lab rankings, I think it's a fine list.
It's interesting that tiny specialist universities like LBS and LSHTM are still reputed among scholars (they are exceptionally good in their fields though).
Reply 3
The overall scores here (the dark blue bars in the RH column) look weird.

Harvard (rank 1) gets 100 - full marks
Oxford (rank 5) gets 67.8 - only 2/3rds as good
Imperial (rank 13) gets 20.9 - 1/5th as good
LSE (rank 25) gets 11.8 - lets be generous and say 1/8th as good
Carnegie Mellon (rank 29) the first in single figures with a score of 9.6
the last uni with a published score is Osaka (rank 50) with a measly 5.7 points

scores for 51-100 are 'data withheld', possibly someone realised it looked silly to have 50 unis splitting the remaining range of marks between 5.7 and 0 out of 100

----
the methodology is to ask selected academics to list upto 15 universities they think are good, regional rep rings are going to flourish whether or not they're being orchestrated
Original post by Joinedup
The overall scores here (the dark blue bars in the RH column) look weird.

Harvard (rank 1) gets 100 - full marks
Oxford (rank 5) gets 67.8 - only 2/3rds as good
Imperial (rank 13) gets 20.9 - 1/5th as good
LSE (rank 25) gets 11.8 - lets be generous and say 1/8th as good
Carnegie Mellon (rank 29) the first in single figures with a score of 9.6
the last uni with a published score is Osaka (rank 50) with a measly 5.7 points

scores for 51-100 are 'data withheld', possibly someone realised it looked silly to have 50 unis splitting the remaining range of marks between 5.7 and 0 out of 100

----
the methodology is to ask selected academics to list upto 15 universities they think are good, regional rep rings are going to flourish whether or not they're being orchestrated


This is right. Beyond the top 50 or so, the number of academics required to list a university as one of their 15 is so vanishingly small that rankings can be massively affected by a local chauvinism.

The top (i.e. most often) listed school, this year it is Harvard, is given a score of 100.0 and all other scores are adjusted against this. That is to say that Oxford was listed not by 68% of respondents but by 68% as many respondents as listed Harvard.

Let's suppose that fully half of those responding included Harvard in their listings. And absent the data I think that's an erring on the side of generosity - bear in mind that academics are allowed to list only in their own subjects and Harvard has no provision in inter alia Geography or Social Work. The survey question "I would send my students to graduate work in X at Y" requires that X is taught at Y. And in some subjects US academics in their fields know that Harvard is not top 15 even nationally.

So, if it is 9000 of the 18000 responding plumping for Harvard, then it's about 6000 listing Oxford (i.e. 68% of 9000). By the time we get to the last one for which the data is revealed, the number 50 ranked school, Osaka University in Japan, with 5.7% as many mentions as Harvard, the number of mentions can be supposed as approximately 500.

My strong suspicion is that responses from Japanese academics were about 500 as well*. Of course, none of them had to suggest Osaka as better than Harvard, the survey asks for only 15 names with no sense of their being ordered. So, 12 from overseas and a cheeky 3 from Japan will be fine. But it will mean as well that Osaka University is listed as number 50 in the world "for international reputation" despite being mentioned exclusively by Japanese.

Another function of the "not needing many to make the top 100" is that a university can be included toward the 'bottom of the top' for excellence in a single subject. And this is seen in the cases of the Karolinska Institute and LSHTM. If there were 200 respondents in epidemiology and public health then every single one of them will have listed LSHTM, it is inarguably top 3 in the world and rivalled only by HSPH and Johns Hopkins. So that puts LSHTM at number 100, which that school might see as unjust - because it is top 3 in the world in everything it does. And that is distorting in a way that perhaps doesn't flatter the university.

As against that, it might be the case that a more general university has a stand-out strength in, random example, Electronic Engineering, we could imagine this of a Korean school, such that it is listed by every Electronic Engineer surveyed but by no-one else. And now this school is in the 'top 100 universities worldwide', despite having just one department that is at all widely regarded.


*Another Fermi calculation would support this as likely the case: 500 is 1/36th of the respondents, which is about as much representation as Japan would be anticipated as having in a survery appropriately distributed.
(edited 10 years ago)
Reply 5
Original post by cambio wechsel
This is right. Beyond the top 50 or so, the number of academics required to list a university as one of their 15 is so vanishingly small that rankings can be massively affected by a local chauvinism.

The top (i.e. most often) listed school, this year it is Harvard, is given a score of 100.0 and all other scores are adjusted against this. That is to say that Oxford was listed not by 68% of respondents but by 68% as many respondents as listed Harvard.

Let's suppose that fully half of those responding included Harvard in their listings. And absent the data I think that's an erring on the side of generosity - bear in mind that academics are allowed to list only in their own subjects and Harvard has no provision in inter alia Geography or Social Work. The survey question "I would send my students to graduate work in X at Y" requires that X is taught at Y. And in some subjects US academics in their fields know that Harvard is not top 15 even nationally.

So, if it is 9000 of the 18000 responding plumping for Harvard, then it's about 5,500 listing Oxford (i.e. 60% of 9000). By the time we get to the last one for which the data is revealed, the number 50 ranked school, Osaka University in Japan, with 5.7% as many mentions as Harvard, the number of mentions can be supposed as approximately 500.

My strong suspicion is that responses from Japanese academics were about 500 as well*. Of course, none of them had to suggest Osaka as better than Harvard, the survey asks for only 15 names with no sense of their being ordered. So, 12 from overseas and a cheeky 3 from Japan will be fine. But it will mean as well that Osaka University is listed as number 50 in the world "for international reputation" despite being mentioned exclusively by Japanese.


*Another Fermi calculation would support this as likely the case: 500 is 1/36th of the respondents, which is about as much representation as Japan would be anticipated as having in a survery appropriately distributed.


I think your assumption is quite possible.

According to THE, there are 10,536 responses in this year (eng 22% phy 18%, med 16%, life 13%, soc 22% and art 9%).

So if 80% of the scholars list up Harvard (it's a more realistic number considering the ratio of fields which respondents participate), then the number of responses might be like this.

therep.png

Below 50th, as you said, regional reputation of a university can significantly affect its ranking position because of the small number of responses. But THE must consider statistical significance of sample data, so 91-100th universities should get at least over 100 responses (more likely over 200 responses).
(edited 10 years ago)
Original post by RussellG
It's not "Times Higher Education University Global Rankings" but "The Times Higher Education World Reputation Rankings", based on a survey among scholars worldwide asking "Which universities do scholars think are the best on the planet?".

World Reputation Rankings 2014 methodology

As research lab rankings, I think it's a fine list.
It's interesting that tiny specialist universities like LBS and LSHTM are still reputed among scholars (they are exceptionally good in their fields though).


Thank you for correcting me, I've edited that now! :smile:
Reply 7
Original post by Origami Bullets
These league tables really should come with a massive health warning for anyone who is currently choosing a university - they have big limitations, which are explained rather well here http://www.thestudentroom.co.uk/showthread.php?t=1096372


University league tables can be dubious but this one is the best imo, since its based on actually asking experts what they think the best universities are, rather than trying to come up with some BS weighting function that tries to trade off "percentage of international faculty" against "number of undergraduates getting a 2:1" or whatever nonsense the other tables use.

The reputation rankings done by employers (rather than academics) yield broadly similar results:

http://www.timeshighereducation.co.uk/news/global-employability-university-ranking-2013/2008497.article
http://www.nytimes.com/imagepages/2012/10/25/world/asia/25iht-sreducemerging25-graphic.html?ref=asia

As others have said, its probably not very trustworthy beyond the top 50-100 or so, since a small number of extra votes will have a very large effect on a university's ranking.
(edited 10 years ago)
Reply 8
The other reason why this table is sensible is that it doesnt change that much year-on-year. In reality, the quality of a university changes very slowly over time, so the tables which claim to have universities dropping/gaining 10 places in less than a year are generally stupid, since these big drops/gains are just usually a result of changes made to the weighting function (which shows how non-robust the whole process is).
Reply 9
Original post by poohat
The other reason why this table is sensible is that it doesnt change that much year-on-year. In reality, the quality of a university changes very slowly over time, so the tables which claim to have universities dropping/gaining 10 places in less than a year are generally stupid, since these big drops/gains are just usually a result of changes made to the weighting function (which shows how non-robust the whole process is).


Well, if you re-sort any national rankings by entrance difficulty, the ranking order is almost the same every year.

THE reputation rankings seem reasonable to some degree from the viewpoint of research standards of universities. But we never know who voted and how much they answered seriously. I'm more than sure that there are many patriots who anyway vote the universities in their countries or/and their own universities, because university staffs are not saints. (In fact, it seems several EA universities are overrated than their actual citation index scores for me.) So I take them with a pinch of salt as usual.
(edited 10 years ago)
Reply 10
Original post by RussellG
Well, if you re-sort any national rankings by entrance difficulty, the ranking order is almost the same every year.

THE reputation rankings seem reasonable to some degree from the viewpoint of research standards of universities. But we never know who voted and how much they answered seriously. I'm more than sure that there are many patriots who anyway vote the universities in their countries or/and their own universities, because university staffs are not saints. (In fact, it seems several EA universities are overrated than their actual citation index scores for me.) So I take them with a pinch of salt as usual.


I think east asian unis being on the list just reflects reality in those countries. You have to remember that Peking and Tsinghua Universities are the equivalent of Oxbridge for China, a country of 1.4 billion people, and as we know, Chinese people are very focused on education, they are extremely competitive to the point that it makes sense their reputation is high. What really pulls them back on the world rankings are citations, and because the language of research is English, there is a massive bias towards English speaking countries. Its difficult for a Chinese university that teaches in Chinese to have many professors who can publish a paper in English and this goes for every non english country in the world.


Look at where most of the world's universities on the top 100 Times list are. The vast majority are English speaking universities. The world reputation reflects reality much more. If rankings reflected student capability and intelligence, the UK would have only Oxbridge in the top 100 because there is an ocean of talent that is not accounted for all around the world that is not recognized because most world rankings rely on research which is written mainly in English.
Reply 11
Original post by ukmed108
I think east asian unis being on the list just reflects reality in those countries. You have to remember that Peking and Tsinghua Universities are the equivalent of Oxbridge for China, a country of 1.4 billion people, and as we know, Chinese people are very focused on education, they are extremely competitive to the point that it makes sense their reputation is high. What really pulls them back on the world rankings are citations, and because the language of research is English, there is a massive bias towards English speaking countries. Its difficult for a Chinese university that teaches in Chinese to have many professors who can publish a paper in English and this goes for every non english country in the world.


Look at where most of the world's universities on the top 100 Times list are. The vast majority are English speaking universities. The world reputation reflects reality much more. If rankings reflected student capability and intelligence, the UK would have only Oxbridge in the top 100 because there is an ocean of talent that is not accounted for all around the world that is not recognized because most world rankings rely on research which is written mainly in English.


Yeah I know how prestigious Peking and Tsinghua are. I know many Chinese people come to UK or US and study in top universities because Peking and Tsinghua (and other top 10) are too competitive to enter.

But the thing is, those reputation rankings are not about how smart students are, but how good research standards they have.

And because of the methodology, those 100 universities must have very strong fields in which their research standards are one of top 15 in the world.

From this viewpoint, Tsinghua has very strong computer science and engineering departments, so no wonder why they are listed in top 15 in those fields.

Tokyo and Kyoto have very strong natural science and life science departments as well, so also I'm not against them if they are listed in top 15 in those fields.

Peking is overrated compared to their publication and citation scores, but they can earn reputation scores in social subjects related to Asia, since they are the top social science and humanity university in China.

But I just felt weird about the ranking positions of NUS, SNU, University of Hong Kong, and Osaka. They have definitely several top 100 research fields, but no way to be listed in top 15. It's a little bit strange if many scholars write down those universities' names with other top research universities, unless respondents have some other intentions. And the ratio of EA scholars in all respondents is only 13%. So I smelled the scent of "patriotism" :rolleyes:. But such thing can also happen to other western universities.
(edited 10 years ago)
Original post by RussellG


According to THE, there are 10,536 responses in this year (eng 22% phy 18%, med 16%, life 13%, soc 22% and art 9%).


I do not know whether engineers are just good at completing surveys, but the make up of responses is not reflective of the composition of faculty in western countries (possibly Germany excepted).

Ultimately you get a Korean engineer's view of the best universities. That Korean engineer is only commenting on his own discipline but what it means is that those US universities with a strong engineering faculty are getting a block vote.

Arts, and to a lesser degree social sciences, travel poorly due to cultural and linguistic factors.

If you are a researcher in English local history, you will neither know nor care where the best research is being undertaken on Portuguese local history.

We study the literature of our own language in literature departments but generally we study the literature of other languages in foreign language departments. If you ask an American professor of Italian what are the best universities for the study of Italian, he may not name any university in Italy although almost certainly the best work on Italian literature is undertaken in Italian universities.

My guess is that possibly Harvard aside, there would be no congruity between the opinions of an American and a Spanish legal academic as to which were the best universities for law.

Interestingly no Russian university made the list at all but equally interestingly Russian was the one major teaching language in which the survey was not distributed.
(edited 10 years ago)
Original post by ukmed108
If rankings reflected student capability and intelligence, the UK would have only Oxbridge in the top 100

Slight exaggeration don't you think! What about Cambridge, Imperial college, UCL, LSE? I certainly wouldn't say the students there are less intelligent or less capable than the students at Oxford.
Reply 14
This is undeniable evidence of how some universities reputations is not warranted.

For instance, Manchester and KCL are considered inferior in the UK to universities like bath and Warwick, Durham is seen as a 3rd rival to Oxbridge.

I hope that people realise just what they're buying into when applying to universities, and not buying into the reputation that can often fool.

tldr: Durham is ****. This proves it.
Reply 15
Original post by nulli tertius
I do not know whether engineers are just good at completing surveys, but the make up of responses is not reflective of the composition of faculty in western countries (possibly Germany excepted).

Ultimately you get a Korean engineer's view of the best universities. That Korean engineer is only commenting on his own discipline but what it means is that those US universities with a strong engineering faculty are getting a block vote.

Arts, and to a lesser degree social sciences, travel poorly due to cultural and linguistic factors.

If you are a researcher in English local history, you will neither know nor care where the best research is being undertaken on Portuguese local history.

We study the literature of our own language in literature departments but generally we study the literature of other languages in foreign language departments. If you ask an American professor of Italian what are the best universities for the study of Italian, he may not name any university in Italy although almost certainly the best work on Italian literature is undertaken in Italian universities.

My guess is that possibly Harvard aside, there would be no congruity between the opinions of an American and a Spanish legal academic as to which were the best universities for law.

Interestingly no Russian university made the list at all but equally interestingly Russian was the one major teaching language in which the survey was not distributed.


I agree with your opinion.
Because scholars are not robots and have linguistic barriers, their perceptions cannot be away from being subjective and limited. And as you mentioned, Social Science and Art and humanity especially have this problem apart from several universal subjects like Economics and Business.

Btw, there is one Russian university in top 100 (51-60th Lomonosov Moscow State University).
Reply 16
Original post by Melthusa
This is undeniable evidence of how some universities reputations is not warranted.

For instance, Manchester and KCL are considered inferior in the UK to universities like bath and Warwick, Durham is seen as a 3rd rival to Oxbridge.

I hope that people realise just what they're buying into when applying to universities, and not buying into the reputation that can often fool.

tldr: Durham is ****. This proves it.


Because the domestic reputation is based on entrance difficulty (how smart the students are).
And the worldwide uni reputation is based on research outcomes (how great researchers and funding they have).

And they are often not equal, even in other countries.

For instance, very few people choose UW(31st) and Davis(52nd) over Brown(81-90th), Dartmouth(out of 100) and all unranked top LACs in the USA.
The offer letter from Sorbonne(71-80th) is not as great as the one from Science-Po(out of 100) in France.

I don't think Durham is ****, since their entrance level is very high (often more difficult than UCL). The problem of Durham is a lack of medical school causing a huge loss of research grants.
Reply 17
Original post by EveryNameIsTaken
Slight exaggeration don't you think! What about Cambridge, Imperial college, UCL, LSE? I certainly wouldn't say the students there are less intelligent or less capable than the students at Oxford.


I'd say that just because if you think of how many countries there are talent wise, the UK is only 60 million people. If there were 100 unis on the list, talent wise a huge chunk of those would be in China, India, US and then there would also be a number from Brazil, Germany, France, Russia, Poland, Turkey etc. You get my point. There would only really be enough room for perhaps 2 but perhaps 3 and possibly 4 because the UK draws in so many international students so yes maybe Imperial and UCL. I already said Cambridge.... at least you now know that Oxbridge means Oxford and Cambridge. A very basic fact on TSR.
Reply 18
Original post by Melthusa
This is undeniable evidence of how some universities reputations is not warranted.

For instance, Manchester and KCL are considered inferior in the UK to universities like bath and Warwick, Durham is seen as a 3rd rival to Oxbridge.

I hope that people realise just what they're buying into when applying to universities, and not buying into the reputation that can often fool.

tldr: Durham is ****. This proves it.


Times Higher Education Reputation gives you a better sense of how the world views UK universities.
Reply 19
hi, so what do you mean by that?
do you think that Manchester and Kcl are superior to Bath?
If we look at world rankings , we always see Manchester , Edinburgh..( i do not mention imperial and oxbridge of course....)

....:smile:
Original post by Melthusa
This is undeniable evidence of how some universities reputations is not warranted.

For instance, Manchester and KCL are considered inferior in the UK to universities like bath and Warwick, Durham is seen as a 3rd rival to Oxbridge.

I hope that people realise just what they're buying into when applying to universities, and not buying into the reputation that can often fool.

tldr: Durham is ****. This proves it.

Quick Reply

Latest

Trending

Trending