Sunday, January 31, 2010

New Report

A comprehensive and interesting report on university rankings from the Swedish National Agency for Higher Education is available here.

Thanks to Beerkens' Blog

Saturday, January 30, 2010

Something from QS

Finally something has appeared on the QS Intelligence Unit Blog. Ben Sowter writes:

"The QS World University Rankings will continue to be published in 2010, albeit through a number of new channels which we are working on. At present, there are no plans to alter the methodology, in fact it seems important to maintain some comparability in a time when a number of new and different interpretations are going to emerge. So in 2010, we are focused on improving our engagement with institutions, redesigning some of our data collection systems to be more user-friendly and intuitive, and our work in specific regional and discipline oriented contexts."

I am not sure that keeping the methodology is a good idea but it is understandable. However, even with the same basic methods there are a couple of minor changes that might help QS find a niche in the "holistic" ranking market as Times Higher appears to focus on making fine distinctions among leading research institutions. One would be to use the academic survey to ask about general excellence or activities other than research. The other would be to remove non-teaching faculty from the faculty totals when calculating faculty student ratio. As it is, the QS rankings are heavily weighted towards research, with an academic survey asking about research, an indicator based on citations and a teaching resources measure that includes researchers who never teach.

Now that QS have done an Asian ranking and are apparently preparing Arab and Latin American ones, they could also also outflank THE by preparing survey forms in additional languages. They offered a Spanish option last year. They ought to have the resources to produce forms in Chinese, French, German. Arabic and Japanese.

Friday, January 29, 2010

Opinion Surveys in University Rankings

In this week's Times Higher Education, Phil Baty discusses the role of reputational surveys in university ranking. It was a distinctive feature of the THE-QS rankings that they devoted 40 % of the weighting to a survey of academic opinion about the research excellence of universities. Baty points out that "The reputation survey used in the now-defunct Times Higher Education-QS World University Rankings was one of its most controversial elements: a survey of a tiny number of academics should not determine 40 per cent of a university's score".

It was not so much that a tiny number of academics was surveyed but that a tiny number responded and that this (relatively) tiny number was heavily biased towards particular countries and regions. A very obvious effect of the survey was to boost the position of Oxford and Cambridge well beyond anything they would have attained on indicators based on other more objective factors.

Whether THE can produce a better survey remains to be seen. But at least they have at last stopped calling it a peer review.

Tuesday, January 26, 2010

The Rise of China

An article in the Financial Times describes the impressive growth of scientific research in China

"China has experienced the strongest growth in scientific research over the past three decades of any country, according to figures compiled for the Financial Times, and the pace shows no sign of slowing.

Jonathan Adams, research evaluation director at Thomson Reuters, said China’s “awe-inspiring” growth had put it in second place to the US – and if it continues on its trajectory it will be the largest producer of scientific knowledge by 2020.

Thomson Reuters, which indexes scientific papers from 10,500 journals worldwide, analysed the performance of four emerging markets countries: Brazil, Russia, India and China, over the past 30 years."

In contrast, the performance of Indian universities and institutes has been rather limp:

" A symptom of this is the poor performance of India in international comparisons of university standards. The 2009 Asian University Rankings, prepared by the higher education consultancy QS, shows the top Indian institution to be IIT Bombay at number 30; 10 universities in China and Hong Kong are higher in the table.

Part of India’s academic problem may be the way red tape ties up its universities, says Ben Sowter, head of the QS intelligence unit. Another issue is that the best institutions are so overwhelmed with applications from would-be students and faculty within India that they do not cultivate the international outlook essential for world-class universities. This looks set to change as India’s human resource minister has stepped up efforts to build links with US and UK institutions. "

A couple of observations. China's research output might not be so impressive if population were taken into account. I also wonder if India's relatively poor performance is the result of a failure to cultivate an international outlook. Is China really so much more international than India? Is it possible that other factors are more important?

Monday, January 18, 2010

Batting for Britain

A notorious feature of the THE-QS rankings was its over-valuation of British and Australian universities. It would seem that Times Higher and Thomson Reuters are not really bothered by this. Indeed it looks like they are set on course to add to this bias in their new rankings, at least as far as British universities are concerned. An opinion piece by Jonathon Adams, the Director of Research Evaluation at Thomson Reuters, echoes previous comments in THE by lamenting the maltreatment of the London School of Economics in the old league table.

"The London School of Economics is generally agreed to be an outstanding institution globally. But how can we judge that? A lot of people would like to study there. If you wanted an informed opinion, you would consult the people who work there. A lot of people who have been there have gone on to great things. These are good indicators that the place is intellectually vibrant and delivers excellent teaching, and those values are endorsed internationally.

Good, but not perfect. Three major problems spring to mind. First, that quick summary tells us there are many ways in which we may value what a university does. It is a knowledge business and a source for teaching, research and dissemination to users. Second, the LSE is a specialist. Its astronomy is weak, so we need to consider subject portfolio. And, third, what will we measure? I need an informed expert to confirm my judgment, but as I can't send my expert to every institution, I need a proxy indicator (not a "metric": an indicator).

Our view of the LSE does not translate readily into anything useful unless we are careful and we make sure our information is appropriate. The LSE stood at only 67th in the last Times Higher Education-QS World University Rankings - some mistake surely? Yes, and quite a big one. LSE academics publish papers in social and economic sciences, which have lower citation rates than the natural sciences; so on the simple "citations per paper" used by QS in analysing the Scopus publications data, it slipped way down the list. Not a good way of comparing it with nearby King's College London, which has a huge medical school.

We need a lot more information than has typically been gathered before we can build an even halfway sensible picture of what a university is doing."

The problem with this is that there are many institutions that scored lower than LSE in the rankings that are agreed by some people somewhere to be outstanding. The “good indicators” raise more questions. A lot of people want to study at LSE. Is that because of its intrinsic merits or shrewd marketing? And who is the "you" who would consult the LSE? A lot of its alumni and alumnae have done great things? No doubt many have become MPs, civil servants, university administrators and CEOs but given the current moral condition of British politics and the performance of the British and European economies that might not be something to be proud of.

It is difficult to concur with the claim that LSE has been treated unfairly in previous rankings. In 2009 they were number five for social sciences and 32nd for arts and humanities. They got top marks for international faculty and international students and in the employer review. They did somewhat less well in the academic survey, which had a disproportionate number of respondents from Britain and Commonwealth countries with large numbers of British alumni and alumae, but that is surely to be expected when LSE excels in a very limited range of disciplines.

LSE also did badly in the citations per faculty indicator (not citations per paper – QS used that for their Asian rankings, not the world rankings) partly because it is a specialist social science institution and it is conventional in the social sciences to produce fewer papers and to cite them less frequently but also because LSE actually does not produce as much social science research, as measured by Scopus and ISi publications, as general institutions such as the Universities of Manchester, Birmingham, Harvard, Yale, Chicago, Toronto, Melbourne and Sao Paulo.

It is difficult to think of changes in the structure or content of the rankings that would benefit LSE but not a host of others. Giving extra weighting to social science publications is an excellent idea and would boost LSE relative to King’s College or Imperial College (I wonder if THE is prepared to let Imperial slip a few places) but it would probably help US state universities and European universities even more. Counting “contributions to society”, such as sitting on committees and commissions and boards of directors would help LSE a bit but might well help Japanese universities and the French grandes ecoles a lot more.

LSE is a narrowly based specialist institution and QS gave it as much as or more than it deserved by ranking it highly in the social science and arts and humanities categories and putting it in the top 100 in the general world rankings. It is good at what it does but it does not do all that much. It would be a shame if the rankings are going to be restructured to promote it beyond its real merits.

The other item is a Rick Trainor’s review of Robert Zemsky’s book Making Reform Work. In the course of his review Trainor, who is president of King’s College London, says that:

"Most fundamentally, while the US debate is premised on a clear and widespread belief in the great, if imperilled, merits of the US system, British opinion often pays too little attention to the successes of UK universities, even in comparison with their US counterparts. For example, British commentators often overlook UK universities' superior completion rates, the greater rigour concerning undergraduate assessment inherent in the existence of an external examiner system, their greater ability (allowing for the much greater size of the US population and its university system) to attract overseas students and, as suggested by the Sainsbury report, their arguably superior record in commercialisation.

Of course, this is not to suggest that the UK higher education system is perfect, any more than US universities are. Nonetheless, there has been too little recognition in the UK of its high international research standing (aided by rises in public investment in recent years), despite persisting American strength and rapidly rising competition from countries such as China and India. Likewise, the UK system receives too little credit domestically for its success in protecting standards despite the huge increase in UK student numbers during the past 25 years. Similarly too few observers on this side of the Atlantic have learned one of the basic lessons propounded by Zemsky: that outstanding achievement in higher education depends on adequate resources - for teaching (which was substantially underfunded, even before the UK's public expenditure crisis began) as well as for research. "

This is a rather odd set of claims. Superior completion rate? I wonder how that happened. Greater rigour because of the external examiner system? Really? Do British universities still have a high international research standing? Just look at their performance on the Shanghai rankings, after removing the cushion of the thirty percent weighting for Nobel and Fields laureates. Have standards really been protected? Would more money make any difference?

It is beginning to look as though an implicit consensus is developing in the British higher educational establishment that the rankings should reflect its self-serving view of the merits of British higher education and that they have an important role to play in fending off the economic crisis. It appears that THE and Thomson Reuters are only too happy to oblige.

Saturday, January 16, 2010

Is anything happening at QS?

Although there has been a lot of activity, so far mainly rhetorical, at Times Higher Education and Thomson Reuters about their forthcoming rankings, nothing has been heard from QS apart from an advert for a manager of a university ranking for Latin America and Iberia.

Nothing has been added to the 2010 ranking news page since December and Ben Sowter’s blog has been silent for a month.

Are they preparing a response to THE or are they just fading away?

Tuesday, January 12, 2010

The Kiplinger Rankings are Out

Kiplinger has produced its 2009-2010 ranking of US universities. This is very much a student consumer ranking that measures the value for money delivered by each institution. It is based on information about student debt, tuition costs, financial aid, gender ratio, class size and average SAT scores, among others.

There is no doubt a lot of room for argument about the validity of the data and how the indicators were weighted but this sort of index does seem very useful.

I am wondering if something like this can be incorporated into existing international rankings. A lot of Kiplingers's data would be difficult or impossible to obtain outside the US but information about things like tuition fees, gender ratio, class size, and number of books in the library is widely available.

The top five private universities are:

1. Caltech
2. Princeton
3. Yale
4. Rice
5. Harvard

The top five public schools (for out-of-state students) are:

1. SUNY Binghamton
2. SUNY College at Geneseo
3. University of North Carolina at Chapel Hill
4. University of Florida
5. College of New Jersey

Sunday, January 10, 2010

The European Union Rankings

The European Union is trying to develop a new ranking system to rival the existing ones. The motivation is fairly transparent. The object, as reported in the EUObserver is "to improve the ranking of European universities and improve Europe's economic power".

The EUObserver provides an excellent and succinct summary of the forces underlying the universities ranking boom.

"This means the rankings are increasingly receiving more attention for different specific purposes: Students use them to short-list their choice of university; public and private institutions use them to decide on funding allocations; universities use them to promote themselves; while some politicians use them as a measure of national economic achievements or aspirations. "

It seems that planning for the new rankings took place in the second half of 2009 and that in the first half of 2010 it will be tested on 150 institutions around the world, but only for engineering and business studies.

At that rate, THE, QS and Shanghai Jiao Tong University have nothing to worry about.

Global Institutional Profiles Project

Thomson Reuters have set up a new site here. It contains information, although not much so far, about the new Times Higher ranking system.

They will "address industry concerns over current profile systems... The 21st century research institution has many fluid layers, and Thomson Reuters is committed to developing an equally robust and dynamic dataset".

Notice that they are talking about research institutions as though universities do nothing but research and that they refer to higher education as an industry.

The page provides some hints about what might be included in the forthcoming rankings: peer review, scholarly outputs, citation patterns, funding levels and faculty characteristics.

I do not know whether there is any significance in the absence of internationalisation and faculty student ratio from the list.

The page could have done with some editing. There are too many barely meaningful adjectives -- robust, dynamic, flexible, data-driven, globally significant. And exactly what is a "fluid layer"?

Saturday, January 09, 2010

Webometrics Out Soon

The next edition of the Webometrics rankings will be published at the end of January. Watch this space.

Friday, January 08, 2010

The THE Reputational Survey

Thomson Reuters, acting on behalf of Times Higher Education, have published an open letter to university administrators announcing the development of a new ranking system. They promise much. The new ranking is the only one that "seeks to fundamentally change the way data is collected and analyzed". They believe "this development underscores a major breakthrough within the rankings dialogue".

There is some good news. Finally, the inaccurate term "peer review" is being dropped to be replaced by "reputational survey". Also, according to a comment on a previous post from Phil Baty, Deputy Editor at THE, "we will be looking to focus the survey more on non-research elements. It allows us to get at the less tangible elements of university activity that can not be measured through numbers." This is very sensible.

The two points above are welcome but I still do not see anything very revolutionary about the forthcoming survey.

There is another question. Thomson Reuters are asking university administrators to encourage their researchers and colleagues to take part. This would seem to introduce an element of bias into the survey from the very start. How many university administrators will read the open letter? How many will act on it? Will there be as many in Japan as in England?

Wednesday, January 06, 2010

What does this say about Nigerian Universities?

According to University World News, Nigerian banks prefer to recruit holders of polytechnic diplomas rather than university graduates. One bank manager said that diploma holders could perform most of the tasks normally done by graduates for less pay and did not require extensive computer literacy training.

I wonder if this would show up in any of the current university rankings.
Something about Ipsos MORI

Anyone interested in Ipsos MORI, the company appointed to conduct a survey of academic opinion for Times Higher Education can go here or have a look at the column to the left.

It seems that they have a number of junior staff outside the UK, or at least a lot of telephone interviewers, so that does to some extent allay one of my concerns about the company.

However the biodata for the senior staff is rather disconcerting. Some snippets:

"after graduating from Oxford University"
"has worked closely with both Conservative and Labour ministers ... as well as a wide range of local authorities and NHS trusts"
"served as Finance Director of BMRB"
"started her career at the BBC"
"has been a User Fellow at the centre for the Analysis of Social Exclusion at LSE and spent time working in the Prime Minister's Stategy Unit"
"a member of the MRQSA council"
"a full member of the Chartered Institute of Marketing"
"member of the council of Roehamton University"
"has chaired a number of round-table discussions with senior peers"
"has a BSc, MSc and MBA from Imperial College"

Very British (and just a little bit cosmopolitan -- "speaks five European languages", "always busy cookng up the next plan to explore to a far flung destination"), very establishment, rather politically correct and perhaps a little inward looking -- in much of the world, working with British government ministers, peers and the NHS is not something you would want to boast about.

Will a survey carried out by such a group reveal that in most respects places like Oxford, LSE and Imperial College are performing increasingly less well than leading American and Japanese universities?

Sunday, January 03, 2010

In case you thought ranking was just a bit of fun


The National University of Taiwan is protesting about a statement on the webometrics site that some universities had resorted to 'bad practices."

The practices consisted of hosting papers written by authors at other institutions. As well as NTU, webometrics referred to the University of Sao Paulo.

Other universities are listed as having more than one webdomain. These include the University of Maryland, the University of Manchester, Yonsei University, Korea University, Chiang Mai University, The Indian Institutes of Technology at Delhi and Kharagpur, Kuwait University and the University of Bahrein