Sunday, September 17, 2017

Criticism of rankings from India

Some parts of the world seem to be increasingly sceptical of international rankings, or least those produced by Times Higher Education (THE). MENA (Middle East and North Africa) and Africa did not seem to be very enthusiastic about THE's snapshot or pilot rankings. Many Latin American universities have chosen not to participate in the world and regional rankings.

India also seems to be suspicious of the rankings. An article by Vyasa Shastri in the E-paper, livemint, details some of the ways in which universities might attempt to manipulate rankings to their advantage.

It is well worth reading although I have one quibble. The article refers to King Abdulaziz University recruiting faculty who would list the university as their secondary affiliation (now 41) when publishing papers. The original idea was to get top marks in the Shanghai Ranking's highly cited researchers indicator. The article correctly notes that the Shanghai rankings no longer count secondary affiliations but they can still help in the Nature and Science and publications indicators and in citations and publications metrics in other rankings.

Also, other Saudi universities do not recruit large numbers of secondary affiliations. There are only four for the rest of Saudi Arabia although I notice that there are now quite a few for Chinese and Australian universities, including five for the University of Melbourne.

Last word, I hope, on Babol Noshirvani University of Technology

If you type in 'Babol University of Technology" rather than 'Babol Noshirvani University of Technology' into the Scopus search box then the university does have enough publications to meet THE's criteria for inclusion the world rankings.

So it seems that it was those highly cited researchers in engineering that propelled the university into the research impact stratosphere. That, and a rather eccentric methodology.

Saturday, September 09, 2017

More on Babol Noshirvani University of Technology

To answer the question in the previous post, how did Babol Noshirvani University of Technology in Iran do so well in the latest THE rankings, part of the answer is that it has two highly cited researchers in engineering, Davood Domiri Ganji and Mohsen Sheikholeslami. I see no reason to question the quality of their research.

But I still have a couple of questions. First THE say that they exclude universities whose research output is less than 1,000 articles between 2012 and 2016. But checking with Scopus indicates that the university had 468 articles over that period, or 591 documents of all kinds including conference papers, book chapters and reviews, which seems way below the threshold level for inclusion. Is it possible that THE have included the Babol University of Medical Sciences in the count of publications or citations? 

Those documents have been cited a total of 2,601 times, which is respectable but not quite on a scale that would rival Oxford and Chicago. It is possible that some or one of those articles have, for some reason, got an unusual number of citations compared to the world average and that this has distorted the indicator score. If so, then we have yet another example of a defective methodology producing absurd results.




Friday, September 08, 2017

Why did Babol Noshirvani University of Technology do so well in the THE rankings?

The THE world rankings and their regional offshoots have always been a source of entertainment mixed with a little bit of bewilderment. Every year a succession of improbable places jumps into the upper reaches of the citations indicator which is supposed to measure global research impact. Usually it is possible to tell what happened  Often it is because of participation in a massive international physics project, although not so much over the last couple of years, contribution to a global medical or genetics survey, or even assiduous self-citation.

However, after checking with Scopus and the Web of Science, I still cannot see exactly how Babol Noshirvani University of Technology got into 14th place for this metric, equal to Oxford and ahead of Yale and Johns Hopkins, in the latest world rankings and 301-350 overall, well ahead of every other Iranian university?

Can anybody help with an explanation? 

Wednesday, September 06, 2017

Highlights from THE citations indicator


The latest THE world rankings were published yesterday. As always, the most interesting part is the field- and year- normalised citations indicator that supposedly measures research impact.

Over the last few years, an array of implausible places have zoomed into the top ranks of this metric, sometimes disappearing as rapidly as they arrived.

The first place for citations this year goes to MIT. I don't think anyone would find that very controversial.

Here are some of the institutions that feature in the top 100 of THE's most important indicator which has a weighting of 30 per cent.

2nd     St. George's, University of London
3rd=    University of California Santa Cruz, ahead of Berkeley and UCLA
6th =   Brandeis University, equal to Harvard
11th=   Anglia Ruskin University, UK, equal to Chicago
14th=   Babol Noshirvani University of Technology, Iran, equal to Oxford
16th=   Oregon Health and Science University
31st     King Abdulaziz University, Saudi Arabia
34th=   Brighton and Sussex Medical School, UK, equal to Edinburgh
44th     Vita-Salute San Raffaele University, Italy, ahead of the University of Michigan
45th=   Ulsan National Institute of Science and Technology, best in South Korea
58th=   University of Kiel, best in Germany and equal to King's College London
67th=   University of Iceland
77th=   University of Luxembourg, equal to University of Amsterdam












Thursday, August 24, 2017

Milestone passed

The previous post was the 1,000th.

Comment by Christian Scholz

This comment is by Christian Schulz of the University of Hamburg. He points that the University of Hamburg's rise in the Shanghai rankings was not the result of highly cited researchers moving from other institutions but the improvement of research within the university.

If this is something that applies to other German universities, then it could be that Germany has a policy of growing its own researchers rather than importing talent from around the world. It seems to have worked very well for football so perhaps the obsession of British universities with importing international researchers is not such a good idea..


I just wanted to share with you, that we did not acquire two researchers to get on the HCR List to get a higher rank in the Shanghai Ranking. Those two researchers are Prof. Büchel and Prof. Ravens-Sieberer. Prof. Büchel is working at our university for over a decade now and Prof. Ravens-Sieberer is at our university since 2008.

Please also aknowledge, that our place in the Shanghai Ranking was very stable from 2010-2015. We were very unpleasent, when they decided to only use the one-year list of HCR, because in 2015 none of our researchers made it on the 2015-list, which caused the descend from 2015 to 2016.

Guest Post by Pablo Achard

This post is by Pablo Achard of the University of Geneva. It refers to  the Shanghai subject rankings. However, the problem of outliers in subject and regional rankings is one that affects all the well known rankings and will probably become more important over the next few years


How a single article is worth 60 places

We can’t repeat it enough: an indicator is bad when a small variation in the input is overly amplified in the output. This is the case when indicators are based on very few events.

I recently came through this issue (again) with Shanghai’s subject ranking of universities. The universities of Geneva and Lausanne (Switzerland) share the same School of Pharmacy and a huge share of published articles in this discipline are signed under the name of both institutions. But in the “Pharmacy and pharmaceutical sciences” ranking, one is ranked between the 101st and 150th position while the other is 40th. Where does this difference come from?

Comparing the scores obtained under each category gives a clue

Geneva
Lausanne
Weight in the final score
PUB
46
44.3
1
CNCI
63.2
65.6
1
IC
83.6
79.5
0.2
TOP
0
40.8
1
AWARD
0
0
1
Weighted sum
125.9
166.6


So the main difference between the two institutions is the score in “TOP”. Actually, the difference in the weighted sum (40.7) is almost equal to the value of this score (40.8). If Geneva and Lausanne had the same TOP score, they would be 40th and 41st

Surprisingly, a look at other institutions for that TOP indicator show only 5 different values : 0, 40.8, 57.7, 70.7 and 100. According to the methodology page of the ranking, “TOP is the number of papers published in Top Journals in an Academic Subject for an institution during the period of 2011-2015. Top Journals are identified through ShanghaiRanking’s Academic Excellence Survey […] The list of the top journals can be found here  […] Only papers of ‘Article’ type are considered.”
Looking deeper, there is just one journal in this list for Pharmacy: NATURE REVIEWS DRUG DISCOVERY. As its name indicates, this recognized journal mainly publishes ‘reviews’. A search on Web of Knowledge shows that in the period 2011-2015, only 63 ‘articles’ were published in this journal. That means a small variation in the input is overly amplified.

I searched for several institutions and rapidly found this rule: Harvard published 4 articles during these five years and got a score of 100 ; MIT published 3 articles and got a score of 70.7 ; 10 institutions published 2 articles and got a 57.7 and finally about 50 institutions published 1 article and got a 40.8.

I still don’t get why this score is so unlinear. But Lausanne published one single article in NATURE REVIEWS DRUG DISCOVERY and Geneva none (they published ‘reviews’ and ‘letters’ but no ‘articles’) and that small difference led to at least a 60 places gap between the two institutions.


This is of course just one example of what happens too often: rankers want to publish sub-rankings and end up with indicators where outliers can’t be absorbed into large distributions. One article, one prize or one  co-author in a large and productive collaboration all of the sudden makes very large differences in final scores and ranks. 

Friday, August 18, 2017

Comment on the 2017 Shanghai Rankings

In the previous post I referred to the vulnerabilities that have developed in the most popular world rankings, THE, QS and Shanghai ARWU, indicators that have a large weighting and can be influenced by universities that know how to work the system or sometimes are just plain lucky.

In the latest QS rankings four universities from Mexico, Chile, Brazil and Argentina have 90+ scores for the academic reputation indicator, which has a 40% weighting. All of these universities have low scores for citations per faculty which would seem at odds with a stellar research reputation. In three cases QS does not even list the score in its main table.

I have spent so much time on the normalised citation indicator in the THE world and regional rankings that I can hardly bear to revisit the issue. I will just mention the long list of universities that have achieved improbable glory by a few researchers, or sometimes just one, on a multi-author international physics, medical or genetics project.

The Shanghai rankings were once known for their stability but have become more volatile recently. The villain here is the highly cited researchers indicator which has a 20% weighting and consists of those scientists included in the  lists now published by Clarivate Analytics.

It seems that several universities have now become aware that if they can recruit a couple of extra highly cited researchers to the faculty they can get a significant boost in these rankings. Equally, if they should be so careless to lose one or two then the ranking consequences could be most unfortunate.

In 2016 a single highly cited researcher was worth 10.3 points in the Shanghai rankings, or 2.06 on the overall score after weighting, which is the difference between 500th place and 386th. That is a good deal, certainly much better than hiring a team of consultants or sending staff for excruciating transformational sharing sessions

Of course, as the number of HiCis increases the value of each incremental diminishes so it would make little difference if a top 20 or 30 university added or lost a couple of researchers.

Take a look at some changes in the Shanghai rankings between 2016 and 2017. The University of Kyoto fell three places from 32nd to 35th place or 0.5 points from 37.2 to 36.7. This was due to a fall in the number of highly cited researchers from seven to five which meant a fall of 2.7 in the HiCi score or a weighted 0.54 points in the overall score.

McMaster University rose from 83rd to 66th  gaining 2.5 overall points. The HiCi  score went from 32.4 to 42.3,  equivalent to  1.98 weighted overall points, representing an increase in the number of such researchers from 10 to 15.

Further down the charts,the University of Hamburg rose from 256th  with an overall  score of 15.46 to  188th with a  score of 18.69, brought about largely by an improvement in the  HiCi score  from zero to 15.4 which was the result of the acquisition of tworesearchers.

Meanwhile the Ecole Polytechnique of Paris fell from 303rd place to 434th partly because of the loss of its only highly cited researcher.

It is time for ShanghaiRanking to start looking around for a Plan B for their citations indicator.









Wednesday, August 16, 2017

Problems with global rankings

There is a problem with any sort of standardised testing. A test that is useful when a score has no financial or social significance becomes less valid when coaching industries workout how to squeeze a few points out of docile candidates and motivation becomes as important as aptitude.

Similarly, a metric used to rank universities may be valid and reliable when nobody cares about the rankings. But once they are used to determine bureaucrats' bonuses, regulate immigration, guide student applications and distribute research funding then they become less accurate. Universities will learn how to apply resources in exactly the right place, submit data in exactly the right way and engage productively with the rankers. The Trinity College Dublin data scandal, for example, has indicated how much a given reported income can affect ranks in the THE world rankings.

All of the current "big three" of global rankings have indicators that have become the source of volatility and that are given a disproportionate weighting. These are the normalised citations indicator in the THE rankings, the QS academic survey and the highly cited researchers list in the Shanghai ARWU.

Examples in the next post.


Monday, August 14, 2017

Some implications of the Universitas 21 rankings

Universitas 21 (U21) produces an annual ranking not of universities but of 50 national university systems. There are 25 criteria grouped in four categories, resources, connectivity, environment and output. There is also an overall league table.

The resources section consists of various aspects of expenditure on tertiary education. Output includes publications,  citations,  performance in the Shanghai rankings, tertiary enrolment, graduates and graduate employment .

The top five in the overall rankings are USA, Switzerland, UK, Denmark and Sweden. No surprises there. The biggest improvements since 2013 have been by China, Malaysia, Russia, Saudi Arabia, Singapore and South Africa.

It is interesting to compare resources with output. The top ten for resources comprise six European countries, three of them in Scandinavia, Canada, the USA, Singapore and Saudi Arabia.

The bottom 10 includes two from Latin America, four, including China, from Asia, three from Eastern Europe, and South Africa.

There is a significant relationship correlation of .732 between resources and output. But the association is not uniform.  China is in 43rd place for resources but is 21st for output.  Saudi Arabia in the top ten for resources but 33rd for output. Malaysia is 11th for resources  but 38th for output.

I have constructed a table showing the relationship between resources and output by dividing  the score for output by resources and we get a table showing how efficient systems are at converting money into employable graduates, instructing students and doing research. This is very crude as is the data and the way in which U21 combines them but it might have some interesting implications

The top ten are:
1. China
2. USA
3. Italy
4. Russia
5. Bulgaria
6. Australia
7. UK
8. Ireland
9. Israel
10. Denmark

We have heard a lot about the lavish funding given to Chinese tertiary education. But it seems that China is also very good at turning resources into research and teaching.

The bottom ten are:

41. Austria
42. Brazil
43. Serbia
44. Chile
45. Mexico
46. India
47. Turkey
48. Ukraine
49. Saudi Arabia
50. Malaysia

At the moment the causes of low efficiency are uncertain. But it seems reasonable that the limitations of primary and secondary school systems and cultural attitudes to science and knowledge may be significant. The results of standardised tests such as PISA and TIMSS should be given careful attention.


Sunday, August 13, 2017

The Need for a Self Citation Index

In view of the remarkable performance of Veltech University in the THE Asian Rankings, rankers, administrators and publishers need to think seriously about the impact of self-citation, and perhaps also intra-institutional ranking. Here is the abstract of an article by Justin W Flatt, Alessandro Blassime, and Effy Vayena.

Improving the Measurement of Scientific Success by Reporting a Self-Citation Index

Abstract

: 
Who among the many researchers is most likely to usher in a new era of scientific breakthroughs? This question is of critical importance to universities, funding agencies, as well as scientists who must compete under great pressure for limited amounts of research money. Citations are the current primary means of evaluating one’s scientific productivity and impact, and while often helpful, there is growing concern over the use of excessive self-citations to help build sustainable careers in science. Incorporating superfluous self-citations in one’s writings requires little effort, receives virtually no penalty, and can boost, albeit artificially, scholarly impact and visibility, which are both necessary for moving up the academic ladder. Such behavior is likely to increase, given the recent explosive rise in popularity of web-based citation analysis tools (Web of Science, Google Scholar, Scopus, and Altmetric) that rank research performance. Here, we argue for new metrics centered on transparency to help curb this form of self-promotion that, if left unchecked, can have a negative impact on the scientific workforce, the way that we publish new knowledge, and ultimately the course of scientific advance.
Keywords:
 publication ethics; citation ethics; self-citation; h-index; self-citation index; bibliometrics; scientific assessment; scientific success


Saturday, August 12, 2017

The public sector: a good place for those with bad school grades

From the Economist ranking of British universities, which is based on the difference between expected and actual graduate earnings.

 That, as Basil Fawlty said in a somewhat different context, explains a lot.  

"Many of the universities at the top of our rankings convert bad grades into good jobs. At Newman, a former teacher-training college on the outskirts of Birmingham, classes are small (the staff:student ratio is 16:1), students are few (around 3,000) and all have to do a work placement as part of their degree. (Newman became a university only in 2013, though it previously had the power to award degrees.)

Part of Newman’s excellent performance can be explained because more than half its students take education-related degrees, meaning many will work in the public sector. That is a good place for those with bad school grades. Indeed, in courses like education or nursing there is no correlation between earnings and the school grades a university expects."