Wednesday, July 29, 2015

Google Scholar Ranking of African Universities

As competition in the ranking world intensifies, Times Higher Education (THE) and Quacquarelli Symonds are  diligently promoting their various regional ranking, data processing and event management projects. The latest is THE's African summit at the University of Johannesburg.

Three weeks ago THE issued what they described as an "experimental and preliminary" ranking which consisted of 15 universities ordered according to the number of citations per paper normalised for field and year. An interesting innovation was that citations were fractionalised so that participants in large collaborative projects would be credited in proportion to their fraction of the total contributors .

This is just one indicator and it is not really a measure of research influence, but rather of research quality and it is still skewed by participation in multi-contributor papers in medicine and particle physics. It is unlikely that the University of Port Harcourt or Universite Cadi Ayyad would  be in the top ten of any other indicator.

THE have indicated that they will add another 15 names to the list at the Johannesburg summit.

The table below was compiled for the purpose of checking on the claims of THE or other rankers that might attempt to evaluate African universities. It simply counts the number of results (2012-2014: exclude citations and patents) from a query to Google Scholar. Data was compiled on the 25th and 26th of July. The criteria for inclusion were being in the top 50 of the Webometrics rankings or the 15 universities in the THE list. The top university in any country not included was added from either the Webometrics or ic4u rankings.

This database includes papers, reports, theses and dissertations, conference proceedings and so on. It is certainly not a measure of research quality but rather of the volume of any activities connected with research.  In the case of the two Kenyan universities it probably reflects the size and inclusiveness of the university repositories.

One thing about the Google scholar list is that it confirms suspicions that the quality of Egyptian universities has been underestimated by the big name rankers. For further evidence one might look at data from social media such as LinkedIn or just contrast the aspirations of Egyptian students in the revolutions of 2011 and 2013 compared with those of students at the University of Cape Town and Durban University of Technology.

RankUniversityCountryGoogle Scholar
1  University of Cape TownSouth Africa17,000
2Cairo UniversityEgypt16,800
3University of PretoriaSouth Africa16,500
4  University of NairobiKenya16,400
5University of the WitwatersrandSouth Africa15,800
6University of Kwazulu-NatalSouth Africa15,500
7Stellenbosch UniversitySouth Africa14,900
8University of IbadanNigeria14,800
9University of South AfricaSouth Africa13,500
10Kenyatta UniversityKenya12,000
11University of JohannesburgSouth Africa11,200
12Makerere UniversityUganda10,400
13North West UniversitySouth Africa10,100
14University of GhanaGhana8,330
15Alexandria University Egypt7,610
16University of LagosNigeria7,220
17Rhodes UniversitySouth Africa7,210
18University of the Western CapeSouth Africa6,870
19Obafemi Awolowo UniversityNigeria6,800
20Mansoura UniversityEgypt6,480
21University of the Free StateSouth Africa6,400
22Addis Ababa UniversityEthiopia6,210
23Zagazig UniversityEgypt6,160
24American University in CairoEgypt5,770
25University of IlorinNigeria5,620
26Assiut UniversityEgypt5,580
27Kwame Nkrumah University of Science and TechnologyGhana5,080
28=University of ZimbabweZimbabwe4,830
28=University of Port HarcourtNigeria4,830
30University of BotswanaBotswana4,260
31University of ZambiaZambia4,240
32University of Dar Es SalaamTanzania4,120
33University of KhartoumSudan4,110
34Suez Canal UniversityEgypt3,670
35Tanta UniversityEgypt3,600
36Jomo Kenyatta University of Agriculture and TechnologyKenya            3,520
37Nelson Mandela Metropolitan UniversitySouth Africa3,490
38Covenant University OtaNigeria2,950
39Helwan UniversityEgypt2,940
40Benha UniversityEgypt2,570
41Minia UniversityEgypt2,390
42University of MalawiMalawi2,340
43Université Abou Bekr BelkaidAlgeria2,290
44Universityof TunisTunisia2,270
45=Université Kasdi Merbah OuarglaAlgeria2,240
45=Cape Peninsula University of TechnologySouth Africa2,240
47Université Cheikh Anta Diop de DakarSenegal1,950
48University of NamibiaNamibia1,760
49Universite de la ReunionReunion1,690
50Durban University of TechnologySouth Africa1,560
51University of MauritiusMauritius1,490
52Université d'Abomey-CalaviBenin1,460
53South Valley UniversityEgypt1,440
54Universidade Eduardo MondlaneMozambique   1,420
55Beni-Suef UniversityEgypt1,400
56Université Cadi Ayyad MarrakechMorroco1,370
57Université de OuagadougouBurkina Faso1,300
58University of RwandaRwanda1,270
59Université des Sciences et de la Technologie Houari BoumedieneAlgeria976
60Université de LoméTogo784
61Université de BamakoMali660
62Kafrelsheikh UniversityEgypt618
63University of SwazilandSwaziland615
64Université Félix Houphouët-BoignyIvory Coast590
65Université de KinshasaDomocratic Republic of the Congo558
66National University of LesothoLesotho555
67Université Constantine 1 Algeria468
68Bejaia UniversityAlgeria413
69Universidade Jean Piaget de Cabo VerdeCape Verde407
70Université Mohammed V SouissiTunisia361
71National Engineering School of SfaxTunisia271
72Université Marien NgouabiRepublc of Congo256
73University of LiberiaLiberia255
74Université Djillali LiabesAlgeria243
75Université Abdou Moumouni de NiameyNiger206
76Misurata UniversityLibya155
77Université Omar BongoGabon138
78University of The GambiaGambia130
79Universidade Católica de AngolaAngola115
80=Université de DschangCameroons113
80=Université de BanguiCentral African Empire113
82Université de NouakchottMauretania108
83Eritrea Institute of TechnologyEritrea76
84Université de DjiboutiDjibouti66
85Université de ToliaraMadagascar59
86Université Hassan II Ain ChockMorocco55
87University of SeychellesSeychelles52
88Mogadishu UniversitySomalia51
89Universidad Nacional de Guinea EcuatorialEquatorial Guinea40
90Universite Gamal Abdel Nasser de ConakryGuinea21
91University of MakeniSierra Leone18
92John Garang Memorial UniversitySouth Sudan12
93Hope Africa UniversityBurundi3
94Universite de Moundou Chad2

Tuesday, July 28, 2015

Would anyone notice if a small, old but awfully clever dog filled in a university ranking survey, and would it make a difference?

The Australian newspaper The Age has a piece by Erica Cervini on how she allowed her dog to complete the QS academic reputation survey on the quality of veterinary schools.

She doesn't elaborate on how the dog chose the schools. Was it by barking or tail wagging when shown pictures of the buildings?

Seriously though, she does have a point. Can QS stop people signing up just to support their employer or outvote their rivals?

To be fair, QS are aware that their surveys might be manipulated and have taken steps over the years to prevent this by such means as forbidding respondents from voting for their declared employer or repeat voting from the same computer. Even so, it seems that some universities, especially in Latin America, are getting scores in the reputation surveys that appear too high,especially when compared with their overall scores. In the employer survey the Pontifical Catholic University of Chile is 56th (overall 167) and the University of Buenos Aires 49th (overall 198). In the academic survey the University of Sao Paulo is 54th (overall 132 and the National Autonomous University of Mexico 55th (overall 175).

QS are apparently considering reforming their citations per faculty indicator and allowing unchanged responses to the surveys to be recycled for five instead of three years. This is welcome but a more rigorous overhaul of the reputation indicators is sorely needed.

Thursday, July 23, 2015

Perfect Storm Heading for Tokyo Metropolitan University

Seen on the Times Higher Education website today:

Tokyo Metropolitan University
World's Best University
Scored a Perfect 100.00 for Two Years in Citations Sector
From TMU to the World

Tokyo Metropolitan University got its perfect score largely because it was one of hundreds of institutions contributing to a few publications from the Large Hadron Collider project. In their recent experimental African rankings THE started using fractionalized counting of citations. If THE use this method in the coming world rankings then TMU will surely suffer a dramatic fall in the citations indicator.

I would not like to be the president of TMU on September 30th.

Recommended Reading

Anybody interested in educational policy, especially the never ending campaign to close gaps of one sort or another or the oddities of university rankings should take a look at chapter four of Jordan Ellenberg's How not to be wrong: The  power of mathematical thinking which is about the obvious -- or ought to be obvious observation -- that smaller populations are more variable.

He notes that South Dakota is top of the league for brain cancer while North Dakota is near the bottom. What makes the difference? It is just that the bigger the population the more likely it is that outliers will be diluted by a great mass of mediocrity. So, extreme scores tend to crop up in small places or small samples.

Similarly when he tossed coins ten at a time he came up head counts ranging from 3 to 9 out of ten.

When he tossed them 100 at a time he got counts ranging from 45 to 60.

When he (actually his computer program) tossed them 1,000 times, the counts ranged from 462 to 537.

It is worth remembering this when a study with a double digit sample is published showing the latest way to close one of achievement gaps or a very small school in a rural state somewhere starts boosting the test scores of underperforming students or a few test takers reveal that the national IQ is imploding. Or the studies fail to be replicated, if indeed anyone tries.

Or university rankings that show very small or very unproductive institutions having an enormous research impact measured by citations.

Saturday, July 18, 2015

In the QS BRICS rankings nearly everybody gets a prize

There is a growing trend towards specialised and regional university rankings. The magic of this is that they can provide something for almost everybody. QS recently published its latest BRICS rankings which combined data from five very different university systems. The result was a triumphant (almost) success for everybody (almost).

Here are some highlights that QS could use in selling the BRICS rankings or an expanded version.

Russian universities are ahead of everybody else for teaching quality.

The top 21 universities in the BRICS for Faculty Student Ratio (perhaps not a perfect proxy for teaching excellence) are Russian, headed by Bauman Moscow State Technical University. Imagine what Russian universities could do if QS recognised the importance of teaching and increased the weighting for this indicator.

India performs excellently for Faculty with a Ph D.

Out of the top 15 for this category, ten are Indian and all of these get the maximum score of 100. Of the other five, four are Brazilian and one Chinese. If only QS realised the importance of a highly qualified faculty, India would do much better in the overall rankings.

South Africa takes five out of the first six places for international faculty.

China has four out of five top places for academic reputation and employer reputation.

Meanwhile a Brazilian university is first for international faculty and another is third for academic reputation.

It seems that with rankings like these a lot depends on the weighting assigned to the various indicators.

Yes, going to the library might be good for you

A study by Felly Chiteng Kot and Jennifer L. Jones has found that

"using a given library resource was associated with a small, but also meaningful, gain in first-term grade point average, net of other factors."

But, correlation does not necessarily mean causation. Could it be that bright people like to go to libraries?

Still, most students are likely to behave better in the library than other places, so let's not quibble too much.

Wednesday, July 15, 2015

Implications of the THE African Pilot Ranking

The most interesting thing about THE's experimental African ranking is the use of fractionalised counting of citations. This means that the total number of citations is divided by the number of institutions contributing to a publication. Previously, the method used in THE rankings was to assign the total citations of a paper to all of the institutions that contributed just as though each one had been the only contributor. This has produced some very questionable results with universities that were excellent but very specialised or just generally unproductive  scoring remarkably high scores for citations. Panjab university, Tokyo Metropolitan University, Scuola Normale Superiore Pisa, Federico Santa Maria Technical University and Moscow State Engineering Physics Institute have all had moments of glory in the THE rankings because of  citation scores that were dramatically higher than their scores for research or any other indicator or group of indicators.

The new method, if applied generally, is likely to see a significant reduction in the scores given to such universities. We can estimate what might happen by looking at the four universities that are included in both the African pilot ranking and last year's world rankings,  Cape Town, Witwatersrand, Stellenbosch and Universite Marrakesh Cadi Ayyad, Morocco.

In the world rankings these universities received scores of 86.6, 67.3, 45.6 and 83 respectively. The score of 83 for Universite Cadi Ayyad resulted very largely from its contributions to several publications from the Large Hadron Collider project, one of which has been cited over 2,000 times, a low overall output of papers and the "regional modification" that gave a big boost to low scoring countries. The scores for the three South African universities reflected a larger total output and citations over a broad range of disciplines.

In the African pilot ranking the scores for citations were Cape Town 90.90, Witwatersrand 99.76, Stellenbosch 95.48 and Cadi Ayyad 78.61. The high scores for the South African institutions reflect a much lower mean score than in the world rankings. 

The fall in Cadi Ayyad's citation score from 3.6 points below Cape Town to 21.3 below and its falling behind Stellenbosch and Witwatersrand presumably reflect the impact of fractionalised counting.

This suggests that if fractionalised counting is used in the coming World University Rankings many small or specialised institutions will suffer and there will be a lot of reshuffling. 

Thursday, July 09, 2015

The Top University in Africa is ...

... the University of Cape Town. What a surprise!

Times Higher Education (THE) has produced another "snapshot" ranking. This one is a list of 15 African universities ranked according to "research influence", that is the number of citations per paper normalised by field and year. It seems that a larger list will be published at a THE summit at the university of Johannesburg scheduled for the end of this month. Then, apparently, there will be discussions about full rankings with a broad range of indicators.

This is a smart move. Apart from diluting the impact of the QS BRICS rankings, this table puts the summit host in the top ten and gets attention from around the continent with three places in the north, two in the west and two in the east in the top fifteen.

Here is the top 15:

1.   University of Cape Town, South Africa
2.   University of the Witwatersrand, South Africa
3.   Makerere University, Uganda
4.   Stellenbosch University, South Africa
5.   University of Kwazulu-Natal, South Africa
6.   University of Port Harcourt, Nigeria
7.   University of the Western Cape, South Africa
8.   University of Nairobi, Kenya
9.    University of Johannesburg, South Africa
10.  Universite Cadi Ayyad, Morrocco
11.  University of Pretoria, South Africa
12.  University of Ghana
13.  University of South Africa
14.  Suez Canal University, Egypt
15.  Universite Hassan II, Morrocco.

This is, of course, just one indicator but even so there will be a few academic eyebrows rising around the continent. Makerere has a good national and regional reputation but does it have more research influence than all but two South African universities?

How come Suez Canal University is there but not Cairo University or the American University in Cairo? And I am sure that in Nigeria there will be a lot of smirking around Ahmadu Bello and Ibadan Universities about Port Harcourt in sixth place.

One very good thing about this "experimental and preliminary ranking" is that THE and data provider Scopus are now using fractionalised counting of citations, so that if 100 universities contribute to a publication they each get credit for one hundredth of the citations.

That has not stopped Makerere and Port Harcourt from getting a boost, perhaps too much of a boost, for taking part in a huge multinational medical study but it has reduced the distortions that this indicator can cause.

So, for once, well done THE!... Now, what about taking a look at secondary affiliations?

Sunday, July 05, 2015

Is this really happening?

If this continues France will be the least intelligent country in the world in a century.

Drawing straight lines on graphs and getting excited about tiny samples can be dangerous. Even so, this is a little frightening.

James Thompson's blog notes a study by Edward Dutton and Richard Lynn that suggests that the French national average IQ declined by nearly 4 points in a decade. The sample size was only 79 so we should not start panicking too much until there are a few more studies. It will be interesting to see if this one is replicated.

Thursday, July 02, 2015

Now the British Academy Sees a Problem

Yesterday I referred to the poor numeracy skills of British (England and Northern Ireland) tertiary graduates reported by the OECD.

Now the British Academy has had its say. It reports that the performance of British school pupils is mediocre and that many undergraduate students are weak in statistics.

But it looks like middling (compared to the OECD average) secondary school students become almost rock bottom (in the OECD) tertiary graduates. Could it be that British universities are actually subtracting relative value from their students?

The Academy notes:
Our school pupils tend to be ranked only in the middle of developed nations in mathematics. Our undergraduates embark on degree courses with varying, and often weak, fluency in statistics. And, in the workplace, demand for more advanced quantitative skills has risen sharply in the past two decades.
Perhaps this has something to do with relatively high graduation rates at British universities so that mediocre students with weak numeracy skills will be recorded as tertiary graduates while their counterparts in most of the OECD will drop out and remain classified as secondary graduate. Even if that were the case the mediocrity of secondary students and tertiary graduates would still need to be addressed.

The Academy proposes a strategy that includes improving the quality of quantitative skills teaching, reviewing school curricula and addressing the early dropping of maths by secondary school students.

I suspect that that will be insufficient.

Wednesday, July 01, 2015

Today India, Tomorrow Japan, Then ....

The ranking businesses are extending their global tentacles. Times Higher Education has produced a "snapshot" MENA ranking that produced interesting results -- Texas A&M University Qatar top for research impact -- and will be announcing their world rankings from Melbourne.

Meanwhile, QS will be in India next week to reveal their latest BRICS rankings and has been getting attention in new places for its subject rankings that get to places other rankings won't go.

With QS getting more international, it is no surprise to hear that Mitsui & Co, Ltd, has purchased shares in QS:

'Nunzio Quacquarelli, CEO of QS said the investment from Mutsui “can especially support our development in Asia” adding, “we were seeking and have found a likeminded company which shares our long term vision” '.
 This is not the first sign of Mitsui's interest in tertiary education:
'Last year the company also invested $5m in Synergis Education, an education company specialising in online and on the ground adult learner programmes.
“We aim to use our experience in the online education field to create new services,” said Takeshi Akutsu, GM of Mitsui Service Business Division in a statement.
“At the same time, through this business we will help to nurture the global human resources needed by the global economy.” '
I wonder if QS will try and start ranking online

What's the Problem with U-Multirank and AHELO?

In a recent post, I discussed the contrast between the poor skills of young people in the UK (strictly speaking Northern Ireland and England) and the high regard in which  British universities are held by the brand name rankers.

There is a piece of data in the skills report from the OECD that is interesting in this respect. Figure 2.2 shows the average numeracy skills of new graduates (age 16-29, 2012). It is depressing reading. The data for tertiary graduates shows that only Italy does worse than the UK and Ireland is either the same or almost the same.The US is very slightly ahead. The top scorers are Austria, Flanders and the Czech Republic.

Something that should have everybody running around doing research and forming  committees is that British tertiary graduates are only very slightly better than most European secondary graduates and slightly better than South Koreans with less than an upper secondary qualification.

It is possible, indeed quite probable, that British tertiary graduates do better on verbal skills and likely that they could conduct themselves well in interviews. Perhaps also, it is places like the University of East London and Bolton University that are dragging down the British average. But this dramatically poor performance is such a glaring contrast to the preening self satisfaction of the higher education establishment that some discussion at least is called for.

We may be seeing an explanation for the reluctance of the Russell Group and its orbiters and the Ivy League to cooperate with U-Multirank and their disdain for the AHELO project that is in marked contrast with their support for the trusted and prestigious THE rankings. They are quite happy to be assessed on reputation, resources, income and citations but comparison with the cognitive skills of graduates from the upstarts of East Asia and perhaps Eastern and Central Europe  is something to be avoided.