Your SlideShare is downloading. ×
Classement Leiden Ranking
Upcoming SlideShare
Loading in...5
×

Thanks for flagging this SlideShare!

Oops! An error has occurred.

×

Introducing the official SlideShare app

Stunning, full-screen experience for iPhone and Android

Text the download link to your phone

Standard text messaging rates apply

Classement Leiden Ranking

329
views

Published on

Methodologie du classement international des institutions d enseignement superieur et de recherche elabore par l'Universite de Leiden (Pays Bas)

Methodologie du classement international des institutions d enseignement superieur et de recherche elabore par l'Universite de Leiden (Pays Bas)

Published in: Education

0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total Views
329
On Slideshare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
2
Comments
0
Likes
0
Embeds 0
No embeds

Report content
Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
No notes for slide

Transcript

  • 1. New developments in bibliometric methods: evaluation, mapping, ranking Ton van Raan Atelier Bibliométrie de l’URFIST de Paris 23 mars 2012 – CNAM Center for Science and Technology Studies (CWTS) Leiden University
  • 2. 2 Leiden University oldest in the Netherlands, 1575 European League of Research Universities (LERU) Within world top-100 12 Nobel Prizes Leiden, historic city (2th, 11th C.), strong cultural (arts, painting) & scientific tradition one of the largest science parks in EU
  • 3. Contents of this presentation: * Bibliometric methodology: • impact • maps * Leiden Ranking 2011-2012: new indicators * Evaluation tools related to the Leiden Ranking
  • 4. 4 WoS/Scopus sub-universe journal articles only,> 1,000,000p/y VOLUME 88, Number 13 PHYSICAL REVIEW LETTERS 1 April 2002 Truncation of Power Law Behavior in “Scale-Free” NetworkModels due to Information Filtering StefanoMossa,1,2 Marc Barthélémy,3 H. Eugene Stanley,1 and Luís A. Nunes Amaral1 1 Center for Polymer Studies and Department of Physics, Boston University, Boston, Massachusetts 02215 2 Dipartimento di Fisica, INFM UdR, and INFM Center for Statistical Mechanics and Complexity, Università di Roma “La Sapienza,” Piazzale Aldo Moro 2, I-00185, Roma, Italy 3 CEA-Service de Physique de la Matière Condensée, BP 12, 91680 Bruyères-le-Châtel, France (Received 18 October 2001; published 14 March2002) We formulate a general model for the growth of scale-free networks under filtering information conditions—that is, when the nodes can process information about only a subset of the existing nodes in the network. We find that the distribution of the number of incoming links to a node follows a universal scaling form, i.e., that it decays as a power law with an exponential truncation controlled not only by the system size but also by a feature not previously considered, the subset of the network “accessible” to the node. We test our model with empirical data for the World Wide Web and find agreement. DOI:10.1103/PhysRevLett.88.138701 PACS numbers:89.20.Hh, 84.35.+i, 89.75.Da, 89.75.Hc There is a great deal of current interest in understanding the structure and growth mechanisms of global networks [1–3], such as the World Wide Web (WWW) [4,5] and the Internet [6]. Networkstructure is critical in manycontexts such as Internet attacks [2], spread of an Email virus [7], or dynamics of human epidemics [8]. In all these problems, the nodes with the largest number of links play an important role on the dynamics of the system. It is therefore important to know the globalstructure of the networkas well as its precise distribution of the number of links. Recent empirical studies report that both the Internet and the WWW have scale-free properties; that is, the number of incoming links and the number of outgoing links at a given node have distributions that decay with power law tails [4–6]. It has been proposed [9] that the scale-free structure of the Internet and the WWW may be explained by a mechanism referred to as “preferential attachment” [10] in which new nodes link to existing nodes with a probability proportional to the number of existing links to these nodes. Here we focus on the stochastic character of the preferential attachment mechanism, which we understand in the following way: New nodes want to connect to the existing nodes with the largest number of links—i.e., with the largest degree—because of the advantages offered bybeing linked to a well-connected node. For a large network it is not plausible that a new node will know the degrees of all existing nodes, so a new node must make a decision on which node to connect with based on what information it has about the state of the network. The preferential attachment mechanism then comes into play as nodes with a larger degree are more likely tobecome known. VOLUME 88, Number 13 PHYSICAL REVIEW LETTERS 1 April 2002 Truncation of Power Law Behavior in “Scale-Free” NetworkModels due to Information Filtering StefanoMossa,1,2 Marc Barthélémy,3 H. Eugene Stanley,1 and Luís A. Nunes Amaral1 1 Center for Polymer Studies and Department of Physics, Boston University, Boston, Massachusetts 02215 2 Dipartimento di Fisica, INFM UdR, and INFM Center for Statistical Mechanics and Complexity, Università di Roma “La Sapienza,” Piazzale Aldo Moro 2, I-00185, Roma, Italy 3 CEA-Service de Physique de la Matière Condensée, BP 12, 91680 Bruyères-le-Châtel, France (Received 18 October 2001; published 14 March2002) We formulate a general model for the growth of scale-free networks under filtering information conditions—that is, when the nodes can process information about only a subset of the existing nodes in the network. We find that the distribution of the number of incoming links to a node follows a universal scaling form, i.e., that it decays as a power law with an exponential truncation controlled not only by the system size but also by a feature not previously considered, the subset of the network “accessible” to the node. We test our model with empirical data for the World Wide Web and find agreement. DOI:10.1103/PhysRevLett.88.138701 PACS numbers:89.20.Hh, 84.35.+i, 89.75.Da, 89.75.Hc There is a great deal of current interest in understanding the structure and growth mechanisms of global networks [1–3], such as the World Wide Web (WWW) [4,5] and the Internet [6]. Networkstructure is critical in manycontexts such as Internet attacks [2], spread of an Email virus [7], or dynamics of human epidemics [8]. In all these problems, the nodes with the largest number of links play an important role on the dynamics of the system. It is therefore important to know the globalstructure of the networkas well as its precise distribution of the number of links. Recent empirical studies report that both the Internet and the WWW have scale-free properties; that is, the number of incoming links and the number of outgoing links at a given node have distributions that decay with power law tails [4–6]. It has been proposed [9] that the scale-free structure of the Internet and the WWW may be explained by a mechanism referred to as “preferential attachment” [10] in which new nodes link to existing nodes with a probability proportional to the number of existing links to these nodes. Here we focus on the stochastic character of the preferential attachment mechanism, which we understand in the following way: New nodes want to connect to the existing nodes with the largest number of links—i.e., with the largest degree—because of the advantages offered bybeing linked to a well-connected node. For a large network it is not plausible that a new node will know the degrees of all existing nodes, so a new node must make a decision on which node to connect with based on what information it has about the state of the network. The preferential attachment mechanism then comes into play as nodes with a larger degree are more likely tobecome known. VOLUME 88, Number 13 PHYSICAL REVIEW LETTERS 1 April 2002 Truncation of Power Law Behavior in “Scale-Free” NetworkModels due to Information Filtering StefanoMossa,1,2 Marc Barthélémy,3 H. Eugene Stanley,1 and Luís A. Nunes Amaral1 1 Center for Polymer Studies and Department of Physics, Boston University, Boston, Massachusetts 02215 2 Dipartimento di Fisica, INFM UdR, and INFM Center for Statistical Mechanics and Complexity, Università di Roma “La Sapienza,” Piazzale Aldo Moro 2, I-00185, Roma, Italy 3 CEA-Service de Physique de la Matière Condensée, BP 12, 91680 Bruyères-le-Châtel, France (Received 18 October 2001; published 14 March2002) We formulate a general model for the growth of scale-free networks under filtering information conditions—that is, when the nodes can process information about only a subset of the existing nodes in the network. We find that the distribution of the number of incoming links to a node follows a universal scaling form, i.e., that it decays as a power law with an exponential truncation controlled not only by the system size but also by a feature not previously considered, the subset of the network “accessible” to the node. We test our model with empirical data for the World Wide Web and find agreement. DOI:10.1103/PhysRevLett.88.138701 PACS numbers:89.20.Hh, 84.35.+i, 89.75.Da, 89.75.Hc There is a great deal of current interest in understanding the structure and growth mechanisms of global networks [1–3], such as the World Wide Web (WWW) [4,5] and the Internet [6]. Networkstructure is critical in manycontexts such as Internet attacks [2], spread of an Email virus [7], or dynamics of human epidemics [8]. In all these problems, the nodes with the largest number of links play an important role on the dynamics of the system. It is therefore important to know the globalstructure of the networkas well as its precise distribution of the number of links. Recent empirical studies report that both the Internet and the WWW have scale-free properties; that is, the number of incoming links and the number of outgoing links at a given node have distributions that decay with power law tails [4–6]. It has been proposed [9] that the scale-free structure of the Internet and the WWW may be explained by a mechanism referred to as “preferential attachment” [10] in which new nodes link to existing nodes with a probability proportional to the number of existing links to these nodes. Here we focus on the stochastic character of the preferential attachment mechanism, which we understand in the following way: New nodes want to connect to the existing nodes with the largest number of links—i.e., with the largest degree—because of the advantages offered bybeing linked to a well-connected node. For a large network it is not plausible that a new node will know the degrees of all existing nodes, so a new node must make a decision on which node to connect with based on what information it has about the state of the network. The preferential attachment mechanism then comes into play as nodes with a larger degree are more likely tobecome known. VOLUME 88, Number 13 PHYSICAL REVIEW LETTERS 1 April 2002 Truncation of Power Law Behavior in “Scale-Free” NetworkModels due to Information Filtering StefanoMossa,1,2 Marc Barthélémy,3 H. Eugene Stanley,1 and Luís A. Nunes Amaral1 1 Center for Polymer Studies and Department of Physics, Boston University, Boston, Massachusetts 02215 2 Dipartimento di Fisica, INFM UdR, and INFM Center for Statistical Mechanics and Complexity, Università di Roma “La Sapienza,” Piazzale Aldo Moro 2, I-00185, Roma, Italy 3 CEA-Service de Physique de la Matière Condensée, BP 12, 91680 Bruyères-le-Châtel, France (Received 18 October 2001; published 14 March2002) We formulate a general model for the growth of scale-free networks under filtering information conditions—that is, when the nodes can process information about only a subset of the existing nodes in the network. We find that the distribution of the number of incoming links to a node follows a universal scaling form, i.e., that it decays as a power law with an exponential truncation controlled not only by the system size but also by a feature not previously considered, the subset of the network “accessible” to the node. We test our model with empirical data for the World Wide Web and find agreement. DOI:10.1103/PhysRevLett.88.138701 PACS numbers:89.20.Hh, 84.35.+i, 89.75.Da, 89.75.Hc There is a great deal of current interest in understanding the structure and growth mechanisms of global networks [1–3], such as the World Wide Web (WWW) [4,5] and the Internet [6]. Networkstructure is critical in manycontexts such as Internet attacks [2], spread of an Email virus [7], or dynamics of human epidemics [8]. In all these problems, the nodes with the largest number of links play an important role on the dynamics of the system. It is therefore important to know the globalstructure of the networkas well as its precise distribution of the number of links. Recent empirical studies report that both the Internet and the WWW have scale-free properties; that is, the number of incoming links and the number of outgoing links at a given node have distributions that decay with power law tails [4–6]. It has been proposed [9] that the scale-free structure of the Internet and the WWW may be explained by a mechanism referred to as “preferential attachment” [10] in which new nodes link to existing nodes with a probability proportional to the number of existing links to these nodes. Here we focus on the stochastic character of the preferential attachment mechanism, which we understand in the following way: New nodes want to connect to the existing nodes with the largest number of links—i.e., with the largest degree—because of the advantages offered bybeing linked to a well-connected node. For a large network it is not plausible that a new node will know the degrees of all existing nodes, so a new node must make a decision on which node to connect with based on what information it has about the state of the network. The preferential attachment mechanism then comes into play as nodes with a larger degree are more likely tobecome known. Refs > non-WoS Total publ universe non-WoS publ: Books Book chapters Conf. proc. Reports Non-WoS journals
  • 5. Aliens from Galaxy M35- 245 approach planet Earth. They have access to the Web of Science…..
  • 6. pa1 pa2 pa3 pa4 pb1 pb2 pb3 pb4 pb5 pb2 pb1 pb3 pb5 pb4 pa2 pa1 pa4 pa3 Co-citation-network Bibliogr.coupl.network Primary Citation Network cited p <> cited p citing p <> citing p citing p <> cited p
  • 7. Primary network is a structure of different items (e.g., citing publ <> cited publ; citing publ <> concepts) In-degree primary network => received citations > impact Secondary network is a structure of similar items (e.g., citing publ<> citing publ; concepts <> concepts) Similarity in the secondary networks => co- occurrence map
  • 8. 1. Bibliometric methodology: impact
  • 9. networks leading, possibly, to different dynamics, e.g., for the initiation and spread of epidemics. In the context of network growth, the impossibility of knowing the degrees of all the nodes comprising the network due to the filtering process— and, hence, the inability to make the optimal, rational, choice—is not altogether unlike the “bounded rationality” concept of Simon [17]. Remarkably, it appears that, for the description of WWW growth, the preferential attachment mechanism, originally proposed by Simon [10], must be modified along the lines of another concept also introduced by him—bounded rationality [17]. We thank R. Albert, P. Ball, A.-L. Barabási, M. Buchanan, J. Camacho, and R. Guimerà for stimulating discussions and helpful suggestions. We are especially grateful to R. Kumar for sharing his data. We thank NIH/NCRR (P41 RR13622) and NSF for support. [1] S. H. Strogatz, Nature (London) 410, 268 (2001). [2] R. Albert and A.-L. Barabási, Rev. Mod. Phys. 74, 47 (2002). [3] S. N. Dorogovtsev and J. F. F. Mendes, Adv. Phys. (to be published). [4] R. Albert, H. Jeong, and A.-L. Barabási, Nature (London) 401, 130 (1999). [5] B. A. Huberman and L. A. Adamic, Nature (London) 401, 131 (1999); R. Kumar et al., in Proceedings of the 25th International Conference on Very Large Databases (Morgan Kaufmann Publishers, San Francisco, 1999), p. 639; A. Broder et al., Comput. Netw. 33, 309 (2000); P. L. Krapivsky, S. Redner, and F. Leyvraz, Phys. Rev. Lett. 85, 4629 (2000); S. N. Dorogovtsev, J. F. F. Mendes, and A. N. Samukhin, ibid. 85, 4633 (2000); A. Vazquez, Europhys. Lett. 54, 430 (2001). [6] M. Faloutsos, P. Faloutsos, and C. Faloutsos, Comput. Commun. Rev. 29, 251 (1999); G. Caldarelli, R. Marchetti, and L. Pietronero, Europhys. Lett. 52, 386 (2000); A. Medina, I. Matta, and J. Byers, Comput. Commun. Rev. 30, 18 (2000); R. Pastor-Satorras, A. Vazquez, and A. Vespignani, arXiv:cond-mat/0105161; L. A. Adamic et al., Phys. Rev. E 64, 046135 (2001). [7] F. B. Cohen, A Short Course on Computer Viruses (Wiley, New York, 1994); R. Pastor-Satorras and A. Vespignagni, Phys. Rev. Lett. 86, 3200 (2001); Phys. Rev. E 63, 066117 (2001). [8] F. Liljeros, C. R. Edling, L. A. Nunes Amaral, H. E. Stanley, and Y. Åberg, Nature (London) 411, 907 (2001). [9] A.-L. Barabási and R. Albert, Science 286, 509 (1999). [10] Y. Ijiri and H. A. Simon, Skew Distributions and the Sizes of Business Firms (North-Holland, Amsterdam, 1977). [11] G. Bianconi and A.-L. Barabasi, Europhys. Lett. 54, 436 (2001). [12] A. F. J. Van Raan, Scientometrics 47, 347 (2000). [13] We consider a modification to the network growth rule described earlier in the paper: at each time step t, the new node establishes m new links, where m is drawn from a power law distribution with exponent gout. [14] For n(I) = const, one recovers the scale-free model of Ref. [9]. [15] It is known [11] that, for an exponential or fat-tailed distribution of fitness, the structure of the network becomes much more complex; in particular, the in-degree distribution is no longer a power law. Hence, we do not consider in this manuscript other shapes of the fitness distribution. [16] L. A. N. Amaral, A. Scala, M. Barthélémy, and H. E. Stanley, Proc. Natl. Acad. Sci. U.S.A. 97, 11 149 (2000). [17] H. A. Simon, Models of Bounded Rationality: Empirically Grounded Economic Reason (MIT Press, Cambridge, 1997). Citing Publications Cited Publications From other disciplines From emerging fields From research devoted to societal, economical and technological problems From industry From international top-groups These all f(t)!! > Sleeping Beauties
  • 10. 1 2 3 4 5 6 7 a b c d e f g Fig. 1: Citing and cited publications Intermezzo: statistical properties of the primary network Out-degree (all)= 1 In-degree (b) = 5
  • 11. Large differences among research fields in citation density Citation counts must always be normalized to field- specific values and therefore: H-index is not normalized and inconsistent
  • 12. 500 Dept. Radiology CPP = 6 Radiology journals JCSm = 4 CPP/JCSm=1.5 P 2007- 2010 150,000 Radiology field FCSm = 3 CPP/FCSm = 2 C 2007- 2010 3000 450,000 25,000 100,000
  • 13. 13 • Two normalization mechanisms: ci: actual number of citations of publication i ei: expected number of citations of publication i • Value of 1 indicates performance at world average          n i i n i i n i i n i i e c ne nc 1 1 1 1 CPP/FCSm   n i i i e c n 1 1 MNCS
  • 14. Applied research, engineering, social sciences Basic research in natural and medical science high FCSm Δ Citation density ~20 ! high FCSm, but low JCSm low FCSm low FCSm, but high JCSm High CPP low CPP
  • 15. PHYSICS 0 5000 10000 15000 20000 25000 30000 35000 Cits91 Cits92 Cits93 Cits94 Cits95 Cits96 Cits97 Cits98 Cits99 Cits00 Cits01 Cits02 Publications from 1991,….1995 time lag & citation window
  • 16. * SOCIOLOGY 0 500 1000 1500 2000 2500 Cits91 Cits92 Cits93 Cits94 Cits95 Cits96 Cits97 Cits98 Cits99 Cits00 Cits01 Cits02 Again: never use journal impact factors!!
  • 17. University Departments Fields ‘bottom-up’ analysis: input data (assignment of researchers to departments) necessary; > Detailed research performance analysis of a university by department ‘top-down’ analysis: field-structure is imposed to university; > Broad overview analysis of a university by field
  • 18. H = 49 H = 27 P C+sc Dept of Radiology, UMC 1997 - 2009 1,045 20,552 1997 - 2000 196 616 1998 - 2001 213 938 1999 - 2002 243 1,151 2000 - 2003 231 1,590 2001 - 2004 221 1,576 2002 - 2005 306 2,097 2003 - 2006 358 2,516 2004 - 2007 417 3,368 2005 - 2008 492 4,254 2006 - 2009 497 3,986 CPP Pnc 16.73 17% 2.32 51% 3.43 38% 3.63 39% 5.61 36% 5.94 29% 5.72 40% 5.88 35% 6.82 35% 7.10 30% 6.42 30% JCSm FCSm 13.63 9.84 2.16 1.92 3.30 2.60 4.00 3.00 4.78 3.14 4.34 3.03 3.94 2.79 3.71 2.65 4.37 3.03 5.08 3.73 5.08 3.52 CPP/ CPP/ JCSm/ JCSm FCSm FCSm 1.23 1.70 1.36 1.07 1.21 1.09 1.04 1.32 1.23 0.91 1.21 1.29 1.17 1.79 1.47 1.37 1.96 1.39 1.45 2.05 1.37 1.58 2.21 1.36 1.56 2.25 1.40 1.40 1.90 1.33 1.26 1.82 1.41 SelfCit 15% 26% 22% 23% 18% 17% 17% 16% 16% 18% 20% Crown indicator
  • 19. Impact trend of two depts. Radiology (UMC A and B) 0.00 0.50 1.00 1.50 2.00 2.50 3.00 1997 - 2000 1998 - 2001 1999 - 2002 2000 - 2003 2001 - 2004 2002 - 2005 2003 - 2006 2004 - 2007 2005 - 2008 2006 - 2009 years CPP/FCSm above below international level P(2006-9)=561: in which fields?
  • 20. 2006 - 2009 Radiology UMC RAD,NUCL MED IM (1.63) CARD&CARDIOV SYS (2.24) SURGERY (1.71) CLIN NEUROLOGY (2.15) ONCOLOGY (1.55) PERIPHL VASC DIS (1.20) PEDIATRICS (1.09) MEDICINE,GEN&INT (4.02) NEUROSCIENCES (2.71) GASTROENTEROLOGY (0.59) FIELD (CPP/FCSm) Research Profile Output (P) and Impact (CPP/FCSm) 2006 - 2009 Dept Radiology UMC
  • 21. PUBLICATIONS AND IMPACT PER SUBFIELD 1992 - 2000 Erkrankungen des Nervensystems 0% 5% 10% 15% 20% 25% 30% 35% 40% NEUROSCIENCES (2.28) BIOCH & MOL BIOL (2.00) CELL BIOLOGY (1.96) DEVELOPMENT BIOL (1.33) MULTIDISCIPL SC (2.84) GENETICS & HERED (2.70) PHYSIOLOGY (1.74) CLIN NEUROLOGY (1.58) PHARMACOL & PHAR (3.94) ONCOLOGY (1.60) BIOLOGY (2.06) ANAT & MORPHOL (2.15) ENDOCRIN & METAB (1.81) PATHOLOGY (2.11) UROLOGY & NEPHRO (1.30) BIOPHYSICS (1.97) IMMUNOLOGY (1.84) MEDICINE, RES (1.88) ZOOLOGY (4.51) OPHTHALMOLOGY (2.21) SUBFIELD (CPP/FCSm) Relative share of impact 1992 - 2000 Erkrankungen des Nervensystems 0% 10% 20% 30% 40% 50% 60% NEUROSCIENCES (2.70) BIOCH & MOL BIOL (1.92) CELL BIOLOGY (2.74) MULTIDISCIPL SC (4.33) EVELOPMENT BIOL (1.32) SUBFIELD (CPP/FCSm) Percentage of Publications IMPACT: LOW AVERAGE HIGH CPP/ FCSm CPP/ FCSm Diseases of the Neurosystem Dept Output in Fields Dept Impact from Fields Knowledge use by very high impact groups Neurosc 2.70 Biochem 1.92 Cell Biol 2.74 multidisc 4.33 Devel Biol 1.32 Neurosc 2.28 Biochem 2.00 Cell Biol 1.96 Devel Biol 1.33 multidisc 2.84 Genetics 2.70 Physiol 1.74 Clin Neurol 1.58 Pharmacol 3.94 Oncology 1.60 Biology 2.06 Anat Morph 2.15 Endocrinol 1.81 Pathology 2.11 Urology 1.30 Biophysics 1.97 Immunol 1.84 Res Medic 1.88 Zoology 4.51 Opthalmol 2.21
  • 22. 2. Bibliometric methodology: maps
  • 23. Major field fields Main field: Medical & Life Sciences BIOMEDICAL SCIENCES ANATOMY & MORPHOLOGY IMMUNOLOGY INTEGRATIVE & COMPLEMENTARY MEDICINE MEDICAL LABORATORY TECHNOLOGY MEDICINE, RESEARCH & EXPERIMENTAL NEUROIMAGING NEUROSCIENCES PHARMACOLOGY & PHARMACY PHYSIOLOGY RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING TOXICOLOGY VIROLOGY ACTA GENETICAE MEDICAE ET GEMELLOLOGIAE AMERICAN JOURNAL OF HUMAN GENETICS AMERICAN JOURNAL OF MEDICAL GENETICS ANIMAL BLOOD GROUPS AND BIOCHEMICAL GENETICS ANNALES DE GENETIQUE ANNALES DE GENETIQUE ET DE SELECTION ANIMALE ANNALS OF HUMAN GENETICS ATTI ASSOCIAZIONE GENETICA ITALIANA BEHAVIOR GENETICS BIOCHEMICAL GENETICS CANADIAN JOURNAL OF GENETICS AND CYTOLOGY CANCER GENETICS AND CYTOGENETICS CARYOLOGIA CHROMOSOMA CLINICAL GENETICS CURRENT GENETICS CYTOGENETICS AND CELL GENETICS CYTOLOGIA DEVELOPMENTAL GENETICS ENVIRONMENTAL MUTAGENESIS EVOLUTION GENE GENETICA GENETICA POLONICA GENETICAL RESEARCH GENETICS GENETIKA HEREDITAS HEREDITY HUMAN BIOLOGY HUMAN GENETICS HUMAN HEREDITY IMMUNOGENETICS INDIAN JOURNAL OF GENETICS AND PLANT BREEDING JAPANESE JOURNAL OF GENETICS JAPANESE JOURNAL OF HUMAN GENETICS JOURNAL DE GENETIQUE HUMAINE JOURNAL OF HEREDITY JOURNAL OF IMMUNOGENETICS JOURNAL OF MEDICAL GENETICS JOURNAL OF MENTAL DEFICIENCY RESEARCH JOURNAL OF MOLECULAR EVOLUTION MOLECULAR & GENERAL GENETICS MUTATION RESEARCH PLASMID SILVAE GENETICA THEORETICAL AND APPLIED GENETICS THEORETICAL POPULATION BIOLOGY EGYPTIAN JOURNAL OF GENETICS AND CYTOLOGY REVISTA BRASILEIRA DE GENETICA ANNUAL REVIEW OF GENETICS JOURNAL OF CRANIOFACIAL GENETICS AND DEVELOPMENTAL BIOLOGY JOURNAL OF INHERITED METABOLIC DISEASE PRENATAL DIAGNOSIS ADVANCES IN GENETICS INCORPORATING MOLECULAR GENETIC MEDICINE CHEMICAL MUTAGENS-PRINCIPLES AND METHODS FOR THEIR DETECTION DNA-A JOURNAL OF MOLECULAR & CELLULAR BIOLOGY EVOLUTIONARY BIOLOGY TERATOGENESIS CARCINOGENESIS AND MUTAGENESIS TSITOLOGIYA I GENETIKA ADVANCES IN HUMAN GENETICS PROGRESS IN MEDICAL GENETICS GENETICS SELECTION EVOLUTION MOLECULAR BIOLOGY AND EVOLUTION SOMATIC CELL AND MOLECULAR GENETICS BIOTECHNOLOGY & GENETIC ENGINEERING REVIEWS EXPERIMENTAL AND CLINICAL IMMUNOGENETICS GENE ANALYSIS TECHNIQUES JOURNAL OF MOLECULAR AND APPLIED GENETICS JOURNAL OF NEUROGENETICS TRENDS IN GENETICS DISEASE MARKERS ANIMAL GENETICS GENETIC EPIDEMIOLOGY JOURNAL OF GENETICS journals Citation relations between journals are used to define fields of science
  • 24. Ja1 Ja2 Ja3 Ja4 Jb1 Jb2 Jb3 Jb4 Jb5 Ja2 Ja1 Ja4 Ja3 Bibliogr.coupl. Journal map based on citation similarity
  • 25. Colors indicate the different subfields Radiology, Nuclear Medicine, & Medical Imaging 111 journals, 2007-2010 Citation-based network map
  • 26. Journal 1. Field = set of journals Defined by WoS or Scopus + reference-based re-definition (expansion) of fields (e.g. Nature, Science)
  • 27. Fa1 Fa2 Fa3 Fa4 Fb1 Fb2 Fb3 Fb4 Fb5 Fa2 Fa1 Fa4 Fa3 Bibliogr.coupl. Field map based on citation similarity
  • 28. Size: World average Science as a structure of 200 related fields World map Colors indicate the different fields
  • 29. 29 Leiden University
  • 30. 30 Delft University of Technology
  • 31. Instead of using citations (references) to calculate publication-similarity and to create citation-based maps one can in exactly the way mathematical way use concepts (keywords extracted from titles and abstracts) to calculate publication- similarity and to create concept-based maps
  • 32. 2.Field = clusters of concept-related publications own definition: new, emerging, interdisciplinary fields, socio-economic themes Evaluation of universities’ ‘third mission’: Role of science for societal demands
  • 33. Urban Studies 2006-2008 Concept map, n=700 Colors indicate the different subfields
  • 34. Urban Studies 2006-2008 Concept density map n=700 Colors indicate the different publication densities
  • 35. Map 2005 Map 2010 Knowledge dynamics described by density and flow ? Map 2015
  • 36. Serious problem in impact assessment by citations mentioned by more and more researchers: Normalization by an average citation density (FCSm) of the whole field might seriously prejudice groups working in a low density subfield How can we measure these within-field citation densities and determine how serious the problem is?
  • 37. Combining concept-based maps with local citation-density measurement
  • 38. Clinical Neurology 2005-2009 Colors now indicate local citation density --> Neuro-surgery has a relatively low citation density within the field of clinical neurology
  • 39. Cardiovasc system 2005-2009 Concept map, n=1,500 Colors now indicate local citation density
  • 40. 3. Leiden Ranking 2011-2012: new indicators
  • 41. New and innovative features! No static tables but a user-friendly menu with (1) selection of universities (2) selection of performance dimension with specific indicators (3) selection of calculation method for the 500 largest universities worldwide
  • 42. Proper normalization is absolutely crucial! Rank of all 250 universities by FCSm 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 0 50 100 150 200 250 300 r FCSm Ranking of the 250 largest European universities by FCSm
  • 43. www.leidenranking.com
  • 44. (1) selection of universities Region: * World * Regions (Europe, Asia, North America, South America, Africa, Oceania) * Countries By number of universities covered in the ranking: *largest 100, 200, 300, 400, 500
  • 45. (2) selection of performance dimension with specific indicators * impact: P, MCS, MNCS, PP(top10%) * collaboration: P, PP(collab), PP(int collab), MGCD, PP(long dist collab) indicators with stability intervals!
  • 46. MNCS (earlier CPP/FCSm) is an average and this is statistically not the best measure for a skew distribution… So we move on to p-top10% this measures takes the whole distribution into account Notice: dimension of MNCS is C/P dimension of p-top10% is n
  • 47. (3) selection of calculation method * full or fractional assignment collaborative publications * all WoS publications or only English language
  • 48. Observations
  • 49. General: * Large effect of all vs only-English publications for D and F * Large effect full vs fractional counting * One-paper-explosion effect * International collaboration
  • 50. Stability intervals clearly show the dramatic influence of just one publication! Examples: Göttingen, Utrecht
  • 51. P[2008]; C[2008-2011*] = 20,485
  • 52. P[2009]; C[2009-2011*] = 2,856
  • 53. The influence of Arts & Humanities on bibliometric university rankings is practically non-existent….
  • 54. Leave out non-English publicationsx Rank based on ALL WoS publications
  • 55. Very influential for D and F univ
  • 56. EUR-50 / language: all / counting: full P MNCS 1 Univ Cambridge 25073 1.76 2 Univ Oxford 26161 1.71 3 ETH Zurich 15890 1.63 4 Imperial Coll London 22280 1.58 5 Univ Edinburgh 13318 1.57 6 Utrecht Univ 18221 1.56 7 Erasmus Univ Rotterdam 12788 1.54 8 Univ Bristol 12199 1.53 9 Univ Coll London 26210 1.51 10 Univ Zurich 14492 1.50 11 King's Coll London 13502 1.46 12 VU Univ Amsterdam 12732 1.46 13 Leiden Univ 12653 1.45 14 Univ Copenhagen 18654 1.43 15 Radboud Univ Nijmegen 12061 1.43 16 Univ Amsterdam 15760 1.43 17 Karolinska Inst 17054 1.41 18 Aarhus Univ 13039 1.41 19 LMUniv München 17672 1.40 20 Katholieke Univ Leuven 19673 1.38 21 Lund Univ 15558 1.37 22 Univ Manchester 18217 1.37 23 Tech Univ München 12452 1.36 24 Univ Groningen 12303 1.36 25 Univ Southampton 10861 1.36 EUR-50 / language: Engl / counting: full P MNCS 1 Univ Cambridge 25037 1.71 2 Univ Oxford 26105 1.65 3 ETH Zurich 15708 1.60 4 Imperial Coll London 22234 1.53 5 Univ Edinburgh 13291 1.51 6 Utrecht Univ 18066 1.51 7 Univ Zurich 13840 1.50 8 LMUniv München 15827 1.48 9 Univ Bristol 12170 1.47 10 Erasmus Univ Rotterdam 12738 1.47 11 Univ Coll London 26166 1.45 12 Univ Glasgow 10144 1.41 13 VU Univ Amsterdam 12697 1.40 14 Leiden Univ 12597 1.40 15 Heidelberg Univ 14552 1.40 16 King's Coll London 13473 1.40 17 Tech Univ München 11660 1.39 18 Radboud Univ Nijmegen 12000 1.38 19 Univ Copenhagen 18633 1.37 20 Univ Amsterdam 15689 1.37 21 Univ Pierre & Marie Curie 19206 1.37 22 Karolinska Inst 17025 1.35 23 Univ Paris-Sud 11 13674 1.35 24 Aarhus Univ 13020 1.35 25 Univ Bonn 10035 1.34 26 Katholieke Univ Leuven 19540 1.34 27 Lund Univ 15528 1.33 28 Univ Manchester 18194 1.32 29 Univ Groningen 12256 1.32 30 Univ Southampton 10843 1.32 30 All <> Eng MNCS: D, F UK
  • 57. Citation Impact German and French universities All versus English-only Publications y = 1,08x R2 = 0,81 0,00 0,20 0,40 0,60 0,80 1,00 1,20 1,40 1,60 1,80 2,00 0,00 0,20 0,40 0,60 0,80 1,00 1,20 1,40 1,60 1,80 2,00 imp-all imp-eng Citation Impact Ranking German and French universities All versus English-only Publications y = 1,02x - 38,31 R2 = 0,83 0 50 100 150 200 250 300 350 400 450 500 0 50 100 150 200 250 300 350 400 450 500 r-all r-eng
  • 58. EUR-50 / language: Engl / counting: full P MNCS 1 Univ Cambridge 25037 1.71 2 Univ Oxford 26105 1.65 3 ETH Zurich 15708 1.60 4 Imperial Coll London 22234 1.53 5 Univ Edinburgh 13291 1.51 6 Utrecht Univ 18066 1.51 7 Univ Zurich 13840 1.50 8 LMUniv München 15827 1.48 9 Univ Bristol 12170 1.47 10 Erasmus Univ Rotterdam 12738 1.47 11 Univ Coll London 26166 1.45 12 Univ Glasgow 10144 1.41 13 VU Univ Amsterdam 12697 1.40 14 Leiden Univ 12597 1.40 15 Heidelberg Univ 14552 1.40 16 King's Coll London 13473 1.40 17 Tech Univ München 11660 1.39 18 Radboud Univ Nijmegen 12000 1.38 19 Univ Copenhagen 18633 1.37 20 Univ Amsterdam 15689 1.37 21 Univ Pierre & Marie Curie 19206 1.37 22 Karolinska Inst 17025 1.35 23 Univ Paris-Sud 11 13674 1.35 24 Aarhus Univ 13020 1.35 25 Univ Bonn 10035 1.34 26 Katholieke Univ Leuven 19540 1.34 27 Lund Univ 15528 1.33 28 Univ Manchester 18194 1.32 29 Univ Groningen 12256 1.32 30 Univ Southampton 10843 1.32 EUR-50 / language: Engl / counting: frac P MNCS 1 Univ Cambridge 14065 1.57 2 ETH Zurich 8632 1.55 3 Univ Oxford 13949 1.49 4 Utrecht Univ 9710 1.49 5 Imperial Coll London 11567 1.41 6 Univ Bristol 6681 1.36 7 Univ Coll London 13745 1.36 8 Univ Edinburgh 7142 1.32 9 VU Univ Amsterdam 6494 1.31 10 Erasmus Univ Rotterdam 6765 1.31 11 Univ Zurich 7754 1.29 12 Univ Amsterdam 8104 1.29 13 King's Coll London 7249 1.28 14 Leiden Univ 6355 1.27 15 Univ Copenhagen 10157 1.25 16 Aarhus Univ 7069 1.24 17 Univ Manchester 10427 1.23 18 Radboud Univ Nijmegen 6478 1.23 19 Tech Univ München 6672 1.22 20 Univ Groningen 6967 1.22 21 Univ Nottingham 6839 1.20 22 Karolinska Inst 8573 1.20 23 Univ Southampton 6017 1.19 24 Katholieke Univ Leuven 10956 1.19 25 Univ Sheffield 6440 1.19 26 LMUniv München 9585 1.16 27 FAUniv Erlangen 5756 1.16 28 Univ Paris-Sud 11 6323 1.15 29 Univ Leeds 6627 1.15 30 Univ Pierre & Marie Curie 9309 1.14 full <>frac MNCS: All: Major changes for the smaller universities
  • 59. 15 29 60! For the smaller universities fractional counting is quite influential, particularly for Delft! EUR-100 / language: Engl / counting: frac P MNCS 1 Univ Göttingen 4776 2.04 2 EPF Lausanne 4790 1.69 3 Univ Cambridge 14046 1.53 4 ETH Zurich 8507 1.53 5 Utrecht Univ 9601 1.45 6 Univ Oxford 13919 1.44 7 Tech Univ Denmark 4287 1.43 8 Imperial Coll London 11547 1.37 9 Univ Bristol 6671 1.31 10 Univ Basel 4263 1.31 11 Univ Coll London 13723 1.31 12 Univ Zurich 7293 1.30 13 Wageningen Univ 4590 1.30 14 Univ Edinburgh 7128 1.28 15 VU Univ Amsterdam 6471 1.27 16 LMUniv München 8261 1.27 17 Tech Univ München 6145 1.26 18 Erasmus Univ Rotterdam 6735 1.26 19 Univ Geneva 4971 1.25 20 Univ Amsterdam 8059 1.24 21 Delft Univ Technol 4531 1.24 22 King's Coll London 7236 1.23 23 Leiden Univ 6321 1.23 24 FAUniv Erlangen 5198 1.22 25 Univ Freiburg 4768 1.22 26 Univ Bern 4717 1.21 27 Univ Pierre & Marie Curie 8416 1.21 28 Karlsruhe Inst Technol 4362 1.21 29 Univ Copenhagen 10145 1.20 30 Aarhus Univ 7060 1.19
  • 60. EUR-100 / language: Engl / counting: frac P MNCS 1 Univ Göttingen 4776 2.04 2 EPF Lausanne 4790 1.69 3 Univ Cambridge 14046 1.53 4 ETH Zurich 8507 1.53 5 Utrecht Univ 9601 1.45 6 Univ Oxford 13919 1.44 7 Tech Univ Denmark 4287 1.43 8 Imperial Coll London 11547 1.37 9 Univ Bristol 6671 1.31 10 Univ Basel 4263 1.31 11 Univ Coll London 13723 1.31 12 Univ Zurich 7293 1.30 13 Wageningen Univ 4590 1.30 14 Univ Edinburgh 7128 1.28 15 VU Univ Amsterdam 6471 1.27 16 Ludwig-Maximilians-Univ Münche 8261 1.27 17 Tech Univ München 6145 1.26 18 Erasmus Univ Rotterdam 6735 1.26 19 Univ Geneva 4971 1.25 20 Univ Amsterdam 8059 1.24 21 Delft Univ Technol 4531 1.24 22 King's Coll London 7236 1.23 23 Leiden Univ 6321 1.23 24 FAUniv Erlangen 5198 1.22 25 Univ Freiburg 4768 1.22 26 Univ Bern 4717 1.21 27 Univ Pierre & Marie Curie 8416 1.21 28 Karlsruhe Inst Technol 4362 1.21 29 Univ Copenhagen 10145 1.20 30 Aarhus Univ 7060 1.19 EUR-100 / language: Engl / counting: frac P PPtop 10% 1 EPF Lausanne 4790 18.8% 2 ETH Zurich 8507 17.6% 3 Univ Cambridge 14046 16.7% 4 Univ Oxford 13919 16.5% 5 Tech Univ Denmark 4287 15.9% 6 Imperial Coll London 11547 15.2% 7 Univ Coll London 13723 14.8% 8 Univ Edinburgh 7128 14.4% 9 Wageningen Univ 4590 14.3% 10 Univ Zurich 7293 14.3% 11 Univ Bristol 6671 14.2% 12 Erasmus Univ Rotterdam 6735 14.2% 13 VU Univ Amsterdam 6471 14.1% 14 Univ Basel 4263 14.1% 15 Univ Geneva 4971 13.9% 16 LMUniv München 8261 13.8% 17 Utrecht Univ 9601 13.6% 18 Tech Univ München 6145 13.5% 19 Leiden Univ 6321 13.3% 20 Univ Amsterdam 8059 13.2% 21 Univ Freiburg 4768 13.1% 22 King's Coll London 7236 13.1% 23 Delft Univ Technol 4531 12.9% 24 Aarhus Univ 7060 12.9% 25 Univ Paris-Sud 11 5866 12.9% 26 FAUniv Erlangen 5198 12.8% 27 Univ Copenhagen 10145 12.8% 28 Karlsruhe Inst Technol 4362 12.7% 29 Univ Pierre & Marie Curie 8416 12.7% 30 Univ Glasgow 5262 12.6%62 !! MNCS > p-top10%: Universities with rank position dominated by just 1 publication change considerably in position
  • 61. 500 largest universities worldwide P[2005-2009], C[2005-2010] y = 0.14x - 0.04 R 2 = 0.97 0 0.05 0.1 0.15 0.2 0.25 0.3 0 0.5 1 1.5 2 2.5 MNCS-engl pp(top10%)
  • 62. EUR-50 / language: Engl / counting: frac P PPint collab 1 ETH Zurich 8507 36.5% 2 Univ Zurich 7293 36.0% 3 Karolinska Inst 8561 35.4% 4 Univ Paris-Sud 11 5866 34.3% 5 Katholieke Univ Leuven 10872 33.9% 6 Aarhus Univ 7060 33.1% 7 Univ Pierre & Marie Curie 8416 32.6% 8 Univ Copenhagen 10145 32.2% 9 Lund Univ 8473 32.0% 10 Univ Oslo 6864 31.4% 11 Univ Helsinki 7988 31.3% 12 Univ Oxford 13919 31.1% 13 Uppsala Univ 6928 30.9% 14 Imperial Coll London 11547 30.5% 15 Leiden Univ 6321 30.1% 16 Univ Cambridge 14046 29.2% 17 VU Univ Amsterdam 6471 28.8% 18 Univ Coll London 13723 28.6% 19 Humboldt-Univ Berlin 6441 28.5% 20 Univ Amsterdam 8059 28.4% 21 Charles Univ Prague 5396 28.2% 22 Univ Edinburgh 7128 27.9% 23 Ghent Univ 8935 27.8% 24 Tech Univ München 6145 27.7% 25 LMUniv München 8261 27.6% Collaboration and, more specifically, international collaboration: * no correlation with impact * weak correlation with MGCD (see Leiden Ranking website)
  • 63. Correlation of p-int-collab with MGCD for the 100 largest European universities y = 0.01x 0.56 R 2 = 0.47 0.100 1.000 1000 10000 MGCD p-int-collab
  • 64. Correlation of p-int-coll with MGCD for Australia y = 0,00x0,92 R2 = 0,94 0,10 1,00 1000,00 10000,00 MGCD p-int-coll
  • 65. 4. Evaluation tools related to the Leiden Ranking
  • 66. Built on the basic data of the Leiden Ranking 2011: Leiden Benchmark Analysis: * all indicators of the ‘total’ university ranking are available for 30 main (NOWT) fields of science * their trends in the past 10 years * world rank by indicator and by main field All this in comparison with 20 other universities by choice * all indicators for each field within each main field (total 200 fields) with distinction between fields above and below university average
  • 67. Current and recent benchmark projects Manchester, Leiden, Heidelberg, Rotterdam, Copenhagen, Zürich, Lisbon UNL, Amsterdam UvA, Amsterdam VU, Southampton Gent, Antwerp, Brussels VUB, UC London, Aarhus, Oslo
  • 68. Output & impact compared to benchmark universities 2002 - 2007 CHEMISTRY Zurich UC London Strasbourg I Paris XI Paris VI Oxford Milano Lunds LMU Munchen Karolinska KU Leuven Helsinki Heidelberg Geneve Freiburg Edinburgh Cambridge Wageningen Utrecht Twente TU Delft Nijmegen Leiden Groningen Eindhoven VU Adam UvA 0 0.5 1 1.5 2 0 500 1000 1500 2000 2500 TOTAL PUBLICATIONS CPP/FCSm
  • 69. Output and impact 2002 - 2007 SOCIAL SCIENCES Zurich UC London Paris XI Paris VI Oxford Milano Lund LMU Munchen Karolinska KU Leuven Helsinki Heidelberg Geneve Freiburg Edinburgh Cambridge Wageningen Utrecht Twente Tilburg TU Delft Nijmegen Maastricht Leiden Groningen Erasmus Eindhoven VU Adam UvA 0 0,2 0,4 0,6 0,8 1 1,2 1,4 0 200 400 600 800 1000 1200 1400 TOTAL PUBLICATIONS CPP/FCSm
  • 70. FIGURE VII.12: RESEARCH AND IMPACT PROFILE COMPARISON CHART 2001 - 2008 0.91 1.15 0.83 1.52 1.45 1.14 1.08 1.41 1.20 1.44 1.42 1.15 1.27 1.25 1.55 0.92 1.37 1.19 1.17 4.70 1.11 1.17 1.10 1.20 1.25 1.20 1.62 0.84 1.06 1.01 1.95 1.10 0.88 1.34 2.00 1.01 1.20 1.13 1.25 2.12 1.17 0.98 1.31 1.98 1.33 1.25 1.00 1.46 2.04 2.44 0.98 1.04 0.70 1.25 1.12 1.10 0.97 0.87 0.94 1.17 7% 5% 3% 1% 1% 3% 5% 7% ONCOLOGY BIOCHEM&MOL BIOL IMMUNOLOGY CARD&CARDIOV SYS PUBL ENV OCC HLT ASTRON&ASTROPH NEUROSCIENCES PSYCHIATRY HEMATOLOGY CELL BIOLOGY MEDICINE,GEN&INT CLIN NEUROLOGY ENDOCRIN&METABOL PERIPHL VASC DIS CHEM,PHYSICAL SURGERY GEOSC,MULTIDISC ECOLOGY DENT,ORAL SURG&M PEDIATRICS GENETICS&HEREDIT OBSTETRICS&GYNEC GASTROENTEROLOGY PHYSICS,COND MAT PHARMACOL&PHARMA PHYSICS,MULTIDIS NUTRITION&DIET RAD,NUCL MED IM PATHOLOGY MEDICINE,RES&EXP Percentage of Total Publication Output IMPACT: LOW AVERAGE HIGH UNIV OSLO LUNDS UNIV
  • 71. Leiden University Major field field P(2005-9) MNCS P>100 CLINICAL MEDICINE MEDICINE, GENERAL & INTERNAL 291.2 3.03 rank by PHYSICS AND ASTRONOMY PHYSICS, MULTIDISCIPLINARY 192.5 2.65 MNCS PHYSICS AND ASTRONOMY PHYSICS, CONDENSED MATTER 204.0 2.10 CLINICAL MEDICINE RESPIRATORY SYSTEM 122.7 1.91 CLINICAL MEDICINE RHEUMATOLOGY 349.0 1.90 CLINICAL MEDICINE CARDIAC & CARDIOVASC SYSTEMS 488.5 1.84 BIOL SCI: HUMANS MEDICINE, RESEARCH & EXPER 100.0 1.81 CLINICAL MEDICINE SURGERY 207.9 1.73 CLINICAL MEDICINE UROLOGY & NEPHROLOGY 158.0 1.67 MOLECULAR BIOL & BIOCHEM BIOTECH & APPLIED MICROBIOL 106.8 1.66 SOCIAL SC RELATED TO MED PUBLIC, ENVIRONM & OCC HEALTH 111.6 1.65 PHYSICS AND ASTRONOMY ASTRONOMY & ASTROPHYSICS 814.5 1.65 BIOL SCI: HUMANS MICROBIOLOGY 179.3 1.58 CLINICAL MEDICINE GENETICS & HEREDITY 394.9 1.57 CLINICAL MEDICINE PERIPHERAL VASCULAR DISEASE 268.0 1.51 Leiden average MNCS = 1.45
  • 72. It appears that the medical fields benefit more and the engineering fields less from the full counting This will be very influential in the relative position of the engineering fields within a university, particularly in performance analyses
  • 73. Shanghai and THES?? Use the Leiden Ranking from now on! What !!??
  • 74. Thank you for your attention more information: www.cwts.leidenuniv.nl