1.
Web of Science, Scopus, Dimensions, and beyond:
The evolving landscape of bibliometric data sources
Ludo Waltman, Martijn Visser, Nees Jan van Eck
Centre for Science and Technology Studies (CWTS), Leiden University
ROI-AV Conference: Visuals and Analytics that Matter
Copenhagen, Denmark
October 3, 2018

2.
Outline
• Bibliometric data sources
• New opportunities for bibliometric visualization
1

3.
Bibliometric data
sources
2

4.
Introduction
• Increasing number of alternatives (Microsoft Academic, Dimensions,
Crossref, OpenCitations Corpus) to traditional bibliographic data sources
(Web of Science, Scopus, Google Scholar)
• Alternative data sources are more open than traditional ones
• How do the various data sources compare in terms of the completeness and
quality of their citation data?
3

5.
Data sources
• Scopus
– May 2018
– Requires subscription
• Web of Science
– SCIE, SSCI, AHCI, CPCI
– June 2018
– Requires subscription
• Dimensions
– June 2018
– Openly available through web interface
• Crossref
– August 2018
– Openly available through API
4

6.
Coverage of publications
5
All publications Publications with DOI
Publications with
unique DOI
Web of Science 40.1 100.0% 18.8 46.9% 18.8 46.9%
Scopus 44.9 100.0% 31.1 69.2% 30.6 68.3%
Dimensions 57.5 100.0% 55.1 95.9% 55.0 95.6%
Crossref 57.3 100.0% 57.3 100.0% 57.3 100.0%
• Publication counts in millions
• Time period 1996-2017

7.
Coverage of publications: Dimensions vs. Scopus
6

8.
Comparison of citation data
7
Scopus-WoS overlap: 460.0M
Only in Scopus: 24.9M
Only in WoS: 15.5M
Scopus-Dimensions overlap: 414.3M
Only in Scopus: 43.5M
Only in Dimensions: 17.9M
Scopus-Crossref overlap: 171.3M
Only in Scopus: 292.1M
Only in Crossref: 6.4M
In these pairwise comparisons of data sources, only
citation links between citing and cited publications
indexed in both data sources are considered

9.
Causes of discrepancies between data sources
• Inaccuracies in references
• Inaccuracies in reference data
• Inaccuracies in citation matching
• Multiple versions of a publication
• Multiple records for a publication
• References not having been deposited, being closed or not having been
matched
8

10.
Example: Discrepancies between Scopus and
Dimensions
9

11.
Example: Discrepancies between Scopus and
Dimensions
10

12.
Example: Discrepancies between Scopus and Web of
Science
11
Group author and/or supplement
seem to cause problems in Web
of Science

13.
Example: Discrepancies within Web of Science
12
September 20, 2017
November 1, 2017
November 8, 2017

14.
Example: Unmatched references in Crossref
13

15.
Conclusions
• Substantial discrepancies between data sources
• Reasonably complete citation data in Dimensions
• Large gaps in citation data in Crossref, due to references not having been
deposited, being closed or not having been matched
• Need for transparent high-quality citation matching algorithm
• Completeness and quality of other metadata?
14

16.
New opportunities for
bibliometric
visualization
15

17.
VOSviewer
16

18.
Users of VOSviewer
• Researchers
• Research institutions
• Research funders
• Scientific publishers
• Industry
17

19.
Data sources supported by VOSviewer
18

20.
Bibliometric networks
19
WoS
Scopus
Dimensions
PubMed
Crossref
Citation network
of pubs / journals / authors / orgs / countries
Co-authorship network
of authors / orgs / countries
Co-citation network
of pubs / journals / authors / orgs / countries
Co-occurrence network
of keywords / terms
Bibliographic coupling network
of pubs / journals / authors
Bibliographic
data source

21.
Types of networks supported by each data source
20
Co-authorship Co-
occurrence
Citation Bibliographic
coupling
Co-citation
Web of Science
    
Scopus
    
Dimensions
    
PubMed
    
Crossref
    

22.
Journal citation network based on Crossref data
21

23.
Demonstration
• Journals:
– Journal of Informetrics (Elsevier)
– Scientometrics (Springer Nature)
• Time period: 2007-2016
• Data sources:
– Dimensions
– Crossref
– OCC (OpenCitations Corpus)
– COCI (OpenCitations Index of Crossref open DOI-to-DOI references)
22

24.
23

25.
24

26.
25

27.
26

28.
Coverage of publications and citations for each data
source
27

29.
Wish list for improving open data sources
• Expanding coverage of publications (OCC)
• Opening citations (Crossref)
• Opening other metadata, e.g. abstracts (Dimensions, Crossref, OCC)
• Improving completeness and standardization of metadata (Crossref)
• Speeding up APIs
28

30.
Toward contextualized scientometrics
• Integrating data collection and visual analysis
• Interactive visual exploration (e.g., drilling down from high-level visual
overviews to underlying data)
• Large-scale visual analyses
• Moving from visualization as a tool for representation to visualization as a
tool for exploration, discussion and reflection
29

31.
Supporting open
citations
30

32.
Supporting open citations
31
www.issi-society.org/open-citations-letter/
April 26, 2018:
• 324 signatories
• 46 countries

33.
Supporting open citations
32
www.issi-society.org/blog/posts/2018/april/open-citations-to-open-science/

34.
Thank you for your attention!
33