Page 114 - Becoming Metric Wise

105
Publication and Citation Analysis

articles in more than 22,000 journals and goes back for most journals to
the year 1996. In Section 5.16 we provide a more complete description
of Scopus. Via SCImago and also via the CWTS Journal Indicators
(http://journalindicators.com) Scopus describes complete journals in a
similar way as the JCR (see Chapter 6, Journal Citation Analysis).
That same year (2004) Google launched a beta version of Google
Scholar (GS). Also this database provides citation data, but its most signifi-
cant disadvantage is that it is unclear which journals or websites are cov-
ered and which are not, see further Section 5.17.
Since the end of the 1980s regional databases have been installed (first
in China). These databases cover the scientific production (or part
thereof) of a country or region. One finds several regional databases in
China (see Jin & Wang, 1999; Su et al., 2001; Wu et al., 2004), Japan,
Taiwan, South-America (SciELO), Russia, India and Spain. Such data-
bases mainly, but not exclusively, cover journals published in a local lan-
guage (Chinese, Japanese, Spanish, Portuguese, Russian). As commercial
databases will never cover the complete scientific literature, such local or
regional databases are a necessary complement.

5.2.4 Advantages of a Citation Index
What are the advantages of a citation index with respect to an index cov-
ering a specific literature, such as medicine or mathematics? (Garfield,
1979).
• Citations and hence, on a higher level, citation indices represent topic
relations.
• Authors are the best qualified persons to link their article with those
written by other scientists. They do this better than professional
indexers.
• Citation indexes are not influenced by specific systems of subject
description.
• A citation index is by construction multidisciplinary.
• A citation index is not influenced by semantic problems, occurring
when the same word means something different in different fields or
when the same property is described by different terms.
Finally an important advantage in the beginning years was the speed
with which a new version of the SCI was made (a new version every 3
months, with yearly compilations). Nowadays, as the publishing process is
done electronically this advantage has largely ceased to exist.