This paper reviews a number of studies comparing
Thomson Scientific’s
Web of Science (
WoS) and Elsevier’s
Scopus. It collates their journal coverage in an important medical subfield: oncology. It is found that all
WoS-covered oncological journals (
n = 126) are indexed in
Scopus, but that
Scopus covers many more journals (an additional
n = 106). However, the latter group tends to have much lower impact factors than
WoS covered journals. Among the top 25% of sources with the highest impact factors in
Scopus, 94% is indexed in the
WoS, and for the bottom 25% only 6%. In short, in oncology the
WoS is a genuine subset of
Scopus, and tends to cover the best journals from it in terms of citation impact per paper. Although
Scopus covers 90% more oncological journals compared to
WoS, the average
Scopus-based impact factor for journals indexed by both databases is only 2.6% higher than that based on
WoS data. Results reflect fundamental differences in coverage policies: the
WoS based on Eugene Garfield’s concepts of covering a selective set of most frequently used (cited) journals;
Scopus with broad coverage, more similar to large disciplinary literature databases. The paper also found that ‘classical’,
WoS-based impact factors strongly correlate with a new,
Scopus-based metric, SCImago Journal Rank (
SJR), one of a series of new indicators founded on earlier work by Pinski and Narin [Pinski, G., & Narin F. (1976). Citation influence for journal aggregates of scientific publications: Theory, with application to the literature of physics.
Information Processing and Management,
12, 297–312] that weight citations according to the prestige of the citing journal (Spearman’s rho = 0.93). Four lines of future research are proposed.
相似文献