Page 162 - Becoming Metric Wise

153
Publication and Citation Analysis

linguistics are best served by including GS results. Yet, studies with wider
disciplinary coverage showed that the coverage of GS is variable and can
be unreliable for some subdisciplines (Kousha and Thelwall, 2007).
Leydesdorff (2012) compared publication trends for China, the USA,
EU-27, and smaller countries as derived from the WoS and Scopus.
Compared with an earlier version of the WoS interface he found that
China no longer grew exponentially during the 2000s, but linearly.
Consequently, the cross-over of the lines for China and the US was post-
poned in time with respect to predictions based on an exponential
growth. He concludes that besides the dynamics in publication trends,
one also has to take into account the dynamics of the databases used for
predictions.

5.19 FINAL REMARKS

Citation data made available by Clarivate Analytics or Scopus (Elsevier)
are behind a paywall. As a consequence informetric papers can rarely
comply with requirements of making data open because of the license
restrictions on which their results are based. One may wonder why they
are not freely available for everyone. Expertise is needed to handle and
interpret them, not to collect them. Hence, it is not surprising that voices
have been raised to consider citation data as part of the commons and
placed in an open repository (Shotton, 2013).
As a cautionary note ending this chapter we would like to point out
that numbers of citations can certainly not be equated to scientific original-
ity, let alone to the mark of geniality. The larger the audience the higher
the citation potential, and conversely, the smaller the audience the smaller
the chance to get cited. Moreover, paradigm changing discoveries have
notoriously limited early impacts (Wang et al., 2013) because the more a
discovery deviates from the current paradigm the longer it takes to be
appreciated by the community. In the context of citation analysis, we also
notice the phenomenon of superspecialization: some topics (e.g., in pure
mathematics) are studied by only a handful of scientists. Articles and scien-
tists dealing with these topics can never become highly-cited on an overall
scale. Moreover, like any human endeavor also science knows topics that
are temporarily “en vogue” and hence articles dealing with these topics
receive—temporarily—much more citations than expected (than
deserved?), see e.g., Rousseau et al. (2013) for the case of the h-index.