Page 187 - Becoming Metric Wise
P. 187

179
                                                         Journal Citation Analysis

              scientific literature of 1974. We note that before the official launch of the
              JCR a preliminary version had been available as computer printouts based
              on 1969 data. This explains why one finds use of the JCR in the litera-
              ture, before the year 1976 (Inhaber, 1974). At first the JCR were printed
              as a volume of the SCI and the SSCI (but with considerable delay), later
              they became available on microfiches, then on CD-ROM and more
              recently as part of the InCites platform. The A&HCI has no accompa-
              nying JCR.
                 The complete matrix of all journal citation relations is very sparse,
              and hence consists largely of empty cells. Any given journal does not cite
              the large majority of the other journals in the database (this observation
              holds for any database of this type, not only for the JCR). If a journal
              cites another one exactly once this is usually not shown in the JCR.
              Taking this restriction into account, approximately 97% of all cells in a
              journal-journal citation matrix are empty (Leydesdorff & Jin, 2005).


              6.13.1 WoS Categories
              Journals included in the JCR are assigned to subject categories (SCs) of
              science, including a category Multidisciplinary Sciences. In this mixed
              group one finds top journals such as Science, Nature and the Proceedings of
              the National Academy of Sciences USA, but also journals with a much lower
              impact factor such as National Academy Science Letters   India and the
              Johns Hopkins APL Technical Digest. One of the categories in the social
              sciences is Information Science and Library Science. Assignment of
              journals to categories is not unequivocal as journals may be assigned to
              more than one subject category. This is illustrated in Table 6.12.
                 This table shows that more than 40% of all journals are assigned to
              more than one subject category. Overlap between subject categories is a
              form of cooccurrence. This aspect has been used by (Morillo et al., 2003)

                          Table 6.12 Division of journals over JCR SCIE subject
                          categories (for the year 2015)
                          6 SCs                        6 journals
                          5 SCs                        25 journals
                          4 SCs                        172
                          3 SCs                        820
                          2 SCs                        2765
                          1 SC                         4989
   182   183   184   185   186   187   188   189   190   191   192