Counting of Citations," by Loet Leydesdorff and Tobias Opthof (2010), deals with an important subject, the construction of "field normalized" indicators of the citation impact of scientific-scholarly journals.
Many authors have underlined that it is improper to make comparisons between citation counts generated in different research fields because citation practices can vary significantly from one field to another. For instance, articles in biochemistry often contain over 50 cited references whereas a typical mathematical paper has perhaps only 10. This difference explains why biochemical papers are cited much more often than are mathematical ones. Eugene Garfield's (1979) view on this is clear: "Evaluation studies using citation data must be very sensitive to all divisions, both subtle and gross, between areas of research; and when they are found, the study must properly compensate for disparities in citation potential" (p. 249). Advanced citation indicators should account for differences among subject fields in the frequency at which articles cite other documents, and the rapidity of maturing and decline of citation impact.
One of the solutions is applying the idea of source normalization, termed by Zitt and Small (2008) as "citing-side normalization." It can be carried out in several ways, but the base idea is that the actual citation rate of a set of target papers in a subject field is "normalized" or "divided" by a measure indicating the frequency at which articles in that field cite other documents. Moed (in press) developed a source normalized indicator of journal impact (SNIP) that was calculated for all journals indexed for Elsevier's Scopus. Since January 2010, this indicator is included in Scopus via the Web site www.journalmetrics.com and also is available at the site www.journalindicators.com hosted by the Centre for Science and Technology Studies at Leiden University, the institute at which SNIP was developed.
Both Leydesdorff and Opthof's (2010) proposed indicator and SNIP are based on source normalization. Moreover, Leydesdorff and Opthof adopted the idea underlying SNIP