Borstel, R. C. von. "An Analysis of Citation Analysis," Letter to the Editor. Science 251 (29
March 1991), p. 1546.
This is an elegant letter to the editor, a scathing attack on citation index analysis. The immediate
object is the stuff published by Pendlebury and his claims about the utility of "citation analysis."
This author argues that citation indexing produces nonsense and he suggests some of the factors
which lead him to draw this conclusion: methods papers should be removed from the analysis
simply because they receive too many citations. Review articles should be removed from the
citation analysis. Three other kinds of papers to be removed: 1) the suspected fraudulent paper; 2)
the controversial paper; and 3) the ‘pizazz' paper. Another class to be removed: papers cited by
friends and those with multiple authorship. All those papers which fail to cite competitors or
unfriendly folk should also be eliminated. Finally, a method of analysis should be found for those
non-authors who failed to write the suggestion which led to a publication which has been issued
by someone who overheard it. There is not much left for analysis.
A lovely blast!