Page 131 - Becoming Metric Wise

122 Becoming Metric-Wise

Using complete-normalized counting assigns 1/3 of a credit to country
C 1 and 2/3 of a credit to country C 2 (based on the number of authors). If
this information is not available (which used to be the case in the WoS
before 2008) then one assigns credits based on the institutional addresses.
Then country C 1 and country C 2 each receive one half of a credit.
Harmonic counting gives country C 1 1/(1 1 1/2 1 1/3)5 6/11 credits
and country C 2 (1/2)/(1 1 1/2 1 1/3)1 (1/3)(1 1 1/21 1/3) 5 3/11 1 2/
11 5 5/11 credits. Again, if this information is not available then C 1 receives
1/(1 1 1/2)5 2/3 credits and country C 2 1/3 of a credit.
Obviously the way in which credits are assigned can influence the out-
come and hence country rankings to a large extent. This has been stressed
by Gauffriau and Larsen in several publications see e.g., Gauffriau and
Larsen (2005a,b). Although assigning credits based on the number of con-
tributing authors of a given country (possibly taking the author rank in the
by-line into account), is the only correct way, it is only since 2008 that this
is possible in the WoS. Before that time WoS records just gave a list of
authors and a list of institutes (including the country of the institute).
Counting citations leads to the same problems as counting publica-
tions. Yet, one may say that the problem is doubled as one may study
citations given and citations received, and the problem occurs in the two
directions. Citations are assigned according to the contributions of each
author. If an article with 4 authors received 10 citations, complete-
normalized counting assigns 2.5 citations to each author. If, however, one
has used first author count, then those ten citations are assigned to the
first author, and the other ones do not receive any citation credit.

5.7 A NOTE ON EPONYMS

It is sometimes said that eponyms present the highest form of scientific
recognition. An eponym is a thing, a theorem or a method named after
a person. Well-known examples are Newton’s Laws, Kepler’s laws,
Pascal’s triangle, the periodic table of Mendeleev, the Erlenmeyer flask,
Einstein’s relativity theory or Alzheimer disease. In the field of infor-
metrics we know the laws of Bradford, Lotka, Zipf and the Hirsch
index (all discussed further on). Yet, some remarks are in order. First, to
eponymize is naming things after persons, rather than scientific content.
Second, many eponyms are given to the wrong person (but this also
happens for citations). In this context one uses the term palimpsests.
Zipf was not the first to observe or describe the regularity that bears his