Page 132 - Becoming Metric Wise

123
Publication and Citation Analysis

name (Rousseau, 2002a) and Pascal’s triangle was known to Jia Xian
(c. 1010 1070) in China and to Omar Khayyam (1048 1131) in
Persia. We even have Stigler’s law (Stigler, 1980) which states that no
scientific discovery is named after its original discoverer (of course,
Stigler’s law was not discovered by Stigler).

5.8 THE ETHICS OF CITING
Citing is also an ethical question. Citing an article, and consequently, its
authors is recognizing an intellectual debt. Honesty requires that this is effec-
tively done. Not citing on purpose is a form of dishonesty, sometimes
referred to as idea-plagiarism (Chaddah, 2014). Yet not trying to retrieve
who was first or who discussed a topic in depth is a form of laziness that
takes from others something they deserve. As with most ethical problems it is
difficult to draw a line. Which references should one include in a reference
list and which are considered not influential enough for one’s research?
Moreover, related to the retrieval question: Should a western scientist ask a
specialist to search in Russian, Persian, Chinese, or Japanese language data-
bases? And should one expect Russian, Iranian, Chinese, or Japanese collea-
gues to search through English, French, or Spanish language databases?
Kelly P. Downing in (Sills, 2014) warns against not checking original
sources and hence perpetuation of invalidated ideas or misinterpretation
of data.
The more citations play a role in evaluations of scientists the more not
citing takes away from colleagues not only honor or prestige, but also
money and possibly progress in their career.

5.9 CITATION NETWORKS AND THE MATHEMATICS OF
CITATION
5.9.1 Generalities on Citation Networks or Graphs

When document d i cites document d j this can be represented by an arrow
linking a node representing d i to a node representing d j (see Fig. 5.3).
In this way the documents of a collection D form a directed graph or
network, which is called a citation network. Using a so-called adjacency
matrix is another way of representing a citation network. In this representa-
tion one considers a matrix (a table consisting of rows and columns). In this
adjacency matrix a 1 is placed in the cell for which the row corresponds to