Page 219 - Becoming Metric Wise

210 Becoming Metric-Wise

• There is a problem finding reference standards, leading to questions
such as “What is a ‘high’ h-index in field F for a researcher who is
active since 7 (or any other number of) years?”
• The number of coauthors may influence the number of citations
received.
• It is rather difficult to collect all data necessary for the determination
of the h-index. Often a scientist’s complete publication list is necessary
in order to discriminate between scientists with the same name and
initial(s). We refer to this problem as the disambiguation problem.
Some other disadvantages are more specifically related to the h-index
itself.
• The h-index is only useful for comparing the better scientists in a
field. It does not discriminate among mediocre scientists. This
follows from the facts that a) it is a rough indicator, always resulting
in a slowly increasing, natural number; and b) when a scientist has
just a few publications, say P, then the possible values of the corre-
sponding h-index are restricted to the natural numbers in the
interval [0, P].
• The h-index lacks sensitivity to performance changes: it can never
decrease and is only weakly sensitive to the number of citations
received. Concretely, once an article belongs to the h-core it does not
matter anymore how many extra citations it receives. Conversely also
for most articles in the tail the exact number of citations or increase
in citations has no influence. In this context an article is called an
h-critical publication (Rousseau and Ye, 2012) if one additional
citation to this article would increase the h-index, and hence would
let this article enter the h-core.
• The lack of sensitivity is the main reason for the introduction of most
other h-type indices.
• From a purely logical point of view, the most important disadvantage
of the h-index is the fact that it is not an independent indicator. This
term and how it applies to the h-index is explained further on, see
Subsection 7.3.3.
The following points have also been mentioned as disadvantages but
we do not fully agree with this point of view.
• The h-index, in its original setting, puts newcomers at a disadvantage
since both publication output and observed citation rates will be
relatively low. In other words, it is based on long-term observations.