[Coral-List] A student's guide to the h-index
fathom5marineresearch at gmail.com
Wed Dec 30 09:25:03 UTC 2020
I am directing this discussion of the h-index primarily at students
contemplating an academic research career. By the time you start research I
hope the h-index will be less influential than it is now, but I’m not
holding my breath. If you’re unsure what the h-index is, see:
The original 2005 paper introducing the h-index is here:
If you haven’t read it, I recommend reading it now so you can form an
Ok, what did you think?
I think this paper demonstrates the pitfalls of applying math to mind.
The first casualties are the concepts of productivity and significance.
English dictionary definitions of these terms require integration of
multiple concepts, each ramifying through further layers of sub concepts.
In h-world, such technicalities are superfluous; productivity = number of
papers (*Np*) and a significant paper = a paper with >*y* citations.
The *reductio *is extended to define three categories of scientist, based
on the rate *m* at which their h-index increases over time: *m* = 1:
successful scientist, *m* = 2: outstanding scientist, *m* = 3: truly unique
scientist. For me, this train has left the rails already, but there’s more:
specific h-values are suggested for advancement to associate professor,
full professor, fellowship in the American Physical Society, and membership
of the National Academy of Sciences.
The author, Jorge Hirsch, lists several caveats that should be kept in mind
when interpreting the h-index, but these nuances don’t register. The paper
is published, it catches like fire, and pretty soon hundreds of thousands
of scientists are heading for true uniqueness.
University administrators and committees love the new index because it
finally brings some much-needed objectivity to their decisions on tenure,
promotion, and grant distribution. This is helpful because, as universities
are now businesses, decision-makers are poorly-equipped to evaluate the
quality of scientists and their research proposals (but remain fully
capable of ranking a sequence of two-digit integers from worst to best).
Is this really such a big problem? I think it is. More importantly, Jorge
Hirsch thinks it is:
What can we do about it?
An engineer colleague of mine has a maxim within his team: don’t bring me a
problem unless you have at least two potential solutions.
I have two potential solutions. I don’t know if they will make a
difference. Let me know if you want to hear them.
More information about the Coral-List