The latest is an article which somebody posted on my Facebook column. It is uninteresting to see what conclusions are drawn from the graphs and lists published there, as the data are quite incomplete - in the h-index-ordered list of Italian researchers I do not appear, for one, but similarly do not dozens of top scientists who have even higher h-indices.
(According to google scholar, my h-index is 137 and I have so far collected over 60,000 citations in my 1000 or so publications, as shown above. That would mean I am the 10th most productive italian scientist overall - yeah, right).Now, the journalists are of course doing a poor job, but are they to blame ? The real blame is on who uses those indicators as anything worth discussing without a careful definition of the boundaries. Let me explain.
The graph below, which shows the discipline of the "top 200" italian scientists, could be taken to show that in Italy, physicists are way more productive than other scientists. But is that true ? Not necessarily, and not by that measure. For you cannot compare the h-index, or any other bibliometric indicator, across disciplines. Even within physics it is far-fetched to do so, as a condensed-matter physicist will never ever have a chance to publish at the rate of scientists who collaborate with 3000-member experiments. [Fisica = Physics; Sc. Biomediche = Biomedical sciences; Chimica = Chemistry]
Alas, the problem of assessing the relative merit of different scientists remains a unsolved one, because no metric can do that in a exhaustive and objective way. And that is a pity. Or maybe not. In fact, assessing with hard numbers the value of scientists would embarrass those who are called to rank them whenever a selection is made to hire somebody or to give her or him a higher qualification within an institute. Better to give selection committees the freedom to choose according to their own taste, or more arcane formulas...
Comments