Notes on measuring notability
editThere has been a lot of discussion at WT:NPROF, see for instance , , . To me h-factors are a bit like grading students; it is an important and generally unbiased metric despite its failings. In the following I will omit HEP, since their h-factors are absurdly high and I avoid them if I can.
To me what matters is a comparison to peers. This can be done by any of (ideally a combination of):
- Others in the areas they list in a Google Scholar profile. Go to their topics, with care that they are not excessively restrictive. Based upon these you can get an idea of what the top people of the area have in terms of numbers. For instance, if they have > 50K cites and h-factors of > 80 then it is a high-citation field; numbers like 10-20K cites indicate a low citation area.
- The h-factors of their co-authors (not grad students or postdocs) who are senior academics, ideally those with WP profiles. Apply similar criteria to 1. above.
- The quality of the publications, i.e. Science (journal) or Journal of the American Chemical Society (field dependent of course) should be favored over low citation, marginal or predatory journals. There are areas where conference articles are the most prestigious places.
- Whether they have any very highly cited publications. I will give "bonus points" for > 1K citations, but not in general for > 100 except in low citation fields. If they have a single publication with > 5K cites that can be enough by itself. Be cautious of large team papers where, sometimes, there are more authors than cites! With these I would reduce their importance unless they are either the first or last authors -- with a caveat if their surname is Aaardvark of course!
- Where they are in the article order. First is most prestigious, then last (the senior author aka boss). Check that there is no "equal contribution" tag. Some journals also require a statement of who did what which requires more work but can be very informative.
- Your knowledge of the area, if there is overlap.
- Past experience, particularly if you have participated a lot in AfD.
- Relevant tenure cases if you have sat on committees (not everyone has).
Some care is needed when evaluating where articles are published. Often editors cite "published in Nature" or similar, implicitly inheriting the notability of the journal -- which is not accepted on Wikipedia (see Wikipedia:NOTINHERITED). It is also common to see the https://topresearcherslist.com/ list or others. (see RfC about Stanford/Elsevier top 2% in Wikipedia talk:Notability (academics)/Archive 15.) My opinion, at least partially supported by this discussion, is that being part if these lists does not by itself confer notability.
Which source is used for citations can matter. In rare cases Google Scholar can have erroneous entries, in very rare cases deliberately so. (Some years ago one of my postdocs found all his papers on someone else's profile.) Scopus is an alternate which can be customized, but gives lower values often because it does not include all articles.
As a rough guide, in solid state-physics/chemistry/materials science I would look for an h-factor if > 50 and more than 8K cites as definite evidence of notability. In math that should be divided by 2, and in HEP probably multiplied by 3. It is very important to check against their peers.
I find that my conclusions generally are close to concensus at AfD, although I have been both more and less accepting in some cases. One case where I seem to be more accepting is AfD Rachid Ouaissa as I feel this is a low citation area. A second low cite area where I was part of the Keep concensus is Kirtiraj Gaikwad. Another case where there were no awards and other issues but the h-factor was decent and the publications are in top journals is Niall J. English. Finally, a case where I did not consider her comparable to her peers (and is in my area) is Hazel Assender, although others disagreed with me.