The following discussion is an archived debate of the proposed deletion of the article below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.

The result was no consensus. Eddie891 Talk Work 15:36, 21 July 2021 (UTC)[reply]

Levon Pogosian[edit]

[Hide this box] New to Articles for deletion (AfD)? Read these primers!

Levon Pogosian (edit | talk | history | protect | delete | links | watch | logs | views) – (View log)
(Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL)

BLP with only references to his own stuff. Is he notable? Rathfelder (talk) 16:24, 26 June 2021 (UTC)[reply]

Note: This discussion has been included in the list of Astronomy-related deletion discussions. Rathfelder (talk) 16:24, 26 June 2021 (UTC)[reply]
Note: This discussion has been included in the list of Canada-related deletion discussions. Rathfelder (talk) 16:24, 26 June 2021 (UTC)[reply]
Note: This discussion has been included in the list of Academics and educators-related deletion discussions. Shellwood (talk) 16:48, 26 June 2021 (UTC)[reply]
Total citations: avg: 14258, median: 7641, Pogosian: 3999.
Total papers: avg: 143, med: 90, P: 86.
h-index: avg: 44, med: 37, P: 34.
Top citations: 1st: avg: 1433, med: 1093, P: 269. 2nd: avg: 905, med: 717, P: 263. 3rd: avg: 756, med: 542, P: 246. 4th: avg: 628, med: 369, P: 224. 5th: avg: 523, med: 336, P: 182.
He is well below the median in all these parameters, which already skew low since 20 papers is apparently super poor for this field (the median out of all 238 coauthors, including those with just 1 publication, is 72 papers!). JoelleJay (talk) 02:15, 27 June 2021 (UTC)[reply]
UPDATE: I have now added in all the coauthors who have attained positions beyond post-doc in physics (so, all 163 professors, staff scientists, senior researchers, etc). As expected with adding in all the people who only coauthored with Pogosian on giant cosmology collaborations, the average (and median) metrics increased across the board.
Citation metrics, 163 coauthors
Caption text
Current position Total citations (avg, med) Total papers (avg, med) h-index (avg, med) 1st (avg, med) 2nd (avg, med) 3rd (avg, med) 4th (avg, med) 5th (avg, med) N
L. Pogosian 3999 86 34 269 263 246 224 182
Profs/lecturers/readers only 18428, 11215 222, 150 53, 49 2012, 1159 1261, 722 1005, 565 789, 449 607, 383 121
Other post-postdocs (e.g. staff scientist, research engineer, senior researcher, etc.)(>20 papers) (avg, med) 7417, 4547 98, 72 32, 28 1292, 625 782, 408 605, 333 377, 256 311, 219 42
All post-postdocs 16746, 10436 217, 111 52, 43 1728, 1093 1101, 738 838, 549 687, 424 540, 377 163
Postdocs 2711, 1217 55, 43 20, 18 464, 239 335, 134 266, 101 196, 99 157, 79 23
JoelleJay (talk) 23:36, 9 July 2021 (UTC)[reply]
Relisted to generate a more thorough discussion and clearer consensus.
Please add new comments below this notice. Thanks, Seraphimblade Talk to me 05:18, 6 July 2021 (UTC)[reply]
I don't know how you can claim I'm comparing him "only to his extremely good co-authors". The cutoff of 20 papers is over 3x lower than the median number of papers of all 238 of his coauthors. 20 is around the number brand-new PhDs in that field have, so if we actually wanted to evaluate him in relation to other professors the cutoff would be much, much higher. And per NPROF For the purposes of satisfying Criterion 1, the academic discipline of the person in question needs to be sufficiently broadly construed. Major disciplines, such as physics, mathematics, history, political science, or their significant subdisciplines (e.g., particle physics, algebraic geometry, medieval history, fluid mechanics, Drosophila genetics are valid examples). limiting analysis to just cosmology is perfectly reasonable. I specifically excluded people who only were only coauthors with him on massive collaborations since they were less likely to be directly comparable (that is, within his sub-subdiscipline) and because they were more likely to substantially increase the average/median citations. But since as you suggest we should be evaluating with a broader scope (cosmology), I will go back through and update the metrics after adding in those authors.
Also, I am going off of what objective information we have available, which is primarily his citation record; claims of how difficult it is to get a cosmology faculty position and assertions that he would have a named professorship if he was in the US are SYNTH/OR and cannot contribute to notability considerations. JoelleJay (talk) 18:34, 9 July 2021 (UTC)[reply]
Thanks. By the way SYNTH/OR are okay for AfD discussions (the WP article about SYNTH/OR explicitly says that). The concern is when it's used for the actual content of an article. Technically speaking, your whole citation analysis is SYNTH/OR. I'll also say that even if his h-index of 43 and 6900+ citations were not considered above average for a cosmology researcher, that alone doesn't mean the article should be deleted. All eligibility criteria have to fail, in order for the subject to be ineligible (not just one of them). Also, sorry that we have missed each other on connecting with my part about "comparing him to only his extremely good co-authors", I did not mean that you're only comparing him to his best co-authors (e.g. high-profile professors rather than postdocs), what I meant was that his co-authors in general (even the grad students) have been extremely successful citation-wise, compared to other people (including grad students) in cosmology. Dr. Universe (talk) 16:45, 11 July 2021 (UTC)[reply]
  • Well, sure, but it's one thing to say "this field is ultra-competitive and unlike notability assessments for every other discipline a professorship in it should be an automatic NPROF pass, here are the references to support that claim" and quite another to just assert this is the case without providing any valid, sourceable reasoning. And of course if someone demonstrates Pogosian clearly meets another criterion I will update my !vote. I'll also reevaluate if you prove his coauthors are like 2 or 3x more successful than those of most other cosmology professors' coauthors. JoelleJay (talk) 19:02, 11 July 2021 (UTC)[reply]
  • I could try to figure out the average citation metrics for a cosmologist, but that would be a research project of its own. It seems that academic notability discussions on AfD have spiraled into something far from the original spirit of Wikipedia. The reason Wikipedia became so successful was because the ease of doing things led it to becoming far more useful to people than other encyclopedias where editing is not so accessible. However to keep this one article online, it seems we're being asked to carry out a full fledged research project as if it were a full-time job. I admire your efforts here, but the amount of effort you put into figuring out which of his 100+ citations are post-docs, which are staff scientists, which are professors, etc., and then figuring out each of their citation metrics, is quite extreme compared to what is done in other regions of Wikipedia. "If the basketball player is getting paid money to play, they're notable". "If the song landed on the Billboard 100 chart, it's notable". My !vote was to keep because I know that the average cosmologist has an h-index lower than 43 and fewer than 6900 citations. hroest's !vote was to keep because he knows/believes that 17 papers with 100+ citations is enough to be considered notable in this field. Your !vote was to delete because you know/believe that an academic can't be notable if their citations are too much lower than their co-authors' based on your metrics. In the end, our !votes are based on what we know, and hopefully there will be enough people !voting that we don't have to carry out a full-fledged research project to figure out whether or not an h-index of 43 and 6900+ citations is 2-3x more than the average cosmologist. Dr. Universe (talk) 08:03, 12 July 2021 (UTC)[reply]
The intent of NPROF is to provide wiki coverage of researchers who are exceptional, whose impact goes beyond that of the average professor in their area of scholarship. Almost all editors !voting in academic AfDs will be unfamiliar with the publishing standards of any given field, and many will be approaching AfDs with preconceived notions of how many citations is "a lot" based on their personal experience in whatever field they're in. But if we use such static, naïve bases for what constitutes a "notable" number of citations or h-index, we will inevitably end up over-covering academics in subfields where citations are much higher on average and under-covering those where it is lower. I saw Pogosian's Scopus h-index of 34 and assumed this would be an easy keep, and was surprised to discover his citation count was so far below the median (and 4 times lower than the average!) of 160+ cosmologists. I don't claim to know the average cosmologist has a higher h-index than 34; my belief is that these metrics demonstrate cosmology overall has such high publication/citation rates that most !voters' internal metrics would perhaps benefit from this additional context. I don't know whether hroest or any of the prior !voters would be swayed by these analyses or put any stock in them at all; but several subsequent participants have considered them relevant enough to inform their choice so it is not like I am the only one who finds this methodology reasonable. JoelleJay (talk) 20:59, 14 July 2021 (UTC)[reply]
Phil Bridger, yep, I absolutely agree (and have brought this up before on wikiproject physics). This is part of why I first looked at the coauthors of Pogosian who weren't on the handful of massive collaborations he was on (but still counted those citations for him). It's also why I look at coauthors in the first place -- to contextualize citation metrics within a particular discipline. C1 values always look super impressive in the ultra-high-publication subfields and are therefore always brought up in AfDs, so if we're going to have those numbers thrown around anyway it's much more informative to show how they compare to those of other people in that subfield than for us to base our !votes on our internal subjective perceptions of what a "typical citation record" looks like (which will of course vary wildly by how familiar we are with a particular topic). JoelleJay (talk) 18:44, 11 July 2021 (UTC)[reply]
Pattern of growth in viable f (R) cosmologies by L Pogosian, A Silvestri in Physical review D, 2008 has 386 citations
Dynamics of linear perturbations in f (R) gravity by R Bean, D Bernat, L Pogosian, A Silvestri, M Trodden - Physical Review D, 2007 has 358
Searching for modified growth patterns with tomographic surveys GB Zhao, L Pogosian, A Silvestri, J Zylberberg - Physical Review D, 2009 has 267
Testing gravity with CAMB and CosmoMC A Hojjati, L Pogosian, GB Zhao - Journal of Cosmology and Astroparticle Physics, 2011 has 223
Bounds on cosmic strings from WMAP and SDSS. M Wyman, L Pogosian, I Wasserman - Physical Review D, 2005 has 208
This amounts to at least 5 papers with refs over 200. The consensus of how many depends on field--recent decisions have been that in biomed science, the most heavily cited field, 2 papers with over 200 citations is is certainly enough (it used to be 1 with over 100, but citation density has increased) ... He has several times that.
JoelleJay is quite correct that judging by h factor alone is meaningless. Eugene Garfield, the inventor of the technique, thought so also,. h =34 could mean 500 400 399....34 or 40, 39....34. Only the first is a record that shows notability. Scientists are not judged by their routine work, but their best work. Using just h factor would be akin to judging an author by their average little known work, rather than having written 2 or 3 best-sellers. That's what the WP warning on just using h factor means. DGG ( talk ) 23:48, 17 July 2021 (UTC)[reply]
  • We just need to remove the ambiguity by looking att he citations to the work other than the many-multi-author papers. That's a reasonable take, although from what I've seen of other cosmologists I'm inclined to believe average citations are high even outside of mega-collaboration papers (which is reflected by my initial analysis, which excluded those coauthors who only worked with the subject on mega-collab papers). I'm always interested in evolving my methods, so I'm going to tweak my scripts to give me the top ten regular-author-number papers for each coauthor. JoelleJay (talk) 03:35, 18 July 2021 (UTC)[reply]
Perhaps citations are high in this field because of the general scientific interest in the work? that would be scientific impact, which is what WP:PROF notability is about. The necessary analysis here is which journals cite them. There is also the possibility that the quality of work in this field is exceptionally high, because of the extremely elite people it attracts (coi note: For some years I had supervisory responsibility for the astronomy library at one of the 3 or 4 leading centers for the subject.)
There is as you say, another source of error--determining the responsibility in multi-author (but not mega-author) papers; there is a method I have used--look at the citation record of each of the authors. That will distinguish between the major scientists and the students. It still has the error that the most senior people may have done none of the actual work besides getting the grant that supports it. This can usefully be analyzed further by seeing where in each of the people's trajectory the papers lies, and in what fields they have done additional work. Then, a further way for analyzing citation data, which I have used only a few times, is to compare with similar people who are clearly notable and clearly non-notable . (I'll note that this is essentially similar to one part of the the way faculty analyze the record of prospective candidates for positions or tenure; And this is in essence why I accept that an associate or full professor at any of the really important universities especially their most important departments will be notable, because the faculty has done the work of determining this, and they, not we, are the experts. As a librarian I helped them find data, but they did the analysis. )
There's a relevant paper on the subject of handling citations on authorship in a similar field, high energy physics: Peter Galison, "The Collective Author" pp. 325-356 in Mario Biagoli and Peter Galison, eds. Scientific Authorship: Credit and Intellectual Property in Science. Routledge, 2003. ISBN 9780415942928 . Based on the conference What is a Scientific Author held at Harvard in March 1997. (also online, but the Google Books preview does not include this paper. There is said to be an authorized available online copy at https://galison.scholar.harvard.edu/files/andrewhsmith/files/galison_collective.pdf but I have not yet been able to successfully download it. DGG ( talk ) 19:20, 19 July 2021 (UTC)[reply]
  • Perhaps citations are high in this field because of the general scientific interest in the work? that would be scientific impact, which is what WP:PROF notability is about. The necessary analysis here is which journals cite them. There is also the possibility that the quality of work in this field is exceptionally high, because of the extremely elite people it attracts
    How is this consistent with NPROF C1 (emphasis mine) The person's research has had a significant impact in their scholarly discipline, broadly construed and For the purposes of satisfying Criterion 1, the academic discipline of the person in question needs to be sufficiently broadly construed... Overly narrow and highly specialized categories should be avoided? If the average number of citations per paper is just scaled up from whatever we think is typical for our fields, that's not an indication that that field is "more important" or that a larger proportion of people in it are notable or more skilled... And I should also note that these papers still have a lot of authors -- I used 20 as a cutoff because otherwise the comparisons would be among a very tiny number of coauthors. So while 300 citations might seem high, it could be perfectly in line with what is expected from 20-author papers, especially if each paper has a high number of references itself. But anyway, we're supposed to be comparing within major subfields, not with professors in general, and within the major subfield of cosmology—even when excluding mega-collab authors—Pogosian comes up below the median for researchers beyond post-doc. And I definitely don't think there's consensus that associate professors are normally notable at major institutions.
    The Galison paper (which I could access from your link) is definitely interesting and gives a very good overview of how HEP collab papers were structured in the late 90s, although it is more philosophical than informative on how we should interpret authorship (other than to disregard alphabetized authorship on mega-collab papers altogether from notability analysis, which I would support). JoelleJay (talk) 22:48, 19 July 2021 (UTC)[reply]
yes, he does not give a solution. Faculty committees consider non-quantitative information also, in a way which we can not, tho we can use the decisions they make on such bases. Though there is indeed no consensus at WP that associate professors (which always implies tenure) at the highest quality institutions are notable at WP, can you really imagine the faculty of such a department appointing to tenure someone who would not be? DGG ( talk ) 07:52, 20 July 2021 (UTC)[reply]
I agree with that in principle, but foresee, if it is enacted, loads of arguments about what is a "highest quality institution". Phil Bridger (talk) 17:40, 20 July 2021 (UTC)[reply]
My evidence about biomed is based on recent debates. I've been folllowing the afds in this area for 15 years now. In 2006, the acceptance of WP:PROF was still uncertain: I recall arguing that if we instead used GNG the literal way itt was being interpreted, then anyone whose had 2 papers, each of which had a substantial comment in a paper from a different author, would meet it., and that this was absurd, because essentially all postdocs even would meet that standard. Garfield when he developed citation analysis considered anyone with 1 paper over 100 to be a "Citation classic, and for many years we followed that rule. But the expected number of citations depends upon the publication density in a field, and , especially, the usual number of citation for each paper, and this has gradually been increasing. For about the last 5 years the practice in biomed was 2 or more papers over 100, but in the last 1 or 2 years some afd debates started asking for higher. (I steadily resisted each increase, so what I'm giving is not my standard, but the higher standard of the community). The people who asked for higher have usually suggest 200 as the cutoff, and were usually satisfied with 1 such paper.I am giving the safe level--there are still biomed people being found notable with considerably lower records, but I do not think any with 2 or more over 200 has been found not to be notable, in biomed or any other field, if the citation numbers were given, with a few rare exceptions. . (There are exceptions to this for people in fields sometimes not taken seriously here, such as education or home economics , or for people against whom there is prejudice based on their having, in addition to their important true scientific work, some connection with pseudoscience, usually in their later year). It would be interesting to do an analysis including the ones where individual citation count isn't mentioned.
I'm accustomed to arguments where there is some realistic doubt about the notability or the significance of the count--of course people can disagree about where to draw the line, and I know I'm relatively inclusive in this field. But this really does seem a very clear keep and I don't see the reason for makign a test case out of it. Years ago I did spend much of my time here defending afds on researchers, but now I'm more concerned with keeping out promotionalism . I'm not sure where my priorities would lie between the two. . DGG ( talk ) 01:02, 21 July 2021 (UTC)[reply]
The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.
More