[HN Gopher] Among top researchers 10% publish at unrealistic lev...
___________________________________________________________________
Among top researchers 10% publish at unrealistic levels, analysis
finds
Author : crescit_eundo
Score : 119 points
Date : 2025-02-18 18:16 UTC (4 hours ago)
(HTM) web link (www.chemistryworld.com)
(TXT) w3m dump (www.chemistryworld.com)
| crescit_eundo wrote:
| https://archive.ph/SnYjB
| seydor wrote:
| note that the 2% list is very large, thus most of the scientists
| are not just the leading edge superstars but also run of the mill
| scientists and principal investigators.
|
| The authors focus on the citation inflation in younger scientists
| but that s unfair (it's only 1000 out of the 20000 in the top
| 10%). The reality is that older established scientists are much
| more advantaged because the people of their lab are sought after
| as collaborators , and they autonatically get an authorship as
| last authors by virtue of being principal investigators, even if
| they don't even take a look at the paper. It's thus not strange
| that they get 35+ authorships per year
| johnea wrote:
| This is pretty much what I came here to post.
|
| Anyone who's even been adjacent to scientific academia knows
| that the lead author is typically the advisor to the grad
| student or post doc that actually did the research and wrote
| the paper.
| asdff wrote:
| In my experience the lead author is always the post doc or
| grad student who actually did the research and wrote the
| paper, while the advisor is the last author and they
| developed the grant to fund both the grad student and post
| doc. If you collaborate with another lab, their post docs go
| towards the front behind your name and the advisor goes
| somewhere at the end. If two people equally wrote a paper
| they might be 1 and 2 and get an asterisk indicating they are
| both co first authors. If anyone was involved at all in the
| work e.g data collection or anyone involved with running an
| assay they go somewhere in the middle. Perhaps it's different
| in other fields
| buildbot wrote:
| Yep, this is what I learned in Grad school as well. First
| author is the primary person doing the work typically, then
| their advisor will be the last, and then you work towards
| the middle. So names in the middle of a paper typically
| contribute less than either edge!
| jltsiren wrote:
| This sounds like a terminological issue. Without further
| context, I would assume that "lead author" means the
| last/senior author, not the first/junior author. But I
| guess that's a life sciences tradition.
|
| "Who actually did the research" is not always an accurate
| description of the first author. There are plenty of
| papers, where the last author had already contributed
| enough to justify authorship before the first author was
| even hired. You often need to develop the idea and get
| preliminary results to convince someone to fund the
| project, before you get the money to hire the first author
| to work on the idea.
| danielthor wrote:
| When I was a grad student the only time my PI was listed
| first on a paper was when he actually did most of the work
| and I just helped with maths. Otherwise the PI was last and
| the grad students and post docs doing the research up
| front. People doing data collection or other ancillary work
| were mentioned in the acknowledgement section at the end.
|
| When the author count is much higher, like in the thousands
| in high energy physics, there are different norms like
| listing the authors alphabetically if the paper summarizes
| many significant achievements.
|
| In the in-between author order is negotiated.
| levocardia wrote:
| ...and imagine how well it would go over for a grad student
| to say "actually Dr Advisor, you don't meet the criteria for
| authorship"
| throwawaymaths wrote:
| I've seen someone who desperately wanted their nobel
| laureate PI on their paper but the professor did not put
| his name on the paper because it was all their idea (even
| though it was his grant, he reviewed and ok'd the paper,
| paper was relevant to his lab, and he really spoke
| effusively at their thesis defense)
| zevon wrote:
| This varies greatly by discipline and tradition - sometimes
| even within the same research groups. For example, some
| people put the senior person / advisor first, some put this
| person last, some go by alphabet.
|
| In my experience, there is even more variation in the amount
| of work expected to be put into a publication by said senior
| person. To put it mildly. ^^
| mvdtnz wrote:
| These researchers compared the publish rates of Stanford Top 2%
| scientists with comparable Nobel laureates and the rate of
| publishing among the anomalous researchers was ludicrously
| higher than their peers at the top of their game - "up to 212
| publications per year, and 5792 new coauthors per year,
| compared to up to 28.3 publications per year and 173 new
| coauthors per year on NL [nobel laureates]".
|
| Do you have any reason to believe the T2 group is composed of
| "older established scientists" who are "much more advantaged"
| than the group of Nobel Laureates? To the tune of 10x the
| number of publications and 33x coauthorships?
| seydor wrote:
| > reason to believe the T2 group is composed of "older
| established scientists"
|
| there were 1000 academics with < 10 years of career and 19000
| with more (in the inflators list). I did not compare their
| advantage to the nobel scientists but to the younger ones.
| (Also the nobel laureates are extreme outliers who don't need
| any more citations to satisfy their ego or funding needs)
| dleeftink wrote:
| It's interesting how the observers of this preferential
| attachment effect have since become, well, _very_
| preferentially attached [0]. The scaling and diffusion of ideas
| is rarely (and unfortunately), equally distributed.
|
| [0]:
| https://scholar.google.com/scholar?q=Emergence+of+scaling+in...
| fph wrote:
| Authorship norms in certain fields seem completely broken: a PI
| shouldn't become an author just for providing money and a lab.
| Yet unfortunately that's the norm in many fields.
| amelius wrote:
| The 10x scientists.
| marginalia_nu wrote:
| To be fair, there has always been extreme outliers in
| scientific output, like Gauss and Euler.
| LudwigNagasena wrote:
| > In our opinion, it is difficult to believe that the majority of
| authors with excessive rates are able to consistently produce
| large quantities of high quality or groundbreaking research, with
| their input to each paper being substantial and within norms of
| what constitutes co-authorship
|
| Huh, I thought providing the general direction and sanity-
| checking the results _is_ the norm of what constitutes co-
| authorship for a top professor.
| mvdtnz wrote:
| Do you think any one person can provide direction and sanity-
| check the work of more than 5,000 papers in a year? I don't.
| thinkingtoilet wrote:
| Where did you get the 5000 number?
| mvdtnz wrote:
| From the research paper that the article is discussing.
|
| https://www.tandfonline.com/doi/full/10.1080/08989621.2024.
| 2...
|
| > Namely, rates on T2 of up to 212 publications per year,
| and 5792 new coauthors per year, compared to up to 28.3
| publications per year and 173 new coauthors per year on NL.
| This is despite that the NL list is weighted with more
| senior researchers, often at the pinnacle of their careers
| - and hence - likely to have higher rates.
| LPisGood wrote:
| That is 212 papers, not 5000. 5000 is the number of new
| authors of those 200 papers.
| Xelynega wrote:
| Looks like the number is off, but is their point any
| different with 212 papers? I still think that it would be
| impossible to give any meaningful support to 212 papers
| in a year. That's a new paper every 1.72 days.
| aaplok wrote:
| Those 200 papers correspond to 4 papers per week, or more
| than 1 paper every other day for the whole year, weekends
| and holidays included. Coauthored with thousands of
| people they'd never published with before. It's a rather
| incredible rate.
| araes wrote:
| And that's before you have to try and actually write for
| grant funding, perform any actual substantial research
| (that took more than 1.5 days), serve as an editor or
| reviewer to do any of that "peer reviewed" part, go to a
| conference on your subject, deal with students, teach, or
| really anything else with academia.
|
| Writing for grant funding alone has already been a big
| complaint about how long and difficult it can be with
| limited success rates in a lot of fields. Time
| constraints of being a reviewer and how much of your own
| academic time you have to give up has been a major
| complaint. They're all time consuming. And then a paper /
| 1.5 days.
|
| Really, from having been in academia for a little while,
| putting out something well thought out once a month seems
| rather fast.
| anticensor wrote:
| My stance is at least one full concept or one data point or
| else no authorship to that person.
| baxtr wrote:
| What if you're the guy who is keeping the lab running, ie the
| guy who is securing the funding?
| dekhn wrote:
| Simply securing funding and being a manager does not
| qualify for authorship (based on the read on multiple
| journals submission policy). I don't think this is
| universally observed or even really policed, so it's fairly
| common to see people who really did not contribute anything
| (even sitting in on a weekly progress report and giving
| feedback) listed as the last/corresponding author.
| sgt101 wrote:
| Which should be treated as misconduct.
|
| Folks who do this should be shunned by their peers and
| driven out of academia. It's exploitative and coercive.
| EA-3167 wrote:
| Unfortunately people like that tend to run academia.
| sieste wrote:
| Not as exploitative as the anonymous reviewer report that
| asked to add 15 additional papers to our lit review ...
| all by the same author.
| hyeonwho4 wrote:
| ICMJE guidelines are frequently cited by other
| publishers, with the caveat that it depends based on the
| field. (Which makes sense. A "substantial contribution"
| to the LIGO collaboration might be much less direct than
| one in ML or field geology.)
|
| > Substantial contributions to the conception or design
| of the work; or the acquisition, analysis, or
| interpretation of data for the work; AND
|
| > Drafting the work or reviewing it critically for
| important intellectual content; AND
|
| > Final approval of the version to be published; AND
|
| > Agreement to be accountable for all aspects of the work
| in ensuring that questions related to the accuracy or
| integrity of any part of the work are appropriately
| investigated and resolved.
|
| So you are right. The corresponding author who got
| funding for the postdoc who designed the experiment for
| the graduate student that supervised undergradutes and
| wrote the paper would have a pretty tenuous commection to
| authorship.
|
| But good luck to everyone else in that chain publishing
| without them.
|
| Ref: https://www.icmje.org/recommendations/browse/roles-
| and-respo...
| margalabargala wrote:
| > even sitting in on a weekly progress report and giving
| feedback
|
| Depending on the specific nature of the feedback, someone
| who does something describable as this could be providing
| anywhere from 0 to 99% of the scientific insight provided
| by the paper.
| dekhn wrote:
| Yes. For one of my hosted interns, I was invited to the
| weekly meeting (the intern was "working for" another
| team) and asked for insight, before being added to the
| paper. I had expected to be listed simply for being the
| intern host but they didn't agree. I have mixed feelings
| about this. Personally, I think if a PI runs a lab,
| brings in all the funding for a grad student, and doesn't
| really get deeply involved, they should still be listed
| because they provided the environment in which the
| research could be done, but I've been told this is a
| controversial opinion.
| margalabargala wrote:
| I'm not sure I understand what it means to be the "intern
| host" in this context or why that matters for authorship,
| if the intern was working for another team, and that team
| was the one working on the paper.
|
| Regardless, if you showed up to a weekly meeting, and
| provided insight into the data provided, then it does
| sound like you deserve authorship.
| godelski wrote:
| In general the last author (in CS) is the one managing.
| I'm sure many a PhD student can tell you that their
| advisors, who are on every paper they write, have at
| times provided little to no input (potentially providing
| negative value, at least with respect to the research
| topic at hand).
| soperj wrote:
| Doesn't sound like you're doing any actual science there.
| What about the guy feeding the scientists or housing them,
| what about the government for providing roads and sewage?
| That's like putting Steve Jobs' name on all the patents
| that Apple produced.
| ant6n wrote:
| Did they? (Put his name in all of Apples patents?)
| bryanlarsen wrote:
| Steve Jobs has 1114 patents. Apple has 16240.
| ryandrake wrote:
| I've worked at at least one company where the managers
| would routinely name themselves as "inventors" on patent
| applications, where their subordinates were the ones
| actually doing the "inventing."
| zamfi wrote:
| "Securing the funding" is much closer to the work than
| "providing roads and sewage".
|
| In most sciences, to actually secure the funding, you
| need to argue for why the problem is important, why the
| team has a shot at solving it, and what possible
| approaches look promising. Then you need to actually
| advise the team in supporting the work.
| leoc wrote:
| You might as well give Donald Trump coauthorship, then.
| Actually this could be an excellent idea: if enough
| scientists kiss his behind in this way Trump might change
| his mind about supporting large cuts to science funding.
| dylan604 wrote:
| Trump has that "you want to be the last person to leave
| the room" type though, as that's as much as he cares to
| remember if even.
| Etheryte wrote:
| If that's all you do then your work is very valuable, but
| you have no business being a coauthor on any of the papers.
| The dean of the university also secures funding for the
| university, hopefully you can see that that doesn't make
| them a coauthor.
| kergonath wrote:
| Also, the guy who does the writing gets to be an author.
|
| I've seen many instances of the new post-doc finishing
| writing up a former PhD student's research after the student
| left and had no interest whatsoever in their former project.
| It's a thankless job, with all the pains of doing science
| without the actual good bits. And yet it is necessary,
| otherwise some good or valuable research would not get
| published.
| zipy124 wrote:
| You are correct, in a large number of fields the PI (principal
| investigator) goes at the end, the big boss of the lab who
| manages it and often handles funding and the likes. Similar to
| the C-suite in a company.
| irrational wrote:
| So, who is the worst? Is there someone pumping out multiple new
| papers per day?
| cgh wrote:
| From the article:
|
| "Similarly, paper mills - organizations that produce fabricated
| or low-quality manuscripts for profit - often rely on inflated
| co-authorship networks to boost the metrics of paying
| researchers..."
|
| The article references this paper on predatory journals as the
| source:
| https://link.springer.com/article/10.1186/s12916-015-0469-2?...
|
| According to this paper more than 75% of the authors published
| in these journals originate in India and Africa (mostly
| Nigeria).
| readyplayernull wrote:
| Among given population X% are suspicious and Y% keep an eye on
| statistic and their heads below the radar.
| recursivedoubts wrote:
| Academia is filled to the absolute brim with scams, news at 11.
|
| (I'm in academia)
| gjsman-1000 wrote:
| Yet they still demand we take them with utmost seriousness...
| matthewdgreen wrote:
| All human activity is full of scams if you cast your net wide
| enough. Insofar as academia contains scams, the distinction is
| that they have a much lower profit margin.
| cherryteastain wrote:
| > Sayre's law states, in a formulation quoted by Charles
| Philip Issawi: "In any dispute the intensity of feeling is
| inversely proportional to the value of the issues at stake."
| By way of corollary, it adds: "That is why academic politics
| are so bitter."
|
| https://en.wikipedia.org/wiki/Sayre%27s_law
| cwmoore wrote:
| How does one compute the profit margin of academia?
| drawkward wrote:
| Got a citation for that, academic?
| TrackerFF wrote:
| If you scrutinize the top publishers, I'm sure you'll find _some_
| duplicate publications, salami slicing, and other things.
|
| When you're that prolific, the number of publications also become
| a goal, especially if you're awarded for it.
|
| Maybe not outright academic fraud, but at least publishing
| pattern that tries to maximize some measure(s).
| YossarianFrPrez wrote:
| I was just thinking about the upper bound on how many papers one
| could publish in a year earlier this morning. Speaking as someone
| who spent years in Tech and is now in the middle of a Ph.D.
| program...
|
| Within the current paradigm, where a post-doc gets hired as a new
| professor and goes about starting the rough equivalent of a
| single private sector team, at least in my subfield, 15-25 (non-
| first author publications) a year is an impressive number. And
| thus the numbers cited in the supplemental materials, the max
| being 136 papers a year, is strange, and I am pretty sure the
| author's points about paper mills etc. hold true.
|
| This is the great thing about the "Contributor Roles Taxonomy"
| system: it provides a lower level of abstraction and gives credit
| for who did what (idea generation, coding, writing, reviewing,
| raising the money, etc.) compared with using "a publication" at
| the unit of measure. It really solves a lot of problems. [1]
|
| [1] https://authorservices.wiley.com/author-resources/Journal-
| Au...
|
| But I'd also like to raise another point. People who wind up in
| Academia tend to go straight from undergrad to grad school (or
| spend a year being a lab manager in academia) and so most if not
| all of the systems we use in software development aren't present.
| Code review, project timeline estimation, building up a lab-wide
| codebase of functions to speed up repetitive tasks, 360 degree
| reviews, lab-wide project management software, an org-chart
| deeper than two (or in rare cases, three) layers, teams with
| differentiated responsibilities multiple teams, etc. are not the
| norm. Every so often I hear of one lab here or there that does
| one or two of these, not all of them. (Though my experience is
| limited.)
|
| My point is that if one were to apply all of the modern systems
| used for coordinating groups of people to produce structured
| forms of writing, etc., then 100 papers a year _sans_ a breadth-
| vs.-depth tradeoff might just be doable. But note that this is
| not "100 papers as year" from an individual, it's "100 papers a
| year from a mid-sized institution." (Six teams of four getting
| out 1.5 papers a month equates to 108 papers a year, near the
| maximum cited above.) Bell Labs' publication / patent rate must
| have been high!
|
| Granted, what I'm saying is not exactly within the current
| paradigm of how science is done, and might not be possible in a
| university setting.
| Steuard wrote:
| It's quite possible that there's a real effect here. But while
| I've only had time to skim parts of the paper, I don't see any
| indication of whether the authors have accounted for the
| different norms in different fields when analyzing their data for
| potentially fraudulent or deceptive behavior.
|
| Just for example, physics papers produced by large international
| collaborations (e.g. every single paper from the Large Hadron
| Collider at CERN) routinely have hundreds of authors (e.g.
| https://www.nature.com/articles/nature.2015.17567): everyone who
| has made substantial contributions to the design and operation of
| the facility is listed, as is everyone on the data analysis
| teams. (My understanding is that people in those specific fields
| all recognize that "number of citations" is a mostly meaningless
| number for those involved, and other metrics for productivity are
| well-known in those communities and routinely used.) I hear that
| some genomics papers have broken 1000 authors as well.
|
| I could easily imagine that the high end of observed publication
| numbers and coauthor counts would be dominated by those giant
| collaborations, even though there is absolutely no attempt to
| mislead anyone in the process. Can anyone tell from this article
| to what degree its conclusions might be influenced by this
| factor?
| daft_pink wrote:
| We're going to penalize people for publishing too mucn now?
| Really?
|
| Just want to say that I love Issac Asimov and he wrote a lot of
| high quality work and people even say that he's a graphomaniac
| with a writing compulsion.
|
| I think people's work should be judged by the output quality not
| some sort of speed limiting judgment.
| kergonath wrote:
| > Just want to say that I love Issac Asimov and he wrote a lot
| of high quality work and people even say that he's a
| graphomaniac with a writing compulsion.
|
| There is a difference between being prolific and what those
| people put out. At least one order of magnitude. Isaac Asimov
| did not write 80 novels each year.
| Xelynega wrote:
| Isaac Asimov wroted/edited ~500 books over his life.
|
| Some of the people mentioned in the research are coauthoring
| ~200 paper a year.
|
| I don't think Isaac would be in the same league as those
| mentioned in this paper.
| userbinator wrote:
| I wonder how many of these papers are now mostly if not
| completely AI-generated.
|
| This happened a year ago, and only gained attention because it
| was bad enough to be noticed:
| https://news.ycombinator.com/item?id=39391034
| alberth wrote:
| Isn't this a well-known issue in the academic and research
| community where researchers include their friends' names on their
| published work?
|
| Since researchers are rated and measured by the number of
| published papers they have, many people game the system by
| exchanging bylines with their friends ... so that it boosts their
| total number of papers published.
| mrweasel wrote:
| This was one of the ethical dilemmas we had to go through at
| university some 20 years ago: "Your professor/supervisor tells
| you to add them to the list of authors on your paper, do you do
| it?". There's a whole subset of "issues" here, one where you're
| more likely to be published with the supervisors name on the
| paper, another where you're travelling to present your paper,
| but you're not likely to be granted the budget and adding a
| professors name will almost ensure an all expenses paid trip.
|
| Low hundreds papers doesn't seem impossible, nor does hundreds
| of collaborators, but it would heavily depend on your work. We
| see the same in software. Some people seems unrealistically
| productive, spawning one successful project after another. I
| would be interested in knowing for how long they can keep up
| the pace though. There's also the question of the quality of
| their work.
| thinkingQueen wrote:
| My eyes were opened when a field called gamification appeared in
| the early 2010s. In a few years many gamification researchers had
| tens of thousands of citations, h-indexes nearing hundred. Well,
| if you think about it they're gamers, they've been grinding their
| RPG characters, sniping skills and whatnot for thousands of
| hours. It's only natural that these guys and girls figure out how
| to reach the maximum scientific high score.
|
| Some of the gamification researchers are near the top 500 of that
| 2% list. Now ask yourself, is gamification something that should
| make you one of the top 500 scientist in the world? I doubt it,
| but modern science is a citation game. Nothing else.
| hansvm wrote:
| That feels like a certain xkcd [0]. If you can game the totally
| bullshit rankings then clearly your papers are worth something.
|
| [0] https://xkcd.com/2385/
| godelski wrote:
| It amazes me how many "high profile" researchers have blatantly
| fraudulent work.
|
| As an example, take this rather recent ICLR submission that is
| egregiously plagiarized[0] (I mean attempts to hide the
| plagiarism are minimal to nonexistent). Clicking on the authors
| you can view their profiles and you'll find that most authors
| have several thousand citations and one north of 25k[1]!!!! They
| also have an i-10 of 400, which is highly abnormal for that
| citation level while the last author has an i-10 of 126 with only
| 7.3k citations.
|
| The problem I see here is that there's lots of suspicious
| activity and clear evidence of fraud, yet little to no action is
| taken. If anything, it seems we mostly ignore it. The ICLR
| submission is a bit odd too, since decisions and comments are
| public, most of the time this happens behind closed doors and we
| don't observe the evidence. As far as I'm aware, no one is making
| network graphs to track fraudulent authors. I think we want to
| sweep it under the rug because at the end of the day our system
| relies heavily upon trust. Unless we strongly incentivize
| reproduction we require high trust and we're afraid to
| acknowledge fraud in the community because we believe it will
| make more distrust science. But I think in reality, the
| reluctance to pursue fraud and the reluctance to have high
| clarity creates more distrust. The truth is that this system
| poorly scales even before we incorporate the reverse
| (bureaucratic) incentives.
|
| I think we lost sight of the goals of research and publishing.
| Publishing exists to facilitate communication among our peers. We
| publish in venues for two reasons: to improve/error check works
| and to facilitate social networks. Publish or perish is
| detrimental as well as the push for highly subjective measures
| like novelty. I strongly advocate for returning to the old
| paradigm, where the conditions for publication are: are there any
| major errors, is there plagiarism, and does the work sufficiently
| provide evidence for the experimental hypothesis (note: I
| intentionally do not say "prove".). This system accepts the
| reality that we cannot sufficiently judge significance.
| Breakthroughs by definition come from research areas that are
| under studied. Frequently from directions that had previously
| been rejected. Unfortunately, this is becoming more common due to
| the speed in which researchers must publish and exacerbated by
| the rise in thresholds to get a work published. Many ideas are
| abandoned not due to evidence that they will not yield progress,
| but that the belief to achieve "sufficient" evidence would
| require too much time. While tenure should "resolve" this, it
| ignores the fact that a tenured professor still has students that
| must publish as well as they've been working in a very different
| way for years. And old habits are hard to break.
|
| I think we need to take a hard long look at our system and I
| suspect we could do so much more if we rebuilt the structure.
|
| [0] https://openreview.net/forum?id=cIKQp84vqN
|
| [1] https://scholar.google.com.hk/citations?user=D-TS1fAAAAAJ
| drawkward wrote:
| Happens in patents too. At one of my gigs, there was an
| engineering department that was perpetually being granted patents
| for their work. The C-level exec that the department laddered up
| to demanded on being on every patent app, despite having nothing
| to do with the invention itself.
|
| My boss insisted that we include other team members on our patent
| apps, even if they were not involved, as a way of improving
| morale. (It backfired. No one on the team who eas uninvolved
| wanted a patent they didnt earn. No one who earned the rigjt to
| be named on the patent app felt it was right to include non-
| inventors.)
___________________________________________________________________
(page generated 2025-02-18 23:00 UTC)