"Sting" catches journals accepting spoof research paper

Discussion in 'Off-Topic Discussions' started by Kizmet, Oct 4, 2013.

Loading...
  1. Kizmet

    Kizmet Moderator

  2. SteveFoerster

    SteveFoerster Resident Gadfly Staff Member

    On the one hand, exposing fraudulent journals is a good thing, so good for John Bohannon for doing so.

    However, the movement toward open access to knowledge is overall an extremely healthy one, which is why it's a very big deal to people who actually know what they're talking about, most notably the Association of Research Libraries. I hope that people who only give this incident a cursory glance don't come away thinking that access to knowledge is somehow bad and that paywalls somehow ensure quality. I'm concerned that because of this we'll see the same sort of mindlessness as those people who say, "Oh, those online schools are all diploma mills."

    I also wonder whether the bad journals were listed in Cabell's. None of the articles I can find seem to say.
     
    heirophant likes this.
  3. I agree. The flaw in this experiment is that it is combative, not inquisitive. Instead of targeting the full spectrum of science publishing, Bohannon chose a narrower target without doing the necessary science to establish that the focus on open-access journals was correct.

    While Bohannon may assert that the result was not a clear indictment of open-access journals, it is clear that was the intent.
     
  4. 03310151

    03310151 Active Member

    It happens all the time I think and not just to journalists. And, sometimes it's not even a "sting".

    This was a big one in Social Science circles a while back:

    DutchNews.nl - Report into fraudulent researcher damns lack of critical culture

    A report into how a Dutch university professor was able to fake research data for years blames the absence of a critical scientific culture at academic institutions.
     
  5. Kizmet

    Kizmet Moderator

  6. FTFaculty

    FTFaculty Well-Known Member

    Where are those journals that will publish your dreck without vetting it when I need them? I have a chair who wants pubs from me--right now. Maybe I need to check out that list, I have some old dead end junk sitting in my documents file.
     
  7. Stanislav

    Stanislav Well-Known Member

    Jeffrey Beall maintained a list, but took it down for some reason.

    In fact, Eastern bloc universities have strict numerical metrics for publications, and an industry of low-quality journals sprung up. Same in India and Pakistan, so I've heard, but I'm not as familiar with these. Ukrainian journal "Actual Problems of Economics" publishes a lot of stuff and has a reputation of being easy; it got prominence by somehow getting into ISI and Scopus indexes, but since lost them (I believe for excessive self-citation). Accepts papers in Ukrainian, Russian and English. Publisher is "National Academy of Management", an accredited private university. http://eco-science.net/index.php?action_skin_change=yes&skin_name=APE_eng
     
  8. Stanislav

    Stanislav Well-Known Member

  9. decimon

    decimon Well-Known Member


    Someone should submit a paper with an English heading and an Urdu body to see if that's noticed. For that matter, the same should be done in the US and Canada.
     
  10. Stanislav

    Stanislav Well-Known Member

    THAT they may notice. Randomly-generated English text, maybe not.

    There's already a perfect sting paper, produced by some frustrated computer scientist: www.scs.stanford.edu/~dm/home/papers/remove.pdf
     
  11. decimon

    decimon Well-Known Member

  12. Helpful2013

    Helpful2013 Active Member

    Steve makes a fine point that open access to knowledge is a good thing, but the reality is far from ideal (see particularly #2 below). My comments below relate to the humanities where I work, and where journal articles definitely aren’t fungible commodities. Note that I’m also not addressing the predatory fake journals, but genuine efforts. The problem is that a quality journal requires lots of costly labour to produce, and those expenses must be met somehow.

    Model #1: The costs are borne by subscribers and so the articles are available to those subscribers, both individuals and university libraries, with others having to pay for article access. Most of the ‘learned societies’ that publish journals follow this model. Some of these maintain paid professional staff in addition to a volunteer overseeing editor that is a scholar in the field.

    Model #2: The costs are borne by the researchers, who pay to publish in the journal. The journal is accessible to the public, but the fee arguably compromises the principle of requiring double-blind peer review before acceptance. Since the fees run into the thousands of dollars, in practice it restricts publication to those working for a large institution that pays the publication fees. Some like the idea of soaking the universities and research grantors, but the matter of paying to publish makes many readers wary of the published results. Most of the ‘open-access’ journals published by corporate publishing houses follow this model.

    Model #3: The journal is electronic and assembled entirely by volunteers with no overhead, and therefore no access costs to pass on to readers. While this model is truly open-access and sounds attractive in principle, there is the matter of credibility. A journal builds a reputation for credibility when it has made contributions for many decades, is produced by a scholarly association, and edited by proven scholars. It gets read, and the readers understand that there is a high standard for what gets published in it. This becomes self-selecting and self-perpetuating, as scholars will seek to publish there, because doing so validates their work. In the end, there’s a tremendous amount of inertia that the new open-access journals following model #3 have to overcome, and in some cases, they just won’t have the resources to build a credible alternative.
     
  13. heirophant

    heirophant Well-Known Member

    I emphatically agree. It's important to put scholarly work product out there where people can see it, instead of hiding it behind pay-walls where it's invisible to people who don't have access to institutional subscriptions or are willing to pay maybe $30/article for full-text.

    Personally, I like PLOS and think that its 10 or so journals (mostly in the biomedical sciences) are entirely credible.

    https://www.plos.org/which-journal-is-right-for-me

    https://en.wikipedia.org/wiki/PLOS
     
    Last edited: Feb 14, 2018
  14. heirophant

    heirophant Well-Known Member

    The controversy about "open access journals" seems to me to revolve around a fundamental ambiguity in what 'open access' is taken to mean.

    1) Journals that aren't hidden behind pay walls and don't charge readers to read them. Hence, open access for readers.

    2) Journals that accept any and all submissions for publication. Hence open access for authors.

    I'm not convinced that 1) and 2) necessarily describe the same set of journals. Nor am I convinced that the same criticisms are applicable to both of them. (My PLOS example seems to illustrate that.)

    There probably is a lot of overlap between 1) and 2) though:

    Observation A: It's probably more common for journals with no quality control to be open access to readers as well, since that kind of journal is unlikely to attract a lot of paying subscribers.

    The Directory of Open Access Journals catalogs upwards of 11,000 journals. A cursory glance at their 'search' function for journal titles suggests that most of them are in what some might call 'third world' countries and their language often isn't English. (So the always-outraged SJW's among us might be able to construct an argument about how opposition to open access publishing is really a plot by Western neo-aristocratic academic elites to preserve their own control of scholarly thought.)

    https://doaj.org/

    DOAJ's definition of 'open access' is squarely in line with my 1). But... their quality control requirement seems to me to be pretty minimal, merely that the journals has more than one editor. So I'm guessing that many of their 11,000 also satisfy my 2). So many of them probably conform to my Observation A.

    https://doaj.org/faq#definition

    But that being said, DOAJ does weed its list pretty aggressively.

    https://docs.google.com/spreadsheets/d/183mRBRqs2jOyP0qZWXN8dUd02D4vL0Mov_kgYF8HORM/edit#gid=1650882189
     
  15. Kizmet

    Kizmet Moderator

    I'm curious about whether any of these journals charge the author for publishing their work. This might be similar to a novelist paying to self-publish a book. Does this also happen?
     
  16. Stanislav

    Stanislav Well-Known Member

    I don't think anyone uses the term "open access" in the second sense. A publication platform with no review is a preprint repository not a journal. Good preprint repositories, like arXiv or SSRN, are excellent things.
    The ostensible problem with open access in the sense #1 is how it tips the incentive structure towards lowering the quality: many journals charge Article Processing Fees on acceptance, and therefore a rejected article is lost revenue. Responsible publishers know enough to build a structure to resist this pressure and preserve credibility. However, OA makes it easy to profit in an unethical way, by accepting all comers while claiming to run peer review. It's a form of academic fraud, not unlike the degree mill industry: selling unearned credentials, in this case those of a "published author". Of course, this is driven by demand and did not start with OA movement (vanity presses exist forever, and so do questionable conferences); OA just makes barriers of entry into this game ridiculously low. Just look at some of the entries on Beall's list.
    I'd say this does not condemn the whole OA movement; Jeffrey Beall kind of jumped the shark when he became increasingly anti-OA, rather than exposing frauds. Example is his feud with MDPI, which appears to be a low-ranked, for profit publisher but not a "predator".
     
  17. Stanislav

    Stanislav Well-Known Member

    Yes, and yes. Both legit OA journals and frauds charge Article Processing Charges (APCs). Some OA journals don't; these can be of differing quality but are not considered frauds. Well, most, but not exactly all: I do have one publication (with my advisor) ran by a publisher Beall condemned, and I didn't pay them. They make their money by running a prolific number of conferences.
    A novel publisher along the same lines is a "vanity press". These things exist.
     
    Last edited: Feb 14, 2018
  18. Stanislav

    Stanislav Well-Known Member

    I should say so. PLoS is basically the golden standard of "Gold AI"; no one says they are unethical. The link above is dead, but I believe the sting authors did send their article to PLoS ONE ("light peer review" "megajournal" of the PLoS stable, which is I believe the biggest science journal by number of articles in the world). They note that PLoS ONE not just rejected it; the reviewers also caught a glaring ethical issue in fake experiment design everyone else missed. And this is a "peer review lite" journal; apparently titles like PLoS Medicine are outright prestigious.
     
  19. decimon

    decimon Well-Known Member

Share This Page