From: Ari Belenkiy <[log in to unmask]>
Date: Thu, 17 Oct 2013 18:18:58 -0700
I made an attempt to find correlation between different columns in the
Excel file provided on the website. The most promising appeared to be
tracing a correlation between the journals, which rejected or accepted the
article, with their appearance at Beall's list - but after "substantial
review".
Indeed, "substantial review" means that a referee did his job properly and
identified the supposed error. Then it is up to editor to say the Final
word. Only this case may point to a "predatory" publisher who must appear
at Beall's list.
Let me immediately report that this correlation appeared to be
inconclusive, to say the least.
Here is a table for this article:
Beall DOAJ
Accepted 9 8
Rejected 6 12
(There are two "mixed" cases where journals are listed as both Beall and
DOAJ but rejected the paper; they are not included here).
The most interesting cases are where after *substantial* review we see
"rejected and Beall" (false positive case) or "accepted and DOAJ" (false
negative).
There are two different ways to proceed.
1. "Rejection" means "innocent" journal (H0 null-hypothesis) while
belonging to Beall's list is a "diagnose" (false positive) while
"acceptance and not-Beall" is false negative.
Then both errors are very high, 33% each.
2. The identity "Beall = predator" is established H0 null hypothesis. Then
rejection of the article by "Beall" journal is Type I error and acceptance
by DOAJ journal is type II error.
Then both errors are very high, 40% each.
Overall, judging from this case being "Beall" does not give much confidence
in the outcome. Nor one can guess "predatory" ("Beall") journal from this
case alone.
Let me list several journals that represent the false positive case
("Beall" but rejected):
.........................................
Advances in Cancer: Research & Treatment (?)
International Journal of Medical Biology (EGY, BEL)
Emerging Issues in Medical Diagnosis and Treatment (CHN)
Journal of Solid Tumors (CAN)
World Journal of Radiology (CHN)
Pharmaceutical Technology & Drug Research (IND)
.......................................................
Is it the case to reconsider their "Beall" status?
Ari Belenkiy
Statistics Department
SFU
Canada
On Thu, Oct 17, 2013 at 3:00 PM, LIBLICENSE <[log in to unmask]> wrote:
> From: "Beall, Jeffrey" <[log in to unmask]>
> Date: Wed, 16 Oct 2013 19:08:59 -0600
>
> Kevin's right; excellent research can indeed appear in predatory
> journals; I have observed this. Predatory publishers do their best to
> appear legitimate, and they do everything they can to fool honest
> researchers into submitting papers to their journals. Sometimes they
> are successful, and a good researcher submits a novel and interesting
> paper to them, which they accept and publish.
>
> Predatory publishers don't discriminate; they want bad papers and good
> ones, as long as they can collect the APC.
>
> Jeffrey Beall, MA, MSLS, Associate Professor
> Scholarly Initiatives Librarian
> Auraria Library
> University of Colorado Denver
> Denver, Colo. 80204 USA
> [log in to unmask]
>
> -----Original Message-----
>
> From: Kevin Smith <[log in to unmask]>
> Date: Wed, 16 Oct 2013 12:40:49 +0000
>
> Surely it is a massive and unjustified leap to go from saying that a
> journal accepted Bohannon's bogus paper to calling that entire journal
> bogus or suggesting that none of the contents of any of these journals
> could have value.
>
> I have been looking at and considering the Journal of Natural
> Pharmaceuticals. A quick look at PubMed and PubMed Central suggest
> that neither index includes that journal, which was the one Bohannon
> focused on in the beginning of his expose. But as I point out in this
> blog post --
> http://blogs.library.duke.edu/scholcomm/2013/10/10/the-big-picture-about-peer-review/
> -- two major indexes for general academic research, one from Gale and
> one from EBSCO, do include it. And we simply have no basis for
> concluding that every article published in that journal is compromised
> by the apparent fact that Bohannon's article was accepted. The web
> site Retraction Watch lists a recent retraction of an article that was
> published in Stem Cells and Development, a journal published by Mary
> Ann Liebert which I am confident is indexed in PubMed and PubMed
> Central. But surely not every article published in that journal is
> tainted by the one retraction?
>
> I am fascinated by some of the cultural assumptions at work in this
> discussion. When the open access community gathered in Stellenbosch
> last year for the Berlin 10 Conference, one of the themes we heard
> repeatedly was that research done in Africa by Africans about African
> issues was unavailable to the people of Africa because it was
> published in Western/Northern journals that were unaffordable for
> African universities. The new business models of open access offer
> opportunities to resolve that problem, but they clearly need to
> develop and work out their problems, just as subscription-based
> journal publishing did several centuries ago. But instead we see
> carefully orchestrated and "cooked" sting operations like Bohannon's
> (who pretended that his article was written by an African) designed to
> undermine those journals before they can get well-established. It is
> ironic that Bohannon controlled for the possibility that his "native
> English" might give the game away (what an assumption!) but not for
> the possibility that subscription-based journals in the developed
> world might also have accepted his paper.
>
> There is an interesting discussion to be had about what exactly
> peer-review can really tell us and how we might resolve the bias in
> current academic publishing for well-capitalized operations in the
> developed world, with their apparent desire to slay all challengers to
> their dominance. There is lots to say. But one thing we cannot say
> is that Bohannon's journalistic sting operation has shown that all of
> the research published in all the journals he targets is bogus.
>
> Kevin L. Smith, M.L.S., J.D.
> Director, Copyright and Scholarly Communication Duke University
> Libraries Durham, NC 27708 [log in to unmask]
>