Fake journals

[sylas]

An amusing cautionary take has come to light.

Anna O. Szust is a young polish researcher, starting out her academic career, and she has made a great start. She has been accepted as an editor in some 40 journals, and indeed became editor in chief at four journals. All this despite having no publications herself indexed in the major academic databases such as Web Of Science, no publications listed in any academic journals, and no citations to her work from any other publications. She had no prior experience as an editor or even a reviewer.

Her C.V. does include various publications, but none in academic journals; and some book chapters. And even those publications listed were actually fakes.... as was her degree.

And her name, in Polish, is unfortunate... Anna O. Szust. "Oszust" is the Polish word for fraud.

Anna is, in fact, an entirely fictitious individual, created specifically for an investigation of predatory "scientific" journals. The results were written up in nature. See: Predatory Journals Recruit Fake Editor. (Nature, v 543, no 7646, 2 March 2017, by P. Sorokowski, E. Kulczycki, A. Sorokowska, and K. Pisanski)

There have also been a couple of reports of this in the regular news. For example, in the Sydney Morning Herald: Scientists outwit predatory publishers by tricking them into appointing a fake editor. Or read an interview with one of the authors at the blog "Retraction Watch" interview with Katarzyna Pisanski. (Retraction Watch tracks and reports on retractions in scientific literature.)

Reading the links will give a bit more background to the state of play in which predatory journals exist. When I was working as an academic (long ago) I often received invitations myself to be an editor at some obscure journal which I generally binned without further ado.

I'm sure other TWeb members working academia can say the same.

Cheers -- sylas

[DesertBerean]

C.v. ?

[rogue06]

C.v. ?

Curriculum vitae

[rogue06]

An amusing cautionary take has come to light.

Anna O. Szust is a young polish researcher, starting out her academic career, and she has made a great start. She has been accepted as an editor in some 40 journals, and indeed became editor in chief at four journals. All this despite having no publications herself indexed in the major academic databases such as Web Of Science, no publications listed in any academic journals, and no citations to her work from any other publications. She had no prior experience as an editor or even a reviewer.

Her C.V. does include various publications, but none in academic journals; and some book chapters. And even those publications listed were actually fakes.... as was her degree.

And her name, in Polish, is unfortunate... Anna O. Szust. "Oszust" is the Polish word for fraud.

Anna is, in fact, an entirely fictitious individual, created specifically for an investigation of predatory "scientific" journals. The results were written up in nature. See: Predatory Journals Recruit Fake Editor. (Nature, v 543, no 7646, 2 March 2017, by P. Sorokowski, E. Kulczycki, A. Sorokowska, and K. Pisanski)

There have also been a couple of reports of this in the regular news. For example, in the Sydney Morning Herald: Scientists outwit predatory publishers by tricking them into appointing a fake editor. Or read an interview with one of the authors at the blog "Retraction Watch" interview with Katarzyna Pisanski. (Retraction Watch tracks and reports on retractions in scientific literature.)

Reading the links will give a bit more background to the state of play in which predatory journals exist. When I was working as an academic (long ago) I often received invitations myself to be an editor at some obscure journal which I generally binned without further ado.

I'm sure other TWeb members working academia can say the same.

Cheers -- sylas

Roughly a year ago I posted something that seems pertinent to this that I'm going to re-post here

International Archives of Medicine for 600 Euros. It was picked up by multiple news outlets often getting front page coverage.

The entire affair can be read about here: I Fooled Millions Into Thinking Chocolate Helps Weight Loss. Here's How as well as here: How the "chocolate diet" hoax fooled millions.

The host of the aforementioned TV show, Adam Conover, had his own fraudulent study "The Possible Irritating Effects of Nutritional Facts" published in a faux journal called Advances In Nutrition And Food Technology to confirm that this does indeed happen. It was pretty obvious that the publisher never read it for it is a blatant spoof.

The problem seems to be especially bad in medical research journals.

[sylas]

Thanks for that one rogue! I hadn't seen it before.

There have been quite a number of spoof articles and the like over the years, as a kind of tweak the nose of the fake journal exercise. Another famous one is the Sokal hoax article.

This one is a bit different. It was done as a research exercise. The authors have deliberately not named any of the journals that accepted the fake editor (as fake journals often use names that sound a bit like some existing legitimate journal and they did not want to spread confusion). All journals accepting the Schust as an editor were later informed of the study and invited to respond. But you can google for Szust and still see her listed as being on the editorial board of several journals.

The article states:


Results can be summarized in the following figure:
nature-editorial-sting-23-mar-17-online.jpg

[sylas]

Further followon.

Rogue, your example of the study showing benefits of chocolate is really interesting. It's a whole 'nother kettle of fish. It wasn't a hoax article so much as an article using deliberately shoddy science with a real study and real investigation... all geared in advance to give a dubious result of benefits for chocolate. It's part of the whole debate about using "p values" as a measure of significance for some result. We've had some threads on that also!

[Sparko]

You can even generate fake papers.

Amphibious Epistemologies
Rogue06 and Sparko
Abstract
In recent years, much research has been devoted to the emulation of the transistor; on the other hand, few have analyzed the construction of lambda calculus. In fact, few cyberinformaticians would disagree with the analysis of von Neumann machines. BITO, our new approach for event-driven modalities, is the solution to all of these challenges.
Table of Contents
1 Introduction

Authenticated technology and courseware have garnered tremendous interest from both cyberinformaticians and futurists in the last several years. An essential issue in cryptography is the study of A* search. Further, a typical question in symbiotic e-voting technology is the understanding of certifiable modalities. Nevertheless, the partition table alone can fulfill the need for "smart" models.

Motivated by these observations, virtual communication and randomized algorithms have been extensively enabled by computational biologists. For example, many applications provide amphibious information. For example, many methodologies prevent distributed communication [7]. The shortcoming of this type of solution, however, is that the foremost relational algorithm for the simulation of kernels by Raman et al. is recursively enumerable. This combination of properties has not yet been analyzed in related work.

To our knowledge, our work here marks the first framework harnessed specifically for the construction of the Turing machine. Two properties make this method optimal: our methodology is not able to be improved to allow introspective models, and also our solution is in Co-NP. The basic tenet of this approach is the improvement of checksums. Thus, BITO is copied from the principles of algorithms.

In this position paper, we concentrate our efforts on showing that the partition table and superpages are usually incompatible. Nevertheless, the development of extreme programming might not be the panacea that scholars expected. Predictably, our heuristic manages the refinement of forward-error correction. As a result, we demonstrate that the foremost highly-available algorithm for the synthesis of I/O automata runs in Ω(n) time.

The rest of this paper is organized as follows. We motivate the need for SCSI disks. Further, to overcome this question, we verify that the Ethernet and superpages can interfere to surmount this obstacle. Our mission here is to set the record straight. We validate the analysis of I/O automata. This result might seem counterintuitive but has ample historical precedence. In the end, we conclude.
http://scigen.csail.mit.edu/scicache...o.Rogue06.html

from https://pdos.csail.mit.edu/archive/scigen/

[sylas]

There's a very interesting followup to that, Sparko. The program you used "SCIgen", has since been reverse engineered with a program to detect papers written by SCIgen. Investigators have discovered a surprising number of nonsense papers written by SCIgen which have been accepted for publication at conferences.

It is reported in Nature. See Publishers withdraw more than 120 gibberish papers, as a news brief in Nature 24 Feb 2014, by Richard Van Noorden.

Cheers -- sylas

[Raphael]

I follow Derek Lowe's pharma industry blog In The Pipeline (his "things I won't work with" posts are brilliant) and he has been talking a lot about this topic, both fake journals and faked research/results and/or shoddy reseach getting through and how they are getting more problematic (including the case in the OP)

http://blogs.sciencemag.org/pipeline.../the-dark-side

[DesertBerean]

Wow.

[Sparko]

There's a very interesting followup to that, Sparko. The program you used "SCIgen", has since been reverse engineered with a program to detect papers written by SCIgen. Investigators have discovered a surprising number of nonsense papers written by SCIgen which have been accepted for publication at conferences.

It is reported in Nature. See Publishers withdraw more than 120 gibberish papers, as a news brief in Nature 24 Feb 2014, by Richard Van Noorden.

Cheers -- sylas

That proves that they were not even read. Even a cursory reading of a scigen paper would tip off even a layman. Unless you are Shunya. I fooled him in a post with one.

[sylas]

That proves that they were not even read. Even a cursory reading of a scigen paper would tip off even a layman.

Quite so. The point of the reverse engineering was not to tell if it was gibberish or not.... but to be able to do an automated search, and also identify the use of a specific tool for producing gibberish. SCIgen gibberish has some specific features that allow a confident identification of tool used.

[Sparko]

Quite so. The point of the reverse engineering was not to tell if it was gibberish or not.... but to be able to do an automated search, and also identify the use of a specific tool for producing gibberish. SCIgen gibberish has some specific features that allow a confident identification of tool used.

so much for "peer reviewed" journals.

[sylas]

so much for "peer reviewed" journals.

Wait a minute! There are certainly some journals (and more often, conferences) which claim to have peer review but pretty obviously have no review at all. That would be the case to publish a SCIgen paper, I think.

I presume and hope that you are not going over the extreme of saying that there's no such thing as a peer reviewed journal!

The predatory journals phenomenon seems exists in part because there's a lot of pressure to get published, but that it's hard work to get into the professional peer reviewed literature, so there's a niche for the fake journals and conferences. Doesn't mean that ALL journals are fake, or that any claim to peer review is meaningless.

Cheers -- sylas

[Sparko]

Wait a minute! There are certainly some journals (and more often, conferences) which claim to have peer review but pretty obviously have no review at all. That would be the case to publish a SCIgen paper, I think.

I presume and hope that you are not going over the extreme of saying that there's no such thing as a peer reviewed journal!

of course not. I am talking about these phony "peer reviewed" journals. They claim they are peer reviewed but in fact they just publish anything. Probably charge a fee too.