The authors of a paper taking a major database to task for including papers from allegedly predatory journals are objecting to the retraction of the article, which followed a request by one of the publishers mentioned in the analysis.
And at least one of the journal’s editorial board members is considering resigning over the move.
The paper, “Predatory publishing in Scopus: evidence on cross-country differences,” was published in Scientometrics, a Springer Nature journal, on February 7. It used Jeffrey Beall’s now-defunct list of allegedly predatory publishers to identify relevant journals. The next day, the study’s findings were the subject of a news story in Nature.
On May 6, Fred Fenter, chief executive editor of Frontiers, a publisher which figured in the analysis, sent Scientometrics editor Wolfgang Glänzel a letter, obtained by Retraction Watch, demanding that the paper be retracted immediately. Much of the letter is a critique of Beall’s list, which has certainly come under fire before. Fenter — whose criticisms of of the list prompted an investigation by Beall’s university, after which Beall eventually retired — writes:
This “data source” is biased, unreliable, unvalidated, and unavailable – and thoroughly unsound for a scientific article. For these, and additional reasons below, this article must be immediately retracted.
The authors, Vít Macháček and Martin Srholec, of Charles University and the Economics Institute of the Czech Academy of Sciences, responded in a May 19 letter, writing, in part:
Mr. Fenter’s letter contains a litany of allegations and insults. His core claim and the only substantial point that in our view deserves consideration is that Beall’s lists (Beall 2016) are not legitimate sources of data for scientific inquiry. In this response, therefore, we focus on this issue.
Beall’s lists are not without limitations, as has been acknowledged in our paper. However, this data source has been established as a relevant input in scientific literature and widely used in empirical research on predatory publishing. In this regard, we follow the suit of a long line of papers published in respected peer-reviewed scientific journals that used Beall’s lists in essentially the same way to identify what has become referred to as potential, possible, or probable predatory journals – see the list of examples below, including papers in Science, Research Policy, Journal of Informetric and three other earlier papers in Scientometrics. We are not using some new, fringe and untested dataset for the first time. In fact, it is fair to say that Beall’s lists have become the dominant source of data in this line of research. It is hard to understand why using this data should suddenly become a reason for retracting a published paper.
Glänzel sent the paper for post-publication review, and provided the authors with referees’ comments on July 2. On July 12, the authors responded to each criticism. They also wrote:
The post-publication peer-reviewers seem to be strongly critical about the use of Beall’s lists and hence also inevitably about the evidence presented in our paper – one can even say that their comments are unusually aggressive. But this is not representative to the overall stance of the scientific community to this matter. It would be fair to consult reviewers who are ex-ante neutral, who do not have predetermined opinions in this regard, who in particular did not publish papers critical of Beall’s lists by themselves in the past, and who could therefore provide assessment of the content and relevance of our paper that is representative and free of emotions.
Let us also point out that at least some of the comments seem to be put forward by a researcher from the field of natural sciences, who is used to deal with controlled experiments but who might not be familiar with how regression analysis is used in social sciences. We hope that this would not lead to a confusion.
On August 17, Glänzel told the authors that the paper would be retracted, and asked whether they agreed. The retraction appeared yesterday (September 6):
The Editor-in-Chief has retracted this article (Macháček & Srholec, 2021) because some of the findings are unreliable. Post-publication peer review indicated the article includes statements about authors from some geographic regions which are unjustified in the generality of the conclusions. Findings are based on a regression analysis; however, this analysis did not include a control group. The regression analysis is, therefore, not complete and the results are unreliable.
Results and findings are based on a so-called blacklist and are not supplemented by any results obtained using a positive control group. In this context, the Scopus database cannot be considered a control group, since it is a comprehensive bibliographic database. Similarly, the analysis was restricted to publications in four languages (English, Spanish, French, and Arabic). The results of the regression analysis are inconclusive for publications in languages not included in the analysis. The authors have been offered the opportunity to submit a new reworked manuscript for peer review. Vít Macháček and Martin Srholec disagree with this retraction.
Srholec confirmed to Retraction Watch that he and Macháček object to the retraction:
Both of the authors strongly disagree with the retraction, because there is no credible academic justification for it. In my view, the retraction is based on false accusations and biased post-publication peer review. It is political and likely driven by business interests. It is deeply disappointing and if fact very hard to believe for me that this could happen in such respected journal.
He added:
Scientometrics has seriously mishandled this retraction. In fact, in my view, this is a rare case of a serious academic misconduct of the journal editor. At the very least, he has turned out to be bluntly incompetent.
Editorial board member Ismael Rafols told Retraction Watch:
I believe that this retraction is problematic in several respects and I can confirm that I am considering resigning because of it — as I expressed to the editor-in-chief and the board during the debate.
According to COPE guidelines, a retraction should only happen if it is shown that the potential errors make the findings unreliable. Otherwise, ‘If only a small part of an article reports flawed data or content, this may be best rectified by a correction.’ Guideline also says that ‘Retraction notices should mention the reasons and basis for the retraction to enable readers to understand why the article is unreliable and should also specify who is retracting the article’. Thus, the authors (as well as Scientometrics readers) should be provided to evidence that the potential flaws render the findings unreliable. I don’t believe this has been the case.In short, I (as well as some colleagues) cannot understand why this case should be a retraction rather than a correction.
Given that the debate on journal publishing (in particular around questionable commercial behaviours) is particularly controversial and stirred by particular interests, I believe that scientific debate should be dealt openly, rather through retractions.
Neither Glänzel nor Fenter immediately responded to requests for comment. We will update this post with anything we learn.
Update, 9/9/21, 1215 UTC: Scientometrics editorial board member Cassidy Sugimoto, who is also president of the International Society for Scientometrics and Informetrics (ISSI), forwarded us a statement she sent to other board members:
Every actor in the scientific system has a role to play. Scholars generate new knowledge and peer reviewers evaluate it. That these individuals are all scholars is critical to the self-organizing nature of scholarly communication. As a community, we have an obligation to uphold scientific principles. Commercial publishers have an incentive to protect their reputation so that they can further monetize scholarship. That they defend these interests in the way that we have observed is disquieting, but not unexpected. Editors, as gatekeepers, have a critical role to play, in defending scientific values against private interests.
I also have a role to play, as President of our main professional society. Although ISSI and Scientometrics bear no formal relationship, their identities are entangled in the public mind, owing to the overlap between leadership and activities (e.g., the celebration of the Derek de Solla Price at the ISSI meetings). The role of societies is to create community, generate shared values, and promote professionalism by asking its members to hold themselves to certain standards.
It is therefore with this lens that I join my colleagues in asking that the retraction be reconsidered. The overwhelming evidence in this case suggests that the interests of private stakeholders have overridden scientific standards. The decision to retract harms our community, violates our values, and undermines the professional integrity of our processes. I cannot, as President of ISSI, condone this behavior.
I acknowledge the burden that this places on the Editor-in-Chief and it is unfortunate that these commercial entities have threatened scientific integrity. However, it is our collective role to defend science and to take actions to sanction those who do not.
Like Retraction Watch? You can make a one-time tax-deductible contribution or a monthly tax-deductible donation to support our work, follow us on Twitter, like us on Facebook, add us to your RSS reader, or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at [email protected].
This is shocking, and sets a terrible precedent. Both the aggressive actions by Frontiers and the apparent easy surrender of Scientometrics are appalling.
Why did Wolfgang Glänzel (Scientometrics editor) go along with this? There should never be corporate interference with publication and editorial decisions. Publishers shouldn’t be able to retract papers with legal threats.
Notably, Springer owns a stake in Frontiers, although they rarely mention this publicly.
Also, based on the authors’ response letter, it’s very likely that Glänzel tapped self-plagiarist crackpot Jaime A. Teixeira da Silva to do post-publication peer review for the article. That is an indefensible editorial position (especially given the intellectual capital on the Scientometrics board), and strongly suggests that Glänzel was looking for someone to kill the article.
http://retractionwatch.com/2016/07/18/great-shock-and-sadness-whistleblower-to-retract-paper-for-duplication/
As a rule, an infamous gadfly like Silva is best ignored. However, there is the disconcerting and appalling possibility that Glänzel chose Silva for post-publication peer review.
In their response letter, Srholec and Macháček referred to a natural scientist reviewer who published numerous anti-Beall articles in the past. This suggests that they knew – or at least strongly suspected – that Silva was a reviewer.
Plus, the reviewer is eager to discuss an obscure Silva and Tsigaris (2018) article published in the Journal of Academic Librarianship, that pretty much only Silva has ever cited: https://scholar.google.com/scholar?cites=13061971235796606354&as_sdt=2005&sciodt=0,5&hl=en
Then there’s the familiar issues of tone, hostility and anti-Beall sentiments in the review that are also suggestive of Silva.
Once again, Silva is best ignored, but in this case, it looks like that for whatever reason, Glänzel chose to give him of all people influence over the publication fate of this article, which I think is appalling and disconcerting.
Spot on! Yes, I strongly suspect that this is the case.
It’s pretty unprofessional if it is. JATdS is a troll – the kind of troll who sends unsolicited complaints to unrelated third parties (e.g. the Elsevier letter from 2014) and has now pivoted to publishing on COVID, despite no qualifications in medicine or epidemiology (or even institutional affiliation). Editors who accept his manuscripts or call on him as a reviewer should be ashamed.
The study by Vít Macháček and Martin Srholec is misleading because it’s not a study on predatory journals, it’s a study on Beall’s list. The authors found that most of the journals included in Beall’s list are connected to certain geographies or languages. In other words, they showed that Beall’s list can appear to be racist. It’s definitely the publisher’s prerogative to avoid carrying racist publications («statements about authors from some geographic regions which are unjustified in the generality of the conclusions»), so the rationale for the retraction is sound. It’s then a matter of opinion whether it’s warranted.
While I think the retraction was horrible, using a loaded word such as “predatory,” especially in the title, was a serious mistake by the authors. You can talk about Beall’s list, and you can cite/quote Beall’s definition of predatory journals with respect to his list, without making the claim directly.
Time to #BoycottSpringer , perhaps?
They are neck deep in their own biowaste, recent story of “Arabian Journal of Geosciences” is just one example…
“…the analysis was restricted to publications in four languages (English, Spanish, French, and Arabic). The results of the regression analysis are inconclusive for publications in languages not included in the analysis.”
That’s 3 more languages than most papers would have considered. This retraction is disgraceful. Glänzel should step down in shame.
Those languages refer to the predominant languages in the researchers’ home countries identified by affiliation data. Other languages are not spoken in enough countries to be included in the analysis. They don’t refer to the languages in which the articles themselves were written. The authors sourced the journals from Beall’s list, which is English-only.
This is a further bias of the paper as a “home country” is not necessarily a valid reason to decide on a language.
You can be in a country but you publish in another language than the local one.
You miss the point. If publication other than in the local language is, at all, done, it’s usually in English.
While other combinations may exist, they are unlikely to be significant enough to lend themselves to statistical evaluation.
I’m confused as to why this is so controversial.
This study would never pass muster in the biological sciences. It seems to be openly based on a list that is (1) curated by someone with questionable motives (at best) (2) is not controlled or otherwise refutable or confirmable and (3) does not transparently share the methodology utilized to classify listed journals.
So can someone explain why is it ok for a list that is created by one highly controversial person to be used to sully the names of many other unsuspecting scientists and journals ? If that paper was published in a blog or tabloid , then that’s fine. But the authors want to publish work in a peer reviewed journal and claim veracity; therefore, unless they’re able to stand on their work’s merits and fully justify all their conclusions and methods, they should have no right to complain and the journal is completely justified in retracting their work which clearly doesn’t pass muster. The least you can do when you wade into a controversial topic is be able to defend your work vigorously, and that is clearly not the case here.
“This study would never pass muster in the biological sciences.”
it’s almost like it is social sciences’ research. go back to primary school for some reading comprehension training (and take Makrams with you)
It seems the Editor in Chief did not fully read the retracted article. As I wrote in another comment, the list of languages “English, French, Spanish, and Arabic” doesn’t refer to the working languages in which the articles in the journals under study were written (which is the reading according to the retraction notice). The journals under study are vastly dominated by English, and this limitation has been clearly stated as such in the article. These four languages refer to the languages of the population where the researchers’ affiliations are based. As a hypothetical example, if a group based in the US, China, and Argentina publishes a work in a journal in Beall’s list, the languages “English, Chinese, and Spanish” are identified for such an article. However, since Chinese is not spoken in enough countries to be useful for the analysis, it is removed, and the outcome becomes “English and Spanish”. These linguistic-demographic data are then used in the analysis.
Given the limitations of the study, one can debate the merit of this part about language & demographics. Nevertheless, the retraction notice seems to contain a profound misunderstanding that (at least in part) undermines the ground for retraction.
Exactly right!
By retracting this paper, does the very definition of “predatory” in question. We will never be able to determine which are the predatory journals? IT is a shame that the real once are getting a mileage now.
It would appear to me that, in these cases involving MDPI and Frontiers, it’s not relevant who’s ‘right’ or ‘wrong’, but rather who has the highest paid lawyers…
An astute RW reader noticed something similar may be going on with an article that dared use the characters “MDPI” and the name “Beall” within the same article: http://retractionwatch.com/2021/08/21/weekend-reads-fraud-in-a-study-of-dishonesty-scrutiny-of-an-open-access-publisher-mit-hhmi-prof-fired-for-sexual-harassment/#comment-1995153
Except the study at issue itself concludes that Frontier is unlike the other journals referred to as predatory.
Correct. But there is some likely more to the story, as I alluded. See the ACKNOWLEDGMENTS of the republished version, https://direct.mit.edu/qss/article/3/3/859/113621/Predatory-publishing-in-Scopus-Evidence-on-cross, specifically:
“Following pressure by Frontiers, the Editor-in-Chief of Scientometrics decided to retract the paper based on dubious claims that some of the findings are unreliable.”
It’s too bad Cabell’s has their curated blacklist behind such a tall paywall. My library declined to subscribe and the company doesn’t support any sort of pay-per-view or pay-per-query by individuals. as far as I can tell. I understand the need for paywalls – someone puts their money into producing something of value, and they need to cover their costs and make a profit. No can give it away for free. But as far as I can tell, it is not accessible in any way to the vast majority of scholars who would use it if they could.
A reason that this so-called list is highly questionable and unreliable with malicious goals behind.
What “institutional affiliations” have to do with the ability to do science and to think, analyse, write and evaluate papers?
If your brain does work outside “institutional affiliations”, those of others may work better outside than inside “oppressive institutions”, enjoying more freedom and critical thinking than you, who are unable to show your name by fear of your institution reprisal.
Neither the paper in question, nor the behavior of Frontiers are acceptable in my humble view.
Frontiers and MDPI are bad publishers that the goal is primarily money, not science.
This retraction shows one thing:
<>.
It was behind the removal of Beall’s list and now it is behind the retraction of this paper.
Look at the dictation tone of Frontiers that proves that Frontiers is now rich enough to pay lawyers to distort science and reprehend any scientist who dares to criticize its questionable behaviors. Regrettable.
The retracted paper itself is of course questionable and should not be published, as it is based on a list of a guy (Beall) who was having some conflicts of interest with open science journals related to library subscriptions issues and library finding.
“Frontiers and MDPI are bad publishers that the goal is primarily money, not science.”
You believe Elsevier and company are in it for charity and the greater good of mankind?
Sorry, but Frontiers had every reason to be upset that a list that smeared them was simply used as reliable data, even IF the study itself showed that they were not at all like the rest of the bunch. Doesn’t mean that dragging the list back into the limelight was something they should be happy about.
Neither the paper in question, nor the behavior of Frontiers are acceptable in my humble view. Frontiers and MDPI are bad publishers that the goal is primarily money, not science. This retraction shows one thing:
Frontiers is powerful enough to impose its rule and vision.
Frontiers was behind the removal of Beall’s list and now it is behind the retraction of this paper. Look at the dictation tone of Frontiers that proves that Frontiers is now rich enough to pay lawyers to distort science and reprehend any scientist who dares to criticize its questionable behaviors. Regrettable. The retracted paper itself is of course questionable and should not be published, as it is based on a list of a guy (Beall) who was having some conflicts of interest with open science journals related to library subscriptions issues and library finding.
What does the autocratic tone of Frontiers mean/hide:
(… This article must be immediately retracted.)
(… We request confirmation of the retraction of this article by Friday 21 May at the latest.)?!!!!!
It seems that Frontiers has no frontiers in their stupidity.
The money Frontiers win easily from publishing transform them into unscrupulous publishers where everything is allowed to earn as much money as possible.
Never seen that people should pay to publish their “scientific Opinions” as is the case with Frontiers that apply fees on publishing Opinions and Viewpoints!
The first to blame on Frontiers issues is Nature Publishing Group on which Frontiers has built its legacy when they started some years ago.
Jaime Teixeira da Silva has a sordid history. Here he is publishing in a predatory journal to attack Retraction Watch for questionable reasons.
https://www.mcser.org/journal/index.php/mjss/article/view/9568
Prof. Glanzel should answer whether he chose this very inappropriate person to conduct peer review duties and why.
“Single-minded” might be the better word. Dr Teixeira da Silva is true to his ideals of ethics in scientific publishing, and critical of anyone who does not share them (i.e. just about everyone).
Beall provided a key public service, whatever his political views (and from what I’ve heard of them, I don’t agree with them at all). It’s utterly shameful that he was forced to retire. Predatory journals are a real phenomenon – they spam me all the time looking for submissions. The scientific community would do better to look to their epistemological defenses against these predators instead of forming a circular firing squad.
Predatory journals are indeed a real phenomenon, that doesn’t mean that smearing any journal somebody doesn’t like as “predatory” should be tolerated.
Just because there are scientists who forge data doesn’t mean suggesting any result not to your liking is forged is in any way acceptable
Paid-open-access journals are the problem, not only predatory journals!
Open access is the nice name of paid predatory publishing.
Indeed, open-access publishers that must take X dollars from Y number of authors to keep in business have a clear economic incentive to maximize accepted papers and minimize rejections. This becomes an unavoidable conflict of interest for them. One cannot trust that their peer review procedures are not influenced, especially for the Swiss-based companies (names withheld so I don’t get sued…) that use editorial office staff (not scientists) to run peer review, and are allowed to overrule associate editor (the actual scientists) decisions.
APC-funded gold open access is a small part of open access. Most open access journals are diamond open access and platinum open access journals funded by their institutions. See DOAJ.
Is this debate at least partly “facing in the wrong direction?” Beall’s list (and perhaps Beall) had plenty of flaws. But, if not like that, how does academia protect itself from predatory journals when the publishing industry itself might not have the right incentives? There are maybe “tricks” that could be used (like sending spoof papers and seeing if they get reviews/acceptance) but this is a time consuming and potentially expensive business. Does a “bad” list mean we should give up on trying to compile a list at all?
On “how does academia protect itself”, there are a few ways:
* counter the “publish or perish” culture, starting at https://sfdora.org/ ;
* invest in open repositories and other open infrastructure (based on free/libre and open source software), cf. e.g. https://investinopen.org/ , to reinforce green open access;
* promote personal responsibility: https://thinkchecksubmit.org/ ;
* help authors find scholar-led gold open access journals, for instance at https://doaj.org/ (where one can find platinum/diamond OA journals and DOAJ seal journals).
This article is now republished at https://direct.mit.edu/qss/article/3/3/859/113621/Predatory-publishing-in-Scopus-Evidence-on-cross
Regression analysis in need of a control group?? And because of that retraction???
Or did I lost my reading glasses.