The Blasé Nature of Retraction Watch ’ s Editorial Policies and the Risk to Sinking Journalistic Standards

Retraction Watch is a distinctly anti-science blog whose primary objective is to smear scientists who hold errors or retractions to their names. Self-appointed watchdogs, Dr. Ivan Oransky and Dr. Adam Marcus lead this profitable charity, which is falsely projected as an “aid-science” project. Claiming to represent the highest standards of science journalism, Oransky and Marcus oversee – with very hawkish eyes – a small team of journalists, who are in charge of researching cases and reporting them. On average, one can observe roughly one blog post every two days in recent months. This indicates that the research required to complete each story takes place under strain, and may thus involve errors. Scientists are very familiar with errors, and there are strict guidelines in dealing with them during the publication process. In science publishing, the process is somewhat bureaucratic and a formal request must be made to the journal and publisher, usually through the editor-in-chief, and a carefully vetted and worded erratum, or corrigendum, is then published. Retraction Watch’s Oransky and Marcus hold scientists to extremely high standards, and anything less than perfection is subjected to ridicule on their blog. Yet, what happens when Retraction Watch errs? This letter aims to highlight a single recent case study in which highly observant readers of a Retraction Watch story pointed out errors in the title and possible ambiguities. Rather than issuing a formal retraction of their old title, or a formal correction (erratum), the Retraction Watch leadership simply changed the title, and even though it was “loosely” acknowledged by Retraction Watch staff as “Fixed, thanks!” in the comment section, this attitude is blasé, and the method of correction is unsatisfactory and unprofessional. Scientists must hold Retraction Watch accountable for their errors, and any unprofessionally processed story, or correction, must be publicly exposed, to increase accountability. The Retraction Watch web-page, and thus its parent organization, The Center for Scientific Integrity, is receiving hundreds of thousands of US$ from at least one philanthropic organization, in support of its anti-science rhetoric, and thus the individuals that make up Retraction Watch must be held accountable for their occasionally poorly vetted and unscrutinized journalistic practices.


Retraction Watch: A Case of Manipulated Journalistic Standards 1.
Retraction Watch (www.retractionwatch.com)claims to be a forum to discuss retractions as a window into understanding science's many problems.Yet, run by two self-declared watchdogs who failed to succeed in the cut-throat world of medical science, Dr. Ivan Oransky and Dr. Adam Marcus, took the easier path of being critical of scientists, focusing their attention on retractions, a bold and worthy initiative, no doubt, by entering the world of science journalism.One of the reasons why a scientific career is so tough -at least at higher levels of research -is because the rules that bind scientists to a select level of ethical standards is very high, much higher than in most areas of society.And the levels of rigor and scrutiny in science publishing are increasing.Admittedly, given the money and power involved in this cut-throat field, science is also open to abuse and fraud, and in that respect, Retraction Watch does a good job of covering many interesting cases worthy of reflection.However, in journalism, especially science journalism, covering facts, as in science, requires the highest level of professionalism, ethics and protocol.Failure to report facts accurately, or failure to correct erroneous or misleading facts can cast a shadow of doubt on a journalist's reputation, or their organization, due to questionable skills.
In this letter I wish to show how Retraction Watch -whose parent organization is The Center for Scientific Integrity (CSI) -not only edited and altered a title of a recent story, but also manipulated it to reflect quite a different factual reality.I will show how their blasé attitude of fixing errors, typically as "Fixed, thanks!", without indicating the error and the corrected form clearly to the public, constitutes a serious violation of journalistic integrity.This is dishonest and misleading to the public.The story was written by Dalmeet Singh Chawla (Chawla 2016a), but approved by the highest rank of the Retraction Watch organization, either Oransky, or Marcus.The Chawla story is indirectly related to Olivier Voinnet, a (formerly) highly acclaimed plant scientist whose figure manipulation and misconduct were exposed in 2014-2016, through whistle-blower web-sites and blogs, including Retraction Watch, in a high-profile scandal that rocked the plant science world, which has been characterized as a "black swan" event (Teixeira da Silva 2015).Several Voinnet papers have been corrected now, either by multiple errata, or retractions.As reported by Keith (2015), two of Voinnet's former and/or current affiliations, Centre National de la Recherche Scientifique (CNRS), in France (the EU), and in ETH Zurich, in Switzerland (non-EU), found varying levels of errors, with the CNRS declaring "the existence of deliberate chart/diagram manipulations" that "amount to scientific misconduct," whereas ETH found that Voinnet "breached his duty of care in the handling of figures as well as in his supervisory duties as a research director", but did not classify his errors as misconduct.In the Keith story, a journalistic error was made in which Keith claimed that the report relied exclusively on work conducted at ETH.The story was corrected as a strikethrough on the web-page to reflect the error (Fig. 1A), but no apology was offered to the public for the original publication of the erroneous fact.Thus, to some extent, Retraction Watch took some responsibility for its error, but the light-headed attitude of that correction -by not apologizing to the public for misleading them with false and incorrect information in the original post -left much to be desired, especially given the fact that -as discussed briefly above -Retraction Watch currently represents one of the most popular blogs (in terms of monthly web traffic) related to science, with hundreds of thousands of hits a month, and the recipient of large charitable funds by philanthropic organizations such as the MacArthur Foundation, totaling hundreds of thousands of US$ (Oransky 2014).Consequently, scientists and the public should demand the highest levels of journalistic standards from Retraction Watch.Thus, errors of fact, even if corrected, reflect poor vetting and research, i.e., sloppy science journalism, and thus should be called out, as equally as Retraction Watch calls out scientists, and their errors, in public.

The Evolution of Error, and Erroneous Correction 2.
Returning to Chawla (2016a), the CNRS and ETH Zurich released a joint press release on September 8, 2016.Although each organization released a separate press release (CNRS 2016), and ETH Zurich ( 2016), but identically worded, both organizations indicated that they were coordinating to set up a commission of inquiry to investigate molecular biology papers.Curiously, the CNRS press release referred to the inquiry as being related to "misconduct" while ETH Zurich did not allude to misconduct.However, these organizations did not specify the exact individual(s) who were being investigated, nor the exact papers or number of papers that were being examined, leaving the press open to widespread speculation.However, given the recent fallout with Voinnet, it was highly likely that this investigation may have been closely related to Voinnet.Clearly hungry for a sensationalist scoop before other news media outlets or competing blogs reported this news, Retraction Watch hurriedly published the story, entitled as "EU institutions are investigating several papers, and that's all we know."Retraction Watch accompanied the launch of its scoop with a Tweet announcing the story with the same title (Fig. 1B).The URL also reflects this title.Two and a half hours later, a reader, Klavs Hansen, pointed out "A minor correction: Switzerland is not an EU country."(Fig. 1C) while another reader, Alan Piljic, pointed out "This title is misleading.I first thought there is an investigation by EC or another major EU institution."(Fig. 1D) five hours after Hansen had made his remark.
An hour later, the title of the story had been changed to "Swiss, French institutions investigating several papers", and possibly one or more parts of the text that may have indicated erroneously that Switzerland was an EU country, were edited.In addition, the comment "Fixed, thanks!" was added by Alison McCook in response to the criticisms made by Hansen and Piljic (Fig. 1C, 1D).Surprised not to have seen any strikethrough to indicate the errors (as done correctly in Keith 2015), and a formal explanation -and apology -to the public and readers for such crass errors, and poor journalistic judgement, I decided to contact Hansen and Piljic, who kindly responded to my request for comment and to verify that the title and/or text content had changed (e.g., Fig. 1E; Chawla 2016b).After a few more exchanges with these readers, in which I expressed my concerns, a bit more evidence of factual story manipulation by Retraction Watch was revealed.

The Correct and Expected Procedure for Correcting Factual Blog Errors 3.
Retraction Watch is no light-weight science blog.Attracting hundreds of thousands of monthly readers to its site, Retraction Watch might currently represent the scientific site, except possibly for publishers' sites or science data-bases, with the highest level of traffic and thus readership.This implies that facts prior to publication should be extremely carefully researched and vetted, and discussed among Retraction Watch staff and management, before publication.Factually rich stories should be approved for publication, confirming accuracy and legitimacy, by members of the CSI ISSN 2039-9340 (print) Mediterranean Journal of Social Sciences MCSER Publishing, Rome-Italy Vol 7 No 6 November 2016 13 board (CSI 2016).Scientists know all too well how easy mistakes can be made, even after careful scrutiny and multiple checks, and even after examination by many individuals, including editors and peers.When scientists err, they are generally expected to correct the literature (Teixeira da Silva 2016), and, depending on many factors, even small errors may require the publication of an erratum, such as an erroneous fact.In the mainstream science literature, there are fairly well established and defined guidelines and procedures for going about correcting erroneous literature and issuing an erratum (Teixeira da Silva and Dobránszki 2016).However, this does not appear to be the case with blogs, and even among scientific journals, large variation exists in dealing with errors.However, as already indicated, Retraction Watch is no ordinary blog.Run by two self-appointed watchdogs who created the CSI and who self-moderate content and errors, Retraction Watch is meant to -at least it is projected to be so by Oransky and Marcus -epitomize the pinnacle of publishing ethics in journalistic blogging, especially related to science.As equally as Retraction Watch holds scientists to extremely high publishing values, including the correction of the literature and published facts, so too should scientists and the public hold Retraction Watch to equally high journalistic standards.
The fact that the title (and possibly content -because there is no way of independently verifying the original content in this blog post or any Retraction Watch post for that matter) was altered to reflect a new "fact" (i.e., the facts had been altered without a formal apology to the readership and public, and without clearly indicating the error as a footnote, at minimum), calls into question the slack standards and lack of journalistic professionalism of the Retraction Watch team.It also calls into question what guidelines are in place to correct the erroneous facts on Retraction Watch, brings doubt regarding oversight and accountability, and thus public responsibility.There are currently no stated guidelines on the blog regarding errata and the correction of erroneous facts, amounting to a clear gap in public accountability and journalistic responsibility, or predatory behavior, using the analogy by Jeffrey Beall, a stanch Retraction Watch ally, and prominent "predatory open access journal" critic.It also calls into question the effectiveness of the vertical or horizontal power structure at Retraction Watch, including the function, use and accountability of the CSI and its board of directors.An organization that claims to represent scientific (i.e., factual) integrity, but that is unable to exercise the same levels of quality that it forcefully imposes on its readership and target group (scientists), is both misleading and deceiving.The CSI thus represents a threat to public and science integrity.
Finally, the fact that a blasé apology was offered, simply as "Thanks, fixed!" (Fig. 1C, 1D), not only undermines the formality of the facts being presented to the public -this is not a trivial case about someone's pet, it is about a formal ethical investigation run by two of the most respectable science research institutes in the EU and Switzerland -but also the attitude, tone and methodology towards corrections.Most scientists who take their (and others') science seriously, and who care about the accuracy of the published record would easily appreciate why this journalistic gaffe is so serious, and why the correction is so light-headed.Yet, anti-science detractors and pro-Retraction Watch supporters will surely quickly work to undermine this gaffe and underplay its importance, using deflectionary tactics.
To appreciate the importance of this issue, one need only ask the following: Imagine a scientist wishes to change the title of their published scientific paper because a reader pointed out an error, or misleading statement.What would happen to science's integrity and the integrity of the published literature if editors or editors in chief could simply correct the academic record within hours after the error was reported, with no formal apology, with no formally documented record of the error, and with no indication of what changed, and into what it had changed?Although the question is rhetorical, it deserves a response.Science publishing and the published scientific record, under such a poor modus operandi -as displayed by Retraction Watch in this case -would become corrupted within the space of weeks or months.This poor level of quality control and correction of factually incorrect -or ambiguous -facts may be acceptable to scientific journalists like Oransky and Marcus, but it is certainly way under par for scientists.
Oransky and Marcus were contacted for a formal response about the change in title of the Chawla (2016a) article.Oransky responded on the same day, in response to the October 15 2016 request to comment on the change in title, as follows (in fact this was the entire content of the email, further highlighting his deplorable public communication skills): "Read the first comment on the post, and the response to it."

Fig. 1 :
Fig. 1: Erroneous and blasé error correction by Retraction Watch.(A) In Keith (2015), an error was correctly corrected as a strikethrough on the web-page (see red box), serving as a responsible public acknowledgement to the public of that error.No public apology was offered for the error.(B) Retraction Watch Tweets the Chawla (2016a) story on September 8, 2016.(C) Comment by a Retraction Watch reader/commentator, Klavs Hansen, pointing out a serious factual error, i.e., that Switzerland is not an EU country.(D) Comment by another Retraction Watch reader/commentator, Alan Piljic, pointing out an ambiguity arising from the factual error pointed out by Hansen.C and D also include the blasé response by a Retraction Watch editor, Alison McCook (Oransky 2015) as "Fixed, thanks!"Comments and fixes in C and D made within hours of the story having been published.(E) Additional proof provided that the original title had been manipulated (Chawla 2016b).