It’s really hard to get papers retracted, police might be best-equipped to handle scientific misconduct investigations, and there’s finally software that will identify likely image manipulation.
Those are three highlights from a number of pieces that have appeared in Nature and Nature Medicine in the past few weeks. Not surprisingly, there are common threads, so join us as we follow the bouncing ball.
First, a piece about the case of Alberto Fusco, which we’ve covered following some stories in the Italian media. Two big themes jump out here: One, Italy has no real official body to deal with allegations of scientific misconduct, so the police have ended up working the case. That fact led to an editorial accompanying the piece headlined “Call the cops.” The editorial concludes, with echoes of investigative reporter Brian Deer:
At the very least, academic investigators could learn from police methods for dealing with allegations of serious misconduct — that word again. And researchers might be less tempted to be cavalier with the truth — and with our money — if they knew who else could knock on their door.
Two, a molecular biologist named Enrico Bucci found likely image manipulation in a quarter of papers he analyzed with a new software program his company has developed. (Bucci has contacted us over the years, but we’ve never had substantial conversations with him. At one point a few months ago, he offered to walk us through the software, but our other priorities — day jobs, an endless stream of retractions — intervened. So we’re glad Nature‘s Alison Abbott ran with this one.)
This sounds like a really big and alarming number. It is, however, dramatically larger than the 1% of papers Rockefeller Press’s Mike Rossner and colleagues found to have image problems when they examined them years ago. [Update: See comment from Rossner below.] The explanation for the difference can be found in the story:
Bucci and his team created a database hosting all accessible biomedical papers published since 2000. But cleaning it of scientific contamination was not the quick job he had imagined. First he removed retracted papers; then he created a network of scientists who had been co-authors at least three times with authors of the retractions.
In other words, this was a highly selected group of which Fusco was a part. In addition to the retraction we’ve reported on, another will appear in Cell Death and Differentiation of a 2006 paper by Fusco and colleagues, Nature reports.
The search for problematic images brings us to a piece in last week’s Nature which featured Clare Francis — the pseudonymous whistleblower whose name will be familiar to Retraction Watch readers. Editors quoted in the piece said “many of her claims did not check out” (we should point out that Clare is actually a “he,” although the confusion is understandable given the pseudonym):
By 2011, editors were growing increasingly frustrated by Francis because — quite apart from her anonymity — many of her claims did not check out. “I have no problem taking time to look at an allegation — but I don’t like people wasting my time,” says Eric Murphy, editor-in-chief of Lipids. Moreover, many of Francis’s complaints are oblique and hard to follow, says Sullenberger. “It is helpful to know specific details about the concerns from a scientific standpoint, not just, ‘The bands in the 10- and 60-minute lanes are geometrical and superimposable’ or ‘Background is silvery smooth’,” she says, referring to some of Francis’s e-mails. Some journal editors have warned Francis that they are less likely to follow up on her requests than other complaints. In September 2011, Wiley’s then legal director, Roy Kaufman, sent her an e-mail saying that the company could “not guarantee that all anonymous allegations sent to us will be investigated”. Francis made the note public, sparking debate over how such allegations should be handled.
Having seen thousands of Francis’s emails, we’d respectfully disagree with these editors’ characterizations. What we’d like to know — and what Nature‘s Richard van Noorden was also trying to figure out — is just how many of Francis’s critiques lead to retractions, if properly investigated. We have a sense that percentage is higher than editors are acknowledging.
That means we must respectfully wonder whether some editors would rather find excuses not to investigate what are actually legitimate complaints. There is a typical quote in the Nature piece from one editor: “One has to wonder about the motivation of the whistle-blower,” says Ulrich Brandt, editor of Biochimica and Biophysica Acta. Brandt has made it clear elsewhere that he thinks editors should ignore anonymous whistleblowers.
But that sort of approach ignores the fact that, well, facts are stubborn things, as we’ve pointed out when calling for editors to pay more attention to such whistleblowers. We’d encourage Cell Death and Differentiation editor Gerry Melino to take a look at an email from Francis about another Fusco paper, for example. (By the way, Elsevier spokesperson Tom Reller, in comments that didn’t make it into the Nature piece but that he posted at Elsevier Connect, raised some other questions about Francis’s methods. His responses are worth a read.)
Finally, that stubborn refusal of many journal editors to properly correct the scientific record brings us to a refreshingly honest Nature Medicine editorial about why it is so difficult to get an obviously flawed paper retracted. (We’re not just praising it because it mentions us in the first paragraph.) The short version: Count the number of times the words “lawyers” (2) and “legal” (5) appear in the piece.
It is very challenging to get after a plagiarism supporting university, probably in any country. The police does not appear to be a useful institution to address with a case of idea theft, or stolen life’s work. It may be wise – but also useless – to have the university investigate any potential scientific misconduct of a famous professor. Unfortunately, in Germany universities appear to be out of any real control – it may take three years to start an investigation of plagiarism. BTW, the legal time for bringing someone in GErmany to court is also three years… In addition, the state attorney receiving the allegations and the proofs may be too close with the university and just ask for an exact date and time of the theft… Therefore, we need a better system to really investigate scientific misconduct in an international manner and to remove plagiators faster out of the system.
I see several paralleles to governments supporting doping in one or the other way. Officially, doping is not welcome, but when it comes to numbers of medals… In sciences it is officially not welcome to steal your guest scientists project and collaborator to support several diploma and PhD students and even put someone not familiar with the project on exactly this project to promote him to full professors salary – and forbid all the people working on the project to even mention the originators name, work and year-long unpaid project management and development. Why would a country help the victim of severest plagiarism, when the professor as part of the governmental education system would support only the plagiarism.
In regards to police investigations, and thereby government prosecution, of scientific fraud, did we already forget about Italy’s L’Aquila earthquake prosecutions? Admittedly the current state of affairs allows a great deal of fraud to be punished, as pointed out in the linked Nature Medicine editorial, by only a slap on the wrist. However I believe it would be a worse fate for science if we allow those with little to no actual knowledge of the scientific process to govern science. If scientists and engineers can imprisoned and convicted of manslaughter for ignoring predictions of relatively inaccurate models, a dangerous legal precedent is established for the prosecution of other instances of “non-reporting.” Would a cancer researcher be obliged to, under threat of prosecution, report to all cancer patients potentially life-saving experimental treatments? Legal systems are completely and entirely based upon the concept of “facts.” This black and white portrayal of reality has no place in the realm of science where facts are replaced by data and theories. The only measuring stick for a scientific idea is acceptance by a panel of peers – let’s not change that to a panel of 12 uneducated civilians.
If a crime (e.g. fraud) has been proposed to have been committed, of course a criminal investigation should take place. Should scientists rot in jail for failing to cite a poorly conducted study? No. But if a scientist fabricates data and uses that to obtain any sort of pay from someone else, that’s fraud and he/she should be persecuted. In America, we regularly jail people for lifestyle choices and petty crime. Shouldn’t we throw away the keys for someone who makes up data entirely or plagiarizes blatantly in order to take money from others and leads people to waste resources on research based on faulty premises?
Yes, scientific fraud is fraud. If you received money to perform research and you, knowingly and with the intention to cover yours mistakes, publish information that is made-up, stolen, plagiarized, yes, IT IS A CRIME and a police investigation should be conducted. In order to do this, many things need to be ready:
1) Whoever finds the misconduct/fraud, needs to come forward and provide all pertinent and detailed information to the proper authority that support the complain .
2) Involved researchers ought to provide all the information / lab results / photographs / statistical results / computers and storage devices, etc. to the investigator, so a fair investigation can be launched. Such information must also include the logging book to the research facility and possible surveillance videotapes.
3) Said involved researches should not be covered by their universities or employers (that would make them accomplices). Even more when the money granted comes from the Government…. In case the employer / universities tries by any means to protect / cover the misbehave, they should be punished as well. How? No more funding!!!
4) The money provider (also known as dad government) or other PROPER authority should find an arbitrator, who must have proper credentialing to perform the investigation. I mean, if the research is about genomics, the arbitrator must be himself an expert in such area. If results in questions are related to image manipulation, an expert on that area should be also be consulted.
5) If during the investigation, the researcher whose work is under the microscope, wants to come clean, it should be allowed to do so and the corresponding penalties kept concealed. But, if this instance is not followed, and after a due process it is confirmed fraud, the results of the investigation and penalties should be publicly exposed both to the research community and funding agencies.The penalties that must be dispensed should be:
For the researcher:
a) Immediate resignation from the academic institution and/or laboratory.
b) Immediate update of the investigation results to the funding agencies, in order to :
b.1. Ban the researcher for his/her misconduct, for at least 5 years, and
b.2. To start the legal process so punished investigator return the totality of the granted money to the funding institute, including the incurred legal fees.
c) Immediate request to the journals where their work was accepted and published, to flag those published articles in which the researches were named as authors, in order to alert the community of possible future notices or concern / retractions to come.
d) ETC, ETC, ETC.
My 3 dollars and 2 cents
RE: “Cell Death and Differentiation editor Gerry Melino”.
This is problematic. There must be an explanation.
https://pubpeer.com/publications/19541840
How does a policeman representing the claim of a publisher or editor investigate an author from a country that lies thousands of miles away? Intra-country prosecutions (or persecutions) may work, but transnational plagiarists, frauds or cheats will always, now and 100 years from now, get a slap on the wrist and, at worst, get a retraction. Plagiarism is not seen as a major crime, and it will not give returns on the investment to investigate. There is no appetite for stirring the pot, unless there is a study that is highly politically or economically motivated. So, understandably Italy’s L’Aquila earthquake case was high-profile because of the real risks to society of an entire nation. So, in some ways, additional scrutiny and criminal prosecution may have been warranted. I guess the same could be said to results that pertain to health, such as medical or pharmaceutical studies. But what about the scientist who studies the proteins in the hairs of the left leg of a water buffalo? No-one actually cares, so that person is literally free to do as they please, if that is their ultimate goal, i.e., to cheat, or to fabricate data, or plagiarize. I have advocated for a long time now that true justice is almost like a return to medieval times, with public hangings, not literally on a rope or a neck under the guillotine, but rather a figurative public exposure of shame. RW serves that function very nicely. If 5 or 10 million scientists are aware of RW, for example, and a cheat knows that if they cheat, their photo and story will be placed here at the scrutiny of the global peer pool, openly and unreservedly, who maybe come to look regularly at the latest scandals, then justice could be served in a simple but effective way. Because one essential element will be sowed into the academic pool: fear. Fear, shame and justice are three key ingredients that need to be carefully used. The problem are modern legal systems that then make the issue wishy-washy, e.g. libel or defamation. I claim these are smoke screens. If a story is based on pure fact, even if against a particular scientist, individual, society, institute, or even nation, then exposing those facts publically is nothing more than freedom of speech. So, although I am personally not against the police getting involved, I don’t see it as being a reliable, or long-term solution. Problems arising from within the system (in this case science and science publishing) must be resolved by players within the system. After all, how many policemen understand basic concepts like experimental design, hypothesis or statistical analysis?
Problematic Gerry Melino pulbication in his own journal.
https://pubpeer.com/publications/18927586
There must be a reasonable explanation.
G Melino holds positions as a professor in Rome, Italy, and the MRC toxicology unit, Leicester, U.K.
Mentioned here: http://www.nature.com/news/image-search-triggers-italian-police-probe-1.14295
“Melino, who has joint appointments at the MRC Toxicology Unit at the University of Leicester, UK, and the University of Rome Tor Vergata.”
A question that needs to be asked: is there a systemic problem at the MRC toxicology unit, Leicester, U.K.?
These high profile publications do seem to have problems on the page.
1. Nature 485, 507–511(24 May 2012) doi:10.1038/nature11058
https://pubpeer.com/publications/22622579
2. Genes & Dev. 2009. 23: 1207-1220. 10.1101/gad.516509
https://pubpeer.com/publications/19451221
As well as these high profile publications there are less high profile publications, which nevertheless are part of the rich tapestry of modern biochemistry.
3. The Journal of Neuroscience, 8 August 2007, 27(32): 8475-8485; doi: 10.1523/JNEUROSCI.0792-07.2007
https://pubpeer.com/publications/17687025
4. J Clin Invest. 2009;119(5):1109–1123. doi:10.1172/JCI35660.
https://pubpeer.com/publications/19363292
5. Molecular Pharmacology May 2011 vol. 79 no. 5 844-854. 10.1124/mol.110.069831
https://pubpeer.com/publications/21325018
thanks David Hardman for bringing this up. I was thinking to bring this up here – your are right there must be an explanation by the editor…
IMHO there should be legal consequences to scientific misconduct committed using taxpayer dollars. This doesn’t necessarily mean police knocking on or down anyone’s door, but it should at the very least entail potentially facing charges in a criminal court. On the surface this may seem a bit draconian, but consider a congressman that diverts taxpayer money or campaign donations into, say a country-club membership or a vacation in the Bahamas. This would clearly be wrong, and this congressman would be subject to a criminal investigation and possible jail time, yes? This is exactly what happened with (former) congressman Jesse Jackson, Jr., who is now serving 30 months in prison for using campaign funds (i.e., taxpayer money) for personal purchases. How are Jackson’s offenses different from a scientist who obtains a federal grant and then commits misconduct (or who commits misconduct to obtain that grant in the first place)? Oh, and part of Jackson’s conviction was due to “making false statements”…which seems awfully analogous to “publishing a paper” containing the fraudulent results.
I see no difference between these two scenarios. The politician uses taxpayer money for personal gain (i.e., wrongful purchases of goods/services) and then lies about it (in statements to investigators), and the scientist ALSO uses taxpayer money for personal gain (i.e., promotion, tenure, stature) and ALSO lies about it (in published papers). The politician goes to jail while the scientist merely gets supervision of federally funded research for three years, and a few other wrist-slaps. I’ll bet if scientists were held criminally liable for taxpayer-funded misconduct there would be a lot less of said misconduct.
I do not know how this works elsewhere, but it is worth checking the short US presentation Grant Fraud.
‘3:
1. Grant recipients are stewards of federal funds.
2. Grant dollars must be used for their intended purpose.
3. Where applicable, grant recipients must account for costs and justify expenditures.
Using federal grant dollars for unjust enrichment, personal gain, or other than their intended use is a form of theft, subject to criminal and civil prosecution under the laws of the United States’
‘4:
‘Federal grant dollars are susceptible to several forms of financial theft, most commonly in the form of specific federal violations, including: …
3. False Statements
4. False Claims
5. Mail Fraud and Wire Fraud
Each of these violations of law are subject to criminal prosecution, fines, restitution, and civil penalties.’
‘6: ‘Grant fraud occurs in many ways, but some of the most common fraud scenarios include:
1. Charging personal expenses as business expenses against the grant.
2. Charging for costs which have not been incurred or are not attributable to the grant.
Relevant US Code includes 18USC1001 (false statements act) and 31USC3729-3733, False Claims.
However, the normal paths rarely get to criminal action, and properly so, since it costs time and effort to prosecute these research misconduct for these things, and complaints have to be prioritized, and the $ amounts are usually much lower than some of the other frauds to be chased. But, criminal actions do happen, and people do go to jail.
The usual path is that complaints go to a university, if one is involved. They may also go to ORI for NIH-funded work, to NSF for NSF-funded work, or to OIGs for other agencies. Universities might ignore complaints , but doing this very often is likely to cause trouble with funders, and folks like ORI can make them redo investigations. Sometimes schools pay fines.
Here is a sample:
Former Penn State Professor Charged in $3 Million Federal Research Grant Fraud,
$640K restitution, 41 months in jail.
Cornell University Guilty of Misusing HIV Grant Funds’
Professor settles false claims case, school agrees to pay $2.5M.
Former MIT professor spared jail for grant fraud ‘A prompt confession, and letters from senior scientists begging for clemency, helped a former MIT professor Luk Van Parijs avoid jail for grant fraud. …’
(That’s by Eugenie Reich, a good science writer who has sometimes written on fraud, including the fine book Plastic Fantastic., which I recommend strongly, despite the sadness of seeing that story about a place I worked for 10 years (Bell Labs).
‘UCLA Professor Gets 2-Year Prison Term in Research Grant Fraud
‘two years in prison and ordered to pay $1.75 million in fines and restitution’
People might try NSF Search Case Closeout Memoranda.
Some complaints get investigated and rejected, but if upheld, usually some action occurs well before it gets to criminal proceedings, and I’d suggest that is an effective way to do things. OIGs don’t do criminal investigations, but they do work with DoJ as needed.
For an example of one that might have, but did not, see this closeout, whose story is described in more detail here, with related court cases.
Basically, the NSF OIG was *very* unhappy with the subject, using stronger language than one usually finds in the hundreds of closeouts I’ve read and recommended a 5-year debarment, although that was reduced to 3. By the time the case was closed (late 2009) the subject had moved to Australia in early 2008, his 501(c)(3) entity was being shut down with cooperation of those left and it was defunct, with no money. Now, perhaps there should have been a criminal complaint, but imagine what that would have required in time and money compared to any plausible restitution.
Anyway, for research fraud, I think criminal complaints are a necessary resort, but only a last resort and one would hope that other institutions better suited to the tasks are doing what they should be, which may or may not be true everywhere.
The laws that exist in the US are admirable (in terms of complexity), and most likely are mirrored by similar laws in about another dozen or so nations. Written laws are one thing, and implementation is another. How many of the nations that have laws or institutes in place actually implement them or have an appetitte to implement them (see Eibl’s example about Germany above)? And then how many of the remaining 180+ nations on this planet lack such laws? So, the balance of justice to injustice is about 1:10. So, while concerted efforts are being made to reign in criminals (sensu lacto) in select spots on the planet, the very same “criminals” can establish shop, or operate quite freely, without the fear of the law or of “police” authorities. With such an imbalanced ratio, expect fraud to increase 10-fold (even if the Fanelli 2013 paper proves otherwise, i.e. a linear increase as assessed by retraction data).
This may well be true: I certainly don’t have the data, and even in the US, have no idea of the fraction of research fraud caught by NSF, ORI, etc. I do know that:
a) One has to have the right laws.
b) One has to be willing to fund enforcement adequately, and I really doubt such enforcement *starts* with police, even if some actions get there. I simply cannot imagine most police forces maintaining scientific expertise of the sort needed.
c) Serious investigations take money and time (often years). That is certainly part of why the US has promulgated a consistent definition of research misconduct, and places like NSF and ORI expect research universities to actually handle such problems well, which some do and some do not. But this is like IRS investigations of fraud: people have to prioritize, as there is never enough staff to chase every single possible case. Personally, I think investing more would have good payoffs, but I don’t set these folks’ budgets.
For over two years now I do repeat at RW that IT’S TIME TO OUTLAW ACADEMIC FRAUD!
Here is briefly the background:
Some academics CONSPIRE with friendly editors to get their papers which are build on fraud (i.e. data fabrication/manipulation and/or plagiarism) published with the ONLY goal to get later more public money in the form of grants. This is per se a conspiracy to OBTAIN PUBLIC MONEY FOR PERSONAL GAIN BY DECEPTION — a crime in all jurisdictions, and should be treated be treated as such.
If politicians, even members of the Parliament, are held accountable, so should be the academics.
Yes, the time to outlaw academic fraud has come!
Ivan, I want to clarify the data that were provided in the piece that I wrote for The Scientist in 2006. Through many years of systematic visual screening of images in The Journal of Cell Biology, we found that 1% of papers accepted for publication had some form of image manipulation in at least one image in the paper that affected the interpretation of the data. About 25% of accepted manuscripts had at least one image that had to be remade because we detected “inappropriate” manipulation, that is, manipulation that violated our guidelines for presenting image data but did not affect the interpretation of the data.
Those numbers remained remarkably consistent throughout my time at The Rockefeller University Press (which ended in May of this year). The latter number is remarkably consistent with the data of Bucci and colleagues, as described in the Nature Medicine article. However, it is unclear from that article exactly how Bucci categorizes the manipulations he has quantified. This is an ongoing problem in discussions of the extent of image manipulation in scientific publications. The 25% figure that I quoted comprised manipulations that violated our guidelines, such as adjustments of contrast that removed all background information in a blot, but when we saw the original data, we determined that the manipulation had not affected the interpretation of the data. In such cases, the author was asked to remake the figure in question with an image that was not manipulated in a way that violated our guidelines, for example, an image of the same blot without a contrast adjustment. In the 1% of cases where we detected manipulation that affected the interpretation of the data, we revoked acceptance of the paper.
Your comment has opened a new and massive can of worms… Can you please explain, in detail, how the editor board came up with this massively large, and odd, 25% of figure manipulation as being “acceptable”. In fact, which individuals exactly formulated this rule and was the rule established based on internal meetings, or in consultation with the readership and wider community? Could one also extend this “rule” to claim that 25% of a figure can be re-used, or duplicated, in the journal (even if the original source if quoted)? There is a heated debate about retraction notices that lack quantification of plagiarism, thus weakening the case made by editors and publishers, and in some cases maybe even annulling the validity of the retraction notice. I susect that your 25% cut-off value is way too high. It should be near zero, or not more than 1-5%. Actually, in my fiel dof study, I see absolutely no need for any figure manipulation, so I am quite stunned to see this 25% value. How widespread is this value among bio-medical journals and why can’t we see such quantification in the Instructions to Authors of so many journals? In fact, how do you measure the edits made? So, one click with the mouse to increase/decrease contrast, for example, is considered as what percentage of “change”? Do you request all authors to provide the original scans of gels and other images so that you can compare? Even so, it is unclear how your quantification is made. For the sake of transparency, and also as a basic requirement for your own defense and that of your journal, a more transparent, quantitative and explcit explanation of these methods would be useful, and encouraging. Sorry if my comment may appear ignorant, but in the decades of publshing in my field of study, I have never seen a single editor or editor in chief analyze figure manipulation, so if you could also provide suitable guidelines and references, we could start to establish guidelines for our field of study. If you values are right, then this has horrific ramifications for the plant and agricultural sciences, where massive figure manipulations (as you describe) take place.
JATdS, the image manipulation guidelines for The Journal of Cell Biology are available here:
http://jcb.rupress.org/site/misc/ifora.xhtml
Click on “Editorial policies” and scroll down to “Image manipulation”. These guidelines were developed in 2002 in close consultation with the academic editors of the journal.
Here are some early articles that I wrote about this issue. The second one is co-authored by one of the academic editors of the journal:
http://jcb.rupress.org/content/158/7/1151.full
http://jcb.rupress.org/content/166/1/11.full
Here is the link to the article to which Ivan referred, which contains some details about the screening process:
http://www.the-scientist.com/?articles.view/articleNo/23749/title/How-to-Guard-Against-Image-Fraud/
Much appreciated. A good start to start adding pressure to editors in plant science journals who seem to be in a state of denial about the importance of post-publication peer review and concepts such as figure manipulation, duplication, and even the importance of retractions.
As David Hardman points out above, the problem with Melino acting in the best interests of scientific integrity at CDD, is that it might bring the spotlight on his own research. As seen on PubPeer and elsewhere, thanks to Clare Francis there are quite a few skeletons in that particular closet which he may not wish to draw attention to.
John Mashey is correct, it is very costly and difficult to prosecute “research fraud” as a criminal matter (as opposed to a civil administrative matter, by universities and federal agencies such as the Office of Research Integrity at HHS and the Office of Inspector General section at NSF). The “federal police” (U.S. Department of Justice) has a high threshold for initiating such a costly prosecution. However, it has been done, in some ORI/DoJ cases leading to prison time for the perpetrator of research fraud:
2005, Dr. Eric Poehlman, Professor at the Univ. of Vermont, formerly at the Univ. of Maryland at Baltimore, falsified massive human subject physiology testing. DoJ imposed a 1 year prison term plus 2 years probation for lying in an NIH grant application, with a $180,000 civil fine, and ORI debarred him for life:
http://www.nytimes.com/2006/10/22/magazine/22sciencefraud.html?pagewanted=1&_r=2
http://ori.hhs.gov/misconduct/cases/press_release_poehlman.shtml
2006, Mr. Paul Kornak, Clinical Research Coordinator at the Stratton New York Veterans Administration Medical Center, falsely claimed to be an M.D. and falsified clinical trial data, including that to enroll an ineligible veteran patient, who died from the treatment. DoJ found negligent homicide with a 6 year prison term, and ORI debarred him for life.
http://web.archive.org/web/20060203101242/http://www.usdoj.gov/usao/nyn/NewsReleases/2005/2005-02/200502031553.htm
http://www.nytimes.com/2005/02/06/nyregion/06vets.html
http://ori.hhs.gov/misconduct/cases/Kornak.shtml
Also see analysis in Science and Engineering Ethics, Volume 16, Number 4, 737-741, “Incarceration, Restitution, and Lifetime Debarment: Legal Consequences of Scientific Misconduct in the Eric Poehlman Case”
From the perspective of someone in an editorial office who has received Clare Francis emails, I’d just like to say that I have some sympathy for those editors who are frustrated with him. The tone of the emails tends to come across as hyperbolic, and it doesn’t help that many irrelevant people tend to be CC’d. The first time I got one, never having heard of Clare Francis, I did wonder how rational the sender was. If you got a wild and rambling email from an anonymous party accusing a colleague of rampant misconduct, wouldn’t you pause and think about how much sense it would make to follow up on it?
That said, I take my responsibility to COPE seriously, and i have investigated every Clare Francis email. It is frustrating that most of the time, there’s a simple explanation of the “plagiarism” (such as a notice on one journal article that it’s being reprinted with permission). I have to admit I do view Clare Francis as having cried wolf a few too many times. I will still investigate, don’t get me wrong, and it’s his right to proceed however he wishes, but I do think Clare Francis could have gotten more sympathy from journals if he had been a little more restrained in the way he brought up these issues, and if he had done a modicum of checking before making accusations.
Perhaps A Managing Editor would care to reveal how many of Ms Francis’s complaints he/she has referred to the institution for clarification?
Because as I recall the COPE guidelines whose responsibilities you take so seriously, editors do not do investigations – institutions do. Editors merely determine if there is a case to answer. Perhaps if Clare Francis had confidence that the editors he contacted would actually refer cases as they are supposed to, he/she would not cc so many people.
In my experience the COPE guidelines and institutional policies are fantastic – but editors don’t refer cases and institutions don’t investigate, so the policies are fantastic but completely pointless in equal proportions.
I apologize, “investigate” was poor word choice; I meant I looked into the accusation of plagiarism to see if it had any merit. I have not yet been in a situation where the editors felt, based on the COPE flowcharts (which in the situations I have faced, advise explaining to the author that secondary papers must refer to the original, and publishing a correction), that contacting the author’s superior was necessary.
All you need to do is take a look. It takes a few minutes at most.
You might also have a rolling program, into the future and into the past, of actually looking at the figures.
Journals need to be pro-active.
I do not believe that “a notice on one journal article that it’s being reprinted with permission” explains away most cases of suspected plagiarism. It might explain a few.
I agree, it is not a huge time sink compared to other responsibilities, but it can be a little more involved than one might think. For instance, in one situation an allegedly plagiarized paper was part of a supplement. I couldn’t be absolutely sure that the supplement didn’t have some sort of notice on its cover or front matter that the articles were being reproduced from another source, so pulling up the article online wasn’t sufficient to look into the matter. But of course the library didn’t have a hard copy of that supplement, so I needed to contact the other editorial office, who had to contact the publisher, and then I needed to follow up several times because I hadn’t heard anything, etc. Multiply that by all the hits in Deja Vu, which Clare Francis seems to work from, and it adds up.
My point isn’t to complain, but to ask if this is really the best use of journals’ time to look at decade-old possible plagiarism, as detected by Deja Vu? Absolutely, we do screen all provisionally accepted papers by CrossCheck, so we’re actively on the lookout for present-day plagiarism, but I do think it’s a legitimate question about how to allocate editorial office time, which regrettably is a limited resource.
(One note: I’ve never received a figure-related query from Clare Francis, it’s all been allegations of plagiarism.)
In reply to A Managing Editor December 11, 2013 at 8:13 am
I think that things have moved on. You might try prioritizing cases.
For example, concentrate on image manipulation, and the reuse of the same (or very similar) images in different papers to represent different things. Doing things like that has never been on the menu.
Together with a few colleagues you should be able to agree on what is more or less important. It does depend on judgement (I do not think that an “app” has been developed for that yet), and looking at the images. That might be the best use of resources. You did ask that question.
Look at the images before reading the text. This is advice from Julio Turrens.
I find this advice helpful.
http://ori.dhhs.gov/images/ddblock/dec_vol21_no1.pdf
“Dr. Julio Turrens, who reported that he and his students found many more examples of inconsistencies when they looked at the image before reading the paper.
Journals should advise reviewers to look at the images first.”
There are claims that computer programs can detect similar images in scientific papers, but they do seem a bit old hat as detecting the individuality of the human iris (or is it the retina?) has been used at some airports for several years. Also you do need to put the degree of similarity together with what might be reasonable practice, for example you could use the same loading control image more than once if it really was the same loading control. Looking at the images and thinking about reasonable explanantions can be done quite quickly by a person, who would have to decide what to do next in any case. If in doubt look at a few more publications. The pattern and frequency are important.
I do not think that it is unreasonable to ask journals to have programs of looking at images in past editions.
Very quickly journals would be able to collect the data. It is possible to score images for what is considered below standard pratice, or if it changes the meaning of the image. Many of the problems would solve themselves. The journals would have a better feel of what was the rate of mistakes. Cases above the “noise” of mistakes would identify themselves. Apart from correcting the record such programs might pay dividends in time saved in that they would discourage submission of manuscripts with below standard practices.
Some fields may be built on sand. It is better to find that out sooner than later.
Journals have a very important part to play.
This is one aspect that keeps bugging me when I see editors saying that they will not investigate from lack of time, lack of confidence on the claim, etc: what everyone calls “investigate” is a 5 to 10min long look into the paper and the presented evidence! It is certainly faster than moaning about it, and I believe should feel better than just ignoring it.
I understand that a proper polite tone is desirable when asking someone to have a look into a business, but I have this feeling that editors are taking anything as an excuse to not take any real responsibility for the output of their journals. And they expect no-one to “ramble” at this…
I once informed an editor about experimental plagiarism and that he should have the paper retracted; he excused himself that this would be the responsibility of the corresponding author (who in this case was not the last author). When I contacted the last author of the plagiarizing paper he mentioned that he was not the corresponding author and did not know anything about that part of the paper. So, it is easier for both, the editor and the plagiarizing authors to keep the paper not retracted.
From now on I will nit even bother contacting editors anymore. If they ignore us, the best to do is to ignore them. Let initiatives like PubPeer become the central for discussing published papers and if they should be taken seriously or cited in my research.
I feel saddened that my comments gave the impression that editors would not look into accusations. Responsible editors will, and I urge you not to give up on them. The original intent of my comment was to try and show the other perspective — why it can be frustrating to receive false alarms that seem to prejudge guilt. But it’s only frustrating because we *do* look into each and every accusation, even if we suspect it’s not going to lead to anything substantive.
If you have been brushed off by journal offices, then I certainly understand your frustration, but please don’t assume we’re all going to react the same way.
In every single case I have followed a whistleblower he/she would be either ignored or promptly responded which seem like two extreme outcomes but usually they both led to inactivity. The prompt response was usually either an attempt to “uncover their nasty personal reasons to bringing this up” or some encouraging “we will take immediate action, be assured” followed by inactivity. The inactivity had to be challenged with insisting on the matter and CCing COPE and the publishers. After that, very few resulted in real editorial measures, yet often interrupted by inactivity again at some stage (like officially announcing an expression of concern or retraction that never came) or resulting in some correction or retraction. In the case of retractions, I have seen cases where after the editors pulled back the retraction or silently did it by removing the paper .pdf without any note to readers of what had happened.
There is only one case in which I have seen a proper line of action, where a reader come openly to exposed two clearly irregular papers and the editor-in-chief promptly retracted the papers and even called the offender’s institution to complain. But in this case the whistleblower was a well-established scientist, the papers were complete copies from older papers, and the offenders were only students from a small institution. Easy to do the right thing heh?
I am afraid most cases of misconduct exposure run exactly the other way around. Where there are no more editors to mediate, then the debate will run directly on the challenged material, in public. I think this is the new situation, and I think it is much more beneficial.
Sometimes people do OK. In this example, a publisher acted properly and correctly, with appropriate due process, and forced a plagiarism ruling & retraction over wishes of the author AND the editor to be allowed to just add some citations.