Two researchers in Denmark are guilty of scientific dishonesty, the Danish Committees on Scientific Dishonesty (DCSD, Danish acronym UVVU) has concluded.
In July, the DCSD said in a draft report that Bente Klarlund Pedersen had acted in a “scientifically dishonest” and “grossly negligent” manner. She — and many of her defenders — responded by saying that while she had made mistakes, she had not committed misconduct. Here’s a key quote from a letter her attorney wrote to the committee:
The errors, contrary to the manipulations, found in some of the articles are regrettable, but they do not make Bente Klarlund Pedersen guilty of scientific dishonesty. That would require a personally committed serious violation of good scientific practice that was willful or grossly negligent.
The DCSD clearly disagreed. In a summary of their ruling yesterday, which was first reported in the Danish press, they write that Klarlund Pedersen — identified as “Defendant” — was “not guilty of any scientific dishonesty” in one case involving four articles and a grant application “because the failure to provide information was due to an error made during the editing of the article.”
In the second case, however, involving 12 studies, they found she was “guilty of scientific dishonesty in six of the 12 articles” because of she was
- “involved in the decision only to take certain readings (results) from a selected part of the total test population” without the article making that clear
- reusing biopsy material in five papers “without this being stated in the articles,” and
- including “an additional group of subjects from a previous study” without indicating that in the relevant article.
Although the DCSD reports are anonymous, Klarlund Pedersen acknowledged that she was the defendant in the two cases. And Milena Penkowa is clearly the defendant in the a third case:
DCSD finds the Defendant guilty of scientific dishonesty in four scientific articles because of manipulation of graphical material. The Defendant was particularly deeply involved in drafting graphics and therefore had particular responsibility for this material in the articles. DCSD rules therefore that the Defendant was guilty at the very least of gross negligence by overlooking the fact that the four articles recycled the same graphical material to report on different research results and by attempting to obscure this fact through image manipulation.
The DCSD also said that Klarlund Pedersen had “joint responsibility” with Penkowa in manipulating images in four studies:
After studying the role played by the Defendant, DCSD finds the Defendant guilty of gross negligence in that she overlooked the image manipulation that had been confirmed in one of the four articles. The graphics in the article concerned are clearly manipulated, and the Defendant should have reacted to this when revising the article, which she would have been expected to do prior to publication.
Not surprisingly, Klarlund Pedersen objected to the final DCSD rulings, the way she did to the draft report. According to a Danish press report, she said she is offended by the outcome, and that
DCSD’s decision is based on an understanding of good research practices out of step with what many scientists consider to be good practice.
Of note, the DCSD wrote:
DCSD is of the opinion that in general the lead author (often called the senior or final author) of a scientific article has special responsibility for all of the article’s content, including reading the final manuscript carefully before submission to a journal.
She says: “DCSD’s decision is based on an understanding of good research practices out of step with what many scientists consider to be good practice.”
She means: DCSD’s decision is based on an understanding of good research practices [that is] out of step with what many scientists [whose work is often cited by the blog Retraction Watch] consider to be good practice.
I find myself at least somewhat sympathetic, as far as the questionable yet commonplace research practices go. I don’t think it’s fair to occasionally crucify a scientist who has followed those practices, which are probably necessary to get anything published. The underlying problems, of being unable to publish good research with unspectactular results, or being unable to publish without omitting some crucial information such as subject selection, should be addressed first and foremost.
Yeah, you just can’t publish these days without “clearly manipulat[ing]” graphics.
And many journals will publish re-used data so long as it is clearly indicated and cited properly.
“The underlying problems, of being unable to publish good research with unspectactular results, or being unable to publish without omitting some crucial information such as subject selection”
People publish with unspectacular results all the time. But if you didn’t earn the publication with high quality science, why do people believe that they should get it. Some semblance of standards is a good thing.
If these really the sorts of practice that are “necessary to get anything published,” then “science” today is nothing but complete bullish.
This seems to be a case of fraud, pure and simple, unless there is something I do not know. On the other hand, there is way too much pressure to get the “right” results.
Not entirely. Pedersen (but not Penkowa, IMHO) had real defenses, both legal and factual. The DCSD found against her. I won’t try to rehash the whole thing; but that finding minimally required the DCSD to move the goalposts on the definition of “misconduct.” Recall, too, that the DCSD are the same folks who issued the poorly-regarded Nyborg decision. That one also involved moving the goalposts (in Nyborg’s case, apparently moving them well outside the comfort zone of many RW readers).
I find the following sentences confusing, because it is unclear who the defendant is.
In the first half of the statement the author of the sentence implies that the defendant is the perpetrator of the graphic fabrication: “DCSD finds the Defendant guilty of scientific dishonesty in four scientific articles because of manipulation of graphical material. The Defendant was particularly deeply involved in drafting graphics and therefore had particular responsibility for this material in the articles.”
In the second half of the statement, the author of the sentence implies that the defendant is not the perpetrator of the fabrication but someone else who is responsible for the paper as a whole, someone who could to some extent be excused for having overlooked the misconduct: “DCSD rules therefore that the Defendant was guilty at the very least of gross negligence by overlooking the fact that the four articles recycled the same graphical material to report on different research results and by attempting to obscure this fact through image manipulation.”
The PI is in charge. From what is written above, whether the goalposts were shifted or not, what was done is misconduct/fraud. When it goes wrong the PI takes it on the chin, be it research misconduct or a lab safety problem. The rewards go to the person with whom responsibility rests. I fail to see how one can accrue rewards whilst taking no responsibility, as this is corrupt and leads us down a very slippery path indeed.
There is further news on the case of Penkowa, as she has just been found guilty of falsification of documents related to animal experiments by a Danish court, and received a 9-month conditional discharge (I think this is the legal term).
A relevant article (in Danish) can be found here http://universitetsavisen.dk/videnskab/penkowa-far-ni-maneders-betinget-faengsel