Why did a chemistry journal fix fraud with a correction instead of a retraction?

Chemical Communciations

Are some cases of research fraud fixable with a correction notice?

A chemistry journal thought so in 2014, when it issued a correction notice for a 2012 paper after the first author admitted to manipulating an image. After an investigation, the publisher agreed the manipulation was a “clear breach” of its ethical guidelines, but decided not to retract the paper since the overall conclusions remain valid.

The last author told us the first author had to repeat the experiments under supervision, and received a “serious warning.”

It’s an older notice, but one we thought interesting enough to cover now. Once you’ve read through the journal’s reasoning, tell us if you agree with the decision to correct (rather than retract) the paper in a poll at the bottom of this post.

Here’s the correction for “A novel route for preparing highly proton conductive membrane materials with metal-organic frameworks,” issued by Chemical Communications:

The Editor has been made aware that Figure 3b in this article is identical to Figure 3a in a previous ChemComm article: Harold B. Tanh Jeazet, Claudia Staudt and Christoph Janiak, Chem. Commun., 2012, 48, 2140-2142. The copied image had been manipulated, with the new version rotated through 180 degrees and the scale removed. No acknowledgement or reference to the original communication was included.

Bin Wu, the first author on the article, has admitted manipulating the original image and including it in the recent communication. In a letter to the journal he apologised for his actions, commenting that through his lack of experience and understanding of scientific ethics he was unaware of the severity of his actions. All the authors, B. Wu, X. Lin, L. Ge, L. Wu and T. Xu, have accepted joint responsibility for this case of scientific fraud and would like to apologise to the authors of the original article, the journal and its readers.

The authors were asked to provide all original raw data relating to the images in the article, which were reviewed by a senior ChemComm referee. While the Editor and senior referee agree that the author’s actions amount to scientific misconduct, and are a clear breach of the Royal Society of Chemistry’s ethical guidelines, we have found that the overall conclusions and scientific findings in the paper are still valid, as they do not rely heavily on the copied and manipulated image. No further data manipulation or fabrication was found. In order to fulfil our duty as a publisher to correct the scientific record, and in accordance with the guidelines from the Committee in Publication Ethics (COPE), we have therefore decided to publish this correction.

The authors’ institution has been informed of the misconduct.

As a result of the circumstances outlined above, the following changes to the article are required.

Figure 3 in this article should be:

The authors have confirmed that the text within the article discussing the image remains accurate.

The paper has been cited 23 times, according to Thomson Reuters Web of Science.

Last author Tongwen Xu, who works at the University of Science and Technology of China, told us:

I agree with this correction. This misconduct of BinWu is on the copy of  Figure 3b from a ChemComm article. He was asked to repeat  all the experiments under the watch of other lab members and provided all original raw data relating to the images in the article. These data  were reviewed by a senior ChemComm referee and  ChemComm Editor. He also got the understanding from the original authors. They all thought that the overall conclusions and scientific findings in the paper are valid, as they do not rely heavily on the copied and manipulated image. No further data manipulation or fabrication was found. The Fig.3b only supports the physical dispersion of MOF in the membrane, not confirms any the chemical reaction or membrane special structure. The new SEM image of his membrane was provided as a correction.

We have checked carefully in the lab and found no other papers were affected.

BinWu was faced to serious warning throughout the lab for his misconduct. He apologized for his actions  and was aware of the severity of his actions.

This isn’t the first time a journal has fixed a paper with clear misconduct — to which the authors admit — using only a correction. Take this correction to a paper by Scott Brodie, a virologist banned from federal funds by the Office of Research Integrity, which explains that he falsified a panel in a figure, but that the rest of the article is valid. Or this correction to a paper by a graduate student at the Albert Einstein College of Medicine, which states that she admitted to faking data; when her co-authors redid the experiments, they found that the new data still supported the article’s conclusion.

COPE guidelines suggest that:

Journal editors should consider retracting a publication if: it reports unethical research

And:

Journal editors should consider issuing a correction if: a small portion of an otherwise reliable publication proves to be misleading (especially because of honest error)

We’ve reached out to Bin Wu, listed at the University of Science and Technology of China, and to the university. We will update this post with anything else we learn.

Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up on our homepage for an email every time there’s a new post, or subscribe to our new daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.

5 thoughts on “Why did a chemistry journal fix fraud with a correction instead of a retraction?”

  1. Assuming the content of the paper remains valid, as stated, misconduct corrected does not change the information and conclusions of value to future scientific endeavours. A salutary lesson for co-authorship since all will have a correction associated with this behaviour to their published record.

  2. Things like this really depend on the field, and although I’m a believer in research integrity and punishments for intentional fakery, I would have to vote either that the journal was correct, or that the journal can do whatever it wants in this case. The picture here matters so much less than a gel, so if the person actually made the material, the fact that they reused a SEM image means they were lazy and didn’t want to waste a few hours and didn’t know that this was wrong. The image is required probably because everyone likes to put shiny SEM images of material in their paper these days, but tells you next to nothing about this particular material as you can be sure it was already crystalline and embedded (i.e. even if there was no sulphonamide you can just mix the MOF cubes and a polymer together and get the same picture) and the picture is not going to tell you about the nature of the bonding interaction.

    The PXRD, IR, and thermogravometric analysis data are the real ‘picture’ of the material here that is comparable to a gel. If those are not faked (they are in the SI by the way), then I think the student was really just ignorant and tried to save a few hours on a side-show technique, and there was no intent to deceive.

    I know this blog is very biology/biochemistry focused, so based on image manipulation there everyone is going to be crying for blood in this case, but there really needs to be a field by field set of standards/punishments.

  3. The question is whether one judges on results or on the process. The misconduct was such that the results/conclusions were not affected. However, the scientific process was incorrect. Deliberately copying and manipulating (it became less recognizable) a picture is plagiarism and falsification. The scientific process is at the end more important.

  4. I guess I will be in the minority on this one. I believe credibility is the only currency in scientific research. If you commit a deliberate act that constitutes fraud, then you have no credibility. I, for one, would not waste my time reading a paper authored by someone who would commit fraud.

    The commenter, blatnoi, above illustrates the problem. In his second paragraph he/she says, “If those [analysis] are not faked…” That’s a big “if”, and the uncertainty exists because the author has already shown a propensity for deceit. I would have to ask the commenter how one would ever determine if analysis data was fake or not if you only had this researcher’s word that ot was valid.

    Determining worthwhile research is difficult enough. There’s already a lot of noise and misdirection in research by people with biases, agendas, sloppy work habits, or just being mistake-prone. Trying to separate deliberate falsification from real results in the same paper is too much to ask. Let some other group repeat the procedure and report their results. Pull the paper.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.