47

I've found a mistake in a published paper, and am unsure what to do now.

I was looking at all cited articles from a paper over 10 years old, and found one that is citing the paper in error.

A variant of a certain protein and a fusion protein made of 2 proteins have been confused, that is the citation points to a paper on a protein which shares an abbreviation but is entirely different - this is a pretty big mistake and suggests they haven't actually read the paper they referenced!

The Elsevier journal published this in 2013 and I don't know how to report it, or what the etiquette is here. I can't see contacting the authors to be of much use as it is already printed, and they could very easily delete my email and forget about the error, whereas an editor or some such person would surely be able to take reasonable steps to address it.

For the record I'm an undergrad and have no professional link/competition/other conflict of interest with this researcher or department!

  • 20
    You'd be amazed how often people incorrectly cite prior research, or butcher their citations. People have even written papers on it (interfaces.journal.informs.org/content/38/2/125.abstract) – JohnnyO Aug 26 '13 at 18:27
  • 11
    You don't say whether that citation being the right one is essential for their results to be correct/meaningful. An occasional incorrect citation, while a mistake, is often not usually a "pretty big mistake". For instance: "Previous work on XGrp-beta indicates that it does not interact with NmpR [3,4,5,7]" with #5 actually not talking about NmpR is pretty minor. "All calculations in this paper were made using the reference diffusion constant of hTsp [8]" with anomalous diffusion being the main point of the paper, and [8] not even being the right protein, is a huge error. – Rex Kerr Aug 26 '13 at 21:16
  • 2
    The first thing is to determine whether this error "destroys" the paper which made the citation. If this is some small thing which can be fixed up, rescuing the paper, it is a different situation from the paper critically depending on the blunder being true. Also, is there any ripple effect: is there other work which depends on the results in that paper. How "big" is this, in other words? – Kaz Aug 26 '13 at 21:31
  • 2
    See also academia.stackexchange.com/q/3445/64 – Joel Reyes Noche Aug 27 '13 at 0:37
47

Small errors that do not affect the results or conclusions of the paper are normally handled through publication of a correction (or erratum). This is handled by the journal editor, who will be in contact with the authors for confirmation that they agree as to it being an error, and provide for the exact correction to be published.

The journal that published paper should have a policy on corrections, check it out! Physical Review Letters’s editorial policies and practices state, for example:

Errata.— The Errata section contains notices regarding errors or omissions in papers previously published. Besides the standard Erratum, several special categories of documents may appear in this section. In the online journal, each of these documents involve bidirectional links between the original article and the document in the Errata section. The category of the corrective document is indicated in its title and in the link from the original article.

The standard Erratum is a statement by the authors of the original paper that briefly describes the correction(s) and, where appropriate, any effects on the conclusions of the paper.

Thus, what you should do is:

  1. Make double sure it is actually an error.
  2. Then ask someone else to double check it again. Preferably someone more senior, i.e. with some experience of academic publication.
  3. Write to the corresponding author, pointing out the error. Be nice, and make real sure not to assume to worst.

    You say “this is a pretty big mistake and suggests they haven't actually read the paper they referenced”: depending on context, it could actually be something minor like a copy-paste mistake (pasting the wrong reference, when they meant another paper)

  4. If you do not obtain a response, or they respond but don't intend to correct the error, then consider contacting the editor.

I can understand why, as an undergrad, you would consider contacting the editor directly without writing to the authors first. However, as long as you remain professional in your correspondence with them, I think it's best to contact the authors first. It's more polite, and they may actually be able to provide you with some explanation you haven't thought of. Contacting the editor before the authors is somewhat overdoing it…

25

Sometime last year, a colleague of mine and I were writing a paper on scientometrics and reading associated articles. We were closely following a certain paper on preferential attachment in networks and realized that the numbers in the published tables do not add up.

We double checked with each other and then asked our faculty adviser to double check. Upon confirmation, we emailed the corresponding author about this error. The author immediately wrote back with a note of thanks. The major conclusions of the paper were not affected by this error - it was merely a calculation error which should have been caught during the reviewing process but sometimes isn't due to so many numbers floating around in a paper.

Since then, I have always assumed that contacting the corresponding author with a note of caution is always the best way to go.

12

I think it would be wise to bring the matter up with a researcher who may have the insights you have and beyond. You (two) can then discuss if something should be done and if so what. The error you describe sounds like it is significant but it could also be a typo or have less impact than what appears because of other factors. So getting someone to backing up your observations could be good.

What can be done? Depending on the severity of the problem, it may be necessary for the journal to retract a paper (if the problem negates the results). More commonly you and the researcher you pair up with could write a "Letter to the Editor" pointing out the problem and perhaps detailing the ramafications of the mistake. This letter could lead to a correction being made or lead to a discussion where the author will have to explain how the problem is not one.

So there are several possible outcomes. As an undegraduate you need the support of a more senior scientist since it is unlikely journal will accept letters from persons who are not considered experts in a field (which is what a PhD can signifiy). That does not reduce the merits of your observation but will ensure the observation gets the attention it may deserve.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.