7

Let’s take a very hypothetical situation. Suppose that a scientific journal (let’s use Scientific Reports for the sake of the example) recently published a paper that seems to be completely off track. By completely, I mean that the phenomenon described in the paper can be completely explained as a technical bias that has been ignored by the authors.

Some papers have an official means to address such concerns. For example, Nature has Matter Arising and Replies formats which can be used to engage in post-publication critics. The thing is, Scientific Reports does not seem to have such a mechanism.

I was wondering about the polite and efficient way to address such issue.

Improve this question
1
  • 2
    The de-facto solution seems to be that you mail it to Andrew Gelman and ask him to publicize it on his blog...
    – darij grinberg
    Commented Oct 3, 2019 at 1:12

2 Answers 2

Reset to default
2

You have to live with the fact that academia is an imperfect system like any else and scientific knowledge underlies evolution. Seeing the bigger picture, in the long term journals that don't allow or implement criticism/replies on their journals papers will likely vanish.

Your (very hypothetical premise) is simply very wrong that you have to find a way to correct wrong/bad content (in your opinion) in every journal.

Technically, this is the same answer as Wolfgang gave, but from a bigger different perspective.

You can also see it like this: Scientific reports doesn't believe that in the long term negative comments on their articles help to avoid more the publication of falsehood than for other journals or just creates "editorial noise/extra-work" (unworthful extra-work probably being the main reason to not offer this option). Do you rememeber facebook offering a downwote button, now it's gone. Or look at stackexchange, upvote +10, downvote -2 karma. When you have scientific branches like biomed/psychology in which over 70% of studies cannot be reproduced, it's questionable if article replies overall and in the long term have much value as this evolution is driven mainly by affirmation, reproduction by other groups instead of scientific discussions in a journal, as the system to create true new scientific knowledge works differently, less efficient and theoretical than in branches like mathematics or physics, which is not surprising if you have a bit of interdisciplinary methodological knowledge.

Improve this answer
2
-1

If it is obvious to you that the paper has a strong bias, then either (i) you are wrong in your opinion, given that apparently all reviewers and editors missed the bias, or (ii) the bias will be apparent to everyone (except the reviewers and editors).

In the first of these cases, there is nothing for you to do except doing some introspection on how you could be so wrong in your assessment. Talking to colleagues about the issue might help.

In the second of these cases, there is also nothing for you to do: The issue is obvious to every potential reader, so it's not necessary to point out. (You could, of course, try to point it out, but why rock a boat that everyone is already seeing as sinking.)

Improve this answer
2
  • 7
    This seems inaccurate. Reviewers and editors do miss things, and some technical issues can be subtle. Worse, once something is accepted in a journal, people are less likely to be as critical as they would with a draft, so whatever issue there is may not jump out.
    – JoshuaZ
    Commented Oct 3, 2019 at 0:02
  • 1
    This is accurate if the paper is completely obviously off track. I up voted. But what if the paper is off track and this is noticed just by OP. You might contact the authors for discussion, comment on the paper if the journal allows that, or write your own paper. But I wouldn't act because the paper is wrong,rather just for the impact it could have. Normally no much for most papers, especially if they are wrong.
    – Alchimista
    Commented Oct 3, 2019 at 11:00

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .