11

So, this is a story that has a lot of minor details and twists and turns that are more complex than I am able to write and keep you, the reader, with your attention. I was invited to participate in a paper, after the initial idea and some execution began to happen. Basically it was an online survey and the group back then realized that they were not going to be able to analyze that amount of data in such a short time before the publishing deadline. The first author contacted me then, to help with data tidying, summarization and visualization, and some basic statistical tests.

By the time I was fully committed, some information was slowly revealed. I realized there were some issues with the data itself: questions that were not properly configured for statistics, wrongly designed questionnaire (questions had non-mutually exclusive answers when they should have) and there were in two languages, meaning the answers of one language had to be translated or converted into one to make proper data analysis. There were several open-ended question that got more than a thousand answers that needed to be categorized by hand. The core scientific questions that could be solved with the data were simply non-existent, they hoped that somehow the questionnaire itself would post-hoc yield some results.

I brought the tools and knowledge in statistical software (which none of the other members of the group had, but the "courtesy author" that had minimal participation in the work). I brought interesting insights that were original and of my authorship. All the tables and figures that were actually useful were initially proposed by me and they ended up in the final manuscript. Over 4000 lines of code were written in R. To be fair, I did not participate in the write-up or the results, nor in the introduction and discussion sections. Whenever the manuscript was sent for revisions, I participated and gave feedback on the topics that were being discussed. I gave the feedback that was required at the time and even gave a final typo and grammar review (500+ mistakes were there). When the manuscript was being recirculated, I was probably too naive to not see that in the authorship order, an "These authors have contributed equally to this work" asterisk was on 3 of the 6 authors and I was not included in that group. I only noticed that in the last review before sending it to the editor for final approval. There was not a meeting where this was discussed and decided.

Do you think that I have the right to ask for a change in the author attributions?

Improve this question
7

4 Answers 4

Reset to default
37

Even with your very important contributions and feedback, what you describe is a (very) poor paper.

You may see the radical improvement, because you know the starting point, however the casual literate reader of such a paper will only see the flaws you mentioned, even when nicely packaged.

Do you want to be identified as an important author of such a paper?

Let it go, you have now a bag of scripts ready to be reused for the next statistical analysis of a survey, wihtout forgetting that you learned a thing or two along the way.

You can surely put it in your cv, the fact that you are not in a prominent position is an advantage that allow you to correctly frame your contribution, if you are asked at an interview about this paper.

Improve this answer
2
  • 1
    Underrated answer.
    – Or4ng3h4t
    Oct 24 at 14:03
  • 15
    After speaking with some other key members in my field that I have a relationship, they arrived at the same conclusion and this how I am going to think about it from now on.
    – throwaway_academic
    Oct 24 at 16:31
20

The best time to negotiate changes in authorship plans would be when the scope of work changes, but before you've actually done the work. That is, "this seems like it is going to take a lot more effort on my part than I initially expected. I'm willing to take on this work, but I'd like to discuss plans for authorship again".

The next best time is as soon as you realize it. An additional "contributed equally" asterisk seems like a small ask, especially if it's already split between three of the authors, what could it possibly hurt to add a fourth? But, if you've already approved a final version for publication you may be a bit too late. Not that it isn't worth asking, but do so as quickly as possible and prepare to be disappointed.

I would not recommend counting your effort by lines of code, though (it's often much more work to do the job in fewer lines), or by number of errors in text when you didn't originally contribute to the writing. It sounds like you contributed substantially to the data analysis and figures/tables for the paper, as well as editing: that's enough to make your case.

Improve this answer
4
  • "I would not recommend counting your effort by lines of code, though (it's often much more work to do the job in fewer lines)" seems a bit inaccurate. A repo with a million lines might not be maximally efficient, but it's still potentially more work than 100k lines. It depends on the scope of what was done. Size indicates that scope at least a bit. Rather than referring to the lines of code refer to the scope of what was done.
    – TheEvilMetal
    Oct 24 at 14:13
  • 3
    @TheEvilMetal You can say it's inaccurate, but then also agree with me entirely in the last sentence of your comment :)
    – Bryan Krause ♦
    Oct 24 at 14:33
  • Totally get that lines of code might not be an absolute marker of effort, but like @TheEvilMetal says, it is 4000 LC versus none from the rest of the team members, and it is also tons and tons of work relative to what the original survey datafiles looked like (came from a propietary platform that made an effort making the analysis outside of the platform very hard, because they offer their own services).
    – throwaway_academic
    Oct 24 at 16:27
  • 2
    @throwaway_academic My advice would be to focus on the degree of change necessary for the original files to be workable, rather than the lines of code. As others have pointed out, the need for these efforts also reduces the value of the entire project to questionable.
    – Bryan Krause ♦
    Oct 24 at 17:33
12

It generally doesn't hurt to ask about authorship changes, but keep these points in mind:

  • in any research project, the amount of work is unexpected until it's finished. If you agreed to data-tidying, it is expected that there would be issues with the data (otherwise, it wouldn't need to be "tidied"). For example, translating or manual categorization seems to fit the 'data-tidying' task (in my opinion).
  • if you're asking for a change in authorship order (or recognition of same-contributions), you should come with a clear description of your own contributions.
  • I would advise against criticizing the contributions of others; only focus on your own contributions, the PI / last author should be the one most aware of each individual's contributions.
  • In this case, it looks like the first author is your link to the paper, so I would ask them about your proposition first, and if they agree, then ask the senior author.
  • Adding an 'authors contributed equally' star is usually not as important as the authorship order. If you're already 4th or more in the list, the star is not going to change much, so it might not be worth asking. (IMO)
Improve this answer
2

Unless you are first author or, in some fields, last (senior) author, the rest of the author ordering doesn't matter.

Nobody will look as closely at author ordering as you do.

Play the long game. If you co-author many great papers, that will help your career. Whether you are listed second or fifth is largely irrelevant.

Little asterisks about so-and-so contributed co-equally are meaningless and will not get picked up in citation ranking metrics.

Improve this answer
1
  • 4
    Little asterisks about so-and-so contributed co-equally will not get picked up in the meaningless citation ranking metrics.
    – EarlGrey
    Oct 25 at 15:09

You must log in to answer this question.

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