I have come across the following situation a couple of times:

After an experiment has been designed and data collected, the first author of the paper wants to ascertain whether a special analysis would work. So he asks a colleague to help/do this special analysis, which requires significant thinking and effort. Alas, the results are either inconclusive or don't add anything useful to the conclusions of the paper, so they never make it into the manuscript.

Should the colleague who worked hard on the specia analysis but produced no visible output be included as an author? I would think so, but I wonder how others would deal with this situation.

  • 27
    If I was trying to expand on your work I would hate to waste my time going down the same paths that your colleague went down only to find out they lead nowhere. The only difference between science and messing around is writing stuff down
    – zoplonix
    Feb 8, 2016 at 19:35
  • 1
    I'm guessing you're the colleague..
    – user541686
    Feb 8, 2016 at 20:23
  • 1
    @Mehrdad Nope, I'm the first author.
    – elisa
    Feb 8, 2016 at 22:56
  • 4
    I'm a coauthor of a paper where I think my biggest contribution was to convince the first author that one section should be dropped because it wasn't sound.
    – Joel
    Feb 9, 2016 at 10:44
  • 2
    Consider also that the analysis, regardless of its results, should at least be mentioned in the paper rather than file-drawered -- if only in a footnote, Appendix, or online repo such as Open Science Framework. See, e.g., Simmons et al.'s brilliant paper "False-Positive Psychology: Undisclosed Flexibility in Data Collection and Analysis Allows Presenting Anything As Significant". Then, because the results have made it into the paper (even if cursorily), this also justifies including the colleague as a coauthor if you so choose.
    – half-pass
    Feb 9, 2016 at 12:36

1 Answer 1


I had a situation like this come up once, where one person did an extensive analysis that turned out not to be particularly useful, so there was no point in including any details of it in the paper. However, we did feel that having the analysis done did lead (indirectly) to a better understanding of what we were studying. Moreover, we also felt that the person who did the analysis deserved some measure of credit in the final paper. So we put a very short mention (one to three sentences) in the manuscript, just stating that we had done the analysis, but that it had not led to any useful conclusions, for such-and-such reasons. Since the paper included this, it was unquestionably appropriate to include the person who had done the bulk of that analysis as an author.

I would suggest that you could do something similar, so your colleague could unambiguously be considered an author.

  • 54
    Agreed. Besides in this way you can prevent a reviewer from asking: "Why don't you carry out also 'that type of analysis'?" Feb 8, 2016 at 14:23
  • 21
    +1. By performing the analysis, they contributed to the paper, even if their contribution cannot be directly measured by anything that made it into the final paper.
    – Cronax
    Feb 8, 2016 at 16:04
  • 13
    +1 My personal philosophy has always been to be inclusive rather than exclusive with co-authorship. Here and there, we may have papers to which one of the co-authors did not end up contributing all that much, but I feel that is better than having papers where people who contributed ended up not been listed authors.
    – xLeitix
    Feb 8, 2016 at 16:29
  • 7
    I agree with this answer and voted it up, but I will point out that asking the person is wise. They might not want formal credit where a favor and good will will do. Also, you can cite a source without author credit. A line or two in the conclusion could reference their work. A footnote credit call out might be allowed or even an acknowledgements section. Some of your options will depend on the journal.
    – The Nate
    Feb 9, 2016 at 5:47
  • 1
    Wow, my answer currently has a score of 95! Now I'd really like to hit 100. ;)
    – Buzz
    Feb 10, 2016 at 21:50

You must log in to answer this question.

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