1

A couple of years ago, a colleague and I collaborated on a research report where we each collected a different set of data for our report. At the time, I conducted a set of interviews with one group of participants (school teachers) and my colleague conducted interviews with a completely different set (college profs). We each analyzed our data independently and wrote our own part of the discussion section pertaining to the data set we collected. The result was a research report that we presented at a conference and which we never published in a journal. My question is: would it be ethical to write a new paper based on the data I collected and excluding my colleague's sdata? Of course, I would have to modify the whole paper starting with the research questions to the literature review and other parts of the paper as well. In other words, I would write a 100% original paper using data collected for a different study. What is your take on this? I appreciate the responses!

4
  • 4
    Choosing the research question after the data is collected is another kind of misconduct. Avoid that.
    – Buffy
    Oct 8, 2021 at 18:48
  • @Buffy I understand your concerns, but I think you're forgetting about grounded theory. Oct 9, 2021 at 11:28
  • @DanielHatton, no, I'm not. Don't mischaracterize how grounded theory works. It doesn't use the data that "sparks" the question to also settle it.
    – Buffy
    Oct 9, 2021 at 11:52
  • @Buffy I'm not sure that's true: "negative case analysis" could be seen as an attempt to "settle the question" with the same data that sparked it. (And there's nothing wrong with that: grounded-theory research avoids the P-hacking problem by not using poorly-thought-through significance tests that are vulnerable to P-hacking.) But in any case, you didn't say "using the data that "sparks" the question to also settle it is another kind of misconduct", you said "choosing the research question after the data is collected is another kind of misconduct". Oct 9, 2021 at 12:11

2 Answers 2

1

Presumably you designed the experiment together. That makes it a joint work. The best way forward is to find a way to work with your co-author on the paper.

In an extreme case, but ethical, you could write the paper yourself, list both of you as authors, and just get the other person to approve of it. More work for you, of course, but it avoids all the pitfalls.

But, perhaps in seeing a draft, they will decide to contribute more.

If you are willing to state a research question that depends only on your own ideas, re-define the methodology, and then re-gather data to test it is another way forward, that might only require acknowledgement of the other person's early contribution. But the existing data is tainted at this point.


See: https://en.wikipedia.org/wiki/Data_dredging

0

It might not be ethical. If you and your colleague discussed the questionnaire design and the methods you would use to analyze the data then even though you are using only your own data this new work is not yours alone.

If you are simply reusing your data to try to address a different question it is probably OK.

You should discuss how to go forward with your colleague. Perhaps just an appropriate acknowledgment in a discussion of the history of the previous project would do. Perhaps coauthorship is really called for.

4
  • Thank you so much for your answer. Such a good point about the importance of writing new a research question: that is exactly my intent as the single set of data I collected would answer a different kind of question than the two sets together. Regarding the instrument, we each developed a different instrument to target each group of participants, so I would say the labour on the original report was divided vertically in half between us.
    – Sam
    Oct 8, 2021 at 18:40
  • See my comment on the question, please.
    – Buffy
    Oct 8, 2021 at 18:49
  • Sorry, but much of this is wrong. The second paragraph and the second sentence of the third.
    – Buffy
    Oct 8, 2021 at 19:02
  • See: academia.stackexchange.com/q/60393/75368
    – Buffy
    Oct 8, 2021 at 19:07

You must log in to answer this question.

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