I work in an environment that's somewhere between academia and the professional world. I'm an analyst within this environment, in addition to my professional responsibilities dictated by the workplace's main mission, I assist the PhD's with their research. (I have an masters in Statistics)

I've been working on a project with one of the PhD's for over a year (it goes slow, professional responsibilities get in the way of research). The project is going nowhere for a variety of reasons (the first is that the methodology he wants to use is demonstrably flawed--from a statistical standpoint).

Several times over this time period he has dangled the carrot of co-authorship on the paper (at times saying "you will be a co-author", and a few weeks later saying "if you contribute maybe you will be a co-author". This is doubly frustrating when it seems to me that I'm the only one contributing...). He also has mentioned several times bringing his friend in as co-author as well (his friend has contributed nothing thus far. I met him, he didn't even seem enthusiastic about the project).

Thus far the main methodology we used is one he suggested, but actually bringing that through to completion has been entirely me. It yields no fruit, though, and I feel this whole enterprise has been an exercise in frustration.

I have another idea, approaching the problem in a different direction. I haven't yet done the modeling, but I believe the idea to be ideologically consistent with the original goal of the project, and should be statistically sound. The trouble is, I'm not sure I want to co-author with this person.

Am I wrong to consider nixing their project and doing my own thing on my own?

  • 1
    Note that to some degree this depends on what the project is - for example, are you using unpublished data collected by your collaborator that you'd need in order to do it "your way"? Are there any original ideas or contributions from your collaborator that might show up (or appear to show up) in your later, independent work?
    – ff524
    Oct 6, 2016 at 2:53
  • 1
    it's using pure public, readily accessible data. You could go and download it yourself right now. It's possible that they could say that there was one original idea that I took, but honestly, if they claim it's an original idea, they lack knowledge of the field.
    – Faydey
    Oct 6, 2016 at 3:02
  • 2
    It's okay to go ahead and try your new approach on your own, but tell your co-author that you are going this way. There will be hard feelings if they keep working or pinning hopes in the combined paper and are surprised when you publish on your own. Oct 6, 2016 at 3:31

1 Answer 1


Do you have a supervisor who would appreciate being kept in the loop?

I think your best bet is to give this colleague a clear message that the methodology he has proposed is flawed and that the results coming from that methodology are not promising. Ask him if he can think of any justification for you to spend more time on this project as currently contemplated.

In other words, pull the plug on the project. But indicate that if he has any other ideas for collaboration about something else, you'd be happy to work with him on a different project.

If you work up your other idea, and publish that work, make sure to mention him in the acknowledgements.

You must log in to answer this question.

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