I am in the data analysis phase of my Phd Dissertation. My supervisor has demanded that I share my transcripts with them so they can "help me" analyze my data. I do not want their help analyzing my data. Must I take this help, and must I share the transcripts with them? I fear they are going to try to use the data in their own research, which is not unheard of for them.

  • 1
    Transcripts of what? Interviews? – Buffy Oct 9 '19 at 18:11
  • 11
    Your phrasing of the question seems to imply that you have a larger problem that can't be addressed with a direct answer. Why are you continuing with an advisor you don't think you can trust to deal fairly with you? Are you just trapped? – Buffy Oct 9 '19 at 18:24

I think there are two questions here: 1) if you need to share your data with your supervisor and 2) what happens if they use it in their own research.

To question #1: The answer to this is yes as they are in a supervisory role and part of their job is to help guide you in your development as a scientist. This is also not an unusual request in my experience. Keeping data secret from your PhD supervisor would also complicated possible publication of said data in a peer reviewed journal. The usual standard around publication ethics hold that all authors are responsible for the entirety of the manuscript and in order for them to bear that responsibility they have access to whatever part they want to see: https://www.etikkom.no/en/library/topics/authorship-and-co-authorship/authorship-and-co-authorship-in-medical-and-health-research/

That said if your supervisor is absolutely not involved in your project and would not be a co-author on any publication or you don't trust your supervisor to act in a fair way then I think that there is a larger issue that is beyond the ability of strangers on the internet to give you advice on. My recommendation would be to go talk to your supervisor about this situation and if you are uncomfortable doing that then I would recommend you seek out someone who you can trust in your department/at your university and try to explore your options.

2) It is not uncommon for data to be reused in different publications. In the extreme example the entire concept of a meta-analysis is reanalyzing existing publications. So having your supervisor use your data for a different analysis/research project is normal. That being said they need to cite your publication as the source for the data and if you significantly contribute to their research then they should add you as a coauthor. If they fail to cite your publication/thesis as the source of the data then they are engaging in misconduct.

  • Or, cite you as the source, if the data hasn't been published yet. – Buffy Oct 9 '19 at 18:28
  • 3
    "The answer to this is yes as they are in a supervisory role and part of their job is to help guide you in your development as a scientist." This entirely depends on what the poster means by "transcripts". If these are transcripts of interviews and the particular ethical approval that the poster has for conducting their research precludes sharing interview transcripts with other researchers, no they cannot share them with their supervisor. This is the case in a lot of research. I have PhD students who cannot share their transcripts with me. – GrotesqueSI Oct 9 '19 at 20:26
  • 2
    I also cannot share my transcripts with anyone else, even when I've replaced all names with pseudonyms due to the possible risks to my interviewees. – GrotesqueSI Oct 9 '19 at 20:30
  • 2
    This was a possibility I had frankly not been aware of because even in the sensitive medical work I'm familiar with the IRB applications include everyone on the team so that data can be shared within research teams. Otherwise how do you expect someone to be a coauthor on a manuscript if they have no ability to see the underlying data? That said I doubt this is the case for the particular poster as she didn't raise said concern but rather concerns about her supervisors behavior with said data if he got it (i.e., the supervisor using it) – user2548575 Oct 9 '19 at 20:47

Not the answer you're looking for? Browse other questions tagged or ask your own question.