For my Computer Science Master Thesis Research I need some large pre-processed datasets.

I know the names of researchers who have those sets (or at least can point me to the source).

My question is: do you have tips on how to approach those researchers to increase the chance of success to share data? What to do absolutely not?


  • 2
    Perhaps a broad question is useful, but you may want to say a little more about the kinds of data to which you are seeking access and what is your relationship with the researchers. For example, have the researchers already published articles using the data? Are they datasets that have a standardised access protocol as is the case with some large longitudinal studies? Do they have any obligations to share the data with you (grant rules; journal rules)? Do they have any existing research relationships with you, your university, or your supervisor? Feb 25, 2015 at 8:19
  • 3
    From my experience in the same field, a polite/friendly e-mail will do. If a researcher is generally unable/unwilling to share his data, e.g. because of confidentiality agreements, your e-mail likely won't change his/her mind. In contrast, being rude might lead to you being denied access to the data, even if the researcher in general would be open to such requests. In short: Use basic common sense, no magic involved here! (Not posting as an answer since it seems a bit vague to me...)
    – anderas
    Feb 25, 2015 at 10:57
  • I would ask your supervisor to establish contact - somebody that these "other" researchers know or have at least already heard of has a much higher chance of getting somewhere than a master student, who will very likely be an unknown quantity.
    – xLeitix
    Feb 25, 2015 at 16:21

1 Answer 1


Email, unless they happen to be at your university.

Just introduce yourself, and the possibility that their data might be useful for advancing your project. Leave open flexibility about the possibility of acknowledgement/authorship (depending on the effort you're asking them to go to), IRB approvals, data sharing agreements, etc.

Be prepared for rejection, but I've had something like a 60% success rate with this approach, and birthed several really productive collaborations.

You must log in to answer this question.

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