I thought it was considered good practice to keep transcripts from qualitative interviews (well locked-up) in case future researchers should want to interrogate the data or check the conclusions. I could have sworn I read this in published papers on research practice - so much so that I didn't even track where I'd read it. Lesson learned - assume nothing - but is it not a matter of reliability and/or validity to ensure transcripts are kept as 'proof' of claims / conclusions? Or is it the case that it's fine to destroy them as soon as an article based on the findings has been accepted for publication? I understand the need to keep them from the general public, but thought this wouldn't apply to colleagues or academics in the same field.
Does anyone have any references for or against this idea?