For instance, in sociolinguistics, data generally consists of transcripts of interviews with subjects, hence these studies require IRB approval, make use of consent forms, and are likely to have stringent requirements for how the data is stored. However, online platforms like ShareLaTeX and the Open Science Framework (OSF) have great potential for collaboration, but primarily if that data is stored on those services. For example, if I store R scripts and csv data files on ShareLaTeX, I can then write up my paper on ShareLaTeX as well and use knitr to directly analyze and link to that data, all at once. This allows collaborators to have access at all times to a version controlled copy of the paper, the data that the paper is describing ,and the exact analyses that have been performed on that data. I'm not sure there is really another way to achieve this sort of interaction with collaborators without using some sort of cloud service and without storing data on that cloud service, but it's also not clear that simply making the data or the whole project private would be sufficient for protecting subjects.
As far as the OSF is concerned, ShareLaTeX can sync to the OSF via Github, which would also allow one to work offline without ruining the version control as RStudio allows one to sync to Github and work on knitr documents. This also has the advantage of providing forums for feedback from non-collaborators who are interested in the work. However, data would then not only be stored on ShareLaTeX, but also on Github and the OSF, which would mean it has to be secure enough to fulfill IRB requirements in both those contexts, as well.
Particular data files can be made private with all of these services, of course, but that's not necessarily the same as having them securely stored in an ethical and responsible way. So how exactly does one deal with this conflict between wanting to participate in open science and facilitate easy collaboration without risking exposing information about subjects?
(Experiences from sociolinguists who have tried an open science approach would be extremely welcome.)
- Badges
- Science topic
- Similar topics
- Higher Education
- Universities