1. That depends on how they are involved. If they are testing, YES. If they are running a model on anonymised data, not necessarily.
2. Yes, it is a problem. However, there are solutions: one of them, which I mentioned in another response, consists in having two lists, one for participants name and one for data, which are stored in different places and can be recouped via a code attributed to each participant.
3. It should not happen. If you see that some minimal changes are needed, apply for an expedited procedure.
2. Yes, it is a problem. However, there are solutions: one of them, which I mentioned in another response, consists in having two lists, one for participants name and one for data, which are stored in different places and can be recouped via a code attributed to each participant.
3. It should not happen. If you see that some minimal changes are needed, apply for an expedited procedure.