While this might be interesting I don't see how you can get a balanced view. If you ask for negative anecdotes then that is what you get, but how will you know how true they are, or how well able the people were to judge what they were seeing? I took part in a demo lesson a couple of years ago where one person (out of 30 watching) decided I was scared of my pony, and my pony was also scared. Absolute rubbish, as the other 29 agreed... but if that one person contacted you with just their view of the truth, then what opinion would you have of what went on?
One trainer tapping the front leg of a horse gently with the toe of his boot, when the horse was attempting to walk all over his head... reported to me as "He's awful, he kicks horses!". I mean, how can you ensure that the anecdotes you collect are objective?
The idea about accident reporting is good in theory, but I worked at an NH yard for 6 months and not one accident in that time was reported to anyone. And there certainly were accidents to working and visiting students.
You mention that you are a future scientist, so can you come up with a fairer, more scientific way to compile your data?
I am not being defensive here, I could offer you a couple of first-hand observations of situations with well-known trainers where I think they damaged the horses, or used unecessary force, but I'm a bit uncomfortable about how they might be used.
As for the way France seem to be embracing Parelli - well, good luck to them! Seems pretty narrow minded to me.