Preregistration – it will change things, but your unexpected findings will not be banished
Lots of activities going on changing science as we know it right now, and I have to seriously procrastinate from my regular job to keep up. (Well, as I am supposedly a scientist, and very actively a university educator involved in teaching students how to do science, I think it does somewhat overlap.)
NeuroSkeptic has a response out to a couple of blog-posts from Sebastiaan Mathôt, intended to calm his worry about the repercussions on pre-registered experiments. (I had considered blogging Sebastiaans posts also, but NeuroSkeptic does it better – but click through to them both and read.)
Dorothy Bishop also chimes in with a long discussion grounded in a talk by Chris Chambers about the new pre-registration requirement for some sections at Cortex. She briefly tweeted yesterday that people seemed worried, and here she discusses it in more detail than 140 characters can, well…
The worry is that this will put the kibbosh on reporting unexpected but interesting data in your findings. As NeuroSkeptic (and others) point out, this is not the case. The way that has been done (pretending that was what we meant to test) has been a major problem. We want to move towards a place where you can clearly report that, nope, what we thought would happen did not. But, we unexpectedly found this neat and interesting thing, that of course needs to be followed up. You didn’t get to India, but ran into the west-indies.