[Yeti DNS Discuss] Friday afternoon thought: Registering Studies
shane at biigroup.cn
Fri Aug 21 09:23:47 UTC 2015
Here's an interesting study:
If scientists were to omit data from a study it would be serious
scientific fraud, but if a study gets quietly put into a drawer
somewhere then it is okay... at least it used to be! This is changing.
The bottom line from this paper is that before policy was changed to
require pre-registering studies, 57% of studies had positive results,
and afterwards only 8% had positive results.
Anyway, this is why I'd like to make sure that when we (soon!) start
running experiments in Yeti that we make sure that we design the
experiments in advance and publish what we are looking for. It seems
like an easy thing to do to get a big boost in trustworthiness of our
More information about the discuss