If you torture the data long enough, it will confess.While this comment can be taken too far and exploratory data analysis is interesting, it is a good warning to medical (and other researchers) not to over-interpret data. Of course, this presumes that decisions are being driven by data and not based on expectations derived directly from a theoretical construct.
The world is a strange place and our minds seem to be pretty poor at "guessing" the correct answers.
No comments:
Post a Comment