There’s been a great furor recently about a study which purports to show that rats fed GM corn develop more tumors than rats fed regular corn.  I’m actually a bit late to this party; scientists and science writers across the web have already picked apart the flaws in this study, from shoddy statistics to poor design, and Carl Zimmer has called the whole thing “a rancid, corrupt way to report about science“.  I don’t have much to add to the chorus; what I’d like to do with this post is to make clear to the layperson what we mean by “bad statistics” and why that makes the study unconvincing.

Continue reading →