Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

When the numbers were small to begin with, a doubled risk ratio could just be the result of statistical noise.

You are right in that the mechanisms may not be well-understood. This does warrant further study, as the paper concludes.



the studies obviously adjusted for variance when considering if their conclusions are significant, you're beating up a strawman




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: