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

I wonder if they ran an a prior power analysis to see if the effect size they were looking for even had a good chance of being discovered.


If they are at all subject to a functional IACUC (i.e. at a university) this is required. The bar for testing things on (especially charismatic) animals is quite high.


I have seen otherwise. And statistical literacy is quite low.


On the second point we agree - but I have seen people absolutely torn apart for sloppy sample size calculations when it comes to animals.




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

Search: