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

That would also show consistent benefits in controlled studies.




No, not in practise. Its hard to controll for something that isn't understood enough. You would have to have a good ekough sampling across the groups, and esch group would have to be big enough.



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

Search: