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

There are objective measures of statistical power, given an effect size a priori ahead of time you can estimate the power of a particular procedure. The trouble is that what a "reasonable" effect size may be is subjective and requires prior knowledge; post-hoc power calculations are widely regarded as misleading and conveying little additional information beyond a p-value.


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

Search: