Hacker Newsnew | comments | show | ask | jobs | submit login

Is this analysis a published scientific research?



Joking aside, the authors of these studies have highlighted the low statistical power of research in a number of fields (see also http://www.nature.com/nrn/journal/v14/n5/abs/nrn3475.html), which is a serious issue that often leads to over-interpretation. It's as if researches willing to publish more are rushing hasty studies out the door.

It's a good thing that people can measure these things and ring the alarm

-----


Depends. By and large, Ioannidis's paper is pointing out the basic consequences of the alpha and beta parameters in Neyman-Pearson hypothesis testing, so there's a very good argument that this paper is merely statistics, and statistics is generally considered to be mathematics, and mathematics not science but its own thing.

-----




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact

Search: