DataTaunew | comments | leaders | submitlogin
Not Even Scientists Can Easily Explain P-values (fivethirtyeight.com)
17 points by platz 3045 days ago | 3 comments


1 point by abe_duarte 3043 days ago | link

As a definition put in words, the one from Stuart Buck in this article is very good. I'm saving it.

-----

1 point by kameo 3044 days ago | link

I don't know that this is necessarily a bad thing. Some topics just naturally do not have a explanation in everyday language. I'd be amazed if you could really explain the nuts and bolts of P vs NP or topological spaces to a random person on the street. Conveying the general idea is possible, but any more than that is asking too much.

-----

1 point by debrouwere 3045 days ago | link

I'm a little bit surprised to read this from the same person who wrote the very eloquent http://fivethirtyeight.com/features/science-isnt-broken/.

P-values are a flawed concept and even people who think they understand them usually don't, but I don't get why a science reporter would want to revel in their ignorance and say "yeah, I mean, I don't understand it either, I give up!"

-----




RSS | Announcements