I am starting this thread as a response to the blog by Captain Refsmat. as I do not get on well with blogs.

 

Two thoughts

 

1) There is another use of statistics that is totally usually ignored in such discussions and revalations. That of engineering statistics, where conditions are carfully controlled to give validity to the technique. This leads to the articles giving the false impression that statistical testing is generally invalid.

 

2) The is another recent populist book on the subject, which is the outcome of a BBC program series into the subject.

 

The Tiger That Isn't

 

Michael Blastland & Andrew Dilnot

 

go well

 

Edit

 

http://www.scienceforums.net/topic/68060-blog-post-capn-refsmmat-statistics-done-wrong-how-scientists-abuse-statistics/page__pid__693980#entry693980

 

Sorry Panic, you were quite right.

I was going to do that and even had the link still on the clipboard, but I forgot.

Edited August 1, 201213 yr by studiot

Be so decent as to at least link to it. Right now, this is meaningless.

 

Also, if this is an SFN blog, you can actually reply on the blog website (comments).