[email protected] wrote:
> Alerted by a colleague, I  recommend an instructive if 
> depressing essay on the problematic use of statistics in science.
>
> http://www.sciencenews.org/view/feature/id/57091/title/Odds_ar
> e,_its_w or http://tinyurl.com/yh7sk7r
>
>   

Yikes. What a simultaneously bizarre and ordinary column. There is 
little here that hasn't been written about and widely known since the 
1960s. Except that the author sensationally attributes the problem to 
"statistics"rather than specifically to the null hypothesis significance 
test (NHST). The fact that he then advocates Bayesian statistic (which 
have also been widely written about, though not much used in psychology, 
since the 1960s) makes the point as effectively as I could.

Correct probabilistic reasoning can be difficult, to be sure (as 
Kahneman & Teversky; Gigerenzer, et al.; etc. have long shown us). But 
many kinds of mathematical and scientific reasoning are difficult. 
That's why we have special institutions with highly-paid instructors to 
teach them.  That doesn't make statistics a "mutant form of math" (an 
accusation that sounds more like it comes form the first half of the 
19th century, than form the start of the 21st). The rise of 
probabilistic reasoning was probably the greatest epistemological 
advance in history since the invention of calculus. Anyone who thinks 
otherwise should try doing, say, evolutionary theory or quantum 
mechanics without it. (Not to mention pretty well all of social and 
behavioral science, of course.)

Chris
-- 

Christopher D. Green
Department of Psychology
York University
Toronto, ON M3J 1P3
Canada

 

416-736-2100 ex. 66164
[email protected]
http://www.yorku.ca/christo/

==========================


---
You are currently subscribed to tips as: [email protected].
To unsubscribe click here: 
http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=1427
or send a blank email to 
leave-1427-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu

Reply via email to