ie, we want our sample data to allow us to REJECT the null ... nullify the null ...
null does not mean zero ... though it can in some cases
if you are testing the null that the population variance for IQ is 225 ... the null is clearly not 0
At 02:20 PM 3/20/2003, Jerry Dallal wrote:
"Richard M." wrote:
>
> I know that it is possible to reject an hypothesis that is not equal to zero
> (like a difference of 0.5 standard deviation or a correlation less or equal
> to .20). But I wonder whether we still call such an hypothesis the "NULL"
> hypothesis.
Yes. We call it the null. Twenty-five years ago I did an unsuccessful literature search to find out why the null hypothesis was called the null. I wish I'd had the good sense to ask Frank Anscombe and alas now he's no longer with us. Stephen Stigler might know. . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
_________________________________________________________
dennis roberts, educational psychology, penn state university
208 cedar, AC 8148632401, mailto:[EMAIL PROTECTED]
http://roberts.ed.psu.edu/users/droberts/drober~1.htm
. . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
