In article <[EMAIL PROTECTED]>,
dennis roberts <[EMAIL PROTECTED]> wrote:
>At 10:16 AM 4/26/01 -0500, Herman Rubin wrote:
>>A p-value tells me nothing of importance.
>i agree if this means practical and of benefit say to society
>> It is in no way
>>a measure of strength of evidence.
>are you saying p tells you nothing?
It tells me something; it tells me the probability
that, if the essentially impossible null hypothesis
happens to be exactly correct, a result "this far
out" would have happened. It does not tell me
how much evidence has been obtained for or against
the null hypothesis compared to any alternatives.
It is generally true that a smaller p value means
stronger evidence for any given situation, but if
one experiment gives a p value of .05 and another
of .01, and these have different sample sizes or
other differences which matter, I cannot say which
gives more evidence. It may even be that if the
null hypothesis or something close enough may be
true, it should be accepted with a very small p
value, and if the sample is small, one should act
as if the alternative is true with a p value of .5.
--
This address is for information only. I do not claim that these views
are those of the Statistics Department or of Purdue University.
Herman Rubin, Dept. of Statistics, Purdue Univ., West Lafayette IN47907-1399
[EMAIL PROTECTED] Phone: (765)494-6054 FAX: (765)494-0558
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
