Re: Astrology & Statistics

"Ray Murphy" <raymur@[EMAIL PROTECTED]
> wrote in message 
news:481d7a09@[EMAIL PROTECTED]
> "Richard Nolle" <nospam@[EMAIL PROTECTED]
> wrote in message
> news:wY2Tj.179773$nr1.87183@[EMAIL PROTECTED]
> RM: You fail to recognise what was being said. Eleven degrees of
> freedom only applies in cases where we are comparing 1 sun sign
> with 12. It's all very elementary 12 scores = 11 degrees of freedom.
> You simply cannot use 11 degrees of freedom if you don't even know
> what the other 10 scores are beacuse you canot even get a Chi square
> value to USE the 11 degrees of freedom.
>
>> The appropriate test must be
>> used or the results are meaningless. That's just the way statistical
>> testing works.
>
> RM: You still don't get it. We are not ~looking~ at 12 sun signs - we
> are only looking at two. To suggest otherwise is completely wrong.
> You won't even be able to provide an alternative accurate p-score
> for what I put forward..
>

Please don't take this personally. Statistics is impersonal. Rules of 
procedure are impersonal. Saying there are only two signs when there are
in 
fact twelve is EXACTLY what is meant by cherry-picking, whether you
realize 
it or not. Taking the two most extreme slots in a twelve-slot distribution

and declaring them to be significant without first testing to see whether 
the results for the whole twelve-slot distribution is significant is
EXACTLY 
what is meant by cherry picking. It's like rolling a di and analyzing the 
results only in terms of the highest and lowest counts per side - ignoring

the fact that there are six sides per di. It produces a meaningless
result.

--Richard

http://www.astropro.com/
phone/fax = 480-753-6261
consultations/orders = 800-527-8761

"Astrology is about time . . . what else is there?"