at the bottom
dennis roberts wrote:
>
> At 11:26 AM 4/11/00 -0400, you wrote:
>
> >dennis roberts wrote:
> > >
> > > this was not about a difference in rhos .. just the rho singly from that
> > > population ...
> > >
> >
> >
> >
> >
> >It can be framed similarly replace mu1-mu2 with rho
> >
> >If the null hypothesis is H0:rho=0
> >and the alternative is H1:rho>0
> >
> >What does the test say about rho if we reject H0 at level
> >alpha(say at the magical 0.05)? Not much on its own.
>
> that's precisely my point ... what you know about rho is essentially zip
>
> > However, what
> >if we plan a statistical experiment as follows:
>
> in other words here ... the significance test as framed, the normal null,
> ... is useless
>
> > ^^^^
> >Given a desired power of 0.90 we would need a sample
> >of size 33 to detect a rho=.5 different from 0
> >at level alpha=0.05.
> >
> >We run the experiment with a sample of size 33 and
> >we reject H0. Now what can we infer about
> >rho in terms of probabilities? Quite
> >a bit more than just running an experiment without
> >considering power.
>
> like what ... that .5 is more likely than 0? but ... could it be .3? or .2
> ... ?
> does this mean that it is CLOSER to .5 than 0?
>
> >If we do not reject H0 given this design
> >what does it say about rho? Not that rho=0,
> >but rho=0 is a more likely scenario than
> >rho=.5(or >.5)
>
> same question ... does this mean it is closer to 0 than .5?
it would be easy to show whether rho is more likely closer to 0 or .5
given the parameters of a designed study
what if the confidence interval for rho was (0,0.5), is rho closer to 0
or .5
or .25 for that matter?
===========================================================================
This list is open to everyone. Occasionally, less thoughtful
people send inappropriate messages. Please DO NOT COMPLAIN TO
THE POSTMASTER about these messages because the postmaster has no
way of controlling them, and excessive complaints will result in
termination of the list.
For information about this list, including information about the
problem of inappropriate messages and information about how to
unsubscribe, please see the web page at
http://jse.stat.ncsu.edu/
===========================================================================
