Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Correlation coefficients and the CORTESTi program - again


From   Ricardo Ovaldia <ovaldia@yahoo.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Correlation coefficients and the CORTESTi program - again
Date   Sat, 11 Sep 2004 05:47:59 -0700 (PDT)

Dear statalist members;

I did not get any replies to my query, so I let me ask
it a different way:

The program -cortesti- compares the correlation
coefficients by constructing a Z-score of the
difference as follows:

      Z1=ln((1+corr1)/(1-corr1))/2
      Z2=ln((1+corr2)/(1-corr2))/2
       Z =(Z1-Z2)/sqrt((1/(n1-3))+(1/(n2-3)))

where corr1 and corr2 are the 2 correlation
coefficients and n1 and n2 the corresponding sample
sizes. Z is the test statistic use to compute a
p-value. Does anyone know of a different approach?

Thank you in advance
Ricardo.

Thank you

--- Ricardo Ovaldia <ovaldia@yahoo.com> wrote:

> Dear all,
> 
> I have been using -cortesti- to compare correlation
> coefficients from two independent samples. I am
> concerned that the test may be too conservative for
> moderate sample sizes (n=85). Is there a different
> way
> to test whether correlation coefficients from two
> independent samples are equal?
> 
> Thank you,
> Ricardo.
> 
> =====
> Ricardo Ovaldia, MS



		
_______________________________
Do you Yahoo!?
Shop for Back-to-School deals on Yahoo! Shopping.
http://shopping.yahoo.com/backtoschool
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index