Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: st: corcor


From   jverkuilen <jverkuilen@gc.cuny.edu>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: corcor
Date   Tue, 10 Mar 2009 12:51:20 -0400

As Nick already noted, this is usually done with Fisher's z because the sampling distribution of the correlation coefficient itelf is decidedly unpleasant when correlations are different than 0 (it is not pivotal if I recall right, and is notably skewed). Fisher's z is just arctanh and the asymptotic SE of a Fisher's z transform is 1/sqrt(n-3). 

James Steiger (Psychology, Vanderbilt) wrote a primer on comparing correlation coefficients. See his web page. The most recent version of the article I know of is: 

J. Steiger (2004). Comparing correlations. In A. Maydeu-Olivares and J.J. McArdle (eds.). Contempoary Psychometrics: A Fetschrift in Honor of Roderick P. McDonald. Mahwah, NJ: LEA.  

-----Original Message-----
From: jasonm@ucla.edu
To: "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Sent: 3/10/2009 3:35 AM
Subject: st: corcor

Are there any functions like corcor (H0: cor(x1x2)=cor(x1x3)) that test
cor(x1x2)=cor(x3x4)?
Thank you.

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   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