Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

Re: st: meta-analysis with correlation coefficients


From   Jonathan Sterne <[email protected]>
To   [email protected], [email protected]
Subject   Re: st: meta-analysis with correlation coefficients
Date   Tue, 15 Jun 2010 14:15:24 +0100

You would input the transformed correlation coefficient and its standard error (1/sqrt(n-3)) for each study.

Best wishes

Jonathan

--On 14 June 2010 20:37 +0200 John Antonakis <[email protected]> wrote:


Hi Jonathan:

I thanked you too soon.

Quick question:

Suppose my first r is .321 for an n of 53. The  the SE of r is .14142136.
The atanh (r to z' transformation) of r is 0.332762; thus, the CI is
0.60994745 to 0.05557572. The tanh of this (reverse the r to z'
transformation) is 0.544090107 to 0.055518572. Nick's program gives the
same thing:

. corrcii 53 0.321,

       n       r      95% confidence limits
      53      0.321      0.056     0.544

OK, this is all easy. However, which numbers do I input for -metan-,
specifically?

TIA,
John.

____________________________________________________

Prof. John Antonakis, Associate Dean Faculty of Business and Economics
Department of Organizational Behavior
University of Lausanne
Internef #618
CH-1015 Lausanne-Dorigny
Switzerland

Tel ++41 (0)21 692-3438
Fax ++41 (0)21 692-3305

Faculty page:
http://www.hec.unil.ch/people/jantonakis

Personal page:
http://www.hec.unil.ch/jantonakis
____________________________________________________



On 14.06.2010 13:40, Jonathan Sterne wrote:
You would simply need to use the usual formula for the standard error
of Fisher's transformation for the correlation coefficient
(1/sqrt(n-3)), then use metan on the transformed correlations and
their standard errors, then transform the summary estimate and 95% CI
back to the original scale.

Best wishes

Jonathan Sterne

--On 14 June 2010 02:33 -0400 statalist-digest
<[email protected]> wrote:

Date: Sun, 13 Jun 2010 23:20:47 +0200
From: John Antonakis <[email protected]>
Subject: st: meta-analysis with correlation coefficients

Hi:

Does anyone know of a Stata user-written ado file that does
meta-analysis with correlation coefficients as inputs (and not only
means and SDs)?

Thanks,
John.

____________________________________________________

Prof. John Antonakis, Associate Dean
Faculty of Business and Economics
Department of Organizational Behavior
University of Lausanne
Internef #618
CH-1015 Lausanne-Dorigny
Switzerland

Tel ++41 (0)21 692-3438
Fax ++41 (0)21 692-3305

Faculty page:
http://www.hec.unil.ch/people/jantonakis

Personal page:
http://www.hec.unil.ch/jantonakis
____________________________________________________





*
*   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/




----------------------

Jonathan Sterne

Department of Social Medicine
University of Bristol
Canynge Hall
39 Whatley Road
Bristol BS8 2PS
UK

Tel:    0117 928 7396
Fax:    0117 928 7325
E-mail: [email protected]
web:    www.epi.bris.ac.uk/staff/jsterne.htm

*
*   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 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index