Re: st: meta-analysis with correlation coefficients

[Jonathan Sterne <Jonathan.Sterne@bristol.ac.uk>]

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 <john.antonakis@unil.ch> 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
> > <owner-statalist@hsphsun2.harvard.edu> wrote:
> >
> > > Date: Sun, 13 Jun 2010 23:20:47 +0200
> > > From: John Antonakis <john.antonakis@unil.ch>
> > > 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: jonathan.sterne@bristol.ac.uk
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/