Dear Donglei Hu,
If you have two correlation coefficients, you may try cordif {multilevel}
and cordif.dep {multilevel} for the independent correlations and for the
dependent correlations, respectively. However, they are both based on the
sampling distribution of correlation coeficient. A better approach may be
the one based on the Fisher z transformation as suggested. The papers by
Olkin and Finn (1995) and Steiger (1980) may be relevant for you.
If you have more than two independent correlation coefficients,
meta-analysis may be a better choice. You may also choose between the
approaches based on correlations (Hunter and Schmidt) or Fisher z
transformation (Hedges and Olkin). If the correlations are dependent,
structural equation modeling (SEM) is a more convenient approach (e.g.,
Cheung & Chan, 2004).
Cheung, M.W.L., & Chan, W. (2004). Testing dependent correlation
coefficients via structural equation modeling. Organizational Research
Methods, 7, 206-223.
Olkin, I., & Finn, J. D. (1995). Correlation redux. Psychological Bulletin,
118, 155-164.
Steiger, J. H. (1980). Tests for comparing elements of a correlation matrix.
Psychological Bulletin, 87, 245-251.
Regards,
Mike
--
-----------------------------------------------------------
Mike W.L. Cheung
Department of Psychology
National University of Singapore
Homepage: http://courses.nus.edu.sg/course/psycwlm/internet/
-----------------------------------------------------------
On 19 Sep 2006 01:22:47 +0200, Peter Dalgaard <[EMAIL PROTECTED]>
wrote:
>
> "David Barron" <[EMAIL PROTECTED]> writes:
>
> > Is cor.test() in the stats packages what you mean?
>
> No, he wants to compare two correlation coefficients, not test that
> one is zero. That's usually a misguided question, but if need be, the
> Fisher z transform atanh(r) can be used to convert r to an
> approximately normal variate with a known variance 1/(N-3) and
> comparing r1 and r2 from two independent samples is straightforward.
> The correlated case (like cor(x,y) vs cor(x,z)) is more complicated.
>
>
>
> > On 18/09/06, Hu, Donglei <[EMAIL PROTECTED]> wrote:
> > > Hi,
> > >
> > >
> > >
> > > I calculated a few correlation coefficients. Now I want to know
> whether
> > > they are different from each other. Is there an R package that can do
> > > such a comparison? Thanks for any suggestion.
> > >
> > >
> > >
> > > Best,
> > >
> > > Donglei Hu
> > >
> > > Department of Medicine
> > >
> > > UCSF
> > >
> > >
> > > [[alternative HTML version deleted]]
> > >
> > > ______________________________________________
> > > R-help@stat.math.ethz.ch mailing list
> > > https://stat.ethz.ch/mailman/listinfo/r-help
> > > PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> > > and provide commented, minimal, self-contained, reproducible code.
> > >
> >
> >
> > --
> > =================================
> > David Barron
> > Said Business School
> > University of Oxford
> > Park End Street
> > Oxford OX1 1HP
> >
> > ______________________________________________
> > R-help@stat.math.ethz.ch mailing list
> > https://stat.ethz.ch/mailman/listinfo/r-help
> > PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> > and provide commented, minimal, self-contained, reproducible code.
> >
>
> --
> O__ ---- Peter Dalgaard Øster Farimagsgade 5, Entr.B
> c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K
> (*) \(*) -- University of Copenhagen Denmark Ph: (+45)
> 35327918
> ~~~~~~~~~~ - ([EMAIL PROTECTED]) FAX: (+45)
> 35327907
>
> ______________________________________________
> R-help@stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
[[alternative HTML version deleted]]
______________________________________________
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
