R help - Jul 2012 - Significance of correlation

Hi,

I understand that to test the significance of correlation between two
PAIRED variables, the function, cor.test () can be used. However, in my
case, I have tested the correlation (i.e., Correlation Coefficient, r)
between two independent (i.e, different) variables, and now I wish to test
for the statistical significance of the correlation. Could you please
suggest how I should do that.

Thank you.

Kind regards,
Chintanu

	[[alternative HTML version deleted]]

If you have pearson correlation coefficients r calculated from n data points
elsewhere, you can use a t test on
t.r <- r * sqrt( (n-2)/(1-r^2) )

You'll need to use pt() to get your p value from t.r, though.

S Ellison

________________________________________


I understand that to test the significance of correlation between two
PAIRED variables, the function, cor.test () can be used. However, in my
case, I have tested the correlation (i.e., Correlation Coefficient, r)
between two independent (i.e, different) variables, and now I wish to test
for the statistical significance of the correlation. Could you please
suggest how I should do that.

Thank you.

Kind regards,
Chintanu

        [[alternative HTML version deleted]]

______________________________________________
R-help at r-project.org 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.

*******************************************************************
This email and any attachments are confidential. Any use...{{dropped:8}}

On 31/07/12 21:13, Chintanu wrote:
> Hi,
>
> I understand that to test the significance of correlation between two
> PAIRED variables, the function, cor.test () can be used. However, in my
> case, I have tested the correlation (i.e., Correlation Coefficient, r)
> between two independent (i.e, different) variables, and now I wish to test
> for the statistical significance of the correlation. Could you please
> suggest how I should do that.
In a word, don't.  If two random variables are independent then they
are necessarily (and obviously) uncorrelated.

Note that "independent" is not synonymous with "different".

     cheers,

         Rolf Turner