Re: [R] Pearson and Spearman correlation coeffcients matrix

From: Jonathan Baron <>
Date: Sun 03 Jul 2005 - 23:33:15 EST

I think "associated statistics" means p-levels.

This post tells how to do it and gives and example.

But I think that nobody wants to make a function for this because it would encourage bad behavior, i.e., snooping through a correlation matrix looking for "significant" correlations and then making up a story about how the 5% of them that were significant at p<.05 were really planned tests.


On 07/03/05 08:12, Douglas Bates wrote:
 On 7/3/05, Nuno Soares <> wrote:
> Hi everyone,
> I've been trying to find a function that outputs the Pearson and/or Spearman
> correlation coefficients for several variables with the associated
> statistics in one single table/matrix. For what I've been able to understand
> the Stats package is only able to compute these coeficients/statistics only
> in defined pairs. This becomes time consuming when we want to determine
> these measures with, say, 10 variables or more. Does anyone knows a solution
> to this?

 I believe that the cor function already does what you want. Check  




 If the first argument to cor is a matrix, it returns the correlation  matrix of all the columns. mailing list  PLEASE do read the posting guide!
Jonathan Baron, Professor of Psychology, University of Pennsylvania
Home page:
R search page:

______________________________________________ mailing list
PLEASE do read the posting guide!
Received on Sun Jul 03 23:35:43 2005

This archive was generated by hypermail 2.1.8 : Fri 03 Mar 2006 - 03:33:11 EST