[R] Rcorr

[Frank E Harrell Jr]

James Allsopp wrote:
> No, that's made no difference, sorry.

Sorry I forgot to check the print method for rcorr. If P<.0001 it prints 
as 0.  To print under your control print the object $P from the list 
created by rcorr:

r <- rcorr(. . .)
r$P

Frank

> 
> Frank E Harrell Jr wrote:
>> James Allsopp wrote:
>>> Hi,
>>> I've just run an rcorr on some data in Spearman's mode and it's just
>>> produced the following values;
>>>       [,1]  [,2]
>>> [1,]  1.00 -0.55
>>> [2,] -0.55  1.00
>>>
>>> n= 46
>>>
>>>
>>> P
>>>      [,1] [,2]
>>> [1,]       0
>>> [2,]  0
>>>
>>> I presume this means the p-value is lower than 0.00005, but is there any
>>> way of increasing the number of significant figures used? How should I
>>> interpret this value?
>>>
>>> Cheers
>>> Jim
>>>
>> Try options(digits=15) before running rcorr().
>>
> 


-- 
Frank E Harrell Jr   Professor and Chair           School of Medicine
                      Department of Biostatistics   Vanderbilt University