[R] digit precision in p value of rcorr

[Rui Barradas]

Hello,

You could post to R-Help, maybe it will of use to others. And since it's 
a follow-up I'll post it there.

To get more (or less) digits use the respective argument to print().

Var <-
structure(list(D.Prime = c(0.17234, 0.14399, 0.14626, 0.035147,
0.058957), T.statistics = c(4.9268, 2.892, 2.6428, 1.1124, 2.7237
)), .Names = c("D.Prime", "T.statistics"), class = "data.frame",
row.names = c("1", "2", "3", "4", "5"))


library(Hmisc)
rc <- rcorr(as.matrix(Var), type="pearson")

print(rc$P, digits = 10)

Hope this helps,

Rui Barradas

Em 12-09-2012 19:39, Jason Love escreveu:
> Hi all,
> Sorry about posting a really novice question.
> I was able to run rcorr after converting the list to a matrix by your help.
> I'm though wondering if there is any way to find out an exact p value as
> the output only gave me 0 for P value as shown below.
> I've added options(digits=10), which doesn't seem to help at all. Any help
> would be appreciated.
>
>
> P
>               D Prime T statistics
> D Prime               0
> T statistics  0
>
>
> On Wed, Sep 12, 2012 at 2:19 PM, Jason Love <jason.love1492 at gmail.com>wrote:
>
>> thanks all for the prompt answer.
>> Yes, I need to go through the R tutorial rather than learning a snippet of
>> codes from googling.
>>
>>
>> On Wed, Sep 12, 2012 at 11:32 AM, Rui Barradas <ruipbarradas at sapo.pt>wrote:
>>
>>> Hello,
>>>
>>> The input must be a matrix, not a list (or its special case data.frame).
>>>
>>> Var <- read.table(text="
>>>    D.Prime    T.statistics
>>>
>>> 1    1.7234e-01     4.926800
>>> 2    1.4399e-01     2.892000
>>> 3    1.4626e-01     2.642800
>>> 4    3.5147e-02     1.112400
>>> 5    5.8957e-02     2.723700
>>> ", header=TRUE)
>>>
>>> # library(Hmisc)
>>> rc <- rcorr(as.matrix(Var), type="pearson")
>>> # from recommended package stats
>>> ct <- cor.test(Var$D.Prime, Var$T.statistics, method = "pearson")
>>>
>>> rc$P
>>>                 D.Prime T.statistics
>>> D.Prime             NA    0.1101842
>>> T.statistics 0.1101842           NA
>>>
>>> ct$p.value
>>> [1] 0.1101842
>>>
>>> To the op: you should say which library you are using. Even if Hmisc is a
>>> very popular one.
>>>
>>> Hope this helps,
>>>
>>> Rui Barradas
>>>
>>>
>>> Em 12-09-2012 16:10, R. Michael Weylandt escreveu:
>>>
>>>   On Wed, Sep 12, 2012 at 4:04 PM, Jason Love <jason.love1492 at gmail.com>
>>>> wrote:
>>>>
>>>>> Hello,
>>>>> I'd like to test a significance of two variables in their correlation
>>>>> using
>>>>> rcorr, which gave me an error of format incompatibility.
>>>>> Below are the lines that I typed in the R window and let me know if
>>>>> anyone
>>>>> knows how to resolve this.
>>>>>
>>>>> Var=read.csv("03apr10ab_corr_**matrix_in_overlaps.csv",**header=F)
>>>>> colnames(Var)=c("D Prime","T statistics")
>>>>>
>>>>>            D Prime    T statistics
>>>>> 1    1.7234e-01     4.926800
>>>>> 2    1.4399e-01     2.892000
>>>>> 3    1.4626e-01     2.642800
>>>>> 4    3.5147e-02     1.112400
>>>>> 5    5.8957e-02     2.723700
>>>>>
>>>>>
>>>>> rcorr(Var, type="pearson")
>>>>>
>>>> Untested (because I'm still without respectable internet after a move)
>>>> I believe rcorr would rather have a matrix than a data.frame(), which
>>>> is what read.csv produces, so try
>>>>
>>>> Var <- as.matrix(Var)
>>>>
>>>> or
>>>>
>>>> rcorr(as.matrix(Var), type = "pearson")
>>>>
>>>>   Error in storage.mode(x) <- if (.R.) "double" else "single" :
>>>>>     (list) object cannot be coerced to type 'double'
>>>>>
>>>> This suggests that the input to rcorr is being converted to a double,
>>>> which isn't a valid storage.mode change for a list (= data frame).
>>>>
>>>> Cheers,
>>>> M
>>>>
>>>>            [[alternative HTML version deleted]]
>>>>> ______________________________**________________
>>>>> R-help at r-project.org mailing list
>>>>> https://stat.ethz.ch/mailman/**listinfo/r-help<https://stat.ethz.ch/mailman/listinfo/r-help>
>>>>> PLEASE do read the posting guide http://www.R-project.org/**
>>>>> posting-guide.html <http://www.R-project.org/posting-guide.html>
>>>>> and provide commented, minimal, self-contained, reproducible code.
>>>>>
>>>> ______________________________**________________
>>>> R-help at r-project.org mailing list
>>>> https://stat.ethz.ch/mailman/**listinfo/r-help<https://stat.ethz.ch/mailman/listinfo/r-help>
>>>> PLEASE do read the posting guide http://www.R-project.org/**
>>>> posting-guide.html <http://www.R-project.org/posting-guide.html>
>>>> and provide commented, minimal, self-contained, reproducible code.
>>>>
>>>