[R] in cor.test, difference between exact=FALSE and exact=NULL

[Peter Dalgaard]

Andrew Yee wrote:
> Pardon my ignorance, but is there a difference in cor.test between
> exact=FALSE and exact=NULL when method=spearman?
>
> Take for example:
>
> x<-c(1,2,2,3,4,5)
> y<-c(1,2,2,10,11,12)
> cor.test(x,y, method="spearman", exact=NULL)
>
> This gives an error message,
> Warning message:  Cannot compute exact p-values with ties in:
> cor.test.default(x, y, method = "spearman", exact = NULL)
>
> However, when exact is changed to FALSE, this seems to run okay.
>
> cor.test(x,y, method="spearman", exact=FALSE)
>
> Question:  should this be exact = FALSE in the documentation and/or the code?
>
>   
No. The default is indeed NULL.

This implies that calculation of exact p-values will be attempted, and
when there are ties you get a warning (NB: not error) message.  Setting
exact=FALSE, no attempt is made and no warning is given.
> Thanks,
> Andrew
> MGH Cancer Center
>
> ______________________________________________
> R-help at 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
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)                  FAX: (+45) 35327907