[R] a bug with Student t-value??

[Peter Dalgaard]

Sunny Srivastava wrote:
> Hi Huaru,
> Are you sure you are looking at 0.95 and NOT 0.975?
> 
> as qt(0.975,9) = 2.262
> 
> generally tables are given for alpha/2 (alpha=0.05 in your case)
> significance.
> 
> Please check carefully.

Or, put differently, tables are often given for two-sided p-values 
(especially in the back pages of textbooks), whereas qt() is for one-sided.

I.e. the t distribution with 9 DF has a probability of 0.95 of falling 
BELOW 1.833 but also of falling BETWEEN -2.262 and +2.262.


-- 
    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