[R] t.tests by unique groupes

[Peter Dalgaard]

Imri wrote:
> My data (slength) look like this:
> Plant Block Treat Genotype Source MPH
> 1       1     1     w   05-AZM sp 160            NA
> 2       2     1     w    12-50    463            NA
> 3       3     1     w    12-51 sp 150     0.0508982
> 4       5     1     w    29-42    567     0.9017094
> 5       6     1     w  KNG-KNG    811            NA
> 6       7     1     w    02-02      1            NA
> 
> Treat column has w or d for each Plant(row). I wont to do t.test of
> MPH~Treat for each of the genotypes (128 total). I tried :
> 
> t.tests<-by(slength, slength[,4], function(x)
> t.test(slength[,6]~slength[,3])) 
> 
> but it is not what I need.
> Do you have any ideas?
>   Thanks
>          Imri
>  

Your function(x)... doesn't depend on x! Try

...t.test(MPH~Treat, data=x)

-- 
    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