Thank you! It's what I want :)
But I can't make it work. I'm working with R version 2.7.1 under Debian
GNU/Linux. Let me paste my results:
> pairwise.wilcox.test(SAND,Organ,p.adj="bonf") -> a
Pairwise comparisons using Wilcoxon rank sum test
data: SAND and Organ
FlowerA FlowerB FlowerC FlowerD LeafA
FlowerB 0.00062 - - - -
FlowerC 0.00062 0.00432 - - -
FlowerD 0.00062 0.00062 0.00062 - -
LeafA 0.00614 0.00614 0.00614 0.00614 -
LeafC 0.00062 0.00062 0.00062 0.00062 0.32425
P value adjustment method: bonferroni
Warning messages:
1: In wilcox.test.default(xi, xj, ...) :
cannot compute exact p-value with ties
2: In wilcox.test.default(xi, xj, ...) :
cannot compute exact p-value with ties
3: In wilcox.test.default(xi, xj, ...) :
cannot compute exact p-value with ties
4: In wilcox.test.default(xi, xj, ...) :
cannot compute exact p-value with ties
5: In wilcox.test.default(xi, xj, ...) :
cannot compute exact p-value with ties
> multcompLetters(a,Letters=c(letters,LETTERS,"."))
Error in do.call(compare, list(x, threshold)) :
(list) object cannot be coerced to 'double'
I though it was because the bad format of the argument "a" (I think the most
similar input may be "A square, symmetric matrix with row names"), so I
tried with one valid input produced by a Tukey test (valid as input but not
as an analysis of my data because its non-normality) with the aim of
checking if this was the problem.
> TukeyHSD(aov(SAND~Organ)) -> b
Tukey multiple comparisons of means
95% family-wise confidence level
Fit: aov(formula = SAND ~ Organ)
$Organ
diff lwr upr p adj
FlowerB-FlowerA -10.392222 -12.12877102 -8.6556734 0.0000000
FlowerC-FlowerA -14.692222 -16.42877102 -12.9556734 0.0000000
FlowerD-FlowerA -5.280000 -7.01654880 -3.5434512 0.0000000
LeafA-FlowerA -3.478889 -5.21543769 -1.7423401 0.0000044
LeafC-FlowerA -2.508889 -4.24543769 -0.7723401 0.0011579
FlowerC-FlowerB -4.300000 -6.03654880 -2.5634512 0.0000000
FlowerD-FlowerB 5.112222 3.37567342 6.8487710 0.0000000
LeafA-FlowerB 6.913333 5.17678453 8.6498821 0.0000000
LeafC-FlowerB 7.883333 6.14678453 9.6198821 0.0000000
FlowerD-FlowerC 9.412222 7.67567342 11.1487710 0.0000000
LeafA-FlowerC 11.213333 9.47678453 12.9498821 0.0000000
LeafC-FlowerC 12.183333 10.44678453 13.9198821 0.0000000
LeafA-FlowerD 1.801111 0.06456231 3.5376599 0.0379548
LeafC-FlowerD 2.771111 1.03456231 4.5076599 0.0002720
LeafC-LeafA 0.970000 -0.76654880 2.7065488 0.5655196
> multcompLetters(b,Letters=c(letters,LETTERS,"."))
Error in do.call(compare, list(x, threshold)) :
(list) object cannot be coerced to 'double'
Además: Warning message:
In if (class(x) == "dist") x <- as.matrix(x) :
la condición tiene longitud > 1 y sólo el primer elemento será usado (I'm
sorry for the spanish warning, it means that the condition has length>1 so
only the first element will be used)
Any idea of what was the mistake I did?
Thank you again!
[[alternative HTML version deleted]]
______________________________________________
[email protected] 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.
