luigi marongiu
2013-Oct-21 23:08 UTC
[R] post hoc test to a significant Kruskal-Wallis test package asbio function pairw.kw()
Dear all,
I am trying to apply a post hoc test to a significant Kruskal-Wallis test.
I eventually found the function ‘pairw.kw()’ of the package “asbio”. This
function has the following syntax:
pairw.kw(y, x, conf)
Arguments
y The response variable. A vector of quantitative responses.
x An explanatory variable. A vector of factor levels.
conf The level of desired confidence, 1 - P(type I error).
The example reported indicated 2 vectors of length 12 arranged in a data
frame:
rye.data <- data.frame(rye = c(50, 49.8, 52.3, 44.5, 62.3, 74.8, 72.5, 80.2,
47.6, 39.5, 47.7,50.7), nutrient = factor(c(rep(1, 4), rep(2, 4), rep(3,
4))))
with(rye.data, pairw.kw(y = rye, x = nutrient, conf = .95))
I tried to arrange the data in this way, as well as using my own dataframe
obtained from the read.table() function and directly with two vectors, but
I always obtained an error. Essentially the two vectors are not of the same
length; this is incorrect since there are both 64 elements.
Is there somebody who could tell me how to use this function?
Many thanks,
Luigi
EXAMPLE GIVEN
Here I am using the vector example.
### define vectors
x<-c( 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 3,
3, 3, 3, 3, 3, 3, 3, 3, 4, 4, 4, 4,
4, 4, 4, 4, 4)
y<-c( 0.002515324, 0.005128344, 0.008474943, 0.009992525,
0.01460982, 0.020567249, 0.069597233, 0.192428254,
0.636910481, 0.673830995, 4.354055891, 4.847755691,
0.403135211, 0.539626363, 0.892722111, 0.993636963,
1.289747804, 3.766986726, 5.623002684, 8.52186547,
17.24684431, 19.84624792, 25.9529251, 35.64632645,
0.437492775, 0.594863773, 0.64377689, 0.88033664,
3.199906591, 5.745723357, 6.339129057, 6.583851172,
11.27921212, 13.01230334, 14.61151433, 26.68041064,
52.42360789, 71.88062442, 91.34132781, 115.1038548,
0.059333604, 0.068746428, 0.1473799, 0.220983138,
0.959677864, 1.21197612, 2.860983172, 4.402992963,
5.122748306, 8.711530662, 25.6245157, 44.0798333,
45.24736747, 66.4490478, 86.60427738, 0.10348061,
0.390758249, 0.634667258, 1.122594755, 2.726868877,
4.15194739, 5.361175006, 8.825635885, 45.75640999)
### confirm length:
length(x) # 64
length(y) # 64
### KRUSKAL-WALLIS to test inter-group differences
kruskal.test(y, x) # p-value = 0.0008701
### Multiple pairwise comparison procedure to accompany a Kruskal-Wallis
test
pairw.kw(y, x, conf = .95)
# Error in outer(levels(x), levels(x), function(x1, x2) { :
# dims [product 0] do not match the length of object [1]
[[alternative HTML version deleted]]
arun
2013-Oct-22 03:12 UTC
[R] post hoc test to a significant Kruskal-Wallis test package asbio function pairw.kw()
Hi,
Try:
pairw.kw(y,factor(x),conf=.95)
A.K.
On Monday, October 21, 2013 7:10 PM, luigi marongiu <marongiu.luigi at
gmail.com> wrote:
Dear all,
I am trying to apply a post hoc test to a significant Kruskal-Wallis test.
I eventually found the function ?pairw.kw()? of the package ?asbio?. This
function has the following syntax:
pairw.kw(y, x, conf)
Arguments
y? ? The response variable. A vector of quantitative responses.
x? ? An explanatory variable. A vector of factor levels.
conf? ? The level of desired con?dence, 1 - P(type I error).
The example reported indicated 2 vectors of length 12 arranged in a data
frame:
rye.data <- data.frame(rye = c(50, 49.8, 52.3, 44.5, 62.3, 74.8, 72.5, 80.2,
47.6, 39.5, 47.7,50.7), nutrient = factor(c(rep(1, 4), rep(2, 4), rep(3,
4))))
with(rye.data, pairw.kw(y = rye, x = nutrient, conf = .95))
I tried to arrange the data in this way, as well as using my own dataframe
obtained from the read.table() function and directly with two vectors, but
I always obtained an error. Essentially the two vectors are not of the same
length; this is incorrect since there are both 64 elements.
Is there somebody who could tell me how to use this function?
Many thanks,
Luigi
EXAMPLE GIVEN
Here I am using the vector example.
### define vectors
x<-c(? ? 0,? ? 0,? ? 0,? ? 0,? ? 0,? ? 0,? ? 0,? ? 0,? ? 0,? ? 0,? ? 0,
0,? ? 1,? ? 1,? ? 1,? ? 1,? ? 1,? ? 1,? ? 1,? ? 1,? ? 1,? ? 1,? ? 1,
1,? ? 2,? ? 2,? ? 2,? ? 2,? ? 2,? ? 2,? ? 2,? ? 2,? ? 2,? ? 2,? ? 2,
2,? ? 2,? ? 2,? ? 2,? ? 2,? ? 3,? ? 3,? ? 3,? ? 3,? ? 3,? ? 3,? ? 3,
3,? ? 3,? ? 3,? ? 3,? ? 3,? ? 3,? ? 3,? ? 3,? ? 4,? ? 4,? ? 4,? ? 4,
4,? ? 4,? ? 4,? ? 4,? ? 4)
y<-c(? ? 0.002515324,? ? 0.005128344,? ? 0.008474943,? ? 0.009992525,
0.01460982,? ? 0.020567249,? ? 0.069597233,? ? 0.192428254,
0.636910481,? ? 0.673830995,? ? 4.354055891,? ? 4.847755691,
0.403135211,? ? 0.539626363,? ? 0.892722111,? ? 0.993636963,
1.289747804,? ? 3.766986726,? ? 5.623002684,? ? 8.52186547,
17.24684431,? ? 19.84624792,? ? 25.9529251,? ? 35.64632645,
0.437492775,? ? 0.594863773,? ? 0.64377689,? ? 0.88033664,
3.199906591,? ? 5.745723357,? ? 6.339129057,? ? 6.583851172,
11.27921212,? ? 13.01230334,? ? 14.61151433,? ? 26.68041064,
52.42360789,? ? 71.88062442,? ? 91.34132781,? ? 115.1038548,
0.059333604,? ? 0.068746428,? ? 0.1473799,? ? 0.220983138,
0.959677864,? ? 1.21197612,? ? 2.860983172,? ? 4.402992963,
5.122748306,? ? 8.711530662,? ? 25.6245157,? ? 44.0798333,
45.24736747,? ? 66.4490478,? ? 86.60427738,? ? 0.10348061,
0.390758249,? ? 0.634667258,? ? 1.122594755,? ? 2.726868877,
4.15194739,? ? 5.361175006,? ? 8.825635885,? ? 45.75640999)
### confirm length:
? ? length(x)? ? # 64
? ? length(y)? ? # 64
### KRUSKAL-WALLIS to test inter-group differences
? ? kruskal.test(y, x)? ? # p-value = 0.0008701
### Multiple pairwise comparison procedure to accompany a Kruskal-Wallis
test
? ? pairw.kw(y, x, conf = .95)
# Error in outer(levels(x), levels(x), function(x1, x2) { :
# dims [product 0] do not match the length of object [1]
??? [[alternative HTML version deleted]]
______________________________________________
R-help at r-project.org 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.