I'm not an advanced user of R.Anyway I just want to do a Wilcox test on two datasets:
library(coin)
GroupA = c(2,4,3,1,2,3,3,2,3,1)
GroupB = c(3,5,4,2,4,3,5,5,3,2)
g = factor(c(rep("GroupA", length(GroupA)), rep("GroupB", length(GroupB))))
v = c(GroupA, GroupB)
ss = wilcox_test(v ~ g, distribution="exact")
What I want is to get the Zvalue and compute the effect size. ss is a S4 object, when I print it I see it reports a Z value, but I can not find it in ss@statistic. The only slot that has the value which I expect is 'test statistic', but even if this is the z-value of the test when I do :
slot(ss, 'teststatistic')
I get the error :
no slot of name "teststatistic" for this object of class "ScalarIndependenceTest"
Can anyone please give me a hint?Thanks
The print equivalent for S4 classes is the show method, which can be inspected using getMethod. In this case,
ss
#
# Exact Wilcoxon-Mann-Whitney Test
#
#data: v by g (GroupA, GroupB)
#Z = -2.1095, p-value = 0.0385
#alternative hypothesis: true mu is not equal to 0
getMethod("show","ScalarIndependenceTest")
#Method Definition:
#
# function (object)
# {
# distname <- switch(class(object@distribution), AsymptNullDistribution = "Asymptotic",
# ApproxNullDistribution = "Approximative", ExactNullDistribution = "Exact")
# RET <- list(statistic = setNames(object@statistic@teststatistic,
# nm = "Z"), p.value = object@distribution@pvalue(object@statistic@teststatistic),
# alternative = object@statistic@alternative, data.name = varnames(object@statistic),
# method = paste(distname, object@method))
#...
#...
# }
ss@statistic@teststatistic
# GroupA
#-2.109531
ss@distribution@pvalue(ss@statistic@teststatistic)
#[1] 0.03850484