Reputation: 1
how can i pull the z ?
Asymptotic Wilcoxon-Pratt Signed-Rank Test
data: y by x (pos, neg)
stratified by block
Z = 2.5113, p-value = 0.01203
alternative hypothesis: true mu is not equal to 0
Upvotes: 0
Views: 1341
Reputation: 119
Just to add to @BenBolker response, as soon as you find out what you want to extract, after using the wilcoxsign_test()
function and assuming the test results are stored in w
, you can use:
# Z value
statistic(w)
# p value
pvalue(w)
Upvotes: 0
Reputation: 226182
The wilcoxsign_test()
function is from the coin
package (this is useful information for people answering).
Most R functions work the way that @ChrisRuehlemann suggests, but wilcoxsign_test()
is an exception; it's stored as an S4 class, which makes things a little more obscure. You have to dig down with some combination of str()
and slotNames()
, and access individual elements via @
library(coin)
w <- wilcoxsign_test(mpg~vs,data=mtcars)
slotNames(w)
## [1] "parameter" "nullvalue" "distribution" "statistic"
## "estimates"
## [6] "method" "call"
w@statistic@teststatistic
## [1] 4.937168
Upvotes: 2
Reputation: 21400
The usual approach is to store the test as an object and extract the value of interest from there. For example, using the wilcox.test
:
test <- wilcox.test(rnorm(100), dnorm(1000))
test
Wilcoxon rank sum test with continuity correction
data: rnorm(100) and dnorm(1000)
W = 54, p-value = 0.9044
alternative hypothesis: true location shift is not equal to 0
Just type the name of your test object, add $
and run your cursor over it to see the options for what you can extract, for example:
test$statistic
W
54
Upvotes: 0