Why do I get an error when I use the "formula" as a keyword argument with wilcox.test in R? The docs say that it has a "formula" parameter.
df = data.frame(A=rnorm(10), D=sample(c('p','q'), 10, replace=T))
wilcox.test(data=df, A~D)
wilcox.test(data=df, formula=A~D)
> df = data.frame(A=rnorm(10), D=sample(c('p','q'), 10, replace=T))
> wilcox.test(data=df, A~D)
Wilcoxon rank sum test
data: A by D
W = 13, p-value = 1
alternative hypothesis: true location shift is not equal to 0
> wilcox.test(data=df, formula=A~D)
Error in wilcox.test.default(data = df, formula = A ~ D) :
argument "x" is missing, with no default
Your arguments are not in the correct order.
> wilcox.test(formula=A~D, data=df)
Wilcoxon rank sum test
data: A by D
W = 13, p-value = 0.6667
alternative hypothesis: true location shift is not equal to 0
42- pointed out that I was incorrect about the purpose of formula arguments, since without the optional data argument the data objects would be inherited from the environment. The function chooses the default method instead of the formula method because it does not see a formula argument in the first argument location. Otherwise, argument order would not matter.