I've been doing Dunns tests as post-hoc and corrections for my thesis data and I've run into an issue. I have 67 variables and 3 groupings that I'm running through, so loops just easily solve the tons of data analysis.
ind1 | var1(6) | var... | var67(73) |
---|---|---|---|
a | value | value | value |
b | value | value | value |
c | value | value | value |
I've done a normal kruskal.test for all my variables and groups with no issue, but "dunnTest" and dunn_test" both stop at 27 variables
library(FSA)
Dunn_df = list()
for(i in names(df[,6:73])){
Dunn_df[[i]] <- dunnTest(formula(paste(i, "~ ind1")), data = df, method = "bonferroni")
}
and
library(rstatix)
Dunn_df = list()
for(i in names(df[,6:73])){
Dunn_df[[i]] <- dunn_test(formula(paste(i, "~ ind1")), data = df, p.adjust.method="BH")
}
both output a nested list that I've able to read with print(Dunn_df), but only do the dunn test on columns 6:32.
Without any reproducible example it is very difficult to help you. Using the example data
mtcars
, the function works just fine on a large number of variables, e.g.