How to change map_signif_level

[jaison75]

I am trying to change the significance intervals and the associated *s for the output. I thought map_signif_level should do the trick. Am I wrong? Is there a way to do it. I would appreciate any help.
Thanks
JJ

bonferroni_n <- 4828
symnum.args <- list(cutpoints = c(0, 0.05/bonferroni_n, 0.05, 1), symbols = c("***", "*", "ns"))

Output of compare_means works when I try the following
compare_means(len ~ dose, data = ToothGrowth,method="t.test", symnum.args = symnum.args)

I thought the following should work for the stat_compare_means, but it doesnot.
ggboxplot(ToothGrowth, x = "dose", y = "len" ) +
stat_compare_means( comparisons = my_comparisons,
method = "t.test",
label="p.signif",
paired = TRUE,
map_signif_level = c("***"=0.05/bonferroni_n , "**"=0.001, "*"=0.05) )

[kassambara]

Hi,

In the current version the function stat_compare_means() don't support the argument map_signif_level. I'll work on that.

Thanks

[elijahrockers]

As far as I can tell, the significance symbols are still hard coded into stat_compare_means(). Any movement on this? I'd love to be able to control the asterisks on my graphs.

[elijahrockers]

See for a solution: https://stackoverflow.com/questions/47839988/indicating-significance-with-ggplot2-in-a-boxplot-with-multiple-groups

[kassambara]

After installing the latest developmental version (devtools::install_github("kassambara/ggpubr")), type this:

df <- ToothGrowth
df$dose <- factor(df$dose)


comparisons <- list(
  c("0.5", "1"), c("1", "2"),
  c("0.5", "2")
)

symnum.args <- list(
  cutpoints = c(0, 0.0001, 0.001, 0.01, 0.05, 1), 
  symbols = c("xxxx", "***", "**", "*", "ns")
  )

ggplot(df, aes(dose, len)) +
geom_boxplot()+
  stat_compare_means(
    comparisons = comparisons,
    label = "p.signif",
    symnum.args = symnum.args
    )

The output should look like this: