Show exact p value in comparisons

[Hemantcnaik]

In example with below

ggplot(diamonds, aes(x = cut, y = carat)) +
  geom_boxplot(aes(fill = color)) +
  geom_signif(comparisons = list(
    c("Fair", "Good"),
    c("Very Good", "Ideal")
  )) +
  facet_wrap(~color) +
  ylim(NA, 6.3)

P value mentioned in D section p <2.22e-16 I want show exact values instead of this value

Thank you

[IndrajeetPatil]

Are you sure you want to such small value?
Because the exact value is going to be smaller than 0.0000000000000002220000000000000081787.

[Hemantcnaik]

p < 2.2e-16 as the p value would indicate a significant result, meaning that the actual p value is even smaller than 2.2e-16 i just want exact value show in figures. In papers which I checked I am not getting anyone representing like this so want to represent exact values

[IndrajeetPatil]

It's basically printing what the Wilcox test produces, but, yes, we can probably try to show the exact p-value if available:

library(ggplot2)
m <- wilcox.test(carat ~ cut, data = dplyr::filter(diamonds, color == "D", cut %in% c("Very Good", "Ideal")))

m
#> 
#>  Wilcoxon rank sum test with continuity correction
#> 
#> data:  carat by cut
#> W = 2554244, p-value < 2.2e-16
#> alternative hypothesis: true location shift is not equal to 0

m$p.value
#> [1] 2.122043e-25
broom::tidy(m)   
#> # A tibble: 1 × 4
#>   statistic  p.value method                                          alternative
#>       <dbl>    <dbl> <chr>                                           <chr>      
#> 1   2554244 2.12e-25 Wilcoxon rank sum test with continuity correct… two.sided

^{Created on 2024-05-14 with reprex v2.1.0}