How to do Pvalue significance labeling in R language?

Boxplots are one of the more common graphs in statistics. This article will describe how to simply compare the means of two or more groups and add significance markers.

Usually, according to the value of the significant p-value, it is divided into four categories:

（1）0.01≤p<0.05，*

（2）0.001≤p<0.01，**

（3）0.0001≤p<0.001，***

（4）p<0.0001, ****

Three methods of adding saliency markers are described next.

Method 1 - Add manually¶

1: Create the data and draw a boxplot.

library(ggplot2)
ggplot(iris, aes(x=Species, y=Sepal.Length))+geom_boxplot()

2: Manually add saliency markers.

library(ggplot2)
### Set the position of the four points of the horizontal line
df2 <- data.frame(a = c(2,2,3,3), b = c(8,8.1,8.1,8))
ggplot(iris, aes(x=Species, y=Sepal.Length))+
   geom_boxplot()+
   ### Draw the line according to the set position
   geom_line(data = df2, aes(x = a, y = b)) +
   ### Add salient marker information
   annotate("text", x = 2.5, y = 8.2, label = "***", size = 8)

The disadvantage of this method is that when adding significance manually, it is necessary to confirm the size of the significant P value.

Method 2 - Using ggsignif¶

1: Download and install the ggsignif package.

install.packages("ggsignif")

2: Create the data and draw a boxplot.

library(ggplot2) 
library(ggsignif)
ggplot(iris, aes(x=Species, y=Sepal.Length)) +
  geom_boxplot()

3: Use the ggsignif R package to add saliency markers.

library(ggplot2)
library(ggsignif)
ggplot(iris, aes(x=Species, y=Sepal.Length)) +
geom_boxplot() +
### Add column information for pairwise comparison
geom_signif(comparisons = list(c("versicolor", "virginica")),
### P value < 0.05, then display
map_signif_level=TRUE)

Method 3 - Using ggpubr¶

1: Download and install the ggpubr package.

install.packages("ggpubr")

2: Create the data and draw a boxplot.

library(ggpubr)
### Add list of pairwise comparisons
my_comparisons <- list(c("setosa", "versicolor"), c("setosa", "virginica"), c("versicolor", "virginica"))
ggboxplot(iris, x="Species", y="Sepal.Length", color = "Species", palette = "jco", add = "jitter")

3: Use the ggpubr R package to add significance markers.

library(ggpubr)
my_comparisons <- list(c("setosa", "versicolor"), c("setosa", "virginica"), c("versicolor", "virginica"))
ggboxplot(iris, x="Species", y="Sepal.Length",color = "Species",palette = "jco", add = "jitter")+
   ### Add statistical results for multiple group comparisons
   stat_compare_means(label.y = 9.5)+
   ### Add saliency marker location information for each pairwise comparison
   stat_compare_means(comparisons=my_comparisons, label.y = c(7.6, 8.4, 8.0), label = "p.signif")

This R package can increase the p-value value between multiple groups, and can also increase the between-group comparison for a specified group.