在分面ggplot中使用stat_compare_means()实现多组t检验对比

你遇到的问题核心是没有明确指定两两对比组合，同时统计量的位置未与点的dodge对齐，导致仅显示默认的整体检验结果。以下是修正后的代码和关键说明：

修正后的完整代码 #

set.seed(13)
df = data.frame(grouping = c(rep("A",24),rep("B",24),rep("C",24)),
                id = c(rep(1,12),rep(2,12),rep(3,12),rep(4,12),rep(5,12),rep(6,12)),
                subject = c(rep(c("One","Two","Three","Four"),72)),
                family = c(rep(c("Red","Red","Red","Red","Blue","Blue","Blue","Blue","Green","Green","Green","Green"),6)),
                percent = sample(1:100,288,replace=T),
                result = "Good")

library(ggplot2)
library(ggpubr)
library(dplyr)

df %>%
    ggplot(aes(x=subject,y=percent, group = interaction(subject,grouping))) + 
    geom_boxplot(outlier.shape = NA) + 
    geom_point(aes(color = grouping),size=1.75,position = position_dodge(width = 0.75))+
    scale_y_continuous(breaks = seq(0, 100, by = 25),limits = c(0,103)) + 
    facet_grid(result~family,switch = "y") + 
    scale_x_discrete("") +
    theme_minimal() +
    stat_compare_means(aes(group = grouping),
                       label="p.signif",
                       method = "t.test",
                       paired=F,
                       tip.length=0,
                       step.increase = 0.1,  # 增大步长避免标签重叠
                       comparisons = list(c("A","B"), c("A","C"), c("B","C")),  # 指定所有两两对比组合
                       position = position_dodge(width = 0.75))  # 与点的dodge位置对齐

关键修改点说明 #

• 指定comparisons参数：必须明确列出所有需要对比的组对，这是触发两两t检验的核心，否则stat_compare_means默认只会执行整体检验，因此仅显示一个"ns"结果。
• 对齐dodge位置：由于散点使用了position_dodge(width=0.75)，统计量标签需匹配相同的位置参数，否则会错位到分组中间，导致显示异常。
• 调整step.increase：增大步长可避免三组对比的显著性标签互相重叠，让每个对比的标签分层显示。

修改后，每个分面下的每个subject分组中，都会显示A vs B、A vs C、B vs C三组的t检验显著性结果。

内容的提问来源于stack exchange，提问作者Jeff238