如何调整facet_wrap面板内箱线图水平间距及单独偏移均值/中位数标签？

使用ggplot2绘制带facet_wrap的箱线图时，均值和中位数标签出现重叠，需完成以下调整：

1. 针对4个场景（Female-No、Female-Yes、Male-No、Male-Yes）单独设置标签位置：
   • Female-No：均值标签4.9上移至均值线上方并右移，中位数标签4.1右移
   • Female-Yes：与Female-No设置相同
   • Male-No：中位数标签8.7右移至箱线图旁的中位数线侧，均值标签8.4下移至均值线下方并右移
   • Male-Yes：均值标签2.5下移至均值线下方并右移，中位数标签2.5水平右移
2. 调整facet_wrap单个面板内箱线图的水平间距，避免标签重叠

解决方案 #

核心思路 #

1. 增大箱线图水平间距：修改position_dodge的width参数拉开箱线图间距，同时缩小箱线图自身宽度，为标签预留空间。
2. 自定义标签位置：在统计量数据框中添加水平/垂直偏移列，通过case_when为每个场景单独设置偏移量，实现标签精准定位。

完整代码 #

library(tidyverse)

datas <- structure(list(Pain = structure(c(2L, 1L, 1L, 1L, 2L, 2L, 1L,
2L, 2L, 2L, 1L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 2L, 1L, 2L, 1L,
1L, 1L, 2L, 2L, 1L, 1L, 2L, 1L, 1L, 1L, 2L, 2L, 2L, 1L, 1L, 1L,
1L, 1L, 1L, 2L, 2L, 1L, 1L, 1L, 1L, 2L, 2L, 1L, 2L, 2L, 1L, 1L,
1L, 1L, 2L), levels = c("No", "Yes"), class = "factor"), Sex = structure(c(1L,
1L, 2L, 1L, 1L, 1L, 2L, 1L, 2L, 1L, 1L, 1L, 2L, 2L, 1L, 1L, 2L,
1L, 2L, 1L, 1L, 1L, 1L, 2L, 1L, 1L, 2L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 2L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 2L, 2L, 1L, 1L, 2L, 1L, 1L), levels = c("Female", "Male"
), class = "factor"), `23` = c(3.7, 9.9, 18.1, 12.8, 2.3, 0,
8.7, 3.7, 1.5, 8.6, 5.7, 5.8, 2.1, 1.5, 6.1, 1.6, 0.6, 2.9, 9.1,
7.4, 0, 0, 9.5, 8.7, 4.6, 0, 0, 8.6, 1.6, 7.3, 6.3, 3, 7.8, 2.2,
0.7, 3.9, 2.8, 0.4, 6.2, 1.1, 2.3, 4.9, 4.5, 8, 8.6, 3.9, 1.5,
1.6, 0, 1.4, 3.2, 3.6, 3, 5.1, 4.1, 8.8, 9.6, 0.7)), row.names = c(NA,
-58L), class = c("tbl_df", "tbl", "data.frame"))

# 计算中位数并添加自定义位置偏移
medians <- datas %>%
  group_by(Pain, Sex) %>%
  summarise(m = round(median(`23`), 1), .groups = "drop") %>%
  mutate(
    x_offset = case_when(
      Sex == "Female" & Pain == "No" ~ 0.2,
      Sex == "Female" & Pain == "Yes" ~ 0.2,
      Sex == "Male" & Pain == "No" ~ 0.3,
      Sex == "Male" & Pain == "Yes" ~ 0.2,
      TRUE ~ 0
    ),
    y_offset = 0
  )

# 计算均值并添加自定义位置偏移
means <- datas %>%
  group_by(Pain, Sex) %>%
  summarise(m = round(mean(`23`), 1), .groups = "drop") %>%
  mutate(
    x_offset = case_when(
      Sex == "Female" & Pain == "No" ~ 0.2,
      Sex == "Female" & Pain == "Yes" ~ 0.2,
      Sex == "Male" & Pain == "No" ~ 0.3,
      Sex == "Male" & Pain == "Yes" ~ 0.2,
      TRUE ~ 0
    ),
    y_offset = case_when(
      Sex == "Female" & Pain == "No" ~ 0.8,
      Sex == "Female" & Pain == "Yes" ~ 0.8,
      Sex == "Male" & Pain == "No" ~ -1.0,
      Sex == "Male" & Pain == "Yes" ~ -0.6,
      TRUE ~ 0
    )
  )

# 绘制调整后的箱线图
datas %>%
  select(Pain, Sex, `23`) %>%
  ggplot(aes(y = `23`, x = Sex, fill = Pain)) +
  geom_boxplot(fatten = 1, outlier.alpha = 0.50, 
               position = position_dodge(width = 1.8),
               width = 0.7) +
  geom_text(data = medians, 
            aes(x = as.numeric(Sex) + x_offset, y = m + y_offset, label = m), 
            color = 'blue', size = 5) +
  geom_text(data = means, 
            aes(x = as.numeric(Sex) + x_offset, y = m + y_offset, label = m), 
            color = 'red', size = 5) +
  stat_boxplot(geom ='errorbar', position = position_dodge(width = 1.8)) +
  stat_summary(aes(ymax = ..y.., ymin = ..y.., color = 'Mean'),
               fun = mean, geom = 'errorbar', width = .75, linetype = 'solid', 
               position = position_dodge(width = 1.8)) +
  stat_summary(aes(ymax = ..y.., ymin = ..y.., color = 'Median'),
               fun = median, geom = 'errorbar', width = .75, linetype = 'solid', 
               position = position_dodge(width = 1.8)) +
  scale_colour_manual('Values', values = c(Median = 'blue', Mean = 'red')) +
  scale_fill_manual(values = c('lightblue', 'lightgreen')) +
  theme(legend.position = 'top',
        text = element_text(size = 12),
        strip.text = element_text(size = 11, face = "bold"),
        legend.text = element_text(size = 12),
        axis.text.x = element_text(size = 16, vjust = 1),
        panel.spacing.x = unit(2, "lines")) +
  facet_wrap(vars(Sex), scales = 'free')

代码说明 #

1. 替换plyr为dplyr：避免包函数冲突，采用tidyverse风格的分组统计。
2. 自定义偏移量：通过x_offset和y_offset为每个场景的标签设置专属位置，解决重叠问题。
3. 调整箱线图布局：增大position_dodge(width)到1.8，缩小箱线图宽度到0.7，提升面板内空间利用率。
4. 标签定位逻辑：将因子型x轴转换为数值后叠加偏移量，实现标签的精准移动。

内容的提问来源于stack exchange，提问作者Andrzej Andrzej