ggplot条形图：列文本数据绑定、可见性优化及替代方案咨询

1. 修正标签与条形的绑定问题 #

你当前代码里用了position_stack(vjust = 0.5)，但分组条形图用的是position = "dodge"，两者位置逻辑不匹配，导致标签错位。需要把文本标签的位置参数改成position_dodge(width = 0.9)，和geom_col的dodge宽度对应，才能让标签和对应分组条形对齐。另外reorder排序时，建议指定按amount的总和排序，避免单值排序的混乱。

修正后的核心代码：

df %>% 
  ggplot(aes(reorder(area, -amount, sum), amount,
             group = type,
             fill = type))+
  geom_col(position = position_dodge(width = 0.9))+
  scale_y_continuous(labels = comma_format(big.mark = ".", decimal.mark = ","))+
  theme(axis.text.x = element_text(vjust = 1, hjust=1))+
  geom_text(aes(label = amount), 
            size = 3, 
            position = position_dodge(width = 0.9), # 匹配dodge宽度
            vjust = 0.5)+ # 垂直居中
  coord_flip()

2. 提升文本标签可见性 #

可以从这几个维度调整：

• 对比色设置：给文本配和条形填充色反差大的颜色，比如白色，添加color = "white"参数
• 数值格式化：用scales::comma把数值转成带千分位的格式，提升可读性
• 文本背景框：用geom_label替代geom_text，自带背景框避免被条形颜色覆盖
• 微调位置与大小：增大字号，或调整vjust/hjust避免标签贴边

示例优化代码：

geom_label(aes(label = comma(amount)), 
           size = 3.5, 
           position = position_dodge(width = 0.9),
           vjust = 0.5,
           fill = "white", # 白色背景增强辨识度
           color = "black")

也可以针对不同填充色设置对应文本颜色：

df %>% 
  ggplot(aes(reorder(area, -amount, sum), amount,
             group = type,
             fill = type, color = type))+
  geom_col(position = position_dodge(width = 0.9))+
  scale_color_manual(values = c("incoming" = "white", "outgoing" = "black"))+
  geom_text(aes(label = comma(amount)), 
            size = 3, 
            position = position_dodge(width = 0.9),
            vjust = 0.5)+
  coord_flip()

3. 除ggplot2外的绘图包选择 #

当然可以，这里给几个常用替代方案：

Base R 原生绘图 #

先把数据转成宽格式，再用barplot实现分组条形图：

# 重塑为宽格式
library(reshape2)
df_wide <- dcast(df, area ~ type, value.var = "amount")

# 绘制分组条形图
bar_pos <- barplot(t(as.matrix(df_wide[, -1])), 
                   beside = TRUE, 
                   names.arg = df_wide$area,
                   col = c("#619CFF", "#F8766D"),
                   legend.text = colnames(df_wide[, -1]),
                   ylab = "Amount")
# 添加文本标签
text(x = bar_pos,
     y = t(as.matrix(df_wide[, -1])) + 200, # 偏移量避免贴边
     labels = t(as.matrix(df_wide[, -1])),
     cex = 0.8)

Plotly 交互式绘图 #

生成可交互图表，鼠标悬停即可查看数值，无需手动添加标签：

library(plotly)

plot_ly(df, x = ~area, y = ~amount, color = ~type, type = "bar", barmode = "group") %>%
  layout(yaxis = list(title = "Amount", tickformat = ",."),
         xaxis = list(title = "Area"))

ggpubr 简化版ggplot #

基于ggplot2封装，语法更简洁：

library(ggpubr)

ggbarplot(df, x = "area", y = "amount",
          fill = "type", palette = "jco",
          position = position_dodge(0.9),
          label = TRUE, lab.pos = "in", lab.col = "white") %>%
  coord_flip() %>%
  scale_y_continuous(labels = comma_format(big.mark = ".", decimal.mark = ","))

内容的提问来源于stack exchange，提问作者Sebastian