如何用ggplot2复现Base R绘制的双柱状图

我已经用Base R画出了下面的双柱状图，现在想改用ggplot2复现一模一样的效果，具体该怎么操作？

以下是我用Base R绘图的代码：

data <- structure(list(
  A= c(53, 39),
  B = c(30, 36), 
  C = c(12, 19), 
  D = c(5, 6)), 
  .Names = c("Poor", "Fair", "Good", "Excellent"), 
  class = "data.frame", 
  row.names = c(NA, -2L))

attach(data)
print(data)
colors <- c("dodgerblue3","red3")
barplot(as.matrix(data), 
        main="",
        beside = T, ylab = "", ylim = range(0,65),  col=colors, border=F)

legend(5,60,  legend=c("Fox News", "CNN"), 
       fill=colors, horiz = T, bty='n')
text(1.5, 55, "53%")
text(2.5, 41, "39%")
text(4.5, 32, "30%")
text(5.5, 38, "36%")
text(7.5, 14, "12%")
text(8.5, 21, "19%")
text(10.5, 7, "5%")
text(11.5, 8, "6%")

对应的图表效果：横轴是评价等级（Poor、Fair、Good、Excellent），纵轴范围0-65；每个等级下有两个并列柱子，深蓝色代表Fox News，红色代表CNN；每个柱子上方都标有对应的百分比；顶部有个无外框的水平图例。

实现方法 #

ggplot2依赖长格式数据，所以先把当前的宽格式数据转成长格式，再绘图：

1. 转换数据格式 #

用tidyr和dplyr包把数据转成长格式，同时添加媒体来源的分组列：

# 加载需要的包
library(ggplot2)
library(tidyr)
library(dplyr)

# 原数据
data <- structure(list(
  A= c(53, 39),
  B = c(30, 36), 
  C = c(12, 19), 
  D = c(5, 6)), 
  .Names = c("Poor", "Fair", "Good", "Excellent"), 
  class = "data.frame", 
  row.names = c(NA, -2L))

# 添加分组列并转成长格式
data_long <- data %>%
  mutate(Source = c("Fox News", "CNN")) %>%
  pivot_longer(cols = -Source, names_to = "Rating", values_to = "Percentage")

2. 绘制匹配的双柱状图 #

用ggplot2构建图表，调整所有参数和原图表对齐：

ggplot(data_long, aes(x = Rating, y = Percentage, fill = Source)) +
  # 画并列柱子，去掉边框
  geom_col(position = position_dodge(width = 0.9), color = NA) +
  # 设置和原图表一致的颜色
  scale_fill_manual(values = c("Fox News" = "dodgerblue3", "CNN" = "red3")) +
  # 设置纵轴范围，去掉纵轴标签
  scale_y_continuous(limits = c(0, 65), name = "") +
  # 添加柱子上方的百分比标签
  geom_text(aes(label = paste0(Percentage, "%")), 
            position = position_dodge(width = 0.9), 
            vjust = -0.5, size = 3.5) +
  # 调整图例和主题，匹配原图表风格
  theme(legend.position = c(0.7, 0.95),
        legend.direction = "horizontal",
        legend.background = element_blank(),
        plot.title = element_blank(),
        panel.grid.major = element_blank(),
        panel.grid.minor = element_blank(),
        panel.background = element_blank(),
        axis.ticks.x = element_blank())

关键参数说明 #

• position_dodge(width = 0.9)：控制并列柱子的间距，和Base R的beside=T效果一致
• geom_text的vjust=-0.5：让标签在柱子上方，避免和柱子重叠
• theme里的设置：去掉网格线、背景，调整图例位置和样式，完全还原原图表的简洁感

内容的提问来源于stack exchange，提问作者Data and AI Nerd