修改ggplot柱状图填充颜色：实现同组连续柱子统一配色

你目前使用以下R代码进行数据可视化：

df <- data.frame(factor = rep("myFactor", 9), filter = rep("Property Price", 9), outcome = rep("Price", 9), filter.value = c(rep("Country X", 3), rep("Country Y", 3), rep("Country Z", 3)), FactorLevel = rep(c("Bungalow", "House", "Apartment"), 3), Outcome_Variable = rep("Price",9), mean = rep(20, 9), min = rep(0, 9), max = rep(100, 9), sd = rep(5, 9), n = rep(50, 9), SE = rep(3, 9)) 
attach(df) 
print.plot <- function (rows) { 
  ggplot(rows, aes(x = FactorLevel, y = mean, fill = filter.value)) + 
    geom_bar(stat = "identity", position = position_dodge()) + 
    geom_errorbar(aes(ymin = mean - SE, ymax = mean + SE), position = position_dodge(.8), width = 0.25)+ 
    geom_text(aes(label = n), vjust = 1.6, color = "black", position = position_dodge(.9), size = 3.5) 
} 
print.plot(df)

你希望生成的柱状图里，最左侧三个柱子（对应Country X）均为红色，中间三个（Country Y）为绿色，最右侧三个（Country Z）为蓝色，也就是每组连续三个柱子使用统一颜色。

解决方案 #

其实这个需求很好实现，你只需要在ggplot的图层中添加scale_fill_manual()函数，手动指定每个filter.value（也就是不同国家）对应的颜色即可。因为你的fill映射的是filter.value，所以每个国家对应的所有柱子都会自动使用你指定的统一颜色。

修改后的完整代码如下：

df <- data.frame(factor = rep("myFactor", 9), filter = rep("Property Price", 9), outcome = rep("Price", 9), filter.value = c(rep("Country X", 3), rep("Country Y", 3), rep("Country Z", 3)), FactorLevel = rep(c("Bungalow", "House", "Apartment"), 3), Outcome_Variable = rep("Price",9), mean = rep(20, 9), min = rep(0, 9), max = rep(100, 9), sd = rep(5, 9), n = rep(50, 9), SE = rep(3, 9)) 
# 这里建议尽量不用attach()，避免环境变量冲突，ggplot会自动从传入的rows数据框读取变量
print.plot <- function (rows) { 
  ggplot(rows, aes(x = FactorLevel, y = mean, fill = filter.value)) + 
    geom_bar(stat = "identity", position = position_dodge()) + 
    geom_errorbar(aes(ymin = mean - SE, ymax = mean + SE), position = position_dodge(.8), width = 0.25)+ 
    geom_text(aes(label = n), vjust = 1.6, color = "black", position = position_dodge(.9), size = 3.5) +
    # 核心：手动指定每个国家对应的填充颜色
    scale_fill_manual(values = c(
      "Country X" = "red",
      "Country Y" = "green",
      "Country Z" = "blue"
    ))
} 
print.plot(df)

额外小提示 #

• 如果你想要更个性化的颜色，可以用十六进制色值代替颜色名称，比如"Country X" = "#FF4500"（橙红色）、"Country Y" = "#2E8B57"（海绿色），能实现更精准的色彩效果。
• 尽量避免使用attach(df)，它可能会导致全局环境里的变量名冲突，ggplot本身支持直接从传入的数据框中读取变量，完全不需要依赖attach。

内容的提问来源于stack exchange，提问作者wwl