如何按各分组累计值调整ggplot堆叠条形图的堆叠顺序

你当前代码中使用reorder(subgroup, value)是基于全数据集的value值为subgroup生成全局统一的因子顺序，ggplot2的堆叠条形图默认遵循全局因子顺序排列所有条形的堆叠层级，因此无法实现不同x分组下独立的子组堆叠顺序。

核心逻辑是先按大组分组，计算每个大组内部各子组的累计值，按累计值为每个大组内的子组排序，再生成独立的分组标识映射到geom_col的group美学，打破全局统一的堆叠顺序限制，同时保留fill映射到原子组保证同子组颜色一致。

library(ggplot2)
library(dplyr)

set.seed(8)
# 构造示例数据
name <- c("a", "b", "c", "d", "e", "f", "g", "h", "i", "j", "k", "l", "m", "n", "o", "p", "q", "r", "s", "t", "v", "w", "x", "y", "z", "aa")
group <- c("1","1","1","2","2", "2", "1","1","1","1","1","1","1","1","1","1","1","1","2","2","2","2","2","2","2","2")
subgroup <- c("C", "C", "C", "C", "C", "C", "A", "A", "A","A","A","A","B","B","B","B","B","B","B","B", "C", "C", "C", "C", "C", "C")
value <- runif(26, min = 0, max = 10)
df <- data.frame(name, group, subgroup,value)

# 数据预处理
df <- df %>%
  group_by(group, subgroup) %>%
  # 计算每个大组内各子组的累计值
  mutate(subg_sum = sum(value)) %>%
  ungroup() %>%
  # 按大组、子组累计值升序排列，决定从下到上的堆叠顺序
  arrange(group, subg_sum) %>%
  # 生成唯一的堆叠分组标识，因子顺序与排序结果一致
  mutate(stack_id = factor(paste0(group, "_", subgroup), levels = unique(paste0(group, "_", subgroup))))

# 绘图
ggplot(df, aes(x = group, y = value)) +
  geom_col(
    aes(fill = subgroup, group = stack_id),
    color = "white" # 可选：添加白色边界区分相邻堆叠块
  )

• 大组1内子组累计值从小到大依次为：C(17.8) < A(30.2) < B(40.8)，堆叠顺序从下到上和该顺序完全一致
• 大组2内子组累计值从小到大依次为：B(10.6) < C(43.1)，堆叠顺序同样匹配组内排序结果
• 如果需要将累计值更大的子组放在堆叠底层，只需要将arrange(group, subg_sum)修改为arrange(group, desc(subg_sum))即可。

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