如何在分组直方图中添加整体直方图并统一坐标轴？

我有一个包含多列的数据集，其中一列是取值为1、2、3的分类变量。我想用ggplot绘制3个分组直方图垂直排列，并在第三个直方图下方添加不区分分类变量的整体直方图。目前编写的代码如下：

#The name of the dataset is dataset.clust

dataset.clust$cluster = factor(dataset.clust$cluster)

library(patchwork)
library(ggplot2)

dataset.clust$cluster = factor(dataset.clust$cluster)

hist_plot1 = ggplot(dataset.clust, aes(x = population, fill = cluster)) +
  geom_histogram(position = "identity", alpha = 0.7) +
  facet_wrap(~ cluster, nrow = 3) +
  scale_fill_manual(values = c("red", "green", "blue")) +
  guides(fill = guide_legend(title = "Cluster")) +
  ggtitle("Grouped Histograms for population")

# Second Histogram
hist_plot2 = ggplot(dataset.clust, aes(x = population)) +
  geom_histogram(fill = "yellow", color = "black")

# Combine histograms and arrange vertically
combined_plot = hist_plot1 / hist_plot2

print(combined_plot)

生成的图中四个子图的x轴和y轴范围不一致，请问如何调整让它们的x轴和y轴保持一致？

要让四个图的x轴和y轴保持一致，核心是统一轴范围、分箱宽度，并通过布局对齐确保视觉统一，具体调整如下：

1. 统一轴范围 #

先计算population列的极值，以及所有直方图的最大计数，然后在两个绘图对象中强制设置相同的轴范围：

# 计算统一的x轴范围
x_min <- min(dataset.clust$population, na.rm = TRUE)
x_max <- max(dataset.clust$population, na.rm = TRUE)

# 计算所有分组及整体直方图的最大y轴计数
y_max <- max(
  ggplot_build(hist_plot1)$data[[1]]$count,
  ggplot_build(hist_plot2)$data[[1]]$count,
  na.rm = TRUE
)

2. 统一分箱宽度 #

为避免自动分箱导致柱子宽度差异，在geom_histogram中显式指定binwidth参数（数值根据数据分布调整）：

# 示例分箱宽度，可根据你的数据修改
bin_width <- 5

3. 调整绘图对象并对齐 #

修改两个绘图对象的轴设置，再用patchwork的plot_layout实现轴对齐：

# 调整分组直方图
hist_plot1 <- hist_plot1 +
  xlim(x_min, x_max) +
  ylim(0, y_max) +
  geom_histogram(position = "identity", alpha = 0.7, binwidth = bin_width) +
  theme(axis.title.x = element_blank()) # 隐藏分组图x轴标题，避免重复

# 调整整体直方图
hist_plot2 <- hist_plot2 +
  xlim(x_min, x_max) +
  ylim(0, y_max) +
  geom_histogram(fill = "yellow", color = "black", binwidth = bin_width) +
  ggtitle("Overall Histogram for population")

# 组合并强制对齐
combined_plot <- hist_plot1 / hist_plot2 +
  plot_layout(align = "hv") # 同时水平、垂直对齐轴

完整代码 #

dataset.clust$cluster = factor(dataset.clust$cluster)

library(patchwork)
library(ggplot2)

# 计算统一参数
x_min <- min(dataset.clust$population, na.rm = TRUE)
x_max <- max(dataset.clust$population, na.rm = TRUE)
bin_width <- 5 # 根据数据分布调整

# 绘制分组直方图
hist_plot1 = ggplot(dataset.clust, aes(x = population, fill = cluster)) +
  geom_histogram(position = "identity", alpha = 0.7, binwidth = bin_width) +
  facet_wrap(~ cluster, nrow = 3) +
  scale_fill_manual(values = c("red", "green", "blue")) +
  guides(fill = guide_legend(title = "Cluster")) +
  ggtitle("Grouped Histograms for population") +
  xlim(x_min, x_max) +
  ylim(0, max(ggplot_build(.)$data[[1]]$count, na.rm = TRUE)) +
  theme(axis.title.x = element_blank())

# 绘制整体直方图
hist_plot2 = ggplot(dataset.clust, aes(x = population)) +
  geom_histogram(fill = "yellow", color = "black", binwidth = bin_width) +
  ggtitle("Overall Histogram for population") +
  xlim(x_min, x_max) +
  ylim(0, max(ggplot_build(hist_plot1)$data[[1]]$count, ggplot_build(.)$data[[1]]$count, na.rm = TRUE))

# 组合对齐并输出
combined_plot = hist_plot1 / hist_plot2 + plot_layout(align = "hv")
print(combined_plot)

内容的提问来源于stack exchange，提问作者Billy