如何将ggplot散点图点大小按组内回答占比设置？

我用ggplot绘制两组调查响应的散点图，当前点大小基于每组的回答计数，但两组样本量差异大，导致其中一组的点始终更大。我需要把点大小改为每组内选择特定回答选项的个体占比：针对21个gain问题，分别计算西班牙裔/非西班牙裔群体对1-5分value的回答占比，并将该占比作为散点图的点大小。

当前代码：

ggplot(data, aes(y = value, x = gain, group = demographic, color = demographic)) +
  geom_count(position = position_dodge(width = 0.9))

数据结构：

structure(list(hispanic = structure(c(2L, 2L, 2L, 2L, 2L, 2L, 
2L, 2L, 2L, 2L, 1L, 1L, 2L, 2L, 2L, 2L, 1L, 1L, 2L, 2L, 2L, 2L, 
1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 1L, 1L, 
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 
2L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 
2L, 2L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 2L, 2L, 1L, 1L, 2L, 2L, 
2L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 2L, 
2L, 2L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 
1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 
1L, 1L, 1L, 1L), levels = c("hispanic", "non_hispanic"), class = "factor"), 
    gain = c("SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16", 
    "SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16", "SU_11_15", 
    "SU_11_16", "SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16", 
    "SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16", "SU_11_15", 
    "SU_11_16", "SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16", 
    "SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16", "SU_11_15", 
    "SU_11_16", "SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16", 
    "SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16", "SU_11_15", 
    "SU_11_16", "SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16", 
    "SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16", "SU_11_15", 
    "SU_11_16", "SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16", 
    "SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16", "SU_11_15", 
    "SU_11_16", "SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16", 
    "SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16", "SU_11_15", 
    "SU_11_16", "SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16", 
    "SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16", "SU_11_15", 
    "SU_11_16", "SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16", 
    "SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16", "SU_11_15", 
    "SU_11_16", "SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16", 
    "SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16", "SU_11_15", 
    "SU_11_16", "SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16", 
    "SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16", "SU_11_15", 
    "SU_11_16", "SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16", 
    "SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16", "SU_11_15", 
    "SU_11_16", "SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16", 
    "SU_11_15", "SU_11_16", "SU_11_15", "SU_11_16"), value = c(5, 
    5, 5, 5, 4, 4, 5, NA, 5, 4, NA, NA, 4, 2, 5, 5, 5, 5, 5, 
    5, 4, 3, 4, 4, 5, 5, 3, 3, 3, 2, 4, 4, 5, 5, NA, NA, 4, 3, 
    3, NA, 4, 3, 2, 4, 3, 1, 5, 3, NA, NA, 4, 2, 5, 4, 3, 3, 
    2, 2, 5, 3, 5, 5, 5, NA, 3, 1, 3, 3, 5, 5, 3, 1, 5, 3, 5, 
    5, 4, 3, 4, 4, 5, 3, 5, 5, 5, 5, 5, 3, 4, 3, 4, 3, 5, 4, 
    4, 2, 4, 3, 5, 3, 5, 4, 4, 3, 4, 2, 5, 5, 5, 5, 4, 4, 4, 
    3, 1, 1, 5, 5, 5, 3, 5, 3, 3, 3, 5, 3, 3, 2, 5, 5, 5, 3, 
    5, 3, 4, 4, 5, 3)), row.names = c(NA, -138L), class = c("tbl_df", 
"tbl", "data.frame"))

核心思路是先预处理数据计算每组内的占比，再用geom_point替代geom_count来指定点大小为计算好的占比。

步骤1：计算每组内的回答占比 #

用dplyr分组统计，过滤缺失值后，按群体、问题、回答分值分组计数，再计算每个分组在所属群体-问题组内的占比：

library(dplyr)

data_summary <- data %>%
  filter(!is.na(value)) %>%
  group_by(hispanic, gain, value) %>%
  summarise(count = n(), .groups = "drop_last") %>%
  mutate(prop = count / sum(count)) %>%
  ungroup()

步骤2：绘制散点图 #

用geom_point将点大小映射为计算好的占比，保留分组颜色映射并设置点偏移避免重叠：

library(ggplot2)

ggplot(data_summary, aes(x = gain, y = value, color = hispanic, size = prop)) +
  geom_point(position = position_dodge(width = 0.9)) +
  scale_size_continuous(range = c(2, 10), name = "组内占比") +
  labs(x = "调查问题", y = "回答分值", color = "群体") +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

关键说明 #

• 替换geom_count为geom_point：geom_count自动统计计数，而我们需要手动传入自定义的占比数据，geom_point更灵活。
• 占比计算逻辑：按群体和问题分组后，每个回答分值的计数除以该组总计数，确保点大小只反映组内相对占比，不受两组样本量差异影响。
• scale_size_continuous：可自定义点的大小范围，让占比差异更直观，同时设置图例名称提升可读性。

内容的提问来源于stack exchange，提问作者Matthew Murray