R ggplot2多线图绘制：无需单独数据框显示百分比

想用R的ggplot2绘制多线图，追踪两个地点（L1、L2）在6个时间点（t1-t6）的人员占比情况。目前能生成双线条图形，但Y轴无法正确显示百分比；现有可行方案需要先创建包含占比的单独数据框，请问有没有更直接的实现方法？

原始数据示例 #

time <- c("t1", "t1", "t1", "t1", "t1", "t1", 
          "t2", "t2", "t2", "t2", "t2", "t2", 
          "t3", "t3", "t3", "t3", "t3", "t3", 
          "t4", "t4", "t4", "t4", "t4", "t4", 
          "t5", "t5", "t5", "t5", "t5", "t5", 
          "t6", "t6", "t6", "t6", "t6", "t6")
location <- c ("L1", "L1", "L1", "L1", "L2", "L2", 
               "L1", "L2", "L2", "L2", "L2", "L2", 
               "L1", "L1", "L1", "L2", "L2", "L2", 
               "L2", "L2", "L2", "L2", "L2", "L2", 
               "L1", "L1", "L2", "L2", "L2", "L2", 
               "L1", "L1", "L2", "L2", "L1", "L2")
data <- data.frame (time, location)
data
table (data$time, exclude = FALSE)
table (data$location, exclude = FALSE)
table (data$location, data$time, exclude = FALSE)

当前无法显示百分比的代码 #

ggplot (data = data, 
        mapping = aes (
          x = time, 
          y = location,
          group = location, 
          color = location
        )) + 
  geom_point (stat = "identity", size = 3) + 
  geom_line (stat = "identity") + 
  ggtitle("各时间点不同地点的人员占比") + 
  xlab("时间") + 
  ylab ("百分比") +
  coord_cartesian( ylim = c(0, 100))

期望效果的测试代码（需单独数据框） #

time <- c("t1", "t1", 
          "t2", "t2", 
          "t3", "t3", 
          "t4", "t4", 
          "t5", "t5", 
          "t6", "t6")
location <- c("L1", "L2",
              "L1", "L2",
              "L1", "L2",
              "L1", "L2",
              "L1", "L2",
              "L1", "L2")
percent <- c(67, 33,
             24, 29,
             35, 45, 
             54, 56, 
             72, 91, 
             83, 23)
test <- data.frame (time, location, percent)

ggplot (data = test, 
        mapping = aes (
          x = time, 
          y =  percent,
          group = location, 
          color = location
        )) + 
  # scale_y_continuous(labels = scales::percent) + 
  geom_point (stat = "identity", size = 3) + 
  geom_line (stat = "identity") + 
  ggtitle("月度人员流向家庭/医院的占比图") + 
  xlab("时间") + 
  ylab ("百分比") +
  coord_cartesian( ylim = c(0, 100))

不需要单独创建占比数据框，以下两种方法可直接实现需求：

方法1：利用ggplot的stat_count直接计算占比 #

通过stat_count的内置变量..prop..计算每个时间点内各地点的占比，再结合scales包格式化Y轴为百分比显示：

library(ggplot2)
library(scales)

ggplot(data = data, aes(x = time, color = location, group = location)) +
  stat_count(aes(y = ..prop.. * 100), geom = "line", position = "identity") +
  stat_count(aes(y = ..prop.. * 100), geom = "point", size = 3, position = "identity") +
  scale_y_continuous(labels = percent_format(accuracy = 1)) +
  ggtitle("各时间点不同地点的人员占比") +
  xlab("时间") +
  ylab("百分比") +
  coord_cartesian(ylim = c(0, 100))

..prop..代表每组内的比例值，乘以100转换为百分比数值；position = "identity"确保计算逻辑是每个时间点内各地点的占比，而非全局占比。

方法2：在ggplot调用中直接用dplyr处理数据 #

如果习惯用dplyr做数据汇总，可直接在ggplot()的data参数里通过管道完成占比计算，无需单独保存中间数据框：

library(ggplot2)
library(dplyr)
library(scales)

ggplot(data = data %>% 
         group_by(time, location) %>% 
         summarise(count = n(), .groups = "drop") %>% 
         group_by(time) %>% 
         mutate(percent = (count / sum(count)) * 100),
       aes(x = time, y = percent, color = location, group = location)) +
  geom_point(size = 3) +
  geom_line() +
  scale_y_continuous(labels = percent_format(accuracy = 1)) +
  ggtitle("各时间点不同地点的人员占比") +
  xlab("时间") +
  ylab("百分比") +
  coord_cartesian(ylim = c(0, 100))

这段代码先按「时间+地点」分组统计人数，再按「时间」分组计算每个地点的占比，直接将处理后的结果传入ggplot，一步完成数据处理与绘图。

内容的提问来源于stack exchange，提问作者Eric Boorman