如何使用 ggplot2 在直方图中 plot 变量的份额

Question

我试过查看旧线程没有成功。 我正在尝试 plot 直方图中不同议会会议中男性立法者的份额。

这是我的代码，它有效，但显示了立法者的数量（不是份额）。 我怎样才能 plot 共享？ 谢谢！

    mergedf %>%
ggplot( aes(x = session, fill = factor(sex))) +
    geom_histogram (binwidth = 0.5)+
theme_minimal()+
  theme(legend.position ="bottom")+
  labs(title = "Share of male legislators by session", x= "Session", y = "Share of legislators", 
       fill ="sex")

编辑：我通过这张表获得了立法者的份额，但我不知道如何将它整合到直方图中。

mergedf %>% 
  tabyl (session, sex) %>% 
  adorn_percentages() %>% 
  adorn_pct_formatting ()

Answer 1

一种选择是使用一些dplyr动词来计算计数和百分比，然后可以通过geom_col显示为条形图（直方图是不同的），如下所示：

mergedf <- data.frame(
  sessions = c( 1, 2, 3, 4, 5, 2, 3, 4, 2),
  sex = c ("female", "female", "female", "male", "female", "female", "female", "male", "male")
)

library(dplyr)
library(ggplot2)

mergedf %>%
  group_by(sessions, sex) %>% 
  summarise(n = n()) %>%
  mutate(pct = n / sum(n)) %>%
  ggplot( aes(x = factor(sessions), y = pct, fill = sex)) +
  geom_col(width = .6)+
  theme_minimal()+
  theme(legend.position ="bottom")+
  labs(title = "Share of male legislators by session", x= "Session", y = "Share of legislators", 
       fill ="sex")
#> `summarise()` has grouped output by 'sessions'. You can override using the
#> `.groups` argument.

Answer 2

您只需在geom_histogram参数中指定position="fill" ：

library(ggplot2)
mergedf <- data.frame(
  session = c( 1, 2, 3, 4, 5, 2, 3, 4, 2),
  sex = c ("female", "female", "female", "male", "female", "female", "female", "male", "male")
)

ggplot(mergedf, aes(x = session, fill = factor(sex))) +
  geom_histogram (binwidth = 0.5, position = "fill") +  # <- HERE
  theme_minimal() +
  theme(legend.position ="bottom") +
  labs(title = "Share of male legislators by session", 
       x= "Session", y = "Share of legislators", fill ="sex")

从技术上讲，您并没有真正构建直方图（计数的合并分布），而是条形图，因此您也可以使用geom_bar ，格式相同：

ggplot(mergedf, aes(x = session, fill = factor(sex))) +
  geom_bar(position="fill") +
  theme_minimal () +
  theme(legend.position ="bottom") +
  labs(title = "Share of male legislators by session", 
       x= "Session", y = "Share of legislators", fill ="sex")