我正在寻找以下问题的解决方案：
我有包含两个因子变量的数据EDU and LEVEL。可重现的数据样本在这里：
structure(list(EDU = structure(c(3L, 1L, 2L, 2L, 3L, 2L, 3L,
2L, 3L, 1L, 1L, 1L, 1L, 1L, 3L, 1L, 2L, 3L, 3L, 1L, 2L, 3L, 2L,
2L, 2L, 1L, 1L, 3L, 3L, 2L, 3L, 2L, 3L, 3L, 2L, 3L, 2L, 2L, 3L,
3L, 1L, 1L, 3L, 3L, 3L, 3L, 2L, 1L, 3L, 1L), .Label = c("A",
"B", "C"), class = "factor"), LEVEL = structure(c(3L, 3L, 4L,
2L, 4L, 3L, 1L, 2L, 2L, 1L, 3L, 2L, 3L, 2L, 3L, 3L, 4L, 2L, 2L,
4L, 1L, 2L, 3L, 3L, 1L, 4L, 2L, 3L, 1L, 1L, 2L, 3L, 1L, 2L, 1L,
4L, 3L, 1L, 4L, 3L, 4L, 1L, 4L, 2L, 4L, 1L, 1L, 4L, 3L, 1L), .Label = c("1",
"2", "3", "4"), class = "factor")), class = "data.frame", row.names = c(NA,
-50L))
使用这些数据我想绘制一个条形图ggplot2显示分组变量EDUx 轴上的累积百分比LEVEL在 y 轴上。另外，我想添加第四个栏，其中包含LEVEL but not分组依据EDU——有点像“整体吧”。此外，我想在图中添加百分比标签，以便每个LEVEL标有相应的相对频率，如下所示thread https://stackoverflow.com/questions/22231124/how-to-draw-stacked-bars-in-ggplot2-that-show-percentages-based-on-group或这个thread https://stackoverflow.com/questions/58744507/how-to-show-percentages-in-stacked-column-with-ggplot-geom-bar。老实说，我尝试使用 stackoverflow 中的不同解决方案来调整我的代码，以将百分比标签放入绘图中，因为有很多关于此主题的线程（特别是当涉及百分比标签时），但卡住了。到目前为止，我的ggplot2代码如下所示：
library(tidyverse)
ggplot(df, aes(x=EDU, fill=LEVEL)) +
geom_bar(position="fill") +
scale_y_continuous(labels = scales::percent)
结果如下图：
到目前为止，该情节看起来不错。但如上所述，我的目标是添加百分比标签，可能是geom_text除了现有的三个栏之外，还有第四个“整体栏”。对于百分比标签我也尝试制作prop.table并添加带有相应道具的百分比标签和annotate:
props <- prop.table(table(df$EDU, df$LEVEL), margin=1)
ggplot(df, aes(x=EDU, fill=LEVEL)) +
geom_bar(position="fill") +
scale_y_continuous(labels = scales::percent) +
annotate("text", x="A", y=.15, label=scales::percent(props[1,4])) +
annotate("text", x="B", y=.10, label=scales::percent(props[2,4])) +
annotate("text", x="C", y=.275, label=scales::percent(props[3,4])) +
annotate("text", x="A", y=.375, label=scales::percent(props[1,3])) +
annotate("text", x="B", y=.275, label=scales::percent(props[2,3])) +
annotate("text", x="C", y=.625, label=scales::percent(props[3,3])) +
annotate("text", x="A", y=.66, label=scales::percent(props[1,2])) +
annotate("text", x="B", y=.5, label=scales::percent(props[2,2])) +
annotate("text", x="C", y=.78, label=scales::percent(props[3,2])) +
annotate("text", x="A", y=.9, label=scales::percent(props[1,1])) +
annotate("text", x="B", y=.9, label=scales::percent(props[2,1])) +
annotate("text", x="C", y=.9, label=scales::percent(props[3,1]))
That results in the following plot:
这对我来说似乎很麻烦，特别是当我想创建多个绘图并且必须单独注释每个百分比时。在这里，问题可能是我如何设置 y 参数annotate以“自动”方式让 R 为我定位标签。
不幸的是，关于“整体酒吧”问题，我不知道如何解决。
我很感激任何帮助！
请放心：您的经验越多，您就越不会害怕提前准备数据。您会发现，首先将数据准备到您想要绘制的内容，然后再进行绘制通常会更容易、更清晰。不要尝试在 ggplot2 中完成所有事情，这会变得非常痛苦。
代码中的注释
library(tidyverse)
## create a percentage column manually
df_perc <-
df %>%
count(EDU, LEVEL) %>%
group_by(EDU) %>%
mutate(perc = n*100/sum(n))
## for the total, create a new data frame and bind to the old one
total <-
df_perc %>%
group_by(LEVEL) %>%
summarise(n = sum(n)) %>%
## ungroup for the total
ungroup() %>%
## add EDU column called total, so you can bind it and plot it easily
mutate(perc= n*100/sum(n), EDU = "Total")
## now bind them and plot them
bind_rows(df_perc, total) %>%
ggplot(aes(x=EDU, y = perc, fill=LEVEL)) +
## use geom_col, and remove position = fill
geom_col() +
# now you can add the labels easily as per all those threads
geom_text(aes(label = paste(round(perc, 2), "%")), position = position_stack(vjust = .5)) +
## you can either change the y values, or use a different scale factor
scale_y_continuous("Percent", labels = function(x) scales::percent(x, scale = 1))