ggplot2: add labels to percentage plot (not position=fill, but just fill)

Question

I would like to add percentage labels to a percentage barplot

I found solutions with position="fill" (Add percentage labels to a stacked barplot) and also here (How to draw stacked bars in ggplot2 that show percentages based on group?), however, I would like to keep relative frequencies for every group.

Here is an example plot:

# library
library(ggplot2)

# data  
df <- data.frame(group=c("A","A","A","A","B","B","B","C","C"),
                   anon=c("yes","no","no","no","yes","yes","no","no","no"))

# percentage barplot
  ggplot(df, aes(group),fill=anon) + 
    geom_bar(aes(y = (..count..)/sum(..count..),fill=anon)) + 
    scale_y_continuous(labels=scales::percent) +
    ylab("relative frequencies")

^{Created on 2020-04-19 by the reprex package (v0.3.0)}

Now I would like to add percentage labels to each red and green portion of each bar,so that I get "relative-relative" (e.g. 25% for "yes" for group A) values. How can this be done? Do I have to change my df for this or is this somehow possible within the ggplot function

Answer 1

A possible solution is to calculate proportion outside of ggplot2, here I use dplyr for calculating those different proportions:

library(dplyr)

df_calculated <- df %>% count(group, anon) %>%
  mutate(Percent_col = n / sum(n)) %>%
  group_by(group) %>%
  mutate(Percent = n/sum(n))

# A tibble: 5 x 5
# Groups:   group [3]
  group anon      n Percent_col Percent
  <fct> <fct> <int>       <dbl>   <dbl>
1 A     no        3       0.333   0.75 
2 A     yes       1       0.111   0.25 
3 B     no        1       0.111   0.333
4 B     yes       2       0.222   0.667
5 C     no        2       0.222   1    

And then to use geom_col instead of geom_bar to draw bargraph and geom_text to add text label of each proportion:

library(dplyr)
library(ggplot2)

ggplot(df_calculated, aes(x = group, y = Percent_col, fill = anon))+
  geom_col()+
  scale_y_continuous(labels=scales::percent) +
  ylab("relative frequencies")+
  geom_text(aes(label = scales::percent(Percent)), position = position_stack(0.5))+
  geom_text(inherit.aes = FALSE, 
            data = df_calculated %>% 
              group_by(group) %>% 
              summarise(Sum = sum(Percent_col)),
            aes(label = scales::percent(Sum), 
                y = Sum, x = group), vjust = -0.5)

Does it answer your question ?