Stacked bar plot, label bars with percentage values [duplicate]

China☆狼群 提交于 2020-01-07 05:35:10

问题


I want to plot a table as a stacked bar plot and label the bars with the percentages. Here is an example:

data <- matrix(c(34, 66, 22, 78), ncol = 2)
data <- as.table(data)
colnames(data) <- c("shop1", "shop2")
rownames(data) <- c("prod1", "prod2")

library(reshape2)
data_m <- melt(data, varnames = c("Product", "Shop"), id.vars = "Product")

library(scales)
library(ggplot2)
ggplot(data_m, aes(x = Shop, y = value, fill = Product)) + 
geom_bar(position = "fill", stat = "identity") + 
scale_y_continuous(labels = percent_format()) +
labs(x = "", y = "")

I tried to add the labels with

geom_text(data = data_m, aes(x = Shop, y = value, 
                         label = paste0((value/100) * 100,"%")), size=4)

but this results in

EDIT: With JanLauGe's answer I get

Now, the percentages are wrongly assigned.

Another remark: what to do if the column sums of the table were not the same, say 91 and 107 instead of 100 as assumed in my above example?


回答1:


Try this instead

geom_text(data = data_m, 
          aes(x = Shop, 
              y = value / max(value), 
              label = paste0(value/100,"%")), 
          size = 4)

The problem: label position is relative to the plot area (0 to 1, 1 = max(value)).

The solution: rescale value accordingly.


EDIT: Duplicate of this post.

What you are looking for is this:

ggplot(data = data_m, 
       aes(x = Shop, 
           y = value, 
           fill = Product,
           cumulative = TRUE)) +
  geom_col() +
  geom_text(aes(label = paste0(value/100,"%")), 
            position = position_stack(vjust = 0.5)) + 
  theme_minimal()


来源:https://stackoverflow.com/questions/43065409/stacked-bar-plot-label-bars-with-percentage-values

易学教程内所有资源均来自网络或用户发布的内容,如有违反法律规定的内容欢迎反馈
该文章没有解决你所遇到的问题?点击提问,说说你的问题,让更多的人一起探讨吧!