在 r 中绘制堆叠和分组的条形图

Question

I'm currently trying to plot some data in R, and not really succeeding. My data looks like this:

My data is stored in a data frame called "Head": My response variable is a factor (called "colour") with only two levels ("black" and "white") My first (and main) explanatory variable is a factor (called "Year" with two levels ("Year 1996" and "Year 2019"). My secondary explanatory variable is a factor (called "DateClass") with three levels ("A", "B" and "C"). So the dataset looks something like this:

Colour Year DateClass

Black 1996 A

Black 2019 B

White 1996 A

Black 1996 C

White 2019 B

White 1996 B

...etc

I want to plot "head colour" as a proportion of white/black vs "Year" and "DateClass". I had thought of a grouped barchart and stacking Black and White, but I don't know how to do it. Any suggestions? Would you plot the stacks of "colour" as percent or as total counts?

Thank you for your answers, best regards

Answer 1

Is this what you are looking for?

DATA :

df <- data.frame(Colour = c("Black", "Black", "White", "Black", "White", "White"),
                 Year = c(1996, 2019, 1996, 1996,2019, 1996),
                 DateClass = c("A", "B", "A", "C", "B", "B"))

EDITED SOLUTION :

First, aggregate the data to obtain the proportions of Colour by DateClass in subsets of df by Year==2019 and, respectively, Year==2019 ; store the results in df1996 and df2019 :

df1996 <- aggregate(
  x = df$DateClass[df$Year==1996],
  by = list(df$Colour[df$Year==1996]),
  function(x) prop.table(table(x)))

df2019 <- aggregate(
  x = df$DateClass[df$Year==2019],
  by = list(df$Colour[df$Year==2019]),
  function(x) prop.table(table(x)))

Now plot the two dataframes side by side using a two-panel layout:

par(mfrow = c(1,2))
barplot(df1996[,-1], cex.names = 0.8, col = c("blue","red"),
        main = "Black/White by DateClass in 1996", cex.main = 0.8,
        xlab = "1996", ylab = "Proportions")

barplot(df2019[,-1], cex.names = 0.8, col = c("blue","red"),
        main = "Black/White by DateClass in 2019", cex.main = 0.8,
        xlab = "2019", ylab = "Proportions")