Search code examples
rggplot2boxplot

Put multiple boxplots per group stratified

I have a question on how to stratify making multiple box plots per group. This is what I have for a sample code

library(ggplot2)

mtcars$vs <- as.character(as.numeric(mtcars$vs))

y6 <- ggplot(mtcars, aes(x=vs,y=hp)) +
  geom_boxplot(aes(group = vs),outlier.shape=NA, size=1, width = 0.6, fatten = 1) + 
  geom_jitter(aes(x=vs, y=hp, pch = factor(cyl)), position=position_jitter(width=.1, height=0), size = 2) + 
  scale_shape_manual(name ="X", values = c(1,2,3))  + 
  coord_cartesian(ylim=c(0, 350))

enter image description here

This is what I obtain from the graph. I hope to stratify the graphs per X axis by the legend making a total of 6 box plots (3 per X axis; 3 for "1" and 3 for "2"). Is there a way to do this? I have attached an image of it below:

enter image description here

Solution

  • Here is the code for you:

    library(ggplot2)
  
ggplot(mtcars, aes(x=vs,y=hp,fill = factor(cyl))) +
  geom_boxplot(aes(fill = factor(cyl)),outlier.shape=NA, size=1, width = 0.6, fatten = 1) + 
  coord_cartesian(ylim=c(0, 350))

    enter image description here

    I have used fill= argument in ggplot() to split/group the data by column cyl.

    If you look closer at mtcars data and your plot, you actually do not have 3 unique values of cyl for vs = 1, just two (cyl 4 & 8)..Therefore you get total of 5 boxes