R Programming Fundamentals

Boxplot

Boxplots are most appropriate when you want to check the distribution of a continuous y variable with some categorical (factor) x variable. We cannot create a boxplot of the mpg variable with the cyl variable in mtcars using the following code:

ggplot(mtcars, aes(cyl, mpg)) + geom_boxplot()

We will get a warning as follows:

Warning message: Continuous x aesthetic -- did you forget aes(group=...)?

The cyl variable is not explicitly declared as a factor variable in the mtcars dataset, so ggplot is confused about what the x variable is supposed to be. This is similar to when we created base plots with factor variables, though as we saw, plot() will still plot a variable not declared as a factor, but it will create a scatterplot instead ...