Rename factors in ggplot2 with Labeller() function in R

Display custom labels for variables during visualization

How do you normally rename factor variables when you plot?

The tidyverse has a function called `labeller`, which is used to conveniently relabel facet (or strip) labels.

It can also be easily modified to create a labeling function for any factor variable.

Step 1

Set up a plot to modify:

iris %>% filter(Sepal.Width == 3.1) %>% 
  ggplot() +
  geom_boxplot(aes(x=Species, y=Sepal.Length)) +
  theme_minimal()

Step 2

create a labeller function to define new labels for each factor.

relabel.Species = as_labeller(c(setosa = "Poppy", 
                   versicolor = "Rose", 
                   virginica = "Daisy"))

the ‘as_labeller’ function packages the named list as a lookup function. You can test this by running `relabel.Species(“versicolor”)` which will return a new label.

Step 3

create a new label column when the plot is created

iris %>% as_tibble() %>% filter(Sepal.Width == 3.1) %>% 
  rowwise() %>% 
  mutate(Species2 = factor(relabel.Species(Species))) %>%  
  ggplot() +
  geom_boxplot(aes(x=Species2, y=Sepal.Length)) +
  theme_gray()

The rowwise() function forces R to mutate each row separately, rather than all the rows. The new relabel.Species function converts the Species variable to a new name and then is converted back into a factor.

ggplot (or any R verb after the operation) will be none the wiser! Alternatively, you could create a new column (Species.Label) instead of replacing the old column.