Add new level to a factor variable that is the combination of two original levels.

[FabianaF]

Hello my goal is to create a plot with 3 box plots. I have a data frame with a column of numerical data and a factor variable with 2 levels. I am trying to group those levels into an additional level. What R code can I use to make a factor variable with 3 levels— “var1”, “var2”, “var1+var2”.

[nirgrahamuk]

library(tidyverse)
set.seed(42)
(exampl_start <- tibble(
  fac = factor(rep(1:2, 5)),
  num = sample(10, 10, TRUE)
))

ggplot(
  data = exampl_start,
  aes(x = fac, y = num)
) +
  geom_boxplot()

(ex1 <- mutate(exampl_start,
  fac =fct_expand(fac, "3")))

#either
(ex2a <- bind_rows(
  ex1,
  mutate(ex1, fac = fct_recode(fac,
    "3" = "1",
    "3" = "2"
  ))
))

#or 
(ex2b <- bind_rows(
  ex1,
  mutate(ex1, fac="3"
  ))
 %>% mutate(fac=factor(fac)))

#result
ggplot(
  data = ex2,
  aes(x = fac, y = num)
) +
  geom_boxplot()