Create New Variables in R with mutate() and case

Often you may want to create a new variable in a data frame in R based on some condition. Fortunately this is easy to do using the mutate() and case_when() functions from the dplyr package.

This tutorial shows several examples of how to use these functions with the following data frame:

#create data frame
df <- data.frame(player = c('a', 'b', 'c', 'd', 'e'),
                 position = c('G', 'F', 'F', 'G', 'G'),
                 points = c(12, 15, 19, 22, 32),
                 rebounds = c(5, 7, 7, 12, 11))

#view data frame
df

  player position points rebounds
1      a        G     12        5
2      b        F     15        7
3      c        F     19        7
4      d        G     22       12
5      e        G     32       11

Example 1: Create New Variable Based on One Existing Variable

The following code shows how to create a new variable called ‘scorer’ based on the value in the points column:

library(dplyr)

#define new variable 'scorer' using mutate() and case_when()
df %>%
  mutate(scorer = case_when(points < 15 ~ 'low',
                           points < 25 ~ 'med',
                           points < 35 ~ 'high'))

  player position points rebounds scorer
1      a        G     12        5    low
2      b        F     15        7    med
3      c        F     19        7    med
4      d        G     22       12    med
5      e        G     32       11   high

Example 2: Create New Variable Based on Several Existing Variables

The following code shows how to create a new variable called ‘type’ based on the value in the player and position column:

library(dplyr)

#define new variable 'type' using mutate() and case_when()
df %>%
  mutate(type = case_when(player == 'a' | player == 'b' ~ 'starter',
                            player == 'c' | player == 'd' ~ 'backup',
                            position == 'G' ~ 'reserve'))

  player position points rebounds    type
1      a        G     12        5 starter
2      b        F     15        7 starter
3      c        F     19        7  backup
4      d        G     22       12  backup
5      e        G     32       11 reserve

The following code shows how to create a new variable called ‘valueAdded’ based on the value in the points and rebounds columns:

library(dplyr)

#define new variable 'valueAdded' using mutate() and case_when()
df %>%
  mutate(valueAdded = case_when(points <= 15 & rebounds <=5 ~ 2,
                                points <=15 & rebounds > 5 ~ 4,
                                points < 25 & rebounds < 8 ~ 6,
                                points < 25 & rebounds > 8 ~ 7,
                                points >=25 ~ 9))

  player position points rebounds valueAdded
1      a        G     12        5          2
2      b        F     15        7          4
3      c        F     19        7          6
4      d        G     22       12          7
5      e        G     32       11          9