How to Select the First Row by Group Using dplyr

Often you may want to select the first row in each group using the dplyr package in R. You can use the following basic syntax to do so:

df %>%
  group_by(group_var) %>%
  arrange(values_var) %>%
  filter(row_number()==1)

The following example shows how to use this function in practice.

Example: Select the First Row by Group in R

Suppose we have the following dataset in R:

#create dataset
df #view dataset
df

   team points
1     A      4
2     A      9
3     A      7
4     B      7
5     B      6
6     B     13
7     C      8
8     C      8
9     C      4
10    C     17

The following code shows how to use the dplyr package to select the first row by group in R:

library(dplyr)

df %>%
  group_by(team) %>%
  arrange(points) %>%
  filter(row_number()==1)

# A tibble: 3 x 2
# Groups:   team [3]
  team  points
    
1 A          4
2 C          4
3 B          6

By default, arrange() sorts the values in ascending order but we can easily sort the values in descending order instead:

df %>%
  group_by(team) %>%
  arrange(desc(points)) %>%
  filter(row_number()==1)

# A tibble: 3 x 2
# Groups:   team [3]
  team  points
    
1 C         17
2 B         13
3 A          9

Note that you can easily modify this code to select the n^th row by each group. Simply change row_number() == n.

For example, if you’d like to select the 2nd row by group, you can use the following syntax:

df %>%
  group_by(team) %>%
  arrange(desc(points)) %>%
  filter(row_number()==2)

Or you could use the following syntax to select the last row by group:

df %>%
  group_by(team) %>%
  arrange(desc(points)) %>%
  filter(row_number()==n())

Additional Resources

How to Arrange Rows in R
How to Count Observations by Group in R
How to Find the Maximum Value by Group in R