Calculating multiple columns from one column with summarise

August 16, 2022

Here is an example of what I’m trying to achieve:

df <- data.frame(label = c(rep("ABC", 5), rep("CDE", 5), rep("FGH", 5)), x = runif(15, 0, 100))

df %>% group_by(label) %>%
  summarise(across(everything(), list(lessthan_10 = ~sum(. < 10), lessthan_20 = ~sum(. < 20), lessthan_30 = ~sum(. < 20), lessthan_40 = ~sum(. < 40))))

In this case, I’m calculating 4 different columns in the summary (counting the entries less than 10, less than 20, less than 30, and less than 40). In reality, I would like to calculate 100 different columns using a custom function that takes in x and 100 different parameters. Is there a way to do this using a loop or a list without writing out every single column I want to calculate?

>Solution :

You can use purrr::map_dfc:

library(tidyverse)
df %>% 
  group_by(label) %>% 
  summarise(map_dfc(seq(10, 40, 10), ~ tibble("x_lessthan_{.x}" := sum(x < .x))))

  label x_lessthan_10 x_lessthan_20 x_lessthan_30 x_lessthan_40
  <chr>         <int>         <int>         <int>         <int>
1 ABC               0             3             3             3
2 CDE               0             2             3             4
3 FGH               1             2             3             3