number of times specific value in each column R

Question

I have:

library(tidyverse)
df <- tibble(one=c(1,1,1,2,2,2,3,3),
       log1 = c(TRUE, TRUE, FALSE, FALSE, FALSE, FALSE, TRUE, TRUE),
       log2 = c(TRUE, TRUE, FALSE, FALSE, FALSE, FALSE, FALSE, FALSE),
       log3 = c(TRUE, TRUE, FALSE, FALSE, FALSE, FALSE, TRUE, FALSE))

I want to find number of times the word "FALSE" appears in each column and group and return a df

I have tried map_df(df, function(x) sum(x==FALSE)) and

df %>%
  group_by(one) %>%
  map_df( function(x) sum(x==FALSE))

but they do not break into separate groups.

this also errors out

df %>%
  group_by(one) %>%
  summarise( function(x) sum(x==FALSE))

Any suggestions?

Andre Wildberg · Accepted Answer · 2022-02-17T15:50:13.503

3

You can use across to work on multiple columns

library(dplyr)

df %>% 
  group_by(one) %>% 
  summarise(across(starts_with("log"), function(x) sum(x==F)))
# A tibble: 3 × 4
    one  log1  log2  log3
  <dbl> <int> <int> <int>
1     1     1     1     1
2     2     3     3     3
3     3     0     2     1

A nice and a little shorter way to do the logic is to use the booleans directly, as mentioned by @RuiBarradas

...
summarise(across(starts_with("log"), function(x) sum(!x)))
...

edited Feb 17 '22 at 15:50

answered Feb 17 '22 at 15:27

Andre Wildberg

12,344
3
12
29

2

Upvote but `sum(!x)` is simpler. – Rui Barradas Feb 17 '22 at 15:42
@RuiBarradas Very true! I always forget to utilize booleans directly! – Andre Wildberg Feb 17 '22 at 15:44

number of times specific value in each column R

1 Answers1