Sumarize many columns by treatment

Question

I have a large data set with many columns and I am trying to find the mean and st of each column per treatment.

Example data

treatment	age	experience	grade	index
a	21	3	4	0,7
a	24	4	4	0,5
b	32	12	5	0,6
b	54	23	3	0,7
c	15	1	4	0,9
c	19	3	5	0,4

I tried:

db%>%
  group_by(treatment) %>% 
  summarise(mean=mean(age),sd=sd(age), .groups = "drop")

Is there a way to summarise all the columns without adding them individually? (my dataset has 56 columns)

Thank you for your help

Try `db %>% group_by(treatment) %>% summarise(across(where(is.numeric), list(Mean = mean, SD = sd)))` — Darren Tsai, May 09 '22 at 12:10
Thank both that the link was very helpful to understand the next step that I need to do. — Gonzalo de Quesada, May 09 '22 at 12:14

score 0 · Answer 1 · answered May 09 '22 at 12:13

0

df %>%
  group_by(treatment) %>%
  summarise(across(1:3, 
                ~sd(.), .names = '{col}_sd'),
            across(1:3, 
                ~mean(.), .names = '{col}_mean'))

answered May 09 '22 at 12:13

Chris Ruehlemann

20,321
4
12
34

Sumarize many columns by treatment

1 Answers1