tidyverse - Using summarise_at in order to sum a column

General
1

Hello,
I want to avoid declaring a loop. I am creating some dummies, and then, using summarise_at I would like just to sum the column "fact" for each vars I created. I tried something like this:

data=data %>% mutate(s1=if_else(clasifier==1,1,0),s2=if_else(clasifier==21,1,0),
                       s3=if_else(clasifier %in% 3:4,1,0),s4=if_else(clasifier %in% 5:6,1,0),s5=if_else(clasifier==7,1,0))
data %>% group_by(zone) %>% summarise_at(vars(s1:s5),sum(fact,na.rm=TRUE))

But I receive a message
Error in is_fun_list(.funs) : objet 'fact_cal' introuvable

What am I doing wrong?
I mean, I plan to sum fact for s1,s2,...,s5 and grouping that operation by zone.
Is it possible? Is there an alternative?
Thanks for your time and interest.

2

What does your dataset look like? Can you provide an example?

1 Like
3

The data is here:
https://www.mediafire.com/file/zs72gqecm3f2cia/datos.zip/file

4

It is blocked for me (from work). Can you post a reproducible example?

5 
datos=datasets::npk
datos=datos %>% mutate(s1=if_else(block==1,1,0),s2=if_else(block==2,1,0),
                       s3=if_else(block %in% 3:4,1,0),s4=if_else(block %in% 5:6,1,0))

datos %>% group_by(N) %>% summarise_at(vars(s1:s4),sum(yield),na.rm=T)

The idea is recreate the results from this code

datos %>% group_by(N,s1) %>% summarise(suma=sum(yield))
datos %>% group_by(N,s2) %>% summarise(suma=sum(yield))
datos %>% group_by(N,s3) %>% summarise(suma=sum(yield))
datos %>% group_by(N,s4) %>% summarise(suma=sum(yield))
datos %>% group_by(N,s5) %>% summarise(suma=sum(yield))

But I don't want to write four times the same code, and I don't want to declare a loop.
I hope my explanation is clear.
Thanks for your time, guys.

6

Try this.

datos %>% 
  pivot_longer(s1:s4) %>%
  group_by(N, name, value) %>% 
  summarize(suma = sum(yield), .groups = "drop")
1 Like
7

I should explain you don't want to summarize_at(vars(s1:s4)) because the column you want to summarize seems to be yield. It seems like instead you want to use s1:s4 as grouping variables, so it makes sense to pivot them so their values appear across rows and can be used in the group_by.

8

Thanks. It worked without error.
I will read the code in order to learn It well.
This way I will avoid writing loops.
Thanks again, arthur.t

9

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.