How to specify function arguments within mutate_at

Question

I want to convert multiple columns of a data.frame from the character class to the factor class while specifying the factor labels.

First I will create a dummy data.frame on which the code should such that others can reproduce the issue as follows:

df <- data.frame(replicate(2,sample(0:1,20,rep=TRUE)),
                 replicate(2,sample(0:2,20,rep=TRUE)))
names(df) <- c("xxx_var1", "xxx_var2", "yyy_var1", "yyy_var2")

I have managed to convert the desired columns as follows:

df <- df %>%
   mutate_at(vars(starts_with("xxx")), factor)

Now I want to specify arguments in the factor() function but I don't know how. If I try the following:

df <- df %>%
   mutate_at(vars(starts_with("xxx")), factor(labels = c("no", "yes"))

it returns the following error:

Error in factor(labels = c("no", "yes")) : invalid 'labels'; length 2 should be 1 or 0

Is it possible to specify the factor labels for all these columns at once using mutate_at?

Are you familiar with map functions from purrr package? Also, can you provide a sample of the data you are using to make the problem reproducible in our machines? You can do that using dput(head(df)) and copying the result from the console. — jpdugo17
– jpdugo17, Commented Dec 1, 2021 at 22:57
It would help to see a representative sample of the data in df as plain text e.g. the output from dput(df). — neilfws
– neilfws, Commented Dec 1, 2021 at 23:01
df %>% mutate_at(vars(starts_with("xxx")), factor, labels = c("no", "yes"))? — jdobres
– jdobres, Commented Dec 1, 2021 at 23:01
library(dplyr); df %>% mutate(across(starts_with("xxx"), ~factor(., levels = c("no", "yes")))) — Jon Spring
– Jon Spring, Commented Dec 1, 2021 at 23:26
@jdobres you solution worked, thanks a lot! Very clean and simple. If you could post your idea as an answer I can accept it as the proper solution. — The Jipsess
– The Jipsess, Commented Dec 2, 2021 at 10:04

jdobres · Accepted Answer · 2021-12-02 12:08:37Z

3

Provide factor's "labels" argument as a named argument to mutate_at:

df %>% 
  mutate_at(vars(starts_with("xxx")), factor, labels = c("no", "yes"))

However, the mutate_xxx functions have been superseded by the use of mutate and the new across function. As Jon Spring commented, the following is equivalent:

df %>% 
  mutate(across(starts_with("xxx"), ~factor(., levels = c("no", "yes"))))

answered Dec 2, 2021 at 12:08

jdobres

12k1 gold badge20 silver badges41 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

How to specify function arguments within mutate_at

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related