How to recode multiple columns that are characters into numeric at once?-CodePudding

I'm new here so apologies if I make a couple mistakes here but essentially, I am trying to recode a lot of columns that have to do with the same content but at different time points. Therefore, I'm trying to find a way to recode multiple columns that have the same corresponding responses (if that makes sense). Hopefully this makes sense but for example, if I was trying to gather peoples' fruit consumption at 5 different time points but I wanted to only focus on 4 types of fruit - apple (1), banana (2), orange (3), and strawberry (4) and this was the data I had:

  id   Fruit_T1   Fruit_T2   Fruit_T3 Fruit_T4
1  1      apple     banana      apple     kiwi
2  2     banana      apple strawberry     <NA>
3  3     orange strawberry       kiwi    apple
4  4 strawberry     orange       <NA>     <NA>
5  5     banana     banana      apple    apple
6  6     orange      apple strawberry  apricot

I am trying to get to this:

  id   Fruit_T1   Fruit_T2   Fruit_T3 Fruit_T4 RFruit_T1 RFruit_T2 RFruit_T3 RFruit_T4
1  1      apple     banana      apple     kiwi         1         2         1         .
2  2     banana      apple strawberry     <NA>         2         1         4      <NA>
3  3     orange strawberry       kiwi    apple         3         4         .         1
4  4 strawberry     orange       <NA>     <NA>         4         3      <NA>      <NA>
5  5     banana     banana      apple    apple         2         2         1         4
6  6     orange      apple strawberry  apricot         3         1         4         .

Where the "NA"'s are distinct from the '.' where the individual did consume fruit but it was not one of the 4 of interest. Sorry for the dumb example but I really appreciate any insight into this situation. Thanks very much!

CodePudding user response：

Using recode with across you could do:

library(dplyr)

rec_vec <- c(apple = "1", banana = "2", orange = "3", strawberry = "4")

dat |>
  mutate(across(!id, recode, !!!rec_vec, .default = ".", .names = "R{.col}"))
#>   id   Fruit_T1   Fruit_T2   Fruit_T3 Fruit_T4 RFruit_T1 RFruit_T2 RFruit_T3
#> 1  1      apple     banana      apple     kiwi         1         2         1
#> 2  2     banana      apple strawberry     <NA>         2         1         4
#> 3  3     orange strawberry       kiwi    apple         3         4         .
#> 4  4 strawberry     orange       <NA>     <NA>         4         3      <NA>
#> 5  5     banana     banana      apple    apple         2         2         1
#> 6  6     orange      apple strawberry  apricot         3         1         4
#>   RFruit_T4
#> 1         .
#> 2      <NA>
#> 3         1
#> 4      <NA>
#> 5         1
#> 6         .

DATA

dat <- structure(list(id = 1:6, Fruit_T1 = c(
  "apple", "banana", "orange",
  "strawberry", "banana", "orange"
), Fruit_T2 = c(
  "banana", "apple",
  "strawberry", "orange", "banana", "apple"
), Fruit_T3 = c(
  "apple",
  "strawberry", "kiwi", NA, "apple", "strawberry"
), Fruit_T4 = c(
  "kiwi",
  NA, "apple", NA, "apple", "apricot"
)), row.names = c(
  "1", "2",
  "3", "4", "5", "6"
), class = "data.frame")