code for mutate_at to recode values in a vector of column names-CodePudding

Im trying to write a code, that will take colnames out of a given vector, and in each of these columns recode 1<-5,2<-4,3<-3,4<-2,5<-1 big5qflip is the vectore of columns that I need to recode out of a questionnaire, I'm having trouble getting it to work, and writing case_when correctly

big5new%>%mutate_at(vars(big5qflip),case_when(
  big5qflip==1~5,
  big5qflip==2~4,
  big5qflip==3~3,
  big5qflip==4~2,
  big5qflip==5~2
))

thank you!

when I tried the code above I got the follwing error Error in mutate_at(): ! .funs must be a one sided formula, a function, or a function name. Run rlang::last_error() to see where the error occurred.

CodePudding user response：

mutate_at has been superseded by across, so here's a solution using this function. Also notice the {{ ... }} to refer to the colnames in big5qflip:

library(dplyr)

big5new <- data.frame(
  X = c(1, 2, 1, 3, 2),
  Y = c(4, 4, 4, 4, 4),
  Z = c(1, 1, 1, 1, 1)
)

big5qflip <- c("X", "Y")

newcodes <- 5:1
names(newcodes) <- 1:5

big5new %>%
  mutate(
    across(
      {{ big5qflip }},
      ~ do.call(recode, append(list(.x =.x), newcodes))
    )
  )

CodePudding user response：

Here is another option. You can use factor to set the levels in the reverse order and then as.numeric to get the level index which is the inverse of the number.

library(tidyverse)

big5new <- data.frame(
  X = c(1, 2, 1, 3, 2),
  Y = c(4, 4, 4, 4, 4),
  Z = c(1, 1, 1, 1, 1)
)

big5qflip <- c("X", "Y")


big5new |>
  mutate(across(all_of(big5qflip), \(x) as.numeric(factor(x, levels = 5:1))))
#>   X Y Z
#> 1 5 2 1
#> 2 4 2 1
#> 3 5 2 1
#> 4 3 2 1
#> 5 4 2 1