using mutate_at with the in operator %in%

问题

I have a data frame with a few variables to reverse code. I have a separate vector that has all the variables to reverse code. I'd like to use mutate_at(), or some other tidy way, to reverse code them all in one line of code. Here's the dataset and the vector of items to reverse

library(tidyverse)
mock_data <- tibble(id = 1:5,
       item_1 = c(1, 5, 3, 5, 5),
       item_2 = c(4, 4, 4, 1, 1),
       item_3 = c(5, 5, 5, 5, 1))

reverse <- c("item_2", "item_3")

Here's what I want it to look like with only items 2 and 3 reverse coded:

library(tidyverse)

solution <- tibble(id = 1:5,
                   item_1 = c(1, 5, 3, 5, 5),
                   item_2 = c(2, 2, 2, 5, 5),
                   item_3 = c(1, 1, 1, 1, 5))

I've tried this below code. I know that the recode is correct because I've used it for other datasets, but I know something is off with the %in% operator.

library(tidyverse)
mock_data %>%
  mutate_at(vars(. %in% reverse), ~(recode(., "1=5; 2=4; 3=3; 4=2; 5=1")))

Error: `. %in% reverse` must evaluate to column positions or names, not a logical vector

Any help would be appreciated!

回答1:

You can give reverse directly to mutate_at, no need for vars(. %in% reverse). And I would simplify the reversing as 6 minus the current value.

mock_data %>% mutate_at(reverse, ~6 - .)
# # A tibble: 5 x 4
#      id item_1 item_2 item_3
#   <int>  <dbl>  <dbl>  <dbl>
# 1     1      1      2      1
# 2     2      5      2      1
# 3     3      3      2      1
# 4     4      5      5      1
# 5     5      5      5      5

If there's a possibility that reverse includes columns that are not in mock_data, and you want to skip those, use mutate_at(vars(one_of(reverse)), ...)

来源：https://stackoverflow.com/questions/59416554/using-mutate-at-with-the-in-operator-in

标签

dplyr

mutate