Recode observation in column for a specific ID

Question

I have a dataset, called 'survey'. In this I have rows of individual IDs, and columns with many questions. I need to recode the value in 1 column as NA and move the observation to the other column.

For example:

ID    Fruit    Vegetable
aaa   NA       grape 
bbb   NA       tomato
ccc   apple    NA
ddd   peach    NA

I want to change the grape and tomato observations, belonging to ID aaa and bbb to put them into the fruit column (survey respondents put them in the wrong column) and leave NA behind.

To look like:

ID    Fruit    Vegetable
aaa   grape    NA 
bbb   tomato   NA
ccc   apple    NA
ddd   peach    NA

Thank you very much

Answer 1

Basically, we'll use 2 conditionals to accomplish this. The first will check if it's NA and in your list of names to recode (fct2recode). The second will remove it from the second column if it was in fct2recode.

library(tidyverse)

df <- read_table("ID    Fruit    Vegetable
aaa   NA       grape 
bbb   NA       tomato
ccc   apple    NA
ddd   peach    NA")

fct2recode <- c("grape", "tomato")

df %>%
    mutate(Fruit = ifelse(is.na(Fruit) & Vegetable %in% fct2recode, 
                          Vegetable, Fruit),
           Vegetable = ifelse(Vegetable %in% fct2recode, NA, Vegetable))

Which results in:

# A tibble: 4 x 3
  ID    Fruit  Vegetable
  <chr> <chr>  <chr>    
1 aaa   grape  NA       
2 bbb   tomato NA       
3 ccc   apple  NA       
4 ddd   peach  NA   

I hope this approach is generalizable enough for your problem.

EDIT:

To specific target obs with IDs "aaa" and "bbb", you can add them as condition in the ifelse:

df %>%
    mutate(Fruit = ifelse(ID %in% c("aaa", "bbb"), Vegetable, Fruit),
           Vegetable = ifelse(ID %in% c("aaa", "bbb", NA, Vegetable)))