Recode values using a wide formatted data dictionary

Package: dplyr

---

Function: case_when()

---

1. Recode values using a wide formatted data dictionary

Review the data (d24)

# A tibble: 3 x 3
  stu_id    q1    q2
   <dbl> <dbl> <dbl>
1      1     1     1
2      2   -99     1
3      3   -99   -99

Review our current wide formatted data dictionary (dict_wide)

# A tibble: 3 x 4
  var_name type    raw_value                        transformation
  <chr>    <chr>   <chr>                            <chr>         
1 stu_id   numeric 1-10                             NA            
2 q1       numeric 1 = Selected, -99 = Not selected -99 -> 0      
3 q2       numeric 1 = Selected, -99 = Not selected -99 -> 0      

Select all variables with the same recoding logic (-99 -> 0)

recode_multi_select <- dict_wide %>%
  dplyr::filter(transformation == "-99 -> 0") %>%
  dplyr::pull(var_name)

recode_multi_select

[1] "q1" "q2"

Now we can use that vector of variable names to recode these variables ll at once.

d24 %>%
  dplyr::mutate(dplyr::across(
    dplyr::all_of(recode_multi_select),
                ~dplyr::case_when(
                  . == -99 ~ 0,
                  TRUE ~ .
                )))

# A tibble: 3 x 3
  stu_id    q1    q2
   <dbl> <dbl> <dbl>
1      1     1     1
2      2     0     1
3      3     0     0

Go to Recode using long formatted data dictionary

Return to Recode