codebreak

Quickly and easily label your data using codebooks saved as a YAML text file.

Installation

remotes::install_github("nt-williams/codebreak")

or

# Install 'codebook' from 'nt-williams' universe
install.packages('codebreak', repos = 'https://nt-williams.r-universe.dev')

Applying a code book

some_data <- data.frame(
     x = c(1, 2, 5, 3, 4, 1),
     y = c(0, 1, 1, 0, 1, 9), 
     z = c(5.2, 3.1, 5.6, 8.9, 9.0, 7.2), 
     w = c(1, 1, 0, 1, 1, 1)
)

Codebooks are created as YAML text files and are saved in the project directory (or somewhere else) as codebook.yml (or as something else).

x:
  label: Variable X # include meaningful variable descriptions
  cb:
    1: These        # convert variable codes to labels
    2: Are
    3: Random
    4: Character
    5: Labels

"y":
  label: Variable Y
  cb: &binary        # reduce repetition with anchors
    0: "No"
    1: "Yes"
    9: null          # account for coded missing values

z:
  label: Variable Z

w:
  label: Variable W
  cb: *binary

Import and apply the codebook to the data:

cb <- codebreak::Codebook$new(system.file("codebook.yml", package = "codebreak"))
cb
#> codebook: /Users/nicholaswilliams/Library/R/arm64/4.4/library/codebreak/codebook.yml 
#> 
#> decode data with `<obj>$decode()`
#> label data with `<obj>$label()`

cb$decode(some_data)
#>           x    y   z   w
#> 1     These   No 5.2 Yes
#> 2       Are  Yes 3.1 Yes
#> 3    Labels  Yes 5.6  No
#> 4    Random   No 8.9 Yes
#> 5 Character  Yes 9.0 Yes
#> 6     These <NA> 7.2 Yes

Rename columns based on the codebook labels:

cb$label(some_data)
#>   Variable X Variable Y Variable Z Variable W
#> 1          1          0        5.2          1
#> 2          2          1        3.1          1
#> 3          5          1        5.6          0
#> 4          3          0        8.9          1
#> 5          4          1        9.0          1
#> 6          1          9        7.2          1

Apply the codebook and rename columns:

cb$decode(some_data, label = TRUE)
#>   Variable X Variable Y Variable Z Variable W
#> 1      These         No        5.2        Yes
#> 2        Are        Yes        3.1        Yes
#> 3     Labels        Yes        5.6         No
#> 4     Random         No        8.9        Yes
#> 5  Character        Yes        9.0        Yes
#> 6      These       <NA>        7.2        Yes

Integration with the labelled package

decode() and label() can return data with the codebook applied using the labelled package by setting as_labelled = TRUE.

some_data <- tibble::as_tibble(some_data)

cb$decode(some_data, as_labelled = TRUE)
#> # A tibble: 6 × 4
#>   x             y             z w        
#>   <dbl+lbl>     <dbl+lbl> <dbl> <dbl+lbl>
#> 1 1 [These]      0 [No]     5.2 1 [Yes]  
#> 2 2 [Are]        1 [Yes]    3.1 1 [Yes]  
#> 3 5 [Labels]     1 [Yes]    5.6 0 [No]   
#> 4 3 [Random]     0 [No]     8.9 1 [Yes]  
#> 5 4 [Character]  1 [Yes]    9   1 [Yes]  
#> 6 1 [These]     NA          7.2 1 [Yes]