djnavarro / codecat

Codebooks For Categorical Data

codecat

The goal of codecat is to provide a minimal toolkit for defining and using codebooks (also called data dictionaries) which specify factor levels and labels for one or more (usually categorical) variables in a data set.

Installation

You can install the development version of codecat like so:

remotes::install_github("djnavarro/codecat")

Example

library(codecat)
library(tibble)

To specify a coding scheme for a variable, use the code() function. A collection of coding schemes that should be applied to a data set is called a codebook, and is created using codebook():

codebook(
  code("SEX", levels = c(0, 1), labels = c("Male", "Female")),
  code("STUDY", levels = c(1, 2, 3), labels = c("S1", "S2", "S3"))
)
#> $SEX
#> # A tibble: 2 × 3
#>   variable level label 
#>   <chr>    <dbl> <chr> 
#> 1 SEX          0 Male  
#> 2 SEX          1 Female
#> 
#> $STUDY
#> # A tibble: 3 × 3
#>   variable level label
#>   <chr>    <dbl> <chr>
#> 1 STUDY        1 S1   
#> 2 STUDY        2 S2   
#> 3 STUDY        3 S3

Apply it to data using encode():

dat <- tibble(
  STUDY = c(1, 1, 1, 2, 2, 2),
  SEX   = c(0, 0, 1, 1, 1, 0),
  AGE   = c(32, 18, 64, 52, 26, 80)
)

labels <- codebook(
  code("SEX", levels = c(0, 1), labels = c("Male", "Female")),
  code("STUDY", levels = c(1, 2, 3), labels = c("S1", "S2", "S3"))
)

encode(dat, labels)
#> # A tibble: 6 × 3
#>   STUDY SEX      AGE
#>   <fct> <fct>  <dbl>
#> 1 S1    Male      32
#> 2 S1    Male      18
#> 3 S1    Female    64
#> 4 S2    Female    52
#> 5 S2    Female    26
#> 6 S2    Male      80

To restore original levels, use decode():

dat2 <- encode(dat, labels)
decode(dat2, labels)
#> # A tibble: 6 × 3
#>   STUDY   SEX   AGE
#>   <dbl> <dbl> <dbl>
#> 1     1     0    32
#> 2     1     0    18
#> 3     1     1    64
#> 4     2     1    52
#> 5     2     1    26
#> 6     2     0    80