Dissect a dataset before regression

Returns a tidy summary of each variable's structure, missingness, uniqueness, and suitability for use in regression models.

Usage

dissect(data, verbose = FALSE, format = c("flextable", "gt", "tibble"))

Arguments

data

A data frame.

verbose

Logical; if TRUE, print the summary and interpretation notes. The tibble is returned invisibly only when printed by the console.

format

Output format. One of "flextable" (default), "gt", or "tibble". Use format = "tibble" for pipeline-friendly raw output.

Value

A tibble, gt_tbl, or flextable, depending on format. The tibble has columns: Variable, Type, Missing ( Levels, Compatibility, and Hint.

Examples

dissect(data_birthwt, format = "tibble")
#> # A tibble: 10 × 7
#>    Variable Type    `Missing (%)` Unique Levels Compatibility Hint              
#>    <chr>    <chr>   <chr>          <int> <chr>  <chr>         <chr>             
#>  1 low      integer 0%                 2 -      maybe         Two numeric value…
#>  2 age      integer 0%                24 -      compatible    Numeric variable …
#>  3 lwt      integer 0%                75 -      compatible    Numeric variable …
#>  4 race     integer 0%                 3 -      compatible    Numeric variable …
#>  5 smoke    integer 0%                 2 -      maybe         Two numeric value…
#>  6 ptl      integer 0%                 4 -      compatible    Numeric variable …
#>  7 ht       integer 0%                 2 -      maybe         Two numeric value…
#>  8 ui       integer 0%                 2 -      maybe         Two numeric value…
#>  9 ftv      integer 0%                 6 -      compatible    Numeric variable …
#> 10 bwt      integer 0%               131 -      compatible    Numeric variable …
dissect(data_birthwt, format = "gt")


  
Dataset dissection before regression
    
Variable
      
Type
      
Missing (%)
      
Unique
      
Levels
      
Compatibility
      
Hint
    
low
integer
0%
2
-
maybe
Two numeric values; confirm binary coding or convert to factor.
age
integer
0%
24
-
compatible
Numeric variable can be used as continuous.
lwt
integer
0%
75
-
compatible
Numeric variable can be used as continuous.
race
integer
0%
3
-
compatible
Numeric variable can be used as continuous.
smoke
integer
0%
2
-
maybe
Two numeric values; confirm binary coding or convert to factor.
ptl
integer
0%
4
-
compatible
Numeric variable can be used as continuous.
ht
integer
0%
2
-
maybe
Two numeric values; confirm binary coding or convert to factor.
ui
integer
0%
2
-
maybe
Two numeric values; confirm binary coding or convert to factor.
ftv
integer
0%
6
-
compatible
Numeric variable can be used as continuous.
bwt
integer
0%
131
-
compatible
Numeric variable can be used as continuous.
Screening aid only; review coding, missingness, sparse levels, and study context before modeling.
    

# Print notes that help interpret the data structure
dissect(data_birthwt, verbose = TRUE, format = "tibble")
#> # A tibble: 10 × 7
#>    Variable Type    `Missing (%)` Unique Levels Compatibility Hint              
#>    <chr>    <chr>   <chr>          <int> <chr>  <chr>         <chr>             
#>  1 low      integer 0%                 2 -      maybe         Two numeric value…
#>  2 age      integer 0%                24 -      compatible    Numeric variable …
#>  3 lwt      integer 0%                75 -      compatible    Numeric variable …
#>  4 race     integer 0%                 3 -      compatible    Numeric variable …
#>  5 smoke    integer 0%                 2 -      maybe         Two numeric value…
#>  6 ptl      integer 0%                 4 -      compatible    Numeric variable …
#>  7 ht       integer 0%                 2 -      maybe         Two numeric value…
#>  8 ui       integer 0%                 2 -      maybe         Two numeric value…
#>  9 ftv      integer 0%                 6 -      compatible    Numeric variable …
#> 10 bwt      integer 0%               131 -      compatible    Numeric variable …
#> 
#> Interpretation notes:
#> - compatible: ready to use in regression or with minimal preparation
#> - maybe: check coding, sparse levels, or whether transformation is needed
#> - incompatible: not usable as-is (e.g., all NA or no variation)
#> # A tibble: 10 × 7
#>    Variable Type    `Missing (%)` Unique Levels Compatibility Hint              
#>    <chr>    <chr>   <chr>          <int> <chr>  <chr>         <chr>             
#>  1 low      integer 0%                 2 -      maybe         Two numeric value…
#>  2 age      integer 0%                24 -      compatible    Numeric variable …
#>  3 lwt      integer 0%                75 -      compatible    Numeric variable …
#>  4 race     integer 0%                 3 -      compatible    Numeric variable …
#>  5 smoke    integer 0%                 2 -      maybe         Two numeric value…
#>  6 ptl      integer 0%                 4 -      compatible    Numeric variable …
#>  7 ht       integer 0%                 2 -      maybe         Two numeric value…
#>  8 ui       integer 0%                 2 -      maybe         Two numeric value…
#>  9 ftv      integer 0%                 6 -      compatible    Numeric variable …
#> 10 bwt      integer 0%               131 -      compatible    Numeric variable …