Render Cell Suppression (Categorical)

This is a custom render for table1 categorical variables which performs GEMINI "cell suppression" for any variable levels which contain fewer than 6 observations.

If the total number of these variable levels with fewer than 6 observations is less than 6, all cells for all levels of the variable must be censored (because it is possible to indirectly deduce the missing counts otherwise).

Usage

render_cell_suppression.categorical(x, ...)

Arguments

x

(character or factor)
A categorical variable to summarize.

...

Optionally accept a named digits (integer) or single_level_binary (logical) argument which specifies the number of digits to round percentages to. Also accepts additional inputs that are passed to prettyNum() to apply additional formatting to shown results (e.g., big.mark = ",").

Value

named (character)
Concatenated with "" to shift values down one row for proper alignment.

Examples

x <- factor(c(rep("a", times = nrow(mtcars)), "b"), levels = c("a", "b"))
render_cell_suppression.categorical(x)
#>                             a              b 
#>             "" "(suppressed)" "(suppressed)" 

x2 <- factor(c(rep("a", times = nrow(mtcars))), levels = c("a", "b"))
render_cell_suppression.categorical(x2)
#>                           a             b 
#>            "" "32 (100.0%)"    "0 (0.0%)" 

y <- factor(
  c(rep("a", times = nrow(mtcars)), "b", "c", "d", "e", "f", "g"),
  levels = c("a", "b", "c", "d", "e", "f", "g")
)
render_cell_suppression.categorical(y)
#>                                                     a 
#>                         ""               "32 (84.2%)" 
#>                          b                          c 
#> "&lt; 6 obs. (suppressed)" "&lt; 6 obs. (suppressed)" 
#>                          d                          e 
#> "&lt; 6 obs. (suppressed)" "&lt; 6 obs. (suppressed)" 
#>                          f                          g 
#> "&lt; 6 obs. (suppressed)" "&lt; 6 obs. (suppressed)" 

z <- factor(
  c(
    rep("a", times = 100),
    rep("b", times = 50),
    rep("c", times = 7),
    rep("d", times = 2)
  ),
  levels = c("a", "b", "c", "d", "e")
)

render_cell_suppression.categorical(z)
#>                             a              b              c              d 
#>             ""  "100 (62.9%)"   "50 (31.4%)" "(suppressed)" "(suppressed)" 
#>              e 
#>     "0 (0.0%)"