square_table: Force a square version of base::table()

In emilelatour/lamisc: Latour's Little Helpers

Force a square version of base::table()

Description

base::table() will not return a square table for a confusion matrix if a fator level is not present in one of the variables. This will return a square table.

Usage

square_table(x, y)

Arguments

x	"X" variable; counts appear in the rows of the table
y	"Y" variable; counts appear in the columns of the table

Value

A table

References

https://stackoverflow.com/questions/5558745/force-table-to-include-all-factors-from-both-arrays-in-r

Examples

library(tibble)

#### Example 1 --------------------------------

df <- tibble::tribble(
  ~a, ~b, ~c,
  1L, 1L, 1L,
  0L, 1L, 1L,
  1L, 1L, 0L,
  0L, 1L, 0L,
  1L, 1L, 1L,
  0L, 1L, 1L,
  1L, 1L, 0L
)

table(df$a, df$b)
square_table(df$a, df$b)


#### Example 2 --------------------------------

set.seed(8675309)

df2 <- tibble::tibble(
  x = factor(sample(0:9, 100, TRUE)),
  y = factor(sample(3:7, 100, TRUE))
)

with(df2, table(x, y))

with(df2, square_table(x, y))
with(df2, square_table(y, x))


#### Example 3 --------------------------------

set.seed(8675309)

df3 <- tibble::tibble(
  x = factor(sample(0:2, 100, TRUE)),
  y = factor(sample(3:5, 100, TRUE))
)

with(df3, table(x, y))

with(df3, square_table(x, y))
with(df3, square_table(y, x))

emilelatour/lamisc documentation built on April 9, 2024, 10:33 a.m.