Formatted 2x2 Contingency Table with Exact Test — table_two_by

Builds an institutional 2x2 contingency table with row counts, row percentages, marginal totals, odds ratio estimates, confidence intervals, and exact test p-values. If any cell count is zero, it automatically defaults to the mid-p version of Central Fisher's exact test (midp = TRUE).

Usage

table_two_by_two(
  data,
  row_var,
  col_var,
  row_label = row_var,
  col_label = col_var,
  midp = NULL,
  conf.level = 0.95
)

Arguments

data: Data frame containing the categorical variables.
row_var: Character name of the row variable.
col_var: Character name of the column variable.
row_label: Optional display label for the row variable.
col_label: Optional display label for the column variable.
midp: Logical or NULL (default). If NULL, defaults to TRUE when any cell in the 2x2 table is zero, and FALSE otherwise.
conf.level: Confidence level for the odds ratio confidence interval (default 0.95).

Value

A list containing:

table: Tibble with formatted cells and marginal totals.
p_value: Exact test p-value.
estimate: Estimated odds ratio.
conf.int: Confidence interval for the odds ratio.
method: Test method name.
midp: Logical indicating whether mid-p adjustment was used.
has_zero: Logical indicating whether any table cell had zero count.
note: Formatted institutional table note.