Counts the distinct combinations of missing and observed values across columns, one row per pattern, most common first. It is the tabular companion to [plot_comissingness_matrix()] and a pre-flight check before multiple imputation (see [pool_imputations()]): it shows which columns tend to be missing together and how many rows are complete.
Value
A tibble with one logical column per variable (`TRUE` = missing), then `n_missing_vars`, `n` (rows with the pattern) and `percent` (of all rows).
Examples
d <- data.frame(a = c(1, NA, 3, NA, 5, 6), b = c(1, NA, 3, NA, 5, NA), c = c(NA, 2, 3, 4, 5, 6))
tbl_missingness_patterns(d)
#> # A tibble: 4 × 6
#> a b c n_missing_vars n percent
#> <lgl> <lgl> <lgl> <int> <int> <dbl>
#> 1 FALSE FALSE FALSE 0 2 33.3
#> 2 TRUE TRUE FALSE 2 2 33.3
#> 3 FALSE FALSE TRUE 1 1 16.7
#> 4 FALSE TRUE FALSE 1 1 16.7