replace NA values in tabyl() counts with 0?

[sfirke]

Only affects calls where there's an unrepresented level in a factor variable, so this is a niche case.

Instead of:

> tabyl(sorted_with_na_and_fac$grp, sort = TRUE)
  sorted_with_na_and_fac$grp  n percent valid_percent
1                          c  2    0.50     0.6666667
2                          a  1    0.25     0.3333333
3                          b NA      NA            NA
4                       <NA>  1    0.25            NA

Return:

> tabyl(sorted_with_na_and_fac$grp, sort = TRUE)
  sorted_with_na_and_fac$grp n percent valid_percent
1                          c 2    0.50     0.6666667
2                          a 1    0.25     0.3333333
3                          b 0    0.00     0.0000000
4                       <NA> 1    0.25     0.0000000

Could likely be done with:

  # replace NA values with zeroes
  result[is.na(result)] <- 0

Though I think I'd want to retain the NA representation of row = NA and column = valid_percent, to show that missing values is not in fact 0% once they've been filtered out.