Question

This seems simple enough but I could not find an easy answer anywhere.

If we have the following data and use table():

df <- data.frame(x=c("a","b","a"),y=c("b","a","c"))

table(df)
   y
x   a b c
  a 0 1 1
  b 1 0 0

whats the easiest way to get:

   y
x   a b c
  a 0 1 1
  b 1 0 0
  C 0 0 0

without adding the c row manually after.

Answer 1

Here is one way. Essentially, you need to make sure that x and y both have the same levels. I do that by computing the union() of their respective levels() and then transform() the original x and y to both have this common set of levels:

> df <- data.frame(x=c("a","b","a"),y=c("b","a","c"))
> lev <- with(df, sort(union(levels(x), levels(y))))
> lev
[1] "a" "b" "c"
> df <- transform(df, x = factor(x, levels = lev), y = factor(y, levels = lev))
> table(df)
   y
x   a b c
  a 0 1 1
  b 1 0 0
  c 0 0 0

Your example has an easy solution – just give x the same levels as y, as the levels of y are complete. In more general situations where neither x nor y has a complete set of levels, the code I show will get that complete set for you and thus works in both situations.