As rkg pointed out briefly above, I think that you should maybe consider more closely the difference between misspellings and typos. I have been working for a very long time in publishing and translation (Perl is Just Another Hobby). To take an example, there is a difference between people who will submit a manuscript containing a spelling mistake such as ‘definately’ and those who might let through a typo such as ‘deifnitely’. There are, of course, borderline cases. But I think that in most cases the difference is clear.

Another example: using pen and ink (admittedly an increasingly less frequent phenomenon), I will never write ‘teh’ (for ‘the’), but this happens frequently when I’m typing on a keyboard. Same thing for the classic ‘from/form’ confusion.

I am not up to date with current research, but back when I was, most efforts were focused on spelling (and how to ‘improve’ it) rather than on typos.

At the risk of straying even further OT, it would be interesting to see research on how spellcheckers have affected spelling habits. Back in the old days, faced with a word that I was unsure about (let’s say ‘vicissitudes’), I would get out a dictionary. Now, I type some approximation into M$ Word, and if it comes out with a squiggly red underline underneath, right-click on it in the hope that the spellchecker will suggest the right spelling...

dave