data file character/word count

I've gotten the basic task set up: participants are required to retype paragraphs (embedded as bitmap backgrounds) without pressing certain keys, while their own typing is invisible off screen. I am successfully using the recorded keystroke parameter (for example parameter k69 to record the letter e) to record participants failure to follow instructions, however I would like to be able to compare this to how many characters are typed overall. The utilities folder/keycodes function does not seem to have an "all keystroke" designation to record every letter typed in a free response (essay format), but only a single individually specified key per question. I do not need RT data for the remaining letters, just a total count to calculate a proportion of mistakes (forbidden keys, already recorded) to total characters/keys typed. I know that I could pull the .txt data files individually, paste into word, and conduct word counts that way, but it seems that there MUST be a simple way to get this number directly out of ML somehow...am I missing something?

Thanks.
Beka

So am I understanding correctly that you want a letter and/or word count for the open ended essay questions? If it were the fill-in-the-blank item type, I think you could do this pretty easily in Excel or SPSS as the data would already be in the primary data files and I'm guessing both programs have a function to calculate the characters or words in a cell or variable. But because essay data are written to separate files (due to their potential length) I don;t see an easy way around your paste into Word idea. Or paste into Excel or SPSS as cells in your data file. Another possibility is using the Fill-in-the-blank item type but that would limit you to 255 characters. If that would work, you could just make the blank multi-line to accommodate the extra text. A custom HTML item could also pull this off.

Some initial thoughts--hope that helps.