Or alternatively self-reporters have overly narrow definition of cheating.

By the way I don't remember a single case where I cheated, but from my clear memory of my total lack of concern for "academic integrity" in high school, I infer that I'm extremely likely to have done so. It might sound weird, applying an outside view to own past, but my memory of things like that is extremely bad.

I considered the confusion to be one of frequency: "Do you cheat" vs "Have you ever cheated" vs "Did you cheat within the last year". I find 2/3rds suspiciously low for the latter. Then again, my friends in school wouldn't believe me that I'd really never shoplifted :)