I like your thesis, and worry about your control group. I think comparing inmates to officers would be interesting, but I don't think prison officers is a good control group since they don't represent the general populace. They are a self-selected, not a random sample. Perhaps you could collect scores from a third group?