STATS MAKE ME CRY

I am testing a theoretical framework suggesting that perceived deception changes mental representations, and that revised mental representations will produce different judgments about misstatement risk, requisite audit effort, and fraud risk. My chairman has asked me to test for a positive association between mental representations and the risk/audit effort measures while controlling for differences in trait skepticism that you measured before the task began.

My participants produced assessments of their mental representations at time 1 and time 2. the responses were classified as positive or negative and converted into an individual score for time 1 and time 2. I have three DVs that are measures of audit risk also measured at time 1 and 2. These measures use 7 point Likert scales. Additionally, a survey score provides a measure of trait skepticism at time 1. How do I go about testing the association between mental representation and its effects on the dependent variables measuring risk most efficiently?