Measuring Change in Regression

#1
We are working on a study to examine the effectiveness of a safety training program. Before and after the training, participants were asked to perform a number of tasks, and they were rated based on the total number of correct tasks observed. Before the training, the participants completed a questionnaire on their attitudes towards workplace safety and training. We want to see if participants’ attitudes affect the amount of pretest-posttest change.

Would it be appropriate to use absolute change (post-test score minus pre-test score) as the dependent variable in the regression analysis?

Your advice is greatly appreciated!