I do not understand question 10.10 (e).
Can I have any clue?
I am not sure how to help. Think about what a propensity score is, and how the mean propensity score in a study or subgroup (e.g. treated or untreated) would be computed: you would use a model to estimate the probability of treatment in each subject, then take the average of those probabilities in the group or subgroup. Does this help? I can't think of more of a hint without telling you the answer.
Tom