It would be helpful to have the system send surveys to random students who were marked as "attended" either each week or even each day. For example, if 10 students came, the survey would go out to 3 of them.