uppose you wanted to evaluate the use of vouchers for students to attend any school they like on outcomes. Could you get a good sense of the benefits by comparing outcomes for 1 students who use vouchers with others (good here means like what you would get if you conducted an experiment)? Why or why not