That wouldn't give you a control group, either, that would give you a new set of students, and would let you answer the question "how successful are students hw

That wouldn't give you a control group, either, that would give you a new set of students, and would let you answer the question "how successful are students hwo graduate without being crippled by student debt compered with dropouts and a control group of students how received no special assistance and graduated anyway." I would be very interested in this experiment, but subsidizing a student's education in the US doesn't give you a 'control' group, but a very privileged group.