Question - A way to deal with the previous issue is to do an instrumental variables estimation where
A) the instrument Z takes the value of 1 if the student was assigned a laptop (and zero otherwise), and the treatment variable X takes the value of 1 if the laptop was actually used (and zero otherwise)
B) the instrument Z takes the value of 1 if the laptop was actually used (and zero otherwise), and the treatment variable X takes the value of 1 if the student was assigned a laptop(and zero otherwise)
In the event that a laptop is assigned to a student, it increases his chances of passing the exam. That is, in both cases (A) and case (B), the presence of such a student is guaranteed.
Comments
Leave a comment