- Explain how a randomized encouragement design leads to an exogenous (random) difference in the proportion of the treatment and control groups which take up an intervention.
a. Provide an example of a situation in which a normal randomized experiment is not feasible but an encouragement design is.
3. Does correlation equal causation? Yes or No?
a. Explain why the fact that more people die in a hospital than those who don’t go to
a hospital doesn’t mean that hospitals kills people.
Categories: