To what extent do you agree or disagree with the following statements about the paper by Evans and Schwab?
(a) The paper demonstrates causation using the data: Catholic schools have an effect on student graduation rates, other things being equal.
(b) The paper assumes causation: Catholic schools have an effect on student graduation rates, other things being equal. The paper assumes a specific functional form to implement the idea of causation and other things being equal-the probit model. The paper uses the data to estimate the size of the Catholic school effect.
(c) The graduation equation tests for interactions among explanatory variables in the selection equation.
(d) The graduation equation assumes there are no interactions.
(e) The computer derives the bivariate probit model from the data.
(f) The computer is told to assume the bivariate probit model. What the computer derives from the data is estimates for parameters in the model.