Nonignorable missing data: in Exercise 9.13, you estimated the effects of incumbency in U.S. congressional elections, discarding uncontested elections.
(a) Construct three “bad” imputation procedures and one “good” imputation procedure for these uncontested elections.
(b) Define clearly how to interpret these imputations. (These election outcomes are not actually “missing”—it is known that they were uncontested.)
(c) Fit the model to the completed dataset under each of the imputation procedures from (a) and compare the results.
Already registered? Login
Not Account? Sign up
Enter your email address to reset your password
Back to Login? Click here