Chapter 10 Selection issues

There are known knowns. These are things we know that we know. There are known unknowns. That is to say, there are things that we know we don’t know. But there are also unknown unknowns. There are things we don’t know we don’t know.

— Donald Rumsfeld

This chapter changes the perspective to the rows of the data matrix. An important consequence of nonresponse and missing data is that the remaining sample may not be representative any more. Multiple imputation of entire blocks of variables (Section 10.1) can be useful to adjust for selective loss of cases in panel and cohort studies. Section 10.2 takes this idea a step further by appended and imputing new synthetic records to the data. This can also work for cross-sectional studies.