One in eight women will get breast cancer, but who gets screened?

Answer: those less likely to get the disease, and when they do, it is caught at an earlier stage.

   

 The video suggests that if you are doing benefit-cost analysis of screening (Chapter 3), you will understate the benefits of screening because of selection bias (Chapter 17). 

To see this clearly, imagine that only people who don't have the disease get tested. In this extreme case, there is no information revealed by the test, i.e., the test has zero benefit and some cost.  The same intuition is behind the selection bias:  less information is revealed by testing a selected group who are less likely to have the disease than if the testing had been done randomly, i.e., over the population.  

The second part of the answer above, catching breast cancer at an earlier stage complicates this simple answer. If the benefits of catching it at an earlier stage are bigger than the benefits of catching it at a later stage, then this would militate in the opposite direction. 

 Here is a good formative quiz based on the video.

HT:  Marginal Revolution University