We have to find a way of reversing the process: instead of going from known populations to saying something about possible samples, we need to go from a single sample back to saying something about a possible population. This is the process of inductive inference outlined in Chapter 3.
