9.8 Next steps

This is a lot to take in. But this stuff is important for everything we do in statistics. Do this stuff before you start fitting all kinds of models because it is important to think about ahead of time! Examination of residual plots should become second nature to you in these analyses because it is the most powerful tool you have for testing assumptions. Don’t freak out if things don’t look perfect (they almost never will), and realize that there may be ways of dealing with violations within the context of linear models. If not, there certainly are other models designed specifically for this purpose.

In Chapter 10 we will continue to unpack the linear model as we talk about how to communicate the results of the models after we fit them and validate assumptions.