ABSTRACT
Wrapping up a book like this is diffi cult. We’ve come a very long way, and you are now well versed in a lot of the statistical techniques that are used in the real world of social research. But social statistics is a huge fi eld, and there is so much more we could cover. So for this last chapter, I’ve decided on a compromise. There are two more topics I’m going to cover, but I’m going to do so at what I call an “awareness” level. I want you to leave this chapter with an awareness of these two topics, but I will not go into the nitty-gritty details as we have done in previous chapters. The fi rst part of the chapter concerns the fi eld of regression diagnostics, which is basically what it sounds like: diagnosing the health of your regression model and, if your model is “sick,” how do you tell what kind of “sickness” it has? This sickness may come from the fact that your regression model violates a regression assumption, such as when you try to use linear regression on a nonlinear relationship. The second part of
the chapter briefl y introduces you to some of the most popular advanced techniques that are being used in contemporary social research. Again, the goal of this chapter is not to teach you the details of these techniques. Rather, it is designed so that, in your future career when you are using statistics, if someone says “multicollinearity” or “multilevel modeling”, you’ll at least know what this person is saying and why they are bringing it up.
