Dod Go Nuts?

The general linear model is a generalized linear model with a normal response distribution and what's called an identity link function. You can have counts, dummy codes or effect codes to capture nominal grouping variables. In my neck of the woods, we have things that are called zero inflated models. So i'm looking at drug use in adolescence. Thank goodness most kids don't use. But what you get is a piling up at zero, a zero inflation. Well, building a model for if you're not zero, then modelling how many are you? I think it's really exciting to think about the ways that we can do this,. It's like, dude, take a deep breath, lean back in