We’ll start talking about regression models. First we will do an activity to suggest variables that could be modeled using a regression analysis. The activity will require students to identify a response variable, the appropriate distribution for the response variable (binomial, Poisson, and normal), and a brainstorm of what explanatory variables would affect the distribution of the response variable.
The main purpose of this activity is to get you to start thinking about what type of regression model you will need to use. The type of regression model will depend on the type of response variable:
- binomial -> logistic regression
- Poisson -> Poisson regression
- normal -> (linear) regression
The secondary purpose of this activity is to get you to start brainstorming what type of explanatory variables would affect the distribution of the response. These explanatory variables should be both measurable and have a relatively large effect on the distribution of the response.