Other ANOVA Models

The anovan function also has name-value arguments that enable you to specify two other types of model terms:

• The nested name-value argument specifies a matrix that indicates which factors are nested within other factors. A nested factor has different levels for each level of the factor it is nested within. A nested model is a model that includes at least one nested factor.

  Suppose an automobile company manufactures six car models. The company has three factories, and each factory makes two distinct models. The car mileage can vary from factory to factory and from model to model. In this example, factory and model are factors that explain differences in the response variable mileage. The model factor is nested within the factory factor because each model is manufactured at only one factory.

  Factory	Car Model
  1	1
  1	2
  2	3
  2	4
  3	5
  3	6

  Nested models commonly number the levels of a nested factor the same way for each value of the factor it is nested within.

• 'continuous' argument specifies that some factors are to be treated as continuous variables. The remaining factors are categorical variables. Although the anovan function can fit models with multiple continuous and categorical predictors, the simplest model that combines one predictor of each type is known as an analysis of covariance model. Analysis of Covariance describes a specialized tool for fitting this model.