Other ANOVA Models

The anovan function has name-value arguments that enable you to specify two types of ANOVA models:

• 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 mileage, which is the response variable. The model factor is nested within the factory factor, because each model is manufactured at only one factory.

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

  A nested model typically numbers the levels of a nested factor in the same way for each value of the factor it is nested within.

• The continuous argument specifies to treat some factors 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. For information about a tool you can use to fit this type of model, see Analysis of Covariance.