Ordinal model

is a model for a response variable which is a category taken from an ordered set of possibilities (for example: absent, slight, moderate, severe). These can be fitted by an extension of the generalized linear model algorithm.