Standard Article

Ordered Categorical Data

  1. Mark P. Becker

Published Online: 15 JUL 2005

DOI: 10.1002/0470011815.b2a10041

Encyclopedia of Biostatistics

Encyclopedia of Biostatistics

How to Cite

Becker, M. P. 2005. Ordered Categorical Data. Encyclopedia of Biostatistics. 6.

Author Information

  1. University of Michigan, Ann Arbor, MI, USA

Publication History

  1. Published Online: 15 JUL 2005

Abstract

Many categorical variables have a natural ordering of the categories. Special methods are available for such data that are more powerful and more parsimonious than methods that ignore the ordering. This article discusses ways of modeling association for ordered categorical data and gives examples of them. These include loglinear models, related association models for two-way tables that have row and column effects, and logit regression models for an ordinal response variable and a set of explanatory variables. The best-known logit model, called a proportional odds model, uses the same effect parameters for each set of logits of cumulative probabilities for the ordered categorical response variable.

Keywords:

  • association model;
  • logit models;
  • loglinear model;
  • multinomial distribution;
  • odds ratio;
  • ordinal variable;
  • proportional odds model;
  • quasi-symmetry