Ordinal Categorical Data

Website for ANALYSIS OF ORDINAL CATEGORICAL DATA, 2nd edition

For the second edition of Analysis of Ordinal Categorical Data by Alan Agresti (Wiley, 2010), this site contains (1) data sets for a couple of examples in the form of complete SAS programs for conducting the analyses, (2) examples of the use of various R functions (especially Thomas Yee's vglm function) for modeling ordinal data (3) examples of the use of Joe Lang's mph.fit R function for conducting various analyses in the book, (4) links for some Stata examples, and (5) corrections of errors in early printings of the book.

1. Data: Datasets contains data sets not shown completely in the text, some of them in the form of SAS programs for conducting the analyses.

2. R: Use of R for ordinal models is a pdf file I have prepared of examples of the use of R, including cumulative logit models, cumulative logit models without proportional odds, adjacent-categories logit model, continuation-ratio logit model, cumulative probit model, complementary log-log model, and use of GEE for repeated measurement data with cumulative logit model. 3. R mph.fit: For examples of the use of the mph.fit R function for the analysis of standardized residuals for a cumulative logit model in Section 3.5.8, the ML partial proportional odds analysis in Section 3.6.5, ML fitting of the mean response model in Section 5.6.2, the ML fitting of the global odds ratio uniform association model in Section 6.6.2, ML fitting of the marginal cumulative logit model for a square table discussed in Section 8.2.1, the ML fitting of the marginal cumulative logit model to the crossover data in Section 8.4.3, and the ML fitting of the paired preference models of Section 8.6.4 that permit responses to be dependent, click on mph.fit R analyses. The function and a detailed manual can be obtained from Joseph Lang's home page.

4. Stata: For those who use Stata, you may find it helpful to refer to examples of categorical data analyses for the first edition of my text "An Introduction to Categorical Data Analysis" at the site set up by the UCLA Statistical Computing Center. In particular, Chap 7 has an example of the linear-by-linear association model, Chap 8 has an example of the cumulative logit model with proportional odds structure, and Chap 9 has an example of ordinal quasi-symmetry. In Stata, the ologit program fits cumulative logit models and the oprobit program fits cumulative probit models. See Stata help for ologit. A program omodel is available from the Stata website for fitting these models and testing the assumption of the same effects for each cumulative probability (i.e., the proportional odds assumption for cumulative logit models). Other ways to fit cumulative link models are with the Stata oglm module. Continuation-ratio logit models can be fitted with the ocratio module. See Stata ocratio search. For information about using GEE in Stata, see Horton article and Stata GEE search. The GLLAMM module for Stata can fit a very wide variety of models, including logit and cumulative logit models with random effects. For further details, see and Stata gllamm search and Chapter 5 of "Multilevel and Longitudinal Modeling Using Stata" by S. Rabe-Hesketh and A. Skrondal (Stata Press, 2005).

5. Corrections: Here is a pdf file showing