Modelling categorical data: Loglinear models and logistic regression

I am a sociologist, with interests in stratification, social mobility, longitudinal data and dynamic social processes. I did a DPhil at Nuffield College, Oxford, formerly worked at the Institute for Social and Economic Research, Essex (home of the British Household Panel Study), and now lecture in sociology at the University of Limerick, Ireland.

Workshop contents and objectives

This course is a practical introduction to categorical data analysis, with the emphasis on loglinear models and logistic regression. Loglinear modelling and logistic regression are two closely related approaches to the analysis of categorical data, which is the most common form of data in the social sciences. "Practical" implies more emphasis on learning to run and interpret analyses, with somewhat less emphasis on maths and statistical theory. The objectives of the course are to enable students to apply these methods in real research contexts, and to become informed consumers of research using these methods.

Bibliography

Basic texts/overview

• Alan Agresti. An Introduction to Categorical Data Analysis. Wiley, New York, 1996.
• Allan McCutcheon and Colin Mills. Categorical data analysis: Log-linear and latent class models. In Elinor Scarbrough and Eric Tanenbaum, editors, Research Strategies in the Social Sciences: A Guide to New Approaches. Oxford University Press, Oxford, 1998.
• David W. Hosmer and Stanley Lemeshow, Applied Logistic Regression, 2nd Edition 2000

Remedial Reading

You should be familiar with this kind of material:

• Alan Agresti and Barbara Finlay, Statistical Methods for the Social Sciences, (chapter on multiple regression).