Latent Variable Modeling of Categorical Data

 
Prof. Dr. Irini Moustaki
Athens University of Economics and Business, Greece
 
The talk is going to give an overview of the available methodology for handling categorical variables within latent variable analysis including recent developments.

Categorical data such as nominal and ordinal are often met in sample surveys. In the literature there have been different traditions for taking into account the qualitative nature of those variables. The different methods are mainly the structural equation modeling approach and the item response theory approach. The two approaches differ in terms of the treatment of categorical variables, model assumptions, estimation methods and model testing.

In the talk emphasis will be given to the methodological advantages of the different methods, goodness-of-fit issues as well as the latest software developments.

presentation Moustaki