Hailemichael Worku, MSc.

Telephone number: +31 (0)71 527 2727
E-Mail: h.m.worku@fsw.leidenuniv.nl
Faculty / Department: Faculteit der Sociale Wetenschappen, Instituut Psychologie, Methodologie & Statistiek
Office Address: Pieter de la Court gebouw
Wassenaarseweg 52
2333 AK Leiden

In social as well as medical studies the interest is often in simultaneous classification of participants/subjects on two or more variables. For example, in the Netherlands Study of Depression and Anxiety (NESDA) the scientific question of interest is how personality factors are related to the prevalence and association of depression and anxiety disorders. We approach this problem using the ideal point classification (IPC) model, a probabilistic multidimensional unfolding model. We are especially interested in the representation of marginal and conditional relationships between personality and depression/anxiety and also modeling jointly with the relationship between personality and the comorbidity. In the IPC model the Euclidean space of the dependent variables is three dimensional, in which the first dimension corresponds to the prevalence of depression, the second to the prevalence of anxiety, and the last for the comorbidity among depression and anxiety. Comparisons of IPC models with logistic regression models for clustered data will be provided.

Last Modified: 10-03-2015