- Latent class methods
- Longitudinal data analysis
- Analysis of preference data
- Categorical and ordered responses
- Microgenetic data
- Spatio-temporal models
This research strand is concerned with changing attitudinal behaviour over time, and how best to model various types of data which might represent changing attitudes. For example, ordinal regression can be used for likert scale data, but how do we model banks of ordinal items at a single point in time, and also if the item responses are measured repeatedly?
One major methodological innovation in phase 2 of this strand has been the development of bivariate ordinal response models with correlated random effects. Such models are not routinely available in standard software and this has led to further development of the software package SABRE (http://www.sabre.lancs.ac.uk). These new models have been applied in two examples: one highligthed above and one in educational research.
.
|
|||||||||
We have analysed dropout and response from 17 waves of the British Household Panel Survey (BHPS) over the period 1991 to 2007. The dependent variables examined are attitudinal and take the form of ordinal responses to a series of questions concerning gender roles. The dataset comprises repeated measurements over time on the same set of individuals and is therefore a longitudinal panel subject to dropout. In order to take into account the repeated measures aspect of the data, we use random effects models to capture the associated residual heterogeneity.
