Multilevel Modelling

Dr Tarani Chandola and Dr. Johan Koskinen

Compulsory for SRMS

Aim
The aim of this unit is to teach students the theory of multilevel models and present applications of multilevel models as well as software for fitting such models.

Objectives:

Students should be able to:

• Recognise when there is a need for more advanced modelling techniques
• Apply multilevel techniques to normal response data, discrete data and repeated measures data
• Acquire knowledge on how to use the MLwiN software for fitting multilevel models
• Understand why multilevel analysis may be more appropriate for certain data designs such as clustered designs
• Discuss the basic underlying theory of multilevel models
• Interpret in non-technical language the results from a multilevel analysis of a large dataset
• Use MLwiN software for multilevel analysis
• Students will develop skills for using multilevel models for their own research and for reading journal papers that very often employ multilevel analysis

Course Content

This unit will teach the theory and applications of multilevel models. Having introduced the basic statistical concepts and modelling tools in Semester 1, in this module, students will be introduced to more advanced modelling techniques. The unit will cover basic and more advanced multilevel models including random intercepts models, random slopes models, inference for multilevel models, the use of contextual variables in multilevel analysis, modelling complex variance structures, binary response multilevel models, modelling repeated measures and multivariate response linear models. All students will gain familiarity with and hands-on experience. Typically this will be managed by having both lectures and practical workshops in each session. Statistical software such as SPSS, MLwiN will be used. A range of prepared data sets will be used, including large-scale surveys and longitudinal studies. Students will achieve, as a minimum, a level of competence that enables them to use more advanced modelling techniques.

Teaching and Learning

The course will consist of lecture-based sessions and practical sessions (MLwiN workshops).

Assessment

The assessment for this module will be based on one piece of coursework.

Key Reading

Rasbash, J., Steele, F., Browne, W. and Goldstein, H. (2009) A user's guide to MLwiN. Centre for Multilevel Modelling, University of Bristol
www.cmm.bristol.ac.uk/MLwiN/download/MLwiN-userman-09.pdf

Additional Reading

Dobson, A. (2002). An introduction to generalized linear models. Chapman and Hall
Goldstein, H. (1995). Multilevel Statistical Models. London: Edward Arnold.
Snijders, T.A.B. and Bosker, R.J. (1999). Multilevel Analysis. London: Sage.