STAT6108 Analysis of Hierarchical (Multilevel & Longitudinal) Data

Module Overview

This module introduces students to the main statistical modelling approaches that can handle hierarchical data structures. The module has mainly an applied scope where basic theory is introduced to ensure understanding. Practical computer sessions using MLwiN and R are conducted where appropriate.

Aims and Objectives

Module Aims

The aims of the module are to introduce students to the main statistical models for hierarchical and longitudinal data, with emphasis on multilevel and marginal models.

Learning Outcomes

Having successfully completed this module you will be able to:

Contrast the main statistical models available for the analysis of hierarchical and longitudinal data.
Apply multilevel and marginal models for the analysis of hierarchical and longitudinal data.
Interpret the results from these statistical analyses in a non-technical language.
Understand the potential of more advanced elements, such as contextual variables and heterogeneous covariance structures for the analysis of hierarchical and longitudinal data.
Recognise the basic statistical theory underpinning multilevel and marginal models.

Syllabus

• Hierarchical data structures • Multilevel models: random intercepts and random slopes; contextual variables, cross-level interactions and heterogeneous variance structures • Model building: estimation; testing; diagnostic checking (specification issues and residual analysis); model selection • Longitudinal data structures • Multilevel and Marginal models for longitudinal data • Growth curve models • Models for hierarchical and longitudinal binary response data The module will integrate the theory with practical application using MLwiN and R.

Special Features

This module combines theory and practice, via lectures and computer sessions.

Learning and Teaching

Teaching and learning methods

This unit is taught via one double lecture per week and supported by practical computer sessions. Additional reading material and formative tests are provided where appropriate.

Type	Hours
Teaching	20
Independent Study	80
Total study time	100

Resources & Reading list

Goldstein, H. (2011). Multilevel Statistical Models.

Bryk, A.S. and Raudenbush, S.W. (1992, 2002). Hierarchical Linear Models: Applications and Data Analysis Methods.

Laboratory space and equipment required. Students require access to computer lab with R and MLwiN.

Singer, Judith D., Willett, John B. (2003). Applied Longitudinal Data Analysis: Modelling Change andEvent Occurrence.

Snijders, T.A.B. and Bosker, R.J. (1999, 2012). Multilevel Analysis.

Goldstein, H. (1995). Multilevel Statistical Models.

Diggle, P. J., Liang, K-Y. and Zeger, S. L. (1994, 2001). The Analysis of Longitudinal Data.

Assessment

Summative

Method	Percentage contribution
Coursework (4000 words)	100%

Referral

Method	Percentage contribution
Coursework (4000 words)	100%

Repeat Information

Repeat type: Internal & External

Linked modules

Prerequisites: STAT6083 OR RESM6004 OR STAT6117