Home | Contacts | FAQ | Graduate School




 

Forgotten your password?

New Users: Register

This Login is for users of the Research Student Log only




 

Multilevel Modelling for Health Research

Course Description

This course aims to provide students with a solid grounding in the theory of multilevel models and enable students to perform and interpret multilevel model analyses using the package MLwiN.

Clustered or hierarchical data is common in health and social sciences research. For example, individuals may be nested within a geographical area or within schools, hospitals or workplaces. Hierarchical data also arises in longitudinal studies where repeated measures on the same individual are collected over time. This course will cover concepts of multilevel models and how to analyse continuous and binary data outcomes in a multilevel framework. Examples will be taken from education and health research.

This three day advanced level course is designed to give participants a good understanding of multilevel modelling. The course will be a mixture of theoretical sessions and practical sessions to illustrate the theoretical concepts. The practical sessions will use the MLwiN software package.

Topics to be covered will include: Introduction to multilevel models, random intercept and random coefficient (slope) models, examining residuals, hypothesis testing, multilevel models for binary data and an introduction to multilevel models for repeated measures/longitudinal data.

This course has been developed jointly with Fiona Steele, Harvey Goldstein and Jon Rasbash from the Centre for Multilevel Modelling, University of Bristol.
Participants with an interest in repeated measures/longitudinal data may note complimentary course on Longitudinal Data Analysis.


Prerequisites: Participants will be expected to have a good understanding and experience of applying and interpreting multiple linear regression models and logistic regression models, for example, have already attended Statistical Analysis Methods for Epidemiology and Social Sciences. Participants should have prior experience of using a statistical package to analyse data such as SPSS, STATA or SAS to gain the most out of the course.

Students will need a UCL computer ID to access course materials and take part in the computer practicals.

Researcher Development Framework Categories

A1) Knowledge base

Course Recommended for

This course is particularly relevant to the following groups:

  • Students in Social & Historical Sciences
  • Students in Medical Sciences

Course Organisers

  • Course Tutor - Ms Jenny Head - (Research Department of Epidemiology & Public Health)
  • Course Tutor - Dr Mai Stafford - (Research Department of Epidemiology & Public Health)
  • Administrator - Ms Kasia Bronk - (Graduate School)

 

2-4 April 2014 expand

Page last updated: 22nd July 2010