The Centre for Longitudinal Analysis in Ireland Workshop: Introduction to Multilevel Modelling using MLwiN and STATA

Venue: ESRI, Whitaker Square, Sir John Rogerson's Quay, Dublin 2

Supported by Department of Children and Youth Affairs

Workshop: Introduction to Multilevel Modelling using MLwiN and STATA

Presenters: William Browne and George Leckie

1st and 2nd April 2014

Course Summary

This introductory workshop is suitable for beginners in multilevel modelling, and covers models for continuous response variables. Topics include multilevel data structures, random intercept and slope models for two-level structures, contextual effects, and modelling variance. Further extensions such as models for three-level and non-hierarchical structures will be discussed.

Theory sessions are accompanied by practical sessions using MLwiN, a package for the analysis of multi-level data developed at the Centre for Multi-Level Modelling in Bristol. No prior knowledge of MLWin is assumed and instruction is provided on how to run MLwiN through STATA (using runmlwin). Participants get the chance to apply what they have learned in the lectures to real datasets. Throughout, there is an emphasis on how to interpret the models and on what kinds of research question they can be used to explore.

Prerequisites : The course assumes that participants are familiar with standard (single-level) multiple regression. No prior knowledge of MLwiN or STATA will be assumed.

Course Level: Beginner
Course Requirements: Participants should have some familiarity with survey data and a good understanding of linear regression.
Course Duration: 2 days
Course Fee: €250 (bursaries are available for PhD students)
Location: ESRI, Whitaker Square, Sir John Rogerson’s Quay, Dublin 2
Target Audience: The course is designed for users of survey data with a multi-level structure, i.e. data where lower level units are nested within higher level units (e.g. children in schools). The course will also be appropriate for those interested in the analysis of longitudinal data as it examines the application of multi-level models to data with repeated measures.
Course Tutors: Prof. Bill Browne is the Professor of Biostatistics in the Veterinary School and the Co-Director of the Centre for Multi-Level Modelling, both at the University of Bristol.

 

 

Dr George Leckie is a Senior Lecturer in Social Statistics at the Centre for Multi-Level Modelling, University of Bristol. His research focuses on the application of multilevel and other latent variable methods to the analysis of educational and social science data

PROGRAMME DAY 1

09.00 Registration
09.30 Introduction: multilevel data structures and overview of multilevel modelling
10.30 Computer practical 1 (Introduction to MLwiN for multiple regression)
11.15 Refreshment Break
11.30 Variance components models; random intercept models
12.30 Lunch
13.30 Computer practical 2 (Random intercept models)
14.30 Residual estimation and shrinkage; random slope models
15.30 Refreshment Break
15.45 Computer practical 3 (Residuals; random slope models)
17.00 RunMLwiN lecture

PROGRAMME DAY 2

09.00 Questions from Day 1
09.30 Contextual effects and cross-level interactions
10.30 Refreshment Break
10.45 Computer practical 4 (Contextual effects)
11.45 Variance functions at level 1 and 2
12.30 Lunch
13.30 Modelling repeated measures data
14.30 Computer practical 5 (repeated measures)
15.30 End of Workshop