Introducing multilevel models and applying them to the Health Survey for England using MLwiN
Wednesday 27 - Thursday 28 October 2010
Humanities Bridgeford Street, University of Manchester
This workshop is jointly organised by ESDS Government and e-Stat, a quantitative node for the ESRC NCeSS programme which brings together academics working in statistics and computer science.
Prerequisites: a working knowledge of multiple regression and the basics of sample design.
The focus of this two day workshop is on multilevel modelling where the multilevel structure is determined by the sample design,in particular on the way the methods can be applied to the Health Survey for England. This structure will then be used to examine the extent to which variables such as height, weight and BMI are clustered by household and area and whether any of this variability can be accounted by background variables such as social class and educational qualifications. The first day will provide a general introduction to multilevel modelling of sample survey data and will introduce the data to be used throughout the practical sessions of the workshop. The second day will be practical, applying MLwiN to a set of substantive questions using the Health Survey for England data.
Get the data:
Unrestricted access materials from this workshop, including the unrestricted access HSE teaching dataset, are available for reuse. Just enter your email address below and click on ‘get the data’.
Programme
| DAY 1 |
|
| |
|
| 9.30 |
Coffee and registration |
| |
|
| 10.00 |
Session 1: Vanessa Higgins, ESDS Government, University of Manchester |
| |
1. Accessing the HSE via ESDS |
| |
2. The HSE sample design and dataset |
| |
3. Descriptive analysis of the HSE |
| |
|
| 11.30 |
Refreshments |
| |
|
| 11.45 |
Session 2: Ian Plewis, CCSR/Social Statistics, University of Manchester |
| |
1. Performing standard single level analyses on the data in Stata, excluding and including options to allow for the survey design |
| |
2. Organising the HSE for multilevel analysis in MLwiN. |
| |
3. Fitting single level models in MLwiN |
| |
|
| 1.15 |
Lunch |
| |
|
| 2.00 |
Session 3: Mark Tranmer, CCSR/Social Statistics, University of Manchester |
| |
Introducing multilevel modelling |
| |
|
| 2.45 |
Refreshments |
| |
|
| 3.00 |
Session 4: Mark Tranmer, CCSR/Social Statistics, University of Manchester |
| |
Normal response multilevel analysis of the HSE using MLwiN part 1 |
| |
|
| 4.00 |
End |
| |
|
| DAY 2 |
|
| |
|
| 10.00 |
Session 1: Mark Tranmer, CCSR, University of Manchester and Bill Browne, School of Clinical Veterinary Sciences, University of Bristol |
| |
1. Normal response multilevel analysis of the HSE using MLwiN part 2 |
| |
2. Exercises based on normal response multilevel analysis of the HSE |
| |
|
| 11.30 |
Coffee |
| |
|
| 11.45 |
Session 2: Bill Browne, School of Clinical Veterinary Sciences, University of Bristol |
| |
Other response types |
| |
|
| 1.15 |
Lunch |
| |
|
| 2.00 |
Session 3: Bill Browne, School of Clinical Veterinary Sciences, University of Bristol |
| |
Exercises with binary and categorical responses part1 |
| |
|
| 3.15 |
Coffee |
| |
|
| 3.30 |
Session 4: Bill Browne, School of Clinical Veterinary Sciences, University of Bristol |
| |
Exercises with binary and categorical responses part 2 |
| |
|
| 4.30 |
End |
|
