Upcoming workshops

Multilevel Modelling courses 7th-11th January 2019, University of Bristol

The Centre for Multilevel Modelling will be running two workshops at the beginning of January. These can be attended as stand-alone courses, however a 25% discount will be given to participants attending both.

Introduction to Multilevel Modelling Using MLwiN, 7-9 January 2019, University of Bristol

This three-day course provides an introduction to multilevel modelling using the MLwiN software. We focus on multilevel modelling for continuous and binary responses (dependent or outcome variables) when the data are clustered or hierarchical. Such methods are appropriate when, for example, analysing the exam scores of students nested within schools, or the health outcomes of patients nested within hospitals. Longitudinal data are also clustered, with repeated measurements on individuals or multiple panel waves per survey respondent.

The course will consist of a 2:1 mix of lectures and hands-on practical sessions applying the taught methods to real datasets using MLwiN. Throughout, there is an emphasis on how to interpret the models and on what kinds of research question they can be used to explore.

Topics covered:

  1. Overview of multilevel modelling
  2. Introduction to MLwiN
  3. Variance-components models
  4. Random-intercept models with covariates
  5. Between- and within-effects of level-1 covariates
  6. Random-coefficient models
  7. Growth-curve models
  8. Three-level models
  9. Review of single-level logistic regression
  10. Two-level logistic regression

Pre-requisites:

We assume no prior knowledge of multilevel modelling or MLwiN. However, participants should be familiar with estimating and interpreting linear regression models, including the writing and interpretation of model equations, hypothesis testing and model selection, and the use and interpretation of dummy variables and interaction terms. Some participants may wish to refresh themselves of this material by reading module 3 of our LEMMA online course. https://www.bristol.ac.uk/cmm/learning/online-course/course-topics.html.

Instructors:

Professor Harvey GoldsteinDr George Leckie

Timing:

The course starts with registration at 10.45 on Day 1 and finishes at 15:00 on Day 3. Participants are expected to attend the full course.

Handling missing data for multilevel models, 10-11 January 2019, University of Bristol

This two-day course can be attended as a stand-alone course for those already familiar with multilevel modelling or as a continuation for the introductory course on multilevel modelling  that immediately precedes it.

The workshop will provide an introduction to this important topic together with a hands-on experience for participants in fitting data with complex patterns of missingness. It will utilise the recently released missing data features in the Stat-JR software distributed along with MLwiN by the Centre for Multilevel modelling, Bristol. This software is designed to handle very general data structures, including multilevel ones, and is applicable to both continuously distributed data and categorical data. It will cover traditional multiple imputation techniques using joint modelling and fully conditional modelling but with an emphasis on more recent and more flexible Bayesian models.

The workshop will consist of a mixture of explanatory talks and demonstrations together with ‘hands-on’ data analysis. If users wish to bring their own datasets (preferably not too large) there will be time to analyse these with help.

Links

General information on Stat-JR: https://www.bristol.ac.uk/cmm/software/statjr/

Stat-JR manuals: https://www.bristol.ac.uk/cmm/software/statjr/manuals/

Missing data in Stat-JR: https://www.bristol.ac.uk/cmm/research/missing-data/

If you have not used Stat-JR before then it is suggested that you work through the quick start guide prior to attending the workshop to ensure that the software is set up correctly. You may also wish to read through the document "Missing Data with Stat-JR".

 

Topics covered:

  1. Introduction to the problem. How do missing data arise? How not to handle missing data. Introduction to Imputation.
  2. Multiple Imputation for general models: multilevel and non-normal variables
  3. Fully conditional approaches to multiple imputation
  4. Joint modelling approaches to multiple imputation – continuous response
  5. Using Stat-JR
  6. Binary responses and the latent normal model
  7. Participant data analyses, including binary data and including participants’ own data if desired.
  8. A joint ‘one-pass’ Bayesian model for missing data.
  9. Participant data analysis using STATJR; single level
  10. One pass method with missing data at level 2
  11. Participant data analysis for multilevel data
  12. Further developments and general discusssion 

Pre-requisites:

This course assumes familiarity with multilevel modelling up to the level of module 5 of our LEMMA online course (https://www.bristol.ac.uk/cmm/learning/online-course/course-topics.html) or that you have attended the introduction to multilevel modelling course preceeding it. Participants are encouraged to gain a basic understanding of missing data techniques beforehand, e.g. by accessing the website: http://missingdata.org.uk/. If you are attending this course you will need to bring your own Windows laptop. If you already have Stat-JR then we advise that you install and test this beforehand, otherwise we will provide a cut-down version for use with missing data on USB stick for the duration of the workshop.

Instructors:

Professor Harvey Goldstein, Mr Christopher Charlton

Timing:

The course starts with registration at 10.15 on Day 1 and finishes at 16:00 on Day 2. Participants are expected to attend the full course. 

 

Course fees

The course fee includes printed materials, lunch, and morning/afternoon refreshments. The course fee does not include travel and accommodation costs. There will be an optional course meal for participants and workshop instructors for each course on the Tuesday and Thursday evening at an additional cost of £30 per meal.

Introduction to multilevel modelling:

  • For UK-registered MSc and PhD students - £180
  • For UK university academics, UK public sector staff, and staff at UK registered charity organisations - £360
  • For all other participants - £660

Handling missing data for multilevel modelling:

  • For UK-registered MSc and PhD students - £120
  • For UK university academics, UK public sector staff, and staff at UK registered charity organisations - £240
  • For all other participants - £440

Both workshops

  • For UK-registered MSc and PhD students - £225
  • For UK university academics, UK public sector staff, and staff at UK registered charity organisations - £450
  • For all other participants - £825

Cancellation/refunds:

A full refund will be given if cancellation occurs three weeks prior to the event. No refund is given after this date. By completing the application form, you are accepting these cancellation terms.

Applications:

Our workshops are now regularly over-subscribed so we have had to introduce an application and selection process. If you would like to attend either workshop, please complete and submit the online application form (see below). Please note the closing date for applications is Sunday 18th November.

Submission of the form and its acknowledgement does not guarantee a place on the workshops. We will email you by Wednesday 21st November to tell you whether or not your application has been successful. If you are offered a place on either workshop, it will not be confirmed until you have accepted and paid the relevant fee.

If you have any queries, please email info-cmm@bristol.ac.uk.

Go to booking form >>

Please click here to read the booking terms and conditions before completing the booking form.

Edit this page