For details of new features and bug fixes go to: https://cran.r-project.org/web/packages/R2MLwiN/NEWS
For information on installation/upgrading see: https://www.bris.ac.uk/cmm/software/r2mlwin/
CMM News 2011-2022
18-Nov-22: MLwiN 3.06 released
The Centre for Multilevel Modelling would like to announce our latest MLwiN release - 3.06. For details of bug fixes go to bug fixes.
17-Oct-22: Introduction to Multilevel Modelling Using MLwiN, R or Stata, 17-19 January 2023
This workshop provides an introduction to multilevel modelling using the MLwiN software. Closing date for applications is 4th December 2022.
09-Mar-22: Introduction to Multilevel Modelling Using MLwiN, R or Stata, 5-7 July 2022
This workshop provides an introduction to multilevel modelling using the MLwiN software. Closing date for applications is 1st May 2022.
24-Mar-22: International Conference on Multilevel Analysis and Workshops: 11-14 April, Utrecht, the Netherlands
The next International Conference on Multilevel Analysis is to be held on April 12th and 13th in Utrecht, the Netherlands. Invited speakers are George Leckie of the Centre for Multilevel Modelling and Terrence D. Jorgensen of the University of Amsterdam.
A conference dinner is organized for April 12th.
A pre-conference workshop on Automated Systematic Reviews is taught by Rens van de Schoot on April 11th. A post-conference workshop on Multilevel SEM using the lavaan package is taught by Terrence Jorgensen on April 14th. Please note that both workshops will be taught online.
You can submit your abstract for an oral or poster presentation until March 20th, and you will be notified within a few days if your abstract is accepted.
For further information about the conference, abstract submission and registration: https://multilevel.fss.uu.nl/
22-Sep-21: Introduction to Multilevel Modelling Using MLwiN, R or Stata, 18-20 January 2022
This workshop provides an introduction to multilevel modelling using the MLwiN software. Closing date for applications is 14th November 2021.
10-Jun-21: New teaching resource: Cross Classified Models
New videos have been developed through a collaboration of the National Centre for Research Methods (NCRM) and the Centre for Multi-Level Modelling at the University of Bristol.
In these Professors William Browne and George Leckie cover a set of statistical models called cross-classified models that extend multilevel models that are used widely in the social sciences to account for dependency structures in data.
The resource comprises of a series of three lectures:
- Introduction
- Model fitting
- Extensions and Further Applications
and a practical walkthrough:
- Practical
The videos can be accessed on the NCRM online resources pages
If you have any questions about these resources please contact Professor William Browne at the Centre for Multi-Level Modelling, University of Bristol.
12-Mar-21: Introduction to Multilevel Modelling Using MLwiN, R or Stata, 13-15 July 2021
This workshop provides an introduction to multilevel modelling using the MLwiN software. Closing date for applications is Sunday 9th May 2021.
12-Nov-20: Introduction to Multilevel Modelling Using MLwiN, 26-28 January 2021
This workshop provides an introduction to multilevel modelling using the MLwiN software. Closing date for applications is Sunday 29th November 2020.
27-Jun-20: R2MLwiN 0.8-7 released
The new version of R2MLwiN includes a variety of bug fixes.
For details of new features and bug fixes go to: https://cran.r-project.org/web/packages/R2MLwiN/NEWS
For more details of known and fixed bugs visit: https://www.bristol.ac.uk/cmm/software/r2mlwin/r2mlwin-knownbugs.html
For information on installation/upgrading see: https://www.bristol.ac.uk/cmm/software/r2mlwin/
26-Mar-20: MLwiN 3.05 released
The Centre for Multilevel Modelling would like to announce our latest MLwiN release - 3.05. For details of bug fixes go to bug fixes.
This new release also includes datasets to accompany the recently published "Multilevel modelling for public health and health services research: Health in context" open access book, providing a practical introduction to multilevel modelling or multilevel analysis (MLA) for researchers working in public health and health services research. The authors, Alastair Leyland and Peter Groenewegen, begin with a compelling argument for the importance of researchers in these fields having an understanding of MLA to be able to judge not only the growing body of research that uses it, but also to recognise the limitations of research that did not use it. The volume also guides the analysis of real-life data sets by introducing and discussing the use of MLwiN with three datasets provided for this purpose. These datasets - lmdp.ws, fysio.ws and cvd_data.ws - are also included in the teaching version of MLwiN. Details of these datasets along with step-by-step descriptions of their analysis are included in the book, available for free download from https://www.springer.com/gb/book/9783030347994.
24-Sep-19: Stat-JR 1.0.7 released
We are pleased to announce a new version 1.0.7 of Stat-JR. This version has all the existing functionality of Stat-JR including its own MCMC estimation engine, interoperability with many software packages, three different interfaces (web, eBook and workflow), its statistical analysis assistant features and tools for to creating statistical teaching materials for SPSS. In addition this new release also showcases work in a recent NCRM funded grant to implement small area estimation techniques.
This release contains many bug fixes.
All manuals have been updated for the new version and are available here.
To download the software version please complete the form at https://www.cmm.bris.ac.uk/clients/softwaredownload/ and select StatJR from the drop-down menu.
Note that Stat-JR is now a 64-bit application, so if you are updating from a version prior to 1.0.5 you will need to download a new C++ compiler.
20-Sep-19: New teaching resource: Quantitative methods e-books
New e-books have been developed through a collaboration of the UK Data Service, National Centre for Research Methods (NCRM) and the Centre for Multi-Level Modelling at the University of Bristol and were created using the StatJR software based on original outputs from a project funded by the British Academy.
These quantitative methods e-books and accompanying quizzes are for direct use in teaching students or for self-study at an undergraduate level. They aim to build skills in quantitative methods and statistical software and use the Living Costs and Food Survey, available from the UK Data Service.
Using SPSS, lecturers and students can use both the practical and quiz elements of each e-book topic:
- Examining Continuous Variables in SPSS
- Correlations in SPSS
- Regression in SPSS
- Multiple Regression in SPSS
Dr Vanessa Higgins, University of Manchester, said "Working together has enabled us to create this new resource to support teaching and will encourage students to learn valuable quantitative skills for the future..."
The e-books can be accessed on the UK Data Service teaching pages
If you have any questions about these resources please contact Professor William Browne at the Centre for Multi-Level Modelling, University of Bristol.
20-Sep-20: Introduction to Multilevel Modelling Using MLwiN, 7-9 January 2020
This workshop provides an introduction to multilevel modelling using the MLwiN software. Closing date for applications is Sunday 10th November 2019.
24-Jun-19: Update to runmlwin released on SSC
An update of runmlwin has been released. runmlwin is a Stata command which allows Stata users to run the powerful MLwiN multilevel modelling software from within Stata (see /cmm/software/runmlwin/ for further details). Released on SSC, the update includes a number of bug fixes (for details of these see https://bitbucket.org/cmm_stata/runmlwin/commits/.
As usual you can update to the most recent version with the following command, in Stata:
. adoupdate runmlwin
If you don't already have it installed you can do so as follows:
. ssc install runmlwin
03-Jun-19: MLwiN 3.04 released
MLwiN 3.04 released
28-May-19: R2MLwiN 0.8-6 released
R2MLwiN 0.8-6 released on CRAN
12-Apr-19: CMM Multilevel Modelling Workshops, 9-12 July 2019
In July the Centre for multilevel modelling will be running two workshops:
Introduction to Multilevel Modelling Using MLwiN (9-11 July 2019)
This provides an introduction to multilevel modelling using the MLwiN software.
Small Area Estimation using R and Stat-JR (12 July 2019)
This provides an introduction to Small Area Estimation and in particular how multilevel models are used as part of this topic.
Closing date for applications is Sunday 26 May 2019.
13-Mar-19: New Stat-JR videos released by NCRM.
The Centre for Multilevel modelling has had lots of funding from the ESRC in the UK through their National Centre for Research Methods (NCRM) programme and this programme also has an online training resource containing videos about research methods training. Professor Bill Browne has created 3 videos related to the work of the centre on the StatJR software. These are firstly an introduction to StatJR, second a talk on eBooks and Statistical Analysis Assistants in StatJR and finally a talk on the more recent StatJR functionality to create bespoke training materials in SPSS. These three videos can be reached at the weblink https://www.ncrm.ac.uk/resources/online/introduction_to_Stat-JR_software/.
11-Feb-19: MLwiN 3.03 released
The Centre for Multilevel Modelling would like to announce our latest MLwiN release - 3.03. For details of bug fixes go to bug fixes.
31-Jan-19: New research shows league tables ‘punish and reward wrong schools’
Secondary school league tables, published on 24th January, ‘punish and reward the wrong schools’ by failing to take pupil background into account, according to new research from the School of Education, University of Bristol.
10-Dec-18: Stat-JR 1.0.6 released and is free for Christmas
We are pleased to announce a new version 1.0.6 of Stat-JR. This version has all the existing functionality of Stat-JR including its own MCMC estimation engine, interoperability with many software packages, three different interfaces (web, eBook and workflow) and its statistical analysis assistant features. In addition this new release also showcases work in a recent British Academy funded grant looking at using Stat-JR to create statistical teaching materials for SPSS.
This release contains many bug fixes.
All manuals have been updated for the new version and are available here.
Our Stat-JR software has been freely available to both the UK academic and MLwiN user communities from its first version. As an early Christmas present we have decided that for the next month or so we will give away Stat-JR version 1.0.6 free to all. This means that if you download Stat-JR in the next month you will have a free version of the software. This version will have all the same functionality as earlier versions of the software and use will not be time-limited although you will not currently be eligible for support or free upgrades. Any Stat-JR interoperability functions that involve other software e.g. SPSS or MLwiN will require you to own such software and have it on your machine. We very much hope you enjoy Stat-JR and have a great festive season.
To download the software version please complete the form at https://www.cmm.bris.ac.uk/clients/softwaredownload/ and select StatJR from the drop-down menu.
Note that Stat-JR is now a 64-bit application, so if you are updating from a version prior to 1.0.5 you will need to download a new C++ compiler.
21-Sep-18: CMM Multilevel Modelling Workshops, 7-11 January 2019
In January the Centre for multilevel modelling will be running two workshops:
Introduction to Multilevel Modelling Using MLwiN (7-9 January 2019)
This provides an introduction to multilevel modelling using the MLwiN software.
Handling missing data for multilevel models (10-11 January 2019)
This provides an introduction to fitting complex patterns of missingness using the Stat-JR software.
Closing date for applications is Sunday 18 November 2018.
24-Apr-18: Introduction to Multilevel Modelling Using MLwiN Workshop, 17-19 July 2018
This workshop provides an introduction to multilevel modelling using the MLwiN software. Closing date for applications is Sunday 3 June 2018.
01-Mar-18: MLwiN 3.02 released
29-Jan-18: Missing data workshop now open to all UK researchers
26-Jan-18: Jon Rasbash prize open to submissions
05-Dec-17: The efficient handling of missing data in statistical models, 20 March 2018
21-Nov-17: Stat-JR 1.0.5 released
Stat-JR 1.0.5 released
10-Oct-17: Introduction to Multilevel Modelling Using MLwiN Workshop, 9-11 January 2018
This workshop provides an introduction to multilevel modelling using the MLwiN software. Closing date for applications is Sunday 19 November 2017.
01-Sep-17: R2MLwiN 0.8-5 released
R2MLwiN 0.8-5 released on CRAN
25-Jun-17: R2MLwiN 0.8-4 released
The new version of R2MLwiN includes updates to facilitate compatibility with recent releases of MLwiN.
31-May-17: MLwiN 3.01 released
The Centre for Multilevel Modelling would like to announce our latest MLwiN release - 3.01. For details of bug fixes go to bug fixes.
02-May-17: New version of Stat-JR missing data template released.
New version of 'one pass' missing data Stat-JR template (2LevelMissingOnePass) released
30-Mar-17: Introduction to Multilevel Modelling Using MLwiN Workshop, 11-13 July 2017
This workshop provides an introduction to multilevel modelling using the MLwiN software. Closing date for applications is Sunday 28 May 2017.
03-Mar-17: MLwiN 3.00 released
MLwiN 3.00 released
24-Nov-16: New Stata and MLwiN practicals for LEMMA online course
New Stata and MLwiN practicals for LEMMA online course
We are pleased to release a Stata practical for our LEMMA online course Module 9: Single-level and multilevel models for ordinal responses and an MLwiN practical for Module 15: Multilevel Modelling of Repeated Measures Data. See https://www.bristol.ac.uk/cmm/learning/online-course/ for further details.
24-Nov-16: Missing data: new Stat-JR functionality to support analyses of incomplete datasets
New Stat-JR functionality to support analyses of incomplete datasets.
There are now three principal Stat-JR templates available to support handling missing data in multilevel generalised linear models; these use two different approaches.
The first two templates use ‘multiple imputation’ which is a widely used procedure that will handle a large number of models: a 2-level (2LevelImpute; available since 2014) and a new N-level (NLevelImpute) version are now available.
The second (one pass) approach is a more recent generalisation with a more robust theoretical justification (see Goldstein et al, 2014 for further details); this has been implemented in the new 2LevelMissingOnePass template.
To use these templates, you will first need to install Stat-JR (https://www.bristol.ac.uk/cmm/
Goldstein, H., Carpenter, J. R. and Browne, W. J. (2014), Fitting multilevel multivariate models with missing data in responses and covariates that may include interactions and non-linear terms. Journal of the Royal Statistical Society: Series A (Statistics in Society). 177(2), 553-564 doi: 10.1111/rssa.12022
19-Sep-16: Introduction to Multilevel Modelling Using MLwiN Workshop
Introduction to Multilevel Modelling Using MLwiN 4-6 January 2017
This workshop provides an introduction to multilevel modelling using the MLwiN software. Closing date for applications is Sunday 13 November 2016.
08-Sep-16: R2MLwiN article published and new version released to CRAN
R2MLwiN article published (08-September-16) and R2MLwiN 0.8-3 released (07 September-16)
The Centre for Multilevel Modelling would like to announce that a journal article for R2MLwiN is now available in the Journal of Statistical Software:
To support this a new version of R2MLwiN (0.8-3) has been released.
For details of new features and bug fixes go to: https://cran.r-project.org/web/packages/R2MLwiN/NEWS
For information on installation/upgrading see: https://www.bris.ac.uk/cmm/software/r2mlwin/
15-Jul-16: Kelvyn Jones elected to Fellow of the British Academy
Kelvyn Jones elected to Fellow of the British Academy
27-Jun-16: Stat-JR 1.0.4 released
Stat-JR 1.0.4 released
14-Jun-16: Update to runmlwin released on SSC
An update of runmlwin has been released. runmlwin is a Stata command which allows Stata users to run the powerful MLwiN multilevel modelling software from within Stata (see https://www.bris.ac.uk/cmm/software/runmlwin/ for further details). Released on SSC, the update includes a number of bug fixes, as well as improved compatibility with MLwiN version 2.36 and later.
As usual you can update to the most recent version with the following command, in Stata:
. adoupdate runmlwin
If you don't already have it installed you can do so as follows:
. ssc install runmlwin
18-May-16: Stat-JR Workflow & eBook Workshop on 4th July 2016, Bristol
A one-day workshop on 4th July 2016, in Bristol, will introduce the Stat-JR software package, and in particular its new workflow system developed as part of our current ESRC-funded project “The use of interactive electronic-books in the teaching and application of modern quantitative methods in the social sciences”. Please note that this is a one-day workshop taking place only on Monday 4th July 2016 (not additionally on 8th July as previously advertised).
12-Apr-16: R2MLwiN 0.8-2 released
R2MLwiN 0.8-2 released
04-Apr-16: MLwiN 2.36 released
MLwiN 2.36 released
18-Mar-16: Stat-JR Workflow & eBook Workshops on 4th & 8th July 2016, Bristol
Two one-day workshops on 4th July and 8th July 2016, in Bristol, will introduce the Stat-JR software package, and in particular its new workflow system developed as part of our current ESRC-funded project “The use of interactive electronic-books in the teaching and application of modern quantitative methods in the social sciences”.
04-Mar-16: A Short Course on Multilevel Modelling in Veterinary and Agricultural Science
Professor Ian Dohoo (Atlantic Veterinary College, Canada) and colleagues will give a short course on Multilevel Modelling in Veterinary and Agricultural Science using MLwiN will run at the The University of Melbourne, Tuesday 29 March to Friday 1 April 2016.
A Short Course on Multilevel Modelling in Veterinary and Agricultural Science
Tuesday 29 March to Friday 1 April 2016 9.00 am to 5.00 pm
Instructors:
Professor Ian Dohoo, Atlantic Veterinary College, Canada
Mark Stevenson, Faculty of Veterinary and Agricultural Sciences, The University of Melbourne
Simon Firestone, Faculty of Veterinary and Agricultural Sciences, The University of Melbourne
Andrew Robinson, Centre of Excellence for Biosecurity Risk Analysis, The University of Melbourne
Where: Kimpton Room, Ground Floor Building 142 (on Royal Parade), Faculty of Veterinary and Agricultural Sciences, The University of Melbourne, Parkville, Victoria, Australia
When: Tuesday 29 March to Friday 1 April 2016
Cost: AU $800 students, AU $1200 others. Includes lunch and morning and afternoon tea.
Purpose: The purpose of this 4-day, application oriented course is to provide the tools for veterinary and agricultural scientists analyse clustered measures of continuous and discrete data. The workshop will also cover variance component estimation.
Primary teaching objectives: The primary objectives are to help veterinary and agricultural scientists identify clustering in their data, and to analyse it appropriately using software packages such as Stata, R and MlwiN.
Teaching methods: The course will consist of lectures and practical exercises using data from the instructors’ own veterinary research. The emphasis is on practical aspects, not the theory, of the methodology.
Required or suggested knowledge base for participants: This course is suitable for veterinary and agricultural scientists who need to these techniques in their work, but may not be primarily interested in theoretical aspects. The course requires no background in matrix algebra, and the statistical prerequisite is basic understanding of regression (including linear and logistic regression).
Please bring a laptop computer. Software, data sets and notes for the course will be provided on USB.
For further details (and on-line registration) see: https://fvas.unimelb.edu.au/
16-Nov-15: Introduction to Multilevel Modelling Using MLwiN Workshop
Introduction to Multilevel Modelling Using MLwiN 6-8 January 2016
05-Nov-15: Stat-JR 1.0.3 released
Stat-JR 1.0.3 released
21-Oct-15: Jon Rasbash prize open to submissions
The Jon Rasbash prize for Quantitative Social Science is awarded biennially for early career achievement in the field of quantitative social science.
04-Sep-15: MLwiN 2.35 released
MLwiN 2.35 released
13-Jul-15: MLwiN 2.34 released
MLwiN 2.34 released
09-Jul-15: Bug in MLwiN 2.33 for MCMC estimation and missing data
Bug in MLwiN 2.33 for MCMC estimation and missing data
11-May-15: MLwiN 2.33 released
MLwiN 2.33 released
18-Mar-15: MLwiN scripting versions for Mac OS X and Linux released
The Centre for Multilevel Modelling would like to announce the availabilty of scripting versions of the MLwiN engine built for alternative operating systems (including Mac OS X and Linux variants), facilitating the easier use of R2MLwiN and runmlwin on these platforms. If you already own a license to MLwiN these new versions can be found on our software download page.
23-Jan-15: MLwiN 2.32 released
The Centre for Multilevel Modelling would like to announce our latest MLwiN release - 2.32. For details of bug fixes go to bug fixes.
29-Sep-14: Stat-JR 1.0.2 released
Stat-JR 1.0.2 released
29-Sep-14: MLwiN 2.31 released
The Centre for Multilevel Modelling would like to announce our latest MLwiN release - 2.31.
01-Sep-14: New LEMMA module: Multilevel Modelling of Repeated Measures Data
New LEMMA module: Multilevel Modelling of Repeated Measures Data
25-Jun-14: New SPSS practicals for LEMMA online course
We are pleased to release SPSS practicals for our LEMMA online course Module 3: Multiple Regression and Module 5: Introduction to Multilevel Modelling.
Sample pdf versions are available here.
02-Jun-15: New job vacancy: Reader /Senior Lecturer / Lecturer in Quantitative Methods / Social Statistics
The Faculty of Social Sciences and Law at the University of Bristol is committed to a substantial expansion of its capacity in quantitative social sciences and is launching an international search for an academic member of staff.
For further details, please see here.
12-May-14: Stat-JR Multiple imputation template & eBook released
There is now a template, and supporting eBook, available for download which conduct multiple imputation for 2-level models in Stat-JR. This incorporates the existing REALCOM procedures, but provides a very much faster implementation.
25-Mar-14: Stat-JR 1.0.1 released
Stat-JR 1.0.1 has now been released. More details about Stat-JR, including how to order it, can be found on the Stat-JR homepage.
Stat-JR 1.0.1 bug fixes.
26-Feb-14: MLwiN 2.30 released
The Centre for Multilevel Modelling would like to announce our latest MLwiN release - 2.30. For details of bug fixes go to bug fixes.
16-Dec-13: MLwiN 2.29 released
The Centre for Multilevel Modelling would like to announce our latest MLwiN release - 2.29. For details of bug fixes go to bug fixes
21-Oct-13: New module for LEMMA online course
Module 14: Missing Data has been added to the LEMMA online course
16-Jul-13: MLwiN 2.28 released
The Centre for Multilevel Modelling would like to announce our latest MLwiN release - 2.28.
08-Jul-13: New job vacancy on LEMMA 3: 2-year research post, causal analysis of longitudinal data
New job vacancy on LEMMA 3: 2-year research post, causal analysis of longitudinal data
08-May-13: New training materials on the LEMMA online course
New training materials have recently been added to the LEMMA online course, covering the following topics:
- Module 10: Single-level and Multilevel Models for Nominal Responses
- Module 11: Three-Level Multilevel Models
- Module 12: Cross-Classified Multilevel Models
- Module 13: Multiple Membership
Go to the LEMMA course >>
12-Mar-13: MLwiN 2.27 released
The Centre for Multilevel Modelling would like to announce our latest MLwiN release - 2.27. For details of bug fixes go to bug fixes
08-Feb-13: Gallery of Multilevel Papers and Multilevel Showcase
The Centre for Multilevel Modelling has recently developed two online resources which we hope will be valuable to multilevel modelling users, teachers and learners. "Gallery of Multilevel papers" is a database of mainly published articles which use multilevel models, searchable by model type and substantive area. The "Multilevel Showcase" is a catalogue of people and organisations whose work has benefited from using our software, attending our workshops or from use of our online training materials.
For further details see https://www.bristol.ac.uk/cmm/research/gallery/
16-Jan-13: Stat-JR: Second beta release
We're pleased to announce the second beta release of our new Stat-JR software system.
This version (0.2) includes an exciting range of new features.
You can request it via Stat-JR's download page.
Stat-JR has been collaboratively developed by the Centre for Multilevel Modelling's MLwiN development team together with researchers from the University of Southampton, as is part of three ESRC funded programme nodes (LEMMA II, e-STAT and LEMMA III).
At present, while the software is a beta release, we are only distributing it in a 30-day limited licence form (with the option to renew every 30 days) so that bug fixes will reach our users quickly. It is then our intention to fully release the software at a later date, when, as with our MLwiN software, Stat-JR will be free to UK academics, with potentially (although not definitely) a small one-off fee for non-academics and non-UK users.
Note that currently, like MLwiN, Stat-JR is designed to run on Microsoft Windows machines.
05-Nov-12: Tony Fielding - Obituary
We regret to report the sad news that Antony (Tony) Fielding, Visiting Professor at the Centre for Multilevel Modelling at the University of Bristol, died on 29 October, aged 68. His obituary from CMM is here.
24-Oct-12: Jon Rasbash Prize winner announced
Paper on the impact of neighbourhood context on individuals' fear of crime awarded the Jon Rasbash Prize
The winner of the Jon Rasbash Prize for Quantitative Social Science in 2012 is Dr Ian Brunton-Smith from the University of Surrey.
The prize was awarded for his paper 'Do Neighborhoods Generate Fear of Crime? An Empirical Test Using the British Crime Survey', published in Criminology (co-authored with Patrick Sturgis). The paper explores the impact of neighbourhood context on individuals' fear of crime, using multilevel modelling applied to data from three years of the British Crime Survey with contextual data from the 2001 census.
16-Oct-12: MLwiN 2.26 released
The Centre for Multilevel Modelling would like to announce our latest MLwiN release - 2.26. For details of bug fixes go to bug fixes
02-Oct-12: R2MLwiN - a new R package to access MLwiN functionality
We are pleased to announce a new R package, R2MLwiN (Zhang et al. 2012) that allows R users access to the functionality within MLwiN directly from within the R package. This package has been developed as part of the e-STAT ESRC digital social research programme grant along with the Stat-JR package.
See the R2MLwiN page for more details, including examples taken from the book MCMC Estimation in MLwiN.
Feedback gratefully received by Professor Bill Brown or Zhengzheng Zhang (z.zhang@bristol.ac.uk).
09-Jul-12: Dr. George Leckie awarded ESRC Future Research Leaders Grant
Dr George Leckie has been awarded an ESRC Future Research Leaders Grant. The three-year project is titled "Multilevel Modelling of the Government's New School Performance Measures, 'Floor Standards' Target and 'Narrowing the Gap' Priority".
Each year, the Government publishes school performance tables that report the achievement and progress of pupils in English secondary schools. The Government's aim is to hold schools accountable and to enable parents to make meaningful choices about where to send their children to school. The latest 2011 school performance tables introduce new school performance measures, 'floor standards' which all schools should pass, and a 'narrowing the gap' priority that all schools narrow the performance gaps between their most disadvantaged pupils and their peers. This project will critically review these high-stakes changes and will develop and apply innovative multilevel modelling statistical techniques to explore potential improvements to the way the Government measures, reports and communicates school performances.
George will visit and collaborate with Professor David Spiegelhalter and the Understanding Uncertainty team at the Statistical Laboratory, University of Cambridge and will undertake a Knowledge Exchange Placement at the Data Analysis Project at the Fischer Family Trust. Professor Fiona Steele will provide mentoring support and additional guidance will be provided by Professors Harvey Goldstein and Bill Browne.
22-May-12: XI Conference of International Society for Quality-of-Life Studies
The XI Conference of International Society for Quality-of-Life Studies (ISQOLS), 1-4 November in Venice, will include a session on "Micro-Macro Analysis-of-Quality of Life Data" (track number 45).
Further details can be found at https://www.aiquav.it/isqols2012/Index.htm
17-May-12: Stat-JR beta release
The MLwiN development team in collaboration with programmers from the University of Southampton are pleased to announce a beta release of our new, exciting Stat-JR software system. This has been developed as part of three ESRC funded programme nodes (LEMMA II, e-STAT and LEMMA III)
At present, while the software is a beta release, we are only distributing it in a 30-day limited licence form (with the option to renew every 30 days) so that bug fixes will reach our users quickly. It is our intention at a later date to then fully release the software when, as with our MLwiN software, Stat-JR will be free to UK academics, with potentially (though not definitely) a small one-off fee for non-academics and non-UK users.
Note that currently, like MLwiN, Stat-JR is designed to run on Microsoft Windows machines.
For more details on the software, documentation and how to apply for a 30-day download of the software, please visit https://www.bristol.ac.uk/cmm/software/statjr/index.html
17-May-12: Gallery of multilevel papers and multilevel showcase
We are pleased to announce our new gallery of multilevel papers and multilevel showcase.
The gallery is a way for individual researchers to share information about their work with others and to find details of papers written by others that satisfy their modelling interests.
For further details see https://www.bristol.ac.uk/cmm/research/gallery/
27-Feb-12: MLwiN 2.25 released
The Centre for Multilevel Modelling would like to announce our latest MLwiN release - 2.25. For details of bug fixes go to bug fixes
22-Dec-11: 4 ESRC AQM Studentships at CMM
We are pleased to invite applications for 4 ESRC '+3' studentships on the South West Doctoral Training Centre's Advanced Quantitative Methods in Social Science and Health (AQM) pathway. The AQM pathway is based at the University of Bristol and these studentships will run from October 2012.
The deadline for applications is 17 February 2012.
For more details about AQM studentships and the applications procedure, visit our website at https://www.bris.ac.uk/cmm/aqm
08-Dec-11: Article about CMM, Autumn 2011
Behind every quantitative research project, there's a statistician waiting to make sense of the data and they could well be using software developed at Bristol's Centre for Multilevel Modelling to do it.
02-Nov-11: The Jon Rasbash Prize for Quantitative Social Science
A new award - The Jon Rasbash prize for Quantitative Social Science has been established to commemorate the contributions to quantitative social science of Jon Rasbash, who was Professor of Computational Statistics and Director of the Centre for Multilevel Modelling at the University of Bristol. Jon was principally known for his development of multilevel methodology and its software implementation and for his research on studying social relationships within families.
The £500 prize recognizes early career achievement in the development and/ or application of advanced quantitative methods in any social science discipline. Applicants should be UK residents and in the first 10 years of their research career (including periods of postgraduate study).
The prize will be awarded on the basis of a research paper published in a peer-reviewed journal in 2010 or 2011. The applicant should be the sole or lead author of the paper. Submissions by 1 December 2011. Further details >>
24-Oct-11: Module 9: Single and multilevel models for ordinal responses
In Module 6 we saw how multiple regression models for continuous responses can be generalised to handle binary responses, and in Module 7 these models were further extended for the analysis of binary data with a two-level hierarchical structure. This module considers standard (single-level) and multilevel models for ordinal categorical response variables, where the numeric codes assigned to categories imply some ordering. We begin with a description of two approaches for the analysis of single-level ordinal data:
- the cumulative logit model which is appropriate for variables such as Likert scale items, where respondents are asked to indicate their strength of agreement with a statement from 'strongly agree' to 'strongly disagree', and educational tests where marks are available as grades rather than percentage scores; and
- the continuation ratio model for ordinal responses that can be viewed as the result of a series of sequential decisions or actions (e.g. highest level of educational qualifications).
We then show how the cumulative logit model can be extended for the analysis of data with a two-level hierarchical structure. Further details >>
27-Sep-11: MLwiN 2.24 released
The Centre for Multilevel Modelling would like to announce our latest MLwiN release - 2.24. For details of bug fixes go to bug fixes
17-Aug-11: Research Assistant/Associate in Social Statistics
The Centre for Multilevel Modelling at the University of Bristol is seeking to appoint an applied statistician to work on a new ESRC-funded project, Longitudinal Effects, Multilevel Modelling and Applications (LEMMA 3).
27-Jul-11: New multilevel modelling course practicals for the R software now online
The Centre for Multilevel Modelling is very pleased to announce the addition of R practicals to our free on-line multilevel modelling course
12-Jul-11: REALCOM-Impute software updated
REALCOM-Impute software has been updated and a bug fixed
10-Jun-11: The Jon Rasbash Prize for Quantitative Social Science
A new award - The Jon Rasbash prize for Quantitative Social Science has been established to commemorate the contributions to quantitative social science of Jon Rasbash, who was Professor of Computational Statistics and Director of the Centre for Multilevel Modelling at the University of Bristol. Jon was principally known for his development of multilevel methodology and its software implementation and for his research on studying social relationships within families.
The £500 prize recognizes early career achievement in the development and/ or application of advanced quantitative methods in any social science discipline. Applicants should be UK residents and in the first 10 years of their research career (including periods of postgraduate study).
The prize will be awarded on the basis of a research paper published in a peer-reviewed journal in 2010 or 2011. The applicant should be the sole or lead author of the paper. Submissions by 1 December 2011. Further details >>
10-May-11: What is multilevel modelling? - short film
CMM's Fiona Steele is featured in a short film introducing multilevel modelling. View the clip on our videos, audio and voice-overs section.
With thanks to Man Luo and the rest of the production team (see the end of the film for the contributors, plus Rebecca Pillinger who also features briefly)
05-May-11: New Realcom user forum
New Realcom user forum - a Realcom section within the MLwiN user forum:
You can view this forum as a guest and you can can log in to post or answer questions about Realcom or Realcom-Impute:
Go to further details about Realcom.
21-Apr-11: MLwiN 2.23 now released
The Centre for Multilevel Modelling would like to announce our latest MLwiN release - 2.23. For details of bug fixes go to bug fixes
29-Mar-11: Research Workshop in Multilevel Modelling using MLwiN (5 - 7 September 2011)
Tutors: William Browne, Kelvyn Jones, George Leckie, Mark Lyons-Amos, Rebecca Pillinger and Fiona Steele
This workshop is designed for researchers who have some experience of multilevel modelling, but now wish to apply it to their own data. The workshop will be limited to 12 participants, with most of the time given to analysis of participants’ own data using the MLwiN software. Participants should be prepared to present their research to the rest of the group, including results from analyses conducted during the workshop. There may also be more formal lectures (e.g. on significance testing, or reporting results of a multilevel analysis), but their specific content will depend on participants’ needs.
Workshop objectives
This is an intensive workshop which aims to take researchers to a point where they have the skills and confidence to specify multilevel models appropriate to their research questions, to estimate models in MLwiN, and to interpret the results. By the end of the workshop, participants should be capable of producing a publication-quality research paper using multilevel modelling.
Applications should not exceed three pages (including the one-page CV) and must be submitted to info-cmm@bristol.ac.uk by Tuesday 31 May 2011. Applicants will be notified of the outcome mid-June 2011.
For further details about this workshop and the application procedure go to: www.bristol.ac.uk/cmm/software/support/workshops/#research
22-Mar-11: Realcom and Realcom-Impute have been updated
Realcom: We no longer support the original mixed-responses module of Realcom as the functionality is included in Realcom-Impute. Because of this the original Realcom installer has been split into realcom-factor and realcom-measerr. More>>
Realcom-Impute: A bug that can affect the covariance estimates for discrete responses has been fixed - an updated version of the software is now available. Go to Realcom Imputation >>
03-Mar-11: Welcome to the Centre for Multilevel Modelling's new web site
We are pleased to announce the publication of our new Centre for Multilevel Modelling web site:
https://www.bristol.ac.uk/cmm/
MLwiN's new address is:
https://www.bristol.ac.uk/cmm/software/mlwin/
If you go to our old web site you be redirected automatically to the new one. Everything that was on our old site is now here, mostly within the same structure, it is just the address that has changed - and we have a new simplified appearance. We apologise for any inconvenience you may have in changing your bookmarks.
If there is anything you cannot find please let us know - email us at hilary.browne@bristol.ac.uk or info-cmm@bristol.ac.uk.
25-Feb-11: ESRC PhD studentships in Advanced Quantitative Methods- applications deadline 5pm, Monday 11th April
The Advanced Quantitative Methods (AQM) pathway of the SWDTC offers ESRC +3 postgraduate research training in advanced quantitative methods in the social sciences and health. The pathway aims to raise the level of quantitative research skills among UK social scientists, in particular in advanced methods for secondary data analysis.
- ESRC PhD studentships in Advanced Quantitative Methods - applications deadline 5pm, Monday 11th April
South West Doctoral Training Centre, University of Bristol.
4 ESRC Studentships in Advanced Quantitative Methods in Social Sciences & Health
We are pleased to invite applications for 4 ESRC '+3' studentships on the South West Doctoral Training Centre's Advanced Quantitative Methods in Social Science & Health (AQM) pathway. The AQM pathway is based at the University of Bristol and these studentships will run from October 2011.
The AQM pathway involves PhD supervision and core training on the application of advanced quantitative methods in the social and health sciences. It is run by leading academics from the University of Bristol's Centre for Multilevel Modelling, Centre for Market & Public Organisation, the School of Geographic Sciences, and the School of Social & Community Medicine.
AQM is for social scientists who wish to develop their quantitative research skills in order to address important substantive social science questions, and in particular those questions involving secondary data analysis. AQM is equally for researchers who wish to specialise in the development of quantitative methods to address important but more generic social science questions, rather than specific applications to a substantive topic.
Applications are welcome from graduates in quantitative social science disciplines and more technical quantitative disciplines like statistics and econometrics. Applicants must have completed a Master's degree satisfying the ESRC Postgraduate Training & Development Guidelines by 30 September 2011.
The deadline for applications is 11 April 2011.
AQM is a priority area and it is expected that successful applicants will be eligible for an additional AQM stipend of £3,000. Thus the total annual maintenance stipend for 2011/12 will be in excess of £16,590.
For more details about AQM studentships and the applications procedure, visit our website at
22-Feb-11: New course module now on-line - Module 8
Module 8: Multilevel Modelling in Practice: Research Questions, Data Preparation and Analysis Lesson
In this module we consider the whole process of conducting a research project using multilevel modelling, taking as an example a study of ethnic differences in educational achievement and progress. You will need to log on or register onto our course to view the full module. Further details…
15-Feb-11: Realcom Impute updated
A bug has been fixed in realcom-impute and realcom mixed response, that could have given incorrect values for some level 2 random effects associated with level 1 responses when level 1 categorical responses were present.
15-Feb-11: Previous CMM newsletters
- newsletters from 2011 onwards - select Advanced search (you can also find this link on on the left-hand side of this page) and search for news based on date or keyword
- email newsletters 2007-2010 >>
- PDF newsletters 1989-2006 >>
Sign up to the Centre for Multilevel Modelling newsletter - email sent to our subscribers about once per month) with our latest news such as MLwiN releases and workshop dates. You can unsubscribe at any time and we won't share your details with anyone else (see our privacy policy)