LEMMA 3: Longitudinal Effects, Multilevel Modelling and Applications

Lead Research Organisation: University of Bristol
Department Name: Education

Abstract

Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.

Publications

10 25 50
 
Description The LEMMA 3 node combined methodological and substantive research on longitudinal data analysis with software development and capacity building. Methodological research included a review of existing approaches (Szmaragd et al., Longitudinal and Lifecourse Studies), and the development of new methods for fitting causal models to incomplete panel data with continuous and categorical outcomes (Clarke et al., Statistical Science). Previous research on multiple imputation for missing data has been extended to situations where the model of scientific interest is for multivariate responses of mixed type and where missing data may be on the responses and/or covariates (Goldstein et al., J. Roy. Stat. Soc. A). Multiple imputation methods have also been applied in record linkage, which can be viewed as a case of partially known (missing) data where a set of candidate records from a secondary file is to be linked to a single record in a basic file of interest (Goldstein et al., Statistics in Medicine).

Methodological work was motivated by substantive questions in education, health and psychology using data from panel studies, ecological momentary assessment and population registers. One project investigated the effects of job loss and re-employment on men's mental health using a simultaneous equations modelling approach to allow for selection effects (Steele et al., Epidemiology). Standard multilevel models have been extended to allow heterogeneity to depend on covariates, with applications to analysis of variation in education attainment within and between U.S. schools in (Leckie et al., J. Educ. Behav. Res.) and patterns of ethnic composition and segregation in England (Leckie and Goldstein, J. Roy. Stat. Soc. A). Swedish educational data were analysed using multilevel behavioural genetic models to decompose environmental and genetic influences on student performances (French et al., in progress). Another strand investigated the impact of stress on cognitive function of telephone helpline nurses, using a multilevel structural equation modelling approach to analyse real-time call record data on stress combined with post-shift outcomes (Steele et al., J. Roy. Stat. Soc. A).

In order to ensure that new methods are accessible to the social science user community, they are implemented in software, including packages developed by the LEMMA team (MLwiN, the new Stat-JR system and the runmlwin and runmixregls Stata commands) and other specialist and mainstream software (e.g. Stata, Mplus and aML). Stat-JR also provides interoperability with other statistical packages such as Stata, R, SPSS, MLwiN, WinBUGS and aML. All software developed under LEMMA is free to UK academics (MLwiN and Stat-JR) or all users (runmlwin and runmixregls). Extensive documentation for each package is available from the Centre for Multilevel Modelling website. In addition, most journal articles include online appendices with syntax and detailed descriptions of how to replicate analyses using available software. Other resources developed to build capacity in the analysis of longitudinal and complex data include six new modules written for the LEMMA virtual learning environment on multilevel modelling. The VLE now has over 16,000 registered users.
Exploitation Route Methods for handling nonresponse in longitudinal and other multilevel data have applications across the social and health sciences. The multiple imputation and data linkage methods developed under the project have been made accessible to other researchers through their implementation in the computationally efficient Stat-JR software, together with an online module on missing data and an electronic book. Further work on interactive electronic-books is underway as part of another ESRC grant ("The use of interactive electronic-books in the teaching and application of modern quantitative methods in the social science", led by Browne). Harvey Goldstein has been advising the Department of Education about the use of multiple imputation in a new longitudinal study of young people (a new LSYPE cohort), and is a member of an advisory group on data linkage in the Avon Longitudinal Study of Parents and Children (ALSPAC).

Leckie is co-authoring applied papers with academics external to the grant, applying the methods and software implementation (runmixregls Stat-JR template) developed on the grant to allow within-cluster heterogeneity to depend on covariates and unobserved cluster heterogeneity.

We anticipate that our methodological reviews and tutorials will help quantitative researchers from any discipline to understand the differences and similarities between alternative approaches to longitudinal data analysis. These resources and online modules provide practical guidance on the analysis and interpretation of longitudinal data which will be useful to both academic and non-academic audiences.

New methodology developed under the project also has the potential for widespread use across the social sciences and beyond, and this is facilitated by describing advanced methods in the context of substantive applications. Further work on causal modelling will be carried out by the ESRC Research Centre on Micro-social Change (MISOC, co-I Clarke). In the short term we expect that methods developed under the project will be used primarily by academic researchers, with impact on non-academic users arising indirectly through substantive research conducted by academic social scientists.
Sectors Communities and Social Services/Policy,Education,Healthcare

 
Description The primary aims of the project were to develop statistical methods for longitudinal data, and to build capacity in longitudinal data analysis through provision of software and training. While all methodological developments were motivated by substantive questions, the applications were mainly illustrative. As such, the impact of the research has been largely scientific to date. In the longer term, research conducted under the project has considerable potential to achieve indirect societal and economic impact through adoption of the new methods by academics engaged in policy research and in research commissioned by government and third sector organisations. Our dissemination strategy and training activities were designed to facilitate adoption by social scientists in the following ways: (i) review and tutorial articles comparing different analytical approaches (e.g. fixed and random effect models, and subject-specific and population-averaged models); (ii) articles in applied statistics and social science journals, including online supplementary materials with annotated software syntax; (iii) provision of free software that implements new methods; (iv) short courses and 'analyse your own data' workshops providing one-to-one support; and (v) provision of extensive free online training materials. RESEARCH (i) Missing data and data linkage As a member of the advisory group for the new cohort of the Longitudinal Study of Young People in England (LSYPE, now Next Steps), Harvey Goldstein gave advice on methods to adjust for missing data. This advice has informed a recent call for proposals on ways to handle missing data. As a member of a steering group on large scale health data linkage at the Health and Social Care Information Centre (HSCIC), Harvey Goldstein has argued for the need for sound data linkage methodology, which has informed their thinking and will be reflected in the future way in which they organise their data linkage work. (ii) Methods for the analysis of longitudinal data where the predictors are measured more frequently than the outcome Fiona Steele, Paul Clarke and George Leckie worked with health psychologists at Aberdeen University (Derek Johnston and Julia Allan) on methods for assessing the effects of stress on cognitive function of telephone helpline nurses in Scotland. Fiona Steele has applied similar methods in another study on whether cardiac-related beliefs and mood from real-time diaries predict subsequent attendance at cardiac rehabilitation (led by Martyn Jones, Professor of Healthcare Research at Dundee). This research has been widely disseminated to clinical groups across Scotland and, based on the findings of the research, further funding is being sought from Health Foundation to alter the message given to acute coronary syndrome (ACS) patients attending rehabilitation. If funded, the intention is to routinize the short form diary questionnaire for all ACS patients across Tayside, which will allow monitoring of changes in key study variables over time once this intervention happens. More generally, the SEM approach has great potential for analysis of real-time data, which are now widely collected in psychology. For example, there are plans for another project on the effect of tooth-brushing (measured repeatedly) on children's dental health. SOFTWARE The MLwiN software continues to be free for UK academics, and has now been downloaded 6524 times. The Stata command runmlwin allows MLwiN to be run directly through Stata, and has been extremely successful with 11,337 downloads and 136 citations in journal articles. The Stat-JR software system has been downloaded 1313 times. TRAINING Five new modules were written for the popular LEMMA online course. Topics include three-level models, cross-classified models, multiple-membership models, models for repeated measures data and missing data (the latter was authored by colleagues at the London School of Hygiene and Tropical Medicine). The course now has 19383 registered users, 2610 (13.5%) of whom are non-academics. A high proportion of users (60.3%) are students, and many are from overseas (74.3%). Users of the online course, and participants at the training workshops, come from a range of disciplines, which increases the reach of the research and potential for future impact.
First Year Of Impact 2013
Sector Education,Healthcare
Impact Types Economic,Policy & public services

 
Description Advisory group for new Longitudinal Study of Young People in England cohort (Goldstein)
Geographic Reach National 
Policy Influence Type Participation in advisory committee
 
Description Evidence to the parliamentary select committee on education about school league tables (Goldstein)
Geographic Reach National 
Policy Influence Type Gave evidence to a government review
 
Description Steering group member for the Department for Education 16-19 Bursary Fund Impact Evaluation, 2012-14 (Leckie)
Geographic Reach National 
Policy Influence Type Membership of a guidance committee
 
Description Steering group of Health and Social Care Information Centre (HSCIC) to prepare recommendations about pseudonymisation of patient records as part of large scale health data linkage (Goldstein)
Geographic Reach National 
Policy Influence Type Membership of a guidance committee
 
Title Stat-JR software: a software environment for promoting interactive complex statistical modelling 
Description Stat-JR is a universal gateway to many specialised statistical packages, providing tools to help users learn about statistical methods and their implementation in a range of statistical packages. As well as offering such inter-operability, Stat-JR has its own estimation engines which can fit models no other statistical software caters for. Stat-JR offers a choice of different interfaces, including a point-and-click menu-driven interface (TREE), a command line interface, and an eBook-reader (DEEP); the latter allows users to produce interactive reports guiding readers through analyses. The modular structure of templates allows users to extend Stat-JR's functionality, facilitating the development, and dissemination, of new statistical methodologies and model families. 
Type Of Material Improvements to research infrastructure 
Year Produced 2012 
Provided To Others? Yes  
Impact The Stat-JR software has been used in several methodological research papers (published and submitted) to estimate new statistical models. Methods implemented in Stat-JR include multiple imputation for multilevel data, data linkage and random effects discrete-choice models for large-scale longitudinal data. Stat-JR has been used in several training workshops attended by approximately 100 researchers to date. The software is free to all UK academics. The Department for Education is considering using Stat-JR for multiple imputation in the new Longitudinal Study of Young People in England (LSYPE) cohort. Charlton, C.M.J.; Michaelides, D.T.; Parker, R.M.A.; Cameron, B.; Szmaragd, C.; Yang, H.; Zhang, Z.; Frazer, A.J.; Goldstein, H.; Jones, K.; Leckie, G.; Moreau, L.; Browne W.J. (2013) Stat-JR version 1.0. Centre for Multilevel Modelling, University of Bristol & Electronics and Computer Science, University of Southampton. MANUALS: Browne, WJ; Charlton, CMJ; Michaelides, DT; Parker, RMA; Cameron, B; Szmaragd, C; Yang, H; Zhang, Z; Goldstein, H; Jones, K; Leckie, G; Moreau, L (2013) A Beginner's Guide to Stat-JR's TREE interface version 1.0.0. Centre for Multilevel Modelling, University of Bristol & Electronics and Computer Science, University of Southampton. Michaelides, DT; Yang, H; Browne, WJ; Charlton, CMJ; Parker, RMA (2013) Stat-JR DEEP eBook Reader and Authoring Guide. Centre for Multilevel Modelling, University of Bristol & Electronics and Computer Science, University of Southampton. Browne, WJ; Charlton, CMJ; Michaelides, DT; Parker, RMA; Cameron, B; Szmaragd, C; Yang, H; Zhang, Z; Goldstein, H; Jones, K; Leckie, G; Moreau, L (2013) An Advanced User's Guide to Stat-JR version 1.0.0. Centre for Multilevel Modelling, University of Bristol & Electronics and Computer Science, University of Southampton. 
URL http://www.bristol.ac.uk/cmm/software/statjr/
 
Description A dynamic analysis of the relationship between employment transitions and mental health among British men (Biostatistics Manchester; Speaker: F. Steele) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Invited presentation at conference organised by Centre for Biostatistics, University of Manchester

None to report
Year(s) Of Engagement Activity 2013
 
Description A dynamic analysis of the relationship between employment transitions and mental health among British men (ICLS London; Speaker: F. Steele) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Invited presentation at conference organised by the ESRC International Centre for Lifecourse Studies, UCL, London

None to report
Year(s) Of Engagement Activity 2012
 
Description A dynamic analysis of the relationship between employment transitions and mental health among British men (LSE; Speaker: F. Steele) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Seminar for Department for Statistics, London School of Economics

None to report
Year(s) Of Engagement Activity 2013
 
Description A dynamic analysis of the relationship between employment transitions and mental health among British men (Leeds; Speaker: F. Steele) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Invited presentation at a Royal Statistical Society conference on "Statistical challenges in lifecourse research", University of Leeds.

None to report
Year(s) Of Engagement Activity 2012
 
Description A dynamic analysis of the relationship between employment transitions and mental health among British men (RMF Oxford; Speaker: F. Steele) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Presentation at ESRC National Centre for Research Methods "Research Methods Festival", Oxford. Presentation was part of a session on Multilevel Modelling: Methods and Applications (which was organised by the project team).

Presentation led to requests to present the work elsewhere.
Year(s) Of Engagement Activity 2012
 
Description A dynamic analysis of the relationship between employment transitions and mental health among British men (RSS conference Telford; Speaker: F. Steele) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Invited presentation in session on "Methods for Longitudinal Data Analysis", Royal Statistical Society Conference, Telford

None to report
Year(s) Of Engagement Activity 2012
 
Description A general procedure for handling missing data (Lisbon; Speaker: H. Goldstein) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Keynote address at Economics and Statistics meeting, Lisbon

None to report
Year(s) Of Engagement Activity 2013
 
Description A longitudinal analysis of residential choice in England: Differential push and pull effects of area characteristics (ISER, Speaker: F. Steele) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact seminar for Institute of Social and Economic Research, University of Essex

None reported
Year(s) Of Engagement Activity 2013
 
Description A longitudinal analysis of residential choice in England: Differential push and pull effects of area characteristics (IoE, Speaker: F. Steele) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact seminar for Department for Quantitative Social Sciences, Institute of Education, London

None reported
Year(s) Of Engagement Activity 2013
 
Description A longitudinal analysis of residential choice in England: Differential push and pull effects of area characteristics (Liverpool, Speaker: F. Steele) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Presentation at joint Geography-Planning and Methods meeting, University of Liverpool

None reported
Year(s) Of Engagement Activity 2013
 
Description A mixed logit model for estimation of push and pull effects in residential location choice (LSE; Speaker: F. Steele) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Presentation to the Statistics Department at London School of Economics.

No known impact.
Year(s) Of Engagement Activity 2013
 
Description A mixed logit model for estimation of push and pull effects in residential location choice (Southampton, Speaker: F. Steele) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Seminar at Southampton Statistical Sciences Research Institute (S3RI), University of Southampton

None reported
Year(s) Of Engagement Activity 2013
 
Description Adjusting for selection bias using multilevel simultaneous equations models (Berlin; Speaker: F. Steele) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Invited presentation at conference on Methodological Innovations in Inequality and Social Policy Research, WZB Social Science Research Center, Berlin

No known impact
Year(s) Of Engagement Activity 2014
 
Description Analysing the effect of stress on information processing among telephone helpline nurses using multilevel modelling (Keele; Speaker: F. Steele) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Presentation to Primary Care Sciences Department at Keele University.

No known impact.
Year(s) Of Engagement Activity 2013
 
Description Cohort change and individual development of aerobic performance during childhood: Results from the Madeira Growth Study (RMF Oxford; Speaker: K. Jones) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Presentation in session on "Multilevel Models: Methods and Applications" at the ESRC National Centre for Research Methods "Research Methods Festival"

None to report
Year(s) Of Engagement Activity 2012
 
Description Dynamic panel-data modelling with structural nested mean models (LDA conference LSE; Speaker: P. Clarke) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Presentation at conference on "Methods for Longitudinal Data Analysis on the Social Sciences", London School of Economics (organised by project team). The conference was attended by 170 people, mainly from social and medical sciences.

None to report
Year(s) Of Engagement Activity 2014
 
Description Efficient fully general Bayesian models for handling missing values in multilevel data (LSHTM; Speaker: H. Goldstein) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Seminar to Medical Statistics group at London School of Hygiene and Tropical Medicine

None to report
Year(s) Of Engagement Activity 2013
 
Description Modeling the residual error variance in two-level random-coefficient multilevel models (World Statistics Congress Hong Kong; Speaker: G. Leckie) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Conference presentation at 59th World Statistics Congress, Hong Kong

None to report
Year(s) Of Engagement Activity 2013
 
Description Modelling heterogeneous variance-covariance components in two-level multilevel models (Amsterdam seminar; Speaker: G. Leckie) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Seminar at Netherlands Institute for the Study of Crime and Law Enforcement, Amsterdam

None to report
Year(s) Of Engagement Activity 2013
 
Description Modelling heterogeneous variance-covariance components in two-level multilevel models (MLM conference Utrecht; Speaker: G. Leckie) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Presentation at 9th International Multilevel Conference, Utrecht

None to report
Year(s) Of Engagement Activity 2013
 
Description Modelling heterogeneous variance-covariance components in two-level multilevel models with application to school effects educational research (RSS conference Newcastle; Speaker: G. Leckie) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Presentation at Royal Statistical Society Annual Conference, Newcastle

None to report
Year(s) Of Engagement Activity 2013
 
Description Modelling repeated measures growth data by aligning significant growth events and modelling changes in within-individual variability over time (LDA conference LSE; Speaker: H. Goldstein) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Presentation at conference on "Methods for Longitudinal Data Analysis on the Social Sciences", London School of Economics (organised by project team). The conference was attended by 170 people, mainly from social and medical sciences.

None to report
Year(s) Of Engagement Activity 2014
 
Description Modelling within- and between- group variance-covariance parameters in multilevel models with applications to educational research (QSS IoE; Speaker: G. Leckie) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Seminar at Institute of Education, Department of Quantitative Social Science Seminar Series

None to report
Year(s) Of Engagement Activity 2013
 
Description Multilevel modelling for longitudinal data (ICLS conference workshop; H. Goldstein) 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Postgraduate students
Results and Impact Half-day workshop at the Society for Longitudinal and Life Course Studies International Conference, Amsterdam

None to report
Year(s) Of Engagement Activity 2013
 
Description New perspectives on multilevel modelling (discussant at RSS meeting; Speaker: P. Clarke) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Discussant at meeting on "New perspectives on multilevel modelling" organised by Social Statistics section of the Royal Statistical Society, London

None to report
Year(s) Of Engagement Activity 2014
 
Description Numerical indigestion: how much data is really good for us? (LSE blog; Goldstein) 
Form Of Engagement Activity A magazine, newsletter or online publication
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Public/other audiences
Results and Impact Contribution to LSE 'Impact of Social Sciences' blog entitled "Numerical indigestion: how much data is really good for us?", 27 November 2012. This draws on research on the limitations of school league tables.

None to report
Year(s) Of Engagement Activity 2012
URL http://blogs.lse.ac.uk/impactofsocialsciences/2012/11/27/goldstein-how-much-data-is-good/#more-8398
 
Description Running MLwiN from within Stata: the runmlwin command (CSSR Manchester; Speaker: G. Leckie) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Seminar at Cathie Marsh Centre for Census & Survey Research (CCSR) Seminar Series, University of Manchester

None to report
Year(s) Of Engagement Activity 2012
 
Description Stat-JR - history, interoperability and eBooks (Reading; Speaker: W.J. Browne) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Seminar to Statistics group at University of Reading

None to report
Year(s) Of Engagement Activity 2014
 
Description Stat-JR and other software developed at the Centre for Multilevel Modelling (Open University; Speaker: W.J. Browne) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Seminar to the Statistics group at the Open University

None to report
Year(s) Of Engagement Activity 2013
 
Description Statistical Software at the Centre for Multilevel Modelling (Bath, Speaker: W.J. Browne) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Presentation to Statistics group at University of Bath

None to report
Year(s) Of Engagement Activity 2012
 
Description Statistical Software at the Centre for Multilevel Modelling (Glasgow, Speaker: W.J. Browne) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Presented to the Statistics group at University of Belfast

Not aware of impact
Year(s) Of Engagement Activity 2011
 
Description Statistical Software at the Centre for Multilevel Modelling (LSHTM, Speaker: W.J. Browne) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Presentation to Medical Statistics group at London School of Hygiene and Tropical Medicine

None to report
Year(s) Of Engagement Activity 2012
 
Description Statistical Software at the Centre for Multilevel Modelling (Nottingham, Speaker: W.J. Browne) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Presentation to Vet School at University of Nottingham

None to report
Year(s) Of Engagement Activity 2012
 
Description Statistical interoperability: What, why, when and how? The Stat-JR experience (RSS computing section; Speaker: W.J. Browne) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Presentation at meeting of the Royal Statistical Society Computing Section, London.

None to report
Year(s) Of Engagement Activity 2013
 
Description Statistical modelling for record linked data (ADLS conference; Speaker: H. Goldstein) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Invited presentation at Administrative Data Liaison Service conference on Data Linkage, Royal Statistical Society, London

None to report
Year(s) Of Engagement Activity 2012
 
Description Stressed nurses and g-computation: problem solved? (ICLS London; Speaker: P. Clarke) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Invited presentation at conference of ESRC International Centre for Lifecourse Studies, UCL, London

None to report
Year(s) Of Engagement Activity 2013
 
Description The Stat-JR package and its interoperability and ebook functionality (Multilevel conference Utrecht; Speaker: W.J. Browne) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Presentation at International Multilevel Conference, Utrecht

None to report
Year(s) Of Engagement Activity 2013
 
Description The Stat-JR package and its interoperability and ebook functionality (spatial modelling Bristol; Speaker: W.J. Browne) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Presentation to Spatial Modelling Group, School of Geographical Sciences, University of Bristol

None to report
Year(s) Of Engagement Activity 2013
 
Description The Use of Interactive E-Books for Teaching Bayesian Statistical Modelling using the Stat-JR package (Biometric Society St Andrews; Speaker: W.J. Browne) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Invited presentation at 4th Channel Network Conference, International Biometric Society, University of St Andrews

None to report
Year(s) Of Engagement Activity 2008,2013
 
Description The Use of Interactive E-Books for teaching Bayesian Statistical Modelling and Missing Data Methods using the Stat-JR package (RSS conference Northumbria; Speaker: W.J. Browne) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Invited presentation at annual conference of Royal Statistical Society, University of Northumbria

None to report
Year(s) Of Engagement Activity 2013
 
Description The relationship between changes in mental health and changes in employment status (CLS conference; Speaker: R. French) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Presentation at Centre for Longitudinal Studies Cohort Conference, Institute for Education, London

None to report
Year(s) Of Engagement Activity 2012
 
Description The relationship between changes in mental health and changes in employment status (SLLC Paris; Speaker: R. French) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Presentation at Conference of the European Child Cohort network and Society for Longitudinal and Life Course Studies, Paris

None to report
Year(s) Of Engagement Activity 2012
 
Description The scope of multilevel models (Beijing; Speaker: K. Jones) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Workshop for the Institute of Geographic Sciences, Chinese Academy of Sciences, Beijing, China, November 6-7th, 2013

None to report
Year(s) Of Engagement Activity 2013
 
Description Trajectories of frailty for the elderly in Britain (BSPS Swansea; Speaker: R. French) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Presentation at the British Society for Population Studies conference, Swansea

None to report
Year(s) Of Engagement Activity 2013
 
Description Using IVs to adjust for unobserved confounding in the mediation analysis of a complex psychiatric trial: potential and limitation (Leeds; Speaker: P. Clarke) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Presentation at Clinical Trials Research Unit, University of Leeds

None to report
Year(s) Of Engagement Activity 2013
 
Description Using interactive eBooks to teach and communicate multilevel analytical theory and practice (MLM conference Utrecht; Speaker: R. Parker) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Presentation at 9th International Multilevel Conference, Utrecht

None to report
Year(s) Of Engagement Activity 2013
 
Description Using panel data to assess the effect of changes in employment status on mental health ('Thinking Futures' festival Bristol; Speaker: Steele) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Public/other audiences
Results and Impact About 25 members of the public attended a talk and panel discussion on the Understanding Society study, as part of the Thinking Futures Festival, University of Bristol, November 2012. The session was entitled "Love, Living and Learning: What 40 000 households can tell us about life if 21st century Britain".

None to report
Year(s) Of Engagement Activity 2012
 
Description Using the STAT-JR software package for statistical analysis (Bristol; Speaker: W.J. Browne) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Presentation at "Structure and Uncertainty" workshop, University of Bristol

None to report
Year(s) Of Engagement Activity 2012
 
Description Using the STAT-JR software package for statistical analysis (RMF Oxford, Speaker: W.J. Browne) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Presentation at the ESRC National Centre for Research Methods "Methods Festival", Oxford

None to report
Year(s) Of Engagement Activity 2012
 
Description Using the STAT-JR software package for statistical analysis (RSS annual conference Telford; Speaker: W.J. Browne) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Invited presentation at the annual conference of the Royal Statistical Society, Telford

None to report
Year(s) Of Engagement Activity 2012
 
Description What is event history analysis? (RMF Oxford; Speaker: F. Steele) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Postgraduate students
Results and Impact Presentation at the ESRC National Centre for Research Method "Research Methods Festival", Oxford. About 20 attended presentation, but video of presentation was made available online after the event.

Received email requests for help from 2 students who had attended presentation (or seen video).
Year(s) Of Engagement Activity 2012
 
Description What is multilevel modelling? (RMF Oxford; Speaker: K. Jones) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Postgraduate students
Results and Impact Invited presentation at the ESRC National Centre for Research Methods "Research Methods Festival", Oxford. The presentation was attended by around 30 people, but the video of the presentation (on the NCRM website) was viewed by many more.

None to report
Year(s) Of Engagement Activity 2012
 
Description runmlwin - a program to run the MlwiN multilevel modelling software from within Stata (RMF Oxford; Speaker: G. Leckie) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Presentation in session on "Multilevel Models: Methods and Applications" at ESRC NCRM Research Methods Festival, Oxford

None to report
Year(s) Of Engagement Activity 2012