STRUCRTUES for Building, Learning, Applying and Computing Statistical Models
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.
Organisations
Publications
Browne W
(2010)
MCMC Sampling for a Multilevel Model With Nonindependent Residuals Within and Between Cluster Units
in Journal of Educational and Behavioral Statistics
Browne WJ
(2009)
The use of simple reparameterizations to improve the efficiency of Markov chain Monte Carlo estimation for multilevel models with applications to discrete time survival models.
in Journal of the Royal Statistical Society. Series A, (Statistics in Society)
Carpenter J
(2011)
REALCOM-IMPUTE Software for Multilevel Multiple Imputation with Mixed Response Types
in Journal of Statistical Software
Cheung C
(2011)
Understanding contextual effects on externalizing behaviors in children in out-of-home care: Influence of workers and foster families
in Children and Youth Services Review
Fiona Steele (Author)
(2009)
e-learning in LEMMA
Garson, G. David
(2012)
Hierarchical Linear Modeling: Guide and Applications
George Leckie (Author)
(2009)
Fitting multilevel models to the English national pupil database to partition family, school and area effects
George Leckie (Author)
(2011)
The social relations model for count data with application to inter-household meat sharing in Nicaragua
George Leckie (Chair)
(2010)
School effect on pupil outcomes : quantitative methods and applications
George Leckie (Author)
(2011)
Running MLwiN from within stata : the runmlwin command
George Leckie (Author)
(2011)
Running multilevel models in MLwiN from within stata : runmlwin
Goldstein
(2010)
Multilevel Statistical Models
Goldstein H
(2009)
Multilevel Multivariate Modelling of Childhood Growth, Numbers of Growth Measurements and Adult Characteristics
in Journal of the Royal Statistical Society Series A: Statistics in Society
Goldstein H
(2009)
Multilevel models with multivariate mixed response types
in Statistical Modelling
Harvey Goldstein (Author)
(2010)
Multilevel multiple imputation using the latent normal model
Jen MH
(2009)
Global variations in health: evaluating Wilkinson's income inequality hypothesis using the World Values Survey.
in Social science & medicine (1982)
Jenkins J
(2012)
The role of maternal factors in sibling relationship quality: a multilevel study of multiple dyads per family.
in Journal of child psychology and psychiatry, and allied disciplines
Johnston R
(2010)
Measuring Segregation-A Cautionary Tale
in Environment and Planning A: Economy and Space
Leckie G
(2010)
School, Family, Neighbourhood: Which is Most Important to a Child's Education?
in Significance
Leckie G
(2011)
Understanding Uncertainty in School League Tables*
in Fiscal Studies
Leckie G
(2012)
Multilevel Modeling of Social Segregation
in Journal of Educational and Behavioral Statistics
Leckie G
(2011)
A Note on 'The Limitations of School League Tables to Inform School Choice'
in Journal of the Royal Statistical Society Series A: Statistics in Society
Leckie G
(2009)
The Limitations of using School League Tables to Inform School Choice
in Journal of the Royal Statistical Society Series A: Statistics in Society
Masaoud E
(2010)
Statistical Modelling of Neighbor Treatment Effects in Aquaculture Clinical Trials
in Journal of Agricultural, Biological, and Environmental Statistics
Description | LEMMA 2 had four interrelated elements: 1) development of multilevel statistical methods to better represent and understand social processes; 2) application of new methods to address a range of important social science questions; 3) development of user-friendly software to implement new methods; and 4) further development of the LEMMA virtual learning environment and other online resources. The project addressed three broad substantive questions: social segregation, school effectiveness, and family dynamics. Methodological research was in three areas: handling missing data in multilevel models; extending multilevel models to allow for dependencies between units at any given level (e.g. between schools who share catchment areas); and methods for studying influences of individuals on other individuals within a social group. A range of software tools were written to make these and other methodological developments accessible to researchers. The findings of one research project indicate that choosing a child's school based on government league tables is inadvisable because the tables ignore the uncertainty that arises from predicting a school's future performance based on past performance. Another project, using data from half a million English children, measured the relative contributions of school, neighbourhood, family and child characteristics on children's educational progress from age 11 to 16. Counter to previous research findings, the project showed that approximately half of the variability in progress that standard approaches would attribute to students actually arises from family influences, while a non-trivial proportion of the variability that would usually be attributed to schools is better described as neighbourhood variability. |
Exploitation Route | SCIENTIFIC IMPACT The project has made significant scientific contributions to the development and application of statistical methods for the analysis of complex social processes, the development of user-friendly software to make advanced statistical methods accessible to social researchers, and the provision of face-to-face and online training. Impacts include: Methods and software - Further enhancements to the MLwiN software (e.g. to improve data entry and efficiency of model estimation). 2630 users have taken advantage of the free UK academic download since the start of LEMMA 2. - Development of multiple-imputation methods for multilevel data structures which can handle missing data on mixtures of continuous and discrete variables defined at different levels, and the associated free REALCOM-Impute software. - Development of a new Stata command, runmlwin. - Development of the beta version of the Stat-JR software system which allows model fitting in multiple software packages through a common interface, allows greater flexibility for estimation of new models, and improves computational efficiency of existing methods. - A multilevel longitudinal structural equation model for estimation of reciprocal effects among individuals in a social group. Applied research - Significant contribution to UK and international debates on the use of school league tables for evaluating school performance and informing parental choice, and on the relative effects of school, area and family on pupil progress. - Further development of a new approach for modelling trends in social segregation. Training - Delivery of 25 days of training, including two 3-day intensive research workshops which supported researchers in analysing their own data. We also contributed to workshops in Belfast, Edinburgh and London. - Further development of the LEMMA Virtual Learning Environment which now has 8519 registered users. - Provision of support and materials to enable other academics to run their own courses. ECONOMIC/SOCIETAL IMPACT Methods, software and training resources developed under LEMMA have achieved economic and societal impacts through use by academics engaged in policy research and in research commissioned by government and third sector organisations. Work on league tables has raised awareness of their limitations for measuring institutional performance. More specifically: - Research on league tables has encouraged debate on their appropriateness for assessing performances of schools and other institutions, and their usefulness for informing parental choice. Goldstein is co-author of a major British Academy policy report on league tables in the public sector (Foley and Goldstein 2012). The research has also contributed to debates in Australia and New Zealand (and discussions with the main teaching union in NZ). - MLwiN has been used in survey methods research, including ESRC Survey Design and Measurement Initiative projects on nonresponse (G. Durrant with Steele, J. Bynner with Goldstein, and P. Lynn) and interviewer effects (P. Sturgis). This work has involved Natcen and ONS and has informed survey practice on strategies for reducing nonresponse and interviewer bias. MLwiN is also widely used in research on educational assessment conducted for assessment agencies and examination boards (e.g. AQA and QCA). MLwiN and REALCOM-impute were used in analysis of a longitudinal survey of prisoners for the Ministry of Justice (I. Brunton-Smith, Surrey). - Since the project start, MLwiN has been purchased by 572 non-academics. New users include the Scottish Government, Surrey Country Council, Matrix Knowledge Group, Statistics Norway, UNESCO Chile and Philips Electronics. Please see the End of Award Report (3b) for other examples of non-academic use of MLwiN. REALCOM-Impute and runmlwin are free to all. - The LEMMA VLE provides free training to all, reducing R&D and training costs for academics and non-academics alike. Uptake among non-academics has been high (1258 users, 15% of total). |
Sectors | Education Healthcare Government Democracy and Justice |
Description | Not yet required (will complete 12 months after end of award) |
Description | Communicating uncertainty in school league tables |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Primary Audience | |
Results and Impact | Paper at CCSR Seminar Series |
Year(s) Of Engagement Activity |
Description | Educational accountability (Queensland; Speaker: H. Goldstein) |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Professional Practitioners |
Results and Impact | Presented at Queensland University of Technology http://eprints.ncrm.ac.uk/2014/ http://eprints.ncrm.ac.uk/2014/ |
Year(s) Of Engagement Activity | 2010 |
URL | http://eprints.ncrm.ac.uk/2014/ |
Description | How to measure social and educational segregation |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Primary Audience | |
Results and Impact | The extent of variation (segregation) in socio-economic factors such as income and education, among groups of people is increasingly seen as an important measure of social progress. Measurement of such inequalities is not totally straightforward, an traditional procedures turn out to have real drawbacks. The session will look at new ways of measuring segregation with an application to differences among schools. |
Year(s) Of Engagement Activity |
Description | Multilevel event history analysis of birth intervals |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Primary Audience | |
Results and Impact | Presented at Seminar for the Scottish Social Survey Network |
Year(s) Of Engagement Activity |
Description | Multilevel event history models with applications to the analysis of recurrent employment transitions (Belfast, 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 | Presented at Northern Ireland Statistics and Research Agency None reported |
Year(s) Of Engagement Activity | 2011 |
URL | http://www.cardi.ie/events/northernirelandlongitudinalstudyresearchforumguestlecturefionasteele |
Description | Parental choice and league tables |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Primary Audience | |
Results and Impact | Presented at Griffith University |
Year(s) Of Engagement Activity |
Description | School league tables, accountability and parental choice |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Primary Audience | |
Results and Impact | Paper at CCCS Seminar |
Year(s) Of Engagement Activity |
Description | Selection effects in family research |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Primary Audience | |
Results and Impact | Paper at Workshop on "Advances in Fertility and Family Research" |
Year(s) Of Engagement Activity |
Description | The STAT-JR package and its potential use with social network models |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Primary Audience | |
Results and Impact | Seminar presentation |
Year(s) Of Engagement Activity |