Grant code: ES/K007246/1

Period of funding: October 2013 - September 2017

• Background and objectives
• Software
• Workshops
• Journal articles
• Conference proceedings
• Other presentations
• Case studies (opens in new tab)
• People (opens in new tab)
• Workflow system

Background and objectives

The importance of the statistically trained researcher in the social sciences has increased dramatically over recent years, with a huge range of opportunities for exploration and understanding that have risen from the enormous amount of data becoming available. To help realise such opportunities, however, there is a need to provide up-to-date, effective training in quantitative methods, enabling applied researchers to tackle data with skill and understanding. Furthermore, computers have revolutionised the process of data analysis over the last fifty years. Nowadays, statistical software can be used to fit complicated statistical models, which has allowed less statistically-literate researchers to access very powerful modelling tools. But as is widely acknowledged, easy access does not bring about understanding, or lead to the close integration of research question and data analysis that marks out high-quality social science research.

The aims of this project are to provide computer-assisted support to increase the skills and understanding of quantitative principles and techniques of researchers in UK social science, and to create innovative tools for reproducible research. Our project proposes a radical solution to these objectives based on interactive electronic books (eBooks) that can be used both for training and research purposes. By building on our work on Stat-JR, carried out under the ESRC Digital Social Research (DSR) programme in our node (eStat), we will take forward the development of eBooks (with a new workflow system to support this work) in three distinct but interrelated directions, which have the potential to revolutionise quantitative methods training and research in UK social science. These three objectives are summarised as follows:

1. Developing eBooks to assist the researcher throughout the process of performing a statistical analysis in a manner tailored to their research questions and the features of their dataset. The eBook environment will reinforce important statistical concepts and good practice in the context of writing high-quality reports.
2. To make the goal of reproducible research a reality by using eBooks to create a record of how published results were obtained, which can be used by the original researcher (e.g. to revise analyses in response to referees’ comments) or by other researchers wishing to replicate the findings: in short to consider the use of eBooks as enhanced journal articles.
3. To train the social science research community in the use of quantitative research methods and statistical software via interactive eBooks, producing an online environment within which researchers can search and find accredited eBooks and published papers as templates for high-quality analyses.

Software

In this project we have developed many new features in the Stat-JR software package and in particular the Statistical Analysis assistant and workflow (see below) features. The Stat-JR software has had two releases in the lifetime of the project. Versions 1.0.4 and 1.0.5. Version 1.0.5. can be downloaded from here and the accompanying manuals including brand new guides on the Statistical Analysis assistant and workflows can be viewed from here. This grant has also supported the further development of the R2MLwiN R package and MLwiN software, as well as the creation of two additional practicals for the LEMMA online course.

Workshops

Stat-JR Workflow and eBook Workshop

For more details of this workshop see the CMM Workshops page:

• Date & location delivered:
  • 4th July 2016, Bristol

Stat-JR Workflow and eBook Workshop

Note that this workshop uses a development (not the release) version of Stat-JR to introduce, and invite feedback on, Stat-JR's new workflow system.

• Dates & locations delivered:
  • 3rd September 2015, Bristol
• Presentations & Practicals:
  • Presentation slides (pdf, 14.2MB, with splitting of animations in Powerpoint courtesy of PPspliT)
  • Practical Part 1 (pdf, 4.9MB)
  • Practical Part 2 (pdf, 4.6MB)
  • Practical Part 3 (pdf, 2.9MB)

eBook Writing in Stat-JR

• Dates & locations delivered:
  • 18th February 2015, Edinburgh
  • 26th September 2014, Bristol
• Practicals:
  • Practical Part 1 (pdf, 4.5MB)
  • Practical Part 2 (pdf, 3.9MB)
  • Practical Part 3 (pdf, 4.9MB)
• Templates:
  • To run through these practicals you will also need to download a zip file of additional Stat-JR templates and unzip it in the templates subdirectory of Stat-JR.

Modelling longitudinal data using the Stat-JR package

• Dates & locations delivered:
  • 7th July 2014, NCRM 6th ESRC Research Methods Festival, Oxford
• Presentations & Practicals:
  • Presentation slides (pdf, 3.3MB)
  • Practical Parts 1, 2 & 3 (pdf, 8.9MB)

Journal articles

2017

• Moreau, L. (2017) A canonical form for PROV documents and its application to equality, signature, and validation. ACM Transactions on Internet Technology, 17, (4). doi:10.1145/3032990
• Goldstein, H., Browne, W.J., Charlton, C (2017) A Bayesian model for measurement and misclassification errors alongside missing data, with an application to higher education participation in Australia. Jounal of Applied Statistics. doi:10.1080/02664763.2017.1322558
• Goldstein, H., Leckie, G., Charlton, C (2017) Multilevel growth curve models that incorporate a random coefficient model for the level 1 variance function. Statistical Methods in Medical Research. doi:10.1177/0962280217706728
• Brunton-Smith, I. Sturgis, P, Leckie, G. (2017) Detecting and understanding interviewer effects on survey data by using a cross-classified mixed effects location-scale model. J. R. Stat. Soc. A, 180: 551–568. doi:10.1111/rssa.12205
• Moreau, L., Batlajery, B.V., Huynh, T.D., Michaelides, D.T. (2017) A Templating System to Generate Provenance. IEEE Transactions on Softwre Engineering. doi:10.1109/TSE.2017.2659745

2016

• Zhang, Z, Parker, R.M.A., Charlton, C.M.J., Leckie, G., Browne, W.J. (2016) R2MLwiN: A Package to Run MLwiN from within R. Journal of Statistical Software 72(10), 1-43. doi:10.18637/jss.v072.i10

2015

• Moreau, L., Groth, P., Cheney, J., Lebo, T., Miles, S. (2015) The rationale of PROV. Web Semantics: Science, Services and Agents on the World Wide Web, 35(4), 235-257. doi:10.1016/j.websem.2015.04.001 
• Kwasnikowska, N., Moreau, L., Van den Bussche, J. (2015) A Formal Account of the Open Provenance Model. ACM Transactions on the Web. doi:10.1145/2734116

2014

• Leckie, G., French, R., Charlton, C (2014) Modeling Heterogeneous Variance–Covariance Components in Two-Level Models. Journal of Educational and Behavioral Statistics, 39(5) 307-332. doi:10.3102/1076998614546494

Conference proceedings

2016

• Michaelides, D.T., Parker, R., Charlton, C., Browne, W.J., Moreau, L. (2016) Intermediate notation for provenance and workflow reproducibility. In: 6th International Provenance & Annotation Workshop (IPAW '16), United States. 06 - 09 Jun 2016. doi:10.1007/978-3-319-40593-3_7

2015

• Moreau, L. & Groth, P. (2015) Provenance of Publications: A PROV Style for LaTeX. In: 7th USENIX Workshop on the Theory and Practice of Provenance, 8-9th July 2015 (TaPP 15), Edinburgh, Scotland.
• Moreau, L. (2015) Aggregation by provenance types: A technique for summarising provenance graphs. In: Graphs as Models 2015 (An ETAPS'15 workshop), Electronic Proceedings in Theoretical Computer Science, pages 129-144, London, UK, April 2015. doi:10.4204/EPTCS.181.9

Other presentations

2016

• Using Computers to Teach Statistics to Reluctant Researchers. Reseach seminar for the "Conversations in Education" seminar series by Bill Browne at the Graduate School of Education, University of Bristol, UK, 30 November 2016
• Creating a Statistical Analysis Assistant using Stat-JR. Presentation by Bill Browne at the RSS conference, University of Manchester, UK. 8th September 2016
• What are Statistical eBooks? Presentation by Bill Browne at the Research Methods Festival, University of Bath, UK, 6th July 2016
• Stat-JR: eBooks, workflows and other software developments at the Centre for Multilevel Modelling. Invited research seminar by Bill Browne for the Statistics and Probability Group at Durham University, UK, 8th February 2016.

2015

• Statistical Software developments at the Centre for Multilevel Modelling. Invited research seminar by Bill Browne at the School of Mathematical Sciences, University of Nottingham, UK, 17th December 2015.
• Stat-JR, eBooks, workflows and other software developed at the multilevel modelling centre. Invited research seminar by Bill Browne at the School of Mathematics, Statistics and Actuarial Science, University of Kent, UK, 29th October 2015.
• Enabling Provenance on the Web: Standardization and Research Questions. Keynote presentation by Luc Moreau at the 14th International Conference on WWW/INTERNET (ICWI) in Maynooth, Greater Dublin, Ireland, 24-26th October 2015.
• Developing Interactive eBooks and an Analysis Assistant to Teach and Apply Modern Quantitative Methods. Presentation by Richard Parker at the Royal Statistical Society (RSS) Annual Conference in Exeter, UK, 7th-10th September 2015.
• Developing Interactive eBooks and an Analysis Assistant to Teach and Apply Modern Quantitative Methods. Presentation by Richard Parker at the International Association for Statistical Education (IASE) 2015 Conference in Rio de Janeiro, Brazil, 22nd-24th July 2015.
• The use of electronic books for teaching statistical ideas with application to statistical ecology. Presentation by Bill Browne at the National Centre for Statistical Ecology (NCSE) Summer Meeting, Penryn Campus, University of Exeter, UK, 29th June - 1st July 2015.
• The Use of eBooks and a Statistical Analysis Assistant to Teach Multilevel Modelling. Plenary talk by Bill Browne at the 10th International Multilevel Modelling Conference in Utrecht, The Netherlands, 9-10th April 2015.

2014

• A Statistical Analysis Assistant – the future or folly? Presentation by Bill Browne at the Royal Statistical Society (RSS) Annual Conference, Sheffield, UK, 1-4th September 2014.
• An Introducation to PROV. Invited presentation by Luc Moreau for the Lotico London Semantic Web group at The Open Data Institute in London, UK, 7th May 2014.
• Stat-JR – history, interoperability and eBooks. Invited research seminar by Bill Browne at the Department of Mathematics and Statistics, University of Reading, UK, 20th February 2014.
• eBook for Causal Modelling and Missing Data Methods. Presentation by Richard Parker at the National Centre for Research Methods (NCRM) Annual Meeting, Windsor, UK, 16-17th January 2014.
• Introducing Stat-JR: a New Software Environment Designed to Promote Interactive Statistical Modelling. Invited research seminar by Richard Parker at the Centre for Behaviour and Evolution, Newcastle University, UK, 14th January 2014.

2013

• Statistical interoperability – what, why, when and how? The Stat-JR experience. Presentation by Bill Browne at a joint meeting of the Statistical Computing Section (SCS) of the Royal Statistical Society (RSS) and the British and Irish region of the Biometric Society (BIR) on Statistical Software Interoperability, London, UK, 14th November 2013.
• Stat-JR and other software developed at the multilevel modelling centre. Invited research seminar by Bill Browne for the Statistics Group at The Open University, 10th October 2013.
• Modeling the residual error variance in two-level random-coefficient multilevel models. Poster spotlight by George Leckie for the 2013 World Statistics Conference in Hong Kong, August 2013