This is a presentation given by Dr Chris Playford as part of the UK Data Service and Office for National Statistics 'Love your Code' event. 14th February, Office for National Statistics, London. This presentation draws on experiences of open social science research practices used throughout this project, as well as through Dr Playford's teaching experience at the University of Exeter Q-Step Centre. 9.30 Registration and coffee 10.00 Welcome Pete Stokes, Office for National Statistics 10.10 Introduction Andrew Engeli, Office for National Statistics and Louise Corti, UK Data Service 10.20 - 11.00 Making the case for reproducibility ADR UK: The case for reproducibility, Paul Jackson, ADR-UK Reproducibility and granular data: What’s the issue?, Stefan Bender, Deutsche Bundesbank The Turing Way: Reproducible, Inclusive, Collaborative Data Science, Kirstie Whitaker, Turing Institute and The UK Reproducibility Network 11.00- 11.15 Coffee break 11.15-12.15 The challenges of reproducibility in statistics: emerging standards Quality Assurance of Administrative Data, Catherine Bromley, Office for Statistics Regulation The Challenges of Reproducible Research and Teaching it, Chris Playford, QStep, Exeter Transparency and reproducibility for linked administrative datasets, Katie Harron, Institute of Child Health, UCL Open analytics at the Health Foundation, Fiona Grimm, The Health Foundation Discussion 12.15 - 13.00 Being reproducible in data publishing: a view from data producers Deriving variables: A Data Managers’ guide to DV best practice, Jess Bailey, NatCen Challenges of creating and documenting derived variables and harmonised datasets in longitudinal studies, Aida Sanchez, Centre for Longitudinal Studies and Dara O’Neill, CLOSER The Reproducibility Analytical Pipeline, Best Practice and Impact, Office for National Statistics Discussion 13.00 Lunch 13.40 Keynote: Ed Humpherson, Office for Statistical Regulation 14.00 Raising the bar: Reproducibility services Reproduction and replication: What we do and what we need, Lars Vilhuber, American Economic Association Reproduction and replication: Where we are going, Joan Llull (UAB), Data editor for Economic Journal How to reproduce: The Cascad Reproducibility Service, Christophe Perignon, Cascad, RunMyCode.org Discussion 14.45 Discussion Groups with tea: Reproducibility for UK national statistics. Facilitators Breakout 1: Quality review and certification for reproducibility: good code, bad method? Facilitators Louise Corti and Lars Vilhuber Breakout 2: Automating processes for reproducibility: benefits and issues around code tracking tools e.g. Gitlab, R Markdown and Jupyter Notebooks. Facilitators: Simon Parker, Cancer Research UK and Fiona Grimm, The Health Foundation Breakout 3: Assessing disclosure risk in code: rules, tools and skills. What about automated output checking? Facilitators: Chris Woods and Valerija Kolbas, UK Data Service 15.15 Roundtable: Towards a common standard for reproducibility with granular data Participants: Paulo Guimares (Bank of Portugal), Debora Price (Professor Cathy Marsh Institute, University of Manchester), Richard Heys (Office for National Statistics) 16.15 Closing note