1. The Approach and ideas of the HLG-BAS: Modernizing Official Statistics

2. Contents Context – Why modernize? The HLG-BAS: Modernizing official statistics Key Standards – GSBPM – GSIM Using GSIM in Practice – Example from Statistics Canada

3. What is the Data Deluge?

4. Prediction for 2020

5. In the last 2 years more information was created than in the whole of the rest of human history!

6. Competition? Private sector understands the value of data Google: – Real-time price indices – Public Data Explorer Facebook, store cards, credit agencies,... – What if they link their data? Will they provide an alternative to official statistics?

7. Efficiency Increasing demands Need for more flexibility Shrinking budgets Response burden constraints “Do more with less”

8. Enablers of change Technological advances – Including better collaboration tools Methodological advances Increasing collaboration between organisations

9. HLG-BAS – High-Level Group for Strategic Developments in Business Architecture in Statistics Created by the Conference of European Statisticians in 2010 10 heads of national and international statistical organisations Official Statistics Response

10. Endorsed by the Conference of European Statisticians in June 2011 We have to re-invent our products and processes and adapt to a changed world HLG-BAS Strategic Vision

11. The challenges are too big for statistical organisations to tackle on their own. We need to work together

12. Common generic statistics production GSBPM GSIM MethodsTechnology Statistical Concepts Information Concepts Statistical HowTo Production HowTo conceptual practical Modernize statistical production

13. Transform vision to reality New products Rationalised processes – “Plug and play” architecture – based on the GSBPM and GSIM Managing organizational change to support change and collaboration A strategy for modernization

14. Transform vision to reality New products Rationalised processes – “Plug and play” architecture – based on the GSBPM and GSIM Managing organizational change to support change and collaboration Conference of European Statisticians, June 2012 A strategy for modernization Endorsed

15. Using common standards, statistics can be produced in a more efficient way No domain is special! Do new methods and tools support this vision, or do they reinforce a stove-pipe mentality?

16. Introducing the GSBPM You are here

17. Why do we need the GSBPM? To define and describe statistical processes in a coherent way To compare and benchmark processes within and between organisations To make better decisions on production systems and organisation of resources

18. The GSBPM

19. Key features Not a linear model Sub-processes do not have to be followed in a strict order It is a matrix, through which there are many possible paths Some iterations of a regular process may skip certain sub-processes

21. The GSBPM is used by more than 50 statistical organizations worldwide to manage and document statistical production

23. Introducing the GSIM You are here

24. GSIM: A complementary initiative Another model is needed to describe information objects and flows within the statistical business process

27. Acquisition Program Methodology Statistical Program Information Request Dissemination Program Statistical Project Data Resource Population Units Concept Variable Classification Value Domain Provider Provision Agreement Statistical Products Data Set Data Structure Unit Data Structure Cube Structure Record Structures Process Step Execution Process Step Design Process Method Separation of Statistical, Acquisition, and Dissemination programs, with central role for Methodology Balanced support for all data acquisition channels Balanced support for multiple dissemination channels Shared Data Resource, maintained corporately, for use by all statistical programs Basic infrastructure for critical base elements Mapping of processes to support managed operations All structures and relationships described in metadata to support automated processes Statistical projects access shared data Provision for formalisation of arrangements for data acquisition and dissemination Process management for all areas of activity Methodology applies to processes in all areas Process Step Definition Process Control Rule Production Activity Conceptual Information

28. STATISTICS CANADA & GSIM How will Statistics Canada use it’s GSIM work to address real-world business problems ?

29. Outline Current Business Reality – the “Opportunity” Our Strategy – Corporate Business Architecture How does GSIM help us? Case Study– Rolling Estimates for Business Statistics Processing Conclusion

30. Current Business Reality Government deficit reduction Information industry relevance Policy leadership decisions

31. What action should one take? Reduce IT ? Do fewer surveys ? Lower quality ? Address fewer areas ? Do a multivariate analysis to decide ? StatCan budget breakdown RPP 2011-12

32. Innovate to transform the business Focused investment for change Common production services – Collection, Dissemination System platforms – business, social processing, Open Data Methodology innovation – e.g. Rolling Estimates Effective management and use of information Business innovation – e-collection, automation, integration

33. Architect cross-business solutions GSBPM GSIM

34. Hypothesis – GSIM will help us by providing the following: A standardized framework to aid in consistent and coherent design capture A foundation for standardized statistical metadata use throughout our systems Increased sharing of system components amongst national Agencies

35. Approach to test the hypothesis Active participation in workshops and sprints leading to GSIM 1.0 Pilot use of evolving GSIM model on a design from our Business Processing Platform – Rolling Estimates Use Case Evaluate and improve

36. Test Case – Rolling Estimates Element of StatCan’s Integrated Business Statistics Program – Common platform for sampling, questionnaire design, post collection of microeconomic surveys – Targets efficiencies, quality, responsiveness – Covers all aspects of the business surveys 130 surveys in 10 different programs integrated by 2016

37. Current State – Change Required Linear business model Collection follow-ups are not prioritized – need to focus on most impact, managed cost Multiple manual interventions Homogeneous editing strategies SamplingDisseminationAnalysisProcessingCollection

38. Iterative business model Optimizing the editing work done by both collection services and post collection Improved timeliness of survey results Re-directing subject matter editing efforts toward analytical activities Future state – Rolling Estimates

39. Rolling Estimates - Details Estimates are produced and analyzed regularly in a cycle until an acceptable level of quality is reached Iterative estimates with quality indicators for each domain of estimation as soon as an acceptable level of survey and administrative data will be available Non-Response and Failed Edits follow-ups and collection cut-off are driven by the quality of the estimates Allows for continuous realignment of the micro and macro- editing strategies based on the most recent available data. Collection, processing and micro/macro editing activities are done in parallel and not in a standard sequential way.

40. The Common Editing Strategy

41. Rolling Estimates – Process Flow

42. Initial application of GSIM framework

43. Conclusion We believe that the innovative use of GSIM across the design, build, and “run” areas will help us reach our business goals StatCan is contributing to its development We are focused on using new developments to validate our approach

44. GSIM: The “sprint’ approach The HLG-BAS decided to accelerate the development of the GSIM “Sprints” – 2 week workshops for 10-12 experts (IT, methodology, statistics,...) Sprint 1 – Slovenia, February 2012 Sprint 2 – Republic of Korea, April 2012 Integration Workshop, Netherlands, September 2012

45. Next steps GSIM v0.8 will be released in the next 2 weeks 3 weeks public consultation – Comments and feedback welcome Discussion at Workshop on Business Architecture in Statistics (Geneva, 7-8 Nov) GSIM v1.0 by the end of 2012

46. The big picture You are here

47. Standards-based Modernization

48. “Grand Unification” GSBPM Generalized Statistical Production System Common Generic Industrialized Statistics Common Generic Industrialized Statistics Methods Technolog y GSIM Practical Conceptual Grand Unification is a new approach that brings together the GSBPM and GSIM to make statistics To be developed...

49. Key points 1. Official statistics organizations have to modernize to survive 2. Modernization is not an IT issue! It is strategic: Defining the future of official statistics 3. GSBPM and GSIM are not software tools – they are new ways of thinking

50. More information GSBPM – http://www1.unece.org/stat/platform/display/metis/The+ Generic+Statistical+Business+Process+Model http://www1.unece.org/stat/platform/display/metis/The+ Generic+Statistical+Business+Process+Model GSIM – http://www1.unece.org/stat/platform/display/metis/Gene ric+Statistical+Information+Model+(GSIM) http://www1.unece.org/stat/platform/display/metis/Gene ric+Statistical+Information+Model+(GSIM) HLG-BAS – http://www1.unece.org/stat/platform/display/hlgbas http://www1.unece.org/stat/platform/display/hlgbas