GSBPM and GSIM in Statistics Norway Prepared by
GSBPM and GSIM in Statistics Norway Prepared by Rune Gløersen and Jenny Linnerud MSIS, Dublin 14 -16 April 2014
The GSBPM
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 development of production systems • To optimize organisation and allocation of resources
4 Statistics Norways Business process model
5 Statistics Norways Business process model Specify needs Develop and design Build Collect Process 1 2 3 4 5 Determine need for information Outputs 1. 1 Consult and confirm need Analyse Disseminate 6 7 Prepare data for dissemination database Build and enhance process components Classify and code Acquire domain intelligence 2. 1 3. 1 Establish frame and registers, select sample 4. 1 5. 1 6. 1 Frame, register and sample methodology Integrate production system with other systems Set up collection Micro-edit Produce statistics Produce product 2. 2 3. 2 4. 2 5. 2 6. 2 7. 2 Establish output objectives Data collection methodology Test production system Run collection Macro-control Quality assure statistics Release and promote product 1. 3 2. 3 4. 3 5. 3 6. 3 7. 3 Check data availability Process and analysis methodology Finalise collection Impute for partial non-response Manage customer queries 4. 4 5. 4 Interpret and explain statistics 1. 2 1. 4 2. 4 Prepare business case Production system 1. 5 2. 5 3. 3 Finalise production system 3. 4 Calculate weights and derive new variables 5. 5 6. 4 Prepare statistics for dissemination 6. 5 Finalise content 6. 6 7. 1 7. 4
Statistics Norways Business process model • Was mapped against GSBPM 4 in the CORA project • Slightly different on detailed level within Build • Some processes on detailed level placed differently within Process and Analyse • Was different with regards to archive as an overarching process, which has been better aligned with GSBPM 5
Complete documentation on our Intranet
GSBPM in Statistics Norway Streamlining Statistics Production
Categorising systems Java SAS Oracle Fame SIV/SIL Blaise Altinn (Idun, Kostra) SFU ssb. no FDM ISEE - statistic register Norsamu ISEE Driller -Google analytics (Trekkbas) Verify Telefinn SMIE SERES Presys Stat. Bank SELEKT X 12 -Arima Tau-Argus Mu-Argus SAS Insight Service Manager (Helpdesk, OTRS) Stat. population registers: “Projectplanning”: - Jira Document centers: Confluence (Trac, Wiki) Windows-server - National register Produktregister Metadataportals: - The Central Coordinating - Vardok Register for Legal Entities - Datadok - GAB – Landed property, - Stabas - ssb. no (About statistics) Address, Dwelling (map) LDA-app MS Office Smart. Draw Arc. GIS Websak SPSS Adobe
Summary • planning new statistics • prioritizing new projects (portfolio management) • improve existing work processes in statistical production • reducing portfolio of IT-systems • reducing risk • making a more complete business architecture • easier training and integration of staff.
11 Introduction to GSIM
We need consistent information • Modernisation of statistics requires: – reuse and sharing of methods, components, processes and data repositories – definition of a shared “plug-and-play” modular component architecture • The Generic Statistical Business Process Model (GSBPM) will help determine which components are required. • GSIM will help to specify the interfaces. 12
GSIM and GSBPM • GSIM describes the information objects and flows within the statistical business process.
GSIM in Statistics Norway - Vision META DATA
GSIM in Statistics Norway - Vision GSIM should lead to: • A foundation for standardised statistical metadata use throughout systems • A standardised framework for consistent and coherent design of statistical production • Increased sharing of system components
Remote Access Infrastructure to Register Data (RAIRD) Statistics Norway and the Norwegian Social Science Data Services (NSD) aim to establish • a national research infrastructure • providing easy access to large amounts of registerbased statistical data • managing statistical confidentiality • protecting the integrity of the data subjects. The work is funded by the Research Council of Norway. See: www. raird. no
RAIRD Information Model (RIM) Based on GSIM v 1. 1 • Design principles • Information objects New information objects for users (producers, administrators and researchers) Less information objects for details of the official production of statistics RAIRD continues out 2017
Research production process to be supported by RIM?
RIM Data Descriptions GSIM information objects • Input Unit Data and Metadata/Event History Data Resource • Analysis Data Sets • Disclosure Control • Final Product • Themes and Subject Fields • Classifications, Concepts, Variables, etc.
GSIM Glossary blue – not in RIM, yellow – in RIM
- Slides: 20