Value of Meta-information System for the Czech Statistical Office Topic 2(i) Advocating for metadata in corporate context

1
Value of Meta-information System for the Czech
Statistical OfficeTopic 2(i) Advocating for
metadata in corporate context

• Joint UNECE/Eurostat/OECD work session
  on metadata Luxembourg, 911 April
  2008 Ebbo Petrikovits
• ebbo.petrikovits_at_czso.cz

2
Introduction

• In 2005 two new and important project were
  launched
• Reform of statistical survey system (SSS)
• Statistical meta-information system (SMS)
• based on the SMS Vision
• In 2006
• Reform of the SSS was transformed into Redesign
  of statistical information system
• SMS became a standard part of the Redesign project

3
Redesign of SIS - major goals

• reducing response burden and boosting respondent
  motivation
• improving quality of statistical information
• optimising production of statistical information
  in the CZSO
• designing a conceptual model of Redesigned SIS
  and of SMS
• defining a unified architecture of statistical
  tasks
• increasing users comfort

4
Redesign of SIS - core principles

• systematic assessment and evaluation of
  statistical data requirements
• increasing share of administrative data
• increasing use of data modelling
• implementation of SMS
• implementation of statistical data warehouse
• freeze of statistical surveys for 2-3 years
• avoiding redundancy in statistical surveying

5
Unification of statistical processes

• Work on the GAS-SIS opened the need for
  description and standardization of the key
  process - production and dissemination of
  statistical information
• We proposed a model of this process
• It consists of 7 main phases
• Inside the phases we defined set of activities

6
Key process

• Phases
• Evaluation of users requirements
• Definition of statistical task
• preparation of data collection and processing
• Data collection
• Data processing
• Data analysis and output production
• Dissemination

7
Links to other processes

• Supporting processes
• Costs controlling
• Work efficiency evaluation based on the
  processing quality

8
SMS goals

• principle goal - to support, standardize and
  describe the key process in statistics
• in this context - support of
• management of methodology-related activities
• assessment of statistical data quality
• monitoring of respondent burden
• integration of SIS with public administration and
  international organizations
• design, implementation and management of
  statistical tasks

9
SMS Architecture

• Based on the SMS Vision
• Global Architecture of SMS (GA-SMS)
• defined the basic principles and rules for design
  and implementation
• set up the IT environment

10
Content of the SMS
Statistical Registers
Statistical Tasks
Statistical Quality
Users
SMS
Time Series
Dissemination
Respondents
Data Fund
GA-SMS
Statistical Classifications
Statistical Variables
11
SMS implementation strategy

• definition and development of individual
  subsystems
• implementation of individual subsystems
• tests of individual subsystems
• integration tests
• semi-operational running
• pilot project on selected statistical task
• operational running
• step-by-step transition of existing statistical
  tasks into SMS

12
Technological environment

• Technological infrastructure
• UNIX operating system
• Oracle database system
• PC with OS Windows/Linux as a client workstations

13
Technological principles

• Work stations independent on operating system
• Internet browser as a basic tool for
  communication
• No supplementary products on the client work
  station
• Oracle Forms as a basic tool for development of
  applications
• Access to the SMS subsystems via SMS Access Portal

14
Subsystem CLASSIFICATION

• Inspired by Neuchâtel Classification Model
• Described objects
• classification
• version of classification
• variant of classification
• code-list
• basic code-list
• combined code-list

15
Subsystem VARIABLES (1)

• Described objects
• statistical variables
• basic
• subject-matter broken-down
• On conceptual level very similar to the Neuchâtel
  Variables Model

16
Subsystem VARIABLES (2)

• Detailed model
• a statistical data is identified by set of
  metadata
• this set we divide into four complex variables
• complex variable consists of elementary variables
• elementary variable consists of
• type of variable
• specification of variable
• type/specification of a elementary variable
  consists of
• code-ist code
• code of code-list item
• valid from

17
Subsystem VARIABLES (3)

• Complex variables
• statistical variable - describes the content of a
  data
• statistical object - describes observed object
• time variable - describes the current time of
  observation
• complementary variable - describes other
  identification attributes which do not belong to
  the above mentioned variables

18
Subsytem TASKS

• Described objects
• statistical task
• structure of a questionnaire
• elements of a questionnaire
• input/output sets
• VIP (virtually identified items)
• time-tables
• program modules and runs
• response duty specification

19
State-of-art in SMS implementation

• CLASS, VAR
• tests of version 1.0 finished,
• preparation of real code-lists, classifications
  and statistical variables needed for the pilot
  test
• tests of version 1.1
• TASKS
• preparation of tests
• training of the member of the test team

20
SMS Management

• management in the implementation phase
• project approach applied
• multi-professional teams
• permanent monitoring from the top management
• management in the operational run phase
• establishment of the SMS administration

21
SMS management in the implementation phase
22
SMS management in the operational phase
SMS Administration
Central Administration
CLASS Administration
VAR Administration
TASKS Administration
QUALITY Administration
S-Administrator
SMS -Methodologist
S-Administrator
S-Administrator
S-Administrator
C-Administrator
C-Administrator
C-Administrator
C-Administrator
S-Methodologist
S-Methodologist
S-Methodologist
S-Methodologist
Technology Administration
S-Administrator - subsystem administrator C-Admini
strator - content administrator T-Administrator -
technology administrator S-Methodologist -
subsystem methodologist
T-Administrators
23
Major findings (1)

• SMS strategy - content and methodology -gt fully
  in the responsibility of the statistical office
• SMS design and implementation should be organize
  in multi-professional teams -gt increasing of
  research capacity
• Development of software applications -gt may be
  outsourced (contract based)
• Testing -gt close cooperation of the project teams
  and the contractor (follow-up the time-schedule
  is necessary)

24
Major findings (2)

• Coordination of time schedules for Redesign
  project and SMS project
• Motivation of project teams - sharing of
  knowledge an information
• Monitoring of the activities by
• the top management - quarterly
• the steering committee - quarterly
• the project task force - monthly
• project teams - weekly

25
Major findings (3)

• Importance of training and transfer of SMS
  know-how
• Focus on the subject matter topics and use of SMS
  tools in statistical practice is advisable

26
Thank you for your attention