Conceptual metadata and process metadata

1
Conceptual metadata and process metadata
WP18

• Max Booleman (Statistics Netherlands)

2
How to improve dissemination (policy) and the
transparency of the statistical system?

• Starting point our mission
• Introduction metadata model
• Dissemination policy
• Coherence (and Comparability)
• Transformation process of StatLine

3
Mission of Statistics Netherlands

• Core-business dissemination of undisputed,
  coherent statistical information on the Dutch
  society.
• This is more than just to publish own results of
  the several statistical inquiries portal
  function

4
Dissemination policy (1)

• Coherent information harmonisation and
  standardization of concepts definitions
  models (equations, accounting systems)
  comparability (time, space, between
  concepts)
• Undisputed information consistent output,
  quality (reliable, timely, complete)
  transparency (communication with users)

5
Metadata model (1) (output oriented)

• Conceptual metadata
• Process metadata
• (Quality metadata)

6
Metadata model (2)
Trade, 2001

Turnover 1000 euro Costs Profit
Size class 1 9
Size class 2 12
total 21 What should be the metadata of 21? What should be the metadata of 21?
7
Metadata model (2)
Conceptual metadata
Trade, 2001

Turnover 1000 euro Costs Profit
Size class 1 9 8 5 3
Size class 2 12 15 10 5
Total 21 23 What should be the metadata of 21? What should be the metadata of 21?
Process metadata
8
Dissemination policy (2)

• How to present inconsistent information?
• Is it understandable for external users?
• Is it in theory possible to publish consistent
  information?
• Could/should we integrate every figure?

9
Metadata model (3)

Class. servers
10
Purpose Statistical Co-ordination (1)
Variable X
Variable close to X
0-10 A
11-20 B
gt20 C



0-15
16-30
gt30
D
E
F
11
Purpose Statistical Co-ordination (2)
Variable X
Variable close to X




0-10 A
11-20 B
gt20 C
0-15
16-30
gt30
D
E
F
12
Purpose Statistical Co-ordination (3)
Standard classification
Variable X
Variable close to X
0-10 A
11-20 B
gt20 C



A G
B H
C I
J D
K E
L F
0-15
16-30
gt30
D
E
F
optional
13
Why a central metadata server?

• Part of the new StatLine datamodel but also
  available for input and throughput processes
• The aims
• Transparency
• Harmonisation, standardization
• Consistency
• Coherence
• Efficiency
• Bureau of Standards (dissemination of metadata
  standards)

14
Metadata server process

1. Convert existing metadata from present StatLine
   into new model
2. New deliveries to StatLine can only point to
   existing metadata
3. New metadata can only be added via special group
   (SC)
4. Existing metadata will be screened (textual
   consistency) and combined by SC
5. Existing metadata can be set to expired by SC.

15
Statistics Netherlands is going to

• have a transparent StatLine and statistical
  system
• have a textual consistent StatLine
• maximize the consistency between figures
• present a clear view on dissemination policy
• be recognized by external organizations as
  Bureau of Standards for statistical concepts
• Thank you