Update on EUROSTAT activities

1
Update on EUROSTAT activities

• A second hand experience

2

• LUCAS
• Census 2010
• SDMX

3
LUCAS data collection process
4
Sampling strategy Second phase sampling design

• Definition of sample size by strata
• Optimal size by NUTS2 and strata based on fixed
  precisions for a set of LC classes targeted by
  country
• Points selection
• LUCAS 2006 sample points included as much as
  possible (land cover/use changes can be detected)
• Maximisation of the distance between points
• Exclusion of remote points and points above 1000m

5
Land Cover nomenclature LUCAS 2009
6
Land Use nomenclature LUCAS 2009
7
Data availability

• Types of data
• Tabular microdata on the first and second phase
  sample (land cover/use on the specific point,
  LC/LU change in the specific point etc.)
• Pictures (four cardinal directions)
• Aggregated estimates (NUTS1/NUTS2 depending on LC
  classes)
• Years
• 2006
• 2008/2009 (from march 2010 on)
• Terms of use
• An agreement has to be signed between DG-ESTAT
  and users about
• Confidentiality only aggregated data can be
  disseminated

8
Data availability per country/year
9
Census
10
EU census goals

• Comparability of census data on the EU level
• Same reference year (first time 2011)
• Same topics (variables)
• Use of harmonized definitions and technical
  specifications
• Use of identical breakdowns of the topics
• Unified dissemination programme (hypercubes)
• gtCommon Baseline across countries
• Transparent quality of census results
• Quality reports
• Detailed tables on quality of the data
• Metadata

11
EU census limits

• What does the regulation not provide?
• No access to microdata
• No possibility to define geographical areas
  flexibly
• No harmonised confidentiality control
• No normative minimum quality requirements
  (quality thresholds)
• No consolidation of census results form different
  Member States
• BUT
• Member States are free to do more!

12
What data for what geographical area?

• NUTS2
• Year of arrival in the country
• Educational attainment
• Location of place of work
• Current activity status
• Occupation
• Industry
• Status in employment
• Tenure status of households
• Housing arrangements
• Type of ownership (of dwellings)
• Water supply system, Toilet facilities, Bathing
  facilities, Type of heating

13
What data for what geographical area ?
LAU 2

• Population topics
• Sex
• Age
• Legal marital status
• Country/place of birth
• Country of citizenship
• Place of usual residence one year prior to the
  census
• (Size of the) Locality
• Household status
• Type of private household
• Size of private household
• Family status
• Type of family nucleus
• Size of family nucleus

Housing topics Occupancy status of conventional
dwellings Number of occupants Useful floor space
and/or Number of rooms Density standard Dwellings
by type of building Dwellings by period of
construction Type of living quarters Location of
living quarters
14
What we can NOT do for GISCO ?

• The municipalities as smallest geographical area
  for the census data to be transmitted to Eurostat
  (LAU 2 level) are fixed. No flexibility to define
  areas freely.
• After long and detailed consultation with the
  Census experts from the Member States, the
  foreseen obligatory statistical programme
  represents a balance between the desirable and
  the feasible.
• Eurostat does not have access to census
  microdata.
• Confidentiality control is done by the NSI.

15
What we can do for GISCO ?

• Usage of common definitions, technical
  specifications and breakdowns makes census data
  better comparable at the European level.
• Intensive description and quality reporting of
  the NSI on the data sources and methodology they
  use to do the population and housing census. This
  might help to develop small area reporting
  systems.
• Key topics will be required for the LAU 2 level.
  It is likely that some of the data might also be
  available for even smaller areas in some Member
  States.
• Eurostat organizes a task force on Census Data
  Disclosure Control which aims at proposing best
  methodology and practice to protect census data
  with minimum damage to disseminated results.
• The Census Hub might be used to exchange and
  disseminate small area data from censuses.

16
Census Hub project architecture
WS
WS
WS
database
database
database
17
The Census Hub project

• The Census Hub project aims to build a new IT
  infrastructure to achieve the data exchange
  between the National Statistical Institutes
  (NSI), Eurostat and the users of census data
  using SDMX standards.
• Data sharing architecture
• Based on the agreed hyper-cubes with harmonised
  data
• Confidentiality problems handled at national
  level
• A data user browses the hub to define a dataset
  of interest via structural metadata (dimensions,
  attributes, measures, code lists, etc). Data are
  retrieved directly from the interested Member
  States systems

18
Present and ongoing activities

• Pilot project in Germany, Ireland, Italy and
  Portugal
• Guideline explaining how to implement an SDMX MSs
  architecture in the Census Hub context available

19
SDMX
20
What is SDMX

• Statistical Data and Metadata Exchange
• SDMX preferred standard for exchange and sharing
  of data and metadata in the global statistical
  community
• Sponsors include
• European Central Bank (ECB)
• Eurostat
• Organisation for Economic Co-operation and
  Development (OECD)
• United Nations Statistical Division (UNSD)

21
Benefits from SDMX standards

• Covers potentially all statistical domains
• Open to all stakeholders
• Are neutral in terms of underlying commercial
  technologies
• Demography and the Census hub already implemented

22
SDMX components
Information model for data and metadata Syntax
for automatic exchange of data and metadata
Guidelines to Harmonise Contents
IT Architectures for data exchange
IT tools to support implementation and to
disseminate SDMX data
SDMX is not just a data transmission
format Similarities with INSPIRE are substantial
23
SDMX Components Information Model

• Statistical data
• Metadata
• Structural
• Conceptual
• Quality
• Methodology
• Data exchange process

24
SDMX Information Model
Provides a way of modelling statistical data,
metadata and data exchange processes.
Dimensions (ex country, variable/topic, year)
Dataset Structure Definition DSD
Code lists
Structural Metadata
Attributes (ex unit of measure)
Describe
Metadata about an individual value, a time series
or a group of time series
Data
25
SDMX Components IT Tools

• SDMX Registry
• Tools to create data definitions and metadata
• Tools to convert and validate data and metadata
• Tools to visualise data and metadata
• Training available from Eurostat
  http//epp.eurostat.ec.europa.eu/portal/page?_page
  id2733,61942355,2733_61942368_dadportal_schema
  PORTAL

26
SDMX Registry
Repository
Graphical User Interface (GUI) for user
interaction over the Web
Structural metadata
Provision of information
DSW standalone Java GUI
CodeLists
Dataflows
ConceptSchemes
Provision agreements
DSDs
Accessible via a Web Service accepting SDMX-ML
messages