|
The data stored in CountrySTAT databases are organized in two major groups:
CountrySTAT Core which contains national data shared with FAOSTAT database and
the CountrySTAT Sub-national, composed of data owned by various national
authorities. This arrangement of CountrySTAT Core and Subnational provide
end-users with the possibility of “navigating” throughout data starting from
either geographic or thematic paths.
| Statistical core framework
|
|
(click to enlarge)
|
|
Following the approach used by the FAOSTAT2 project, CountrySTAT categorizes
statistical information into:
-
Core data
on items with national and international (FAO) relevance.
-
Related data
on items with subnational relevance. They may be aggregated into a major group
of items.
-
Linked data
on items outside the relevant scope and immediate responsibility of national
statistical bureaus.
-
Linked metadata on national and international codes and
classification systems.
The CountrySTAT methodology recommends statistical offices to define and use
data matrixes with a practical level of granularity in order to reduce to a
minimum the number of personal in charge of each matrix. Such a strategy
enables full tracing and controlling of matrixes from the moment of their
structure definition to the final stage of populating with data. Once these
building blocks are in place, accurate and comparable indicators are based on
them.
Among other features, the pilot version of CountrySTAT allows statistical
offices to enrich metadata storage such as footnotes at cell level, matrix
level, or at any matrix variable level. In addition, CountrySTAT has the
advantage of supporting multiple languages on both graphical user interface and
data variable labels. As variable labels may vary from one language to another,
data processing (merge/join matrixes) will strongly rely on codes instead of
labels. Thus, it is vital to apply a consistent and comprehensive coding system
(classification). CountrySTAT will provide methods and techniques to convert
national classifications into international ones, in particular FAOSTAT
classifications.
As part of its framework, CountrySTAT is committed to create protocol and
procedures for exchanging data via human-readable and platform-independent
documents written in languages such as XML. The pilot version of CountrySTAT
will export documents using GESMES/TS(Generic Statistical Message for Time
Series) format and future versions of CountrySTAT will aim at exporting
documents in SDMX – Statistic Data and Metadata Exchange (www.sdmx.org) format.
|
