Page 149 - Contributed Paper Session (CPS) - Volume 7

CPS2049 Mohammed Al Rifai et al.
 Databases that can be integrated with sample surveys, and used in
the imputation of missing values for sample surveys.
 Construct new sampling frames or update the available frames, for
the design and selection survey samples.
 Regular time series for the proposed statistical variables.

3. Result
The implementation of the concept of quality is not limited to censuses
and sample surveys, but also is a comprehensive concept with standards and
conditions that apply to all official statistical databases, including
administrative data. In fact, the process of achieving quality in administrative
data may be more complicated and difficult than for sample surveys, because
administrative databases are not usually constructed for the purpose of
providing statistical indicators, but only to serve the administrative objective
of the data‐producing institutions.
Harmonization of administrative data between official statistics agencies’
requirements and the requirements of the producing institutions is a big
challenge. This challenge varies for different countries. Many countries have
been working on harmonizing their administrative databases with their official
statistics outputs for some time and have successful models, while other
countries are still working through the challenges of conceptual
harmonization, to develop and improve their administrative databases. The
following are common challenges face the administrative data:
Relevance: Relevance is the degree to which administrative data meet
the official statistics needs. A comprehensive evaluation could be
implemented to see whether all statistics that are needed are produced
and the extent to which concepts (definitions, classifications etc.) reflect
user needs. It is important to note that, while the administrative data will
be relevant for the users of the data‐producing entity, the data may not
be relevant for official statistics indicators.
Inclusion and Coverage: In some situations, administrative data bases
may exclude variables that fall within the requirements of official
statistics. Data‐producing institutions are focused on collecting and
producing data that serves their own purposes. This is one of the biggest
challenges for official statistical agencies’ use of administrative data.
Coverage of statistical units can also be an obstacle. Statistical agencies
require extensive coverage of the populations being measured in order
to output statistical indicators for small geographic areas and
administrative regions.
Accuracy: The accuracy of data is one of the key data quality dimensions.
It is well known that the accuracy of the statistical indicator is inversely
proportional to the amount of statistical error in its estimated value,

136 | I S I W S C 2 0 1 9