Page 150 - Contributed Paper Session (CPS) - Volume 7

CPS2049 Mohammed Al Rifai et al.
while the error is the amount of the difference between the actual and
estimated value. In sample surveys, sample error can be measured
statistically and controlled through the adoption of probability sampling
methods. Non‐sample error can be minimized by following optimal
procedures throughout the various stages of survey implementation.
For administrative data, it may be difficult to optimize the organizational
and technical procedures for constructing the databases in order to
minimize qualitative measurement error because these procedures are
controlled by the data‐producing agency rather than the statistical
agency. Thus, it may be difficult to accurately evaluate the quality of
indicators sourced from administrative data.
Consistency of methodologies, statistical concepts and
classifications: Official statistics agencies depend on the concepts,
definitions, manuals, and statistical classifications recommended by the
international statistical organizations for many categories of economic,
social and demographic data. Using the administrative data as a source
for official statistics requires consistent concepts, definitions, manuals,
classifications, and appropriate methodologies to achieve integration
between data from statistical surveys and administrative data, which
maximize the benefit from the entire database in optimum way.
Timeliness: Lack of timeliness in the delivery of the data to the national
statistical agencies is a very common problem. Some delays are
inevitable whereas others are because of inefficiencies in processes, lack
of priority given to statistical needs or insufficient resources.
The large size of administrative files can mean significant processing
costs with a subsequent impact on timeliness.
Comparability: Administrative statistics should be consistent internally,
over time and comparable between region and geographic areas. On
other hand it should allow to combine and make joint use of related data
from different sources.
Interpretability: The interpretability of statistical data reflects the
availability of supplementary information and metadata. The provision
of metadata is essential to enable public evaluation of the
appropriateness of the data for various uses. Metadata also provides
explanations of the procedures and methods of treatment that have
been implemented in the construction of the data. A major challenge for
statistical agencies that use administrative data as their source for
statistical indicators is ensuring the adequacy of metadata about the
administrative databases.

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