=20
Statistical metadata include content oriented =
;and technological metadata. Both groups are nee=
ded for design, implementation and running of STs.
Content oriented metadata are presented in the fol= lowing groups:
1. Metadata on statistical concepts and models
This group describes models of statistical classifications and statistic= al variables.
2. Metadata on statistical methods
This group describes imputation methods of missing values/data, grossing= up to the whole population of observed units, methods for time series conv= ersions, seasonal adjustment, methods for expert estimates, analytical, mat= hematical and statistical methods of evaluation, etc.
3. Metadata on processing procedures<= /p>
This group describes processing procedures for individual stages of STs = life cycle. For example the data collection, respondent burden measurement,= preparation of statistical questionnaires, data validation, quality assess= ment, and aggregation, preparation of statistical tables, etc.
4. Metadata on use of statistical information
This group describes user satisfaction, use of statistical information b= y respondents, analysis of users' requirements for information, FAQ, users'= opinions, use of web pages, etc.
5. Metadata on SBP assessment and evaluation=
This group provides source materials for assessment and evaluation of ef= fectiveness in individual phases of SBP and source material= s for financial controlling in the CZSO.
The table below shows a placement of the above groups of metadata in the=
SMS architecture.

Classifications 
Variables 
Tasks 
Quality 
Dissemination 
Users 
Respondents 
Time series 
Data fund 
Statistical models 
x 
x 







Statistical methods 


x 






Processing procedures 


x 
x 
x 
x 

x 
x 
Use of statistical information 




x 
x 
x 
x 

Assessment and further development 
x 
x 
x 
x 
x 
x 
x 
x 
x 
The aim of the SMS is to support all phases of SBP as set out in point 2=
.3. The SMS is an endtoend system.
The subsystems store, update and =
use metadata items in individual phases of SBP, as shown in the following t=
able.
SBP phases 
Definition of statistical task 
Processing preparation  Data collection 
Primary processing 
Data analysis 
Dissemination 
Classifications 
x 


x 

x 
Variables 
x 




x 
Tasks 
x 
x 
x 
x 
x 

Quality 
x 

x 
x 
x 

Dissemination 
x 




x 
Users 
x 




x 
Respondents 
x 
x 
x 


x 
Time series 
x 



x 
x 
Data fund 
x 
x 

x 


Due to progress of work on the SMS (s=
ee point 4.1) we take an example of detailed information on metadata kept i=
n subsystems CLASS, VAR and TASKS that are being developed.
Subsystem CLASS is based on the Neuchât= el model of statistical classifications. It allows creation, storage, updat= e and use of statistical classifications, which are necessary for data proc= essing. There is basic metainformation kept on each classification incl. it= s history, e.g. the title and coordinator of classification, validity and c= ontents of classification/codelist in language versions (CZ, EN) etc.These= characteristics on codelists are stored in the SMS database.
Subsystem VAR is based on a unique model for = description of statistical variables at micro data and macro data level. Th= e model was developed in the CZSO using experience of Work Session on Metad= ata UNECE. Metadata are aimed at description of the contents of statistical= data. The most important metainformation is the statistical concept, stati= stical function, title, definition, and unit of measure and subjectmatter = breakdown. Also metainformation on the coordinator, subjectmatter area, va= lidity, etc., is kept on statistical variable.
. Following characteristics of a variable are kept in SMS database:
=E2=80=A2 identifier,
=E2=80=A2 s=
tructural description,
=E2=80=A2 full and short name=
,
=E2=80=A2 definition,
=E2=80=A2 &=
nbsp;validity (fromto),
=E2=80=A2 set of compulsory=
and voluntary attributes,
=E2=80=A2 names for prese=
ntation,
=E2=80=A2 remarks
Subsystem TASKS contains metainformation on f=
unctional and technological specifications of STs. Mainly the following met=
ainformation is kept:
=E2=80=A2 basic characteristic=
s of a task,
=E2=80=A2 statistical questionnai=
re content a structure definition,
=E2=80=A2 i=
nput data sets,
=E2=80=A2 annex to the decree on ann=
ual programme of statistical surveys (list of surveys with response duty),<=
br />=E2=80=A2 data item validation rules, autocorrectio=
n rules, transformation rules, derivation rules
=E2=80=A2 &=
nbsp; definition of statistical samples,
=E2=80=A2 s=
pecifications of imputation methods,
=E2=80=A2 quali=
ty requirements,
=E2=80=A2 aggregations rules,
=
=E2=80=A2 estimates procedures,
=E2=80=A2 =
; specification of users,
=E2=80=A2 timetable=
s for preparation of a task, for user tests and for statistical production,=
=E2=80=A2 applied codelists,
=E2=80=A2 &=
nbsp; legislation base for a task,
=E2=80=A2 d=
ata flow and organization of the collection and processing,
=E2=80=A2&=
nbsp; documentation (user and technological).
The division of life cycle of the statistical task (part 2.3) will also = be used for cost controlling purposes. Metainformation on the history of a = statistical task, especially timetable of processing, will be used when co= nsidering labour intensity of individual phases/activities of the process.<= /p>
Information on statistical data quality obtained in the history of proce= ssing will be used for quality of work assessment of the CZSO's departments= responsible for design and implementation of statistical tasks. This infor= mation can also be used to assess the quality of work of the entire Office.= The method applied is European Foundation for Quality Management (EFQM).
Subsystem TASKS is designed to allow specification of nonstatistical ta= sks such as controlling, other administrative subsystem or development task= s.