Quality indicators for the GSBPM

1
Quality Indicators for the Generic Statistical Business
Process Model (GSBPM) - For Statistics derived from
Surveys
(Version 1.0, May 2016)
This work is licensed under the Creative Commons Attribution 3.0 Unported
License. To view a copy of this license, visit
http://creativecommons.org/licenses/by/3.0/. If you re-use all or part of this work,
please attribute it to the United Nations Economic Commission for Europe
(UNECE), on behalf of the international statistical community.

2
Contents
Abbreviations .......................................................................................................................................3
Acknowledgements ..............................................................................................................................3
I. Introduction...............................................................................................................................4
II. Quality indicators for the GSBPM phases and sub-processes..............................................5
A. Specify Needs Phase ...................................................................................................5
1.1. Identify Needs......................................................................................................5
1.2. Consult and confirm needs...................................................................................6
1.3. Establish output objectives ..................................................................................6
1.4. Identify concepts..................................................................................................7
1.5. Check data availability.........................................................................................7
1.6. Prepare business case...........................................................................................7
B. Design Phase ...............................................................................................................8
2.1. Design outputs .....................................................................................................9
2.2. Design variable descriptions..............................................................................10
2.3. Design collection ...............................................................................................10
2.4. Design frame and sample...................................................................................11
2.5. Design processing and analysis .........................................................................12
2.6. Design production systems and workflow.........................................................13
C. Build Phase................................................................................................................13
3.1. Build collection instrument................................................................................14
3.2. Build or enhance process components...............................................................15
3.3. Build or enhance dissemination components.....................................................18
3.4. Configure workflows .........................................................................................18
3.5. Test production system ......................................................................................18
3.6. Test statistical business process.........................................................................19
3.7. Finalise production systems...............................................................................20
D. Collect Phase.............................................................................................................20
4.1. Create frame and select sample..........................................................................20
4.2. Set up collection.................................................................................................22
4.3. Run collection....................................................................................................23
4.4. Finalise collection ..............................................................................................24
E. Process Phase ............................................................................................................25
5.1. Integrate data......................................................................................................25
5.2. Classify and code ...............................................................................................27
5.3. Review and validate...........................................................................................27
5.4. Edit and impute..................................................................................................28
5.5. Derive new variables and units..........................................................................30
5.6. Calculate weights...............................................................................................31

3
5.7. Calculate aggregates ..........................................................................................31
5.8. Finalise data files ...............................................................................................32
F. Analyse Phase ...............................................................................................................33
6.1. Prepare draft outputs..........................................................................................33
6.2. Validate outputs .................................................................................................33
6.3. Interpret and explain outputs .............................................................................34
6.4. Apply disclosure control....................................................................................34
6.5. Finalise outputs..................................................................................................35
G. Disseminate Phase.....................................................................................................36
7.1. Update output systems .......................................................................................36
7.2. Produce dissemination products ........................................................................36
7.3. Manage release of dissemination products ........................................................37
7.4. Promote dissemination products........................................................................39
7.5. Manage user support..........................................................................................40
H. Evaluate Phase...........................................................................................................41
8.1. Gather evaluation inputs ....................................................................................41
8.2. Conduct evaluation ............................................................................................42
8.3. Agree an action plan ..........................................................................................43
III. Over-arching processes ..........................................................................................................44
Annex 1 ...............................................................................................................................................46
Abbreviations
CoP - Code of Practice
ES - European Statistics
ESMS - Euro SDMX Metadata Structure
ESS - European Statistical System
ESQRS - Standard for Quality Reports Structure
GSBPM - Generic Statistical Process Model
ISO - , International Organization for Standardization
NQAF - National Quality Assurance Framework
QPI – Quality and Performance Indicators
SDMX – Statistical Data and Metadata eXchange
SIMS - Single Integrated Metadata Structure
Acknowledgements
Contributors to this work include: Alice Born, Kaveri Mechanda and Laurie Reedman,
(Statistics Canada, Canada); Giovanna Brancato, Marina Signore and Giorgia Simeoni
(ISTAT, Italy); Bilge Çorumlu, Nilgün Dorsan, Gülhan Eminkahyagil, Nezihat Keret, and
Deniz Özkan (TurkStat, Turkey); and August Götzfried and Zsuzsanna Kovacs (Eurostat,
European Union)

4
I. Introduction
1. Quality concerns organisations, processes and products. This document outlines a set
of quality indicators that have been developed for the production of statistics from survey
sources, with reference to the different stages of the Generic Statistical Process Model
(GSBPM) Version 5.0
2. The main goal of quality management within the statistical business process is to
understand and manage the quality of the statistical products. There is general agreement
among statistical organisations that quality should be defined according to the ISO 9000-2005
standard: "The degree to which a set of inherent characteristics fulfils requirements".1
Thus,
product quality is a complex and multi-faceted concept, usually defined in terms of several
quality dimensions. The dimensions of quality that are considered most important depend on
user perspectives, needs and priorities, which vary between processes and across groups of
users.2
3. A fundamental role in quality management is played by a set of quality indicators that
should be implemented within the sub-processes to prevent and monitor errors. The first
version quality indicators for surveys are presented here, and complement the quality
management process of the GSBPM.
4. Quality indicators are mapped for each phase (Phases 1 to 8) and sub-process of the
GSBPM. Within version 5.0 of the GSBPM, the quality management over-arching process
refers mainly to product and process quality. The following guiding principles in mapping the
quality indicator to the GSBPM were used:
 Indicators were limited to direct surveys with the intention to extend the work to
administrative and Big Data in the future;
 Develop generic indicators to reflect the nature of the GSBPM as a reference model;
 Be consistent with existing quality assurance frameworks when selecting the quality
indicators and determining their related quality dimension;
 No formulas are used to express the indicators, only descriptions or explanations;
 Quantitative indicators were used whenever possible;
 Qualitative indicators in the form of yes/no or large/medium/low were proposed when
appropriate;
 Map indicators to the phase they measure even if they might be calculated at a later
stage; and
 Allow for a certain degree of redundancy by mentioning the same indicators in
different phases or sub-processes.
5. Quality indicators were determined by examining practices within national statistical
agencies, United Nations’ Statistical Commission National Quality Assurance Framework,
European Statistics (ES) Code of Practice, Euro SDMX Metadata Structure (ESMS), national
and Eurostat quality assurance frameworks, European Statistical System (ESS) Standard for
Quality Reports Structure (ESQRS) and Single Integrated Metadata Structure (SIMS).
6. While mapping the quality indicators to the GSBPM, the related quality dimensions
were determined for each indicator. A suitable global framework is the National Quality
Assurance Framework (NQAF) developed by a global expert group under the United Nations
Statistical Commission. Each of the quality indicators is mapped to one of the dimensions of
NQAF. In addition, the quality indicators are mapped to the quality dimensions of the ES
Code of Practice Principles.
1
ISO 9000:2005, Quality management systems – Fundamentals and vocabulary, International Organization for
Standardization.
2
Generic Statistical Business Process Model (GSBPM) Version 5, UNECE, December, 2013.

5
7. The quality indicators are presented after each sub-process in a table format – Column
1: Quality dimension; Column 2: Quality indicator; and Column 3: Notes. The NQAF quality
dimension is presented in Column 1 in the order of the (nineteen) dimensions of the NQAF.
The quality dimension of the ES CoP is noted in column 3 when this differs from the NQAF
dimension.
8. The current overarching quality management layer illustrates the importance of the
benchmarking and peer review approaches to evaluation. In addition to the quality indicators
for each phase and sub-process of the GSBPM, a set of quality indicators are attached to the
quality management overarching process to address this aspect of overall quality
management. These indicators are related to the availability of a quality policy, quality
assurance plan, monitoring procedures and organizational structure for managing quality.
II. Quality indicators for the GSBPM phases and sub-processes
9. This section considers each phase in turn, identifying the various sub-processes
within that phase, and describing their contents.
A. Specify Needs Phase
10. This phase is triggered when a need for new statistics is identified, or feedback about
current statistics initiates a review. It includes all activities associated with engaging
customers to identify their detailed statistical needs, proposing high level solution options and
preparing business cases to meet these needs.
11. In this phase, the organisation:
 identifies the need for the statistics;
 confirms, in more detail, the statistical needs of the stakeholders;
 establishes the high level objectives of the statistical outputs;
 identifies the relevant concepts and variables for which data are required;
 checks the extent to which current data sources can meet these needs;
 prepares the business case to get approval to produce the statistics.
12. This phase is broken down into six sub-processes. These are generally sequential,
from left to right, but can also occur in parallel, and can be iterative. The sub-processes are:
1.1. Identify Needs
13. This sub-process includes the initial investigation and identification of what statistics
are needed and what is needed of the statistics. It may be triggered by a new information
request, an environmental change such as a reduced budget. Action plans from evaluations of
previous iterations of the process, or from other processes, might provide an input to this sub-
process. It also includes consideration of practice amongst other (national and international)

6
statistical organisations producing similar data, and in particular the methods used by those
organisations. It may involve consideration of specific needs of different user communities,
such as the disabled or different ethnic groups.
Quality
Dimension
Indicator Notes
Relevance To what extent have stakeholders been
identified and included in discussions
about statistical needs?
To what extent has relevant supporting
documentation been gathered?
1.2. Consult and confirm needs
14. This sub-process focuses on consulting with the stakeholders and confirming in detail
the needs for the statistics. A good understanding of user needs is required so that the
statistical organisation knows not only what it is expected to deliver, but also when, how,
and, perhaps most importantly, why. For second and subsequent iterations of this phase, the
main focus will be on determining whether previously identified needs have changed. This
detailed understanding of user needs is the critical part of this sub-process.
Quality
Dimension
Indicator Notes
Relevance To what extent have stakeholders
confirmed the detailed statistical needs
(what, when, how and why) as
documented by the NSO?
Could be a two part indicator;
proportion of stakeholders who
have confirmed, and proportion of
statistical needs confirmed.
1.3. Establish output objectives
15. This sub-process identifies the statistical outputs that are required to meet the user
needs identified in sub-process 1.2 (Consult and confirm needs). It includes agreeing the
suitability of the proposed outputs and their quality measures with users. Legal frameworks
(e.g. relating to confidentiality), and available resources are likely to be constraints when
establishing output objectives.
Quality
Dimension
Indicator Notes
Statistical
confidentiality
and security
To what extent have legal constraints
regarding statistical outputs been
considered, for example but not limited
to ensuring confidentiality of data and
preventing the disclosure of sensitive
information?
Relevance To what extent have all statistical needs
been addressed by the proposed
outputs?

7
Accuracy and
reliability
To what extent are the proposed outputs
and their quality measures suitable to
user needs?
1.4. Identify concepts
16. This sub-process clarifies the required concepts to be measured by the business
process from the point of view of the user. At this stage the concepts identified may not align
with existing statistical standards. This alignment, and the choice or definition of the
statistical concepts and variables to be used, takes place in sub-process 2.2.
Quality
Dimension
Indicator Notes
Relevance Compliance rate of concepts and
definitions of variables with existing
standards
1.5. Check data availability
17. This sub-process checks whether current data sources could meet user requirements,
and the conditions under which they would be available, including any restrictions on their
use. An assessment of possible alternatives would normally include research into potential
administrative or other non-statistical data sources, to determine whether they would be
suitable for use for statistical purposes. When existing sources have been assessed, a strategy
for filling any remaining gaps in the data requirement is prepared. This sub-process also
includes a more general assessment of the legal framework in which data would be collected
and used, and may therefore identify proposals for changes to existing legislation or the
introduction of a new legal framework.
Quality
Dimension
Indicator Notes
Statistical
confidentiality
and security
To what extent have legal constraints
regarding data collection, acquisition
and use been assessed and any
necessary changes been proposed?
Relevance To what extent do current data sources
meet user requirements, taking into
consideration the conditions under
which they would be available and any
restrictions on their use?
If current data sources do not fully meet
user requirements, to what extent has a
strategy been proposed to fully meet
user requirements?
1.6. Prepare business case
18. This sub-process documents the findings of the other sub-processes in this phase in
the form of a business case to get approval to implement the new or modified statistical
business process. Such a business case would need to conform to the requirements of the
approval body, but would typically include elements such as:

8
 A description of the "As-Is" business process (if it already exists), with information on
how the current statistics are produced, highlighting any inefficiencies and issues to be
addressed;
 The proposed "To-Be" solution, detailing how the statistical business process will be
developed to produce the new or revised statistics;
 An assessment of costs and benefits, as well as any external constraints.
Quality
Dimension
Indicator Notes
Adequacy of
resources
To what extent have resource
requirements for the proposed outputs
and their quality measures been
considered?
Relevance To what extent does the business case
conform to the requirements of the
approval body?
Relevance To what extent does the business case
reflect the findings, recommendations
and proposals from steps 1.2 to 1.5?
B. Design Phase
19. This phase describes the development and design activities, and any associated
practical research work needed to define the statistical outputs, concepts, methodologies,
collection instruments3
and operational processes. It includes all the design elements needed
to define or refine the statistical products or services identified in the business case. This
phase specifies all relevant metadata, ready for use later in the statistical business process, as
well as quality assurance procedures. For statistical outputs produced on a regular basis, this
phase usually occurs for the first iteration, and whenever improvement actions are identified
in the Evaluate phase of a previous iteration.
20. Design activities make substantial use of international and national standards, in order
to reduce the length and cost of the design process, and enhance to comparability and
usability of outputs. Organisations are also encouraged to reuse or adapt design elements
from existing processes. Additionally, outputs of design processes may form the basis for
future standards at the organisation, national or international levels.
21. This phase is broken down into six sub-processes, which are generally sequential,
from left to right, but can also occur in parallel, and can be iterative. These sub-processes are:
3
For GSBPM purposes, collection instruments are defined broadly to include any tool or routine to gather or
extract data and metadata, from paper questionnaires to web-scraping tools. In GSIM version 1.1, collection
instruments are “exchange channels” used for incoming information.

9
2.1. Design outputs
22. This sub-process contains the detailed design of the statistical outputs, products and
services to be produced, including the related development work and preparation of the
systems and tools used in the "Disseminate" phase. Disclosure control methods, as well as
processes governing access to any confidential outputs are also designed here. Outputs should
be designed to follow existing standards wherever possible, so inputs to this process may
include metadata from similar or previous collections, international standards, and
information about practices in other statistical organisations from sub-process 1.1 (Identify
needs).
Quality
Dimension
Indicator Notes
Statistical
confidentiality
and security
Have the confidentiality rules and
micro data access procedures been
designed?
yes/no indicator
Relevance Percentage of/Extent to which
outputs fulfill users’ needs (and/or
priority needs)
Link to “identify needs” (sub-
process 1.1) and to the “evaluate”
phase
Relevance Percentage of/ Extent to which
outputs changed as a result of
improvement actions or as a result of
user satisfaction surveys/analyses (
for outputs produced on a regular
basis)
Link to “identify needs” (sub-
process 1.1) and to the “evaluate”
phase
Relevance Planned data completeness rate:
extent to which the planned outputs
will satisfy requirements (e.g. from
Regulations or other agreements with
users)
Could be calculated as the ratio of
the number of data cells planned to
the number of data cells required
ESS QPI - R1. (Planned) Data
completeness rate
Coherence
and
comparability
Expected length of comparable time
series.
Breaks in statistical time series may
occur when there is a change in the
definition of the parameter to be
estimated (e.g. variable or
population) or the methodology used
for the estimation. Sometimes a
break can be prevented, e.g. by
linking.
The length of comparable time
series is applicable:
- to all statistical processes
producing time-series;
- to users and producers, with
different level of details given.
ESS QPI - CC2. Length of
comparable time series
Accuracy Data revisions are planned (Yes/No) ESS QPI - A6. Data revision -
average size.

10
2.2. Design variable descriptions
23. This sub-process defines the statistical variables to be collected via the collection
instrument, as well as any other variables that will be derived from them in sub-process 5.5
(Derive new variables and units), and any statistical classifications that will be used. It is
expected that existing national and international standards will be followed wherever
possible. This sub-process may need to run in parallel with sub-process 2.3 (Design
collection), as the definition of the variables to be collected, and the choice of collection
instrument may be inter-dependent to some degree. Preparation of metadata descriptions of
collected and derived variables and classifications is a necessary precondition for subsequent
phases.
Quality Dimension Indicator Notes
Cost effectiveness Percentage of/ Extent to which
concepts, definitions and
classifications associated to
(key) variables and populations,
are re-used from other similar
surveys
Managing metadata Percentage of/Extent to which
concepts, definitions and
classifications associated to
(key) variables and populations
follow international or national
standards
See also5.5 for derived
variables
Corresponds to accessibility
and clarity principle in the
ES Code of Practice
Managing metadata Percentage of/Extent to which
new concepts, definitions and
classifications are introduced
(provide motivation for it)
ES Code of Practice
Managing metadata Percentage of metadata
adequately archived (easily
retrievable; properly labelled;
retention period indicated)
ES Code of Practice
Managing metadata Percentage of / extent to which
collected and derived variables
and classifications have
metadata descriptions.
ES Code of Practice
2.3. Design collection
24. This sub-process determines the most appropriate collection method(s) and
instrument(s). The actual activities in this sub-process will vary according to the type of
collection instruments required, which can include computer assisted interviewing, paper
questionnaires, administrative data interfaces and data integration techniques. This sub-
process includes the design of collection instruments, questions and response templates (in
conjunction with the variables and statistical classifications designed in sub-process 2.2
(Design variable descriptions)). It also includes the design of any formal agreements relating
to data supply, such as memoranda of understanding, and confirmation of the legal basis for
the data collection. This sub-process is enabled by tools such as question libraries (to
facilitate the reuse of questions and related attributes), questionnaire tools (to enable the
quick and easy compilation of questions into formats suitable for cognitive testing) and
agreement templates (to help standardise terms and conditions). This sub-process also
includes the design of process-specific provider management systems.

11
Soundness of
implementation
Is the process re-using known
methods and collection systems,
e.g. according to
guidelines/recommendations?
yes/no indicator
Corresponds to the
appropriate statistical
procedures principle in the
ES Code of Practice
Soundness of
implementation
How well does the collection
method suit the nature and volume
of the information to be gathered?
fully/partly/no indicator
Corresponds to the
ES Code of Practice
Soundness of
implementation
When has the data collection
technique last been
revised/improved?
For outputs produced on a
regular basis.
Corresponds to the
ES Code of Practice
Soundness of
implementation
Appropriateness of questionnaire to
the pre-specified standards.
Corresponds to the
ES Code of Practice
Managing respondent
burden
Percentage of questions used to
collect information which will not
be published (and motivation).
Managing respondent
burden
Indirect evaluation of response
burden: number of questions on the
questionnaire
To be evaluated taking into
account the complexity of
each question, the
questionnaire paths and the
expected fraction of the
sample/population that
should fill in each path.
Managing respondent
burden
Trend in respondent burden with
respect to the previous iteration
For outputs produced on a
regular basis.
Managing respondent
burden
Is there a communication plan
encouraging response by informing
potential respondents about the
survey and the importance of their
contribution?
Managing respondent
burden
The extent to which the respondent
can choose among different data
collection modes
2.4. Design frame and sample
25. This sub-process only applies to processes which involve data collection based on
sampling, such as through statistical surveys. It identifies and specifies the population of
interest, defines a sampling frame (and, where necessary, the register from which it is
derived), and determines the most appropriate sampling criteria and methodology (which
could include complete enumeration). Common sources for a sampling frame are
administrative and statistical registers, censuses and information from other sample surveys.
This sub-process describes how these sources can be combined if needed. Analysis of

12
whether the frame covers the target population should be performed. A sampling plan should
be made: The actual sample is created in sub-process 4.1 (Create frame and select sample),
using the methodology, specified in this sub-process.
Quality
Dimension
Indicator Notes
Methodological
soundness
Extent to which the survey
population matches the target
population
See also phase 4 “collect”
Methodological
soundness
Timeliness of the frame: how
recently was the frame last updated?
Methodological
soundness
Impact of coverage errors: assess the
likely impact of coverage error on
key estimates.
ESS QPI - A2. Over-coverage - rate
Methodological
soundness
Key indicators for sample design
(e.g. estimated size,
expected/planned sampling errors for
key variables, domains, costs,…)
ESS QPI - A1. Sampling error -
indicators.......
Methodological
soundness
Feasibility of estimation (e.g. a
complex sample design might force
the use of bootstrap variance
estimation while a simpler design
might not be as efficient but the
design based variance might be more
desirable)
2.5. Design processing and analysis
26. This sub-process designs the statistical processing methodology to be applied during
the "Process" and "Analyse" phases. This can include specification of routines for coding,
editing, imputing, estimating, integrating, validating and finalizing data sets.
Quality
Dimension
Indicator Notes
Cost
effectiveness
To what extent is the process
planning to re-use systems for
coding, E&I, data integration,
weighting, estimation
Soundness of
implementation
To what extent is the business
process using standard or well-
known methods for subsequent
phases (e.g. coding, E&I, data
integration, weighting, estimation,
revision,…), in a transparent way?
See also phase 5 and 6
yes/partly/no indicator
Corresponds to the appropriate
statistical procedures principle in the
ES Code of Practice
Soundness of
implementation
When have the methodologies for
subsequent phases (e.g. coding,
See also phase 5 and 6
for outputs produced on a regular

13
E&I, data integration, weighting,
estimation,…) last been assessed?
basis
ES Code of Practice
2.6. Design production systems and workflow
27. This sub-process determines the workflow from data collection to dissemination,
taking an overview of all the processes required within the whole statistical production
process, and ensuring that they fit together efficiently with no gaps or redundancies. Various
systems and databases are needed throughout the process. A general principle is to reuse
processes and technology across many statistical business processes, so existing production
solutions (e.g. services, systems and databases) should be examined first, to determine
whether they are fit for purpose for this specific process, then, if any gaps are identified, new
solutions should be designed. This sub-process also considers how staff will interact with
systems, and who will be responsible for what and when.
Soundness of
implementation
Percentage of identified and documented
GSBPM processes (with sub-processes)
with their flows
Corresponds to the
procedures principle in
the ES Code of Practice
Cost effectiveness Percentage of/Extent to which corporate
solutions (e.g. tools, processes,
technologies) are reused in subsequent
phases and sub-processes
Cost effectiveness Percentage of/Extent to which
responsibilities for subsequent phases and
sub-processes have been set
Cost effectiveness Estimated cost for producing and
disseminate designed outputs/Key
Performance Indicators (KPIs)
Accuracy and
reliability
Percentage of/ Extent to which quality
indicators are planned to be calculated for
subsequent sub-processes of GSBPM
Accuracy and
reliability
Amount/percentage of quality indicators
used as Key Performance Indicators
Timeliness and
Punctuality
Planned time frame for subsequent phases
and sub-processes
ESS QPI- TP2. Time lag
- final results
C. Build Phase

14
28. This phase builds and tests the production solution to the point where it is ready for
use in the "live" environment. The outputs of the "Design" phase direct the selection of
reusable processes, instruments, information, and services that are assembled and configured
in this phase to create the complete operational environment to run the process. New services
are built by exception, created in response to gaps in the existing catalogue of services
sourced from within the organisation and externally. These new services are constructed to be
broadly reusable within the statistical production architecture.
29. For statistical outputs produced on a regular basis, this phase usually occurs for the
first iteration, and also following a review or a change in methodology or technology, rather
than for every iteration.
30. It is broken down into seven sub-processes, which are generally sequential, from left
to right, but can also occur in parallel, and can be iterative. These sub-processes are:
3.1. Build collection instrument
31. This sub-process describes the activities to build the collection instruments to be used
during the "Collect" phase. The collection instrument is generated or built based on the
design specifications created during the "Design" phase. A collection may use one or more
modes to receive the data, e.g. personal or telephone interviews; paper, electronic or web
questionnaires; SDMX hubs. Collection instruments may also be data extraction routines
used to gather data from existing statistical or administrative data sets. This sub-process also
includes preparing and testing the contents and functioning of that instrument (e.g. testing the
questions in a questionnaire). It is recommended to consider the direct connection of
collection instruments to the statistical metadata system, so that metadata can be more easily
captured in the collection phase. Connection of metadata and data at the point of capture can
save work in later phases. Capturing the metrics of data collection (paradata) is also an
important consideration in this sub-process.
Quality
Dimension
Indicator Notes
Soundness of
implementation
Has the questionnaire been tested
using appropriate methods (e.g.
questionnaire pre-test, pilot in real
situation, in depth - interviews,
focus groups, interviewer
support,…)?
ES Code of Practice
Soundness of
implementation
Have the test results been taken into
account in the process of
implementing the final
questionnaire, and documented in a
report?
ES Code of Practice
Soundness of
implementation
Has the data collection
tool/instrument (electronic
questionnaire, acquisition web site,
SDMX hub) been tested and how?
This indicator refers to the tests of
the IT instruments used for data
collection (e.g. functionality test,
stress test…)
ES Code of Practice
Soundness of
implementation
To what extent have the test results
been taken into account in the
process of implementing the final
ES Code of Practice

15
data collection tools
Managing
respondent
burden
Estimated reporting burden (e.g. the
time needed to: obtain internal or
external expertise; retrieve the
required information; handle
sensitive information; and answer
the questionnaire.)
Managing
respondent
burden
Estimated response time (i.e. the
interview length)
Can be a proxy indicator of
respondent burden
Managing
respondent
burden
Percentage of questions used to
collect information which will not
be published (and motivation).
See also 2.3
Managing
respondent
burden
Trend in respondent burden with
respect to the previous iteration (for
outputs produced on a regular basis
See also 2.3
Accuracy and
reliability
If mixed or multiple data collection
modes are adopted, has the mode
effect on data quality been tested?
Accuracy and
reliability
Have the test results been taken into
account in the process of
implementing the final data
collection modes?
Accuracy and
reliability
Extent to which paradata can be
captured at the data collection
stage?
The collection instrument(s) should
allow for capturing paradata to be
used for quality assessment
Accessibility
and clarity
Extent to which metadata can be
captured at the data collection stage
and stored in metadata management
systems?
The collection instrument(s) should
allow for capturing metadata at an
early stage
Managing
metadata
Do collection instruments capture
what is needed to create variables
agreed upon in design phase?
See also 2.2
Yes/No indicators; There could be
one for each variable and
classification
Corresponds to the accessibility and
clarity principle in the ES Code of
Practice
Managing
metadata
Do collection instruments allow for
coding to the lowest level of the
classifications agreed upon in
design phase?
See also 2.2
Yes/No indicators; There could be
one for each variable and
classification See 2.2
Corresponds to the accessibility and
clarity principle in the ES Code of
Practice
3.2. Build or enhance process components
32. This sub-process describes the activities to build new and enhance existing
components and services needed for the “Process” and “Analyse” phases, as designed in the
"Design" phase. Services may include dashboard functions and features, information

16
services, transformation functions, workflow frameworks, provider and metadata
management services.
Quality
Dimension
Indicator Notes
Soundness of
implementation
What proportion of functions in the statistical
process are built using corporately supported
software tools, components or services?
See also 2.5
Yes/No indicator.
Corresponds to the
ES Code of Practice
Soundness of
implementation
Have Enterprise Architecture best practices for
software development been followed?
See also 2.5
Yes/No indicator.
Corresponds to the
ES Code of Practice
Soundness of
implementation
Has testing been done throughout the Building
process?
See also 2.5
Yes/No indicator.
Corresponds to the
ES Code of Practice
Soundness of
implementation
Have corporate requirements for dashboards
and information services been incorporated?
See also 2.5
Yes/No indicator.
Corresponds to the
ES Code of Practice
Soundness of
implementation
Was the testing strategy designed when the
process and its components were designed?4
Corresponds to the
ES Code of Practice
Soundness of
implementation
Was additional testing done by someone other
than the person(s) who did the programming?
Yes/No indicator on
testing if software and IT
tools are working properly
and not affecting
quality/introducing errors.
Corresponds to the
ES Code of Practice
Soundness of
implementation
Were the different types of testing designed, Yes/No indicator on
4
Definition from the LEG on Quality (2001): Efficiency - produces the desired outcomes cost efficiently;
Effectiveness - successful in delivering the desired outcomes; Robustness - delivers results against challenging
demands; Flexibility -readily adaptable to changing needs and demands; Transparency - open, visible and
easily understood and Integration - complementary and consistent, both with other processes, and with meeting
business needs

17
executed, documented and signed-off:
Functional testing; volume testing; stress
testing; (end-to-end testing and user testing to
be done in 3.5 Test production system)
and not affecting
Corresponds to the
ES Code of Practice
Soundness of
implementation
Was testing done specifically to ensure that the
software produces the correct results?
Yes/No indicator on
and not affecting
Corresponds to the
ES Code of Practice
This could be either a
parallel run, or in the case
where a parallel run is
impossible, testing against
known results, for example
from a prototype or
simulations of theoretical
results.
Soundness of
implementation
Extent to which process components that have
complete documentation, support staff, and
user training, all available at the same time that
the software is put into use.
Yes/No indicator on
and not affecting
Corresponds to the
ES Code of Practice
Accuracy and
reliability
Has the quality of the data after the test of the
coding procedure been assessed (e.g. quality
indicators such as “recall rate” have been
calculated)?
The recall rate is calculated as the ratio
between the number of values automatically
coded and the total number of values submitted
to coding.
This is an indicator of the
quality of the data
obtained by the coding
procedure.
Indicator of the efficacy of
the automated coding
procedure
Accuracy and
reliability
Have the assessment results been taken into
account in the implementation of the final
procedure?
Accuracy and
reliability
Has the output of the E&I procedure been
assessed? (e.g. by simulation and by calculating
indicators, analysing distributions,…)
Accuracy and
reliability
Have the assessment results been taken into
account in the implementation of the finale
procedure?

18
3.3. Build or enhance dissemination components
33. This sub-process describes the activities to build new and enhance existing
components and services needed for the dissemination of statistical products as designed in
sub-process 2.1 (Design outputs). All types of dissemination components and services are
included, from those that are used to produce traditional paper publications to those that
provide web services, open data outputs, or access to micro-data.
Quality
Dimension
Indicator Notes
Managing
metadata
Extent to which relevant metadata can be
linked to output data
Pre-condition for 7.1
See also phase 7
Corresponds to the
accessibility and clarity
principle in the ES Code of
Practice
Accessibility
and clarity
Extent to which user requirements are
fulfilled in terms of dissemination formats,
information systems, graphical supports,…
Pre-condition for 7.1
See also phase 7
3.4. Configure workflows
34. This sub-process configures the workflow, systems and transformations used within
the statistical business processes, from data collection through to dissemination. It ensures
that the workflow specified in sub-process 2.6 (Design production systems and workflow)
works in practice.
Quality
Dimension
Indicator Notes
Soundness of
implementation
Ratio of the number of sub-processes
automated through an IT tool to the
total number of sub-processes
specified in 2.6
This quality indicator assumes that
processes have been specified in
BPMN or using another tool in 2.6
statistical procedures principle in
Timeliness and
punctuality
Planned timeliness of all subsequent
phases and sub-processes
See 2.6
ESS QPI- TP2. Time lag - final
results
3.5. Test production system
35. This sub-process is concerned with the testing of assembled and configured services
and related workflows. It includes technical testing and sign-off of new programmes and
routines, as well as confirmation that existing routines from other statistical business
processes are suitable for use in this case. Whilst part of this activity concerning the testing of
individual components and services could logically be linked with sub-process 3.2 (Build or
enhance process components), this sub-process also includes testing of interactions between
assembled and configured services, and ensuring that the production solution works as a
coherent set processes, information and services.

19
Quality
Dimension
Indicator Notes
Soundness of
implementation
Have all programs, routines and configured
services been individually tested and signed
off prior to the start of testing the production
system?
While 3.2 refers to newly
built programs and
routines, 3.5 includes
testing new and previously
existing programs and
routines.
Soundness of
implementation
Has the entire production system been tested
and signed off, ensuring that data correctly
enters and exits each program, routine and
configured service, and that the functionality
of each program, routine and configured
service has been executed according to
expectations?
This assumes that a there is
a business standard in the
statistical agency for the
system testing.
Corresponds to the
ES Code of Practice
3.6. Test statistical business process
36. This sub-process describes the activities to manage a field test or pilot of the statistical
business process. Typically it includes a small-scale data collection, to test collection
instruments, followed by processing and analysis of the collected data, to ensure the
statistical business process performs as expected. Following the pilot, it may be necessary to
go back to a previous step and make adjustments to instruments, systems or components. For
a major statistical business process, e.g. a population census, there may be several iterations
until the process is working satisfactorily.
Quality
Dimension
Indicator Notes
Cost
effectiveness
Estimated costs for producing and disseminating
outputs and divergences from planned costs in
design phase
See 2.6
Accuracy and
reliability
Pilot has been carried out and results have been
taken into account in final implementation
Dimension of the test/field pilot compared to real
survey
Accuracy and
reliability
Assessment of major error sources from the
pilot(e.g. coverage, nonresponse, measurement, and
process errors)
ESS QPI - A2.
Over-coverage –
rate
A4. Unit non-
response - rate
Timeliness and
punctuality
Estimated time frame for subsequent phases and sub-
processes and divergences from planned one in
design phase
See 2.6
ESS QPI - TP2.
Time lag - final
results
Accuracy and
reliability
Pilot has been carried out and results have been
taken into account in final implementation

20
Dimension of the test/field pilot compared to real
survey
3.7. Finalise production systems
37. This sub-process includes the activities to put the assembled and configured processes
and services, including modified and newly-created services into production ready for use by
business areas. The activities include:
 producing documentation about the process components, including technical
documentation and user manuals
 training the business users on how to operate the process
 moving the process components into the production environment, and ensuring they
work as expected in that environment (this activity may also be part of sub-process
3.5 (Test production system)).
Quality
Dimension
Indicator Notes
Accessibility and
clarity
Percentage of materials adequately archived
(easily retrievable; properly labelled; retention
period indicated)
D. Collect Phase
38. This phase collects or gathers all necessary information (data and metadata), using
different collection modes (including extractions from statistical, administrative and other
non-statistical registers and databases), and loads them into the appropriate environment for
further processing. Whilst it can include validation of data set formats, it does not include any
transformations of the data themselves, as these are all done in the "Process" phase. For
statistical outputs produced regularly, this phase occurs in each iteration.
39. The "Collect" phase is broken down into four sub-processes, which are generally
sequential, from left to right, but can also occur in parallel, and can be iterative. These sub-
processes are:
4.1. Create frame and select sample
40. This sub-process establishes the frame and selects the sample for this iteration of the
collection, as specified in sub-process 2.4 (Design frame and sample). It also includes the
coordination of samples between instances of the same statistical business process (for
example to manage overlap or rotation), and between different processes using a common
frame or register (for example to manage overlap or to spread response burden). Quality
assurance and approval of the frame and the selected sample are also undertaken in this sub-

21
process, though maintenance of underlying registers, from which frames for several statistical
business processes are drawn, is treated as a separate business process. The sampling aspect
of this sub-process is not usually relevant for processes based entirely on the use of pre-
existing sources (e.g. administrative sources) as such processes generally create frames from
the available data and then follow a census approach.
Quality
Dimension
Indicator Notes
Accuracy
and
reliability
The rate of over-coverage:
The proportion of units accessible via the
frame that do not belong to the target
population (are out-of-scope).
The rate of over-coverage is applicable:
− to all statistical processes (including use
of administrative sources);
− to producers.
If the survey has more than one unit type, a
rate may be calculated for each type.
If there is more than one frame or if over-
coverage rates vary strongly between sub-
populations, rates should be separated.
Need auxiliary data to assess
coverage; often cannot assess
coverage until after collection
has happened.
ESS QPI - A2. Over-coverage -
rate
Accuracy
and
reliability
Rate of duplicate records identified and
corrected during frame creation
Accuracy
and
reliability
Rate of missing or suspicious stratification
and classification variables; rate of missing
contact variables; time elapsed since last
successful contact
Accuracy
and
reliability
Relative discrepancy between expected
and observed sample size; relative
discrepancy between expected and
observed response, attrition and out of
scope rates
Can only be assessed after
collection is finished
ESS QPI - A4. Unit non-
response - rate.
Accuracy
and
reliability
The sampling error can be expressed:
a) in relative terms, in which case the
relative standard error or, synonymously,
the coefficient of variation (CV) is used.
b) in terms of confidence intervals.
Sampling errors indicators are applicable:
- to statistical processes based on
probability samples or other sampling
procedures allowing computation of such
information.
- to users and producers, with different
level of details given.
ESS QPI - A1. Sampling error
- indicators.......
Timeliness
and
punctuality
Delay between expected and actual
creation of frame
Timeliness
and
creation of sample

22
punctuality
4.2. Set up collection
41. This sub-process ensures that the people, processes and technology are ready to
collect data and metadata, in all modes as designed. It takes place over a period of time, as it
includes the strategy, planning and training activities in preparation for the specific instance
of the statistical business process. Where the process is repeated regularly, some (or all) of
these activities may not be explicitly required for each iteration. For one-off and new
processes, these activities can be lengthy. This sub-process includes:
 preparing a collection strategy;
 training collection staff;
 ensuring collection resources are available e.g. laptops;
 agreeing terms with any intermediate collection bodies, e.g. sub-contractors for
computer assisted telephone interviewing
 configuring collection systems to request and receive the data;
 ensuring the security of data to be collected;
 preparing collection instruments (e.g. printing questionnaires, pre-filling them with
existing data, loading questionnaires and data onto interviewers' computers etc.).
42. For non-survey sources, this sub-process will include ensuring that the necessary
processes, systems and confidentiality procedures are in place, to receive or extract the
necessary information from the source.
Quality
Dimension
Indicator Notes
Confidentiality
and security
Risk of a breach while data is being
transferred
Adequacy of
resources
Rate of HR requirements fulfilled;
rate of IT requirements fulfilled
Adequacy of
resources
Success rate for collection staff to
perform collection tasks after having
been trained
Test collection staff before and
after training to assess
effectiveness
Soundness of
implementation
Success rate for testing collection
systems, under expected as well as
high volume and extreme situations
End to end system testing.
statistical procedures principle in
Timeliness and
punctuality
sign-off of collection systems
(including data transmission, security,
collection management systems, and
QC systems)
Timeliness and
punctuality
sign-off of collection materials
(questionnaire, training materials,
etc.)

23
4.3. Run collection
43. This sub-process is where the collection is implemented, with the different
instruments being used to collect or gather the information, which may include raw micro-
data or aggregates produced at the source, as well as any associated metadata. It includes the
initial contact with providers and any subsequent follow-up or reminder actions. It may
include manual data entry at the point of contact, or fieldwork management, depending on the
source and collection mode. It records when and how providers were contacted, and whether
they have responded. This sub-process also includes the management of the providers
involved in the current collection, ensuring that the relationship between the statistical
organisation and data providers remains positive, and recording and responding to comments,
queries and complaints. For administrative and other non-statistical sources, this process is
brief: the provider is either contacted to send the information, or sends it as scheduled. When
the collection meets its targets, it is closed and a report on the collection is produced. Some
basic validation of the structure and integrity of the information received may take place
within this sub-process, e.g. checking that files are in the right format and contain the
expected fields. All validation of the content takes place in the Process phase.
Managing respondent
burden
Are there enough staff
responsible for dealing with
the respondent's questions
Managing respondent
burden
Support is provided to
respondents (e.g. toll free
number,…)
Accuracy and
reliability
Quality control is used to
manage the quality of data
collection and data capture
processes
Accuracy and
reliability
Meaningful feedback is
provided to interviewers and
fieldworkers on a regular
basis
Accuracy and
reliability
Monitoring of fieldwork
operations are done during
data collection
Accuracy and
reliability
Interviewer performance is
measured for CATI, CAPI,
PAPI surveys, e.g.
interviewers’ productivity
Accuracy and
reliability
Domain response rates;
representativity indicators;
achieved CVs of key
variables in domains of
interest
ESS QPI - A1. Sampling error –
indicators
A4. Unit non-response - rate
Accuracy and
reliability
Unit nonresponse rate; item
nonresponse rate; proxy rate
ESS QPI - A4. Unit non-response –
rate
A5. Item non-response - rate
Accuracy and
reliability
Mode effect when more
than one collection mode
Can only be assessed after
estimation

24
Accuracy and
reliability
Outgoing error rates;
estimate of non-sampling
error
Data capture is covered in 4.4
Timeliness and
punctuality
Delay between expected and
actual start and close of
collection
4.4. Finalise collection
44. This sub-process includes loading the collected data and metadata into a suitable
electronic environment for further processing. It may include manual or automatic data take-
on, for example using clerical staff or optical character recognition tools to extract
information from paper questionnaires, or converting the formats of files received from other
organisations. It may also include analysis of the process metadata (paradata) associated with
collection to ensure the collection activities have met requirements. In cases where there is a
physical collection instrument, such as a paper questionnaire, which is not needed for further
processing, this sub-process manages the archiving of that material.
Quality
Dimension
Indicator Notes
Cost-
effectiveness
Discrepancy between planned
versus actual collection costs
Percentage of collection activities
that met requirements (assessed
through analysis of paradata)
Accuracy and
reliability
Outgoing error rates; estimate of
non-sampling error
Accuracy and
reliability
The proportion of units accessible
via the frame that do not belong to
the target population (are out-of-
scope).
The rate of over-coverage is
applicable:
− to all statistical processes
(including use of administrative
sources);
− to producers.
If the survey has more than one
unit type, a rate may be calculated
for each type.
If there is more than one frame or if
over-coverage rates vary strongly
between sub-populations, rates
should be separated.
Need auxiliary data to assess
coverage; often cannot assess
coverage until after collection has
happened.
ESS QPI - A2. Over-coverage - rate
Accuracy and
reliability
ESS QPI - A4. Unit non-response –
rate
A5. Item non-response - rate

25
Accessibility
and clarity
Percentage of materials adequately
archived (easily retrievable;
properly labelled; retention period
indicated)
E. Process Phase
45. This phase describes the cleaning of data and their preparation for analysis. It is made
up of sub-processes that check, clean, and transform input data, so that they can be analysed
and disseminated as statistical outputs. It may be repeated several times if necessary. For
statistical outputs produced regularly, this phase occurs in each iteration. The sub-processes
in this phase can apply to data from both statistical and non-statistical sources (with the
possible exception of sub-process 5.6 (Calculate weights), which is usually specific to survey
data).
46. The "Process" and "Analyse" phases can be iterative and parallel. Analysis can reveal
a broader understanding of the data, which might make it apparent that additional processing
is needed. Activities within the "Process" and "Analyse" phases may commence before the
"Collect" phase is completed. This enables the compilation of provisional results where
timeliness is an important concern for users, and increases the time available for analysis.
47. This phase is broken down into eight sub-processes, which may be sequential, from
left to right, but can also occur in parallel, and can be iterative. These sub-processes are:
5.1. Integrate data
48. This sub-process integrates data from one or more sources. It is where the results of
sub-processes in the "Collect" phase are combined. The input data can be from a mixture of
external or internal data sources, and a variety of collection modes, including extracts of
administrative data. The result is a set of linked data. Data integration can include:
 combining data from multiple sources, as part of the creation of integrated statistics
such as national accounts
 matching / record linkage routines, with the aim of linking micro or macro data from
different sources
 prioritizing, when two or more sources contain data for the same variable, with
potentially different values
49. Data integration may take place at any point in this phase, before or after any of the
other sub-processes. There may also be several instances of data integration in any statistical
business process. Following integration, depending on data protection requirements, data may
be anonymised, that is stripped of identifiers such as name and address, to help to protect
confidentiality.

26
Quality
Dimension
Indicator Notes
Accuracy and
reliability
The proportion of units covered by both
the survey and the administrative sources
in relation to the total number of units in
the survey.
The proportion is applicable
− to mixed statistical processes where
some variables or data for some units come
from survey data and others from
administrative source(s);
− to producers.
ESS QPI - A3. Common units -
proportion
Accuracy and
reliability
Existence of linkage variables (unique
identifier) of the register (yes/no question)
Linking of microdata to other microdata.
Future development needed in this
area, since no agreed upon
international/corporate indicators at
this point.
Accuracy and
reliability
Degree of linkability of the linkage
variables (high, medium, low)
this point.
Accuracy and
reliability
Reliability of the linkage results This indicator will depend on the
linkage key, its accuracy in the
linked data sets, the validity of the
linkage procedure , as well as the
comparison between the finally
resulting linked records and the
expected ones
this point.
Accuracy and
reliability
Percentage of errors comes from
identification and transformation of
population, units or data items.
 It is possible that the meaning of a
population, a unit or data items
changes in the course of the
process. Errors may occur in this
transformation process.
 The conversion of one statistical
concept into another.
For example; measurement units for
imported and exported products collected
from administrative sources could be
different than the measurement units for
statistically required data. This type of
errors should be measured during the
integration of data.

27
Accuracy and
reliability
If record linkage is required, percentage of
records that were successfully matched
this point.
Accuracy and
reliability
Extent to which manual intervention was
needed in data linking
this point.
5.2. Classify and code
50. This sub-process classifies and codes the input data. For example automatic (or
clerical) coding routines may assign numeric codes to text responses according to a pre-
determined classification scheme.
Quality
Dimension
Indicator Notes
Methodological
soundness
Compliance rate of classifications
of input data to the pre-determined
standard international classification
and national versions of
international classification scheme
All international or national
classifications and breakdowns
which are used for the data set are
produced
e.g. although NACE Rev2 is
introduced as international
classification, using the older
version or using a different
classification than the proposed
classification.
Methodological
soundness
Compliance rate of coding of input
data to the pre-determined standard
coding scheme
The standard coding scheme in this
indicator refers to the compliance
with the local codes used in these
variables.
Accuracy and
reliability
It is calculated as the ratio between
the number of values automatically
coded and the total number of
values submitted to coding.
It measures the efficiency of the
automatic coding procedure
Accuracy and
reliability
Quality control is used to manage
the quality of automated and
manual coding processes
Timeliness and
punctuality
timing of adaptation of
correspondence tables
5.3. Review and validate
51. This sub-process examines data to try to identify potential problems, errors and
discrepancies such as outliers, item non-response and miscoding. It can also be referred to as
input data validation. It may be run iteratively, validating data against predefined edit rules,

28
usually in a set order. It may flag data for automatic or manual inspection or editing.
Reviewing and validating can apply to data from any type of source, before and after
integration. Whilst validation is treated as part of the “Process” phase, in practice, some
elements of validation may occur alongside collection activities, particularly for modes such
as web collection. Whilst this sub-process is concerned with detection of actual or potential
errors, any correction activities that actually change the data are done in sub-process 5.4.
Quality
Dimension
Indicator Notes
Accuracy
and
reliability
Rate of actual errors
Identification of incorrect data (actual
errors) in the processing stage - Missing,
invalid or inconsistent entries or that point
out data records that are actually in error.
5.4. Edit and impute
52. Where data are considered incorrect, missing or unreliable, new values may be
inserted in this sub-process. The terms editing and imputation cover a variety of methods to
do this, often using a rule-based approach. Specific steps typically include:
 the determination of whether to add or change data;
 the selection of the method to be used;
 adding / changing data values;
 writing the new data values back to the data set, and flagging them as changed;
 the production of metadata on the editing and imputation process.
Quality
Dimension
Indicator Notes
Accuracy
and
reliability
Imputation rate
- The indicator is expressed as the
ratio of the number of replaced
values to the total number of
values for a given variable.
The imputation rate is applicable:
− to all statistical processes (with
micro data; hence, e.g., direct data
collection and administrative
data);
− to producers.
Information on the extent to
which imputation is used and the
reasons for it should be noted. A
short description of the methods
used and their effects on the
estimates.
This indicator is influenced both
The un-weighted rate shows, for a
particular variable, the proportion of units
for which a value has been imputed due to
the original value being a missing,
implausible, or inconsistent value in
comparison with the number of units with
a value for this variable.
The weighted rate shows, for a particular
variable, the relative contribution of
imputed values to the estimate of this
item/variable. Obviously this weighted
indicator is meaningful when the objective
of a survey is that of estimating the total
amount or the average of a variable. When
the objective of the estimation is that of
estimating complex indices, the weighted
indicator is not meaningful.
ESS QPI - A7. Imputation - rate

29
by the item non-response and the
editing process. It measures both
the relative amount of imputed
values and the relative influence
on the final estimates from the
imputation procedures.
The un-weighted imputation rate
for a variable is the ratio of the
number of imputed values to the
total number of values requested
for the variable.
The weighted rate shows the
relative contribution to a statistic
from imputed values; typically a
total for a quantitative variable.
For a qualitative variable, the
relative contribution is based on
the number of units with an
imputed value for the qualitative
item.
Accuracy
and
reliability
An indicator of an edit's
effectiveness would be the rate of
false negative or false positive
assessments.
One way to verify this would be to re-
interview the respondents of a sample of
units to confirm the reported values, and
see what proportion of true values were
flagged as errors and what proportion of
errors were not flagged as errors.
Accuracy
and
reliability
Edit failure rates can be calculated
for key variables and by domains
of interest. A sub-class of edits
could be those designed to detect
outlier observations.
A high/very high edit failure rate for a
given variable would be suggest possible
errors in previous phases (e.g. in the
questionnaire or in data collection).
Accuracy
and
reliability
Rate of robustness of outliers for
key variables
Robustness of Outliers =
Corrected/Discarded Outliers /
Total detected outliers
This indicator will measure the
quality of outlier detection
process
Accessibility
and clarity
Percentage of metadata
adequately archived (easily
retrievable; properly labelled;
retention period indicated)

30
5.5. Derive new variables and units
53. This sub-process derives data for variables and units that are not explicitly provided in
the collection, but are needed to deliver the required outputs. It derives new variables by
applying arithmetic formulae to one or more of the variables that are already present in the
dataset, or applying different model assumptions. This activity may need to be iterative, as
some derived variables may themselves be based on other derived variables. It is therefore
important to ensure that variables are derived in the correct order. New units may be derived
by aggregating or splitting data for collection units, or by various other estimation methods.
Examples include deriving households where the collection units are persons or enterprises
where the collection units are legal units.
Quality
Dimension
Indicator Notes
Accuracy and
reliability
Rate of model assumptions and associated
errors can be expressed as errors due to
domain specific models needed to define
the target of estimation.
- Where models are applicable in relation
to a specific source of error, they should be
presented in the section concerned. This is
recommended also in the case of a cut-off
threshold and model based estimation.
- Domain specific models, for example, as
needed to define the target of estimation
itself, should be thoroughly described and
their validity for the data at hand assessed.
(SIMS)
- Model assumption errors occur with the
use of methods, such as calibration,
generalized regression estimator,
calculation based on full scope or constant
scope, benchmarking, seasonal adjustment
and other models not included in the
preceding accuracy components, in order
to calculate statistics or indexes. (OECD
Glossary)
- Model assumption errors are errors
caused by models used. Models are based
on assumptions. (Statistics Netherlands)
In case of model based seasonal
adjustment; the diagnostics like
autocorrelation test, seasonal
autocorrelation test, skewness, kurtosis and
normality test for model residuals provides
the opportunity of checking model
assumptions satisfied such as Best Linear
Unbiased Estimator.
Another example can be given as Small
A short description of the
methods used and their effects
on the estimates.

31
Ares Estimation which is the estimation of
main indicators for small domains in case
of non-representability of the survey for
them. The diagnostics can be given for
example Haussman test, and residual-
based test depends on the model used.
Coherence and
comparability
Rate of comparability for derived variables
Definitions, classifications and units of
derived variables will be taken as reference
for the comparability and coherence
checks.
5.6. Calculate weights
54. This sub process creates weights for unit data records according to the methodology
created in sub-process 2.5 (Design processing and analysis). In the case of sample surveys,
weights can be used to "gross-up" results to make them representative of the target
population, or to adjust for non-response in total enumerations. In other situations, variables
may need weighting for normalization purposes.
Quality
Dimension
Indicator Notes
See 2.5
Accuracy and
reliability
The weights are adjusted for coverage
and non-response error (yes/no
indicator)
5.7. Calculate aggregates
55. This sub-process creates aggregate data and population totals from micro-data or
lower-level aggregates. It includes summing data for records sharing certain characteristics,
determining measures of average and dispersion, and applying weights from sub-process 5.6
to derive appropriate totals. In the case of sample surveys, sampling errors may also be
calculated in this sub-process, and associated to the relevant aggregates.
Quality
Dimension
Indicator Notes
Accuracy
and
reliability
The sampling error can be expressed:
a) in relative terms, in which case the
relative standard error or, synonymously, the
coefficient of variation (CV) is used.
b) in terms of confidence intervals.
Sampling errors indicators are applicable:
- to statistical processes based on probability
samples or other sampling procedures
allowing computation of such information.
- to users and producers, with different level
of details given.
This indicator is also included
in 4.1 (in 4.1 you haven’t yet
collected the data so you can’t
actually calculate them yet:
now you can)
ESS QPI - A1. Sampling error
- indicators

32
Accuracy
and
reliability
The following indicators are proposed to
analyse the revision:
1) Mean Absolute Revision (MAR), i.e.
average of absolute revisions over a time
period (useful to analyse stability in terms
of size). If a relative measure, useful for
comparisons and to analyse levels is
sought: Relative Mean Absolute Revisions
(RMAR), i.e. relative average of absolute
revisions over a time period
2) Mean Revision (MR), i.e. average of
revisions over a time period (useful to
analyse directions in terms of sign) and its
significance (Yes/Not)
Standard Deviation of Revisions (SDR), i.e.
a measure of the variability of the revisions
The “revision” is defined as the
difference between a later and
an earlier estimate of the key
item.
The proposed indicators are the
OECD core/basic measures.
Indicators MR, MAR and
RMAR are Eurostat Quality
and Performance Indicators
included in the “A6. Data
revision - average size” group.
A t-test to assess if MR is
significantly different from 0
exists and the second indicator
proposed in 2) results from it.
ESS QPI - A6. Data revision -
average size..
5.8. Finalise data files
56. This sub-process brings together the results of the other sub-processes in this phase
and results in a data file (usually of macro-data), which is used as the input to the "Analyse"
phase. Sometimes this may be an intermediate rather than a final file, particularly for business
processes where there are strong time pressures, and a requirement to produce both
preliminary and final estimates.
Quality
Dimension
Indicator Notes
Relevance data completeness rate: extent to which the
outputs satisfy requirements (e.g. from
Regulations or other agreements with
users)
Could be calculated as the ratio of
the number of data cells obtained
to the number of data cells
required
ESS QPI - R1. Data completeness
– rate
Accuracy
and
reliability
Degree of closeness of computations or
estimates to the exact or true value
Accuracy: closeness of computations or
estimates to the exact or true values that
the statistics were intended to measure.
(SIMS)
Reliability: closeness of the initial
estimated value to the subsequent

33
estimated value. (SIMS)
Timeliness
and
Punctuality
integration of data
F. Analyse Phase
57. In this phase, statistical outputs are produced, examined in detail and made ready for
dissemination. It includes preparing statistical content (including commentary, technical
notes, etc.), and ensuring outputs are "fit for purpose" prior to dissemination to customers.
This phase also includes the sub-processes and activities that enable statistical analysts to
understand the statistics produced. For statistical outputs produced regularly, this phase
occurs in every iteration. The "Analyse" phase and sub-processes are generic for all statistical
outputs, regardless of how the data were sourced.
58. The "Analyse" phase is broken down into five sub-processes, which are generally
sequential, from left to right, but can also occur in parallel, and can be iterative. The sub-
processes are:
6.1. Prepare draft outputs
59. This sub-process is where the data are transformed into statistical outputs. It includes
the production of additional measurements such as indices, trends or seasonally adjusted
series, as well as the recording of quality characteristics.
Quality
Dimension
Indicator Notes
Soundness of
implementation
To what extent is the business process
using standard or well-known methods
(e.g. calculating indices, trends, seasonal
adjustment?
statistical procedures principle
in the ES Code of Practice
Accuracy and
reliability
Quality Control methods can be applied to
ensure that the accuracy of the
transformation process itself is sufficient.
Indicators could be percentage of outputs
reviewed (manually or automated),
percentage of errors detected.
Timeliness and
punctuality
Delay between the anticipated and actual
completion of this step.
6.2. Validate outputs
60. This sub-process is where statisticians validate the quality of the outputs produced, in
accordance with a general quality framework and with expectations. This sub-process also

34
includes activities involved with the gathering of intelligence, with the cumulative effect of
building up a body of knowledge about a specific statistical domain. This knowledge is then
applied to the current collection, in the current environment, to identify any divergence from
expectations and to allow informed analyses. Validation activities can include:
• checking that the population coverage and response rates are as required;
• comparing the statistics with previous cycles (if applicable);
• checking that the associated metadata and paradata (process metadata) are present and in
line with expectations
• confronting the statistics against other relevant data (both internal and external);
• investigating inconsistencies in the statistics;
• performing macro editing;
• validating the statistics against expectations and domain intelligence.
Quality
Dimension
Indicator Notes
Accuracy and
reliability
Proportion of overall budget dedicated
to validation activities; number of
validation measures applied;
As an example of validation
measure the indicator
“Asymmetry for mirror flows
statistics” can be calculated (ESS
QPI- CC1. Asymmetry for mirror
flows statistics - coefficient)
Accuracy and
reliability
Number or amount of changes made to
the data based on validation results;
Coherence and
comparability
Availability of backcasting procedures
where there is a break in the series
Coherence and
comparability
Degree of coherence with other sources,
with provisional data, with quick
estimates, with previous results of the
same process
6.3. Interpret and explain outputs
61. This sub-process is where the in-depth understanding of the outputs is gained by
statisticians. They use that understanding to interpret and explain the statistics produced for
this cycle by assessing how well the statistics reflect their initial expectations, viewing the
statistics from all perspectives using different tools and media, and carrying out in-depth
statistical analyses.
Quality
Dimension
Indicator Notes
Accuracy and
reliability
Proportion of overall budget dedicated to
interpretation and explanation activities;
extent to which a report is produced and
accepted
6.4. Apply disclosure control
62. This sub-process ensures that the data (and metadata) to be disseminated do not
breach the appropriate rules on confidentiality. This may include checks for primary and
secondary disclosure, as well as the application of data suppression or perturbation
techniques. The degree and method of disclosure control may vary for different types of
outputs, for example the approach used for micro-data sets for research purposes will be
different to that for published tables or maps.

35
Quality
Dimension
Indicator Notes
Statistical
Confidentiality
and security
To which extent is the business
process using standard or well-known
methods identification and protection
of sensitive information
Corresponds to the
appropriateness of statistical
procedures principle in the ES
Code of Practice
Statistical
Confidentiality
and security
To what extent is the data protected
from the risk of disclosure of sensitive
information?
Some software provide a
diagnostic indicating the level of
protection
Statistical
Confidentiality
and security
To what extent is the data actually
protected? What is the residual risk of
disclosure?
Statistical
Confidentiality
and security
To what extent has the usability of the
data been degraded? What is the loss
in precision or level of detail?
6.5. Finalise outputs
63. This sub-process ensures the statistics and associated information are fit for purpose
and reach the required quality level, and are thus ready for use. It includes:
 completing consistency checks;
 determining the level of release, and applying caveats;
 collating supporting information, including interpretation, commentary, technical notes,
briefings, measures of uncertainty and any other necessary metadata;
 producing the supporting internal documents;
 pre-release discussion with appropriate internal subject matter experts;
 approving the statistical content for release.
Quality
Dimension
Indicator Notes
Relevance Number of planned outputs that were not
disseminated
Accuracy and
reliability
Number of errors that were detected and had to
be corrected
Accessibility
and clarity
Metadata completeness - rate
The rate of completeness of metadata is the
ratio of the number of metadata elements
provided to the total number of metadata
elements applicable.
The rate of completeness of metadata is
applicable:
- to all statistical processes;
- to producers
ESS QPI - AC3. Metadata
completeness – rate

36
G. Disseminate Phase
64. This phase manages the release of the statistical products to customers. It includes all
activities associated with assembling and releasing a range of static and dynamic products via
a range of channels. These activities support customers to access and use the outputs released
by the statistical organization.
65. For statistical outputs produced regularly, this phase occurs in each iteration. It is
made up of five sub-processes, which are generally sequential, from left to right, but can also
occur in parallel, and can be iterative. These sub-processes are:
7.1. Update output systems
66. This sub-process manages the update of systems where data and metadata are stored
ready for dissemination purposes, including:
 formatting data and metadata ready to be put into output databases;
 loading data and metadata into output databases;
 ensuring data are linked to the relevant metadata.
67. Formatting, loading and linking of metadata should preferably mostly take place in
earlier phases, but this sub-process includes a final check that all of the necessary metadata
are in place ready for dissemination.
Quality
Dimension
Indicator Notes
Accessibility and
clarity
Date of last update of the content of the
metadata.
- The date of the latest dissemination of
the metadata should be specified.
- The date on which the metadata
element was inserted or modified in the
database should be specified.
Managing
metadata
Extent to which metadata are available
and accessible
Corresponds to the
accessibility and clarity
principle in the ES Code
of Practice
7.2. Produce dissemination products
68. This sub-process produces the products, as previously designed (in sub-process 2.1),
to meet user needs. They could include printed publications, press releases and web sites. The

37
products can take many forms including interactive graphics, tables, public-use micro-data
sets and downloadable files. Typical steps include:
 preparing the product components (explanatory text, tables, charts, quality statements
etc.);
 assembling the components into products;
 editing the products and checking that they meet publication standards.
Quality
Dimension
Indicator Notes
Quality
commitment
Ratio of statistical products that are
disseminated with quality
statements/quality reports
Relevance The rate of available statistics
The indicator is the ratio of the number
output data elements provided in
accordance to a relevant regulation to
those required by the regulation.
- The extent to which all statistics that
are needed are available.
Relevance Percentage of/Extent to
which “statistical outputs/products”
meets users’ needs
- Description of users and their
respective needs with respect to the
statistical data.
This indicator is also included in
2.1.
It shall be considered in 7.2
Accessibility
and clarity
The extent to which relevant metadata
is linked to output data
See also 3.3
7.3. Manage release of dissemination products
69. This sub-process ensures that all elements for the release are in place including
managing the timing of the release. It includes briefings for specific groups such as the press
or ministers, as well as the arrangements for any pre-release embargoes. It also includes the
provision of products to subscribers, and managing access to confidential data by authorized
user groups, such as researchers. Sometimes an organization may need to retract a product,
for example if an error is discovered. This is also included in this sub-process.
Quality
Dimension
Indicator Notes
Impartiality
and objectivity
Availability of revision policy (Yes/No) See 2.1 and 5.7
Impartiality
and objectivity
Time lag between the release of an output and
announcement of the error to the users
Transparency Number of press meetings held before and
after the release of outputs
Corresponds to the
impartiality and objectivity
principle in the ES Code of
Practice

38
Accuracy and
reliability
Number of errors corrected in disseminated
products
Excludes planned revisions.
Can include non-data errors.
Timeliness
and
punctuality
Punctuality of statistical outputs
Punctuality is the time lag between the
delivery/release date of data and the target
date for delivery/release as agreed for
delivery or announced in an official release
calendar, laid down by Regulations or
previously agreed among partners.
The punctuality of statistical outputs is
applicable:
- to all statistical processes with fixed/pre-
announced release dates,
- to users and producers, with different
aspects and calculation formulae.
ESS QPI - TP3. Punctuality -
delivery and publication
Timeliness
and
punctuality
Time lag - first results
General definition:
The timeliness of statistical outputs is the
length of time between the end of the event or
phenomenon they describe and their
availability.
Specific definition:
The number of days (or weeks or months)
from the last day of the reference period to the
day of publication of first results.
This indicator is applicable:
- to all statistical processes with preliminary
data releases;
- to producers.
ESS QPI - TP1. Time lag -
first results
Timeliness
and
punctuality
Time lag - final results
General definition:
The timeliness of statistical outputs is the
length of time between the end of the event or
phenomenon they describe and their
availability.
Specific definition:
The number of days (or weeks or months)
from the last day of the reference period to the
day of publication of complete and final
results.
ESS QPI - TP2. Time lag -
final results

39
of details given.
Timeliness
and
punctuality
Availability of a dissemination policy
defining dissemination practices and its
availability on the web site
Accessibility
and clarity
Availability of a release calendar and its
availability on the web site
Accessibility
and clarity
Number of analytical and data products
accessed
Website metrics for
publications and data tables
available on the
organization’s website.
Include page views and
downloads.
Accessibility
and clarity
Percentage of website visitors who found the
information that they were looking for
Five-point scale: all, most,
half, some, none
Coherence and
comparability
Length of comparable time series
Number of reference periods in time series
from last break.
Comment
Breaks in statistical time series may occur
when there is a change in the definition of the
parameter to be estimated (e.g. variable or
population) or the methodology used for the
estimation. Sometimes a break can be
prevented, e.g. by linking.
The length of comparable
series is applicable:
- to all statistical processes producing time-
series;
of details given.
ESS QPI- CC2. Length of
comparable time series
7.4. Promote dissemination products
70. Whilst marketing in general can be considered to be an over-arching process, this sub-
process concerns the active promotion of the statistical products produced in a specific
statistical business process, to help them reach the widest possible audience. It includes the
use of customer relationship management tools, to better target potential users of the
products, as well as the use of tools including web sites, wikis and blogs to facilitate the
process of communicating statistical information to users.
Quality
Dimension
Indicator Notes
Relevance User satisfaction about the metadata
availability
-user satisfaction surveys shall include

40
questions on the opinions of users about
metadata availability
Accessibility
and clarity
The number of social media visitors/followers
Accessibility
and clarity
Metadata - consultations
Number of metadata consultations (ESMS)
within a statistical domain for a given time
period.
By "number of consultations" it is meant the
number of times a metadata file is viewed.
Some information is available through the
monthly Monitoring report on Eurostat
Electronic Dissemination and its excel files
with detailed figures.
- to producers
ESS QPI - AC2. Metadata -
consultations
Accessibility
and clarity
Number of consultations of data tables within
a statistical domain for a given time period
ESS QPI - AC1. Data tables
– consultations
7.5. Manage user support
71. This sub-process ensures that customer queries and requests for services such as
micro-data access are recorded, and that responses are provided within agreed deadlines.
These queries and requests should be regularly reviewed to provide an input to the over-
arching quality management process, as they can indicate new or changing user needs.
Quality
Dimension
Indicator Notes
Relevance - User satisfaction index
- Length of time since most recent user
satisfaction survey
Measures to determine user satisfaction.
Relevance The percentage of unmet user needs
Relevance Time since last user consultation, in terms of
years or months
Accessibility
and clarity
Availability of an information service/unit or
a call centre to users to answer enquires about
data and metadata issues

41
H. Evaluate Phase
72. This phase manages the evaluation of a specific instance of a statistical business
process, as opposed to the more general over-arching process of statistical quality
management described in Section VI. It logically takes place at the end of the instance of the
process, but relies on inputs gathered throughout the different phases. It includes evaluating
the success of a specific instance of the statistical business process, drawing on a range of
quantitative and qualitative inputs, and identifying and prioritising potential improvements.
73. For statistical outputs produced regularly, evaluation should, at least in theory occur
for each iteration, determining whether future iterations should take place, and if so, whether
any improvements should be implemented. However, in some cases, particularly for regular
and well established statistical business processes, evaluation may not be formally carried out
for each iteration. In such cases, this phase can be seen as providing the decision as to
whether the next iteration should start from the Specify Needs phase, or from some later
phase (often the Collect phase).
74. This phase is made up of three sub-processes, which are generally sequential, from
left to right, but which can overlap to some extent in practice. These sub-processes are:
8.1. Gather evaluation inputs
75. Evaluation material can be produced in any other phase or sub-process. It may take
many forms, including feedback from users, process metadata (paradata), system metrics, and
staff suggestions. Reports of progress against an action plan agreed during a previous
iteration may also form an input to evaluations of subsequent iterations. This sub-process
gathers all of these inputs, and makes them available for the person or team producing the
evaluation.
Quality
Dimension
Indicator Notes
Output
quality
Extent to which quality indicators
have been collected for all phases and
sub-phases including costs and
timeliness of phases and sub-phases.
Indicators and feedbacks should
have been collected in previous
phases (and some of them probably
also analysed)
Output Quality gathers all
dimensions related to the quality of
statistics (e.g. relevance, accuracy,
timeliness, coherence,…)
Output
quality
Types and relative weight of different
measures gathered (e.g. quantitative
indicators, feedback from users,
paradata or other metrics derived by
procedures, staff suggestions,
interviewers/supervisors follow ups)
Indicators and feedback should have
been collected in previous phases
(and some of them probably also
analysed)

42
8.2. Conduct evaluation
76. This sub-process analyses the evaluation inputs and synthesises them into an
evaluation report. The resulting report should note any quality issues specific to this iteration
of the statistical business process, and should make recommendations for changes if
appropriate. These recommendations can cover changes to any phase or sub-process for
future iterations of the process, or can suggest that the process is not repeated.
Quality
Dimension
Indicator Notes
Soundness of
implementation
To which extent process
components satisfy process quality
requirements such as Efficiency,
Effectiveness; Robustness;
Flexibility; Transparency and
Integration
See also phase 3. Build.
For a new process, such an
assessment has been carried out in
phase 3. Build.
For regular processes this stage
could represent the opportunity to
assess both process components and
outputs.
ES Code of Practice
Cost
effectiveness
Percentage of GSBPM phases and
sub-processes for which there were
no gaps between planned and
attained costs
Output quality Extent to which quality indicators
are close to target values (includes
all indicators and metadata such as
those needed for quality reporting)
assessment is based on information
from 8.1
Output quality Trends in quality indicators (e.g.
improvements/worsening) for
recurring processes.
Output quality Percentage of quality dimensions
and sub-dimensions (e.g. for
accuracy) that was not possible to
assess and why.
Output quality If an evaluation report has been
produced and on which basis (e.g.
overall assessment of quality
indicators calculated during the
process, application of a quality
assessment procedure, e.g. self-
The indicator can assume values like
:
0 (no evaluation report produced)
1 (evaluation report produced on
currently available quality
indicators)

43
assessment, audit…) 2 (evaluation report produced on the
result of an ad hoc analysis, e.g. a
study to estimate MSE)
3 (evaluation report produced on the
result of a self-assessment
procedure)
4 (evaluation report produced on the
result of an audit procedure)
Timeliness and
punctuality
sub-processes for which there were
no gaps between target and achieved
timeliness
ESS QPI - TP2. Time lag - final
results
Output quality Have evaluated substantial changes
in quality indicators (Yes/No)
Significant changes in quality
indicators and other measures such
as edit failure rates, response rates
etc. may signal quality issues.
8.3. Agree an action plan
77. This sub-process brings together the necessary decision-making power to form and
agree an action plan based on the evaluation report. It should also include consideration of a
mechanism for monitoring the impact of those actions, which may, in turn, provide an input
to evaluations of future iterations of the process.
Quality commitment Extent to which the action plan contains
mechanisms for monitoring the impact of
improvement actions
Quality commitment Assuming that an evaluation report was
prepared in 8.2 for quality indicators of
previous GSBPM phases, and the gaps
were identified between the expected and
actual quality of the output, cost
effectiveness and timeliness; then the
decision needs be made to take action for
areas where the gaps are identified.
The quality indicator is the ratio of: the
number of actionable quality issues
(quality indicators where problems are
identified or targets are not met) / to the
total number of quality issues
Also a plan can be made to not take an
action for all actionable items but for some
of them. In that case the quality indicator
is: number of quality issues to take action
for divided by the number of all actionable
quality issues

44
Quality commitment Completion rate of the action plan is: the
number of successfully fixed or improved
quality issues divided by total number of
quality issues planned to be fixed
III. Over-arching processes
78. The GSBPM also recognises several over-arching processes that apply throughout the
production phases, and across statistical business processes. The processes of quality
management and metadata management are further elaborated in this Section.
79. Quality concerns organisations, processes and products. In the present framework,
quality management over-arching process refers mainly to product and process quality.
80. The main goal of quality management within the statistical business process is to
understand and manage the quality of the statistical products. There is general agreement
among statistical organisations that quality should be defined according to the ISO 9000-2005
standard: “The degree to which a set of inherent characteristics fulfils requirements"5
. Thus,
product quality is a complex and multi-faceted concept, usually defined in terms of several
quality dimensions. The dimensions of quality that are considered most important depend on
user perspectives, needs and priorities, which vary between processes and across groups of
users.
81. In order to improve the product quality, quality management should be present
throughout the statistical business process model. It is closely linked to Phase 8 (Evaluate),
which has the specific role of post-evaluating individual instances of a statistical business
process. However, quality management has both a deeper and broader scope. As well as
evaluating iterations of a process, it is also necessary to evaluate separate phases and sub-
processes, ideally each time they are applied, but at least according to an agreed schedule.
Metadata generated by the different sub-processes themselves are also of interest as an input
for process quality management. These evaluations can apply within a specific process, or
across several processes that use common components.
82. In addition, a fundamental role in quality management is played by the set of quality
control actions that should be implemented within the sub-processes to prevent and monitor
errors. The strategy could be reported in a quality assurance plan.
83. Within an organisation, quality management will usually refer to a specific quality
framework, and may therefore take different forms and deliver different results within
different organisations. The current multiplicity of quality frameworks enhances the
importance of the benchmarking and peer review approaches to evaluation, and whilst these
approaches are unlikely to be feasible for every iteration of every part of every statistical
business process, they should be used in a systematic way according to a pre-determined
schedule that allows for the review of all main parts of the process within a specified time
period6
.
5
ISO 9000:2005, Quality management systems -- Fundamentals and vocabulary. International Organization for
Standardization
6
A suitable global framework is the National Quality Assurance Framework developed by a global expert group
under the United Nations Statistical Commission. See: http://unstats.un.org/unsd/dnss/QualityNQAF/nqaf.aspx
Quality Management

45
84. Broadening the field of application of the quality management over-arching process,
evaluation of groups of statistical business processes can also be considered, in order to
identify potential duplication or gaps.
85. All evaluations result in feedback, which should be used to improve the relevant
process, phase or sub-process, creating a quality loop.
86. Examples of quality management activities include:
 Setting and maintaining of the quality framework;
 Setting of global quality criteria;
 Setting process quality targets and monitoring compliance;
 Seeking and analysing user feedback;
 Reviewing operation and documenting lessons learned;
 Examining process metadata and quality indicators;
 Internal or external auditing on statistical processes.
87. Quality management also involves institutional and organisational factors. Such
factors are included in other GSBPM over-arching processes (e.g. Human resources
management, Statistical programme management) although they can have an impact on
quality.
Quality
Dimension
Indicator Notes
Quality
commitment
Availability of a quality assurance plan, or any
other similar scheme,
describes the working standards, the formal
obligations (such as laws and internal rules) and
the set of quality control actions to prevent and
monitor errors, to evaluate quality indicators
and to control different points at each stage of the
statistical process.
- This indicator is valid for the institutional level
Quality
commitment
Availability of a quality policy and its availability
on the web site
Improve Run
Evaluate
Plan

46
- A Quality Commitment Statement is made
publicly available, laying out principles and
commitments related to quality in statistics which
are consistent with the goals set out in the mission
and vision statements.
- This indicator is valid for the institutional level
Quality
commitment
Availability of procedures to plan and monitor the
quality of the statistical production process.
Quality
commitment
Availability of a clear organizational structure for
managing quality within the statistical authority.
Examples of such a structure are:
Quality Committee;
Quality Manager;
Centralized Quality unit;
Other structures (e.g. a selected group of staff
trained as “quality pilots” to act as
project/processes coach/advisers).
Quality
commitment
For what proportion of GSBPM sub-processes are
standardised corporate solutions used?
Quality
commitment
Is a process of risk identification and management
in place? (Yes/No)
Time since risk management plans were last
reviewed? (Years and Months)
Quality
commitment
Extent of HR requirements fulfilled
E.g., Training, Staffing
Quality
commitment
Extent to which quality indicators, metadata and
paradata are compliant to standards
Managing
respondent
burden
Is there a communication strategy encouraging
response by informing potential respondents about
the survey
Managing
respondent
burden
Percentage of statistics produced from
administrative data and other data sources instead
of survey
Covers all statistical
domains
Annex 1
European Statistical System Quality and Performance Indicators – ESS QPI
 R1. Data completeness – rate
 A1. Sampling error - indicators.......
 A2. Over-coverage - rate
 A3. Common units - proportion
 A4. Unit non-response - rate
 A5. Item non-response - rate
 A6. Data revision - average size..
 A7. Imputation - rate
 TP1. Time lag - first results

47
 TP2. Time lag - final results
 TP3. Punctuality - delivery and publication
 CC1. Asymmetry for mirror flows statistics - coefficient
 CC2. Length of comparable time series
 AC1. Data tables – consultations
 AC2. Metadata - consultations
 AC3. Metadata completeness – rate
ESS QPI Name in GSBPM Phase and sub-process
R1. Data completeness –
rate
Planned data completeness rate: Design phase. 2.1.
Design outputs
Data completeness rate Process phase. 5.8
Finalise data file
A1. Sampling error -
indicators.......
Key indicators for sample design
(e.g. estimated size,
expected/planned sampling errors
for key variables, domains,
costs,…)
Design phase. 2.4.
Design frame and sample
The sampling error can be
expressed:
a) in relative terms, in which case
the relative standard error or,
synonymously, the coefficient of
variation (CV) is used.
b) in terms of confidence intervals
Collect phase. 4.1.
Create frame and select
sample
achieved CVs of key variables in
domains of interest
Collect phase 4.3. Run
collection
The sampling error can be
expressed:
a) in relative terms, in which case
the relative standard error or,
synonymously, the coefficient of
variation (CV) is used.
b) in terms of confidence
intervals.
Process phase. 5.7.
Calculate aggregates
A2. Over-coverage - rate Impact of coverage errors: assess
the likely impact of coverage error
on key estimates.
Design phase. 2.4.
Design frame and sample
Assessment of major error sources
from the pilot(e.g. coverage,
nonresponse, measurement, and
process errors)
Build phase. 3.6. Test
statistical business
process
scope).
Collect phase. 4.1.
sample
Collect phase. 4.4.
finalise collection

48
scope).
A3. Common units -
proportion
The proportion of units covered
by both the survey and the
administrative sources in relation
to the total number of units in the
survey
Process phase. 5.1.
integrate data
A4. Unit non-response -
rate
Assessment of major error sources
from the pilot(e.g. coverage,
nonresponse, measurement, and
process errors)
Build phase. 3.6. Test
process
Relative discrepancy between
expected and observed sample
size; relative discrepancy between
expected and observed response,
attrition and out of scope rates
Collect phase. 4.1.
sample
achieved CVs of key variables in
domains of interest
collection
collection
Collect phase. 4.4.
finalise collection
A5. Item non-response -
rate
collection
Collect phase. 4.4.
finalise collection
A6. Data revision - average
size..
Data revisions are planned
(Yes/No)
Design phase. 2.1 design
outputs
The following indicators are
proposed to analyse the revision:
Mean Absolute Revision (MAR),
i.e. average of absolute revisions
over a time period (useful to
analyse stability in terms of size).
If a relative measure, useful for
comparisons and to analyse levels
is sought: Relative Mean Absolute
Revisions (RMAR), i.e. relative
average of absolute revisions over
a time period
Mean Revision (MR), i.e. average
of revisions over a time period
(useful to analyse directions in
terms of sign) and its significance
(Yes/Not)
Standard Deviation of Revisions
(SDR), i.e. a measure of the
variability of the revisions
Process phase.5.7.
Calculate aggregates

49
A7. Imputation - rate Imputation rate
- The indicator is expressed as the
ratio of the number of replaced
values to the total number of
values for a given variable.
Process phase 5.4. Edit
and impute
TP1. Time lag - first results Time lag - first results Disseminate phase. 7.3.
Manage release of
dissemination products
TP2. Time lag - final results Planned time frame for
subsequent phases and sub-
processes
Design phase 2.6. Design
production systems and
workflow
Build phase. 3.4.
Configure workflows
Estimated time frame for
subsequent phases and sub-
processes and divergences from
planned one in design phase
Build phase. 3.6 Test
process
Time lag - final results Disseminate phase. 7.3.
Manage release of
sub-processes for which there
were no gaps between target and
achieved timeliness
Evaluate phase. 8.2.
Conduct evaluation
TP3. Punctuality - delivery
and publication
Punctuality of statistical outputs
Punctuality is the time lag
between the delivery/release date
of data and the target date for
delivery/release as agreed for
delivery or announced in an
official release calendar, laid
down by Regulations or
previously agreed among partners.
Disseminate phase. 7.3.
Manage release of
CC1. Asymmetry for mirror
flows statistics - coefficient
Asymmetry for mirror flows
statistics
Analyse phase. 6.2
Validate outputs
CC2. Length of comparable
time series
Expected length of comparable
time series.
Design phase. 2.1 design
outputs
Length of comparable time series Disseminate phase. 7.3.
Manage release of
AC1. Data tables –
consultations
Number of consultations of data
tables within a statistical domain
for a given time period
Disseminate phase. 7.4.
Promote dissemination
products
AC2. Metadata -
consultations ................
Metadata – consultations Disseminate phase. 7.4.
Promote dissemination

50
products
AC3. Metadata
completeness – rate
Metadata completeness - rate Analyse Phase. 6.5.
Finalise outputs

Quality indicators for the GSBPM

More Related Content

What's hot (20)

Similar to Quality indicators for the GSBPM (20)

More from Светла Иванова (20)

Recently uploaded (20)

Quality indicators for the GSBPM