European Statistical System Sponsorship on Standardisation
On the management of statistical standards in the ESS
The Sponsorship on Standardisation is a strategic task force of the European Statistical System (ESS) consisting of six Member States (DE, FR, HU, IT, LV, NL) and Eurostat. The aim is to produce proposals that will improve quality and/or efficiency of the ESS through integration of processes. This requires a certain level of standardisation. The Sponsorship analyses the current standards, scenarios for further standardisation and instruments to support that standardisation process. This paper will focus on three elements of the work of the Sponsorship:
1) A framework to describe future scenarios for standardisation and integration;
2) A SWOT instrument to assess the business case;
3) A process and organisational structure for standardisation efforts.
1. Framework for mapping initiatives for integrating ESS business processes
The Sponsorship has devised a framework for analysing and mapping either individual integration and standardisation initiatives or full standardisation scenarios in the European Statistical System (ESS). The scope of the framework concerns the system of EU NSIs (and other statistical authorities) and Eurostat involved in the production of EU statistics.
The framework is based on a simplified/adapted model of the business process. It defines four main business areas (policy, design, management and implementation) and the related activities , including meta-processes, so as to go beyond the flat and mechanical representation of statistical production processes provided by the GSBPM. This framework is inspired by the Statistics Netherlands (CBS) enterprise architecture and involves stakeholders such as the NSIs, Eurostat or other public administrations at the different stages. It is compatible with the principle of subsidiarity.
The process is supported by services and infrastructures which may be common (EU) or specific (national), as shown in Figure 1 below.
For each of the four main business areas and the related activities, the framework tentatively proposes four integration scenarios . The four options differ as to the level of integration of the resources and processes and range from a set of autonomous entities (= level A) to a fully integrated situation (a complete set of common infrastructure = level D). The intermediate levels (levels B and C) would correspond respectively to a minimum standard approach to integration and to the sharing of some infrastructures. Level D would require a significant reorganisation of some of the ESS business areas.
Figure 1. An architectural view of the business process for producing EU statistics and the supporting infrastructure
The choice of the appropriate level of integration in each specific sub-domain should be based on a careful SWOT analysis. The coherence of the approach over the different business areas, however, is critical. For instance, appropriate integration of operational aspects should be driven by agreed common methods and by appropriate governance.
2. SWOT analysis method for evaluation (business case) of standardisation scenarios
On various occasions, for example at ESS Committee and Partnership Group meetings (governance bodies of the European Statistical System), there has been a clear demand for the business case to be provided for further ESS standardisation. In particular, we need a tool to assess the relative merits and costs of the main investments for building the infrastructure necessary for the different scenarios proposed above. This assessment involves many different factors, most of which are not easily quantified. In order to overcome these difficulties it would be useful to consider the categories (Strengths, Wea k nesses, Opportunities and Threats) of a qualitative SWOT analysis, developing a SWOT instrument  which consists of two key components.
A) A fixed list of concrete SWOT aspects to be assessed. Proposed aspects are:
Improved process and systems quality (less risk)
Easy development of new statistical processes and systems
Easy incorporation of new data sources
Easy incorporation of new dissemination channels
(Re)use of existing ESS standards, systems and/or approaches
Reduced personnel costs
Costs of development (one-off)
Costs of transition (one-off)
Costs of support and maintenance (recurring)
Loss of autonomy (enforced vs. voluntary standards)
Lack of flexibility (rigid standards)
Lack of room for differences between parties (e.g. national differences)
(Re)use of standards, systems and/or approaches from non-ESS parties
Improved quality of individual data sets for strategic and other users
Increased consistency of data over statistical domains
Easier development of new statistical products
Reduced burden on respondents
Better communication with users and stakeholders
Loss of identity for ESS partners
Proprietary standards that hamper cooperation with non-ESS partners
High entry costs for new parties
Lack of coherence with national (government) policies
Lack of synergy with other statistical communities (UN, OECD, …)
Lack of support from stakeholders/ funding providers
B) A scale for scoring each aspect. We propose a scale composed of two factors:
Relevance . For example, the threat ‘High entry costs’ might be considered less relevant than ‘Lack of support’ (or vice versa).
Effect . For example, a scenario with a lot of law-enforced rules has a large impact on the aspect ‘High entry costs’.
Relevance might be scored on a scale from 1 to 4, while e f fect might be scored on a scale from 0 to 3. For a given scenario, the value of an aspect is the product of both factors.
After fixing the SWOT aspects and scales, different standardisation scenarios/initiatives can be scored and analysed. For example, their relative performance can be compared. An important open question is at what level of detail the SWOT instrument should be applied. A too detailed level may require a lot of work without providing a lot of insight.
A major strength of the SWOT instrument is that many a s pects can be studied simultaneously in a standardised, but still flexible, way. The key weakness is that the approach is very qualitative and does not provide insight into the various cost categories of a standardisation scenario. An impo r tant opportunity is that SWOT results can be presented in an easy and intu i tive way to stakeholders to give them a flavour of the directions taken. On the other hand, a threat is that the SWOT aspects can be criticised as incomplete or the SWOT scores considered arb i trary.
It is possible to combine the SWOT instrument with more quantitative ways to present business cases for standardisation, for example by calculating ‘guesstimates’ of development and maintenance costs or expected efficiency gains.
Lastly, the SWOT analysis should be carried out by the different organisations involved, and aggregated to a European business case, also identifying the underlying spread in merits and costs.
3. Process for setting and implementing ESS standards
Figure 2 below presents a proposal for a standardisation process that can be used for setting and implementing ESS standards  . The process consists of five main phases that each contain a number of activities. It describes the full lifecycle of a standard and is cyclical: a standard may become obsolete and then needs to be replaced by a new standard.
Figure 2. The standardisation process
The standardisation process should be managed in a formal and structured way according to the principles listed below:
Consensus . Acceptance by c onsensus ensures that all views are heard and the resulting standard is generally agreed to by those involved.
Transparency and openness . Involvement of all parties ensures transparency of the process, and provides a public notice of a proposed standard in advance.
Balance means that no one group’s interest dominates the approach.
Due process ensure s that anyone with a ‘direct and material interest’ has a right to express a position and to have that position considered (where necessary including the right to appeal).
The preconditions and infrastructure for a standardisation process are the following:
A) Organisation and responsibilities
The proposal is for the ESS Committee to be responsible for strategic issues and approval, and for Directors of Methodology Group in close cooperation with the IT Directors Group to look after managerial tasks. The directors of methdology could be supported by a steering group on standardisation and a secretariat. Tasks requiring special expertise can be delegated. Decisions on funding are made by Eurostat and Member States as regards EU and national resources respectively.
The initiative of setting up a new standard can be bottom-up (e.g. it comes from working groups or NSIs) or top-down (e.g. it comes from the ESSC, Eurostat or a directors group). The figure above shows the path taken by the potential new standard.
The new standardisation procedure also has to deal with the transition. A set of potential standards currently in use has to be adopted formally as ‘standard’. A similar question is the adoption of standards developed by other organisations. In these cases a shorter procedure can be used, e.g. starting with the adoption phase.
B) A standards repository (inventory/register)
An accurate register and repository for standards and pre-standards (for which the process is on-going) and an archive of standards withdrawn is necessary. The design of this component is the task of another pillar of the Sponsorship on Standardisation.
 The current proposal is based on a similar SWOT instrument by Kloek, Szűcs and Vereczkei (2011)
 This proposal is based on the results of ESSnet Preparation of Standardisation