Tasks of DA
Tasks: Overview
- Introduction
- Task 1: Shortlist of the possibilities of the structure and architecture of a European network
- Task 2: Elaboration of pros and cons of the solutions on the shortlist
- Task 3: Guideline for safe centres
- Task 4: Evaluation of assignability of general rules of output checking to the ECHP data
- Task 5: Documentation for researchers to use EU microdata through a safe centre
- Task 6: Feasibility and cost analysis
- Task 7: Information and involvement to non-participating NSI’s
The project procedure is described in different tasks. Each task will be arranged and elaborated by a designated MS, respectively a MS team.
At first it is necessary to give an overview concerning the present ways of access to microdata and also the demand of the scientific community in the different MS. Hence, task 1 evaluates the actual situation of safe centres in the MS and the request to European datasets. After that comprehension, a shortlist of possibilities to create an European network of safe centres will be worked out. Task 2 deals with the elaboration of the pros and cons of the solutions on the shortlist, including a detailed discussion about technical, legal and cost aspects of the different possibilities to develop a structure and architecture of a European network. Also based on the results of task 2, in the context of task 3 the construction of a guideline for European safe centres is intended. In task 1 to 2 it is necessary to discuss whether the implementation of a decentralised access to EU microdata sets is practicable and which legal frames of the countries in the European Statistical System (ESS) must be regarded.
Besides the technical and legal requirements, anonymisation procedures need also to be focused. Thus, task 4 handles with the evaluation of the assign ability of general rules of output checking to the ECHP data. Task 5 develops a consistent guideline and documentations for researchers to use safe centres. Task 6 includes the feasibility and cost analysis for a future implementation. In task 7 the non-participated NSI’s will be informed about project of the decentralised access to EU microdata sets.
Task 1: Shortlist of the possibilities of the structure and architecture of a European network
In the first period of the project an evaluation of the current situation regarding safe centres in Europe is necessary. How many Member states offer access to microdata via a safe centre? Which MS are planning to build up a safe centre in the next years? By answering those questions, it is essential to get to know more about the different structures and managements of safe centres in the several MS.
Imaginable is to use the results of the brief questionnaire created by another ESSnet project, namely “the ESSnet on Statistical Disclosure Control”, to evaluate the state of the art in Europe as far as RDCs are concerned. The analysis of the questionnaire will give the picture of the situation of the RDCs in Europe, the future nodes of a network. Starting from this the work will go on in identifying different possible forms of a European network.
The aim of this task is to investigate the different schemes of interactions between the nodes (the national RDCs) and therefore providing ideas of possible architectures. Besides this, the aim of the task is to produce a complete list of issues of organisational, legal and procedural nature to be solved in order to network RDC in different Member states. From a first analysis of such issues for each type of identified architecture a shortlist of candidate architectures will be provided to be further studied in the subsequent task 2.
Task 2: Elaboration of pros and cons of the solutions on the shortlist
Task 2 will investigate the different solutions on the shortlist in detail and give the actual scoring of these possible solutions on the defined aspects.
The pros and cons of the different approaches have to be investigated in respect to the technical implementation and also with regard to its conversion as well as the legal and financial aspects. The results of this task can be the basis for an informed decision on the best way to implement local access to European datasets. Depending on the actual outcome, it will even permit decision-makers to describe a phased approach.
For instance one solution, although suboptimal, might be very suited for immediate implementation while other, possibly better, solutions take more time and resources to implement and can be seen as a future goal. In this way a long-term vision can be presented while also making actual progress in the short term.
One of the most central issues of the project refers to the idea of implementing a structure and architecture of the network. There are several ways that might be considered. Depending on the network architecture there are technical problems and legal issues to be solved. Every type of network has advantages and disadvantages in aspects of technical, managerial and legal requirements.
Task 3: Guideline for safe centres
Independent from the different ways of network architecture the safe centre itself has to fulfil certain criteria regarding security and confidentiality aspects. For the sharing of the data it is necessary that the security requirements of the safe centres will be accepted by the single MS. For this reason a certification system in terms of a guideline shall be developed, where a safe centre has to fulfil certain criteria to count as safe to provide EU microdata. On the one hand this offers for present safe centres the possibility to upgrade to a certified safe centre to host European datasets. On the other hand this is also helpful for countries which have not implemented a safe centre yet. They can build it in a way that it’s possible to guarantee access to their own national and even European datasets.
Task 4: Evaluation of assignability of general rules of output checking to the ECHP data
A determining necessity for providing access to microdata on a European scale is consistency in the way each Member State checks output against disclosure of data on an individual level. A common set of guidelines is therefore needed.
The development of this set of guidelines is currently being dealt within another ESSnet project, namely the “Guidelines project on output checking”, that is part of the ESSnet project on Statistical Disclosure Control. The result of this guidelines project will be a set of guidelines that can be applied to all kinds of microdatasets (business, households, individuals). In the context of the present proposal, task 4 aims to evaluate the transferability of the anonymisation procedures to the ECHP data exemplarily. Due to the fact, that this dataset holds a persistent demand and involves a plurality of states, it seems sufficient and adequate for a comparable achievement.
This guidelines project started in the beginning of 2008 and will be completed at the end of 2009. Since the time-frame of this project coincides with the project on “Decentralised access to EU microdata sets”, it would be inefficient to tackle the same problem again independently in this project on decentralised access. Especially since all countries involved in the guidelines group are also participating in this project on decentralised access, i.e. Germany, Italy, UK, and The Netherlands (and for the most part even with the same people).
Task 5: Documentation for researchers to use EU microdata through a safe centre
In addition to the scenario of architecture, technology, management, procedures and legality, as a result of the earlier tasks, task 5 will provide a scheme of documentation set and a system of metadata to be provided for the researcher and guidelines for the availability and practical use of this documentation.
First of all researcher need general information how to access EU microdata sets. Second, a detailed description of the microdata sets, and a correctly structured test data file is necessary. The description shall contain for example the following information:
• The circumstance of the survey,
• Guidelines for the use of the data files
• List of variables (in the order of the questionnaire and also in alphabetical order)
• The calculation of derived variables
• Description of sampling and weighting
• The related questionnaires and guidelines for the questionnaires and the related publications should be available
Microdata access shall operate in accordance with a standardised procedure. A clear description and eventual help should be provided for the users. The first step is to work out a standardised questionnaire for access requests. It should cover description of research purposes, data needs, information about researcher and research institution.
A standardised form of contract is needed too. Signing of the contract and granting access has to be in accordance with the legal regulation. The standard of SDMX will be considered for providing metadata. And if possible the metadata will be compatible with the SDMX metadata standard.
Task 6: Feasibility and cost analysis
One part of the project is to explore the feasibility to offer a viable, Hcost-effectiveH and (regarding to the state-comparability) standardised access to European microdata in safe centres of the MS. Since there are many different legal and technical restrictions to consider, one central aim is to pose a best practice proposal for a comparable way of data-access. On the bases of the outcomes of this feasibility study it should be possible to devise central requirements for a future implementation of the achieved results. The ways and possibilities to implement the best practice solutions are then to be evaluated in follow-up projects. The feasibility and cost analysis refers to the possible solutions developed on the shortlist in task 1.
For this task it is needed to have an overview of the salaries of the different MS. Depending on this and the costs of technical infrastructure in a certain MS the costs of running a certified safe centre to host European datasets could be estimated per year. It should be taken in account that salaries and IT costs can vary from year to year. The HcostsH HtherebyH HincurredH must be calculated and a financial disposition of the certified safe centre has to be defined.
Task 7: Information and involvement to non-participating NSI’s
The project shall also inform other non-participating countries about the content of the project and the results received so far. For this reason a website will be developed where information about the project and documents can be provided. This website is also helpful for the participants of the project to provide and share documents and information in a special login area. The website will contain a communication tool like a feedback form or a contact e-mail-adress, where other non-participating countries are invited to give feedback on the project and where comments are expected for example on a special legal situation in their country. An invitation e-mail with the link of this website will be send to the responsible institutions in all Member states of the EU to inform about the project and they are welcome to make remarks and give input. The MS will be informed as soon as first results in the project are achieved and the website is established.
In addition to the communication tool it is planned to arrange a session about the project on decentralised access to EU microdata sets within a European conference or workshop i.e. the UNECE conference “work session on statistical data confidentiality”. We will contact the organizer of the conference to contribute a session on “access to European datasets”. If the conference takes place in the fourth quarter of the project year first results of the project can be reported in this session. At this moment it is not clear yet where and when the conference will take place. There will probably be a decision in the first two month of the year 2009. This topic will also be discussed in the first meeting of the project group.
It is also planned to have a feedback from countries which are interested in this area and which are trying to establish there own safe centre. The observer PT will participate on the meetings to comment the single steps of the project and to give feedback on the development.
