 |
ESSnet on common tools and harmonised methodology for SDC in the ESS |
Contact:Peter-Paul de WolfStatistics Netherlands P.O. Box 24500 2490 HA The Hague The Netherlands Phone: +31 70 337 5060 
Last update: 10 Oct 2011 |
Task 1.1. Framework for the harmonisation of microdata protection.ObjectivesThe aim is to test the methodological framework for micro data protection for multiple member states as proposed in Ichim and Franconi (2010) on a real case study stemming from a European survey. The model proposes to achieve harmonisation of SDC process through the harmonisation of methodology at the input and the provision of harmonised and objective measures for the output allowing de facto some flexibility to the whole process and allowing for better adaptation to national context and better global efficiency.Description of workThe case study chosen is the Continuing Vocational Training Survey (CVTS) for whom the release of micro data for research is enabled by the legal framework and anonymisation methods are not yet finalised. The output will be the development of techniques to solve each step of the SDC process with default options and variations to reach the harmonised anonymisation of CVTS microdata. Software routines will be developed as well. The methods and the routines will be sent to CVTS representatives in all MS to get feedback. If a dataset is available at Eurostat with microdata from various MS the operational implementation of the method on few cases will be delivered.In case of cancellation of the WG/TF, or indeed of the CVTS survey itself, another survey will be considered in this project to test the feasibility of the proposed framework. The alternative survey will be taken from the list of surveys in Reg. 831/2002 among those which are harmonised and running; it will be either one for whom no anonymisation method has been agreed yet or one for whom the release is already taking place but the number of participating countries to the CD-Rom distribution is limited. We expect to be informed about the cancellation of the CVTS to the earliest notice in order to make proper arrangement for the other survey. Task 1-1: Choice of measure for the outputA crucial step in the definition of possible methodologies to be applied to the data is the definition of the set of objective measures for the output that will be maintained by all possible candidate methods. Identification of such benchmarking statistics will involve in particular the participation of survey experts. Views on possible candidate statistics will be shared with CVTS experts to reach a general agreement from the side of the data producers.Task 1-2: User needsUsers needs are essential when identifying possible anonymisation methods. The analysis of the projects undertaken by researchers on the data and contacts with users of CVTS data will contribute to the definition of the benchmarking statistics and to prioritise features.Task 1-3: Definition of methodologiesDestatis and Statistics Austria have already presented proposals to reach protection of CVTS data ((Brandt and Hafner, 2007; Meindl and Templ, 2008, Lenz et al. 2006) as well as other (Belsby et al. 2005). These methods will be reviewed to study ways to achieve anonymisation taking into account the new framework and the benchmarking statistics. Istat will also contribute with the definition of another methodology.Task 1-4: Implementation and reportingThe method will be implemented for CVTS data from few participating MS to show feasibility. (Sweden, Austria, Germany, Italy and The Netherlands). A report on the implementation process of the above framework will be produced at the end of the project detailing the pros and cons and critical points encountered by MS, viability of general proposal and future directions..
|