Challenges in Survey Taking for the Next Decade APPLICATIONS OF STATISTICAL DISCLOSURE CONTROL METHODS

National Statistical Institutes (NSIs) and Market Research Bureaus conduct surveys about many different topics. To reach this aim they have developed a fully-equipped statistical production process. It is a long way from collecting raw data to publishing public information. This paper discusses one of the methodological aspects of the statistical production process. The aim of a NSI or Market Research Bureau is always to publish as much as possible. However, the privacy of individual respondents should be respected. Therefore, statistical disclosure control techniques have been developed to protect sensitive information that can be attributed to individual respondents. Statistical Disclosure Control (SDC) is a topic that can be approached from many viewpoints. In this paper it is explained how results of European research projects on SDC can be used in the production of official statistics. Two related software packages are described that can be applied for producing safe data. The package τ-ARGUS is used for tabular data and its twin µ-ARGUS for microdata. The main techniques used to protect sensitive information are global recoding and local suppression. Both τ-ARGUS and its twin µ-ARGUS are products developed in the SDC project under the Fourth Framework Programme of the European Union. New versions (that include results of the on-going research) of both packages have been released in the CASC (Computational Aspects of Statistical Confidentiality) project that is funded under the Fifth Framework Programme of the European Union. Also other methods that allow use of data are discussed. The most important one is that of restricte