1. Statistical offices collect information about persons, businesses, institutions, etc. through censuses and surveys. The data collected are ultimately released in a suitable form to policy makers, researchers and the general public for statistical purposes. The release of such information may have the undesirable side-effect that information on individual entities instead of on (sufficiently ...

1. The task of statistical offices is to produce and publish statistical information about society. The data collected are ultimately released in a suitable form to policy makers, researchers and the general public for statistical purposes. The release of such information may have the undesirable effect that information on individual entities instead of on sufficiently large groups of individua...

The paper discusses how a statistical office could strike a satisfactory balance between confidentiality protection and freedom of information. Flexible use of statistical data is of vital interest for researchers and for the democratic process. On the other hand, the willingness of respondents to provide data is dependent on the ability of the statistical office to guarantee their anonymity. T...

1. The accurate and reliable assessment of statistical disclosure risk is a necessary pre-cursor of the efficient application of disclosure control techniques. Research on statistical disclosure risk assessment with respect to microdata files has two major themes. The first uses the concept of uniqueness, examining the level of unique individuals or households within populations and samples (Se...

1. National Statistical Institutes are able to release microdata sets only on the condition that the privacy of respondents is safe and that the event of breach of confidentiality is extremely unlikely. Different institutions adopt different definitions of disclosure, of disclosure risk and use different models to estimate such risk and protect microdata set. In all cases, the aim is the same: ...

1. The election of a suitable threshold value (n) in table protection is one of the main points of discussion for a statistical agency, when fixing generic rules to preserve confidentiality. Nowadays, different criteria are applied in statistical offices and institutes. These cell limit values vary from 3 to 5, in some cases. In Spain, the most expanded rule is not to publish frequency cells un...

An overview of papers submitted on Topic 2 (Impact of new technogical developments in software, communications and computing on SDC) is first given. Then a list of key issues for substantive discussion are identified.

1. The paper presents brief descriptions of various cell suppression software packages, focussing on availability, costs, platforms, and the underlying methodology. The systems performances will be compared with respect to information loss, and computing time requirement; other key qualities of the software will be discussed.

In the course of the EU funded project CASC, the software τ-ARGUS shall be extended to become a generally applicable standard tool for tabular data protection. The required modifications will affect the facilities provided for (residual) disclosure risk statement, the data structure and the user interface. Methods will have to be implemented for protection of complex hierarchical tables and for...

1. Statistical offices collect large amounts of data for statistical purposes. Respondents are only willing to provide the statistical offices with the required information if they can be certain that these offices will disseminate their data with the utmost care. This implies that the confidentiality of the data must be assured. Therefore the amount of detail in the publications has to be limi...