Symbolix

We provide statistical analysis and disclosure control services to assist data providers to make their data available to the public in a safe and understandable way. We work a lot with Space-Time Research, who provide software to allow huge amounts of data to be accessed through database and online software. Statistical disclosure controls are the techniques we use to reduce the possibility that individuals or businesses can be uniquely identified in published data sets. This is a critical to modern data dissemination to safely satisfy the demand for online access to public data. Data collectors and providers must undertake to protect the privacy of all respondents. In a web 2.0 world, this goes well beyond just removing names and credit card numbers from the database. In this project, we reviewed a number of surveys, censuses and official registries of public and enterprise data, collected by a European Statistics Organisation. This collection of data is being prepared for online release, but requires careful data preparation to ensure that privacy is protected, but the data can still be accessed and used by the public. To achieve this, we developed a risk profile for each data set, and the combined group. This allowed us to compare and evaluate a range of possible confidentialisation techniques in use around the world. The recommenations considered the likelihood and consequence of a breach, the utility of the final data set, the development time and whether each method could be applied in a web 2.0 setting.