Firms´ Panel Data and Factual Anonymisation

The aim of the project is to supply the scientific community with economic panel data without violating confidentiality. Therefore these micro data can be released only in anonymized form. Since so far most data files of the German Statistical Office have only been available in cross-sectional form, the data have to processed as follows: In the first step cross-sectional data of the statistical offices will be linked over time. In the second step the analytical validity of these data will be improved by additional information from other data resources. The last step will be the analysis of the possibility to realize factual anonymised panel data and the creation of so called scientific-use-files.

Access to micro and individual data held by the statistical offices is regulated by the Federal Law for Statistics (BStatG). It most recent version allows scientists with the aim of independent scientific research to obtain micro data which are only ‘moderately’ This new regulation is associated with the term factual anonymity as defined by the European Science Foundation, and it is also referred to as research-data secrecy.Such factually anonymised database is also called a scientific-use-file due to its exclusive disposability to science, in opposition to so called public-use-files available to everyone.

If a statistical office wants to provide the scientific community with anonymised microdata, the microdata have to satisfy two conditions. On the one hand, they must preserve the analytical validity in the best possible way (information loss as low as possible), on the other hand, the reidentification risk must be low (guarantee of anonymity).

The project work is based on the results of the project “Factual Anonymisation of Firms’ Micro Data”. In this project the factual anonymisation of cross-sectional firms´data could be realised by using data perturbating methods. The Institute for Applied Economic Research will be responsible for the evaluation of the effects of data perturbating methods on the analytical validity of firms´ panel data. The other project work will be done by the Institute for Research and Development in Federal Statistics, Institute for Research and Development of the Federal States´ Statistics and the “Institut für Arbeitsmarkt- und Berufsforschung (IAB)”, Nürnberg.