Obligation to statistical confidentiality
Under the terms of the Federal Statistics Act, the RDC are obliged to ensure statistical confidentiality when offering access to microdata. The safeguarding of confidentiality follows a given set of criteria.
Statistical confidentiality implies for the RDC that they may not release results that allow conclusions on individual cases. For this reason, the RDC check every result that is produced within an on-site use of microdata for statistical confidentiality. They do this by applying the set of criteria that is valid for the according statistic.
We ask you to ensure that the results of your analyses meet these rules and do not contain confidentiality issues. If your results allow conclusions on individual cases, these results have to be blocked. These blocks and the related loss of information can transfer to future analyses if blocked values could be revealed by subtraction.
The leaflet 'Regulations on the analysis of microdata’ contains important information that helps you with your analysis:
- Legal and methodological basics of statistical confidentiality
- Requirements concerning the program code
- Criteria on the admission of output
- Valid confidentiality rules per statistic
- Terms of use
To help you preparing a program code that meets the rules, we kindly ask you to use the sample program code. We provide those for the analysis software packages SAS, SPSS and Stata.
Please note: If you don’t follow the rules described in the regulations, the RDC can make use of the possibility to reject the program code you sent for remote execution resp. the check and release of the results you generated at a safe centre. Since 1 September 2017, this option is laid down by default in all contracts.
You may find further information on the topic of “confidentiality” in RDC working paper No. 50 "Statistische Geheimhaltung - Der Schutz vertraulicher Daten in der amtlichen Statistik".