Challenges for statistical disclosure control in a world with big data and open data
Peter‐Paul de Wolf, Kees Zeelenberg · Munich Personal RePEc Archive (Ludwig Maximilian University of Munich) · 2015
National statistical institutes (NSIs) produce tables and microdata files; these data are typically checked for unwanted disclosure of data of individual persons and enterprises. In recent years many other data sources, in the form of open data and big data, have become available, for the general public, for researchers, and for and from data collectors and providers other than NSIs. This poses new problems for statistical disclosure control. We discuss several of these problems, such as should NSIs protect their tables and microdata files against linking to other datasets so as to prevent detailed profiling of their respondents, should we entertain different disclosure scenarios and switch to other disclosure-control methods?