Overview of the register-based census
Two thirds of Germany’s population were not questioned in the 2011 Census. Thanks to the new register-based method, however, this has not been a problem. The people concerned are nonetheless covered by the census. Here you will find information on how the new census approach works.
2011 Census goal - more than just counting the population
The goal of the 2011 Census is to provide the most accurate snapshot possible of basic data on the country’s population and the employment and housing conditions. The reference date of that snapshot was 9 May 2011.
What makes the census results so special is that they will allow evaluations for very small areas. The 2011 Census will provide not only the Federation, Länder, administrative regions and districts but also the municipalities with relevant data for planning purposes. Results relating to buildings and dwellings can be evaluated by municipalities with a safe statistical unit and for scientific purposes, even for individual street formations.
Comparability across Europe despite different methods
In accordance with a European Union requirement, all Member States have to conduct censuses of the population and of buildings and housing at ten year intervals from 2011. To ensure comparability of the results across Europe, all Member States have to supply data for a specified range of variables. However, they have considerable freedom to choose the methodology applied to obtain the information needed.
Overview of the 2011 Census model in Germany
Unlike in the past when the approach was that of complete enumeration, Germany decided to apply a register-based method to conduct the current EU-wide census. This means that existing administrative registers are used as data sources. In certain areas, they were supplemented by a combination of complete counts and sample surveys.
Using data from administrative registers
Technological progress has enabled the use of data that are available from administrative registers today. As there is an obligation to register in Germany, all municipalities maintain population registers that include comparable information. In addition, the Federal Employment Agency has information on all employees subject to social insurance contributions and on all people registered as unemployed or as jobseekers. And finally, public employers can provide data on public officials, judges and soldiers.
These administrative data, however, do not provide reliable information, for instance, on the educational attainment, precise occupation or living conditions of an individual. In addition, the relevant registers do not contain employment data for certain groups of the population (such as self-employed people). Regarding buildings and dwellings, register data are not at all available for Germany as a whole.
For the above reasons, supplementary surveys such as the census of buildings and housing, household survey and survey in residential establishments and collective living quarters had to be integrated into the 2011 Census. The latter survey was required because the census test had shown that the information on people living in these establishments was rather vague in the population registers.
Household generating procedure
A particular challenge has been to link the individual data sources. The results of the census of buildings and housing as such are appropriate for answering questions about the number and size of dwellings. The data from the population registers provide information for instance on the number of children at kindergarten age. However, the 2011 Census must also give an answer to questions like how big is the average useful floor space available to families with three and more children or to people living alone. Using data from the various census components, these household relationships can be generated in the course of a separate so-called household generating procedure.
Before deciding upon the new census procedure, a detailed census test was carried out between 2001 and 2003. It was designed, among other things, to find out whether the register-based method would yield reliable results. In fact, the test confirmed that the new procedure was suited to link data from the different registers and to eliminate, in statistical terms, inaccuracies contained in them. In this way it could be ensured that the results of the register-based 2011 Census would be precise.