Determining the numbers of inhabitants
Determining the new official numbers of inhabitants is based on the data transmitted from the population registers (as at 9 May 2011) to the statistical offices. However, as was shown by the census test of 2001, not all data from the population registers are precise and up to date. Therefore, the statistical offices cannot simply count the data from these registers to obtain the numbers of inhabitants, they rather have to carry out a number of correction measures.
In the census test of 2001, the statistical offices discovered that the distribution of overcoverage and undercoverage in the population registers varies among municipalities and that larger towns and municipalities tend to have not only higher absolute values but also higher proportions of overcoverage and undercoverage. Therefore, the question of which measure should be chosen to correct the number of inhabitants depends on whether the municipality concerned is small (less than 10,000 inhabitants) or large (10,000 inhabitants or more).
Extrapolation of overcoverage and undercoverage
In larger municipalities with at least 10,000 inhabitants, a sample-based household survey was conducted. In this random sample survey, the extent of overcoverage and undercoverage in the population registers was determined at the selected addresses and was then extrapolated for the relevant municipality.
Survey to clarify discrepancies
In smaller municipalities, however, this method is not efficient enough. This is why, in municipalities with less than 10,000 inhabitants, the survey to clarify discrepancies was conducted. Here, the addresses were not selected at random, as was the case for larger municipalities. Instead, addresses were ascertained where there were implausible constellations of data from the population registers and data from the questionnaires of the census of buildings and housing. The survey was conducted only at addresses with just one dwelling. Undercoverage and overcoverage discovered for the addresses included in the survey were corrected directly in the statistical data and were not extrapolated, as was the case for larger municipalities.
Check for multiple registration
There are persons who, in the registers, are registered several times with a (main) residence. But there are other persons who are registered only with a second residence. Such incorrect information in the population register data had to be corrected, too, during statistical processing. Again, various methods were applied, depending on the municipality size. In large municipalities with at least 10,000 inhabitants, the data were corrected in an automated way. In smaller municipalities with less than 10,000 inhabitants, the relevant persons were questioned in a written survey (survey to clarify the residence) to determine their actual housing status.
Complete enumeration at special-facility addresses
At special-facility addresses, which include for example residential establishments and old people's homes, all people living there were questioned. Then the persons determined in the survey were compared with the population register data to rule out possible double counting and to clearly identify the residence to be counted.
All these measures allow to correct the personal data records of the population registers and to determine the official numbers of inhabitants.
The corrections are used only to determine the numbers of inhabitants. No data will be transferred back to the population registers. This was prohibited by the Federal Constitutional Court in its population census judgment of 1983 where it defined the prohibition to transfer data back.
