Page 16 - Invited Paper Session (IPS) - Volume 1

IPS30 Stefan D. et al.
was collected using a complete enumeration. Addresses carrying stigmatising
information, e.g. in the case of psychiatric hospitals or prisons (“confidential
special facilities”), were distinguished from non-confidential special facilities,
e.g. student dormitories. In confidential special facilities, the privacy of data
collection was secured by a special procedure and only a reduced set of
variables was collected.
• As there are no registers of buildings and dwellings covering the whole of
Germany, the variables of the housing census needed to be obtained through
a postal survey of buildings and dwellings that was conducted among all
property owners (for the total of just under 20 million buildings with residential
space, data were collected at approximately 19 million owners). In addition,
the census of buildings and housing covers auxiliary variables (number of
persons living in a dwelling and names of two persons) which were used in the
household generation procedure (see below).
• Information on the employment of the population has been taken from
registers of the Federal Employment Agency (for about 36 million employees
subject to social insurance contributions) and from the administrative files of
the public service agencies with personnel (for about 3 million public officials,
judges and soldiers). These registers were equally used to supplement the
demographic information obtained from the population registers, the
household sample survey and the survey of addresses with non-confidential
special facilities. Together with the register of addresses and buildings, this
information constituted the reference data stock.
• To obtain information about household and family structures and their
housing conditions (that information is not included in registers) data from
the various census components had to be combined in a so-called household
generating procedure. In this multistage procedure, information about
persons from the population registers, the household sample survey and the
survey conducted at special facilities was used to form households and to link
2
them to dwellings collected in the housing census.
Merging data sets from different sources for individual persons was one
of the great challenges of the 2011 census, because it had to be accomplished
without an existing personal identification number available in the different
registers. An already existing set of ID numbers for the purpose of the tax
authorities was available in some of the registers, but could not be used due
to legal restrictions. Therefore individual and address-based information such
as name, sex, date of birth, municipal code, post code, street name, and house
number were used to link respective records of different data sets.

2 For further details, see
https://www.zensus2011.de/EN/2011Census/Methodology/Methodology_household_generati
ng_procedure_node.html
5 | I S I W S C 2 0 1 9