GESIS Leibniz Institute for the Social Sciences: Go to homepage
MISSY: Go to homepage
Metadata for Official Statistics

Expand / Collapse

Use the - and -icon left of the headlines to open and close the different sections of the content area.

The - and -icon open / close all sections at once.



Explanatory notes are shown when the mouse cursor is moved over the field headlines.



The myMetadata Box collects variables which can be used in the functions of the work with myMetadata section.

Add variables
via drag & drop into the box in the top right corner or by clicking the -icon which appears when you hover over the variables name or headline.

Add variable lists
On pages displaying variable lists you can add complete lists of variables as well by using the -icon which is placed next to the headline.

Remove a selected variable
with the -icon which appears on the right when hovering the variable in the list.


Study: MZ 2003

Additional Programs

  • Additional Programs
    • Health insurance (Sampling Fraction: 1% )
    • Need for long-term care, benefit take-up of the long-term care insurance (Sampling Fraction: 0.45% )
    • Vocational, general and further education (Sampling Fraction: 0.45% )
    • Complementary information on economic activity (Sampling Fraction: 0.45% )
    • Labour force participation one year before (Sampling Fraction: 0.45% )
    • Questions on health, diseases and accidental injuries, smoking habits (Sampling Fraction: 0.45% )
    • Previous economic activity (Sampling Fraction: 0.45% )

Specific Features

Methodical Notes

  • When constructing the derived variables EF657 to EF670, the characteristic "Spouse or partner of reference person in the household" mentioned in the variable label was obviously not used, but "Spouse or partner of reference person in the family/living arrangement " (EF514=2). The variables labels of the derived variables EF657 to EF670 were therefore changed to "(...) spouse or partner of reference person in the family/living arrangement ". (Label change: February 2019).

Year Specific Documentation

  • Microcensus Scientific Use File 2003 Documentation and Data Preparation [in German] [.pdf]
  • Comparison of the Microcensus Scientific Use Files 2002 and 2003 [in German] [.pdf]
  • Microcensus Scientific Use File 2004 A Comparison to Previous Years [in German] [.pdf]

Target Sample Size

Cross-sectional Data
Sampling Units Districts
Achieved Sample Size 502873


Units Of Observation

  • Persons (in private households and collective dwellings)
  • Households

Units Of Analysis

  • Persons
  • Living arrangements
  • Families
  • Households

Data Collection

Start Date


End Date


Participation Mandatory


Interview Mode

  • 7.3 % CAPI
  • 78 % PAPI
  • 14.7 % Self-administered

Percentage Of Proxy Interviews

25.82 %


Design Weight: Target

Dwelling, Household, Persons

Design Weight: Method

The Microcensus is a single-stage, stratified cluster sample with a uniform sampling fraction for all strata. As a rule, the sampling fraction is 1 %; it applies to all sampling units (dwellings, households, persons). Until 2004, by contrast, variables of the supplementary and additional programmes and the European Union Labour Force Surveys (EU-LFS) were implemented using smaller sampling fractions, which sometimes varied from stratum to stratum. For example, from 1996–2004 the national average was around 0.45% but the sampling fractions at administrative-district (Regierungsbezirk) level varied (0.4 %, 0.6 %, 0.8 %, or 1 %).

Because the selection probability of the Scientific Use File (SUF) basically remains constant at 70%, design weights can be created on the basis of the inverses of the sampling fractions of the Microcensus (1 %) and the Scientific Use File (70 %): w = 1 / (0.01 * 0.7).

However, for the supplementary and additional programmes, no information is available about the variable sampling fractions until 2004.

Non-response Weight: Method

In the Microcensus, as in all surveys, the sampling plan cannot be realised fully. Hence, undercoverage occurs in the form of non-response on the part of the households to be surveyed. Because households and their members are obliged by law to provide information for the Microcensus, non-response is due mainly to the fact that households could not be reached during the survey. Cases of non-response are corrected using the information that is available about the non-responding households.

From 1990 to 2004, 19 compensation classes were created at the level of regional subgroups (at least 100,000 inhabitants). These compensation classes were made up of a combination of the following variables: household size, nationality of the reference person of the household, right of occupancy of the reference person of the household (for Germans), or, in the case of single-person households, sex and age group. The compensation factor is the ratio of the households to be surveyed to the surveyed households (target figure divided by actual figure).

Coverage Adjustment Weight: Method

In contrast to compensation, adjustment is carried out at person level.

From 1990 to 2004, the sample results were adjusted to conform to benchmark figures from the intercensal population estimates at the level of the regional adjustment strata (at least 500,000 inhabitants on average) for the civilian population according to four classes: nationality (German, foreign national) x sex. The adjustment for regular soldiers/temporary career volunteers and persons doing military or community service is carried out at the level of the administrative districts of the federal states on the basis of inventory data of the Federal Ministry of Defence or the Federal Ministry of the Interior. The adjustment factor is the ratio of the target figure to the actual figure (i.e. target divided by actual), where the target figure is the intercensal population estimate and the actual figure is the Microcensus figure.

Final Weighting: Method

The final extrapolation factors result from the combination of the two-stage procedure of compensation and adjustment.

From 1990 to 2004, the final extrapolation factors were calculated by multiplying the household-related compensation factor by the person-related adjustment factor. In addition, a household factor was introduced that was calculated as the arithmetic mean of the person-related adjustment factors.

In the Scientific Use Files from 1989 or 1991 onwards, only the final extrapolation factors are included.

For the years 1996 to 2004, an extrapolation factor for persons is available in addition to the extrapolation factor at household level. Further extrapolation factors are available for the extrapolation of the variables of the European Union Labour Force Survey (a 0.45% subsample, on average) and additional programmes, some of which were conducted as a subsample with a variable sampling fraction of 0.45%, on average.

See [.pdf] (in German)