Metadata for Official Statistics

First steps in the data analysis

These instructions for the preparation of microcensus scientific use files is addressed to interested “newcomers”. By means of a number of questions, attention is drawn to points that are central to the data analysis and are, in part, microcensus-specific.

Although the compilation is not exhaustive, it draws attention to issues that are of importance when working with the microcensus scientific use files.

  1. What is the population of interest?
    In the case of the microcensus, it must be borne in mind that both persons in private households and persons in non-institutional and institutional collective dwellings are surveyed. Non-institutional and institutional collective dwellings can be student halls, hospitals, or prisons. [Population: private households] Moreover, the microcensus is an area sample, which may lead to people being surveyed twice, namely at their principal and their secondary place of residence. [Population: Principal and secondary place of residence] Depending on the question, it may be appropriate to restrict the population to persons at their family residence. [Population: family residence]
  2. Should weighting be applied?
    In the case of the microcensuses prior to 1989, it is necessary to consider how duplicated persons or households in the data set should be handled. [Adjustment duplication/deletion]
  3. How should Überhänge (i.e., households who should have been interviewed in the previous year but who were not interviewed until the following year using the previous year’s questionnaire) be handled?
    In the case of the microcensuses from 2006 onwards, it is necessary to consider how Überhänge from the previous year should be handled. Such Überhänge arise as a result of the changeover to quarterly surveys in 2005. Respondents who cannot be reached in the selected quarter (usually the last quarter) are sometimes interviewed in the first quarter of the following year. As a result of this delay in conducting the interview, it can happen that households are surveyed twice in one year [Jahresüberhang from the previous year]..