Target Population and Samples¶
The target population covered in the SOEP is defined as the population of private households residing within the current boundaries of the Federal Republic of Germany (FRG). Because of changes in these boundaries (in 1990) and changes in the population due to migration, various adaptations have been made to the initial sampling structure to maintain the sample’s representativity. In addition, certain groups have been oversampled to increase the statistical power.
The different SOEP-Core subsamples constitute the centerpiece of the SOEP.
In 1984, the survey started with a sample covering the entire population of then West Germany (FRG), where the five biggest groups of foreigners (“guest workers”) were oversampled.
The SOEP was expanded to the territory of the German Democratic Republic in June 1990, only six months after the fall of the Berlin Wall. In 1994/95, a boost sample of migrants who came to Germany after 1984 was added to take the influx of ethnic Germans from former Soviet countries into account. In 2013 another sample of migrants which includes individuals who immigrated to Germany after 1995 or second-generation immigrants was added. Since then, multiple migration or refugees samples were added in cooperation with the IAB (Institut für Arbeitsmarkt- und Berufsforschung) or the BAMF (Bundesamt für Migration und Flüchtlinge)
Now and then samples that were representative of the entire population in Germany were added to counter effects of panel attrition and to increase the overall sample size.
The different samples in the SOEP are identified by letters: sample “A” refers to the German sample drawn in 1984, “C” to the East Germans from 1990, and so on. Even though these samples are kept separate, the respondents have received identical questionnaires for the most part, and distinctions by sample are usually not necessary in an analysis.
However, one of the ideas of the SOEP is that the users have full information available about survey methodological issues and survey design, which in this case means that you can identify the corresponding sample for each observation. In the following section, we present details on each of the samples, which unless stated otherwise are multi-stage random samples with regional clusters. The households are selected by random-walk routines.
For an extensive discussion on sampling (and weighting), see: Survey methods.