The Socio-Economic Panel (SOEP) is a representative, multi-cohort survey that has been running since 1984. Every year, individuals in households throughout Germany are surveyed by our survey institute on behalf of DIW Berlin. These respondents provide information on topics such as their income, employment history, education, and health. Because the same people are surveyed every year, it is possible to track long-term psychological, economic, societal, and social developments. To keep pace with changes in society, random samples are added regularly and the survey is adapted accordingly.
Title: Socio-Economic Panel (SOEP), data from 1984-2019, EU Edition
DOI : 10.5684/soep.core.v36eu
Collection period: 1984-2019
Publication date: 2021-03-31
Principal investigators: Stefan Liebig, Jan Goebel, Markus Grabka, Carsten Schröder, Sabine Zinn, Charlotte Bartels, Alexandra Fedorets, Andreas Franken, Martin Gerike, Florian Griese, Jannes Jacobsen, Selin Kara, Johannes König, Peter Krause, Hannes Kröger, Elisabeth Liebau, Maria Metzing, Jana Nebelin, Marvin Petrenz, David Richter, Paul Schmelzer, Christian Schmitt, Jürgen Schupp, Daniel Schnitzlein, Rainer Siegers, Hans Walter Steinhauer, Knut Wenzig, Stefan Zimmermann
Contributor: Kantar Deutschland GmbH (Data Collector)
Population: Persons living in private households in Germany
Amount of households: 19.032
Amount of individuals: 32.050 + 3476 Children
Special samples: Citizens of the GDR (1990), Immigration/Migration (1994/95, 2013, 2015), Refugees (since 2016). See the chapter SOEP-Samples in Detail on the SOEPcompanion for a description of all our samples.
Selection method: All samples of SOEP are multi-stage random samples which are regionally clustered. The respondents (households) are selected by random-walk or register sample.
Collection Mode:The interview methodology of the SOEP is based on a set of pre-tested questionnaires for households and individuals. Principally an interviewer tries to obtain face-to-face interviews with all members of a given survey household aged 12 years and over. Additionally one person (head of household) is asked to answer a household related questionnaire covering information on housing, housing costs, and different sources of income. This covers also some questions on children in the household up to 12 years of age, mainly concerning attendance at institutions (kindergarten, elementary school, etc.)
Citation of the Data Set: Socio-Economic Panel (SOEP), data for years 1984-2019, SOEP-Core v36, EU Edition, 2021, doi:10.5684/soep.core.v36eu
Publications using this file should refer to the above DOI Find an explanation on the usage of DOI here.and cite one of the following references
For the SOEP-Core data 1984-2019 (v36) - waves A to BJ - we provide the following editions:
soep.core.v36eu (EU Edition, 100%)
soep.core.v36i (International Scientific Use Version, 95%)
soep.core.v36t (Teaching Edition, 50%)
soep.core.v36at (Add-on: Area types)
soep.core.v36pr (Add-on: Planning regions)
soep,core.v36r (Remote Edition)
soep.core.v36o (Onsite Edition)
Find detailed information on the SOEPcompanion.
New Sample P
“Top Shareholder Sample”: Sample P was conceptualized as a sample of highly affluent households in Germany. Against the backdrop of the increasing income and wealth inequality in Germany over recent decades, despite economic growth, there has been a growing need for data on wealthy populations in the social sciences. Sample P was created to improve the empirical base for the German government’s poverty and wealth report and to lay the foundation for medium- and long-term cross-sectional and longitudinal analysis. The gross sample consisted of 23,259 households.
New Sample Q
“LGB* Sample”: Sample Q is a boost sample of a hard-to-survey population: lesbians, gays, bisexuals, transgender people, and those who identify as non-binary. While the actual percentage of LGBTQ+ people in the general population is unknown, this group was too scarcely represented in the SOEP to allow for meaningful analysis. 835 households were recruited through an approximately 9-month-long telephone screening process. Of these households, 477 participated in the survey between April and November.
Dataset BIOL - Variables on recognition of occupational qualifications in samples M3-M5 and the CAMCES module (identifiable in the variable label by the abbreviation AA/AAC) were corrected. The slightly different biographical questionnaire for samples M1-M2 is no longer used, and variables on migration history have been added to the SOEP-Core biographical questionnaire, which is now used for all samples. The variables have been integrated, versioned, and harmonized in biol accordingly. The religious affiliation of the father and mother has been reversioned and harmonized to include the response option “konfessionslos” (no religious affiliation). Additional variables with occupational codes have been added. Some variables at the federal state level were included as East-West variables with the suffix _ew. Since bioresid and biosoc will no longer be part of the data distribution, some data processing steps for the variables in these datasets have been included in the versioning and harmonization routines for biol.
Dataset PL - Variables on balance of assets (identifiable in variable label by abbreviation VB) have been corrected, re-sorted, and labeled. Inheritance variables have been re-versioned. Religious background has been re-versioned and harmonized to include the response option “konfessionslos” (no religious affiliation).
Dataset HBRUTTO - Some regional variables at the federal state level have been included as East-West variables with the suffix _ew. New variables on incentive type, incentive model, and variables describing screening process for the LGB sample have been added. Residential environment variables (wum) will no longer be included in the survey starting in 2019.
Dataset PBRUTTO - Some regional variables at the federal state level have been included as East-West variables with the suffix _ew. New variables on DRV record linkage and IAB record linkage have been added. Variables have been added indicating which questionnaire was used.
Dataset JUGENDL - Since bioage17 will no longer be part of the data distribution, some of the data processing steps for bioage17 variables have been included in the versioning and harmonization routines for jugendl.
Dataset KIDLONG - k_nrkid has been corrected to count only 16-year-old children in the household. Households with children without a stated birth year have been assigned a missing value of -1. - bgk93_r/kd_cty_r included incorrect values and has been corrected.
Dataset PLUECKEL - lpid has been removed.
Dataset HBRUTT - Some regional variables at the federal state level have been included as East-West variables with the suffix _ew.
Due to changes in data protection and privacy law, variables containing information on Germany’s federal states (Bundesländer) may not be transmitted to recipients outside the European Union. We have developed a new concept with different editions for the different data access procedures resulting from the change in law (listed in ascending order by the amount of information contained in each edition):
The default edition that we transmit to European users by sending them a personalized download link is the EU Edition. Some datasets may not be available in more restricted editions. If variables are not available in a more restricted edition, they are recoded to -7, a new missing value labeled “only available in less restricted edition”.
Originally, BIOIMMIG was generated by appending each new wave of data to the data from the previous years. This practice bears potential for errors since the SOEP includes a large number of variables that need to be comparable over time. In order to minimize this potential for error, v36 of BIOIMMIG is the first version of the dataset that has been generated using longitudinal data.
Variable bireason “Main reason for moving to Germany”
Variable biscger “Attended school in Germany”
There has been a significant increase in cases due to the addition of further variables and the new migration questions from the biographical questionnaire. The variable “country of last school attendance” was only included as an indicator from wave bd (2013) onwards to generate the variable biscger. The corresponding long variable, lb0186_v1, also has values for the years 2001-2012.
Variable bicamp “Refugee residence Y, N”
Due to two newly added variables (lr3440, plm0679), there are significantly more cases with a value of "2 No".
Variables birelh[p|gp|c|sb|sh|dr|fr] “Family in Country of Origin”
Previously, in the generation process, it was not defined whether these variables should represent country of origin or country of residence or both. In the previous versions, the two were arbitrarily merged. The decision to include only the country of origin leads to a significant reduction of cases.
Variable birelhc2 "Underage children not in Germany”
Due to two newly added years (1997, 1999), there are significantly more cases with the value "2 No".
Variable biwfam "Already had family in country”
The variable biwfam was changed from a category (y/n) to a binary variable, since otherwise there would be a distortion of the content. The data generated previously were too imprecise, since the recoding to value “2 No” does not unambiguously exclude cases in which respondents have no family members in Germany.
For a closer look at the changes and variables used, see the bioimmig documentation.
Variable hgtyp2hh in hgen
Datasets BIOSOC, BIORESID, BIOAGE17
The datasets biosoc, bioage17, and bioresid will no longer be provided. Most of the information from biosoc and bioresid will be maintained in the biol dataset with different variable names. In jugendl, the variables from bioage17 are retained. In order to reduce the number of datasets and to avoid redundant information, we decided to include the variables from biosoc and bioresid in biol and bioage17 in jugendl. The generated data from biosoc, bioage17, and bioresid are reproduced in the best possible way in biol and jugendl by applying versioning and harmonization. Users who have used biosoc, bioage17, or bioresid should use this table to facilitate transition.
The datasets with the suffixes mig and refugees—for instance, bep_mig and bgp_refugees—are no longer available. This information from the migration and refugee samples is fully integrated into the associated “raw” and “long” files.
Individual (PAPI) 2019: Field-de Field-en
Household (PAPI) 2019: Field-de Field-en
Biography (PAPI) 2019: Field-de Field-en
Catch-up Individual (PAPI) 2019: Field-de
Youth (16-17-year-olds, PAPI) 2019: Field-de
Early Youth (13-14-year-olds, PAPI) 2019: Field-de
Pre-teen (11-12-year-olds, PAPI) 2019: Field-de
Mother and Child (Newborns, PAPI) 2019: Field-de
Mother and Child (Newborns, CAPI) 2019: Var-de
Mother and Child (2-3-year-olds, PAPI) 2019: Field-de
Mother and Child (2-3-year-olds, CAPI) 2019: Var-de
Mother and Child (5-6-year-olds, PAPI) 2019: Field-de
Mother and Child (5-6-year-olds, CAPI) 2019: Var-de
Parents and Child (7-8-year-olds, PAPI) 2019: Field-de
Parents and Child (7-8-year-olds, CAPI) 2019: Var-de
Mother and Child (9-10-year-olds, PAPI) 2019: Field-de
Deceased Individual (PAPI) 2019: Field-de
Please find all sample specific questionnaires of this year and all questionnaires of previous years on this site
15) Die Vercodung der offenen Angaben zur beruflichen Tätigkeit nach der International Standard Classification of Occupations 2008 (ISCO08) - Direktvercodung - Vorgehensweise und Entscheidungsregeln bei nicht eindeutigen Angaben
All documentation for filtering can be found on this page