Fairness Measures

Datasets and software for detecting algorithmic discrimination



Ranking Algorithms

Classification Algorithms



Code (on GitHub) 🔗



DIW SOEP Data (Socio-economic Panel), Germany

The German Socio-Economic Panel (SOEP) is a wide-ranging representative longitudinal study of private households, located at the German Institute for Economic Research, DIW Berlin. Every year, there were nearly 11,000 households, and about 30,000 persons sampled by the fieldwork organization Kantar Public Germany.

The data provide information on all household members, consisting of Germans living in the Old and New German States, Foreigners, and recent Immigrants to Germany. The Panel was started in 1984 (access to the SOEP data).

Some of the many topics include household composition, occupational biographies, employment, earnings, health and satisfaction indicators.


Data Types: Multivariate (Panel, multiple data sets)

Attribute Types: Categorical, Integer, Real

Format: ASCII (csv), Stata, SPSS, SAS (German/English versions available)

Number of Entries: depending on wave (year)

Range: 1984 - 2016 (as of 2017)

Number of Datasets: variable per year

Source: German Institute for Economic Research / Deutsches Institut für Wirtschaftsforschung (DIW)


Attributes: for a complete list of variables, please visit paneldata.org We recommend starting with the SOEP-Core Study to determine your own target variables

Some variables with special interest to fairness measures*:

* These Variables are not the only ones and are intended to provide only insight into the data set. Please consider also the relevant questionnaires. sample here

How to obtain this dataset?

This dataset is not available for public download. If you are interested in obtaining it, please get in touch with the dataset provider.