Dataset for field experiments analyzing discrimination in amateur soccer

This paper presents data of field experiments that analyze discrimination in amateur soccer. The studies created fake accounts and asked amateur soccer coaches to come for a trial practice. The fake accounts had either a native- or a foreign-sounding name. The dataset is based on three published studies that analyzed discrimination in 23 countries. The dataset contains 24,915 observations and several variables that are interesting for further research. This data can be used to compare discrimination in amateur soccer with discrimination in diverse fields, such as migration, economics, or political science. For a more detailed discussion, please see the published articles.


Value of the Data
• The dataset provides future research with the possibility to analyze discrimination in amateur soccer and examine within-and between-country differences • The dataset contains response rates to inquires in amateur soccer, which could be compared with discrimination outcomes in other social domains (e.g., housing, labor, or sharing economy) • The dataset shows specific foreign groups that suffer from discrimination (e.g., people with a Turkish-sounding name in Germany or with a German-sounding name in Denmark). This makes it possible to compare the extent of discrimination of specific foreign groups in different countries. • The dataset contains information to validate or contradict the usefulness of covariates that are thought to have an influence on discrimination (e.g., share of foreign-born population or share of right-wing voters) • Researchers in similar field experiments can append their data.
• Educators can use the dataset to give students a broad understanding of field experiments.

Data Description
The dataset presented in this paper combines the dataset of three published papers. All papers analyze discrimination in amateur soccer from a similar perspective [1] examine discrimination towards people with either a Swiss-French, Swiss-German, Swiss-Italian, Eastern European, Turkish, or Arabian-sounding name. The authors contact amateur soccer clubs in Switzerland with fake accounts using the following email (   The authors then categorize the response of the coach into negative, positive, or positive with additional inquiries. This is captured in the variable Response. All variables are listed in Table 1 . Afterwards, the authors simplify the variable into either a positive response or a negative response (see variable SimpleResponse ).
Dietl et al. [2] replicate the study of [1] . However, they contact a subsample of the amateur soccer clubs that did not respond to the first email again. They call this process "resilience" (the variable has the same name).
Gomez-Gonzalez et al. [3] extend the study of [1] and reach amateur soccer clubs in 22 European countries. They contact the clubs in each country with either a native-sounding name or a name from one of the three largest foreign-sounding groups. The countries for all three studies are listed in Table 2 .
The variables ForeignGroup1, ForeignGroup2, ForeignGroup3 show the country of origin for the foreign group for each country (e.g., Ukrainian-, Vietnamese-, and Russian-sounding names in the Czech Republic).

Experimental Design, Materials and Methods
The initial data was gathered through three field experiments. Parts of the data were made public after the publication of each study. We received the missing data from the authors of the studies. However, still not all studies had the information for the same variables. For example, the data for the original unmodified Response is not available for [2] . Additionally, this data is missing for a few countries in [3] (see footnote for Table 1 ). The simplified response rate is available for all countries (see Fig. 1 ).
We then labeled and named each variable in each study in accordance with each other. It is important to note that the foreign groups for each country are specifically listed in the dataset but also available in a numeric format. This ensures that future research can focus on specific foreign groups but also on specific countries. This is similar for the variable Sender . While the variable Native shows if a request was sent from a native-or foreign-sounding name, Sender shows the nationality of the person that sent the request.

Ethics Statement
All experiment received an ethical approval from the University of Zurich. [1] had the approval date 14.07.2017 (no IRB number available), [2] had the IRB #2019-053, and [3] had the IRB #2019-006.

CRediT Author Statement
All authors contributed equally to this study and share the responsibility for its content.

Declaration of Competing Interest
The authors declare no competing interests.

Acknowledgments
This work was financially supported by the Swiss National Science Foundation (SNSF), grant no. CRSK-1_190264 and Stiftung für wissenschaftliche Forschung an der Universität Zürich.

Supplementary Materials
Supplementary material associated with this article can be found, in the online version, at doi: 10.1016/j.dib.2021.107751 .