We started from a simple idea: We wanted to create fake email accounts with typical foreign or native names. These fake aliases would contact coaches of amateur football (soccer) clubs in Switzerland and ask if they could come for a trial practice. The results could, potentially, show the level of social integration for various groups in Switzerland.
But whenever we tried to convince ourselves or others that this idea was worth the effort, we were scared off by the enormous amount of initial work and the potentially worthless results.
Why would someone not be allowed to attend a trial practice because of their name? Sure, in other fields (e.g., labor economics or housing), researchers frequently find discrimination. But we wanted to examine amateur football clubs! Coaches and clubs benefit from every additional player who is willing to be part of the team — what could be the harm of a player with a foreign name? Most colleagues, friends and family members we talked to were convinced that the results would be uninteresting: there would be no difference between foreigners and natives.
If this was not enough, we knew that the data gathering part would take ages. Construct names for typical foreign-sounding names for the three largest Swiss language groups (German, French, and Italian), get official translations of the application text, apply for an ethical approval, and, of course, find the email of each coach.
But we were convinced that the effort was worth the cost. Amateur football clubs are an important part of everyday social life. If people with foreign-sounding names suffer from discrimination in their first step to join a society — joining a social group such as a football club — how can one assume that they would fare any better when entering the labor market, asking for a loan, or renting an apartment?
Figure 1: Amateur football clubs and their primary language in Switzerland.
After years of doubt, we applied for a grant and were lucky to receive generous funding from the University of Zurich. Finally we conducted the experiment.
Our research group still remembers the day when we sent out the first emails. We were all sitting in front of our computers and waiting to see if the experiment actually worked. It did.
The results show that not only people with foreign-sounding names receive statistically significantly fewer responses, but also that native groups in certain regions (e.g., French-sounding names in the German-speaking part of Switzerland) receive fewer responses.
You can find the paper here: