https://orcid.org/0000-0001-8610-4828
ABSTRACT
In this paper, we examine if a slight behavioural modification has an influence on discrimination. Specifically, we use a field experiment based on emails to observe if persistence increases the chances of joining a social group. We find that native- or foreign-sounding names are equally successful when being persistent. However, non-persistent individuals with foreign-sounding names suffer from discrimination. The results show that persistent individuals can be especially interesting in the context of field experiments.
I. Introduction
Contacting someone for the first time is part of daily life. Applying for a job, asking to rent an apartment, using the sharing economy, or trying to join a social group are all situations in which people often do not know each other beforehand. Research finds that certain groups frequently suffer from discrimination in these circumstances.
For example, African Americans in the US receive fewer responses when applying for a job (Bertrand and Mullainathan Citation2004), when asking to look at an apartment (Hanson and Hawley Citation2011), or when trying to book an AirBnB apartment (Edelman, Luca, and Svirsky Citation2017) or an Uber ride (Ge et al. Citation2016). However, this is not exclusively the case for African Americans but also for minorities in numerous other countries. For example, Turkish-sounding names receive fewer call backs in the German labour market (Kaas and Manger Citation2012), as do homosexual couples in the Swedish housing market (Ahmed and Hammarstedt Citation2009), or foreigners when trying to join a social activity in Switzerland (Nesseler, Gomez-Gonzalez, and Dietl Citation2019).
In the majority of these field experiments, applicants experience a one-shot attempt. After a first contact, a no response or a negative response is evaluated as a rejection. However, in more realistic scenarios some applicants could be persistent, i.e. trying again after the first failure. In scenarios of imperfect information, a reminder email may be perceived as a signal of effort and commitment (Connelly et al. Citation2011).
The influence of persistence is largely overlooked in this type of experiment. Only studies interested in increasing survey response rate report a positive influence of reminder messages, e.g. Misra, Stokols, and Marino (Citation2012). For this purpose, we extend the study by Nesseler, Gomez-Gonzalez, and Dietl (Citation2019), in which the authors measure access to social activities in Switzerland, by including the influence of persistence. Nesseler et al.’s study is ideal because it is recent, the data is publicly available, and social integration is especially interesting in Switzerland.Footnote1
II. Experimental design and methods
We constructed a list of names for each groupFootnote2 ; Swiss-French, Swiss-German, Swiss-Italian, Serbo-Croatian, Turkish, and Arabic. We explicitly chose these foreign groups as they represent a sizable part of the foreign-born population in Switzerland (9.6%, 3.9% and 1.8%, respectively).Footnote3 We cannot include other foreign groups such as French, German, and Italian because the official languages in Switzerland prevent Swiss citizens from identifying names from these groups as foreign names. To ensure that we have a representative sample we generated each name based on the frequency of their first- and last-name. A table with the full list of first- and last-names is available in the appendix (Table A1).Footnote4
Regarding email profiles, we generated random eight-digit gmail.com accounts. The receiver could see the sender’s full name not only in the signature but also in the account from which the email was sent. While other email providers are also popular in Switzerland, we choose Gmail because these emails have a high probability of not being labelled as spam.Footnote5
We gathered the data of 607 amateur football clubs in Switzerland. We randomly selected one team for every club that had more than one team. We contacted the coach of a team. That ensured that the email was not lost between administrative personnel.
Similar to Nesseler, Gomez-Gonzalez, and Dietl (Citation2019), we sent the email in the respective language of the club. We used official translations for the text in French, German, and Italian (the versions of these emails are included in the appendix, -A4). The email read as follows:
Subject: Trial practice
Hello,
I am looking for a team that I can join. I think that I have already played on a similar level. Could I come for a trial training session?
Many thanks
Name 1-30
We sent out all emails on 3 February 2020 before noon. We categorized all responses into four categories: no response, negative response, positive response, and positive response with further inquiries. A negative response meant that the sender was not invited to join a training session. A positive response meant that the sender was invited for a training session. A positive response with further inquiries meant that respondents wanted to know more (e.g. age or playing position). Once we received a response, we waited between 24 and 48 hours and wrote that the applicant was no longer interested in participating in a training session.Footnote6 We wrote this response email in accordance with the guidelines of the ethical committee of the University of Zurich.
We waited one week for clubs to answer. Among the clubs that did not respond, we randomly selected 100 clubs to send a second email to. The ethical committee of the University of Zurich did not allow us to send more than 100 new emails as being persistent could potentially reinforce negative stereotypes directed at foreigners. The reminder included 50 native-sounding names and 50 foreign-sounding names that were randomly assigned to clubs. In the reminder email we added that the sender is looking for a club in the region. Thus, clubs would have had the chance to respond to the second mail without necessarily inviting applicants to a trial practice. This opened a fifth response possibility: forwarding the applicant to another club. However, no club chose this option. The email reads as follows:
Subject: Trial practice
Hello,
I am still looking for a team in Region XY. Could I come for a trial training session?
Many thanks
Name 1-30
provides an overview of the covariates used in the analysis. We distinguish between the first and the second (persistence) email. To control for external factors that could have an influence on the decision to respond, we use the following control variables: club has a youth section, foreigner rate, total inhabitants, right wing votes, and number of other amateur football clubs in the area. Here, area is defined by the first three digits of the postcode.Footnote7
Table 1. Summary statistics
III. Results
We sent out 310 emails with native-sounding names and 297 with foreign-sounding names. Native-sounding names received a positive response in 60.9% of the cases, foreign-sounding names in 51.3% of the cases. These results are not significantly different from the findings of Nesseler, Gomez-Gonzalez, and Dietl (Citation2019).Footnote8
The results from show that foreign-sounding names receive significantly fewer responses.Footnote9 While the covariates influence the response rate, e.g. a higher rate of foreigners decreases the chance to receive a response, foreign-sounding names consistently receive fewer responses. Models 3 and 4 show that this is mainly due to Arabic-sounding names.
Table 2. Regression results. First email to clubs
After sending out 100 reminder emails, native-sounding names received a positive response in 30.0% of the cases and foreign-sounding names in 32.0% of the cases. The response rate for the reminder emails (31.0%) is significantly lower than for the first emails (56.0%) and does not show differences by name origin. shows that neither foreign-sounding names nor Arabic-sounding names received fewer responses.Footnote10
Table 3. Regression results. Persistence (second) email to clubs
Different mechanisms may explain the positive role of persistence. One possibility is that the observed discrimination (after the first email) responds to a problem of imperfect information and perceived distance (Eriksen Citation2002), which is reduced with the reminder email (if understood as a signal of commitment and interest by the receiver). However, more research is needed to clarify the implicit meaning of the signal in this context (Connelly et al. Citation2011). Other factors related to omitted variables or receiver selection bias are unlikely to explain the difference after the second email with the randomized experimental design (Gaddis Citation2018).
IV. Conclusion
The results show that individuals with foreign-sounding names suffer from discrimination when trying to integrate. The magnitude is similar to previous research examining social integration in Switzerland (Nesseler, Gomez-Gonzalez, and Dietl Citation2019). Additionally, we find that a reminder email is perceived as a positive signal (Connelly et al. Citation2011), which increases the chances of receiving a response to join a club. This result is especially interesting for field experiments, which have largely overlooked the influence of persistence. Our study highlights the empirical significance of persistence.
Disclosure statement
No potential conflict of interest was reported by the authors.
Correction Statement
This article has been republished with minor changes. These changes do not impact the academic content of the article.
Additional information
Funding
Notes
1 Switzerland has the highest share of foreigners for every European country with more than 1 million inhabitants.
2 Switzerland has three major language groups: French, German, and Italian. Romansh, the fourth recognized language, is spoken by less than 1% of the population.
3 We used the numbers provided by Eurostat. The numbers do not include naturalized citizens or foreign-born citizens younger than 15. We included Serbia, Croatia, and Bosnia Herzegovina as countries with Serbo-Croatian sounding names; Turkey with Turkish sounding names; Algeria, Iraq, Jordan, Lebanon, Morocco, Libya, Somalia, Sudan, Syria, Tunisia, and Yemen as Arabic sounding names. Eurostat does not provide data for several other foreign groups in Switzerland that have Arabic sounding names (e.g. Bahrain, Oman, Qatar, or Saudi Arabia). Thus, the percentage of foreigners with Arabic sounding names might be higher.
4 To confirm the validity of the names we performed surveys at various public places in Switzerland and asked respondents whether the names sounded native or foreign. The majority of the respondents was able to correctly categorize the names.
5 Prior to the experiment start, we sent test emails and verified that these were not labelled as spam.
6 The 24–48 h window aimed to prevent receivers that responded to emails from getting upset with an immediate negative rejection from the sender while minimizing the uncertainty generated by the experiment.
7 Postcodes have four digits in Switzerland. Every additional digit separates the region of the previous digit in several parts.
8 The public dataset provided by Nesseler, Gomez-Gonzalez, and Dietl (Citation2019) is available in: https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/PZLYUN. This allows us to perform a Mann-Whitney two-sample analysis. The results show that neither the results for foreign-sounding names (n = 616; z = −0.897; Probability > |z| = 0.3695) nor the results for native-sounding names (n = 1,179; z = −0.398; Probability > |z| = 0.6908) are significantly different.
9 Using a logit regression with marginal effects yields the same results.
10 We do not include control variables in Table 3 as the results would suffer from overfitting.
References
- Ahmed, A. M., and M. Hammarstedt. 2009. “Detecting Discrimination against Homosexuals: Evidence from a Field Experiment on the Internet.” Economica 76 (303): 588–597. doi:10.1111/j.1468-0335.2008.00692.x.
- Bertrand, M., and S. Mullainathan. 2004. “Are Emily and Greg More Employable than Lakisha and Jamal? A Field Experiment on Labor Market Discrimination.” American Economic Review 94 (4): 991–1013. doi:10.1257/0002828042002561.
- Connelly, B. L., S. T. Certo, R. D. Ireland, and C. R. Reutzel. 2011. “Signaling Theory: A Review and Assessment.” Journal of Management 37 (1): 39–67. doi:10.1177/0149206310388419.
- Edelman, B., M. Luca, and D. Svirsky. 2017. “Racial Discrimination in the Sharing Economy: Evidence from a Field Experiment.” American Economic Journal: Applied Economics 9 (2): 1–22. doi:10.1257/app.20160213.
- Eriksen, T. H. 2002. Ethnicity and Nationalism: Anthropological Perspectives. London, UK: Pluto Press.
- Gaddis, S. M. 2018. Audit Studies: Behind the Scenes with Theory, Method, and Nuance. Cham, Switzerland: Springer International Publishing.
- Ge, Y., C. R. Knittel, D. MacKenzie, and S. Zoepf. 2016. “Racial and Gender Discrimination in Transportation Network Companies.” National Bureau of Economic Research. Working Paper No. w22776. doi:10.3386/w22776
- Hanson, A., and Z. Hawley. 2011. “Do Landlords Discriminate in the Rental Housing Market? Evidence from an Internet Field Experiment in US Cities.” Journal of Urban Economics 70 (23): 99–114. doi:10.1016/j.jue.2011.02.003.
- Kaas, L., and C. Manger. 2012. “Ethnic Discrimination in Germany’s Labour Market: A Field Experiment.” German Economic Review 13 (1): 1–20. doi:10.1111/j.1468-0475.2011.00538.x.
- Misra, S., D. Stokols, and A. H. Marino. 2012. “Using Norm–Based Appeals to Increase Response Rates in Evaluation Research: A Field Experiment.” American Journal of Evaluation 33 (1): 88–98. doi:10.1177/1098214011414862.
- Nesseler, C., C. Gomez-Gonzalez, and H. Dietl. 2019. “What’s in a Name? Measuring Access to Social Activities with a Field Experiment.” Palgrave Communications 5 (160): 1–7. doi:10.1057/s41599-019-0372-0.
Appendix
Data availability
The data for this study is available at https://doi.org/10.7910/DVN/7XVFFJ
Table A1. Names used
Table A2. Content of the emails in German
Table A3. Content of the emails in French
Table A4. Content of the emails in Italian