Do Swiss Citizens of Immigrant Origin Face Hiring Discrimination in the Labour Market?

Abstract

The extent to which discrimination in employment disadvantages children of immigrants is a major question both in economic research on labour market and in sociological studies of integration. This working paper contributes to the debate by reporting findings of a correspondence test in which pairs of equally qualified Swiss citizens – one from the majority group and one from one minority immigrant group – apply at job openings all over the country. Results show that children of
immigrants holding Swiss qualifications and dual nationality need to send 30% more applications to receive a call-back for an interview when applying for apprenticeship level occupations. Chances of dual citizens to be invited to a job interview are largely the same across linguistic regions. These results partially confirm the findings of a correspondence test conducted in 2003. Unequal treatment is less frequent in occupations requiring tertiary level education compared to occupations requiring a completed apprenticeship; this is however not the case for Swiss candidates of Kosovar descent. The results suggest the need for further inquiry in occupational differences and ethnic hierarchies in the Swiss labour market.

Full text

Working Paper #20
February, 2019
Eva Zschirnt and Rosita Fibbi
Do Swiss Citizens of
Immigrant Origin
Face Hiring Discrimination
in the Labour Market?

Eva Zschirnt
(European University Institute)
and Rosita Fibbi
(Swiss Forum for Migration
and Population Studies)
Do Swiss Citizens of
Immigrant Origin
Face Hiring Discrimination
in the Labour Market?
The Natio nal Cent res of Comp etence in Research (NCCR)
are a research instrument of the Swiss National Science Foundation
National Center of Competence in Research –
The Migration-Mobility Nexus
nccr-onthemove.ch
The nccr – on the move is the National Center of Competence in Research (NCCR) for migration and mobility
studies and aims to enhance the understanding of contemporar y phenomena related to migration and mobility in
Switzerland and beyond. Connecting disciplines, the NCCR brings together research from the social sciences,
economics and law. Managed from the University of Neuchâtel, the network comprises fourteen research projects
at ten universities in Switzerland: The Universities of Basel, Geneva, Lausanne, Lucerne, Neuchâtel, Zurich,
ETH Zurich, the Graduate Institute Geneva, the University of Applied Sciences and Arts of Western Switzerland,
and the University of Applied Sciences and Arts of Northwestern Switzerland.
The Working Papers Series is an online platform for academic debates by members and cooperating partners
of the nccr – on the move. The authors are responsible for the analyses and arguments, which do not necessarily
reflect those of the nccr – on the move.
nccr – on the move, University of Neuchâtel, Rue Abram-Louis-Breguet 2, 2000 Neuchâtel, Switzerland
Contact for the Working Paper Series: info@nccr-onthemove.ch

nccr – on the move, Working Paper #20
3
Abstract
The extent to which discrimination in employment disadvantages children of immigrants is a major
question both in economic research on labour market and in sociological studies of integration. This
working paper contributes to the debate by reporting findings of a correspondence test in which
pairs of equally qualified Swiss citizens – one from the majority group and one from one minority
immigrant group – apply at job openings all over the country. Results show that children of
immigrants holding Swiss qualifications and dual nationality need to send 30% more applications to
receive a call-back for an interview when applying for apprenticeship level occupations. Chances of
dual citizens to be invited to a job interview are largely the same across linguistic regions. These
results partially confirm the findings of a correspondence test conducted in 2003. Unequal treatment
is less frequent in occupations requiring tertiary level education compared to occupations requiring
a completed apprenticeship; this is however not the case for Swiss candidates of Kosovar descent.
The results suggest the need for further inquiry in occupational differences and ethnic hierarchies in
the Swiss labour market.
Keywords
Labour Market, Switzerland, Hiring Discrimination, Correspondence Test, Second Generation
Acknowledgments
This work was supported by the nccr – on the move, which is financed by the Swiss National
Science Foundation. The results of the German-speaking part come from Eva Zschirnt’s PhD
dissertation (Zschirnt 2018), and are now complemented with the results of the French-speaking
area. We would like to thank Robin Stünzi for his excellent research assistance in conducting the
experiment in the French-speaking cantons and Didier Ruedin for his precious advice and
competent support throughout the study.
Contacts
eva.zschirnt@eui.eu, rosita.fibbi@unine.ch
European University Institute, Department of Political and Social Sciences
Via dei Roccettini 9, 50014 San Domenico di Fiesole, Italy
University of Neuchâtel, Swiss Forum for Migration Studies (SFM)
Rue A.-L. Breguet 2, 2000 Neuchâtel, Switzerland
!

nccr – on the move, Working Paper #20
4
Contents
1!Introduction 5!
2!The Swiss Context 6!
3!Theoretical Background 6!
4!Methodology 7!
5!Designing a Correspondence Test in a Bilingual Country 8!
5.1 Choice of groups 8
5.2 Signalling Group Membership 9
5.3 Choice of Locations 10
5.4 Choice of Occupations 10
5.5 Identifying Vacancies 11
5.6 Application Materials 11
5.7 Research Ethics 13
6!Findings 14!
6.1 Do Swiss Citizens of Different Immigrant Origin Face Hiring Discrimination? 14
6.2 Are There Differences by Occupation? 15
6.3 Are There Differences by Gender (in Sales)? 16
6.4 Are There Regional Differences in Hiring Discrimination? 17
6.5 Is Discrimination Lower in Professions Requiring Tertiary Level Training? 19
7!Discussion 21!
7.1 Occupational Differences 21
7.2 Ethnic Hierarchies 22
7.3 Regional Differences 23
7.4 New Insights into Hiring Discrimination In Switzerland 24
7.5 Swiss Results in an International Perspective 25
7.6 Limitation of the Study 26
8!Summary and Conclusions 26!
9!Bibliography 29!
!
!Appendix 35!

nccr – on the move, Working Paper #20
5
1 Introduction
Ethnic discrimination in hiring decisions has been well-documented in many countries, with ethnic
minority jobseekers writing about 50% more applications before being invited for a job interview
than their majority competitors (Zschirnt and Ruedin 2016). However, it is not only first generation
immigrants that are affected by ethnic discrimination, but also their offspring. Meta-analysis of field
experiments on hiring discrimination, as well as research specifically addressing the difference in
discrimination experienced by the first and second generation, finds evidence that discrimination is
often due to the foreign name per se, ignoring the degree of integration of job applicants (Carlsson
2010; Zschirnt and Ruedin 2016). Yet, in societies that usually portray themselves as meritocratic
and maintain that hard work will be rewarded independent of ethnic or social origin, discrimination
against immigrants and their offspring is worrisome and poses a threat to social cohesion. It is
worthwhile studying this issue in an immigration country like Switzerland, where, in 2017, 37% of
the resident population is of an immigrant background (Bundesamt für Statistik 2018 c), 25% are
foreigners (Bundesamt für Statistik 2018 a) and 17% of Swiss nationals aged 15 and over residing
in the country hold dual citizenship1 (Bundesamt für Statistik 2018 b).
Since their emergence in the late 1960s, field experiments have become a well-established way to
study the extent of discrimination that minority job candidates encounter in the labour market.
These allow researchers to vary the important characteristic, for example by varying the name to
portray a different ethnic background. While such experiments have been conducted in most OECD
countries and often also in more than one study, the knowledge about labour market discrimination
on grounds of ethnicity is still scarce in Switzerland. So far, only Fibbi, Kaya and Piguet (2003)
have conducted a correspondence test in Switzerland, in which they studied youths born abroad but
who completed their education and apprenticeship in Switzerland and were now looking for a first
job. They found significant rates of discrimination against minority candidates that varied
depending on their ethnic origin2.
This paper provides information from a new correspondence test conducted in the two main
language regions of Switzerland3. It studies to what extent jobseekers with a non-Swiss name
encounter discrimination on the Swiss labour market. Looking at three groups (Swiss dual nationals
with German/French, Kosovar, or Turkish names), two regions (German- and French-speaking
Switzerland) and four professions (sales assistant, electrician, HR clerk, nurse), the study assesses
to what extent jobseekers with a non-Swiss name encounter hiring discrimination in the Swiss
labour market and whether discrimination varies depending on the ethnic background of a candidate
or the position he or she is applying for.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
1 Both statistical sources providing information on dual citizens, Swiss Labour Force Survey (SLFS) and Structural Survey, cover
persons aged 15 and over; therefore the figure of dual nationals underestimates their number among the resident population.
2 There is a second correspondence test in a Swiss study conducted by Diekmann et al. (2014). However, unlike the majority of
correspondence tests, the authors did not apply for advertised vacancies but sent out unsolicited applications.
3 The Italian-speaking region, representing about 4% of the resident population of the country, was excluded because of the small
number of jobs posted there and the high work load of preparing application material in a third language.

nccr – on the move, Working Paper #20
6
2 The Swiss Context
Switzerland makes for an interesting case study on the labour market discrimination of foreign
named applicants, due to the high number of immigrants and their offspring – in particular from
European Union or EFTA countries – living and working in the country. In fact, two-thirds of the
Swiss immigrant population holds passports of EU or EFTA countries, with Italy, Germany,
Portugal and France being the most important countries of origin, followed by countries such as
Kosovo, Macedonia, Serbia, Croatia and Bosnia that were part of the Former Yugoslavia, and
Turkey (Bundesamt für Statistik 2018d; OECD 2012). While employment rates and wages of
immigrants are high in international comparison, immigrants are still disadvantaged: their
unemployment rates are higher and salaries are lower than those of native Swiss (SECO 2018).
Furthermore, there is evidence that the second generation in particular faces discrimination in the
Swiss labour market, e.g. when looking for apprenticeships (Imdorf 2008). Such constraints on
foreign born residents’ social mobility can have long term repercussions for a meritocratic society
(Zschirnt, 2018).
It is also a case worth studying for a number of structural reasons: First, due to the federal structure
of the country, policy making is highly decentralised. Having four official languages further
amplifies the level of heterogeneity between the different Swiss regions. Second, the labour market
in Switzerland is characterised by high-female labour force participation rates, high salaries, and
unemployment rates that have been consistently low in international comparison. Third, and closely
related, the economy is very internationally oriented (OECD 2012). In addition to these structural
factors, Switzerland does not have a comprehensive anti-discrimination law. Since it is not part of
the European Union, it was not obliged to implement the EU’s anti-discrimination directives.
According to the results of the Migrant Integration Policy Index (MIPEX) in 2015, Switzerland is
“one of the very few countries without a comprehensive anti-discrimination law and equality body
with legal standing; a sizeable number of potential victims are poorly protected against racial,
ethnic, religious and nationality discrimination” (Huddleston et al. 2015, 40).
Furthermore, the only correspondence test data available on ethnic discrimination in the labour
market is already 15 years old and focused on the very specific situation of foreign born but Swiss
educated youths looking for their first position after completing their apprenticeship (Fibbi, Kaya
and Piguet 2003). Finally, a strong anti-immigration sentiment has been portrayed by the Swiss
media and politics (see e.g. Boulila 2018), which raises the question to what extend these anti-
immigration attitudes are reflected in the hiring decisions of Swiss employers.
3 Theoretical Background
Studies of labour market discrimination have a long tradition, particularly in the field of economics.
In an early work on “The economics of discrimination” Gary Becker developed his theory of taste
based discrimination (Becker 1957). According to this theory, employers are prepared to choose
one candidate over another because they have a preference to work with someone of a specific
gender, race, or other characteristic. Employers are even prepared to incur a cost (e.g. pay higher
wages) to hire their preferred candidate and avoid working with a candidate they are averse to.
According to Becker, discrimination not only occurs when the employer has discriminatory tastes,

nccr – on the move, Working Paper #20
7
but also when he/she anticipates that other employees or customers will avoid such an employee. In
response to Becker’s theory, Phelps (1972), Arrow (1973), and Aigner and Cain (1977) proposed a
theory of statistical discrimination. They argue that, because of a scarcity of information about job
candidates, employers use group information as signals in their hiring decisions and will avoid
hiring members of a certain group, if they are expected “to be less qualified, reliable, long-term, etc.
[…] and if the cost of gaining information about the individual applicants is excessive” (Phelps
1972). Thus, following statistical discrimination theory, employers do not necessarily have a taste
for discrimination, but try to avoid uncertainties and losses in their selection decisions by focusing
on applicants’ group memberships. According to both of these theories discrimination should not be
a long term phenomenon, but should disappear or at least diminish over time if markets are
competitive (e.g. Darity and Mason 1998). However, despite such predictions, meta-analyses of
field experimental data have shown that ethnic discrimination in the labour market is very persistent
over time (Quillian et al. 2017; Zschirnt and Ruedin 2016). The debate over why discrimination
occurs and which theory is better suited to explain its occurrence is still ongoing.
4 Methodology
Field experiments have been deemed to be the “methodological gold standard” in discrimination
research (Protsch and Solga 2017), because they allow researchers to observe real employers’ hiring
decisions in the labour market, while these employers are not aware that they are part of an
experiment. They have proven particularly useful in the field of discrimination studies, where it
would often not be possible to obtain data of the same quality by using other research methods, e.g.
due to problems with social desirability biases. The field experiments quantify hiring discrimination
against minority groups (e.g. immigrants, women, sexual or religious minorities, etc.) by submitting
equally qualified fictitious applications to real advertised vacancies. Contact with employers can
take place in person (so-called audit studies) or in writing (so-called correspondence studies), and
the number of applications submitted to each employer can vary; what all experiments have in
common is that employers’ responses are carefully recorded and that the differences in invitation
rates for majority and minority candidates can then be attributed to discrimination.
Conducting field experiments on ethnic discrimination in hiring decisions has several advantages
compared to other methodologies (e.g. surveys or interviews with employers), but the methodology
also faces limitations. The biggest advantage (in particular in correspondence tests) is that they give
researchers complete control in creating the fictitious candidate profiles, which can be matched on
numerous criteria (e.g. gender, age, socio-economic status, family status, work experience, or
education) while providing sufficient variation to avoid detection by employers. However, field
experiments, and in particular audit studies, have been criticised for several reasons. Heckman and
Siegelman (1993) and Heckman (1998) criticise them on a methodological basis, for example on
the problem of unobserved variables: they doubt that researchers can really match fictitious
candidates and ensure that they are equivalent and that characteristics unobserved by the researchers
do not influence the observations. Furthermore, they caution that employers might be forced to
focus on very minute differences to break the tie between two very well qualified applicants.
Another point of criticism is that field experiments are usually conducted for a limited number of
positions, usually lower status positions, which makes the generalisation of results very
problematic. Some of these points have been alleviated by conducting testing in writing rather than

nccr – on the move, Working Paper #20
8
in person, as this makes the matching of candidates easier and allows testing in a greater variety of
jobs – the requirement being that applications can be submitted in writing.
A major strand of argumentation against conducting field experiments is the issue of research
ethics. The nature of their research design makes audit studies and correspondence tests examples
of covert research, where participants are not aware that they are taking part in an experiment. Field
experiments thus infringe the basic research ethics principles of voluntary participation and
informed consent by the participants, and non-deception. However, several research ethics
guidelines in the social sciences have recognized that this infringement of research ethics standards
might be justified in some situations, e.g. when it would not be possible to obtain data of the same
quality by other means4. Furthermore, numerous courts have deemed evidence obtained by audit or
correspondence tests as admissible evidence in legal proceedings, thus strengthening the position of
researchers using the methodology (for a detailed discussion on research ethics see Zschirnt
(2019)). It is therefore important to take these concerns seriously and address them in the research
design of a field experiment.
5 Designing a Correspondence Test in a Bilingual Country
Planning a correspondence test is very resource and time consuming and requires a lot of attention
to detail. In the case of Switzerland, this challenge was exacerbated by the bilingual character of the
research design. The individual steps and decisions made from the start to the preparation of the
application material are now discussed in detail.
5.1 Choice of Groups
The number of minority groups tested in a correspondence test varies quite a lot between
experiments. While many researchers choose to test only one – often the largest or most visible
immigrant group of a country, some experiments have included several immigrant groups (e.g.
Booth, Leigh and Varganova, 2012; Weichselbaumer, 2016; Wood et al. 2009). Like in the
correspondence test conducted by Fibbi, Kaya, and Piguet (2003), we chose to look at descendants
of three immigrant groups5. We kept the groups of immigrants with Turkish or Kosovar origin (in
Fibbi, Kaya and Piguet (2003), these were labelled according to their formal status in those days
“Albanian speaking former Yugoslavs”), but replaced the group of the European Union immigrant.
Instead of Portuguese we tested groups originating from neighbouring countries, i.e. people with
German origin in the German-speaking part of Switzerland and with French origin in the French
language region. These choices were based on the size of the respective immigrant groups as well
as on findings from research on attitudes towards foreigners in Switzerland. These attitude surveys
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
4 Zschirnt (2018, 2019) includes the following examples: The American Sociological Association (2018), the British Sociological
Association (2002), the American Psychological Association (2017), the National Committees for Research Ethics in Norway
(NESH) (2006), the Deutsche Gesellschaft für Soziologie (2017), or the non-binding EU Code of Ethics for Socio-Economic
Research (Dench, Iphofen and Huws 2004)
5 The field experiment discussed in this paper focuses on Swiss-born children of immigrants. They are portrayed as dual citizens and
are assumed to have acquired Swiss citizenship through an individual – usually quite strict – naturalisation procedure, in the absence
of a jus soli provision in Switzerland. In the paper, they are designated in reference to their ancestry as having a ‘Turkish/Kosovar/
German/French-origin/descent’, or as being ‘minority’ applicants or foreign named candidates.

nccr – on the move, Working Paper #20
9
consistently show strong negative stereotypes towards people from former Yugoslavian countries
and some negative attitude towards Turks, while immigrants from EU countries rarely encounter
negative attitudes (e.g. Krings and Olivares, 2007; Longchamp et al., 2014; Rapp, 2015; Raymann,
2003). However, some academic studies conducted in recent years have shown that immigrants
from neighbouring EU countries can also encounter negative stereotypes and that the small
differences between Germans and Swiss-Germans, or French and Swiss-French, are often
exaggerated (e.g. Helbling, 2011; Krings et al. 2014; Matser et al., 2010; Oudenhoven, Selenko and
Otten, 2010). The different findings from attitude research, depending on the immigrant groups
analysed, calls for a study that investigates whether such differences in attitudes also translate into
behaviour, such as hiring practices.
5.2 Signalling Group Membership
In contrast to Fibbi, Kaya and Piguet’s study (2003), in which fictitious applicants attended Swiss
schools but indicated a birthplace abroad and held a long-term residence permit, all minority
applicants in this correspondence test pointed out in their CVs that they had dual nationality (Swiss
and nationality of parents’ country of origin), had completed all their education in Switzerland, and
listed German or French as one of their native languages. The résumés thus portray the applicants as
Swiss dual nationals, members of the second generation, fluent in the local language and well
integrated (not only in the labour market). Employers are therefore presented with “ideal” ethnic
minority candidates. Next to indicating the citizenship of the country of origin, ethnicity was also
signalled by the names chosen, as well as the listing of a second native language (e.g. Turkish or
Albanian).
The names of the fictitious candidates were chosen very carefully, since employers should
recognise them and attribute them correctly to the ethnic background of the applicants. However,
names can signal more than just ethnicity, e.g. socio-economic status or religion, and might
introduce a number of unintended and unobserved variables (for discussions on names used in
correspondence testing see e.g. Crabtree and Chykina 2018; Fryer and Levitt 2004; Gaddis 2017a;
Gaddis 2017b; Pager 2007). Using information on the frequency of names provided by statistical
offices, language societies, newspapers, online sources or previous studies, frequent first and last
names that were common in each country of origin were selected. The origins of the names were
then checked against popular baby naming websites to exclude names with strong religious
connotations. Finally, all combinations of names were looked up on Facebook, to see if they existed
(i.e. were plausible) and whether numerous profiles with these names existed, so that employers
would not be able to immediately identify a candidate. Table 1 provides an overview of the names
used in this correspondence test.

nccr – on the move, Working Paper #20
10
Table 1. Names used in the correspondence test
Origin
Female
Male
German
Kathrin Hoffmann
Alexander Hoffmann
Swiss German
Fabienne Kälin
Pascal Kälin
Swiss French
Céline Favre
Sébastien Favre
French
Inès Dubois
Raphaël Dubois
Kosovo
Shpresa Krasniqi
Bekim Krasniqi
Turkey
Filiz Yilmaz
Cem Yilmaz
Our study reveals whether hiring discrimination turns Swiss citizens holding immigrant-sounding
names and dual citizenship into ‘ethnic’ minorities. Therefore, we indicate the tested groups as
based either on names, or origin, or ethnicity, those terms being in this text used as synonyms.
5.3 Choice of Locations
Many correspondence tests focus on specific cities within a country or a specific language region,
yet some studies also used vacancies from all over the countries studied (examples include: Austria
(Weichselbaumer 2016b), Germany (Kaas and Manger 2012; Schneider, Yemane and Weimann.
2014), The Netherlands (Andriessen et al. 2012; Blommaert, Coenders and Van Tubergen 2014),
Sweden (Agerström et al. 2012), or the US (Widner and Chicoine 2011)). While Fibbi, Kaya and
Piguet (2003) focused on specific urban regions in the German and French-speaking parts of
Switzerland, this study included vacancies from all over Switzerland, both in urban and rural
regions, excluding only the small Italian-speaking part for logistical reasons. The number of
vacancies published in the Italian-speaking area is too low to make it worthwhile investing the huge
effort of creating application materials in Italian. Data by the Stellenmarktmonitor Schweiz
(Buchmann et al. 2015) also shows that most vacancies are published in the German-speaking part
of Switzerland (almost 80%), followed by the French-speaking area (almost 15%). Including both
language regions allows us to examine whether discrimination is different across the larger Swiss
language areas.
5.4 Choice of Occupations
Occupations tested in the correspondence tests were carefully selected and had to fulfil certain
criteria to make their inclusion in the correspondence test feasible. These criteria were:
- Applications for the position are made in writing,
- Applications are submitted online (by email or through an online application form),
- Application materials can be standardised to fit a great number of positions,
- Vacancies for the occupation are published steadily and in sufficient numbers, and
- Positions do not require too much work experience or specialised qualifications.
These criteria led to the focus on apprenticeship and intermediate skilled positions, excluding
highly skilled positions, where work experience, networks and specialised qualifications are more
important.

nccr – on the move, Working Paper #20
11
In order to identify occupations for which the creation of credible application materials was feasible
and where a sufficiently high number of vacancies was available, we used the data set of the
Stellenmarktmonitor Schweiz (Buchmann et al. 2015), and compared the frequently listed
occupations of this data set to search results on major Swiss job search websites. Thus, we created a
shortlist of the most frequently advertised positions. Subsequently, we checked whether it was
possible to create credible and convincing applications for our fictitious candidates, excluding
positions such as architects, which would require a portfolio of previous work. We chose two
positions requiring a completed apprenticeship (sales assistant and electrician) and two intermediate
skilled positions (nurse and HR clerk); the intermediate skilled positions were, however, only tested
in the German-speaking part of Switzerland (Zschirnt 2018). Furthermore, occupations dominated
by male or female workers only received fictitious applicants of the respective gender (e.g. only
male fictitious candidates applied for a position as electrician) (Table 2).
Table 2. Overview of occupations and gender tested as well as regional coverage
Occupation
Gender
Regional Coverage
Sales Assistant
(Detailhandelskaufmann/-frau, conseiller/conseillère client)
Male and
female
French and German-speaking regions
Electrician
(Elektroinstallateur, installateur électricien)
Only male
French and German-speaking regions
Nurse
(Pflegefachfrau)
Only female
German-speaking region
HR Clerks
(HR-Fachmann/-frau)
Male and
female
German-speaking region
5.5 Identifying Vacancies
As mentioned in the previous paragraph, a steady rate of advertised positions is essential to the
conduct of a correspondence test. Open positions were identified using the major job search
websites in Switzerland. In the German-speaking part, positions were found using www.jobs.ch and
www.job-room.ch (the latter being the website of the Swiss Federal employment agency). To
identify positions in the French-speaking area, a greater number of websites was used because of
the smaller size of the labour market. These included (among others): www.jobs.ch, www.jobup.ch,
www.indeed.ch, or www.job-room.ch.
5.6 Application Materials
Creating credible applications for the fictitious candidates is the most important step in the
preparation of a correspondence test. Applications have to be perceived as realistic, fitting the
vacancies they are submitted to, and, in paired applications, have to be sufficiently different but
substantially equal to avoid raising suspicion on the part of the employer. While the preparation of
CVs and cover letters meeting these requirements is always challenging, conducting a
correspondence test in Switzerland involves even more preparatory work as a complete application
dossier usually includes also a photograph, work certificates or diplomas. Conversations with HR
professionals in Switzerland have shown that applications that miss any of these elements are often

nccr – on the move, Working Paper #20
12
considered incomplete and are immediately discarded. Furthermore, the Swiss context provides
another challenge, since the labour market is divided by the language regions of the country,
requiring researchers to prepare the application packages in two languages. To limit the preparatory
burden on the research team, positions that were advertised in Italian or English were excluded
from the experiment.
In a first step, we designed CVs for each occupation for the fictitious applicants. These had to be
realistic, but could not be the CV of actual job seekers or employees. Applicants had to show a
credible career trajectory and meet the requirements of the posted positions. The fictitious
candidates in this correspondence test were born between 1990 and 1996, had a completed
apprenticeship and had continued to work in the same company for some years. We changed this
procedure slightly for the HR clerk résumé as this position requires more work experience; this was
shown by a change of employer shortly after completing his or her apprenticeship. To create the
résumés, we drew on LinkedIn profiles that were publically accessible, résumés uploaded on job
search websites, and information from career advice websites. We also looked at jobs posted for the
selected occupations to see which skills or qualifications employers were looking for. The CVs that
were created based on this information and the cover letters matching these candidates were then
discussed and fine-tuned with HR specialists. Finally, all CVs and cover letters were prepared in
two different layouts to avoid easy detection by the employers.
In a second step, reference letters from previous employers were created. Examples retrieved
online, as well as from text books explaining how to write reference letters, were used to draft
reference letters for our fictitious candidates. These, too, were then fine-tuned with the help of HR
specialists. Once the first set of cover letters, CVs and reference letters for the German-speaking
part of Switzerland was created, all documents were translated into French.
The next element required for a complete application is a photograph of the candidate, which is
usually included on the CV. There have been several correspondence tests that have used
photographs showing the fictitious candidates (e.g. Arceo-Gomez and Campos-Vazquez 2014; Kaas
and Manger 2012; Schneider, Yemane and Weinmann 2014; Weichselbaumer 2016a;
Weichselbaumer 2016b), even though the use of photographs poses its own challenges. A detailed
discussion of the problems that the use of photographs can introduce in the context of a
correspondence test can be found in Rich (2018). We are very thankful that Doris Weichselbaumer
granted us the permission to use the carefully prepared and pretested photographs from her Austrian
correspondence test (Weichselbaumer 2016b). The pictures can be used for both the majority and
the minority candidate as they portray the same phenotype and were randomly assigned to the
candidates6.
The diplomas applicants habitually include in their application dossiers were another important part
of the application package. Examples of diplomas were collected using our personal networks,
google image searches or websites such as LinkedIn or Instagram, where people uploaded copies of
their diplomas and made them accessible for all users. The documents obtained this way were then
manipulated in Photoshop to fit the fictitious candidates.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
6 The construction of the photographs as well as the pre-tests regarding their comparability are discussed in detail in Weichselbaumer
(2016b).

nccr – on the move, Working Paper #20
13
The last elements to be created were contact details for the candidates. All of them had to be able to
provide an actual address, an email address, and a phone number. The addresses used in this
correspondence test were real street addresses of apartment houses in several Swiss cities, the email
addresses were created using gmail.com, and the phone numbers were mobile phone numbers set up
with a voice mail box with the provider’s default message. We had one phone number for the
majority candidates and one phone number for the minority candidates and were able to match the
calls to the vacancies using the messages left or a reverse phone number search.
For each vacancy all the individual elements of the application (i.e. the cover letter, the CV, the
layout, the photograph, etc.) were randomly assigned to the two fictitious candidates and complete
application packages were compiled. These random assignments of applicants’ characteristics were
carefully recorded in the data. Furthermore, the order in which applications were sent was also
randomized.
The fieldwork was carried out in the German-speaking cantons from October 2017 and lasted 7
months, while the French data collection took place from February 2018 and lasted 9 months.
5.7 Research Ethics
The research ethics concerns inherent in correspondence testing were briefly mentioned in the
methodology section above. In order to take full account of the ethical issues, the ethical challenges
involved in the present research project were very carefully analysed. In accordance with the
relevant literature, we argue that infringing the non-deception principle can be counterbalanced by
the greater societal interest in the research topic and the lack of alternative methods to obtain data of
the same quality.
The project description, including a detailed account of how the ethical challenges were addressed
in the research project, was presented to the NCCR Ethical Commission composed of three
international scholars (2 external and 1 internal to the University of Neuchâtel) and was submitted
for approval to the Ethical Commission of the University of Neuchâtel. The Commission approved
the project in May 2016. This ethical approval process also laid the foundation for a publication on
research ethics in correspondence testing that emerged from this project (Zschirnt 2019).
The details of the research design were strongly influenced by compliance with the ethical
standards discussed in the above-mentioned documentation and the concern to minimise
inconvenience for employers. For example, each employer received only one set of paired
applications to keep inconvenience to each individual employer as low as possible. In the cases
where employers contacted the fictitious candidates for an interview, these invitations were quickly
and politely declined. Furthermore, results are analysed only in an aggregated form, which ensures
respect for the privacy of the employers. In the name of transparency, we inform the companies of
their participation in the study at the end of the experimental research.
Despite the thorough preparatory work, some employers in the health care system identified the
prepared diplomas for nurses as fictitious. These employers informed the police, so that the
prosecutor’s office investigated their complaints for forgery of documents and the unauthorized use
of the logo of the Red Cross, which appeared on the nurses’ diplomas. The Neuchâtel Public

nccr – on the move, Working Paper #20
14
Prosecutor's Office eventually closed the case on the grounds that no profit was meant to be drawn
from the forgery of documents and that there was no intention to misuse the Red Cross emblem and
name as the team was not aware of its special protection. In the near future the authors of this
Working Paper will present and discuss those methodological, ethical and legal aspects of the study
in a scientific paper.
Because of the judicial case, the testing of intermediate skilled positions (i.e. nurses and HR clerks)
was no longer pursued in the French speaking part of the Swiss labour market.
6 Findings
For the sake of readability, the main findings are presented in separate tables corresponding to the
research question they provide an answer for. The few redundancies that this choice of a meticulous
presentation entail are largely compensated by a more transparent message. All the tables have
strictly the same structure (following Zschirnt 2018), detailed in the legend of Table 1 and
thoroughly discussed in the first subchapter.
6.1 Do Swiss Citizens of Different Immigrant Origin Face Hiring Discrimination?
The first step of the analysis addresses the question whether Swiss citizens of immigrant origin are
confronted by ethnic discrimination in hiring decisions in the Swiss labour market. The data
presented in Table 3 provides the descriptive statistics from the experiment in both language
regions for the two occupations of sales assistant and electrician, the two professions that were
tested nationwide and require an apprenticeship level qualification.
Table 3. Correspondence testing results for the Swiss labour market (nationwide) in apprenticeship
level occupations (sales assistant and electrician), by ethnicity
[1]
Number
of jobs
[2]
None
invited
[3]
Both
invited
[4]
Only
majority
invited
[5]
Only
minority
invited
[6]
Net
discrimination
rate in %
[7]
Callback
majority
in %
[8]
Callback
minority
in %
[9]
Ratio
[10]
χ²
France and
Germany
179
109
44
18
8
14.3
34.6
29.1
1.2
**
Kosovo
182
111
39
26
6
28.2
35.7
24.7
1.4
***
Turkey
180
125
31
15
9
10.9
25.6
22.2
1.2
Total
CH all
541
345
114
59
23
18.4
32.0
25.3
1.3
***
Explanatory notes:
Column [1] shows the total number of vacancies used. Columns 2 to 5 express the treatment of the pair of applications sent.
Column [2] specifies in how many cases both applications were rejected or did not receive an answer.
Column [3] shows the number of cases where both candidates received a positive reply.
Column [4] lists the cases in which only the Swiss-origin candidate received a positive reply.
Column [5] are the cases in which only the Swiss minority candidate received a positive reply.
Column [6] shows the net discrimination rate: ([4]-[5])/([3]+ [4]+ [5]).
Columns [7] and [8] report the success rates of the majority ([3]+[4])/[1] and minority candidates([3]+[5])/[1],
Column [9] presents the ratio between the success rates of the majority and minority candidates, also known as the relative call-back
rate. Column [10] shows the statistical significance of the difference between [4] and [5] using a chi-square test (significant at the
* 10%, ** 5% and ***1% level). The different rows present the results for all applicants combined (1st row), and for applicants from
France and Germany, from Kosovo, and Turkey separately. (Table based on Zschirnt, 2018)

nccr – on the move, Working Paper #20
15
Paired applications were sent in response to vacancies posted online by 541 firms looking for a
sales assistant or an electrician (Column 1). In 85% of the cases, the two applications were treated
equally (sum of N=345, none invited in Column 2 and N=114, both invited in Column 3).
However, in other cases the two fictitious applicants encountered different treatment (Column 4,
only majority invited and Column 5, only minority invited). If unequal treatment were random,
cases in favour of the majority candidate should equate cases in favour of the minority candidate:
the values in Column 4 and Column 5 should be the same. Cases that differ from this symmetry are
considered as unequal treatment. The net discrimination rate (18.4%, Column 6) indicates that
minority candidates stand about a one-in-five chance of advancing as far as a majority candidate in
the hiring process because they are not invited the interview. The outcome of interest is whether
candidates get invited for a job interview or not, and call-back rate – the ratio between interview
invitations and applications sent – shows the success rate of each candidate to receive a positive
reply. Applications of majority candidates resulted in an invitation for a job interview in 32.0% of
the cases (Column 7) while applications of minority candidates received such an invitation in 25.3%
of the cases (Column 8). In the following the results are discussed mainly on the basis of the call-
back rates.
Column 9 of Table 3 expresses these outcomes as a ratio, the relative call-back rate, which
compares how often a majority applicant is called for an interview to how often a minority
applicant is invited for an interview. Majority applicants for the two tested apprenticeship level
occupations need to send 10 résumés to get one call-back, whereas minority applicants need to send
about 13 résumés. In other terms: minority candidates have to write 1.3 times as many applications
as majority candidates to be invited for a job interview. This 30-percent gap in call-back is
statistically significant at the 1% level (Column 10).
These findings demonstrate that Swiss children of immigrants educated in the country and holding
Swiss qualifications are significantly less likely to be invited for a job interview compared to their
majority competitors. A comparison of three different origin groups shows, however, that there are
substantive variations according to origin. Swiss of Turkish origin have lower chances than a
majority candidate, yet the gap between the two is not wide enough so as to exclude the null
hypothesis, i.e. that this gap is due to hazard. Swiss children of Kosovar origin are the most exposed
to hiring discrimination as they need to submit 1.4 times as many applications as majority
candidates to be invited for a job interview, whereas Swiss citizens originating from the two
neighbouring countries need to write 1.2 applications to be invited to a job interview.
6.2 Are There Differences by Occupation?
In a second step, the analysis looks at the selected occupations in more detail, asking whether the
welcoming of Swiss applicants of immigrant origin in Swiss firms varies according to the type of
occupation they apply for.

nccr – on the move, Working Paper #20
16
Table 4. Correspondence testing results for the Swiss labour market (nationwide) in apprenticeship
level occupations, by occupation and ethnicity
[1]
Number
of jobs
[2]
None
invited
[3]
Both
invited
[4]
Only
majority
invited
[5]
Only
minority
invited
[6]
Net
discrimination
rate in %
[7]
Callback
majority
in %
[8]
Callback
minority
in %
[9]
Ratio
[10]
χ²
Sales assistant
270
198
26
33
13
27.8
21.9
14.4
1.51
***
France and
Germany
89
61
10
14
4
35.7
27.0
15.7
1.71
**
Kosovo
91
66
10
11
4
28.0
23.1
15.4
1.50
*
Turkey
90
71
6
8
5
15.8
15.6
12.2
1.27
Electrician
271
147
88
26
10
12.9
42.1
36.2
1.16
***
France and
Germany
90
48
34
4
4
0.0
42.2
42.2
1.00
Kosovo
91
45
29
15
2
28.3
48.4
34.1
1.42
***
Turkey
90
54
25
7
4
8.3
35.6
32.2
1.10
Total CH all
541
345
114
59
23
18.4
32.0
25.3
1.3
***
The respective minority candidates stand different chances of being invited to a job interview
according to the occupation involved, in spite of the fact that the difference in overall relative call-
back ratios for the two trades is not statistically significant (Table 4). Swiss professionals of
Kosovar origin present similar unfavourable relative call-back ratios in the two occupations tested.
By contrast, Swiss applicants whose parents come from neighbouring countries face significantly
lower chances than a majority candidate to advance in the recruitment process when applying for a
sales assistant position than when applying for a position as an electrician. Yet, they are treated on
par when applying as electricians. Swiss candidates with a Turkish background are the least
exposed to hiring discrimination: in both occupations the findings for Swiss candidates with a
Turkish background were not statistically significant.
6.3 Are There Differences by Gender (in Sales)?
The next sept of the analysis examined whether the gender of the applicant influences the chances
of being invited for a job interview. The answer to this question is based on data relative to the sales
assistant position only, as only these job openings received applications from both male and female
candidates.
The relative call back rate (ratio) is 1.5 for both male and female candidates who apply for sales
assistant positions. However, the invitation rates are higher for female candidates (28% vs. 18%)
compared to male candidates (15% vs. 10%) (Table 5).

nccr – on the move, Working Paper #20
17
Table 5. Correspondence testing results for the Swiss labour market (nationwide) in sales
occupations, by gender
[1]
Number
of jobs
[2]
None
invited
[3]
Both
invited
[4]
Only
majority
invited
[5]
Only
minority
invited
[6]
Net
discrimination
rate in %
[7]
Callback
majority
in %
[8]
Callback
minority
in %
[9]
Ratio
[10]
χ²
Male
122
100
8
10
4
27.3
14.8
9.8
1.5
Female
148
98
18
23
9
28.0
27.7
18.2
1.5
**
Total sales
assistants
CH all
270
198
26
33
13
27.8
21.9
14.4
1.5
***
Are female minority candidates more or less exposed than men to hiring discrimination? Empirical
evidence from similar studies delivers contrasting conclusions. Andriessen et al. (2012) found
women somewhat less targeted by hiring discrimination, corroborating the hypothesis that minority
men are perceived as more threatening than minority women. On the other hand, in their meta-
analysis of correspondence testing studies in OECD countries, Zschirnt and Ruedin (2016) found no
indication of systematic gender differences on a large scale. Bovenkerk et al. (1995) came to similar
conclusions.
In our field experiment, we find no statistically significant gender difference among minority
groups. As our experiment was designed to assess ethnic hiring discrimination, the data do not
allow a more detailed discussion of this issue.
6.4 Are There Regional Differences in Hiring Discrimination?
The national scope of the experiments allows us to focus on the question of whether the treatment
of Swiss citizens of immigrant descent varies according to the language region in which they live.
In other words, is there a difference between the French and the German-speaking parts of the
Swiss labour market when it comes to ethnic discrimination in hiring decisions?

nccr – on the move, Working Paper #20
18
Table 6. Correspondence testing results for the Swiss labour market in apprenticeship level
occupations, by ethnicity and language region
[1]
Number
of jobs
[2]
None
invited
[3]
Both
invited
[4]
Only
majority
invited
[5]
Only
minority
invited
[6]
Net
discrimination
rate in %
[7]
Callback
majority
in %
[8]
Callback
minority
in %
[9]
Ratio
[10]
χ²
Germany
90
40
30
14
6
16.0
48.9
40.0
1.2
*
Kosovo
92
44
31
13
4
18.8
47.8
38.0
1.3
**
Turkey
90
55
22
9
4
14.3
34.4
28.9
1.2
Total
CH DE
272
139
83
36
14
16.5
43.8
35.7
1.2
***
France
89
69
14
4
2
10.0
20.2
18.0
1.1
Kosovo
90
67
8
13
2
47.8
23.3
11.1
2.1
***
Turkey
90
70
9
6
5
5.0
16.7
15.6
1.1
Total
CH FR
269
206
31
23
9
22.2
20.1
14.9
1.4
***
Unequal treatment for Swiss candidates of immigrant origin is a reality common in both language
regions (Table 6). This is notably the case for applicants of Kosovar origin, the only group
significantly affected by hiring discrimination in both language regions. The gap between majority
applicants and candidates with a Kosovar name is especially wide in the French-speaking region:
the minority applicant has to write twice as many applications to be invited for a job interview.
Moreover, Table 6 shows that Swiss candidates of German descent applying for a position in the
German-speaking area of the country are also significantly affected by the risk of unequal treatment
(at the 10% level), while this is not the case for French-origin candidates in the French-speaking
cantons (here the findings are not statistically significant). In both regions Turkish-origin applicants
experience unequal treatment to a lesser, non-significant, degree. Furthermore, in our experiment
call-back rates for majority and minority applicants are generally higher in the German-speaking
region than in the French-speaking one, which could affect the results.
We have seen that regional relative call-back ratios for both trades are statistically significant at the
aggregate level. Are the differences by trade that were observed at the national level (Table 7) also
important at the regional level?

nccr – on the move, Working Paper #20
19
Table 7. Correspondence testing results for the Swiss labour market in apprenticeship level
occupations, by occupation, ethnicity and language region
[1]
Number
of jobs
[2]
None
invited
[3]
Both
invited
[4]
Only
majority
invited
[5]
Only
minority
invited
[6]
Net
discrimination
rate in %
[7]
Callback
majority
in %
[8]
Callback
minority
in %
[9]
Ratio
[10]
χ²
Sales assistant
136
83
19
24
10
26.4
31.6
21.3
1.5
**
Germany
45
24
7
11
3
38.1
40.0
22.2
1.8
**
Kosovo
46
28
8
7
3
22.2
32.6
23.9
1.4
Turkey
45
31
4
6
4
14.3
22.2
17.8
1.3
Electrician
136
56
64
12
4
10.0
55.9
50.0
1.1
**
Germany
45
16
23
3
3
0.0
57.8
57.8
1.0
Kosovo
46
16
23
6
1
16.7
63.0
52.2
1.2
*
Turkey
45
24
18
3
0
14.3
46.7
40.0
1.2
*
Total CH DE
272
139
83
36
14
16.5
43.8
35.7
1.2
***
Sales assistant
134
115
7
9
3
31.6
11.9
7.5
1.6
*
France
44
37
3
3
1
28.6
13.6
9.1
1.5
Kosovo
45
38
2
4
1
42.9
13.3
6.7
2.0
Turkey
45
40
2
2
1
20.0
8.9
6.7
1.3
Electrician
135
91
24
14
6
18.2
28.1
22.2
1.3
*
France
45
32
11
1
1
0.0
26.7
26.7
1.0
Kosovo
45
29
6
9
1
50.0
33.3
15.6
2.1
**
Turkey
45
30
7
4
4
0.0
24.4
24.4
1.0
Total CH FR
269
206
31
23
9
22.2
20.1
14.9
1.4
***
In both regions, call-back ratios for electricians tend to be lower than for the sales occupation,
signalling less discrimination in this occupation. Kosovar-origin electricians are the most exposed
to unequal treatment in the two regions, Turkish-origin electricians are also affected in the German-
speaking area but not in the French-speaking one; finally, electricians whose parents come from a
neighbouring country have similar or even identical call-back rates to majority candidates in the
two regions.
The picture is quite different in the sales sector. In the German-speaking region, the only minority
group affected by statistically significant unequal treatment is the German-origin sales applicants,
with a quite high relative call-back rate. Candidates with a German background experience the
highest relative call-back rate when applying for one occupation, sales, and the lowest for the other
tested trade, electrician. Contrasting with the general trend, Kosovar-origin applicants are not the
most affected by hiring discrimination in the sales position. In the French-speaking region, the
ethnic group ranking of candidates for a sales position resembles the overall picture.
6.5 Is Discrimination Lower in Professions Requiring Tertiary Level Training?
In the German-speaking region the field experiment included not only apprenticeship level
occupations but also professions requiring a tertiary qualification (nurses and HR clerks, for more
details, see Zschirnt 2018). This allows us to answer the question of whether unequal treatment
differs according to the level of qualification required for the occupation.

nccr – on the move, Working Paper #20
20
Table 8. Correspondence testing results for the Swiss-German labour market, by educational level
of occupation and ethnicity
[1]
Number
of jobs
[2]
None
invited
[3]
Both
invited
[4]
Only
majority
invited
[5]
Only
minority
invited
[6]
Net
discrimination
rate in %
[7]
Callback
majority
in %
[8]
Callback
minority
in %
[9]
Ratio
[10]
χ²
Germany
90
40
30
14
6
16.0
48.9
40.0
1.2
*
Kosovo
92
44
31
13
4
18.8
47.8
38.0
1.3
**
Turkey
90
55
22
9
4
14.3
34.4
28.9
1.2
Total
secondary
ed. level
CH DE
272
139
83
36
14
16.5
43.8
35.7
1.2
***
Germany
96
57
26
3
10
-17.9
30.2
37.5
0.8
*
Kosovo
96
61
23
10
3
19.4
34.4
27.1
1.3
*
Turkey
96
50
34
8
4
8.7
43.8
39.6
1.1
Total
tertiary
ed. level
CH DE
288
168
83
21
17
3.3
36.1
34.7
1.0
Overall, the relative call-back rate for tertiary level occupations is 1.0, indicating that the treatment
of applications tends to be identical between majority and minority groups. Data reveal a specific
ethnic ranking. While applicants with a Kosovar background are exposed to unequal treatment, with
a relative call-back rate of 1.3, candidates with a German background are even preferred to Swiss-
origin applicants, as indicated by a relative call-back rate of 0.8. Such a situation contrasts with the
observations for secondary level occupations (Table 8). The gap between relative call-back rates,
according to the educational requirements of jobs, is statistically significant (p<0.05) only for
majority candidates but not for minority applicants.
As shown in Table 9, the lowest relative call back rate was measured for the position of HR clerks
in the German speaking part of Switzerland. On the aggregate level combining all ethnic origin
groups, the results are, however, not statistically significant. Findings for the HR clerk position are
only statistically significant in the case of German-named applicants. For the second intermediate
skilled position, that of nurses, the results also show almost equal treatment when considering all
candidates, but here, too, the results are not statistically significant. In the case of nurses, the only
significant finding is discrimination against Kosovar named applicants (Zschirnt 2018).

nccr – on the move, Working Paper #20
21
Table 9. Correspondence testing results for the Swiss-German labour market in tertiary level
occupations, by occupation and ethnicity
[1]
Number
of jobs
[2]
None
invited
[3]
Both
invited
[4]
Only
majority
invited
[5]
Only
minority
invited
[6]
Net
discrimination
rate in %
[7]
Callback
majority
in %
[8]
Callback
minority
in %
[9]
Ratio
[10]
χ²
Nurse
144
62
63
12
7
6.1
52.1
48.6
1.1
Germany
48
18
22
3
5
-6.7
52.1
56.3
0.9
Kosovo
48
27
16
5
0
23.8
43.8
33.3
1.3
**
Turkey
48
17
25
4
2
6.5
60.4
56.3
1.1
HR clerk
144
106
20
9
10
-2.6
20.1
20.8
1.0
Germany
48
39
4
0
5
-55.6
8.3
18.8
0.4
**
Kosovo
48
34
7
5
3
13.3
25.0
20.8
1.2
Turkey
48
33
9
4
2
13.3
27.1
22.9
1.2
Total
tertiary
ed. level
CH DE
288
168
83
21
17
3.3
36.1
34.7
1.0
A closer look at the two different professions shows that in both professions Swiss applicants of
German origin are more readily invited for a job interview than majority candidates, the gap being
statistically significant for HR clerks. Moreover, Swiss applicants of Kosovar descent experience a
significantly lower call-back rate even in the very tight labour market for nurses.
7 Discussion
This correspondence test uncovers major features of the treatment of minorities on the Swiss labour
market concerning occupational differences, ethnic hierarchies, and regional variations. A
comparative perspective reveals the implications of the findings.
7.1 Occupational Differences
International evidence attests “that more discrimination is found in the lower segments of the labour
market” (Andriessen et al. 2012: 256; also Bovenkerk et al. 1995; Carlsson 2010). Our experiment
confirms this at the aggregate levels and for German applicants. Yet, the experiment finds evidence
that hiring discrimination is similar across levels of qualification for Kosovar candidates. Our data
shows that education reduces exposure to discrimination for German candidates but does not seem
to protect Kosovans from discrimination in accessing the labour market.
This finding is consistent with the matching hierarchies theory. Auer et al. argue that employers’
hiring decisions are influenced by their perception “of a candidate’s nationality within the
framework of pre-existing ethnic hierarchies” (2018: 4), on one side, and by their understanding of
occupational hierarchies, on the other. The model maintains that employers seek “to maximize the
fit between an applicant’s position within the ethnic hierarchy and the occupation’s position within
the social status scale” (Auer et al. 2018: 4).

nccr – on the move, Working Paper #20
22
We find the highest relative call-back rates in the secondary level occupation as sales assistant,
indicating a considerable gap between majority and minority candidates. This result might point to
Becker’s theory (1957) of taste-based discrimination, which predicts higher discrimination rates in
jobs involving customer contact. Andriessen et al. (2012) also find most discrimination in the retail
trade in the Netherlands; they designed their experiment specifically to study the impact of
customer contact. They do find some evidence for customer discrimination, but conclude that
relatively little of the discrimination observed can be attributed solely to this phenomenon.
In tertiary level occupations discrimination is lowest; the difficulty in finding qualified personnel
might make employers more inclined to overlook criteria, such as group membership, and
concentrate more on qualifications (Andriessen et al. 2012). According to ethnic conflict theory,
this situation should lead to lower levels of competition and hence reduced discrimination. Lower
discrimination rates are expected especially for the nursing profession, which is characterised by
widespread labour shortage in Switzerland. Nevertheless, the experiment points to one unique case
of hiring discrimination at the tertiary educational level, affecting Swiss nurses with a Kosovar
background. This finding supports the hypothesis of an especially strong negative attitude towards
this ethnic group.
7.2 Ethnic Hierarchies
The study documents different degrees of exposure to discrimination among the offspring of the
tested immigrant groups; it thus highlights the existence of a differentiated intergroup bias towards
the various outgroups, designing ethnic hierarchies. “In such ethnic hierarchies, the ethnic in-group
is typically ranked first (most preferred) and ethnic out-groups are ranked in a specific order further
down” (Andriessen et al. 2012: 243, emphasis in the original). Ethnic hierarchies reflect the
majority perception of threats rooted in two types of factors. On one side, cultural factors like
immigrants’ degree of similarity in values and, on the other side, economic factors like their social
status and their qualifications (Bobo and Hutchings 1996; Hagendoorn 1995). Ethnic ranking of
immigrant groups mirrors majority attitudes towards minorities. Andriessen et al. (2012) also refer
to Snellman (2007), who argues that discrimination gets more pronounced the lower the standing of
an ethnic group is in the ethnic hierarchy.
Our correspondence test on apprenticeship level occupations points towards a clear disadvantage
for Swiss citizens of Kosovar descent. Research on attitudes towards (first generation) immigrants
in Switzerland reveals a consistent pattern of ethnic hierarchy: groups from the Balkan region are
confronted with the least favourable attitudes followed by Turks, while European Union citizens are
usually considered favourably (Auer et al. 2018; Binggeli, Krings and Sczesny 2014; Hainmüller
and Hangartner 2013; Raymann 2003; Ruedin et al. 2013). Other studies document the relevance of
the religious category of Muslim as a target of negative attitudes (e.g. Helbling 2010). The
differences between the ethnic groups in our experiment are only partially in line with the expected
order from the ethnic hierarchy: therefore, further investigation about the position of Turkish-origin
applicants and the negative score for EU citizens is needed.
Although hierarchies are specific to a place and probably time, it is worth underlining that in
European studies Turkish-origin immigrants and their offspring do not appear as the most targeted
group by hostile behaviour in hiring, according to meta-analyses of correspondence testing research

nccr – on the move, Working Paper #20
23
(Rich 2014; Zschirnt and Ruedin 2016). An experimental study in the Netherlands also finds that,
after controlling for individual characteristics, ethnic penalties in unemployment are smaller for
Turkish than for Moroccan second generation applicants (Andriessen et al. 2012).
The role of the media and the way minorities are politicised in public debate plays an important role
in the perception of minorities (Brug et al. 2015). “The mediatised debate provides and reinforces
stereotypes that can be used as shortcuts in statistical discrimination” (Zschirnt and Ruedin 2016:
1118) or affect taste-based discrimination (Pecoraro and Ruedin 2016). While both non-EU groups
tested in our study share a Muslim background, a hypothesis that might account for the Turkish case
in the Swiss labour market is that public debate has focused very little on the Turkish immigrant
group representing about 3% of the foreign population in the country in 2017. Such a hypothesis
needs further investigation.
A closer look at the features taken by hiring discrimination for applicants with an EU background in
our experiment shows that it does not extend to both EU-origin groups tested, the French and the
Germans. The result is due to a rather specific constellation of factors: it is attributable to German-
origin candidates in the German-speaking part of the country for one precise trade, sales. As a
matter of fact, in all other positions German named candidates face no discrimination or are even
preferred to the Swiss candidates. Among EU citizens in Switzerland, Germans do not enjoy a
favourable image. Citing a study by Stolz (2005: 558-559), Helbling indicates that negative
attitudes towards foreigners from South European countries and France do not exceed 5 per cent
while around 11 per cent find that Germans are little or not at all “likeable” (Helbling 2010: 7).
In her dissertation, Zschirnt discusses this issue thoroughly: “While readers who are not familiar
with the Swiss context might find this result puzzling, work conducted by Helbling (2011) in the
city of Zurich as well as research by Matser et al. (2010) have shown negative attitudes towards
German immigrants. Both argue that in the face of many similarities the minor differences between
Germans and Swiss Germans are strongly emphasised. The fact that we find higher relative call-
back rates for Germans applying for sales positions, could also be due to perceived language skills
(dialect) and/or expected customer discrimination. Since all sales positions involve customer
contact, employers might expect German candidates to speak only high-German and not the Swiss
German dialect, which could be regarded as negative by Swiss customers. However, since all
candidates had completed their education in Switzerland, where the local dialect is also spoken in
schools, they should be expected to be proficient in the local dialect. (…) As Krings et al. (2014)
have shown, the Swiss perceive Germans as highly competent, but lacking in warmth. This might
be beneficial for the positions of electricians, nurses, or HR clerks, while the lack in warmth might
contribute to the high discrimination these candidates experience in the sales positions.
Interestingly, a preference for German candidates even before national candidates also seems to
emerge from research on labour market discrimination in the Netherlands suggesting that German
applicants might benefit from positive stereotypes in the labour market (Philippen and van Eldert
2017).” (2018: 217).
7.3 Regional Differences
Two sets of factors may influence discrimination: attitudes and the political environment
surrounding the issue of immigration on the one hand, and the labour market situation on the other.

nccr – on the move, Working Paper #20
24
Rather negative attitudes and an unfriendly political environment are associated with higher
minority discrimination (Blommaert, Van Tubergen and Coenders 2012; Quillian 2006). On this
basis, differences could be expected, with higher discrimination rates in less immigrant friendly
regions. On the other hand, lower unemployment rates are signs of labour shortages, and could be
considered conducive to lower minority discrimination (Baert et al. 2013), although empirical
evidence does not always support this hypothesis (Carlsson et al. 2018; Zschirnt and Ruedin 2016).
Attitudes towards immigration (Ackermann and Ackermann 2015; Green et al. 2011), votes on
immigrant related issues (Manatschal 2015) as well as unemployment rates (SECO 2018) are
dimensions differentiating the two main Swiss language regions: the German-speaking region
presents a less immigration friendly attitude and political environment while it experiences a lower
unemployment rate. All these studies are fraught with difficulties as we are dealing with
differentiating between language regions, missing variables, and technical issues like translation
effects or differences in social desirability. The field experiment is not completely free from these
problems either, but the focus on behaviour can overcome many of these differences.
Countervailing forces presumably explain the absence of noticeable regional differences. If
anything, the gap in relative call-back rates, barely significant (p>.1), tends to be lower in the
German-speaking than in the French-speaking region. Indeed, applications from both majority and
minority receive a more positive response in the former area, characterised by a lower
unemployment rate.
7.4 New Insights into Hiring Discrimination in Switzerland
Hiring discrimination affecting Swiss educated offspring of immigrants was also studied in a
correspondence testing experiment in the early 2000s (Fibbi, Kaya and Piguet 2003). How does the
present research modify our knowledge of ethnic ranking and regional differences?
The previous study concerning entry-level jobs requiring upper secondary education showed a clear
ethnic ranking, placing Kosovar-origin applicants in the most unfavourable position, followed by
Turkish-origin candidates, while the tested EU-origin group, the Portuguese, appeared the least
affected by unequal treatment. The present study shows convergent evidence of a higher exposure
to discrimination of Kosovar-origin candidates in comparison to Turkish-origin applicants. At the
same time, it reveals that Turkish-origin applicants are often in a comparable position to EU-origin
candidates, who are specifically targeted in one particular trade as sales assistants.
The 2003 study documented comparable discrimination rates across regions when looking at
maximal discrimination rates7. In the same vein, the present research shows comparable results in
the two language regions of the country, and possibly a more minority favourable situation in the
German-speaking region. On the whole, results point to no substantive regional differences.
The comparison between the two studies must be taken with caution: while they use the same basic
methodology, there are some relevant differences in the design. They both test jobs requiring an
apprenticeship, yet in the present study observations are concentrated on two occupations while in
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
7 According to the ILO methodology (Bovenkerk 1992), the maximal discrimination rate includes also the cases where both
candidates are invited to a job interview yet the second applicant is invited only after the first declined a similar invitation.

nccr – on the move, Working Paper #20
25
the previous study applications were sent for all sorts of trades. In the previous study, application
packages consisted of a CV and a motivation letter whereas now they also included pictures and
diplomas, providing the employers with more extensive information. The EU-origin group tested
then were Portuguese while now these groups are from neighbouring countries. Finally, the legal
and social status of fictitious candidates differed to a certain extent. In the previous study they
appeared as born abroad, holding a foreign nationality and a long-term residence permit but having
followed Swiss schools all the way through; in the present study fictitious applicants are portrayed
as Swiss citizens, holding dual nationality and having been schooled entirely in this country.
A comparison of the net discrimination rates in the two studies shows a flattening of the ethnic
hierarchy. It could be attributed to very different factors: on one side, the improved legal status of
the minority candidates as Swiss citizens, as a consequence of their longer presence in the country
affecting their acceptance; on the other side, it could be a consequence of a greater amount of
information provided to employers by candidates, reducing statistical discrimination. The research
design does not allow us to elaborate further; new research is needed in this field.
Two further results concerning the structure of the ethnic ranking are especially salient. First, the
significant discrimination attested for Swiss candidates of EU-origin is unexpected on the basis of
cultural distance considerations as well as socio-economic position. And second, the result
attributable to unequal treatment of candidates with German background in one specific trade.
The second striking finding concerns the position of Swiss candidates of Turkish-origin in the
Swiss ethnic ranking, where they appear placed close to EU-citizens. Further research is needed to
assess the robustness of this trend. Diachronic comparisons of discrimination rates are rather rare in
the literature. Zschirnt and Ruedin (2016) find no changes over time at the aggregate level. One
study documents a decline in discrimination rates for Latinos in the US (Quillian et al. 2016); yet,
considering the empirical evidence insufficiently broad, it waives further exploration of the
magnitude and the mechanisms behind this trend. Changes in the perception and declining ethnic
disadvantage are documented in the literature (Alba and Foner 2015; Fibbi 2018; Guglielmo 2003;
Ignatiev 1995), yet over a longer period of time.
7.5 Swiss Results in an International Perspective
The relative call-back rate of 1.3 resulting from the Swiss experimental study indicates that
minority candidates have to send 30% more applications than majority candidates in order to pass
the first yet crucial stage of the hiring process and be invited for a job interview. To assess the
relevance of this research result, we put it in perspective with international studies. We resort in
particular to meta-analyses, i.e. quantitative reviews of the findings of independent empirical
studies on the same topic using the same methodology with the purpose of providing a synthetic
overall result.
Two recent meta-analyses are useful for our purpose. Combining the findings of 43 different studies
on hiring discrimination against ethnic minority groups conducted in OECD countries between
1990 and 2015, Zschirnt and Ruedin (2016) report a mean call-back rate of 1.6. Moreover,
observing that the relative call back rates in German speaking countries are often lower in
international comparison, they suggest that the finding may be attributed to the comprehensive

nccr – on the move, Working Paper #20
26
information of the standard application files received by the employers which reduces the likelihood
of statistical discrimination in hiring. In the second meta-analysis Quillian et al. look at 28
experimental studies conducted between 1989 and 2015 on hiring discrimination against African
Americans or Latinos in the US labour market, and observe that “white applicants receive 36%
more call-backs than equally qualified African Americans and (…) on average 24% more call-backs
than Latinos” (2016: 2).
Such comparisons show that hiring discrimination against immigrant-origin Swiss citizens reaches a
substantial level, comparable with that found for Blacks in the US. While it is not as high as the
mean observed in the OECD countries, which generally focus on ‘problematic’ groups, it should be
kept in mind that the Swiss average includes also immigrant-origin groups from neighbouring
European countries, which are usually not included in the choice of the groups in other
correspondence tests. Zschirnt and Ruedin observe that the reported rates of discrimination in the
studies included in their meta-analysis “may thus overestimate the extent of discrimination”, due to
the choice of minority groups studied (2016: 10).
7.6 Limitation of the Study
Correspondence testing is often claimed to be the best method to measure discrimination; yet it does
so with a number of limitations that entail an underestimation of the overall hiring discrimination.
On the one hand, it focuses on the very initial stage of the job-seeking process, yet it captures 90%
of the discrimination occurring at this stage of the hiring process (Rich 2014; Riach and Rich 2002).
Furthermore, a recently published meta-analysis of audit studies in the US indicates that
discrimination is much higher when it comes to job offers at the interview stage (Quillian et al.
2018). Correspondence tests thus present a minimum rate of discrimination. Furthermore,
correspondence tests can take into account only publically announced vacancies, a fraction of the all
the job openings, excluding informally or internally filled positions. The method is suitable only for
written applications, thus excluding many entry-level and unskilled jobs, although this is less of an
issue in the formalized job market in Switzerland.
On the other hand, the study assesses discrimination of specific groups in specific occupations,
while acknowledging that this evidence is not sufficient for a generalization to the whole labour
market. The experimental design allows us to control for the compositional differences in the labour
supply (i.e. the candidates’ characteristics) to isolate the variable whose impact is to be tested. But it
does not uncover demand side mechanisms, i.e. employers’ hiring behaviour. “Determining which
characteristics influence hiring decisions would require the researcher to know not only the
successful candidate but also the entire applicant pool” (Auer et al. 2018: 10). Moreover, while the
method focuses on the assessment of discrimination it neglects the motives and the mechanisms
(Reskin 2003). Research with other methodological approaches is necessary to better grasp how
hiring discrimination is produced in everyday working life.
8 Summary and Conclusions
Today, field experiments are considered as “the gold standard” when it comes to measuring
discrimination in the market place. More than 15 years after the first assessment of hiring

nccr – on the move, Working Paper #20
27
discrimination in Switzerland it was necessary to take stock. The experimental study conducted in
2017-18 in the two main language regions aimed at assessing and measuring the hiring
discrimination of Swiss citizens of immigrant origin, educated in the country and holding Swiss
qualifications. The findings show significant differences in the chances of applicants passing the
first crucial selection phase of the recruiting process, namely the invitation to a job interview,
depending on their parents’ country of origin. The extensive application files, a standard in this
country, provide employers with far-reaching information, which should moderate statistical
discrimination. Assessing discrimination among Swiss citizens, who are equally educated and
skilled, on the basis of their ancestry, the experiment’s results make the case that some of this
discrimination could be attributed to taste-discrimination.
Hiring discrimination for occupations requiring an upper secondary level of qualification concerns
mostly applicants whose parents came from a non-EU country and reveals a systematic ethnic
hierarchy. Swiss candidates with a Kosovar background face the highest, statistically significant
discrimination across occupations and regions. On the contrary, Swiss applicants of Turkish origin
experience the lowest, statistically non-significant, hiring discrimination. Less expectedly, hiring
discrimination affects also applicants whose parents came from EU neighbouring countries.
German-origin candidates experience the highest discrimination rate in one specific occupation and
the lowest in other occupations. Regional comparison on upper secondary level occupations shows
that hiring discrimination is clearly a reality in both regions, revealing similar ethnic hierarchies.
In the German-speaking region, fictitious candidates applied for jobs both at an upper secondary
and at a tertiary level of qualification. The comparison shows significant lower risks of unequal
treatment in occupations requiring higher qualifications for applicants with a German background,
while other minority candidates do not seem protected by education attainment. Here, the matching
hierarchies model finds fresh confirmation.
The number of correspondence tests on hiring discrimination has increased sharply since 2000,
showing that this issue has become a widespread concern, not only in Europe or North America.
Field experiments have now been conducted in almost all of the EU15 countries, as well as
Switzerland and Norway. In all those countries children of immigrants are increasingly coming of
age and are confronted in the labour market with inequalities that can no longer be attributed to
migrants' deficits, as was the case for their parents.
Switzerland is no exception to this trend of a rising “second generation” and of more stringent
evidence of its unequal treatment in the labour market. Similarly, while reflecting the ethnic
composition of the immigrant-origin resident population, the ethnic ranking observed in
Switzerland echoes findings in other European countries. However, contrary to other European
countries, there is no acute awareness of this issue in Switzerland. The relatively low
unemployment rate in international comparison may make hiring discrimination less visible, to the
extent that it does not necessarily lead to unemployment, as is the case elsewhere. Trajectories of
children of immigrants from the Balkans and Turkey (Schnell and Fibbi 2016) do not show long
unemployment spells; qualitative fieldwork on those trajectories reveals, however, heavy job-search
phases with numerous applications and a certain perception of discrimination in accessing the
vocational and professional labour market.

nccr – on the move, Working Paper #20
28
The findings of this experimental study suggest the need for further research to close some gaps.
Studies on attitudes towards “outgroups” have recently enlarged their focus to “racial” and religious
groups, but ethno-national groups became a blind spot; the evolution of attitudes over time has also
been neglected. Moreover, analyses usually concentrate on first generation immigrants and little is
known on stereotyping and attitudes towards their offspring. Furthermore, the assessment of hiring
discrimination towards well-established groups with an immigrant background calls for research
designed to uncover demand-side mechanisms underpinning employers’ hiring behaviour. At the
meso level, analysis of work organisation within the firms (Baron and Bielby 1980) could
contribute to explain the occupational variations that emerged from this study and develop an
understanding of the link to segmentation and stratification.
!

nccr – on the move, Working Paper #20
29
9 Bibliography
Ackermann, Kathrin and Maya Ackermann (2015). "The big five in context: Personality, diversity and
attitudes toward equal opportunities for immigrants in Switzerland". Swiss Political Science Review, 21(3):
396-418. doi:10.1111/spsr.12170
Agerström, Jens et al. (2012). "Warm and Competent Hassan = Cold and Incompetent Eric: A Harsh
Equation of Real-Life Hiring Discrimination". Basic and Applied Social Psychology, 34(4): 359-366.
doi:10.1080/01973533.2012.693438
Aigner, Dennis J. and Glen G. Cain (1977). "Statistical theories of discrimination in labor markets".
Industrial and Labor relations review, 30(2): 175-187. doi:10.1177/001979397703000204
Alba, Richard and Nancy Foner (2015). Strangers no more: Immigration and the challenges of integration in
North America and Western Europe. Princeton: Princeton University Press.
American Psychological Association (2017). Ethical Principles of Psychologists and Code of Conduct.
Retrieved from: http://www.apa.org/ethics/code/ethics-code-2017.pdf
American Sociological Association (2018). Code of Ethics. Retrieved from:
http://www.asanet.org/sites/default/files/asa_code_of_ethics-june2018.pdf
Andriessen, Iris et al. (2012). "Ethnic discrimination in the Dutch labor market: Its relationship with job
characteristics and multiple group membership". Work and Occupations, 39(3): 237-269.
doi:10.1177/0730888412444783
Arceo-Gomez, Eva O. and Raymundo M. Campos-Vazquez (2014). "Race and marriage in the labor market:
A discrimination correspondence study in a developing country". American Economic Review, 104(5): 376-
380. doi:10.1257/aer.104.5.376
Arrow, Kenneth J. (1973). "The Theory of Discrimination", in Ashonfelter, O. and A. Rees (Eds.),
Discrimination in Labor Markets. Princeton: Princeton University Press. pp. 15-42.
Auer, Daniel et al. (2018). "The Matching Hierarchies Model: Evidence from a Survey Experiment on
Employers’ Hiring Intent Regarding Immigrant Applicants". International Migration Review (Online first).
doi:10.1177/0197918318764872
Baert, Stijn et al. (2013). Do employers discriminate less if vacancies are difficult to fill? Evidence from a
field experiment. Rochester, NY: Social Science Research Network. Retrieved from:
https://papers.ssrn.com/abstract=2214896
Baron, James N and William T Bielby (1980). "Bringing the firms back in: Stratification, segmentation, and
the organization of work". American sociological review, 45(5): 737-765. doi:10.2307/2094893
Becker, Gary S. (1957). The economics of discrimination. Chicago: Chicago University Press.
Binggeli, Steve, Franciska Krings and Sabine Sczesny (2014). "Stereotype content associated with
immigrant groups in Switzerland". Swiss journal of psychology, 73(3): 123–133.
Blommaert, Lieselotte, Marcel Coenders and Frank van Tubergen (2014). "Discrimination of Arabic-Named
Applicants in the Netherlands: An Internet-Based Field Experiment Examining Different Phases in Online
Recruitment Procedures". Social Forces, 92(3): 957-982. doi:10.1093/sf/sot124

nccr – on the move, Working Paper #20
30
Blommaert, Lieselotte, Frank van Tubergen and Marcel Coenders (2012). "Implicit and explicit interethnic
attitudes and ethnic discrimination in hiring". Social science research, 41(1): 61-73.
doi:10.1016/j.ssresearch.2011.09.007
Bobo, Lawrence and Vincent L Hutchings (1996). "Perceptions of racial group competition: Extending
Blumer's theory of group position to a multiracial social context". American Sociological Review, 61(6):
951-972. doi:10.2307/2096302
Booth, Alison L, Andrew Leigh and Elena Varganova (2012). "Does ethnic discrimination vary across
minority groups? Evidence from a field experiment". Oxford Bulletin of Economics and Statistics, 74(4):
547-573. doi:10.1111/j.1468-0084.2011.00664.x
Boulila, Stefanie Claudine (2018). "Race and racial denial in Switzerland". Ethnic and Racial Studies
(Online first). doi:10.1080/01419870.2018.1493211
Bovenkerk, Frank (1992). A Manual for International Comparative Research on Discrimination on the
Grounds of "Race" and Ethnic Origin. Geneva: ILO.
Bovenkerk, Frank et al. (1995). Discrimination Against Migrant Workers and Ethnic Minorities in Access to
Employment in the Netherlands. Geneva: ILO.
British Sociological Association (2002). "Statement of Ethical Practice for the British Sociological
Association". Retrieved from: http://www.britsoc.co.uk/media/27107/StatementofEthicalPractice.pdf
Brug, Wouter van der et al. (Eds.) (2015). The Politicisation of Migration. London: Routledge.
Buchmann, Marlis, Helen Buchs, Ann-Sophie Gnehm, Debra Hevenstone, Urs Klarer, Marianne Müller,
Stefan Sacchi & Alexander Salvisberg (2015). Swiss Job Market Monitor 1950 - 2014. Institute of
Sociology, University of Zurich. Distributed by FORS, Lausanne.
Bundesamt für Statistik (2018a). "Struktur der ständigen Wohnbevölkerung, am 31.12.2017 (in 1000) ".
Retrieved from: https://www.bfs.admin.ch/bfs/de/home/statistiken/kataloge-
datenbanken/tabellen.assetdetail.5866879.html
Bundesamt für Statistik (2018b). "Ständige Wohnbevölkerung ab 15 Jahren nach doppelter
Staatszugehörigkeit und Kanton ". Retrieved from:
https://www.bfs.admin.ch/bfs/de/home/statistiken/kataloge-datenbanken/tabellen.assetdetail.4242682.html8
Bundesamt für Statistik (2018c). "Ständige Wohnbevölkerung ab 15 Jahren nach Migrationsstatus und
verschiedenen soziodemografischen Merkmalen, 2013-2017 ". Retrieved from:
https://www.bfs.admin.ch/bfs/de/home/statistiken/bevoelkerung/migration-integration/nach-
migrationsstatuts.assetdetail.6046031.html
Bundesamt für Statistik (2018d). "Ständige ausländische Wohnbevölkerung nach Staatsangehörigkeit, 1980-
2017 ". Retrieved from: https://www.bfs.admin.ch/bfs/de/home/statistiken/bevoelkerung/migration-
integration/auslaendische-bevoelkerung.assetdetail.5866926.html
Carlsson, Magnus (2010). "Experimental Evidence of Discrimination in the Hiring of First-and Second-
generation Immigrants". Labour, 24(3): 263-278. doi:10.1111/j.14679914.2010.00482.x
Carlsson, Magnus, Luca Fumarco and Dan-Olof Rooth (2018). "Ethnic discrimination in hiring, labour
market tightness and the business cycle-evidence from field experiments". Applied Economics, 50(24): 2652-
2663. doi:10.1080/00036846.2017.1406653
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

nccr – on the move, Working Paper #20
31
Crabtree, Charles and Volha Chykina (2018). "Last Name Selection in Audit Studies". Sociological Science,
5: 2-21. doi:10.15195/v5.a2
Darity, William A. and Patrick L. Mason (1998). "Evidence on discrimination in employment: Codes of
color, codes of gender". Journal of Economic Perspectives, 12(2): 63-90. doi:10.1257/jep.12.2.63
Dench, Sally, Ron Iphofen and Ursula Huws (2004). An EU Code of Ethics for Socio-Economic Research.
Falmer, Brighton: The Institute for Employment Studies. Retrieved from:
http://www.respectproject.org/ethics/412ethics.pdf
Deutsche Gesellschaft für Soziologie (2017). "Ethik-Kodex der deutschen Gesellschaft für Soziologie (DSG)
und des Berufsverbandes deutscher Soziologinnen und Soziologen (BDS)". Retrieved from:
https://www.soziologie.de/de/die-dgs/ethik/ethik-kodex/
Diekmann, Andreas, Ben Jann and Mathias Näf (2014). "Wie fremdenfeindlich ist die Schweiz? Fünf
Feldexperimente über prosoziales Verhalten und die Diskriminierung von Ausländern in der Stadt Zürich
und der Deutschschweiz". Soziale Welt, 65: 185-199.
Fibbi, Rosita (2018). "Non più 'Cincali': i processi all'origine della metamorfosi della percezione degli
italiani in Svizzera", in Scaffai, Niccolò and Nelly Valsangiacomo (Eds.), À l'italienne. Narrazioni
dell'italianità dagli anni Ottanta a oggi. Roma: Carocci editore. pp. 199-214.
Fibbi, Rosita, Bülent Kaya and Etienne Piguet (2003). Le passeport ou le diplôme ? Etude des
discriminations à l’embauche des jeunes issus de la migration. Neuchâtel: Forum suisse pour l'étude des
migrations et de la population.
Fryer, Roland G. and Steven D. Levitt (2004). "The causes and consequences of distinctively black names".
The Quarterly Journal of Economics, 119(3): 767-805. doi:10.1162/0033553041502180
Gaddis, S. Michael (2017a). "Racial/Ethnic Perceptions from Hispanic Names: Selecting Names to Test for
Discrimination". Socius, 3: 1-11. doi:10.1177/2378023117737193
Gaddis, S. Michael (2017b). "How Black are Lakisha and Jamal? Racial Perceptions from Names Used in
Correspondence Audit Studies". Sociological Science, 4: 469-489. doi:10.15195/v4.a19
Green, Eva G. T. et al. (2011). "Nationalism and patriotism as predictors of immigration attitudes in
Switzerland: A municipality level analysis". Swiss Political Science Review, 17(4): 369-393.
doi:10.1111/j.1662-6370.2011.02030.x
Guglielmo, Thomas (2003). "'No color barriers’’: Italians, race, and power in the United States", in
Guglielmo, Jennifer and Salvatore Salerno (Eds.), Are Italians White? New York: Routledge. pp. 29-43.
Hagendoorn, Louk (1995). "Intergroup biases in multiple group systems: The perception of ethnic
hierarchies". European review of social psychology, 6(1): 199-228. doi:10.1080/14792779443000058
Hainmueller, Jens and Dominik Hangartner (2013). "Who Gets a Swiss Passport? A Natural Experiment in
Immigrant Discrimination". American Political Science Review, 107(01): 159-187.
doi:10.1017/S0003055412000494
Heckman, James J. (1998). "Detecting discrimination". The Journal of Economic Perspectives, 12(2): 101-
116. doi:10.1257/jep.12.2.101

nccr – on the move, Working Paper #20
32
Heckman, James J. and Peter Siegelman (1993). "The Urban Institute Audit Studies: Their Methods and
Findings", in Fix, Michael and Raymond Struyk (Eds.), Clear and Convincing Evidence: Measurements of
Discrimination in America. Washington, DC: The Urban institute Press. pp. 187-258.
Helbling, Marc (2010). "Islamophobia in Switzerland: A new phenomenon or a new name for xenophobia",
in Hug, Simon and Hanspeter Kriesi (Eds.), Value change in Switzerland. Lanham: Lexington Books. pp. 65-
80.
Helbling, Marc (2011). "Why Swiss-Germans Dislike Germans". European Societies, 13(1): 5-27.
doi:10.1080/14616696.2010.533784
Huddleston, Thomas et al. (2015). Migrant Integration Policy Index 2015: Switzerland. Barcelona/Brussels:
CIDOB and MPG.
Ignatiev, Noël (1995). How the Irish became white. New York: Routledge.
Imdorf, Christian (2008). "Migrantenjugendliche in der betrieblichen Ausbildungsplatzvergabe–auch ein
Problem für Kommunen", in Bommes, Michael and Marianne Krüger-Potratz (Eds.), Migrationsreport 2008:
Fakten - Analysen - Perspektiven. Frankfurt am Main: Campus Verlag. pp. 113-158.
Kaas, Leo and Christian 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
Krings, Franciska et al. (2014). "Selective incivility: Immigrant groups experience subtle workplace
discrimination at different rates". Cultural Diversity and Ethnic Minority Psychology, 20(4): 491-498.
doi:10.1037/a0035436
Krings, Franciska and José Olivares (2007). "At the doorstep to employment: Discrimination against
immigrants as a function of applicant ethnicity, job type, and raters' prejudice". International Journal of
Psychology, 42(6): 406-417. doi:10.1080/00207590701251721
Longchamp, Claude et al. (2014). Verbreitung und Entwicklung von Rassismus, Fremdenfeindlichkeit,
Muslimfeindlichkeit und Judenfeindlichkeit - Schlussbericht zur Studie "Zusammenleben in der Schweiz
2010-2014". Bern: Gfs Bern. Retrieved from: http://www.gfsbern.ch/de-ch/Detail/zusammenleben-in-der-
schweiz-2010-2014
Manatschal, Anita (2015). "Switzerland–Really Europe's Heart of Darkness?". Swiss Political Science
Review, 21(1): 23-35. doi:10.1111/spsr.12148
Matser, Carine et al. (2010). "Impact of relative size and language on the attitudes between nations and
linguistic groups: The case of Switzerland". Applied Psychology, 59(1): 143-158. doi:10.1111/j.1464-
0597.2008.00369.x
National Committees for Research Ethics in Norway (NESH) (2006). "Guidelines for Research Ethics in the
Social Sciences, Law and the Humanities". Retrieved from:
https://www.etikkom.no/globalassets/documents/publikasjoner-som-pdf/guidelines-for-research-ethics-in-
the-social-sciences-law-and-the-humanities-2006.pdf
OECD (2012). Jobs for Immigrants (Vol. 3): Labour Market Integration in Austria, Norway and
Switzerland. Paris: OECD Publishing. Retrieved from: http://dx.doi.org/10.1787/9789264167537-en
OECD (2018). "Harmonised unemployment rate". Retrieved from: https://data.oecd.org/unemp/harmonised-
unemployment-rate-hur.htm

nccr – on the move, Working Paper #20
33
Oudenhoven, Jan Pieter van, Eva Selenko and Sabine Otten (2010). "Effects of country size and language
similarity on international attitudes: A six-nation study". International Journal of Psychology, 45(1): 48-55.
doi:10.1080/00207590902914069
Pager, Devah (2007). "The Use of Field Experiments for Studies of Employment Discrimination:
Contributions, Critiques, and Directions for the Future". Annals of the American Academy of Political and
Social Science, 609: 104-133. doi:10.2307/25097877
Pecoraro, Marco and Didier Ruedin (2016). "A Foreigner Who Doesn't Steal My Job: The Role of
Unemployment Risk and Values in Attitudes towards Equal Opportunities". International Migration Review,
50(3): 628–666. doi:10.1111/imre.12162
Phelps, Edmund S (1972). "The statistical theory of racism and sexism". The American Economic Review,
62(4): 659-661.
Phlippen, Sandra and Bart van Eldert (2017). "Nederlander heeft liefst een Duitser op de werkvloer".
Algemeen Dagblad, Retrieved from: https://www.ad.nl/economie/nederlander-heeft-liefst-een-duitser-op-de-
werkvloer~a7e6d172/
Protsch, Paula and Heike Solga (2017). "Going across Europe for an apprenticeship? A factorial survey
experiment on employers’ hiring preferences in Germany". Journal of European Social Policy, 27(4): 387-
399. doi:10.1177/0958928717719200
Quillian, Lincoln (2006). "New Approaches to Understanding Racial Prejudice and Discrimination". Annual
Review of Sociology, 32: 299–328.
Quillian, Lincoln, John Lee and Mariana Oliver (2018). Meta-Analysis of Field Experiments Shows
Significantly More Racial Discrimination in Job Offers than in Callbacks. Evanston: Northwestern
University
Quillian, Lincoln et al. (2016). "The persistence of racial discrimination: A meta-analysis of field
experiments in hiring since 1972". Unpublished paper, Northwestern University.
Quillian, Lincoln et al. (2017). "Meta-analysis of field experiments shows no change in racial discrimination
in hiring over time". Proceedings of the National Academy of Sciences, 114(41): 10870-10875.
doi:10.1073/pnas.1706255114
Rapp, Carolin (2015). "More diversity, less tolerance? The effect of type of cultural diversity on the erosion
of tolerance in Swiss municipalities". Ethnic and racial studies, 38(10): 1779-1797.
doi:10.1080/01419870.2015.1015582
Raymann, Ursula (2003). Meinungen und Einstellungen gegenüber Ausländerinnen and Ausländern in der
Schweiz. Zürich: GfS-Forschungsinstitut.
Reskin, Barbara F. (2003). "Including Mechanisms in Our Models of Ascriptive Inequality". American
Sociological Review, 68: 1-21.
Riach, Peter A. and Judith Rich (2002). "Field Experiments of Discrimination in the Market Place". The
Economic Journal, 112(483): F480-F518. doi:10.2307/798458
Rich, Judith (2014). What Do Field Experiments of Discrimination in Markets Tell Us? A Meta Analysis of
Studies Conducted since 2000. Bonn: IZA.
Rich, Judith (2018). "Do photos help or hinder field experiments of discrimination?". International Journal
of Manpower, 39(4): 502-518. doi:10.1108/IJM-09-2017-0242

nccr – on the move, Working Paper #20
34
Ruedin, Didier et al. (2013). Migrationsängste der Schweizer Bevölkerung. Neuchâtel: Swiss Forum for
Migration and Population Studies. Retrieved from:
https://www.unine.ch/files/live/sites/sfm/files/listes_publicationsSFM/Studie_Migrationsaengste_sfm.pdf
Schneider, Jan, Ruta Yemane and Martin Weinmann (2014). Diskriminierung am Ausbildungsmarkt:
Ausmass, Ursachen und Handlungsperspektiven. Berlin: Forschungsbereich beim Sachverständigenrat
deutscher Stiftungen für Integration und Migration (SVR).
Schnell, Philipp and Rosita Fibbi (2016). "Getting Ahead: Educational and Occupational Trajectories of the
‘New’ Second-Generation in Switzerland". Journal of International Migration and Integration, 17(4): 1085-
1107. doi:10.1007/s12134-015-0452-y
SECO (2018). "La situation sur le marché du travail en décembre 2017". Retrieved from:
https://www.seco.admin.ch/seco/fr/home/Publikationen_Dienstleistungen/Publikationen_und_Formulare/Ar
beit/Arbeitslosenversicherung/Die_Lage_auf_dem_Arbeitsmarkt/Lage_Arbeitsmarkt_2017/alz_dezember_2
017.html
Snellman, Alexandra (2007). Social hierarchies, prejudice, and discrimination. Acta Universitatis
Upsaliensis, Uppsala. Retrieved from: http://uu.diva-portal.org/smash/get/diva2:170901/FULLTEXT01.pdf
Stolz, Jörg (2005). "Explaining Islamophobia. A Test of Four Theories Based on the Case of a Swiss City".
Swiss Journal of Sociology 31(3): 547-566.
Weichselbaumer, Doris (2016a). Discrimination Against Female Migrant Wearing Headscarves. Bonn:
Institute for the Study of Labor (IZA).
Weichselbaumer, Doris (2016b). "Discrimination Against Migrant Job Applicants in Austria: An
Experimental Study". German Economic Review, 18(2): 237-265. doi:10.1111/geer.12104
Widner, Daniel and Stephen Chicoine (2011). "It's All in the Name: Employment Discrimination Against
Arab Americans". Sociological Forum, 26(4): 806-823. doi:10.1111/j.1573-7861.2011.01285.x
Wood, Martin et al. (2009). A Test for Racial Discrimination in Recruitment Practice in British Cities.
Norwich: Department for Work and Pensions. Retrieved from:
http://webarchive.nationalarchives.gov.uk/20130314010347/http:/research.dwp.gov.uk/asd/asd5/rrs-
index.asp
Zschirnt, Eva (2018). Ethnic discrimination in the Swiss labour market - Ethnic hierarchies in
correspondence test results (PhD). Université de Neuchâtel, Neuchâtel. Retrieved from:
http://doc.rero.ch/record/323279
Zschirnt, Eva (2019). "Research Ethics in Correspondence Testing: An Update". Research Ethics (Online
first). doi:10.1177/1747016118820497
Zschirnt, Eva and Didier Ruedin (2016). "Ethnic Discrimination in Hiring Decisions: A Meta-Analysis of
Correspondence Tests 1990 –2015". Journal of Ethnic & Migration Studies, 42(7): 1115-1134.
doi:10.1080/1369183X.2015.1133279
!

nccr – on the move, Working Paper #20
35
Appendix
Figure 1: Example of CV and the cover letter of the Swiss-origin candidate
!

nccr – on the move, Working Paper #20
36
!

nccr – on the move, Working Paper #20
37
Figure 2: Example of the CV and the cover letter of the immigrant descent candidate

nccr – on the move, Working Paper #20
38