Lincoln Quillian, Anthony Heath, Devah Pager, Arnfinn H. Midtbøen, Fenella Fleischmann, Ole Hexel
Sociological Science, June 17, 2019
10.15195/v6.a18
Abstract
Comparing levels of discrimination across countries can provide a window into large-scale social and political factors often described as the root of discrimination. Because of difficulties in measurement, however, little is established about variation in hiring discrimination across countries. We address this gap through a formal meta-analysis of 97 field experiments of discrimination incorporating more than 200,000 job applications in nine countries in Europe and North America. We find significant discrimination against nonwhite natives in all countries in our analysis; discrimination against white immigrants is present but low. However, discrimination rates vary strongly by country: In high-discrimination countries, white natives receive nearly twice the callbacks of nonwhites; in low-discrimination countries, white natives receive about 25 percent more. France has the highest discrimination rates, followed by Sweden. We find smaller differences among Great Britain, Canada, Belgium, the Netherlands, Norway, the United States, and Germany. These findings challenge several conventional macro-level theories of discrimination.
Comparing levels of discrimination across countries can provide a window into large-scale social and political factors often described as the root of discrimination. Because of difficulties in measurement, however, little is established about variation in hiring discrimination across countries. We address this gap through a formal meta-analysis of 97 field experiments of discrimination incorporating more than 200,000 job applications in nine countries in Europe and North America. We find significant discrimination against nonwhite natives in all countries in our analysis; discrimination against white immigrants is present but low. However, discrimination rates vary strongly by country: In high-discrimination countries, white natives receive nearly twice the callbacks of nonwhites; in low-discrimination countries, white natives receive about 25 percent more. France has the highest discrimination rates, followed by Sweden. We find smaller differences among Great Britain, Canada, Belgium, the Netherlands, Norway, the United States, and Germany. These findings challenge several conventional macro-level theories of discrimination.
 | This work is licensed under a Creative Commons Attribution 4.0 International License. |
- Citation: Quillian, Lincoln, Anthony Heath, Devah Pager, Arnfinn H. Midtbøen, Fenella Fleischmann, and Ole Hexel. 2019. “Do Some Countries Discriminate More than Others? Evidence from 97 Field Experiments of Racial Discrimination in Hiring.” Sociological Science 6: 467-496.
- Received: March 7, 2019
- Accepted: April 23, 2019
- Editors: Jesper Sørensen, Olav Sorenson
- DOI: 10.15195/v6.a18
If the study were conducted in other continents the results would be the same. Why the narrow focus on North America and Europe? Are you trying to suggest racism is a “western” concept?
The countries are selected based on data availability. The nine countries in our analysis have at least three field experiments of racial or ethnic discrimination in hiring – these are all the countries in the world that meet this criterion. See page 474 of the article.
Based on what was written, it wasn’t meant to suggest it is a “western” concept. That was more of an assumption you got. As you go through what was written, you see other countries are named. It’s not like Europe and North America were the only countries the writer listed. Those are the ones the writer probably just had more information on at the time. So it is everywhere that discrimination happens. Some places more than others. You can’t just assume or pick apart a text and only see what you are wanting to see. Many other places were listed.
Europe and North America is where the majority of white people live and the largest amount of discrimination is not against WHITEs but it’s white against other ethnicities particularly blacks. Whites pretty much are the authors of racism. there’s a difference between people not liking each other culturally they don’t agree with different practices of another ethnicity however whites are the one who created that superior inferior aspect that is what is classified as racism that’s why there’s so many white supremacists white nationalists and they believe that blacks are inferior furthermore other races are inferior but they really have a thing for Black people.
Here in Brazil we also discriminate immigrants and refugees. As a teacher in a public school I have to protect students from Venezuela who suffer discrimination and also some muslim girls who cover her heads. Now we are receiving refugees from Syria who suffer discrimination and I’m always teaching my students to be more tolerant and polite with them.
In endnote 4, this article incorrectly states that a “problematic aspect of Zschirnt and Ruedin’s (2016) analysis is that they treat subeffects from studies as though they are independent discrimination estimates” (p.490). The note expands on the dependencies within studies and mentions similarities among “testers”. In our meta-analysis, we did account for the described lack of independence, and the remark on testers would not apply because we only included (written) correspondence tests.
Hi Didier.
Can you describe how dependence was accounted for in your study? Was there something like standard errors calculated accounting for clustering at the study level? I just looked again at your study and I cannot find any mention of clustering of standard errors in the text or supplementary materials (am I missing it?) In some analyses in your paper (e.g. figure 1) you seem to use the study-level N, which is appropriate. But in other analyses (e.g. tables 2, 3, 4, 5) you seem to use the much larger sub-effect n, and these sub-effects within a study will not be statistically independent because of the use of many common procedures (e.g. similar elements of resumes, similar ways of finding jobs, etc.) across subgroups within the same study.
When we refer to testers in the footnote, the points we make apply just as well to correspondence tests. In retrospect, our use of “testers” in the footnote was not the best choice of term because in some uses it suggests in-person audits, but our points apply as well to correspondence studies. Many studies that had multiple field sites, for instance, follow a lot of the same procedures in how they generate resumes across field sites.
If we did not understand your procedures correctly, I am happy to be corrected on the procedures you actually followed. And in general we think your study is important and (other than this one issue) well-conducted.
Thanks for your response, Lincoln! When I wrote “account for”, this refers to multilevel models (random intercepts for each study), as provided by the metafor package in R. However, we did choose to present raw numbers (mean, median) in the tables because the substantive results remain. I’ll have to take the blame for this misunderstanding since I did not make this clear enough in our paper.
Every country has some level of discrimination. Surprised France was first. Overall probably the least diverse countries discriminate the most.
Doing comparative research is highly commendable, and studies of labour market discrimination of minorities are important in light of recurrent signs of racism around the world. However, comparing discrimination across countries is a challenging task, and one that Quillian et al. inadvertently show to be all but impossible, given the data at hand.
Knowing that discrimination varies between minority groups, the authors attempt to show comparable figures from ostensively similar studies in different countries. As someone who is doing comparative research on minorities, I was curious about how different European countries would come out. So, I read to my astonishment that “France has the highest discrimination rates, followed by Sweden.” In fact, Sweden has “…a discrimination ratio about 30 percent higher than that of the United States”. That does sound bad. But all studies in Sweden pertain to MENA (mostly Arabic) names, and Figure 1 reveals that the only US study of MENA minorities finds a discrimination rate (the call-back ratio between two groups of job applicants) of 2.8, while it is 1.65 in Sweden and 1.78 in France! So, in this case – the only one that is comparable across these three countries – discrimination in the US appears much greater, although less precisely measured. Moreover, it is sobering to compare the estimate across countries for the MENA category: There are hardly any significant differences between the countries that have done studies on this minority: USA, Belgium, Canada, Germany, France, Sweden, or the Netherlands (all confidence intervals but France’s cover 1.5, France being very close, too).
So what is the trick? Well, the US discrimination estimate turns out to depend heavily on studies of the black minority, with its particular historical and institutional connotations, and which is hardly comparable to any European immigrant group (blacks in the US turn out to experience relatively low levels of discrimination, which is interesting in itself). The upshot is that the country differences estimated in Table 3 are based on information on different minority groups in different countries, making the reported country differences either fictitious (e.g., the coefficients for Germany, Sweden and Norway in Table 3 give their difference to the US in the log discrimination rate of Blacks, despite the fact that there are no studies of Blacks [of any sort] in these countries) or in any case highly uncertain (cf. the confidence intervals in Figure 2).
But the fact that the results are not supporting the conclusions is not the biggest problem. Instead, it is the apparent failure of the authors to recognise that discrimination studies of this type target one, or a few, select minority groups in each destination country; groups chosen because researchers assume discrimination to be particularly likely against them, or because the group is of political or other importance. The authors claim to measure “national discrimination levels” but the groups covered by the included studies do NOT represent ‘immigrants’ or ‘minorities’ in general. In some cases, they do not cover more than a fraction of all minorities. By letting the measure of discrimination depend on which groups happen to appear in existing studies, discrimination against some minorities, such as Muslims in the US, is made invisible, and the overall level of discrimination is unknown and may be vastly underestimated – in other cases overestimated. To draw general conclusions about cross-country differences in discrimination on the basis of these data is thus highly misleading. The (unqualified) conclusion that “France is the country with the highest level of discrimination, followed by Sweden. By contrast, Germany, Norway, and the United States have lower rates of discrimination” does not only defy the results presented, they represent a generalisation far beyond the scattered tests that exist: Sweden with (only) MENA immigrants (mostly refugees), Germany with (only) Turkish and Norway with (only) Pakistani labour market immigrants, France with black and non-black North African immigrants – all compared with the numerous US studies of their black minority (and some of Latinos).
It is clear that studies of non-white minorities, in all countries, show evidence of blatant discrimination, but it is equally clear that comparing “national discrimination levels” on the basis of these studies is simply not possible. In light of this, the policy speculations in the article may send policy makers in totally erroneous directions. This is serious and worrying: While it may seem easy to shrug these conclusions off, the article is published in a scientific journal and even unfounded conclusions may travel far beyond its borders.
There may be other/complementary factors explaining the relatively low discrimination in Germany beside detailed requirements for information about applicants’ qualifications and the close articulation between education, training, and employment. Access to the German labour market is a two-stage process for skilled workers in case they have completed an apprenticeship: They first have (had) to find an employer how offers (offered) them an apprenticeship place before they transition from apprenticeship to a skilled job. In contrast to skilled job markets, apprenticeship markets in German-speaking countries (Germany, Switzerland, Austria) are hardly regulated and the information available to employers about (rather young) candidates for apprenticeship is less robust compared to the information they get from candidates for skilled jobs.Therefore apprenticeship markets can be expected to be more prone to discrimination compared to skilled job markets in the above mentioned countries, especially when competition for apprenticeships is high. There are however only very few experimental studies on discrimination in apprenticeship markets so far.
Interesting point. Thank you.
In Trinidad blacks vote for one party and the indians vote for another. They indians say the blacks cant run buisness6and are stupid. The Blacks say Indians are corrupt and commit fraud. In Africa white people are killed for being white and their land is taken. In India whites and blacks sre seen as less then and are targeted in sex trafficking. Dont even get me started on China with black people. Or No go zones in sweden and france where if your white or black don’t enter, middle eastern only- They patrol the streets at night if your seen drinking or dressing sexually (goes against their muslim culture) they harass and sometimes assult you. No wonder they are higher for discrimination. France and sweden also have admitted they are taking in more immigrants then they can/should because of the EU causing areas to change drastically over short periods of time. To focus only on white racism, is cherry picking and frankly discriminatory. Racism happens everywhere from EVERY race. However in western society and Europe we take racism mkre seriously and have many laws about hate and discriminatory speech. It has gotten much much better in those societies.
Ciara Jenks in which part of the continent are you from…because clearly you are not from Africa. I am from South Africa and i can affirm that white(boers) are not being killed and stripped off their land. It is unwise to speak about something that you know nothing of. I am a black woman and let me tell you what being a black woman means for certain minorities individually. I am not only judged and discriminated by the colour of my skin but i am also judged for being a woman…In South Africa like in other countries being black means that you are an Ape, it means you are incapable, it means you are not worthy. I have experienced first hand on racism from some white people (particularly boers) drastically hanging and killing our people(fathers and mothers) who work on the farms and discriminating against some who work as domestics or labourers, white farmers are not being intentionally targeted, black people are just as tired Ciara… Indians at work dominating the entire company and isolating black people including myself regardless of holding a superior position in the company.
It is easier to speak without knowing the harsh realities that people face daily. So if people are conducting such reports, i feel like collecting information based on hearsay is not enough…Highly ranked officials and these thesis doctors amaze me because when one reads this information that they provide, they speak in big words about realities that they know nothing about.
So yes Ciara…about one thing that you are correct about, racism does exist everywhere. Boers immigrated to our lands. Africa belongs to the Apes but because the Apes were too kind and welcoming, they were invaded. The Boers have no claim over these lands…if i were to immigrate to Europe or any western or eastern country, I’d have no claim over those lands because i belong to the African continent. That is where my ancestral clan comes from. So maybe America and Europe and whatever western countries reserve the rights to admit into their whoever that they want to admitt.
But need i remind you that no black person in history anciently went to these countries willingly, they were taken from their lands (these African lands) to be sold as slaves to masters.
White people deliberately invaded African because they were after wealth. We knew nothing about wealth, we were just a bunch of fools busy sticking spears into each other, “as so they say”. We had natural resources and minerals and lands and everything. Ohhh and than we were invaded. We were invaded by whites, indians, Chinese…
Look at us now, come to our african communities and we are the poorest among every other race, followed by colored people.
So what’s the solution…it’s to eradicate discrimination and racism, to love and care for each other and to see each other as human beings. But do you think that will ever happen. No it will never!
White boers in SA tormented black people for years and just imagine, when Democracy came black people never not even once retaliated. Yet we still suffer through the hands of the same people. Wouldn’t you be tired and perhaps angry that even after everything the same people show absolutely no remorse.
The West is rich and wealthy particularly because of the wealth that they gain from our markets. China always calls our stupid so called african leaders to patronize them by pleading that they open our markets. Yet Chinese even here in SA isolate against our Africans and when you shop in their shops, they treat you as if you are a beggar in your own lands with your own money that they need.
On top of that, all these invaders divide us by employing our African brothers and sisters and saying that they are better.They work better than South Africans etc. Than they come back to suggest that foreigners are taking up our jobs as South Africans. And because we are a bunch of fools that cannot even see when we are being played and used. The invaders know that if African were to trully unite, it would be powerful but we can’t see that.
Same thing happens…if an African american comes to Africa, Africans do not regard him or her to be African enough. Again if an African from Africa goes to America…African americans see an African with an empty head and not equally worthy because they come from dirt (A place of sticking spears into each other and wearing cow skin), a place of poverty and slums…honestly you would swear that you stink. African English is not as eloquent or fluent in pronunciation so they are not as superior to African Americans.
Thank you Anitta for the time and compassion you have expressed in your response.
I agree with all that you have shared and find it disheartening that within the Western world those who choose to fight anti-black racism get applauded and treated as heroes for finally admitting that maybe they do have privilege. Then get awarded jobs as Diversity and Inclusion officers over Black citizens, whose lived experiences and knowledge about what organizations need to do to correct their policeis, remain more relevant. When and how will minorities actually get the opportunity to advance?
Academia awards themselves degrees in order to take positions over minorities. When Blacks obtain the same graduate degrees they still cannot find employment. One study showed that 13.4% of Blacks in Montreal, Canada with graduate degrees remained unemployed; which was similar only to 12% of non-Blacks without a high school diploma. It’s a sad world Anitta; where hate becomes justified and validated through policies and systems. Writing a paper about it doesn’t fix it. I wish it did.. if so I’d write forever…