I notice that often a lot of Chinese mathematicians collaborate with other Chinese, and Americans collaborate with other Americans. Many of the journal papers that I look up seem to indicate this sort of self-segregation too.

Is this real or perceived?

If it's real, is it due to racism, familiarity with people of one's own background, or perhaps a combination of various factors?

  • 3
    What are you looking for in an answer to this question? I suspect it's mostly a combination of geography and language. But my guess does not an answer make.
    – Thomas
    Sep 19, 2018 at 5:33
  • 2
    All my collaborators are Italian, simply because my supervisor is Italian. If he were Chinese, or if we were working in China, I suspect I'd have a lot more Chinese collaborators. There's no deeper cause or reason here. Sep 19, 2018 at 7:16
  • 3
    A slightly provocative question: do you see the same happening outside of academia? I doubt academia is any special in this regard.
    – Ambicion
    Sep 19, 2018 at 9:19
  • 3
    I don't think it makes a lot of sense to answer a request for data and sources with "your claims are unsourced and lack data", @JonCuster... It's incredible that everyone is jumping on the bandwagon, attacking OP, defending their turf, finding excuses... It's a valid question.
    – user9646
    Sep 19, 2018 at 19:42
  • 3
    @JonCuster I don't see what the problem is here, the OP clearly started by asking if this perception was real or perceived. I don't think asking if something could be linked to racism is "playing the racism card". Sep 19, 2018 at 22:22

5 Answers 5


I seriously doubt that racism plays much part in any effect you might think you see. I also question what you think is the evidence.

There might be some small effect based on language, not race. If a couple of academics at a US institution share a language other than English, they may be drawn together to ease communication. But such an effect should be short lived.

The fact is that the majority of US academics in STEM fields in the US are white males. The person in the next office is likely to be a white male. While there are issues with that, I don't see much racism in evidence within the academy. As with anything, there can be exceptions and a few noted scientists and mathematicians were notably racist. But they stand out.

The article cited by user Carl Christian on general self segregation can be easily explained. In the college first year student example, the self selection was going on before the students actually had a chance to meet anyone. So their self segregation was a result of factors outside the university not inside it. In the US, much of the force driving racism is fear of the other. We fear people who we haven't met and don't know. When a white and a black person pass one another on the street at night, both feel fear. But this fear is caused by the pervasive separation of the races in neighborhoods and schools. People have been actively discouraged from interacting in normal ways.

Fear of the other is probably much influenced by evolution. People in truly ancient times had to be wary when the met another person (from a different tribe) or an unfamiliar animal. Until you had some knowledge of how they would react, you had no basis for trust.

But within a university department, people do know one another and see one another each day and so have no reason to fear them. So, if you need to collaborate with someone you look first to the next office over, not thinking of them as different. They are just fellow academics. Likely they will be white men, of course. But if they have similar interests, regardless of race or sex, you will be glad to collaborate.

If racism and sexism exists in academia, and they do, it is much more likely to be evident at the point of entry, either for students or academic jobs. But after people come to know one another, assuming that is permitted and encouraged, the fear of the other goes away and we can just accept that people are people. The army doesn't give you the option of not participating. Initially you are forced to work with people not like you. But mutual trust is required for the job.

Since it wasn't referred to in the question, I didn't discuss the notion of racism toward students by faculty. Such racism might be conscious or unconscious, but it has a damaging effect when it occurs, regardless of cause.

  • 4
    Your post mentions fear of the other, but sort of misses the counterpart which is comfort with the familiar. The under-representation of non-white males in positions of power within academia has long been identified as a problem not just for the inequities it represents but for the subsequent generations of academics that are lacking mentors that look like them and share more similar life experiences.
    – Bryan Krause
    Sep 19, 2018 at 21:12
  • @BryanKrause, correct, certainly after your first sentence. I don't think comfort explains racism, however. It would explain indifference, perhaps, but the racists we see on the screen don't seem comfortable in any sense. But I still think that the effect is primarily at the "gate". Fortunately, that problem is, in theory, easier to address, though there are forces that oppose even positive moves to be more open in education.
    – Buffy
    Sep 19, 2018 at 21:43
  • 2
    I didn't suggest it explains racism, I meant that it's an explanation for the phenomenon the question asker is talking about. Your answer says you think racism only happens at the point of entry and is moot after that, but you don't give much evidence or support for the latter.
    – Bryan Krause
    Sep 19, 2018 at 21:48
  • 2
    I think that's a hard statement to make unless you are a person of color, which you may be, I can only presume you have a preference towards canines. I don't have as much time as you in academia and am not a person of color, but I'd be extremely wary to say racism doesn't exist within academia (even if limited to 'past the gates'). Though I do agree with you that it's probably not the cause of the phenomenon the OP is describing.
    – Bryan Krause
    Sep 19, 2018 at 22:12
  • 1
    @BryanKrause you've made an assumption. I'm not, in fact, a dog person.
    – Buffy
    Sep 19, 2018 at 22:25

There probably is some self-segregation, but nothing above what people normally do in everyday society.

I think it's worth keeping in mind that much of the dialogue regarding segregation deals with the actions of government, an extremely powerful 3rd party. Self-segregation is at best an ethical issue, not legal. Moreover, everyone has a right to free association and can decide who they want to deal with.

Arguably, I would say that academia self-segregates less than society at large, due to more liberal attitudes of academics and students.

  • 3
    Is there data to support the assertion that self-segregation in academia is less, or no greater, than in society at large? Everyone has their own observations and experiences, but it's hard to know whether or not they are representative. Sep 22, 2018 at 19:06
  • @NateEldredge It appears there is ample data to attest that segregation in STEM is most likely greater than in any other segment of society (finance, entertainment etc...). It shouldn't be too hard for instance to obtain the data about the number of PhD's in mathematics in any given year by race, unless you do not consider this to be data. There are also several articles out there from African American mathematicians about the virulence of racism in mathematics department. Have fun reading! Jul 18, 2020 at 20:20

From my observations, there's some self-segregation. It's not so much because of racism but rather the feeling that [outsider] is not [one of us]. For example, if you collaborate with some Chinese nationals who talk to each other in Chinese (and you don't speak that language), then that's a subtle sign that you're not one of them. This applies even if they speak fluent English and default to that language when speaking to you. There's nothing unique about Chinese either - it could equally be German, French, or whatever language.

Another example of this is religious beliefs, when they manifest in real life. If I'm Muslim (and some races are certainly more likely to be Muslim), then there're certain things I cannot eat, and I certainly cannot take alcohol. If I go to a social event where everyone is drinking alcohol, then I'll definitely get the feeling that I'm not one of them.


I am not sure if people are "segregating" their collaborators. However, I do see some patterns in research departments and groups in STEM fields which may lead to what you see in those collaborations.

My particular research group had a lot of ethnic diversity, however, I have noticed that some other groups or even department have a tendency to have a lot of senior researchers and students of very consistent ethnic background (to the point of having similar names).

Some salient examples that I observed in my particular university.

  • Some groups in theoretical computer science seems to have a huge amount of researchers with Greek background and they all collaborate with each other. Just look up some research paper in this area.

  • In applied computing, there is a lot of Chinese researchers, and they all collaborate with each other or industries in East Asia like Huawei and Baidu.

  • In electronics, there seems to be a lot of researchers of Eastern European or Italian researchers or Middle Eastern backgrounds. I do see that, often, these professors taking on students with the same ethnic background.

  • In mathematics, a lot of researcher from Germany, also a lot of researchers from Israel. But it gets more interesting when you break down the particular fields...

Of course, this is just what I observed for some research groups repeatedly during my time in academia as a student and as a researcher.

What could be the explanation for what I saw?

There could be a confluence of factors here beyond bias or prejudice in picking collaborators or professors or students.

Cultural factors. For example, I think Italian engineering school place great emphasis on mechatronics. However, I don't know why there are so many famous Greek computer scientists (maybe someone can enlighten me). I think there is a strong tradition of math education in Germany, which may explain why there would be many German researchers in this field. Finally, I think East Asian culture place great value on tech entrepreneurship, which may explain why many would be doing applied computing.

This still doesn't resolve some of the other more interesting things that I saw, for example,

  1. the head of department being the same ethnicity as his replacement, who he mentored for a decade,

  2. a professor with a mixed-race child taking interest in a student who is also mixed-race (she was a fellow classmate), then taking that student as his student.

  3. a professor in applied computing with wife who is Asian having large amount of Asian female research students. Again, is it racial ethnic preference or self-selection? We'll never know.

  4. The funniest one. I actually remember two distinct research groups where all the members look like Abercrombie & Fitch models. In one group photo, where conference was taking place near a beach, the members of the research group even posed as if they were models. I might add that the research lead for this group was gay. It did cross my mind that whoever is leading the research group might have a bias for certain "look" to his/her students.

But at least it give me more assurance that this game is not totally rigged in North America.


Even in mathematics, it greatly varies from area to area etc. E.g. I co-authored papers with (in no particular order, mentioned by the country of origin rather than residence/citizenship) Australian, Indonesian, Indian, Chinese, Japanese, South African, Iranian, Israeli, Italian, French, Croatian, Greek, Belgian, Spanish, Dutch, German, UK, US, Canadian, Mexican, Bulgarian, Polish, Estonian, Russian mathematicians and computer scientists.

On the other hand I am probably not typical; I imagine a majority still stays in one country throughout the career and only collaborate in their mother tongue.

Disclaimer: I grew up in USSR, did my PhD in Australia, hold a Dutch passport and work in UK.

You must log in to answer this question.