Join 3,433 readers in helping fund MetaFilter (Hide)


Racial and gender biases in faculty mentoring
April 22, 2014 12:07 PM   Subscribe

"Faculty at private schools were significantly more likely to discriminate against women and minorities than faculty at public schools. And faculty in fields that were very lucrative were also more likely to discriminate. So there was very little discrimination in the humanities. There was more discrimination among faculty at the natural sciences. And there was a lot of discrimination among the faculty at business schools." (link to NPR story). Katherine Milkman and colleagues conducted a field experiment in which professors were contacted by fictional prospective doctoral students and found that "faculty ignored requests from women and minorities at a higher rate than requests from White males, particularly in higher-paying disciplines and private institutions."
posted by needled (95 comments total) 29 users marked this as a favorite

 
Sad but not totally surprising. It was years ago, but I had good experiences when I was calling and emailing professors at grad schools I was considering; it's depressing to consider how someone's name alone could change that.
posted by Dip Flash at 12:15 PM on April 22


The comments are interesting:
Finally, I have to note that Milkman's methodology is, at least in my case, extremely haphazard. When I received one of her 'scam' emails, it was from a student (whose name I do not recall - no, seriously!) praising my scholarship and seeking to pursue a Ph.D. under my direction. Flattering, but completely misinformed. While my department does offer a Ph.D. in History, it does so only in the field of US History, and I am not a US historian. So when I received the 'scam' email from this alleged student, I filed it away in my mind in the category of 'students who are so unserious as to not merit further consideration'; after all, had the alleged student spent any time on our department website (as all serious students do), he/she would have realized that doctoral work under my supervision was impossible. Of course my case may be exceptional. But I'm sure there are lots of comparable stories to be found among Milkman's data. After all, she is not interested in human beings and the complex motivations for their behavior; instead, she is interested only in comparing percentages of response rates. That, I'm afraid, is not very satisfying.
posted by Pogo_Fuzzybutt at 12:18 PM on April 22 [12 favorites]


The history professor fails to explain why that would lead to the observed differential between response rates, I notice.
posted by Mogur at 12:22 PM on April 22 [37 favorites]


Re the comment above: yeah, I, too ignore those emails when they come from someone who obviously hasn't done their homework. But if if I found out I'd been receiving identical emails, one per year, with different names, and I answered it if and only if the person's name didn't sound black, then yeah, I'd be pretty shaken by that comparison of percentages of response rates.
posted by escabeche at 12:24 PM on April 22 [8 favorites]


I don't know if one of my academic experiences would be considered discrimination, or at least strange. I majored in English (creative writing) in college and the Director of the program asked if I would be interested in becoming an assistant somewhere in the Ethnic Studies division, a totally separate college of the university. I was like ??? It didn't make any sense to me at the time and I refused. I wonder if I was right to not be interested and a little offended, or if I should not have looked a gift horse in the mouth.
posted by ChuckRamone at 12:25 PM on April 22


I agree, Pogo_Fuzzybutt, the comments on the NPR article are interesting. Someone named "Richard Barton" is super defensive about his failure to respond to an email from "Juanita Martinez."
posted by Bentobox Humperdinck at 12:25 PM on April 22


I think their use of language is appropriate. No explanation of the differential, but the expression of dissatisfaction with the methodology, which certainly doesn't help make a convincing case (as if one were needed). Something like that.
posted by emmet at 12:26 PM on April 22


The history professor fails to explain why that would lead to the observed differential between response rates, I notice.

Exactly. It's pretty safe to assume the white male applicant's query emails looked just as sketchy as the ones for the black or female candidates. So even if the candidates were perceived as "not serious," the black and female candidates were more likely to be dismissed outright on the basis of that lack of seriousness. I'm suprised an academic would be so sloppy with his/her reasoning, but then again, it is defensive reasoning, which is rarely a useful kind.
posted by saulgoodman at 12:28 PM on April 22 [22 favorites]


The history professor fails to explain why that would lead to the observed differential between response rates, I notice.

Something doesn't add up here. The linked NPR piece claims that the letters sent out to faculty were identical except for the name of the student, the comment that Pogo_Fuzzybutt quotes suggests that the email made errors about the Prof's research--which would mean that the emails were tailored (well or poorly depending on circumstances) for particular targets. Either that Prof. is remembering an email which was not, in fact, part of Milkman's research or Milkman has badly misdescribed the nature of the experiment.

Tailoring the emails to different profs would be a disaster, of course, because it would be impossible to tell if the response rates were reflecting the nature of that tailoring rather than the gender/race of the supposed sender.
posted by yoink at 12:28 PM on April 22 [1 favorite]


First off, we shoudn't discriminate against minority groups either, the paper is with coauthors Modupe Akinola and Dolly Chugh!

Also, I know Katy and her work well, and she is really good at designing experiments looking at these issues (and she has a lot of cool papers) with a behavioral economic lens. I hadn't read this paper before, but she isn't trying to get at causes; her research is a really solid way of getting at discrimination rates, not so much motivations.

Here is a fun talk she gave on how to get people to exercise more by bundling sins and pleasures together.
posted by blahblahblah at 12:30 PM on April 22 [14 favorites]


How would you even know if that comment quoted by Pogo_Fuzzybutt is from a real professor?
posted by ChuckRamone at 12:30 PM on April 22


the comment that Pogo_Fuzzybutt quotes suggests that the email made errors about the Prof's research

No, it suggests a generic letter based on department.
posted by kmz at 12:31 PM on April 22 [12 favorites]


Also, according to the paper, this is the text of the letter, only names changed. This should provide a random sample, controlling for department issue errors, since those would fall across all departments randomly:

Subject Line: Prospective Doctoral Student (On Campus Today/[Next Monday])

Dear Professor [Surname of Professor Inserted Here],

I am writing you because I am a prospective doctoral student with considerable interest in your research. My plan is to apply to doctoral programs this coming fall, and I am eager to learn as much as I can about research opportunities in the meantime.

I will be on campus today/[next Monday], and although I know it is short notice, I was wondering if you might have 10 minutes when you would be willing to meet with me to briefly talk about your work and any possible opportunities for me to get involved in your research. Any time that would be convenient for you would befine with me, as meeting with you is my first priority during this campus visit.

Thank you in advance for your consideration.
Sincerely,
[Student’s Full Name Inserted Here]
posted by blahblahblah at 12:32 PM on April 22 [18 favorites]


Oh yeah, this is a well-done letter from my perspective, because it's exactly the kind of thing I might answer and might not.
posted by escabeche at 12:33 PM on April 22 [11 favorites]


Thanks blahblahblah--clearly the Prof Pogo_Fuzzybutt cites was misremembering the email or remembering an unrelated email. This says nothing about wanting to pursue a Ph.D. in "US history" or anything like that. This is a very nicely designed research instrument. The results of the experiment are truly disturbing.
posted by yoink at 12:35 PM on April 22 [6 favorites]


Ah, yeah, that totally makes sense.

And actually rereading the original comment above there's literally zero indication that there was any tailoring to specific professors. The mistake wasn't about the professor's specific research, the mistake was that the department in question didn't offer a doctorate in the professor's field.
posted by kmz at 12:35 PM on April 22 [1 favorite]


This says nothing about wanting to pursue a Ph.D. in "US history" or anything like that.

You are a professor of US History. You receive a letter from a student interested in your research. The logical leap isn't so much a leap as a tiny tiny step.
posted by kmz at 12:37 PM on April 22 [5 favorites]


My institution has one Ph.D currently. It's in psychology.

If you were to email a business faculty and say you were interested in doing a PhD under them, you wouldn't get a response because you clearly didn't do your homework on this institution. And doing a PhD in business would be impossible because it doesn't exist.

Also, people who write these letters also try to sell themselves by saying things like: earned my undergraduate in blah blah blah, conducted an honors thesis on this this this, and then worked in the field as a that that that.

The email does look very scammy. So if that's what was sent out, and it was sent out to the top faculty in their fields in the country and there aren't doctoral programs under some of those faculty, then no, the email is probably not going to get a response.

There could have been value done in this study if that letter had been sent out by an undergraduate student because some lack of specificity would be acceptable coming from an 18 or 19 year old incoming/prospective student compared to a 22+ adult looking into a doctoral program.
posted by zizzle at 12:38 PM on April 22 [2 favorites]


And there was a lot of discrimination among the faculty at business schools.

Every time one tries to talk to hardcore libertarians about inequality, one hears about how a truly free market would correct for something so irrational as racism or sexism... I remain skeptical.
posted by showbiz_liz at 12:38 PM on April 22 [11 favorites]


The mistake wasn't about the professor's specific research, the mistake was that the department in question didn't offer a doctorate in the professor's field.

That must be a weird-ass department. How do you get to be a professor in a department with a Ph.D. program and not be able to direct any doctorates pursued in your department? In any case, even if that is what was going on in that specific case, it's such a wacky outlier that it clearly does nothing to call Milkman's research into question.
posted by yoink at 12:39 PM on April 22 [1 favorite]


The mistake wasn't about the professor's specific research, the mistake was that the department in question didn't offer a doctorate in the professor's field.
The email doesn't assume that the department does offer a doctorate in the professor's field. That's an email that a student might send if he or she wanted a position as a research assistant, which is helpful for applications to PhD programs.

At any rate, the point here is that professors were more likely to respond to some names than to others. The emails were identical. I'm not sure why this professor's particular objections are relevant.
posted by ArbitraryAndCapricious at 12:39 PM on April 22 [11 favorites]


So if that's what was sent out, and it was sent out to the top faculty in their fields in the country and there aren't doctoral programs under some of those faculty, then no, the email is probably not going to get a response.

That would be a reasonable account of a low response rate across the board. It is not a reasonable account of differential response rates according to the presumed race/gender of the email sender.
posted by yoink at 12:40 PM on April 22 [36 favorites]


Yeah, no individual rejection by a professor can be taken as evidence that that specific professor is racist/sexist, there could be any number of legitimate reasons the letter was not responded to. But when the data in aggregate indicate a difference in response rates, it means that there is some bias, which could, and likely does, directly affect who is getting PhDs. I don't expect a US History professor to be a statistics expert, but the basic concept here shouldn't be that hard.
posted by Tsuga at 12:40 PM on April 22 [12 favorites]


Of course the whole comment is moot anyway... yes, there's always going to be exceptions and particular perfectly good reasons why some singular examples may have gone one way or another. But the aggregate statistics are still there, and the statistics are damning.
posted by kmz at 12:40 PM on April 22 [4 favorites]


Well, I for one would like to here from more of the faculty this letter was sent to.
posted by zizzle at 12:40 PM on April 22


zizzle,

Not of what you say invalidates the study. For there is no reason to assume nor is there any reason to suggest that the mismatch between the letter being sent and the receive is unevenly distributed, and thus skewing the results.
posted by MisantropicPainforest at 12:41 PM on April 22 [2 favorites]


This says nothing about wanting to pursue a Ph.D. in "US history" or anything like that.

You are a professor of US History. You receive a letter from a student interested in your research. The logical leap isn't so much a leap as a tiny tiny step.


Okay, here is the methodology, which addresses all your questions:
1) Identified 6,300 doctoral programs from the US News List
2) Randomly selected two tenure-track faculty from each (over 6,500 people)
3) Mail them the letter with randomly varying name
4) Half got the letter saying that people would be on campus today, half next week

So, some faculty may have been misidentified (not part of a PhD program, not on faculty anymore, etc), but any errors of this sort would be randomly distributed, so it would not effect results (it would be uncorrelated with race or gender). The design here is solid, since you are looking at differences by gender/race.
posted by blahblahblah at 12:41 PM on April 22 [34 favorites]


The design here is solid.

Oh, totally agreed. I was just trying to clarify something which honestly wasn't even relevant and was pretty much a derail. Apologies.

And doing a PhD in business would be impossible because it doesn't exist.

I guess all those Business schools with doctoral programs are just imaginary then.
posted by kmz at 12:45 PM on April 22 [4 favorites]


That must be a weird-ass department. How do you get to be a professor in a department with a Ph.D. program and not be able to direct any doctorates pursued in your department? In any case, even if that is what was going on in that specific case, it's such a wacky outlier that it clearly does nothing to call Milkman's research into question.

Because you hire people to teach various required service courses, while only offering PhDs in specific areas? They could also be offering terminal MA degrees, as well (Public History, for example, of in support of an education program).
posted by GenjiandProust at 12:49 PM on April 22


Also, people who write these letters also try to sell themselves by saying things like: earned my undergraduate in blah blah blah, conducted an honors thesis on this this this, and then worked in the field as a

For what it's worth, my emails and phone messages were very similar, something like:

Dear professor X, I am considering applying to your program and had some questions. When would be a good time to talk?

Best,
Whiteguy McWhiteyson

My response rate was 100 percent.

If I was writing those emails today I'd know to add more information (to be helpful if nothing else) but I didn't think of that at the time.
posted by Dip Flash at 12:49 PM on April 22 [6 favorites]


"VEDANTAM: Yeah. And one of the ways they try to do it David, is by attracting diverse faculty. And Milkman and her colleagues looked to see whether having diverse faculty protected against this kind of bias. Here she is again.

MILKMAN: There's absolutely no benefit seen when women reach out to female faculty, nor do we see benefits from black students reaching out to black faculty or Hispanic students reaching out to Hispanic faculty."


Wait, am I reading this wrong or does it say that women and black faculty showed the same tendencies as the rest of the faculty?
posted by asra at 12:49 PM on April 22


By the way, this was the name list. They checked with 38 people first to make sure the names identified people by gender/race

Brad Anderson
Steven Smith
Meredith Roberts
Claire Smith
Lamar Washington
Terell Jones
Keisha Thomas
Latoya Brown
Carlos Lopez
Juan Gonzalez
Gabriella Rodriguez
Juanita Martinez
Raj Singh
Deepak Patel
Sonali Desai
Indira Shah
Chang Huang
Dong Lin
Mei Chen
Ling Wong
posted by blahblahblah at 12:50 PM on April 22 [9 favorites]


Yes, just like that recent study where everyone, men and women both, were biased against women applicants for lab positions.
posted by Dip Flash at 12:51 PM on April 22 [8 favorites]


Although mentoring of any kind is uncommon in PhD programs for any student, beyond lip service. Mentoring that is meaningful and good and sustained is very rare overall.
posted by Bwithh at 12:51 PM on April 22 [1 favorite]


I guess all those Business schools with doctoral programs are just imaginary then.

An institution can offer PhDs without offering PhDs or even graduate degrees in specific disciplines or sub-disciplines.
posted by GenjiandProust at 12:52 PM on April 22


Yes, just like that recent study where everyone, men and women both, were biased against women applicants for lab positions.

Yeesh, depressing.
posted by yoink at 12:54 PM on April 22


Also, people who write these letters also try to sell themselves by saying things like: earned my undergraduate in blah blah blah, conducted an honors thesis on this this this, and then worked in the field as a that that that.


One thing to think about is who are the people you're thinking about when you say "the people who write these letters"?

I know my department has been working on trying to increase minority (defined in a lot of different ways, including race/gender/socioeconomic status/non-native, etc.) applicants/students, and one thing that we've been concerned about is that these students are also less likely to be acclimated into academic culture and might not know/might not have had a lot of guidance in terms of what these types of interactions are "supposed to be like". We don't want smart, capable people who are going to do kick-ass research to have a rougher time in the program because they don't know they're are/ are not supposed to do XYZ in terms of personal interactions with their advisors, other faculty members, other students, etc.

"These are the rules of interaction in academia and of course everybody knows them" is not a true statement, and the people excluded from "everybody" are already likely to be underrepresented in academia.
posted by damayanti at 12:54 PM on April 22 [17 favorites]


There was also the Swedish study of foreign name CV bias.
posted by ChuckRamone at 12:55 PM on April 22


Although mentoring of any kind is uncommon in PhD programs for any student, beyond lip service.

Whut?
posted by yoink at 12:55 PM on April 22 [1 favorite]


The study as a whole is really good and important work, but this crucial detail is actually the most interesting, to me:

Counterintuitively, the representation of women and minorities and bias were uncorrelated, suggesting that greater representation cannot be assumed to reduce bias.

This is a very important finding. I think people tend to blithely assume that representation actually does reduce bias, but that's not clearly the case in any field (including politics, where I think that misperception is exploited to an absurd degree). There are deeper logics at work, and those need to be recognized if we're ever going to move our institutions in the direction of real meritocracy.
posted by clockzero at 12:57 PM on April 22 [7 favorites]


Although mentoring of any kind is uncommon in PhD programs for any student, beyond lip service.

Whut?


It's not as common as you might think for professors to really mentor graduate students, for a number of complex reasons.
posted by clockzero at 12:57 PM on April 22 [2 favorites]


Wait, am I reading this wrong or does it say that women and black faculty showed the same tendencies as the rest of the faculty?
You're reading it right.
posted by ArbitraryAndCapricious at 1:01 PM on April 22 [1 favorite]


clockzero,

another interesting thing is that they varied the letter also by when the student would be on campus, either today or next week. I thought this was to make it more random, but having read the paper now they were also trying to find out how temporal distance affects discrimination. interesting stuff.
posted by MisantropicPainforest at 1:01 PM on April 22 [3 favorites]


I think people tend to blithely assume that representation actually does reduce bias, but that's not clearly the case in any field (including politics, where I think that misperception is exploited to an absurd degree). There are deeper logics at work, and those need to be recognized if we're ever going to move our institutions in the direction of real meritocracy.

I actually have a paper looking at exactly this issue in crowdfunding, where it turns out that women do 13% better than men, all other things being equal. Women are generally discriminated against in any sort of fundraising, so we tried to figure out why.

It turns out that the answer is that a small proportion of women (less than 30%) are activists, who actively support other women, especially in fields where women are underrepresented. The majority of women did not offer any particular help to people of similar gender. So, I agree that representation alone doesn't always reduce bias, but activists do tend to uplift others of their minority group.
posted by blahblahblah at 1:02 PM on April 22 [19 favorites]


It's not as common as you might think for professors to really mentor graduate students

That's a very different claim than "mentoring of any kind is uncommon in PhD programs for any student." I know this to be untrue of my department and untrue, generally, of my discipline. It may be true in other disciplines--I don't know--but it is certainly false as a blanket statement about the universal experience of almost all PhD students.
posted by yoink at 1:02 PM on April 22 [5 favorites]


So, I agree that representation alone doesn't always reduce bias, but activists do tend to uplift others of their minority group.

Although, shouldn't a similar effect have been discernible in Milkman's data if this was a generalizable finding? It seems surprising that there isn't even a subset of the diverse faculty who are "activist" in this sense and who would thereby skew the average response rate a little higher for that subgroup as a whole.
posted by yoink at 1:05 PM on April 22


Although, shouldn't a similar effect have been discernible in Milkman's data if this was a generalizable finding? It seems surprising that there isn't even a subset of the diverse faculty who are "activist" in this sense and who would thereby skew the average response rate a little higher for that subgroup as a whole.

I haven't spent a huge amount of time going over the paper, but, since they are looking at field salary, they treat fields such as fine arts (negative discrimination) and humanities (almost zero discrimination) as low salary fields. I might recontextualize these as "activist fields" where there are more politically active academics who are showing no, or even negative, discrimination. It is an interesting question though.
posted by blahblahblah at 1:09 PM on April 22 [3 favorites]


It seems surprising that there isn't even a subset of the diverse faculty who are "activist" in this sense and who would thereby skew the average response rate a little higher for that subgroup as a whole.

They probably did, but a 'little higher' may not be statistically significant.
posted by MisantropicPainforest at 1:09 PM on April 22 [1 favorite]


It's not as common as you might think for professors to really mentor graduate students

That's a very different claim than "mentoring of any kind is uncommon in PhD programs for any student." I know this to be untrue of my department and untrue, generally, of my discipline. It may be true in other disciplines--I don't know--but it is certainly false as a blanket statement about the universal experience of almost all PhD students.


That's true, those two claims are different.

I think the former is almost certainly true. The latter is true to the extent that it's a characterization of a broad trend rather than an ironclad law which attains in every case, and to be fair, it is expressed that way here.

So I think you're right that there's variation, but how much that matters depends on whether you're interested in broad trends or in exceptions, I guess.
posted by clockzero at 1:09 PM on April 22


Perhaps the professors aren't racist, but their spam filters are...
posted by pwnguin at 1:11 PM on April 22 [1 favorite]


I actually have a paper looking at exactly this issue in crowdfunding, where it turns out that women do 13% better than men, all other things being equal. Women are generally discriminated against in any sort of fundraising, so we tried to figure out why.

It turns out that the answer is that a small proportion of women (less than 30%) are activists, who actively support other women, especially in fields where women are underrepresented. The majority of women did not offer any particular help to people of similar gender. So, I agree that representation alone doesn't always reduce bias, but activists do tend to uplift others of their minority group.


That is super interesting! I was thinking about bias in terms of how power-holders evaluate the impacts of their decision-making on disparate groups, but that's clearly not the only kind of bias that matters, as for instance when we're trying to make sense of the patterns that determine how people get into power in the first place.
posted by clockzero at 1:14 PM on April 22


It's not correct to say that random distribution of the potential stimulus error means that it can't affect the outcome. This error essentially acts as another experimental stimulus, which is not being accounted for in the analysis. It's entirely possible that it interacts with the intended stimuli such that response patterns are different regarding legit- and illegit-seeming letters. Maybe if the letter seems on the level, there are no or only weak race/gender effects, but given a seemingly bogus letter, those effects become very strong. Or maybe the reverse! There's no way of knowing without being able to know which letters went to people who could possibly mentor doctoral students and which ones didn't.
posted by aaronetc at 1:46 PM on April 22


I guess all those Business schools with doctoral programs are just imaginary then.

Doctorates from business schools are DBAs, not PhDs.
posted by Jacqueline at 1:48 PM on April 22


Jacqueline : Doctorates from business schools are DBAs, not PhDs.

Not really true. Except for Harvard, most academic business schools grant PhDs, not DBAs.

aaronetc
It's not correct to say that random distribution of the potential stimulus error means that it can't affect the outcome. This error essentially acts as another experimental stimulus, which is not being accounted for in the analysis. It's entirely possible that it interacts with the intended stimuli such that response patterns are different regarding legit- and illegit-seeming letters. Maybe if the letter seems on the level, there are no or only weak race/gender effects, but given a seemingly bogus letter, those effects become very strong. Or maybe the reverse! There's no way of knowing without being able to know which letters went to people who could possibly mentor doctoral students and which ones didn't.

I think this is a good point, though I think it is hard to come up with a strong theoretical reason to believe that a potential interaction is changing the patterns significantly based solely on randomly distributed errors based on legitimacy. However, if I were reviewing the paper, it would be something I would want them to check. They do look for interactions, though their approach (Table 4) doesn't really answer the question you raise.
posted by blahblahblah at 1:55 PM on April 22 [3 favorites]


The reason minority professors did not respond in higher numbers is because the rationale for discrimination need not be irrational (that is to say, driven by some base hate or dislike).

I have held positions of small authority in the academy. Not (yet) a professor, but in positions where my opinion/support can make a big difference in what job/research students might receive.

There are attributes correlated with some race/gender specifically, and people often use these in their heuristic models of how to treat a person. This might fail the test of certain (many) ethical frameworks, but it is not unfounded. For example, there appears to me to be a consistent trend of a subset of Chinese students who (perhaps through no fault of their own in not understanding a new culture) to be relentless and annoying in asking for help/recommendations to get a job in banking. I've had so many shameless, short, and bland emails from Chinese students. I'm NOT SAYING that this makes it morally acceptable to discriminate on their race. I am saying that these attributes can be correlated with information.

There is a latent bias against Chinese in US institutions. I see it and I'm honest enough to admit I even see it sometimes in myself. There is the stereotypical poor-English Chinese student who looks great on paper, and then can't speak well and looks at his shoes when you try to meet with him. I have done my best to fight this stereotype internally, and challenge myself to understand what it means to move from your home and country to pursue a better life in a foreign land (admirable!! and an overlooked virtue among Chinese students).

Still, the point being it is very hard to force ourselves to ignore information correlated with race/gender. That's just a base fact, but it is, frankly, dangerous to talk about. Which is too bad, I think.
posted by jjmoney at 1:56 PM on April 22 [3 favorites]


I'm surprised by their results. If I got their letter (and I do routinely get letters similar to paragraph 1),

Dear Professor [Surname of Professor Inserted Here],

= Ah, mass mailing spam. (I am not a professor.)

I am writing you because I am a prospective doctoral student with considerable interest in your research. My plan is to apply to doctoral programs this coming fall, and I am eager to learn as much as I can about research opportunities in the meantime.

= Standard mass mailing boilerplate. Meh. I don't have time for this, look at the website.

I will be on campus today/[next Monday], and although I know it is short notice, I was wondering if you might have 10 minutes [...]

= Automatic yes, yes, yes. I always have 10 minutes to chat with a prospective student of any sort, even if it is just about how bad the weather sucks here, or how gorgeous summer is. Although it is very odd that if their first priority is to chat with me, they've left it till the day of their visit to ask me - what if I'm out of town?

In any case, if I get to para 2, I have no doubt that I'd say yes, as long as it is physically possible. I guess the test is, did I hit Delete before reading to the end or not. And given this subject line:

Subject Line: Prospective Doctoral Student (On Campus Today/[Next Monday])

I would have read the whole email for sure.

So I'm very surprised that the sender's name (with its implied race/gender) made that much difference to the response rate on an email that said a prospective student was going to be on campus and wanted to chat. What do these non-responders think their job is?

In the physical sciences, if you're getting funding from the NSF, you're already making efforts to recruit underrepresented minorities - a black or Hispanic or female-sounding name should boost response rates, if anything.
posted by RedOrGreen at 2:00 PM on April 22 [1 favorite]


For example, there appears to me to be a consistent trend of a subset of Chinese students who (perhaps through no fault of their own in not understanding a new culture) to be relentless and annoying in asking for help/recommendations to get a job in banking. I've had so many shameless, short, and bland emails from Chinese students.

These weren't ostensibly letters from chinese nationals, they were just letters from people with different sounding names.
posted by MisantropicPainforest at 2:03 PM on April 22 [1 favorite]


Dear Professor [Surname of Professor Inserted Here],

= Ah, mass mailing spam. (I am not a professor.)


But surely a prospective student would rather err on the positive side rather than not? Besides, in latin america, the word Professor has a different (lighter) heft to it than in English, and it's an easy mistake to make.
posted by dhruva at 3:10 PM on April 22 [1 favorite]


Also somewhat anecdotally, when I was searching for PhD programs in the West, I wrote to about 10-15 people (with my Indian name) and got one response. From a prof in Israel.
posted by dhruva at 3:12 PM on April 22 [1 favorite]


Still, the point being it is very hard to force ourselves to ignore information correlated with race/gender.

I would argue that the tragedy of racism and sexism is that it isn't that hard. When people are made aware of the phenomenon of implicit bias and how it can affect their behavior, that awareness can often be enough to produce better outcomes, even though the discovery of such biases, understandably, makes people upset.

But I think far too many of us just want to check the box in our self-produced psychological profiles that says "I am not racist/sexist; I have never had a racist or sexist thought or done anything that is racist/sexist" and call it a day.

The box we should be checking is "You know what? If I'm not careful, the constant bombardment of negative stereotypes I get in popular culture about ethnic minorities and women will affect how I perceive of them and treat them. This is true even if I am an ethnic minority and/or a woman. So, you know, I should take like a second or two to think about that in certain situations."

That's it -- just a second or two. But we usually fail to do even that much. Which I guess means another box we should all be checking is "Yeah, I'm probably at least a little bit sexist/racist."
posted by lord_wolf at 3:30 PM on April 22 [26 favorites]


when I was searching for PhD programs in the West, I wrote to about 10-15 people (with my Indian name) and got one response.

Yeah, but you'd be surprised at how many others wrote similar messages.

It's the "I will be visiting campus and would like to meet you" aspect of this letter that sets it apart, not the fact that someone is "interested in my research". (I mean, I think it's great stuff, but I don't flatter myself that much.)
posted by RedOrGreen at 3:31 PM on April 22 [1 favorite]


I teach at a college with only undergraduates, so I don't get these particular emails. But I do mentor undergrads, some of them women and/or with non-whitebread names, and this just fills me with despair. We certainly had a perception of this bias when I was in grad school, but it's very awkward to call out the people with total power over your future, especially when they are women or people of color. How do I tell my students "You have to email professors to have a shot of admission to grad school but your response rate may seem really low because the gatekeepers suffer from unacknowledged implicit bias."?
posted by hydropsyche at 3:40 PM on April 22 [3 favorites]


My Baha'i mother gave me a biblical name for this reason -- college professors discriminate based on names. Sad to read that things haven't changed much over the course of my existence.
posted by oceanjesse at 3:42 PM on April 22 [1 favorite]


= Ah, mass mailing spam. (I am not a professor.)

We had a long discussion around here (it might've been on the green) about what to call people who teach college students, and in the end the consensus was that there is no consensus, and it is mostly bound by field and local custom. The best practice is to ask, but that isn't really going to work for a first-contact email.
posted by rtha at 3:57 PM on April 22


They only sent the email to tenure-track professors, so there shouldn't have been an issue with wrongly calling the recipients professors.
posted by ArbitraryAndCapricious at 4:01 PM on April 22 [2 favorites]


Pretty bad. The clear take-away is that academia needs more conservative faculty members.
posted by IndigoJones at 4:15 PM on April 22


I noticed in my Master's program (Accounting) that several of the female professors had some very weird ideas about how THEIR husbands and children had never distracted their attention away from the importance of their careers, but that the female graduate students didn't take their future prospects sufficiently seriously and this was why the women tended to have a harder time finding internships/jobs. The idea that even female and minority professors would show a preference towards white males in business schools is not even vaguely surprising.
posted by Sequence at 4:45 PM on April 22 [2 favorites]


And this, dear meFites, is why this queer gender-variant person dropped out of an Ivy League PhD program and into IT. Which even then was no haven of non-discrimination, but at least had a high tolerance for weird-os.
posted by Dreidl at 4:57 PM on April 22 [3 favorites]


How do I tell my students "You have to email professors to have a shot of admission to grad school but your response rate may seem really low because the gatekeepers suffer from unacknowledged implicit bias."?

I hope you do share this kind of research with them, because they need to know what they are facing. That's the easy part; the hard part is to help them find effective work-arounds to the problem -- do they need an introductory email from a current professor, or should they be focusing on making personal contact at conferences rather than following the one-size-fits-all advice of sending an email to an individual professor, maybe? In some departments they might do better to make first contact with a DGS or graduate admissions chair, but that is going to vary too much to generalize.

It's the same when any solid research comes out that shows entrenched bias -- the effective response is neither to ignore it nor to become disheartened. There was a piece being sent around recently about the myth of meritocracy in academia, which this research highlights.

To me, this kind of study only emphasizes the importance of those programs that focus on attracting and retaining diverse students into the pipeline, because they explicitly try to resolve the problems of institutional bias that these studies describe.
posted by Dip Flash at 5:17 PM on April 22 [3 favorites]


Maybe I'm an oddity, but when I get an email like this I pass them on to our grad chair, who actually is the person who can help them get admitted, tell them about the program, provide them with a link to the SGS website, etc, etc. I can do nothing.

I don't think I change the response because of the name at the bottom of the email, although this kind of mass-email is more likely to be written by a foreign student, only because they're more likely to be ignorant of the application process. At least in my field, here, you apply to the program, not to your hoped-for supervisor. A depressing number of these kinds of emails are students who are hoping to get into law or business, which means that they're spam.

I would be depressingly eager to have a student come to talk about writing an MA or PhD thesis...I've only had one supervisee, mostly because I'm an Early Modern prof in a department that's strong in other areas, and when I get a good undergrad student I send them to places that are strong in my area.
posted by jrochest at 5:43 PM on April 22 [1 favorite]


>>For example, there appears to me to be a consistent trend of a subset of Chinese students who (perhaps through no fault of their own in not understanding a new culture) to be relentless and annoying in asking for help/recommendations to get a job in banking. I've had so many shameless, short, and bland emails from Chinese students.

>These weren't ostensibly letters from chinese nationals, they were just letters from people with different sounding names.

That's an interesting point. For the Indian and Chinese sounding names, I wonder if the response rate would have been better had they been given either westernised or Biblical first names. (eg. David Patel, Alice Singh, Michelle Chen, Florence Lin). I suspect the response rate would have been a tad better if you're able to mitigate the heuristic aspect of professors having had had poor experiences with students from totally different cultural upbringing.

Could even throw in middle names and hyphenated names to connote greater cultural assimilation: David Richard Patel, Henry Albert Chan, Jenny Anderson-Wang, Patty Roberts-Shah.
posted by tksh at 6:09 PM on April 22 [1 favorite]


That's an interesting point. For the Indian and Chinese sounding names, I wonder if the response rate would have been better had they been given either westernised or Biblical first names. (eg. David Patel, Alice Singh, Michelle Chen, Florence Lin). I suspect the response rate would have been a tad better if you're able to mitigate the heuristic aspect of professors having had had poor experiences with students from totally different cultural upbringing.

Could even throw in middle names and hyphenated names to connote greater cultural assimilation: David Richard Patel, Henry Albert Chan, Jenny Anderson-Wang, Patty Roberts-Shah.


This is a great idea for subsequent work.
posted by clockzero at 6:29 PM on April 22 [3 favorites]


I noticed in my Master's program (Accounting) that several of the female professors had some very weird ideas about how THEIR husbands and children had never distracted their attention away from the importance of their careers, but that the female graduate students didn't take their future prospects sufficiently seriously and this was why the women tended to have a harder time finding internships/jobs.

Internalizing the values of a sexist system is sometimes part of surviving it. As a woman in science, I know for a fact that I am sexist. It's something I struggle with constantly.
posted by BrashTech at 6:40 PM on April 22 [10 favorites]


Anglicizing first names has somewhat helped in studies of hiring practices, and it's helped Natalie Portman. I'm not saying I haven't felt like changing my name. But you know, try as I might, I can't seem to make myself feel like a Jen.

This why immigrants still push their kids to the professions. It's easier to rock an LSAT or MCAT than it is to be a method actor.
posted by cotton dress sock at 7:21 PM on April 22


Boom, she changed her last name, sorry. I guess she thought Natalie on its own mightn't be so helpful.
posted by cotton dress sock at 7:26 PM on April 22


The only flaw I can really see in the methodology, and it'd be a hard one to fix without screwing up the standardization, is that the form letters do seem a bit generic. What this means is that the researcher may have been unintentionally selecting for professors who are willing to respond to a generic email – perhaps professors who have higher standards are less biased, but selected themselves out of the study because the "person" writing them didn't seem to have done their homework. A lot of professors I know would probably not respond to an email from a prospective grad student if the email didn't make any reference to the professor's own research; their minimum standard for a response includes a requirement that the student have taken the time to read a couple of their papers and check out their faculty webpage.

That's a bit of a stretch though, to be honest. I imagine that the selection bias was there at some level, but perhaps not at a statistically significant level even if we assume that pickier professors are less biased than less-picky professors. Also, we have no particular reason to make that assumption; numerous studies of academic hiring practices have shown a consistent race and gender bias across practically every slice of the academic landscape, and I think that at this point the existence of that bias is pretty much the null hypothesis, unfortunate though that may be. I'd be much more surprised if this study hadn't found such a bias, though the stratification by department (and lucrativeness of field) is an interesting wrinkle.

I really wish that this sort of thing wasn't true. However, I don't see any crippling flaws in the methodology for this paper and it's pretty much in keeping with related research on the subject. A lot of work needs to be done, clearly. At least academics might respond faster than some other groups; if there's any group of people whose worldview can be changed through academic research, it's academics. Departments want to be inclusive, though they often don't know how to do it. Of course, the tenure system also means that the old guard (who are likely to be the most biased) are deeply entrenched and occupy most of the positions of power.

The newer researchers, though – the assistant and associate professors, and the newer full professors – are very aware of these issues and are pretty keen to see something done about them, in my own experience. This kind of research is pretty high profile; even outside of the social sciences, a lot of professors and grad students hear about these studies because they are the ones being studied which is a novel and somewhat unsettling feeling for many of us. Plus, for those of us who are of a more social activist bent, they give us valuable ammunition that we can use in trying to create change in our departments.

Things are changing, as far as I can see. For instance, I can think of at least two programs in my university that specifically try to identify promising women and PoCs among the undergraduates and either target them for graduate work or else give them a leg up by offering them opportunities to work in the labs or get other CV-boosting experience. There are probably more (I'm not really keyed into that stuff, though my PI is) and I don't imagine that they're terribly unusual programs nationwide. Over time, things are improving. Unfortunately, there is a lot of institutional inertia to overcome even if everybody is willing to work toward a more equitable academy. It can't happen overnight, but that doesn't mean we shouldn't keep trying to hurry it along.
posted by Scientist at 7:56 PM on April 22 [3 favorites]


There's a program like that at the university where I work, Scientist, and I think they do really great work. One thing they do is formalize some of the process of finding mentors, which can overcome some of the unconscious biases that this research seems to reveal. (And I think it's incredibly important to note that this is unconscious bias, and I think that few professors want to discriminate against underrepresented groups, although more on that later.) However, it has some limitations. The first one is that it's designed to help students get into PhD programs, and when students opt for professional programs instead that's taken as a failure and puts their funding in jeopardy. A lot of students, particularly those from relatively disadvantaged backgrounds, want a relatively stable career path, and the more they learn about research careers the less attractive those careers sound. We could really use some similar programs for students who want to go to medical school or get a grad degree in public health or eventually get an MBA. There are good summer programs, but I don't think those are as helpful for finding faculty mentors, plus a lot of my students have other commitments during the summer.

The second issue is that our program is only for underrepresented minorities, and Asian-Americans aren't underrepresented. In a lot of disciplines, they're over-represented. But they still seem to be targets of unconscious discrimination, and I would say that there's also some conscious bias against them, particularly against Asian immigrants who aren't hyper-assimilated. (And that's not even getting into issues with attitudes towards Chinese international students, who represent a very large chunk of the student body at my institution.) I haven't looked into this carefully, but my understanding is that there is some evidence that these failures of mentoring hurt Asian-Americans' career prospects. I honestly think it would be worthwhile to think more about issues that affect students of East and South Asian descent, because those issues tend to get a little lost in (also vitally important) discussions about underrepresented minorities.
posted by ArbitraryAndCapricious at 5:43 AM on April 23 [2 favorites]


You could maybe get close to these results with a spam filter that a) was triggered only when an email contained no mention of a specific research area, b) incremented the score when $name had over the past ten years triggered (a), c) assigned a multiplying factor based on [(average salary five years after graduation)/(average hours worked per week * standardized test scores)]. You could probably get closer if it d) assigned a multiplying factor based on the inverse of the proportion of $name in the applicant pool over the last five years, and e) assigned a multiplying factor of lesser magnitude than (d) based on the inverse of the incidence of $name in the client population - not practitioner - of the field.

There is such thing as prejudice, but the chart on p56 of the paper looked an awful lot like the likelihood of applicant cluelessness mitigated by desirability of ethnicity/gender in terms of both diversity and population served. Various biasing factors before application of the filter could (ie, would) cause these results, but that's separate from bias in the filter itself.

The cluelessness of the email immediately suggested a problem. It's human - animal - nature to compare experience when there's a suggestion of a problem. This would be true of a white male assistant professor just as much as it would be true of a black female assistant professor as they're identical in that regard. Also, while a black female professor may well want there to be more black female professors, she probably gets more email from clueless black women - she might just be sick of it.
posted by c10h12n2 at 6:56 AM on April 23


Are you saying that the results this study produce could have been caused by a spam filter that is racially/ethnically biased?

There is such thing as prejudice, but the chart on p56 of the paper looked an awful lot like the likelihood of applicant cluelessness mitigated by desirability of ethnicity/gender in terms of both diversity and population served.

I'm having trouble parsing this. Why would black people be considered more clueless than white people if not for discrimation/bias?
posted by MisantropicPainforest at 7:02 AM on April 23


The cluelessness of the email immediately suggested a problem. It's human - animal - nature to compare experience when there's a suggestion of a problem.

Can you explain why the clueless people with the non-ethnic-type names got more responses than those with ethnic-type names? Because if the problem really boiled down to some combination of spam filter* + ignore clueless people, then we should not see such an enormous disparity. But we do.

* seriously? spam filter?
posted by rtha at 7:07 AM on April 23 [5 favorites]


It's okay. The Supreme Court says racism is best dealt with on the principle of "out of sight, out of mind," so it's all good now.
posted by saulgoodman at 8:21 AM on April 23


MisantropicPainforest, how would the spamfilter know if someone was black, or what it was to be black?

rtha, anecdotally/experientially, yes. While there's little practical difference between someone at MIT and someone at Tsinghua or IIT-Bombay - regardless of skin color or gender - there's very often a significant difference between someone at MIT and someone at Hanyang Mining University or Guru Mukhund College. Students and researchers at the latter (imaginary) schools are far less likely to be academically well-trained or familiar with the conventions or products of current research. And there's a lot of them: China and India have large populations, so they generate a lot of clueless email. If I'm working 100+ hours per week, trying to triage my incoming mail, and I'm already sick of this steady large volume of clueless mail from Asian-sounding people - yeah, I might be a little more likely to delete them.

The study, apart from creating a problem condition, also mixed test mail with the recipients' other email. To see why this might be relevant, imagine getting a bunch of various email from Cousin Han, Miss Smithers, Dr Singh, and Prince Mgumbo - all mentioning the urgent need to transfer funds. The study was also hardly neutral: Kevin is not somehow less black a name than others, they had to go pretty far into racial stereotypes. Even the language of your question refers to "ethnic-type-names" - I'm pretty sure crackerboy is its own special ethnicity rather than some absence of ethnicity. Those guys eat peanut-butter and jelly sandwiches and milk, right?

The study gave some raw figures, and they're sort of interesting. But it's not even clear how people can reliably tell the difference between Chinese men and women just from the name, because the names are not often gender-specific (maybe tonally, but not in ASCII). Yet there are some large variations there. For any hope of insight, we'd need with a larger study with more variations of input, such as more clueful emails, emails with errors, students named James with Spelman email addresses, students names Antwoine with Harvard addresses, etc, to cross-check biasing factors with some specificity. And some way of accounting for previous and existing states, since they'd necessarily load the results. Otherwise we get empty defensivenesses and handwringings, which shouldn't be what research is about even when discussed at internet water coolers.
posted by c10h12n2 at 9:51 AM on April 23


how would the spamfilter know if someone was black, or what it was to be black?

It wouldn't. Only a human would, hence the conclusion that there is racial bias.

Are you saying the opposite?
posted by MisantropicPainforest at 9:53 AM on April 23


If the filter kept a memory of the names associated with emails flagged as possibly clueless, and over time some names were more likely to be flagged as clueless - as a raw number but ideally in proportion (eg, 93% of name A gets flagged, 37% of name B gets flagged, ...) it could reasonably assign higher spam scores without ever knowing anything about the names, and obviously without making unfounded assumptions based on names. If the world is so wacky that pools of names are small and definitively indicate skin color, income, and educational level, then the filter would work nearly perfectly.

Racism is indefensible because it's not reasonably possible to draw conclusions about an individual's character, ability, or preferences based on skin color, hair type, or geographic origin (let alone ancestral geographic origin). This does not mean that people named Sun, Sunil, or Suneequa who are too clueless for very selective graduate programs can't grossly outnumber people with those same names who are extremely capable.

I'm not too worried about people making snap judgments based on direct indication and bulk experience, especially relative to an extremely high standard. I'd be worried if a Lakeesha wrote brief, intelligent, and engaging emails to 20 professors and was ignored any significant portion of the time. In reality - and this is unfortunate, not even because of connnoted skin color but because of connoted socioeconomic level - she'd cause a riot of interest.
posted by c10h12n2 at 11:13 AM on April 23



If the filter kept a memory of the names associated with emails flagged as possibly clueless, and over time some names were more likely to be flagged as clueless - as a raw number but ideally in proportion (eg, 93% of name A gets flagged, 37% of name B gets flagged, ...) it could reasonably assign higher spam scores without ever knowing anything about the names, and obviously without making unfounded assumptions based on names. If the world is so wacky that pools of names are small and definitively indicate skin color, income, and educational level, then the filter would work nearly perfectly.


This is, to put it lightly, exceedingly unlikely and not a valid criticism against the study unless you have some evidence that this type of thing is not uncommon.

I'd be worried if a Lakeesha wrote brief, intelligent, and engaging emails to 20 professors and was ignored any significant portion of the time.

Thats....exactly what happened here. Jane was ignored a lot less.
posted by MisantropicPainforest at 11:42 AM on April 23


I'd be worried if a Lakeesha wrote brief, intelligent, and engaging emails to 20 professors and was ignored any significant portion of the time. In reality - and this is unfortunate, not even because of connnoted skin color but because of connoted socioeconomic level - she'd cause a riot of interest

To be very direct: the research in this study explicitly contradicts this. Contrary to statements like this about how being black and/or female is a plus ("a riot of interest"), this shows the opposite. Given identical (and totally appropriate and normal) emails, my name will garner responses and Lakeesha's will not, and this effect varies considerably by discipline -- it's not something that is simply part of academia, but is rather part of how specific fields operate.

The spam filter stuff is too inane to deserve a response.
posted by Dip Flash at 12:26 PM on April 23 [2 favorites]


If it wasn't clear: the spam filter suggestion was a thought experiment. It shows both a plausible way by which a non-thinking/associating mechanism with zero initial knowledge could come to a similar result and a simple model of how a decision process in a human might work upon encountering the email. The wacky-world addendum is indeed unlikely; it's also a remark about how {all black|yellow|brown people} aren't being acted against.

I'm sure Jane was ignored less. "Jane" doesn't say anything except probably female and from an English speaking part of the world. There are black people who satisfy that, also white, also yellow, and also brown. For directness: Jane doesn't mean white, or not-black, or anything at all. Lakeesha means poor, and in the US, poor means dumb.

Something very similar is true of the large number of Chinese and Indian students who want to escape but are at bad schools. It's not only that they haven't been trained well for research - often they don't understand what research is. They're fundamentally not part of that world.

Results varying by discipline does not necessarily mean that people in some disciplines are more or less racist than others. Some disciplines may be more of interest to some groups of people whether or not they know what's actually involved - the generic email rather implies the writer doesn't know what's involved.

Professors should care about the capability of their prospective students, and thus care less about a prospective applicant who raises more clueless-flags. A selection process is about selection. The study doesn't contradict anything, or show anything but response to common time-wasting emails with some other factors often/usually associated with wastes of time. For directness: all of these emails were - in actuality, as none came from an actual prospective student - wastes of time. Some raised more time-wastey flags than others and were likelier to be ignored. None of these circumstances was part of the study, making sure conclusions of endemic racism impossible.
posted by c10h12n2 at 1:48 PM on April 23


If you're arguing that a generically white sounding name like Jane doesn't carry any negative connotations, but a generically black sounding name like Lakeesha "means poor, and in the US, poor means dumb", then you are more right than you know.

For directness: all of these emails were - in actuality, as none came from an actual prospective student - wastes of time. Some raised more time-wastey flags than others and were likelier to be ignored. None of these circumstances was part of the study, making sure conclusions of endemic racism impossible.

Oh what a crock of shit. The things you are proposing, like your asanine thought experiment, would require any attempt at social science research to include every possible confounding factor or any possible alternative explanation before any inference is made. Which does nothing more than dismiss a whole body and method of learning about the world and betray your profound misunderstanding of how social science research is conducted. In short, if you want to lob methodological criticisms of this study, then those criticisms need to be built on the assumptions of social science.
posted by MisantropicPainforest at 1:55 PM on April 23


Dip Flash: "The spam filter stuff is too inane to deserve a response."

If your research sends unsolicited email, a confounding factor analysis of the prevalent use of systems to weed unsolicited mail out of inboxes seems reasonable. Maybe it's addressed in the pile of references? 419 scam emails often use race-apparent names, so it's not impossible that this is a thing that spam filters pick up on.

As always, the appropriate response to a publication like this is more research =)
posted by pwnguin at 2:12 PM on April 23


Some raised more time-wastey flags than others and were likelier to be ignored.

No, those flags were actually "raised" by the recipients' implicit bias: their bias that a female name would be a waste of time to respond to more often than a male name; their bias that a "black" female name would be a waste of time to respond to more often than a non-"black" female name, and so on. The emails themselves were identical, and yet there were statistically outrageously significant disparities, and those disparities were because of the assumptions the recipients made about who the senders were, and how serious they must be, based only on how ethnic the name seemed.

Why you are so insistent that there *must* be other explanations besides implicit race and gender bias (to the extent that you had to come up with an absurd thought experiment involving a spam filter!) is really, incredibly weird to me. Are you somehow unfamiliar with the history of (in this case) the United States? The concept that academia (like other industries and fields) suffers from racism and sexism is not really all that new.
posted by rtha at 2:21 PM on April 23 [5 favorites]


 it's not impossible that this is a thing that spam filters pick up on.

Yes, it is. That's not how Spam filters work in real life, okay?
posted by saulgoodman at 8:21 PM on April 23


jjmoney, I'm with you on it being too dangerous to talk about. Everyone's a little bit racist, most of us don't want to be. But when a lot of folks with the same characteristics (whatever they may be, perhaps they have purple skin, one eye, one horn, and wings) are doing the same kind of annoying behavior (say, people-eating) at your work and you can't avoid dealing with it it, repeatedly, your brain is going to start noticing and categorizing these things. And when it seems like most of the time whenever you run into someone in a nice shade of lavender with one eye, one horn and wings, they're invading your personal space, drooling on your head and breathing down your neck, and their stomach is making growling noises.... after awhile you do kind of learn that a lot of the one-eyed, one-horned, flying purple guys tend to want to eat people. And you will kinda want to avoid all of them because it feels inevitable that most of the time, they will want to take a chunk out of your arm. And you won't exactly feel joy when you see another one come in because you don't know if the purple guy's already had lunch or not.

Of course, this is how bad stereotypes start in the first place. I don't really know how you can totally avoid thinking like this. I'm trying to NOT do that, but it can feel difficult at times.

And in this case, well, it's e-mail and it's easy to blow off the e-mail of a total stranger who's not even around and you don't have to care about giving them a chance or not, or dealing with consequences of blowing them off.
posted by jenfullmoon at 10:38 PM on April 23


Thanks for posting this. It looks like a thoughtful study, and the results are pretty damning. (And the lame criticisms and attempts to derail in comments are, as always, astonishing.)

Leaving aside the really important result of the study - the undeniably significant and large race and gender based differences in response rates across intuitions - because it's already been well discussed above, I'm struck by a couple random things.

First, the average response rate seems surprisingly high. 65% is pretty close to the response rate I get from the average professor who knows my name and interacts with me several times a year. (To be fair, I'm not sure my own response rate to personally addressed email is any better.) That strikes me as a *very* good thing.

Second, the distinction between public and private universities is unexpected. (By which I mean not that it's wrong, but that my intuition is clearly skewed.) Among every non-emeritus faculty member I know, the difference between a professor at a public and a private PhD granting university amounts to the question, "was the job offer in a city in which I'm willing to live from a private or public institution." In my own experience at a few of both kinds of institutions, the faculty seem indistinguishable, while there tends to be slightly more resources and discussion of diversity at private universities. (Possibly because their demographic biases are more obvious.) Given the care the authors gave to correcting for school ranking, I'm amazed the discrepancy remains.

Third, I'm curious what return email addresses and domain names were used by the students. (Seems weird to leave it out of the very thorough paper, but I suppose not everything can be included.) I wonder if indian-name@school-you've-heard-of.edu would show the same difference. There really is an astonishing amount of spam directed at US academics from people who seem to be in China and India, and their first two sentences look a lot like these messages. On the other hand, even that hand-waveyist apologism can't explain several of the largest discrepancies. (Fig 1b is pretty amazing.) And, of course, for an individual student of Indian or Chinese descent, writing this off as unread assumed-spam doesn't address the injustice or long term consequences.

Forth, does it really make sense to include "Legal professions," "theology," "social work," "fire protection" and "physical education" in the same category? I know these authors didn't make up the broad categories. . . but, compared to the minute distinctions between engineering and the physical sciences, it seems like the might as well have called that category "other." One might imagine social work, law, and fire protection inhabit different academic cultures. I'm curious to know which ones are responsible for the results.
posted by eotvos at 11:46 PM on April 23


I have a brown-seeming gender-indeterminate name (my surname is more common as a woman's given name), and upon visual inspection, I am indeed a brown-skinned male. I went to grad school at an Ivy, did grad admissions for my department at an Ivy for five years, and have for the past six years been associated with a fairly prestigious academic journal - the last is relevant because I get a lot of clueless email there as well.

I'm not suggesting there is no racism, sexism, or classism. I deal with these on a daily basis. It's an old man saying tattoos are stupid and assuming both that I would agree and that I didn't have any. It's a checkout person at the supermarket assuming my parsley was cilantro (when I corrected I was informed that my people ate cilantro, not parsley). It's people assuming I'm a doctor in certain circumstances and a take-out delivery guy in others. It's people telling me I have an English or Indian accent pretty much because I have brown skin and my accent+idiom is close to Eastern Standard rather than regionalistic. It is in fact so pervasive that when a white South African man once told me I really didn't seem like an American, I thought Oh, here we go, and asked, "Why is that?" The man shifted uncomfortably and said, "Well, you seem... educated." And maybe there was even some racism in that, but I felt ashamed for what I'd assumed.

People in the US are still often taken aback when they ask what nationality I am and I answer - on purpose - "I'm American." They want to hear that I'm Indian, or Pakistani, or Iranian, because nationality for them is not a matter of citizenship but of skin color, ancient geographic origin, and a presumed culture. Even though this doesn't make sense - South Asia is hardly a single-hued monoculture. But hey, in my own country it's more often seen as reasonable to call Denzel Washington African American, but not Charlize Theron. If she spoke with her American accent to someone in the US who didn't know who she was, that person probably wouldn't ask her nationality, either.

Once I say I'm American, the next question is often, “Really? When did you come here?” Because I can't really be from here. So I get to say, “I was born here – they've been letting brown people breed here for quite some time now.”

With all that said, I've experienced remarkably little racism within academia, at least within the fields I've been in. Maybe my threshold is high due to daily life, or my skin too thick, but it's been pretty negligible. I haven't witnessed much racism either. And the biggest difference I could probably describe as far as who's racist is whether a person is young enough (and to some extent privileged enough) to have watched Sesame Street as a kid. It's true: if as a little kid it seems totally normal for some genderless thing with blue fur to be a person, with feelings and individuality, just like you, you're probably less likely to be fazed as an adult by different skin color – less likely to assume that skin color or anatomy are fundamental bases of difference.

I've also had remarkably little trouble getting responses from professors, although I've certainly heard this happens often to many people. Non-colloquial email isn't that difficult, but there are some good rules: be specific, be brief, and be dead clear in what you're about and what you want the person to do. Some people do this naturally, others have to be told. I've certainly told students how to write email.

And this is why I have a problem, not so much with the study - it's just some data, with pretty clear parameters - but with hard conclusions about bigotry from that data without any allowance for alternate mechanisms.

Clearly, there exists bigotry in the world, including academia. And clearly, responses in the study varied with certain indicators. What's not clear is the motivation.

Academics (it should be unnecessary, but I'll include the disclaimer that my experience is of course limited) have tremendous demands on their time. There are jokes like, "Sure, the schedule is very free, you're free to work any 100 hours of the week that you like." This starts in grad school and doesn't let up until long after you're a full professor. And apart from work, there's family, and maybe something else you do sometimes just for fun (except for Caltech people, they only do work and family – those people are freakshows).

Academics are also public figures. They get a lot of email from people with whom they have no relationship. And since they are in a sense gatekeepers to better lives, they get lots of email from people who want better lives. And who wants better lives? Darned near everybody. So that's a lot of email wanting stuff from you. And a lot of those people are in bad places. What distinguishes the people in bad places is not location, race, sex, or whatever but lack of access to information, not participating in what the relatively small portion of the global population that actually does things is doing.

That is the problem, not so much bigotry. Now, bigotry can create that problem. If you're little girl in Afghanistan, maybe people will stop you from going to school, even if there is a school. If you're a little girl in the US, maybe people make you play with dolls or assume you're bad at math. You're totally screwed if you're a little girl and poor, because poor people are screwed. Heck, maybe there's a school and you're not poor, but your local educational bureaucrat saw some research that suggested female brains are less adept with numbers and is too ignorant to understand that even if that's true (I haven't noticed), mathematics is not about numbers but relationships - and most sexists would agree that women are great at relationships.

So it's not the little girl's fault for having not done something to not be dumb. It's not that girls or poor black people are by nature dumb. But they are going to be disproportionately dumb. In order to avoid being dumb, you're going to have to have opportunity and do some hard work.

This is not to say that poor people don't work hard. Being poor is incredibly hard work, and incredibly expensive. But it doesn't make a difference: whether you spend your day working three US minimum-wage jobs, collecting welfare to buy chicken nuggets, working in fields and then walking 10k for clean water, losing a week of time and spending a month's pay for travel to get medicine for your sister - medicine that might be available to someone more privileged by going to a store, except they won't have that condition so they won't need the meds - you are not doing the specific work that will allow you to participate.

And we haven't even talked about people who are just dumb. I don't know why, but that's a lot of people. It's sad and ugly to say, but: lots of people are just dumb. Lots of people are dumb even though they have money and access. Lots of Indian schools are full of dumb people with money, not least because the elite institutions are essentially free. But these students are undergraduates with money, and they're dumb. The Indian middle class - people with cars, houses, jobs - is on the order of 400M people. Most of them, like most of any other large population, are dumb.

But back to the work. You're an academic. Your workload is massive. The only way to deal with that massive workload, that onslaught of demands for your time, is not to do some of it. Randy Pausch has video lectures about this. Just as working three jobs so you can maybe pay rent is not going to leave you time to do things more important to you, offering your time to every person who wants something from you, including clueless time-sinks, is going to adversely impact the time you have to do whatever you think might actually make the world a better place, let alone do what fascinates you.

I don't ignore emails. Ever. This is personal; I don't even ignore texts, and even as a middle aged guy who understands not everybody is just like me, it hurts my feelings a tiny bit when a friend doesn't respond to one of my texts. But I know people who do ignore emails, and I don't think less of them for it.

When I get email similar to that of the study, I don't ignore it, but I don't offer my time either. I generally point them to the website and say to get back to me (usually they don't). If there's some piece of information I can offer to ameliorate dire cluelessness (eg, No, you do not apply to a graduate program in network engineering so you can learn to use your computer better, or even to fix other people's computers) - I do. But I'm firm about not offering time. I don't make a determination about my response from cues offered by the sender's name or even institution. If I think you are clueless I am going to tell you to go away.

So, Prof Lakeesha Williams gets this email. She knows before reading it that she doesn't know the sender. She reads it, and the person clearly doesn't know what they want, and this is probably a mass email sent to lots of people. That's the point at which I'd respond negatively and many people would delete, because there are already two reasons to get rid of it, the second reason being huge. But she doesn't. She looks at the sender's name. And she sees that the name belongs to a group of people, or is similar to the names of a group of people, that sends her lots of clueless email. So now there's another mark against it. Now, since that email has multiple marks against it and doesn't seem to be written by or intended for an individual, she deletes it.

Is the third mark racist, sexist, or classist? Yes, but the reaction comes from legitimate experience - a lot it - rather than ignorance, so yeah, it seems different to me. Are the circumstances that create such a condition racist, sexist, or classist? Yes, absolutely, in addition to the circumstance that most people are just dumb, regardless.

And I just don't buy - I don't even see how anybody does - that Lakeesha indicates just black. It indicates black, poor, non-participatory, disenfranchised. I have no idea what color Jane is. Call me crazy, but just as I don't believe in things for which I have no proof, I don't make unnecessary unprompted assumptions.

It's not surprising at all that nonmale or nonwhite people would behave the same as white males - the experience with the sending population is the same. Unless you consider being poor and black more of your identity than that you're an academic, you're not going to behave otherwise just out of empathy/pity It's not surprising that private schools discriminate more than public - the majority of the most selective institutions in the US are private (unlike, say, India, where if you pay significant tuition through undergrad it's specifically because you're untalented).

So yeah, there was some data here. There some evidence of some kinds of bigotry, and the results of other kinds of bigotry. But there's no data to isolate conclusions that professors think anything in particular about white males.

Establishing that something happens is a necessary first step to finding why it happens. The study did the first part, if imperfectly. I can see reasons to keep things simple and standardized, but for social studies more and more varied data seems a really good idea. I hope and assume that Prof Milkman et al will go further.

But seriously, there's nothing here that supports hard conclusions about generic nonwhite nonmales are weird and sucky bigotry is entrenched in academia, and I'm really not sure what place shrieking and unsubstantiated name-calling have in a civilized world. Go teach, go find out useful information and make it available as broadly as possible. Go donate money, go hold politicians and bureaucrats to task. Or at least speak with civility and rationality, with genuine respect for ideas, individuals – for difference and disagreement. Have a real respect for real diversity. If everyone who shouted “asinine immoral slutty crock murderer” at a woman having an abortion would instead go adopt some babies, they'd find themselves with much less of that problem they hate so much.

(And if anyone read all this: thank you, I appreciate it. But I have to go earn my money now.)
posted by c10h12n2 at 8:56 AM on April 24 [1 favorite]


Partners in study gave legal memo a lower rating when told author wasn’t white

"The reviewers gave the memo supposedly written by a white man a rating of 4.1 out of 5, while they gave the memo supposedly written by a black man a rating of 3.2 out of 5. The white Thomas Meyer was praised for his potential and good analytical skills, while the black Thomas Meyer was criticized as average at best and needing a lot of work."
posted by rtha at 8:15 AM on April 25 [4 favorites]


« Older Controversial education tech company InBloom has s...  |  So you can get 4096 in 2048, a... Newer »


This thread has been archived and is closed to new comments