Racial Bias, Even When We Have Good Intentions | Hacker News Reader

Racial Bias, Even When We Have Good Intentions

nytimes.com

36 pointsjeremynixon52 comments

I do a lot of hiring. It drives me crazy that applicant names (which signal gender and ethnicity) are literally the first thing I see in our candidate screening system.

So I wrote a GreaseMonkey script to replace candidate names with dictionary words. It isn't perfect -- names show up on resumes, and every other stage if the hiring process. I wish name-substitution were a standard feature of these systems!

Awesome. In the Netherlands there have been quite a few studies where they'll send exactly the same resume and application to different companies but change the name, and they'll see Mohammed invited a lot less often than say John, or whatever, despite again exactly the same application.

On HackerNews, iirc, there was an article a while ago about someone writing a script to automatically respond to housing ads, as he had a lot of trouble finding an apartment somewhere in Germany, Munich I believe. Ah I found it. So what he'd do is send out applications where he could just change out the name and see which ones get the best response. Here's the quote I wanted to highlight:

> The conclusion from this little study, is that a girl with an italian name, gets an 90% answer rate, a guy with an arab name and is younger than 25 gets, 1% answer rate. The master of all, is the young munich guy who is around 25

As someone with an Arab name, this can get a little bit depressing :p And while I try to stay away from conjuring up victimization plots against me (I'd absolutely hate to go through life feeling as if my ethnicity is holding me back and that any trouble finding an apartment or a job must be at least in part related to this), I can't help but say that it feels familiar to personal experiences, as much as I hate to say it about myself.

So what you're doing is absolutely awesome :) Thank you.

Great thinking.

Perhaps its me but I know for me, a woman in software engineering, I constantly worry about what my name signals to potential employers. I wish my gender could be erased while being evaluated as I don't wish for it to be unintentionally or intentionally a factor in evaluating my job qualifications. Or the quality of any work I might publish online. No matter if it works out in my favor or not. I am kinda disturbed by the trend of LinkedIn profiles because I believe you are (apparently? I think...) encouraged to upload a picture of yourself. We did away with including a picture with your resume for a reason, now it seems to be coming back?

Seems as if employer tend to hire "potental friends" over potentally over more qualified applicants [http://www.forbes.com/sites/susanadams/2012/12/03/employers-...] which I think is part of the gender/race/ethnicity divide. People want to surround themselves with others like them.

As an aside, my experience seems to be that many employers of middle class jobs seem to expect candidates to come from a middle class upbringing and perhaps don't understand the opportunities that weren't available to some candidate due to growing up poor.

Just my opinions and thoughts on the matter.

It's not just race. Ugly people are incredibly discriminated against. Attractive politicians receive two and a half times as many votes as unattractive ones. Attractive people are far more likely to get hired and get paid 14% more. Unattractive people are twice as likely to be convicted for the same crimes in court, and have to pay damages twice as high on average.

http://lesswrong.com/lw/lj/the_halo_effect/

Good point. Whenever I read articles like this, I often wonder if we are over-optimizing for racial discrimination and under-optimizing for other other forms. Interestingly, the people who cry out the loudest for racial discrimination often could care less about the type of discrimination you just pointed out (from my experience).

The two concepts are IMO not so easily separable - see the widespread finding of blacks in the US to be "unattractive". Attractiveness is, despite the loud griping of evo-psych proponents, largely cultural. Our perceptions of race, of gender roles, of etc etc, contribute greatly to attraction.

Discrimination in all its forms are often hard to disentangle.

Not always. Short people, for example, are biased against in job interviews and most people don't care.

The short people thing IMO supports my argument. There's a lot of evolutionary psychology drivel about how we're genetically predisposed to disliking short people due to lack of fitness, etc etc.

But a lot of the bias against short people is cultural in origin and hasn't always been the case. See the historically shifting attitudes towards fat people (from envy to disdain, as societies have industrialized).

We're taught by our upbringing and our surroundings to be biased against certain features in certain contexts - whether that's skin color, height, gender, orientation, or anything else. Some of this is colored by racism, sexism, or any number of other -isms we suffer from.

The shortness-bias is fairly culturally-universal, actually. And your point does little to refute that no one makes a big deal about this particular bias. (Not that I'm saying they shouldn't).

There are pleanty of groups fighting to end other forms of discrimination (incl. unconsciencous), such as ending gender, or sexuality based discrimination.

"Intersectionality" is a term for what happens when you're at the short end of the stick of several forms.

And height. People are more willing to vote for a tall person. Tall people get paid more. etc...

Agreed. I always find it baffling that people are deeply concerned about one type of bias, but not the other. Explanations, I find, are usually rationalized.

And gender too. Studies with a male name/female name, show men get offered more money & jobs.

If you want to test your own _implicit_ biases (sexism, agism, homophobia, racism), there is a nice battery of tests https://implicit.harvard.edu/implicit/, which base on reaction times.

There are two interesting things about it:

- even if you are liberal, implicit biases may persist (because of various biological and cultural factors),

- its about you (its easy to excuse oneself that implicit x-isms are only about others, less rational or something).

Being caucasian I have a hard time finding objective examples of etnic discrimination which I could give to people who thinks that it's a solved issue.

This article provided me few of them, thanks.

Blacks make up about 13% of the population but they make up 40% of male inmates in jail or prison. With such a huge disparity one can start thinking cultural issues (whatever they are) may be at play. Unless you subscribe to 19th Century "scientific racism."

That is scientific evidence for the existence of racism people.

To be real scientists, and rational, we must be willing to admit what the evidence says, even if it makes us uncomfortabl. Creationists think evolution means their religion is wrong, so they fight the evidence. People don't like to think they act racist, so fight the evidence.

Interesting article.

For the "black names" thing, though, I would be curious to know if there were any studies also comparing "stereotypical white-but-not-anglophone" names, for example Greek, East-European, or South-European names.

And also, for example, non-traditional names that have no racial implications, like "Rainbow" or similar.

I don't know about other ethnic names, but there have been studies with male name versus female name, which show men get offered more jobs and more money.

Yeap. And regardless of the gender of the person evaluating the candidate.

"non-traditional names that have no racial implications, like "Rainbow" or similar"

Aren't the races behind these names usually clear? In Africa such names are along the lines of Pretty, or Precious, and in the west they're Apple, or Peaches.

I read about a study once where college students in the US were asked to provide their ethnicity on the front of their test papers. This had a notable negative effect on the scores of African-Americans due to unconscious bias, I believe it was presented as an example of priming

Yes, the effect is called "stereotype threat"[1] in the psychology literature. A professor of psychology who participates in the journal club I participate in is deeply skeptical of how large the effect size of stereotype threat is in the real world. On my part, I like the interventions that have been suggested by the main stereotype threat researchers, to prime students with self-expressed affirmations of their core values to help them boost their own performance.

[1] http://mrnas.pbworks.com/f/claude%20steele%20stereotype%20th...

https://teachersh.scis-his.net/tmrak/wp-content/uploads/site...

This article raises a question: how does one train their "fast" thinking? If someone reads this article and recalls all of the times that at least internally they passed a wrong judgement based on another's race, what can they do to prevent immediately thinking such things?

I don't much like this intuitive answer: if you're simply aware of your tendency to erroneously judge, you can consciously act against it. Is there any way to train the subconscious? The lack of time to consciously address racism in some situations could lead to mortality.

Awesome article; science is great.

One way is to hide the information from yourself. I've heard of orchestras hiring by getting the applicant to play their instrument behind a screen. You can't see them, can't see their race/ethnicity, or gender. The only information you can use is the only information you should be using: how they play.

I wasn't satisfied with this answer for a whole week, but now I'm trying to figure out ways to integrate that hiring technique to hiring in other fields (programming, especially). Thanks!

A must read paper related to this topic is "Cultural biases in economic Exchange? [2003,2008]" [1]

[1] http://www.kellogg.northwestern.edu/faculty/sapienza/htm/cul...

This is quite interesting. In my relations, I have subconsciously witnessed racial bias at play without people seemingly aware. I guess a lot of associations we make are based off things we see and not consciously decide on in a forceful way that reinforces those beliefs to us as something we stand by.

Much of it is down to nature. Minorities usually don't get preferential treatment when dealing with people who just don't relate naturally with someone they don't share natural similarities with, first-off physically.

I think it's really great that we have strong movements against racism and other forms of stigma stereotypically-related but doing away with it totally remains a job of nature...and people just being naturally different. Am I painting a gloomy-picture, I hope not. I haven't come to a good conclusion on what needs to be done to shake the associations we make in our subconscious minds that affect others unfairly.

Articles like this always leave out Asians because it does not fit the narrative and forces you to think about culture as a main factor.

Bias goes both ways and biases effecting social status for Asians are typically positive. For example, if hiring for a Mathematician from three resumes, 1 White, 1 Black, and 1 Asian and being unaware of actual skill, the Asian would likely be considered first. In this case bias helped the Asian get a job that will raise their social status.

There are negative biases towards Asians as well, but they don't have much effect on social status, such as being bad drivers.

There are also positive biases towards Blacks, but the those positive biases don't have much effect on social status. An example would be if basketball team had to be constructed from a pool of available players, again unaware of actual skill, the Black players would likely be picked first.

I think the real point is that bias does have an effect on social status. Depending on what those biases are it could be a positive or negative effect.

I'm not sure that "positive" is the right word here. Biases seem to be frequently double-edged at a minimum.

The negative consequence is more obvious in your basketball example. Picking blacks for a basketball team simply because they are black is not positive. It is part of a system of thinking, which says that "blacks are good at these things". The corollary is that "blacks are not good at these other things". So, even what you might consider a "positive" bias has a reinforcing effect on the way we classify people and form other biases that are objectively and obviously negative in socially significant ways.

Likewise, classifying Asians as good mathemeticians is part of a system of thinking that says they are less good at other things. All of these preconceived notions have some value attached to them which, collectively, have a direct bearing on social status.

In short, biases tend to have a profound negative side even, or especially, when we are unaware of as much and/or believe they are OK.

You are correct positive may not be the best term, but I think we are in agreement otherwise

>positive may not be the best term, but I think we are in agreement otherwise

I didn't think we were in agreement, which was why I commented.

My point was that bias generally has an overall negative effect, even when some biases appear to be positive (as you asserted). If you agree, but simply think that "positive" is not the best term, then I don't understand your original comment:

>Depending on what those biases are it could be a positive or negative effect.

That seemed to be your point, in summary, and I don't know what term you'd substitute for "positive" that would be qualitatively different, while not completely dismantling your premise.

In context of the comment I was initially replying, regarding how Asians are considered a counter example, I think positive is the correct term.

My initial post was strictly to point out the original posters bias towards Asians.

So my point, in context of the original post, is that bias towards Asians give then a higher social standing. Which I do consider a positive when compared to other groups.

Your initial post seemed to be more in general, and I agree that all bias has some negativity to it.

Understood. Thanks for clarifying.

> I'm not sure that "positive" is the right word here. Biases seem to be frequently double-edged at a minimum.

I've seen the term "benovolent sexism" used for gender roles that are, in theory, complimentary, but in reality are bad for society. e.g. "women are good with children" is in theory a compliment. But in reality it means men who try to work in child care are viewed suspiciously, and women are expected to give up their careers to look after children (because they are better at it!). Benovolent sexism (and hostile sexism (like "women are too emotional to wield power")) are bad.

Likewise, "benovolent racism" is bad.

The narrative is "People are still judged based on their race/ethnicity". If Asians are hired more, then that's further evidence.

And "culture" is one component of ethnicity.

I'm Asian - are you? Personally, I am pretty tired of my race being weaponized against other racial minorities in the USA, and doubly sick of it being done by people who aren't even aware of the Asian-American experience.

The fact that Asians have "done well" in no way implies the lack of racism - and we face plenty of it ourselves, though certainly not to the same tune as blacks and hispanics in this country.

While we're not stereotyped as criminals or parasites, we are stereotyped as weak and single-minded. Asian-American representation in management (or even just technical leadership) is disproportionate to their representation in the rank and file as a result. We also suffer from a whole boatload of "positive" stereotypes that ultimately prove limiting as second- and third-generation Asian-Americans try to exceed the stereotypically associated roles for them.

Try talking to an Asian artist or musician and see how the "logical and good at math" stereotypes works for them.

Why is the belief that the human species has races just like cats, dogs, cows, horses, etc. so widespread in the US culture?

Even anti-racism rhetoric takes this absurd proposition as a fact. Take this article for example: all the cases where the author talks about ethnicity he uses "race" instead.

No, there's the belief that the human species has races unlike cats, dogs, etc.

Just because the word is shared with the scientific concept doesn't mean it's believed to be the same thing. It's simply a remnant of the times when that was the general belief.

"There is great genetic diversity within all human populations. Pure races, in the sense of genetically homogenous populations, do not exist in the human species today, nor is there any evidence that they have ever existed in the past." - American Association of Physical Anthropologists, 1996 - http://www.virginia.edu/woodson/courses/aas102%20%28spring%2...

You missed my point. No, the human species doesn't have races in the scientific sense. What you shouldn't assume is that all mentions of the word "race" are referring to the scientific concept. Quoting census.gov:

The racial categories included in the census questionnaire generally reflect a social definition of race recognized in this country and not an attempt to define race biologically, anthropologically, or genetically.

If this "social definition" of race is accepted and somehow washed of its original meaning, why keep on using "racism" with the obviously disconnected meaning? How is racism defined in this brave new world of semantic chameleonism? As a synonym for xenophobia?

I don't think it was this definition that has mutated from the original, but the scientific definition that got more precise and diverged from the colloquial use. But I'm hardly an expert, so don't assume I know what I'm talking about.

As for xenophobia and racism, from UNESCO: Xenophobia and racism often overlap, but are distinct phenomena. Whereas racism usually entails distinction based on physical characteristic differences, such as skin colour, hair type, facial features, etc, xenophobia implies behaviour based on the idea that the other is foreign to or originates from outside the community or nation.

This is far from universally agreed on.

>when examining the frequency of different alleles (variants of a particular gene) at an individual locus (the location of a particular gene) between individuals, it is nonetheless possible to classify individuals into different racial groups with an accuracy that approaches 100 percent when one takes into account the frequency of the alleles at several loci at the same time. This happens because differences in the frequency of alleles at different loci are correlated across populations — the alleles that are more frequent in a population at two or more loci are correlated when we consider the two populations simultaneously. Or in other words, the frequency of the alleles tends to cluster differently for different populations.

https://en.wikipedia.org/wiki/Human_Genetic_Diversity:_Lewon...

In other words people's genetics do naturally cluster into different groups.

The quantitative aspect is more important than the qualitative one. Take dog races (breeds) for example:

"We also showed that the genetic variation between dog breeds is much greater than the variation within breeds. Between-breed variation is estimated at 27.5 percent. By comparison, genetic variation between human populations is only 5.4 percent. Thus the concept of a dog breed is very real and can be defined not only by the dog's appearance but genetically as well." - http://www.americanscientist.org/issues/feature/2007/5/genet...

No one is disputing that dogs are far more genetically diverse than humans. But saying that humans don't cluster into different genetic groups is factually wrong.

I'm saying that genetic clusters that flow gradually one into the other and have minimal differences between them are a distinct notion from race as we see it in other animals.

American-English is weird like that.

Race is a social construct, constructed socially in the United States.

The term "white names" drives me insane. It usually exemplifies the cultural ignorance and Anglocentrism of the speaker more than anything.

I'm sure mr. Sendhil Mullainathan, born in a small farming village in India, is completely ignorant about non-anglophone cultures.

Reply on news.ycombinator.com