Settings

Theme

What Researchers Discovered When They Sent 80k Fake Résumés to U.S. Jobs

nytimes.com

19 points by gaius_baltar 2 years ago · 40 comments

Reader

karaterobot 2 years ago

I'm not thrilled with a study that includes human subjects without their informed consent, and which resulted in them being both identifiable and called out as racists in the New York Times.

  • JambalayaJim 2 years ago

    So the impact to the employers is low; whereas the potential to help disadvantaged racialized people is high. That’s a perfectly acceptable trade off.

hermannj314 2 years ago

I took one of those cognitive bias quizzes that showed I have racial bias. It was a hard pill to swallow, to have to acknowledge I am empirically a racist and can't change that. (it dealt with reaction times or some such)

But what I can change is my behavior that doesnt involve millisecond reaction times. I can use empirical data to make hiring choices. I can stop using phrases like "culture fit" or making decisions based on intuition of how I feel about someone.

I can do my part to fight cognitive bias by acknowledging it exists and by choosing to do better and make decisions with data, but I can't even trust myself to create the criteria for evaluation without polluting it with my own biases.

I am not sure why I am posting this. Probably because I want to remind myself that growing up in the rural Midwest put ideas in my head that while I reject them rationally, they have polluted deep parts or my brain that form a large part of my intuition-based decisions.

  • JumpCrisscross 2 years ago

    > cognitive bias quizzes that showed I have racial bias...have to acknowledge I am empirically a racist

    Implicit association [1]. It means you may have an inbuilt racist tendency (or lack of familiarity), not that you are a racist. Given you proceeded to correct for it, I would strongly reject the notion that you are--empirically or otherwise--a racist.

    [1] https://en.wikipedia.org/wiki/Implicit-association_test

    • Sirizarry 2 years ago

      Yes this commenter isn’t racist and frankly I find it sad that they think they are. I grew up a minority in Chicago (admittedly very minority friendly in many ways (except the police)) and saw a lot of racist stuff against a wide variety of ethnicities. None of these moments were millisecond biases or subconscious remarks. Hell I’m probably more biased than this person yet I know for a fact that I am not racist. I hope they figure that out for themselves soon, sounds sad to constantly think of yourself in such a negative way.

      • JumpCrisscross 2 years ago

        Totally agree. There has been a neo-academic attempt to redefine racism independent of intent. Structural racism is real, but it can exist without every person within it necessarily being a racist.

gaius_baltarOP 2 years ago

Archived version: https://archive.is/qn32R

1vuio0pswjnm7 2 years ago

Feel sorry for "real" applicants having to compete with thousands of "fake" ones.

JumpCrisscross 2 years ago

“On average, they found, employers contacted the presumed white applicants 9.5 percent more often than the presumed Black applicants.

Yet this practice varied significantly by firm and industry. One-fifth of the companies — many of them retailers or car dealers — were responsible for nearly half of the gap in callbacks to white and Black applicants.

On average, companies did not treat male and female applicants differently. This aligns with other research showing that gender discrimination against women is rare in entry-level jobs, and starts later in careers.

However, when companies did favor men (especially in manufacturing) or women (mostly at apparel stores), the biases were much larger than for race.

...

Being gay, as indicated by including membership in an L.G.B.T.Q. club on the résumé, resulted in a slight penalty for white applicants, but benefited Black applicants — although the effect was small, when this was on their résumés, the racial penalty disappeared.”

That’s more positive than I expected from the headline! The problems are improving and concentrating.

  • Dig1t 2 years ago

    White people are underrepresented (and paid less) in big tech companies by a considerable margin now, according to the companies' own ESG reports that they publish. Our industry is overcompensating for these imbalances.

    I work at the biggest FAANG and just last week our team decided we are rejecting all applications for a manager position we have open until we get more diverse candidates. I'm pretty sure that's somehow illegal, but I don't want to get fired for saying that it seems wrong.

    Here is Apple's ESG report showing that white people are underrepresented (40% of the company vs 60% of the country):

    https://investor.apple.com/esg/

    Here is Microsoft's blog post about paying white people less:

    https://blogs.microsoft.com/blog/2023/11/01/microsofts-2023-...

    "Inside the U.S., all racial and ethnic minority groups who are rewards-eligible combined earn $1.007 total pay for every $1.000 earned by U.S. rewards-eligible white employees with the same job title and level and considering tenure."

    • JumpCrisscross 2 years ago

      > our team decided we are rejecting all applications for a manager position we have open until we get more diverse candidates. I'm pretty sure that's somehow illegal

      Yes, it is and should be.

      > Here is Microsoft's blog post about paying white people less

      The tone is weird, but I'll be damned if they have three digits of significance to those figures, even at a company as large as Microsoft. The correct thing to say would be that they are statistically indistinguishable.

ortusdux 2 years ago

Quantifying subtle biases is very difficult. There was a great study published in 2015 that showed that the skin color of the hands holding baseball cards in eBay listings had a measurable effect on the sale price.

https://onlinelibrary.wiley.com/doi/abs/10.1111/1756-2171.12...

sylware 2 years ago

Did they evaluate regilious in-between-ing? I don't think they could do that, it seems it is very accute. In my country, it seems so much accute, laws were setup to fight just that...

  • beAbU 2 years ago

    Like the other poster I'm also confused by the term "in-between-ing" - are you asking if they tested for religious discrimination?

    If that's what you are asking, how is it possible to discriminate if there is no religion information on the resume to begin with. Unless name/surname/location is used as a proxy?

    • JumpCrisscross 2 years ago

      > how is it possible to discriminate if there is no religion information on the resume

      You'd throw in religiously-coded social club affiliations and/or school names.

      Given OP's other comment, I think the spirit of their question points to coding the resume with where someone is on a generic political spectrum.

  • bediger4000 2 years ago

    Could you elaborate on what "in-betweening" means here?

    • sylware 2 years ago

      Looking at google, to help you understand, you could use the word "communalism", here it would mean to "significantly prefer/select from a specific religious group".

      Here an example in India: https://vajiramandravi.com/quest-upsc-notes/communalism/

      • JumpCrisscross 2 years ago

        I’m genuinely more confused after reading that. Are you asking if they tried sending resumes with religiously-coded names? (They obviously didn’t.)

        • sylware 2 years ago

          Yeah I think leaving out such accutely critical non-orthogonal "parameter" is damaging significantly the study.

          • JumpCrisscross 2 years ago

            > I think leaving out such accutely critical non-orthogonal "parameter" is damaging significantly the study

            That's a totally separate question. Does America have an entry-level religious wage gap in need of explanation?

            Like, if I were doing this study in India, sure, that would be a target variable. Perhaps moreso than race.

            • sylware 2 years ago

              I think the impact of religion "in-between-ing" is not orthogonal and not neglictible.

      • Y_Y 2 years ago

        That article has many Indian English-isms! I think the closest word I know to describe that phenomenon would be "sectarianism", though that has a pile of other connotations too.

instagib 2 years ago

hacker news finally did something about reposts and automatically upvoted the post. I tried to post this and was instead taken here. I actually tried searching for a prior post but it didn’t show up.

Working paper doi: 10.3386/w32313

  • NicoJuicy 2 years ago

    That was always the case, as far as I'm aware of.

    Edit: it's probably possible to resubmit an old article and resurface it back.

    Which works as intended

  • mikestew 2 years ago

    hacker news finally did something about reposts and automatically upvoted the post.

    "Finally"? Yeah, like what, ten years ago? Sorry for the pedantry, but maybe not ten years, but it's been a number of years that if one submits something that's already submitted, it upvotes the first submission and takes you to the comments.

    • instagib 2 years ago

      Idk how well it works but changing the title slightly or website gets right past it as I’ve seen before.

      Just discovered a new to me feature. Thanks for being a dick about it.

      • mikestew 2 years ago

        Thanks for being a dick about it.

        I'm just happy to have my efforts recognized.

Keyboard Shortcuts

j
Next item
k
Previous item
o / Enter
Open selected item
?
Show this help
Esc
Close modal / clear selection