6 ms·

Blind reviews (and even interviews) are great ways of making hiring more fair. They are explicitly the inverse of DEI approaches. DEI is predicated on outcome d

by dbspin 2y ago

Blind reviews (and even interviews) are great ways of making hiring more fair. They are explicitly the inverse of DEI approaches. DEI is predicated on outcome diversity, rather than treating applicants equally irrespective of background. That's the E and I part. The entire premise is that certain groups require special support (fair - e.g.: blind people, wheel chair users), and have been historically excluded because of bias (sometimes true, often wholly false - much of the time differential hiring is path dependent with fewer qualified applicants from a given group).

scott_w 2y ago
> They are explicitly the inverse of DEI approaches. This is essentially a No True Scotsman fallacy. If it's DEI, it's bad so any good approach is, by definition, not part of DEI. > DEI is predicated on outcome diversity, rather than treating applicants equally irrespective of background. The first part of this is incorrect. Good DEI is about creating a level playing field (as you correctly point out for blind people or wheelchair users). Obviously, this isn't possible in all cases: I think everyone agrees we wouldn't want a blind taxi driver. > The entire premise is that certain groups require special support This is correct. Fair criticism of DEI initiatives can be levied at those which don't do this effectively and instead shortcut by using, say, hiring quotas. I've said multiple times that things like this are lazy and stupid because they don't address the lack of opportunity for disadvantaged backgrounds. > and have been historically excluded because of bias (sometimes true, often wholly false This is an inaccurate stating of the situation. Some groups (e.g. black people in the USA) are excluded due to bias. Some have been excluded due to situational factors (young white men in the UK have worse outcomes due to poverty). Good DEI initiatives attempt to counter these, with varying levels of success. Let me take the article as an example. They identified an advantage for people on CTI programmes, which also happened to turn out good ATC operators. This may have advantaged people who could afford to attend the programmes, which could have skewed white male. A good DEI initiative might have been to put the work into outreach in under-represented areas to get more people of colour into CTI programmes. Instead, the FAA banned CTI programmes, threw the students there to the wolves, and seemed to sneak in a test designed to hit hiring quotas. Not only was this discriminatory, it also actively reduced the number of qualified ATC operators. Nowhere in this scenarios did I need to fall back on "DEI bad," because I tried to discuss the specific issues within the article.
- thot_experiment 2y ago
  These are really good points, it's depressing as hell to see the the quality of discussion around this stuff. Obviously DEI is great when it's trying to fix things on the input side. Perhaps I can simplify this argument. If you have a lift heavy things job, which we can agree that women on average are worse at, you shouldn't hire more women by quota, but you could provide free weight training for women. Both things are DEI, the latter is the kind of DEI we want.
  - ryandrake 2y ago
    I think in your example, you shouldn't hire by quota, but you also shouldn't exclude women or introduce obstacles that exclude them. It's so weird that this has turned into such a controversial statement!
  - int_19h 2y ago
    The problem is that DEI in practice tends to be the other kind of stuff. I think at this point it's actually kinda disingenuous to pretend that "DEI" is "just diversity, equity, and inclusion" (i.e. that you can just point at the dictionary definitions of these words to explain what it is). No, it's a very specific political mindset, and the label is now firmly associated with it. You can't say that "DEI is just equality" anymore so than you can say that about "all lives matter".
    - vkou 2y ago
      > The problem is that DEI in practice tends to be the other kind of stuff And what does the political opposite of those initiatives look like in practice? What does it look like in practice when you don't stop and wonder why women make up 20% of your qualified candidate pool, but only 7% of your workforce? (As another poster observed.) Do you just shrug your shoulders, assume that your perfectly meritocratic (By whose definition?) system is free of any form of systemic or personal bias, and move on, without wondering why?
      - int_19h 2y ago
thijson 2y ago
Your mention of blind reviews reminds me of a social experiment I read about several years ago. All of this is anecdotal though. The article was written by someone that administered a web site that paired candidates with employers. Employers would conduct a phone screen via the web site to choose candidates. The web site saw that females had a lower chance of being selected, and based on the assumption that it was their gender being the reason, decided change the pitch of voices to mask their gender. This experiment actually backfired and lowered the chance of women being hired though. The author's conclusion in the end was that women had a lower chance of being hired because they gave up too easily, they couldn't handle rejection as well as men.
- thijson 2y ago
  https://news.slashdot.org/story/16/06/30/2035225/women-interviewing-for-tech-jobs-actually-did-worse-when-their-voices-were-masked-as-mens https://news.slashdot.org/story/16/06/30/2035225/women-inter...
- derangedHorse 2y ago
  I like this method of interviewing. If it results in more men initially then that's fine. As long as the mechanism for hiring is such that it reduces discrimination for everyone, then it's one worth pushing. If there are traits employers reject candidates on en-masse, then at least this data would help us analyze what these traits are. Once we know what the determining traits for hiring are, we can either debate whether their importance in the job at hand (if there are doubts) or find ways to encourage these traits in underrepresented communities.
- fwip 2y ago
  Simply pitch shifting somebody doesn't make them sound like a normal male/female speaker. There's a lot more to it, including musicality of speech, word choice, resonant frequencies, etc. If you pitch shifted the average American woman, you'd probably get a voice that sounded like a gay (camp) man.
BeetleB 2y ago
> and have been historically excluded because of bias (sometimes true, often wholly false - much of the time differential hiring is path dependent with fewer qualified applicants from a given group). It's very hard to find a company that does real "blind" interviews. And by blind, I mean where networking doesn't positively impact your application. As long as networking boosts your chances of getting hired somewhere, you've got a very wide open door to biases, because networks are almost always biased. I should not be able to give me resume to a friend to ensure the hiring manager gets to see it. Yet I haven't found a company where that behavior is detrimental.

by dbspin 2y ago

scott_w 2y ago

> They are explicitly the inverse of DEI approaches. This is essentially a No True Scotsman fallacy. If it's DEI, it's bad so any good approach is, by definition, not part of DEI. > DEI is predicated on outcome diversity, rather than treating applicants equally irrespective of background. The first part of this is incorrect. Good DEI is about creating a level playing field (as you correctly point out for blind people or wheelchair users). Obviously, this isn't possible in all cases: I think everyone agrees we wouldn't want a blind taxi driver. > The entire premise is that certain groups require special support This is correct. Fair criticism of DEI initiatives can be levied at those which don't do this effectively and instead shortcut by using, say, hiring quotas. I've said multiple times that things like this are lazy and stupid because they don't address the lack of opportunity for disadvantaged backgrounds. > and have been historically excluded because of bias (sometimes true, often wholly false This is an inaccurate stating of the situation. Some groups (e.g. black people in the USA) are excluded due to bias. Some have been excluded due to situational factors (young white men in the UK have worse outcomes due to poverty). Good DEI initiatives attempt to counter these, with varying levels of success. Let me take the article as an example. They identified an advantage for people on CTI programmes, which also happened to turn out good ATC operators. This may have advantaged people who could afford to attend the programmes, which could have skewed white male. A good DEI initiative might have been to put the work into outreach in under-represented areas to get more people of colour into CTI programmes. Instead, the FAA banned CTI programmes, threw the students there to the wolves, and seemed to sneak in a test designed to hit hiring quotas. Not only was this discriminatory, it also actively reduced the number of qualified ATC operators. Nowhere in this scenarios did I need to fall back on "DEI bad," because I tried to discuss the specific issues within the article.

thot_experiment 2y ago
These are really good points, it's depressing as hell to see the the quality of discussion around this stuff. Obviously DEI is great when it's trying to fix things on the input side. Perhaps I can simplify this argument. If you have a lift heavy things job, which we can agree that women on average are worse at, you shouldn't hire more women by quota, but you could provide free weight training for women. Both things are DEI, the latter is the kind of DEI we want.
- ryandrake 2y ago
  I think in your example, you shouldn't hire by quota, but you also shouldn't exclude women or introduce obstacles that exclude them. It's so weird that this has turned into such a controversial statement!
- int_19h 2y ago
  The problem is that DEI in practice tends to be the other kind of stuff. I think at this point it's actually kinda disingenuous to pretend that "DEI" is "just diversity, equity, and inclusion" (i.e. that you can just point at the dictionary definitions of these words to explain what it is). No, it's a very specific political mindset, and the label is now firmly associated with it. You can't say that "DEI is just equality" anymore so than you can say that about "all lives matter".
  - vkou 2y ago
    > The problem is that DEI in practice tends to be the other kind of stuff And what does the political opposite of those initiatives look like in practice? What does it look like in practice when you don't stop and wonder why women make up 20% of your qualified candidate pool, but only 7% of your workforce? (As another poster observed.) Do you just shrug your shoulders, assume that your perfectly meritocratic (By whose definition?) system is free of any form of systemic or personal bias, and move on, without wondering why?
    - int_19h 2y ago

thijson 2y ago

Your mention of blind reviews reminds me of a social experiment I read about several years ago. All of this is anecdotal though. The article was written by someone that administered a web site that paired candidates with employers. Employers would conduct a phone screen via the web site to choose candidates. The web site saw that females had a lower chance of being selected, and based on the assumption that it was their gender being the reason, decided change the pitch of voices to mask their gender. This experiment actually backfired and lowered the chance of women being hired though. The author's conclusion in the end was that women had a lower chance of being hired because they gave up too easily, they couldn't handle rejection as well as men.

thijson 2y ago
https://news.slashdot.org/story/16/06/30/2035225/women-interviewing-for-tech-jobs-actually-did-worse-when-their-voices-were-masked-as-mens https://news.slashdot.org/story/16/06/30/2035225/women-inter...
derangedHorse 2y ago
I like this method of interviewing. If it results in more men initially then that's fine. As long as the mechanism for hiring is such that it reduces discrimination for everyone, then it's one worth pushing. If there are traits employers reject candidates on en-masse, then at least this data would help us analyze what these traits are. Once we know what the determining traits for hiring are, we can either debate whether their importance in the job at hand (if there are doubts) or find ways to encourage these traits in underrepresented communities.
fwip 2y ago
Simply pitch shifting somebody doesn't make them sound like a normal male/female speaker. There's a lot more to it, including musicality of speech, word choice, resonant frequencies, etc. If you pitch shifted the average American woman, you'd probably get a voice that sounded like a gay (camp) man.

BeetleB 2y ago

> and have been historically excluded because of bias (sometimes true, often wholly false - much of the time differential hiring is path dependent with fewer qualified applicants from a given group). It's very hard to find a company that does real "blind" interviews. And by blind, I mean where networking doesn't positively impact your application. As long as networking boosts your chances of getting hired somewhere, you've got a very wide open door to biases, because networks are almost always biased. I should not be able to give me resume to a friend to ensure the hiring manager gets to see it. Yet I haven't found a company where that behavior is detrimental.