[untitled]
20 comments
Judge people by the content of their character and their ability to do the job. Giving certain people advantages does help them, yes, but it leads to resentment.
> Judge people by the content of their character and their ability to do the job.
Given that resumes with black-sounding names get fewer callbacks than the identical resume with a different name (http://www.nber.org/papers/w9873), it's pretty clear we're not very good at doing that.
Given that resumes with black-sounding names get fewer callbacks than the identical resume with a different name (http://www.nber.org/papers/w9873), it's pretty clear we're not very good at doing that.
I can't see how you made the jump from "character and ability" to "resume"–one's name is not one's character nor one's ability.
In fact the submission by the founder of interviewing.io seems to imply that more technical evaluations (focus on ability) lead to more diversity:
> At interviewing.io, we rely entirely on performance in anonymous technical interviews - not resumes - to surface top-performing candidates, and 40% of hires are people from non-traditional backgrounds and underrepresented groups
In fact the submission by the founder of interviewing.io seems to imply that more technical evaluations (focus on ability) lead to more diversity:
> At interviewing.io, we rely entirely on performance in anonymous technical interviews - not resumes - to surface top-performing candidates, and 40% of hires are people from non-traditional backgrounds and underrepresented groups
> I can't see how you made the jump from "character and ability" to "resume"–one's name is not one's character nor one's ability.
How does one judge someone's "character and ability" if the resume went into the trash because of their name?
How does one judge someone's "character and ability" if the resume went into the trash because of their name?
What if there is no resume, as in the case I cited? I did not find original poster's stated goals to be disagreeable: hire based on character and ability. If it means the tech industry moves towards blind auditions, as the case is with certain music auditions, so be it.
The alternative is to artificially add advantages to one group, and like the poster said, it leads to resentment and people questioning if a job was given for merit or quota. And I would add that the focus in jobs becomes more of a political and social one, as your in-group fights for process advantages, versus focusing on skill.
The alternative is to artificially add advantages to one group, and like the poster said, it leads to resentment and people questioning if a job was given for merit or quota. And I would add that the focus in jobs becomes more of a political and social one, as your in-group fights for process advantages, versus focusing on skill.
> What if there is no resume, as in the case I cited?
I suspect you find out the person's name at some point.
> I did not find original poster's stated goals to be disagreeable: hire based on character and ability.
I don't find the goals disagreeable.
I find our ability to do so - even when we've the best of intentions - very questionable.
I suspect you find out the person's name at some point.
> I did not find original poster's stated goals to be disagreeable: hire based on character and ability.
I don't find the goals disagreeable.
I find our ability to do so - even when we've the best of intentions - very questionable.
So you've given up completely? It seems to me that we've only scratched the surface of what we could be doing process-wise. The US isn't even that bad, in lots of other countries, a photo gets attached to your resume. I go back to the blind audition idea because this actually worked: https://www.theguardian.com/women-in-leadership/2013/oct/14/...
So what similar ideas can we do in tech? I don't need to know someone's name to read a resume, in fact a unique identifier would be more useful than a name.
So what similar ideas can we do in tech? I don't need to know someone's name to read a resume, in fact a unique identifier would be more useful than a name.
Why didn't they control for poor-sounding names? Rightly or wrongly, some names have an association with low income. What if the white names were "Cletus" and "Brandy Lynn"? And instead of "Lakisha", they used "Jasmine" or "Kiara"?
There are various controls detailed in the study.
For example, they checked for some class associations:
> Applicants living in better neighborhoods receive more callbacks but, interestingly, this effect does not differ by race.
and this bit:
> First, we examine whether the race-specific names we have chosen might also proxy for social class above and beyond the race of the applicant. Using birth certificates data on mother’s education for the different names used in our sample, we find little relationship between social background and the name specific callback rates.
> ...
> We also argue that a social class interpretation would find it hard to explain all of our results, such as why living a better neighborhood does not increase callback rates more for African American names than for White names.
This bit likely controls indirectly for class, as well:
> Whites with higher quality resumes receive 30 percent more callbacks than Whites with lower quality resumes, a statistically significant difference. On the other hand, having a higher quality resume has a much smaller effect for African Americans.
For example, they checked for some class associations:
> Applicants living in better neighborhoods receive more callbacks but, interestingly, this effect does not differ by race.
and this bit:
> First, we examine whether the race-specific names we have chosen might also proxy for social class above and beyond the race of the applicant. Using birth certificates data on mother’s education for the different names used in our sample, we find little relationship between social background and the name specific callback rates.
> ...
> We also argue that a social class interpretation would find it hard to explain all of our results, such as why living a better neighborhood does not increase callback rates more for African American names than for White names.
This bit likely controls indirectly for class, as well:
> Whites with higher quality resumes receive 30 percent more callbacks than Whites with lower quality resumes, a statistically significant difference. On the other hand, having a higher quality resume has a much smaller effect for African Americans.
Sure, I would agree if by "quota" you mean quota in hires. I agree that the hiring decision should be based on some "ideal of meritocracy" (with the caveat you should be careful about bias). However, putting a quota like "there should be at least one woman among the interviewees" seems perfectly reasonable and has proven to improve both performance and diversity (a great example is the last UN secretary's policy). Basically, if you ensure that you are not excluding people at the interview stage (which regrettably is very much the status quo), you do get both great hires and (accidentally) diversity.
What happens when you don’t get any women CVs passed from the recruiters for that opening? Do you wait? Do you mandate recruiters to forward a woman’s CV regarding of qualifications? There are a lot of very fine lines to walk here without a clear good answer for any.
That’s all nice and dandy until you realize that many companies don’t look outside of their bubble for recruitment. Either internal referrals or only recruiting from a few colleges.
Tech companies should not use quotas at the time of hiring. HOWEVER they should absolutely use quotas on their upstream pipeline. They should impose those quotas on recruiting firms, and fire those who can’t meet the bar. At some point in their history, Google did this for gender diversity, with the founders enforcing the policy themselves. It worked. If they had persisted in that courageous policy as they grew, and extended it to other under-represented groups, Google would not be in the representatives rut it is today.
This makes much more sense indeed. If I'd try to hire 25% minority groups into my company from a pipeline that currently tops out at 5% applications, it wouldn't do any good for my company. I'd LOVE to hire more women for the team, but it's rare they apply.
Part of me wonders if the diversity quota debate could disappear if passionate tech companies invested educational resources in communities where they are seeing a lack of diversity. For example, instead of biasing your recruitment or hiring process, invest money in underprivileged communities for tech education...computer labs, free teaching resources etc. Then your pool of diverse qualified candidates will be larger by virtue of helping them at an earlier stage, and the sigma of "you only got hired because of your race/gender" will largely go away.
Merit, tech skill based please.
They shouldn't. Now, you should make sure that your employees and hiring managers aren't making hires based on race. And that goes for favoring _any_ racial group.
This is pretty naive. Everybody who is not a raging racist thinks they are "colorblind" when it comes to race, but the fact that we let biases unconsciously guide our hiring decisions is pretty well established (Take the John/Jennifer study from Yale for instance). Having a "mechanical" way to fight the bias helps: sure, base the hiring decision only on some magically objective measure of merit that you come up before the interviews that has nothing to do with non-job-related characteristics, I agree with that; but also just ensure that minority groups are actually represented in a reasonable relative proportion in the applicant pool, otherwise there is definitely bias happening somewhere in your pipeline.
no, they should not.