> I strongly recommend looking up the word "bias" in the dictionary.
From the fist Google result [1], "prejudice in favor of or against one thing, person, or group compared with another, usually in a way considered to be unfair." Representation is not mentioned at all.
> Systemic and institutional bias in tech hiring isn't disproven by of the existence of female-dominated fields. Yes, some fields are female dominated while others are male-dominated. (Such a revelation!)
The point is, do you think it's valid to say that female dominated fields are biased against men with no evidence for bias aside from the disparity in representation? This is what it seems like you're claiming when you write that underrepresentation relative to the population is the same thing as bias.
> It sounds like you're angling for the "male nurses are oppressed by the matriarchy!" trope, or some kind of gender essentialist nonsense.
I agree, concluding that male nurses are subject to bias just because they're substantially outnumbered by female nurses is incorrect. But you wrote "underrepresentation of the population is literally the definition of bias." Well, male nurses are substantially underrepresented. If underrepresentation relative to the population is "literally the definition of bias" then it's hard to argue that male nurses aren't subject to bias when they are unambiguously underrepresented.
The point is not that male nurses are oppressed - the point is that they aren't oppressed and thus your definition is wrong. Underrepresentation is not bias. An institution can exactly representative of the general population but also immensely biased - like a tech company that threw out 3/4th of male applicants to make tech roles 50% women - and it can be not at all representative of the population and not biased at all - like a hospital that hires 90% female nurses because nurses are 90% women. Prejudice in favor or against one group can occur or not occur regardless of representation relative to the population.
You've been breaking the site guidelines by perpetuating this flamewar. Please stop, and please don't perpetuate flamewars on HN in the future.
It doesn't look like you have a pattern of doing this in the past on HN, and (although I've only read a small portion of this thread) I don't see how your comments deserved abuse. But it's necessary to recognize when an argument is going bad fast, and to step away from those sorts of exchanges. HN is for curious conversation, not ideological battle, and when it comes to these tedious tit-for-tat spats the crowd the right hand side of the page because neither party can let go, it really doesn't matter who wins the argument. Actually it would be better to define 'winning' as being first to let go.
> Selection bias pushes men towards high-status, high-paying fields and women towards low-paying, pink collar fields. You do realize that if one sector disproportionately hires men, then there are a disproportionately high number of women in the remaining sectors, yes?
Yes, you're right! Contrary to what you previously wrote, "underrepresentation of the population is literally the definition of bias" underrepresentation is not the definition of bias. Bias is being more favorable towards some groups as compared to others. The fact that nursing is overwhelmingly women does not mean that nursing is biased against men, because men and women make different choices. If we tried to make hospitals have 50% male nurses they would likely have to institute immense bias in favor of men to achieve representation equal to the general population.
> But beyond the statistical biases, the highly-gendered nature of tech prevents women from being seen as a "good cultural fit" by hiring managers. It prevents women from being encourages to pursue CS in school, and pushes women out of tech when then do manage to get a foothold. Whereas nursing actually encourages men to apply much more proactively because people don't get hilariously defensive if it's trying to be rectified. But I'm certain you have lengthy, sexist rebuttals prepared for how sexism isn't real, with analogous comparisons about on the job sexual harassment and diversity hiring.
Again, you've demonstrated that you fully comprehend that "underrepresentation of the population is literally the definition of bias" is not at all the case. If a tech company rejects women wrongfully then they are biased against women - regardless of whether women make up 10% of engineers at the company or 90%. Representation is not bias. Bias is bias.
If you witnessed your company wrongfully reject women candidates from the hiring process then by all means your company was biased against women. I don't claim to deny what you have witnessed. But I've witnessed my current and former companies both set up policies that explicitly bias the hiring process to push the representation of women in tech roles to be closer to 50%, even though the current representation of women in tech roles was larger than the average in our metro area. Specifically it was 19% vs 23%. Were there any women that were wrongfully rejected due to "culture fit"? Maybe. But the end result is still that women in tech were more likely to work at the company than men in tech. Is this biased relative to the general population? Yes. But is the general population applying to engineering positions Dropbox? No, only the subset of those that seek a job in tech are applying to engineering positions at tech companies.
Saying tech companies are biased against female engineers because they make up 20% of their engineers is erroneous and simplistic in the same vein that saying hospitals are biased against male nurses because nurses are 10% men. Hospitals usually only have 10% male nurses because only 10% of people that want to be nurses are men. Likewise, tech companies usually have 20% women engineers because women make up 20% of people that want to be engineers. The people studying engineering are 20% female. Are these disparities due to biases in different party of society? Probably. But saying that a tech company is biased because 20% of its engineers are women, is like saying that a hospital is biased for a 10% male workforce. They reflect the population that has decided to enter these fields. The bias, if it exists, is not in the tech company is in whatever social mechanisms to which you attribute these disparities in nursing and technology. If a hospital takes measures to make men account for 20% of the workforce then they're biasing their hiring process to favor men. The fact that biases in society may be responsible for the fact that male nurses are so rare does not change this fact.
Can you elaborate on what you meant by, "underrepresentation of the population is literally the definition of bias." Your response seems to be contracting this definition rather than supporting it.
> This is a part of it. There is also an unconscious bias against women in hiring. There's plenty of overt sexism, but unconscious biases are significant too. That whole "gut feel"-based hire is remarkably sexist and racist. As are resume callbacks in favoring male and anglophile names.
Again, when we actually try to identify this bias we rarely find the alleged bias against women.
> As are mentoring and promotional opportunities should one successfully land the job. As is unaddressed sexual harassment on the job.
Neither of which are related to bias in the company's hiring process.
> Re-read my point regarding statistical biases. The existence of a 20/80 ratio is maintained by biased hiring. You seem to believe 20% out of 50% is the "correct" ratio because the current ratio is 20%. And on we go again with your wonderful sexist circular logic.
How do hospitals know that 10% male nurses is the "correct" ratio? They don't, they just give people equal opportunity and because 10% of nurses are male 10% of their hired nurses are male. The whole point is that there is no "correct" ratio. Companies should (and are theoretically obligated to) extend equal opportunity regardless of race or gender. Be biased against no one. If you think your hiring managers are erroneously passing on women, measure this bias and rectify it. Bias isn't eliminated by picking a percentage and instructing recruiters to try and reach that percentage.
When a company tries to pick a "correct" ratio and that ratio is different from the representation of people who work in that rule, bias is usually the result. If a hospital one day decided to institute a diversity program that sets a target of 15% male nurses chances are the hospital would end up being biased in favor of men. Likewise, a tech company that decides to set a target of 30% women engineers when women make up 20% of the engineering workforce is going to end up being biased in favor of women.
You might believe that this bias is justified because of bias women might face in the jobs themselves, or bias earlier in life that you believe is responsible for the fact that only 20% of aspiring engineers are women - all that is your own opinion. But bias you feel is justified is still bias.
The level of abuse that you stooped to in this thread is shocking. I'm going to ascribe it to going on tilt, which happens sometimes when another commenter is extremely wrong, or it extremely feels like they are. We've all been there. But please don't behave like this, or anything remotely close to this, on Hacker News again.
From the fist Google result [1], "prejudice in favor of or against one thing, person, or group compared with another, usually in a way considered to be unfair." Representation is not mentioned at all.
> Systemic and institutional bias in tech hiring isn't disproven by of the existence of female-dominated fields. Yes, some fields are female dominated while others are male-dominated. (Such a revelation!)
The point is, do you think it's valid to say that female dominated fields are biased against men with no evidence for bias aside from the disparity in representation? This is what it seems like you're claiming when you write that underrepresentation relative to the population is the same thing as bias.
> It sounds like you're angling for the "male nurses are oppressed by the matriarchy!" trope, or some kind of gender essentialist nonsense.
I agree, concluding that male nurses are subject to bias just because they're substantially outnumbered by female nurses is incorrect. But you wrote "underrepresentation of the population is literally the definition of bias." Well, male nurses are substantially underrepresented. If underrepresentation relative to the population is "literally the definition of bias" then it's hard to argue that male nurses aren't subject to bias when they are unambiguously underrepresented.
The point is not that male nurses are oppressed - the point is that they aren't oppressed and thus your definition is wrong. Underrepresentation is not bias. An institution can exactly representative of the general population but also immensely biased - like a tech company that threw out 3/4th of male applicants to make tech roles 50% women - and it can be not at all representative of the population and not biased at all - like a hospital that hires 90% female nurses because nurses are 90% women. Prejudice in favor or against one group can occur or not occur regardless of representation relative to the population.
1. https://www.lexico.com/en/definition/bias