3 ms·

I am with you 100%. We keep trying to invent ways to skew the hiring towards women. Let's just stop trying to candy-coat it. You want gender ratios, so just hir

I am with you 100%. We keep trying to invent ways to skew the hiring towards women. Let's just stop trying to candy-coat it. You want gender ratios, so just hire based on gender ratios and deal with the fallout. This narrative that women are somehow grossly discriminated against in tech doesn't hold water anywhere I see. Every tech company is desperate to hire more women engineers to increase their diversity quotient. The trouble is that most applicants, by a 95%-to-5% margin, are men. It only makes sense that 95% of the hires are men. If we want more women engineering talent in the office, we simply need more women applicants. Focusing on the interview stage is the wrong part of the pipeline to start at.

umanwizard 10y ago
> You want gender ratios, so just hire based on gender ratios and deal with the fallout. That would be illegal, at least in the US.
- cderwin 10y ago
  It also ought not be done, for its moral consequences are quite grave. While it's not an argument made above, I suspect the moral consequences of "candy-coat[ing] ... gender ratios" are equally grave, and the just solution is not to force a "fair" outcome but to instead work towards attaining a fair process. Adding artificial biases towards women does not solve a sexism problem -- at least not directly.
- existencebox 10y ago
  So I realize this is delayed to your post and likely to be buried, but this brought such a fascinating thought that I couldn't not mention it. I'm going to assert something first; you may chose to disagree, but my statement requires this lemma: We are (as a society/science/etc) absolutely _shit_ at most things psychology. Next to "food science" it's the field I give some of the least credence to findings in unless they are so massively replicated and built in an unrealistically airtight way. The very nature of these fields, the number of convoluting variables and our inability to even observe outcomes accurately makes the process of study extremely difficult, and "p-hacking" (as another comment elsewhere referenced in a very appropriate noun choice) becomes very feasible and less detectable. (As a "data-engineer/scientist/whatever we're called these days" I see this effect daily; it's a common topic in my office that often the desire from management is not to find a result, but for us to find the result that says what they want, and that's in often FAR more measurable and empirical spaces) From this, my core question; if it is illegal to hire based on gender ratios, what about the process of adjusting your hiring practice, adjusting which findings you chose to perscribe to, such that you achieve the end ratio you desire? To illustrate this I'll give an absurd example to demonstrate "feasibility": Let's say I'm actively racist and don't want to hire white people. I start by putting my hiring centers around inner city low cost residences; I preferentially hire lower income candidates "as a social program", and I hire with a bias based on referrals of existing candidates. I argue that my motive is achieved through a layer of "legal" "massage-able" legitimate decisions. So the questions I'd leave with: First, when do these stop being "reasonable choices" in isolation and start becoming a discriminatory strategy Second, and regardless of my above, can you see why I might find the cut and dry statement of "hiring based on a desired gender ratios is illegal" VERY hypocritical in the current environment of EXTREMELY STRONG DESIRE to hire based on gender ratio, even if it's a "rectification"? (To this topic, I'm also curious tangentially in how affirmative action fit into this framework in the past, I understand the purpose and don't want to turn this into an argument of it's merits, but it seems like a prime example where the courts pulled off making discrimination legal, so the black and white statement of "IS ILLEGAL" seems like it's been dealt with in the past) Finally, do you see the parallels I'm trying to draw, or can you find a flaw in the correlation?

umanwizard 10y ago

> You want gender ratios, so just hire based on gender ratios and deal with the fallout. That would be illegal, at least in the US.

cderwin 10y ago
It also ought not be done, for its moral consequences are quite grave. While it's not an argument made above, I suspect the moral consequences of "candy-coat[ing] ... gender ratios" are equally grave, and the just solution is not to force a "fair" outcome but to instead work towards attaining a fair process. Adding artificial biases towards women does not solve a sexism problem -- at least not directly.
existencebox 10y ago
So I realize this is delayed to your post and likely to be buried, but this brought such a fascinating thought that I couldn't not mention it. I'm going to assert something first; you may chose to disagree, but my statement requires this lemma: We are (as a society/science/etc) absolutely _shit_ at most things psychology. Next to "food science" it's the field I give some of the least credence to findings in unless they are so massively replicated and built in an unrealistically airtight way. The very nature of these fields, the number of convoluting variables and our inability to even observe outcomes accurately makes the process of study extremely difficult, and "p-hacking" (as another comment elsewhere referenced in a very appropriate noun choice) becomes very feasible and less detectable. (As a "data-engineer/scientist/whatever we're called these days" I see this effect daily; it's a common topic in my office that often the desire from management is not to find a result, but for us to find the result that says what they want, and that's in often FAR more measurable and empirical spaces) From this, my core question; if it is illegal to hire based on gender ratios, what about the process of adjusting your hiring practice, adjusting which findings you chose to perscribe to, such that you achieve the end ratio you desire? To illustrate this I'll give an absurd example to demonstrate "feasibility": Let's say I'm actively racist and don't want to hire white people. I start by putting my hiring centers around inner city low cost residences; I preferentially hire lower income candidates "as a social program", and I hire with a bias based on referrals of existing candidates. I argue that my motive is achieved through a layer of "legal" "massage-able" legitimate decisions. So the questions I'd leave with: First, when do these stop being "reasonable choices" in isolation and start becoming a discriminatory strategy Second, and regardless of my above, can you see why I might find the cut and dry statement of "hiring based on a desired gender ratios is illegal" VERY hypocritical in the current environment of EXTREMELY STRONG DESIRE to hire based on gender ratio, even if it's a "rectification"? (To this topic, I'm also curious tangentially in how affirmative action fit into this framework in the past, I understand the purpose and don't want to turn this into an argument of it's merits, but it seems like a prime example where the courts pulled off making discrimination legal, so the black and white statement of "IS ILLEGAL" seems like it's been dealt with in the past) Finally, do you see the parallels I'm trying to draw, or can you find a flaw in the correlation?