4 ms·

we should be promoting equality of opportunity not equality of outcome - i dont understand how this chart with %of white people makes any sense at all - did you

we should be promoting equality of opportunity not equality of outcome - i dont understand how this chart with %of white people makes any sense at all - did you do multivariate analysis to see what factors are a concause for someone to be hired?

darawk 6y ago
I think he is promoting that. This article is written in response to the steady drumbeat of "diversity in tech" hit pieces in the tech media over the last few years. Its just pointing out the hypocrisy of the tech media on this issue.
swyx 6y ago
the problem with this framing of equality is that you forget some outcomes ARE opportunities. opportunities compound, throughout lifetimes. affirmative action hiring is "two wrongs make a right" type thinking. it's not ideal. doesn't mean its better than the alternative, which is "oh let's just fix every injustice in society". won't see that in our lifetime, won't see that in 100 lifetimes. what if equality of opportunity STARTS with mildly rebalancing equality of outcome?
- luckylion 6y ago
  > what if equality of opportunity STARTS with mildly rebalancing equality of outcome? What if it doesn't? Then you're sowing aninmosity by unjustly favoring some people over others, spend billions of dollars, throw back humanity's progress by not chosing the best person for the job etc, all for nothing. Before you start, you better be damn sure that you're 100% right. The recent examples with the gender pay gap all but vanishing when you calculate it scientifically do not bode well for these kinds of ideologies even getting the fundamental data right - hard to believe they'll get the therapy right if they fail on that stage already.
  - 6gvONxR4sf7o 6y ago
    The pay gap might largely vanish when you control for different factors, but that doesn't mean there's no gap or no issue. That just means it's driven by gaps in those other factors. If you pressure different people to make different career choices, for example, you can't just then control for those choices and say it's all good and everything's fair.
    - luckylion 6y ago
      And if you remove all agency from people, they can't make any choices, everything they do is considered a choice they were pressured into. The gap vanished as a "gender gap". The "if you work less hours you will get less money" gap isn't quite the same.
6gvONxR4sf7o 6y ago
Equality of opportunity is really close to equality of outcome. If two groups are offered the same opportunity, but one is pressured to take it and the other is pressured to take something else, the opportunity is the same but the fact that the outcomes are different speaks to something else going on, like the pressures in my example. Given the history we have, inequal outcomes can pretty frequently be directly tied to issues that still don't provide the kinds of agency we want. I think of it as equal opportunity being a lever, while equal outcome is a measurement. For example, suppose you find some data about outcomes in different groups with covariate X. Maybe there's an outcome gap between groups which goes away when you account for X. That doesn't mean the disparity in outcomes is fair. It just punts that question to a new question of whether the disparity in X is fair. And so on.
- manfredo 6y ago
  This is a vast assumption, and one that often produces inequalities more than it reduces it. In reality, people's preferences often do vary between groups even without discrimination or inequalities of opportunity. E.g. over 90% of pediatricians are women, and there's no evidence that this is due to discrimination against men. The representation of women in technology is actually highest in countries with less opportunities for women, and lowest in countries with the greatest equality between men and women. Efforts to force an equal outcome in the context of unequal preferences does not reduce inequalities of opportunity, it produces inequalities.
  - 6gvONxR4sf7o 6y ago
    I would say the same about the case that you’re making. It’s a vast assumption that produces more inequality than it reduces. It’s a question of nature versus nurture. Our preferences are a function of our expectations and the expectations of us. I agree that the underrepresentation of men in pediatrics, for example, is a function of preferences. But I think those preferences are functions of nurture more than nature. If men were the primary caregivers for children, I expect the representation would be flipped. Taking today’s society and culture as a given, yeah I agree with you. But I think that’s a problem with the prevalent culture.

darawk 6y ago

I think he is promoting that. This article is written in response to the steady drumbeat of "diversity in tech" hit pieces in the tech media over the last few years. Its just pointing out the hypocrisy of the tech media on this issue.

swyx 6y ago

the problem with this framing of equality is that you forget some outcomes ARE opportunities. opportunities compound, throughout lifetimes. affirmative action hiring is "two wrongs make a right" type thinking. it's not ideal. doesn't mean its better than the alternative, which is "oh let's just fix every injustice in society". won't see that in our lifetime, won't see that in 100 lifetimes. what if equality of opportunity STARTS with mildly rebalancing equality of outcome?

luckylion 6y ago
> what if equality of opportunity STARTS with mildly rebalancing equality of outcome? What if it doesn't? Then you're sowing aninmosity by unjustly favoring some people over others, spend billions of dollars, throw back humanity's progress by not chosing the best person for the job etc, all for nothing. Before you start, you better be damn sure that you're 100% right. The recent examples with the gender pay gap all but vanishing when you calculate it scientifically do not bode well for these kinds of ideologies even getting the fundamental data right - hard to believe they'll get the therapy right if they fail on that stage already.
- 6gvONxR4sf7o 6y ago
  The pay gap might largely vanish when you control for different factors, but that doesn't mean there's no gap or no issue. That just means it's driven by gaps in those other factors. If you pressure different people to make different career choices, for example, you can't just then control for those choices and say it's all good and everything's fair.
  - luckylion 6y ago
    And if you remove all agency from people, they can't make any choices, everything they do is considered a choice they were pressured into. The gap vanished as a "gender gap". The "if you work less hours you will get less money" gap isn't quite the same.

6gvONxR4sf7o 6y ago

Equality of opportunity is really close to equality of outcome. If two groups are offered the same opportunity, but one is pressured to take it and the other is pressured to take something else, the opportunity is the same but the fact that the outcomes are different speaks to something else going on, like the pressures in my example. Given the history we have, inequal outcomes can pretty frequently be directly tied to issues that still don't provide the kinds of agency we want. I think of it as equal opportunity being a lever, while equal outcome is a measurement. For example, suppose you find some data about outcomes in different groups with covariate X. Maybe there's an outcome gap between groups which goes away when you account for X. That doesn't mean the disparity in outcomes is fair. It just punts that question to a new question of whether the disparity in X is fair. And so on.

manfredo 6y ago
This is a vast assumption, and one that often produces inequalities more than it reduces it. In reality, people's preferences often do vary between groups even without discrimination or inequalities of opportunity. E.g. over 90% of pediatricians are women, and there's no evidence that this is due to discrimination against men. The representation of women in technology is actually highest in countries with less opportunities for women, and lowest in countries with the greatest equality between men and women. Efforts to force an equal outcome in the context of unequal preferences does not reduce inequalities of opportunity, it produces inequalities.
- 6gvONxR4sf7o 6y ago
  I would say the same about the case that you’re making. It’s a vast assumption that produces more inequality than it reduces. It’s a question of nature versus nurture. Our preferences are a function of our expectations and the expectations of us. I agree that the underrepresentation of men in pediatrics, for example, is a function of preferences. But I think those preferences are functions of nurture more than nature. If men were the primary caregivers for children, I expect the representation would be flipped. Taking today’s society and culture as a given, yeah I agree with you. But I think that’s a problem with the prevalent culture.