First, the section is "Non-discrimnatory ways to reduce the gender gap".
You cannot reduce the "gap" by aiming at the women who are already at Google. This should be obvious. They are already there. To make a dent in the gap, you need to attract more women who currently do not choose to work in tech or not at Google. So nothing he writes is about his colleagues at Google, or by extension women who are already in or interested in tech. It is about potentially attracting those who are not currently interested in that career.
Second, this is about subtle differences in the distribution of preferences in populations with huge overlaps. So when you have an individual in front of you, you simply cannot tell whether they will exhibit this trait more or less by inference from the population statistic. And you certainly can't say "all X have more of this property than all Y". He even has a nice graph for this.
Third, it's not about ability, but about preference. "Women on average show a higher interest in people..." How do you get from "preference" to "ability"?
Fourth, you are assuming a symmetry where "higher X" implies "lower Y that's complementary to X". This would probably be true for interests (if I a am more interested in working alone, I am probably less interested in working in groups). But it does not translate to abilities.
In fact, one of the documented areas where gender differences in ability do show up is exactly a case of this lack of symmetry: you've probably heard that tend to score higher on the math SATs and women higher on the verbals. This is apparently not the whole story: men with high math scores do tend to not also have high verbal scores (and therefore prefer STEM). However, for women the two are correlated, not anti-correlated: those with high math scores tend to also have high verbal scores, so they have more options. And people with more options tend to prefer non-STEM options, regardless of gender.
Fifth, I thought pair programming was a Good Thing™?
Only if you reduce their turnover to be below turnover for men. Which at first thought seems to be worse than discriminating only during the hiring phase.
Sure, if you explicitly exclude any way that the "gap" could be reduced by aiming at the women who are already at Google it is clear that you can not reduce the "gap" by aiming at the women who are already at Google.
> You cannot reduce the "gap" by aiming at the women who are already at Google. This should be obvious. They are already there. To make a dent in the gap, you need to attract more women who currently do not choose to work in tech or not at Google. So nothing he writes is about his colleagues at Google, or by extension women who are already in or interested in tech. It is about potentially attracting those who are not currently interested in that career.
If that were the case then he would not lead his essay implying that the current hiring policies were hurting Google. He seems to imply standards have somehow been damaged, but right now that's not what Google does.
> Third, it's not about ability, but about preference.
There are so many potential sources of preference, from negative experience to personal ability. It seems very suspect to point to gender stereotypes that a large number of women have been arguing against for decades are in fact the root cause here.
> you've probably heard that tend to score higher on the math SATs and women higher on the verbals. This is apparently not the whole story: men with high math scores do tend to not also have high verbal scores (and therefore prefer STEM). However, for women the two are correlated, not anti-correlated: those with high math scores tend to also have high verbal scores, so they have more options
Sources, please.
> Fifth, I thought pair programming was a Good Thing™?
That depends. If I told you the only way you're going to be competitive and not drag down the standard of my organization is by pairing up with another person, should you take that as a compliment?
> That depends. If I told you the only way you're going to be competitive and not drag down the standard of my organization is by pairing up with another person, should you take that as a compliment?
> Women on average show a higher interest in people and men in things
> We can make software engineering more people-oriented with pair programming and more collaboration.
That was literally the only mention of pair programming the the entire memo and it was as a suggestion to make positions more appealing to women who were more interested in people and more social. Not once does he equate pair programming as some remedial for your perceived statement of lower quality ability. The guy was trying to suggest ways to attract more women and you flip it around to demonize him. Do you not see that you are large part of the problem? Instead of trying to engage and explain to him why his points are perhaps wrong or misguided, you instead twist his words to simply eviscerate him instead.
Please excuse the delay. I'll get to breaking your links down shortly. You're asking me to source papers during my commute hour and on mobile it's difficult to grab and source the resources I'll cite.
First, the section is "Non-discrimnatory ways to reduce the gender gap".
You cannot reduce the "gap" by aiming at the women who are already at Google. This should be obvious. They are already there. To make a dent in the gap, you need to attract more women who currently do not choose to work in tech or not at Google. So nothing he writes is about his colleagues at Google, or by extension women who are already in or interested in tech. It is about potentially attracting those who are not currently interested in that career.
Second, this is about subtle differences in the distribution of preferences in populations with huge overlaps. So when you have an individual in front of you, you simply cannot tell whether they will exhibit this trait more or less by inference from the population statistic. And you certainly can't say "all X have more of this property than all Y". He even has a nice graph for this.
Third, it's not about ability, but about preference. "Women on average show a higher interest in people..." How do you get from "preference" to "ability"?
Fourth, you are assuming a symmetry where "higher X" implies "lower Y that's complementary to X". This would probably be true for interests (if I a am more interested in working alone, I am probably less interested in working in groups). But it does not translate to abilities.
In fact, one of the documented areas where gender differences in ability do show up is exactly a case of this lack of symmetry: you've probably heard that tend to score higher on the math SATs and women higher on the verbals. This is apparently not the whole story: men with high math scores do tend to not also have high verbal scores (and therefore prefer STEM). However, for women the two are correlated, not anti-correlated: those with high math scores tend to also have high verbal scores, so they have more options. And people with more options tend to prefer non-STEM options, regardless of gender.
Fifth, I thought pair programming was a Good Thing™?