Everyday News Update

Google’s Diversity Rumpus – The Fun Thing Is He’s Right About The Cause Of The Gender Disparity – Forbes

BERLIN, GERMANY – APRIL 25: The Google logo hangs among plants at a juice stand at the W20 conference on April 25, 2017 in Berlin, Germany. (Photo by Sean Gallup/Getty Images)

There is the most wondrous rumpus going on about what are, at root, some fairly mild comments on the gender diversity within the tech world. What makes it all even the more delicious is that the basic observation being made is correct–and it’s also the basic observation which explains most of the gender pay gap that people like shouting about. That gender pay gap is indeed there when we look at the average incomes of men and women, what we want to know is why is it there? Three possible explanations appear, one is that women get paid less than men because they’re women. This is unlikely as we’ll come to. The second is that women are specifically discriminated against in the sort of jobs they can get. There’s no doubt that this used to happen and there’s very little reason to think that it does today as Gary Becker pointed out. The third is that either there are different aptitudes and or desires, again on average, across the populations of men and women. That most certainly used to be true too and it’s possible, obviously, that it still exists. Any of these would explain the gender pay gap and only one of them seems likely upon examination.

The flip side of this is that for it to be a useful and correct explanation for the existence of the gender pay gap then we should be seeing an imbalance in the gender numbers in certain occupations and professions. Which is exactly what the general complaints about women in tech are, are they not?

Of course, it’s not quite politically correct in certain quarters to be making these sorts of points which is why there is the rumpus:

It started as an anti-diversity memo on Google’s internal mailing list. Then it tumbled—first in bits and pieces, then in its entirety—into public view.
“Google’s Ideological Echo Chamber,” is a 10-page manifesto published by an anonymous Google software engineer that argues several points, chiefly among them that the search giant’s left-leaning biases are shutting down conversations about their flawed diversity agenda. “[W]hen it comes to diversity and inclusion, Google’s left bias has created a politically correct monoculture that maintains its hold by shaming dissenters into silence.”

What appears to enrage the most is this snippet from the document:

Note, I’m not saying that all men differ from women in the following ways or that these differences are “just.” I’m simply stating that the distribution of preferences and abilities of men and women differ in part due to biological causes and that these differences may explain why we don’t see equal representation of women in tech and leadership. Many of these differences are small and there’s significant overlap between men and women, so you can’t say anything about an individual given these population level distributions.

This is exactly the same as the point we all need to understand about IQ distributions. Yes, it is true, there are differences in average IQs across time, place and even race. Leave aside that IQ tests might well just be measuring the societal biases of whoever created them, it’s still true that those averages are useless to us in trying to evaluate an individual. Because the variations across individuals within any one of those groupings are hugely greater than any difference in the averages. It’s just about possible to both walk and breathe as a human being with an IQ of 60 and there are entire clubs for people with IQs up at 150, 160 and so on. That’s very much larger than any societal difference that anyone’s ever observed–the average is useless to us, and please do note it’s useless not just not a good guide, in evaluating an individual–so we shouldn’t be using the average in our decision making.

There are those who get very angry at people pointing this sort of thing out of course:

You have probably heard about the manifesto a Googler (not someone senior) published internally about, essentially, how women and men are intrinsically different and we should stop trying to make it possible for women to be engineers, because it’s just not worth it.

That is not what was said of course. As above, it was quite specific about talking about propensities, not individuals. Probabilities, not certainties. And we really are pretty certain such propensities exist. Here’s Simon Baron Cohen (No, Sacha is the cousin), one of the main researchers in the area, on the point:

On the empathy quotient (EQ), which asks a range of questions about how interested you are in people and their emotional lives, and how involved you become in other people’s feelings, women as a group score higher than men. On the systemising quotient (SQ), which asks you how interested you are in systems of different kinds (maps, gadgets, car engines, forecasts, structures), men as a group score higher than women. This has given rise to the idea that in a typical female brain, interest in empathy is stronger than interest in systems, whilst the typical male brain is more interested in systems than in empathy. Of course, a proportion of both sexes are equally interested in emotions and systems.

Our research group has recently analysed the proportion of each sex with each of these profiles, and the results are striking. For every 10 men, six will have a male brain, two will have a balanced brain, and two will have a female brain. In contrast, for every 10 women, four will have the female brain, four will have the balanced brain, and two will have a male brain. This leads to certain conclusions. First, the sexes do differ on average. Second, women seem to have specialised more (as a group) to be better at empathy, and men seem to have specialised more as a group to become better systemisers.

Engineering is rather about that systemising. Again, before anyone starts shouting, this is again about probabilities and propensities, not a determination of the skills or desires of any one individual.

Coding, engineering, is that very systemising. Note that this does not, by any means, lead to the conclusion that all men are better at coding or engineering than all women (I am fully in touch with my feminine side on this very point). Rather, that in any random group of people who happen to be good at engineering we would expect to find more men in that group than women. So when we look at those who happen to be doing engineering for a living we might make two points. The first being that we would indeed rather like it that people do for a living what they have an aptitude for: this division and specialisation of labour thing works better that way. The second is that we really shouldn’t be surprised at all at there being a gender imbalance here.

Which brings us back to our three possible explanations for the gender pay gap. That people aren’t being paid different sums for the same job just on the grounds of gender should be obvious for what we don’t see. Doing that has been illegal for decades and we do not see the near entire laywerly population of America storming the court rooms to take win only fee cases. And given American lawyers if such law breaking were happening upon any scale then we would see them doing so.

The second idea is that employers are discriminating on what jobs people are hired to do. That’s certainly possible and it has most definitely happened before in tech:

At which point a little story from British history. Way back in 1962 Stephanie Shirley noted that there was an awful lot of that taste discrimination going on in the programming workforce. Specifically, women with children were simply not getting employed. So she specifically set out to hire women with children as programmers for her newly founded FI Group. Made an absolute fortune doing so too. Because that female programming talent was being discriminated against she could get talent cheap: at one point of her 300 programmers 298 were female. And there very definitely was gender discrimination at the time. What is now Dame Stephanie called herself Steve for many years simply so as to be able to speak with the potential buyers of the company’s services.

Which leads us to an interesting thought experiment. If you believe in the self-selecting into occupations thesis then there’s no to little underpaid female programming talent out there. If you believe in the taste discrimination one then there is. Which should mean that it would be possible to hire that female programming talent at a bargain price and thus make a fortune. So, to test your thoughts: do you think that a deliberate campaign to preferentially hire female programmers would make extra profits for a firm?

We do not see people preferentially hiring women because they are cheaper. Thus working through the logic here we should be assuming that there is not that taste discrimination against women in hiring practices. All of which really only leaves us with the third explanation as the only viable one. The one which this Googler is being so excoriated for pointing to, the Simon Baron Cohen (recall, an actual academic researcher into this issue) one, that there are indeed differences across the populations. Again, to insist again, this tells us nothing whatsoever about an individual, only that if we’re selecting for some skill or mindset the percentage of which varies across gender then we’re going to end up with a workforce that looks gender skewed.

Google’s Diversity Rumpus – The Fun Thing Is He’s Right About The Cause Of The Gender Disparity – Forbes

Leave a Reply

Featured Links

    Search Archive

    Search by Date
    Search by Category
    Search with Google

    Photo Gallery

    @2012 Designed By Quick News Update