re: "Women in tech, where are you?" VIEW POST

re: This whole bit is standing on James Damore's memo and the research he cites (note: I only read the abstract of the paper) basically stating that th...

First off, some (honestly!) friendly advice: if you want to have a productive conversation about feminist matters, go very carefully on positively quoting Damore or using phrases like SJW as you do in another comment. Even the moderate liberal audience has a high chance of being turned off by such things. I try very hard to be open to all viewpoints, so I'm not instantly walking away here, but some people will.

When I was quickly going through your comment, quite frustrated at the time for other reasons I might add, my initial emotional reaction was steering me towards being defensive (read "being a dick"). Literally reading the first paragraph in annoying patronizing voice. And as you say, it was difficult to read more from there. It did not feel good. And that's my problem! You are being genuine and to the point!

Of course, it does not feel good having your faults pointed out, it does not feel good to be wrong, it does not feel good to be offended by something, may it be even be a little thing. And that's fine, natural even. We do not know each others intent, emotional state, history and hardships we had to go through. And in an open discussion, it is necessary to calmly read again and really 'listen' to what the other side has to say. And one would do that, if they're truly interested in the discussion (you just did that!) - that's where I'm getting at with PC being a deterrent in my other comment.
Well, and if simply mentioning someone else's name reduces the number of participants, so be it. I can't imagine having an open discussion with such person. A person convinced against their will is of the same opinion still.

As I mentioned, I'm not taking any stance. I'm simply taking a premise that the research suggest and bulding from there.

Sorry, back to the point. I could argue PC all day.

First, let's note a few points that, I think, we can agree on and build from there:

  1. (A) Women are just as intelligent as men (confirmed by studies).

  2. (B) There are differences of deeper nature affecting the career choice of men and women.

  3. (C) The difference (B) is small, even insignificant, on average.

you are making a big leap in assuming (1) the difference is large

Consider a measure of 'being interested in computer science'.
Let's take a random man and a random woman out of the population. We ask the question "Who is more interested in CS and by how much"? And the answer is, statistically, that the man is more interested in CS, but only by a little, as expected. Now let's look at the numbers - how much are both interested in CS? Well not much - because average person is not interested in CS! I hope that's not surprising and we can denote it as point (D). We learned not to look at the average, because the average does not matter in this case. We need to look at the people, that are interested in CS. Simplified:

  1. Denote the measure as C - 0 being not interested in CS and 9 being very interested in CS.
  2. Assume C follows normal distribution.
  3. Let's say only people with C > 7 will pursuit CS.
  4. Let's say the difference in C between genders is only 1 in favor of men.
  5. Plot the data for each gender - horizontal axis being C and vertical axis being percent of population of men/women in given C column.
  6. We are now presented with two overlapping bell curves, men's curve being shifted by 1 to the right.

(2) it causes [significant] differences in career choices;

Now sum everything from C > 7 - that small difference in the middle results in significant difference at the end (5) because C does not scale linearly and follows normal distribution.

(3) it causes differences in skills that have a [significant] effect on how capable people are at succeeding in a career.

Similar statistical 'proof' can be used on other, already mapped, personality traits.

Math doesn't lie. But we can question this if:

  1. Personality traits do not follow normal distribution or

  2. C does not correlate with personality traits in expected manner or

  3. C does not follow normal distribution.

But evidence seem to suggest otherwise.

I'm still working on getting data about enrollment, or rather dropoff rates during college. I have higher priority data I'm digging up first, like exposure rates in high school. Follow me if you want to keep an eye on what I find. :-)

Wil do!

Also note: the percentage of women in software has changed over the years. From an all-time high of 37% in the 80s to a steady-state around 27% in the 90s to a new steady-state of 18% for the last decade or so. You can't say that's just "human nature" - there are societal factors involved in those changes.

There is an explanation for this that is aligned with the claim.

  1. (E) The study found, that more equal society gets, the bigger the difference between men and women. Scroll down to Discussion and continue from there.

  2. (F) Specifically for CS: to not be interested in CS, I have to know what CS is. Awarness of what CS is has definitely gone up since the 80s.

  3. (G) I can also argue the other side: If (E) does not hold for CS, then following (F), one could predict, that CS will be on the rise, since tech world is becoming a part of more and more people's lives and businesses. That CS will go through social 'normalization'.

Now if we are using 'CS' in general sense, then I believe that (G) holds even if (E) holds as well, because it will be quite difficult to avoid 'CS' in the tech-enabled future (it's becoming an essential skill).

If we are talking CS as (basically) a branch of mathematics, then the representation of men and women might stay more or less the same. My uni for example is very open, very theoretical and mostly full of dudes.

Now, in context of all of this, it might be interesting to see the data from the universities. From applicants to continued career in the field of study. That's why I'm fishing.

Thank you for taking my advice in the friendly spirit in which it was offered. It was not intended to be a criticism or judgment on you. I totally understand how it can be easy to interpret this kind of discussion in a negative kneejerk way. Discussions on the internet are hard.

I think the main thing you and I differ on is the degree to which we believe people's behavior is governed by nature vs. nurture. Let me offer as a counter-argument, my own discussion of the nurture side of it.

I am not nearly as interested in trying to determine what % is nature and where that leaves us, as I am in figuring out how to balance out the nurture side of the equation and THEN seeing where nature leaves us. Because I do not think we are "done" fixing the environmental factors involved. Let's spend our time on those.

The good thing is, to whatever extent environmental factors ARE part of the imbalance, we can DO things about them. So my post attempts to start addressing that.

Thank you for taking my advice in the friendly spirit in which it was offered.

That's not what I'm doing at all, quite the opposite in fact. I agree that discussions are hard but I have no intention of adjusting my speech (which I consider to be mild) for sake of not offending anyone.

balance out the nurture side of the equation and THEN seeing where nature leaves us.

Did you read the findings in the paper? The countries with more advanced 'nurture' side observed, that it has the opposite effect than expected. The gender ratios across different fields has gone down instead of up.

Now I totally agree with you that we have a lot of work in terms of making women feel more welcome and cultivating interest in women (for STEM fields specifically for example).

My point is: You are most likely focusing on the wrong number. And the (reputable) research supports that.

So what happens when there is an enormous amount of pressure on businesses and institutions on the topic of gender equality? Policies gets introduced - and thats not good for anybody. And I believe, that was the message of the memo.

If we want to move forward, we need to figure out what the relevant metric is and how to effectively measure it.
From the top of my head Strive for equal and maximal ratios of men and women, who are happy with where they are..

And for me, 'happines' in this context is:

  • doing something fulfiling, something that makes me feel good
  • being secure (financially and otherwise) and safe
  • time-dependent variable

Did you read the findings in the paper? The countries with more advanced 'nurture' side observed, that it has the opposite effect than expected. The gender ratios across different fields has gone down instead of up.

Until now I only read the abstract of the paper. Because you said that was all that you had read. I just spent some more time with it. I think you need to, also. Even in the abstract it says (emphasis mine):

Overall, higher levels of human development—including long and healthy life, equal access to knowledge and education, and economic wealth—were the main nation-level predictors of larger sex differences in personality. Changes in men’s personality traits appeared to be the primary cause of sex difference variation across cultures. It is proposed that heightened levels of sexual dimorphism result from personality traits of men and women being less constrained and more able to naturally diverge in developed nations.

In the Discussion section you suggested, it repeats this message on page 176 under the graph:

The larger contribution to this correlation came from men’s shifts in personality as the mean value of their averaged scores on four dimensions— Neuroticism, Extraversion, Agreeableness, and Conscientiousness—was significantly correlated with HDI, r(53) = .56, p = .001, whereas the same correlation for women was insignificant, r(53) = .17, p = .22.

And page 178.

Moreover, these changes appear to result from men’s cross-cultural personality variation. In more traditional and less developed cultures a man is, indeed, more like a woman, at least in terms of self-reported personality traits.

In other words, in the little bit of the paper you suggested I read, it looks like the differences are coming from men's changes in behavior, not women's. Yes, I'm cherry-picking science here, which I'm not actually proud of. But the main point I take away from this paper is that personality traits are partially determined by your environment, which is actually saying there is a lot less nature and more nurture in our behavior than you might think.

Look, I've been in software for 20 years and known many people. I have discussed career stuff with many men and women alike. I have been working with high school kids for 13 years, trying to get more of them interested in software. And I say again, there are many factors involved in the gender ratios we see today, but nature is far down the scale - there's a lot of stuff we need to fix still.

code of conduct - report abuse