Hacker News Reader

Comment by stickfigure | Hacker News Reader

Sure, the grey zones might be hard to pick out but it does seem like there are some clear extremes. "Locomotive engineers and operators" is 92% male yet "Registered nurses" is 87% female. If you look at all the male-dominant and female-dominant occupations, there does seem to be some sort of pattern which (for lack of more precise terminology) seems to be well-described as "people-oriented" and "thing-oriented".

Go through the list, look at the entries which are either <20% female or >80% female, and tell me if you don't see the same pattern:

https://www.bls.gov/cps/cpsaat11.htm

View on HN

Isn't software development a gray zone, yet is male dominated?

Software development is playing with legos for a living, except with a billion pieces and the ability to make your own pieces, with infinite floor space. It’s “thing-oriented” and only grey if you start mingling in the secondary designers, managers, and such external requirements. But it’s fundamentally “thing oriented” like nursing is fundamentally “people oriented” despite the variety of equipment and tools a nurse must be familiar with to do their job.

Look at the chart. "Computer and mathematical occupations" is male majority, but not male-dominated like say train drivers or mechanical engineers. Some subfields (web and interface designers, notably) are pretty balanced.

We study diseases by (first) looking at people who have severe forms. Ie it would be tough to come up with an understanding of autism just by studying the brains of people who are "a little aspy". If you can figure out what causes the gender segregation of train drivers / pediatricians, maybe you can apply that knowledge to relatively-mildly-imbalanced tech fields.

I'm still not convinced, the amount of "social pressure" from day 0 (or even before) for a kid is just shocking (starting with: I won't buy a pink doll to my newborn baby boy).

Obviously with animals is different, but there might also be other things at play that we are not aware.

Happy to be wrong, but I think there are so many layers to make a call at this point that sure, there might be an influence, but society is probably what affects this the most.

Though it does seem obvious, it's pretty dangerous to go looking for patterns in the data. More scientific to define the criteria before knowing the results so you can have a hypothesis to test, not just an observation to make.

No, looking at the data is observation or analysis by definition...whether or not you make a prediction about it. Experimentation is when you make a hypothesis first (after observation) and then test it. Looking at data is not a test--it's either observation or analysis.

Dangerous? That's how science... Works? You look for patterns, you get some data, you find more patterns, you create a hypothesis and you test it.

What is dangerous about analysing data?

Or do you mean it's dangerous to form absolute conclusions just on a subset of available data?

The danger is that you will come to a conclusion that is objectively wrong.

P-hacking is a known issue in many scientific fields. So is drawing conclusions from over collection of data without repetition of the study... As the number of data points you collect approaches infinity, the chance of finding at least one meaningful-seeming correlation approaches 100% because having an unlimited number of data points to bash against each other makes you more likely to observe an improbable correlation.

This seems like a political argument, not a scientific one.

Reply on news.ycombinator.com