This is sleight of hand. Even supposing all that "data" to be accurate, you assume your conclusion by acting as if you know what the reason for it is. But, as we all know, nothing is more highly disputed.

This is not the way to an open-minded and open-hearted discussion about the truth; it's a tangent to yet another generic ideological flamewar. That violates the HN guidelines.

Since you've done this repeatedly before and don't appear to have any other purpose in commenting here, we've banned your account. Please stop creating new accounts to break the HN guidelines with.

What why?

The SAT scores provided by the parent comment is official data. It is not "data" as you connote it to be. Just follow the Government stats linked he published.

The conclusion of the parent comment is not unwarranted. Tech companies pride themselves in hiring only the best and there is strong statistical correlation between SAT scores or any other standardization test vs academic/professional performance.

Tech companies want the best now, it doesn't matter if some group of people are best due to inherent awesomeness or lot of privilege or both or neither. Similarly it doesn't matter if some group of people are not among the best due to financial hardship in their families, racial biases against them.

In the current paradigm, tech companies should either hire the best or hire for diversity. They cannot do both until the underlying society has already changed.

Is pointing out racial/gender differences in SAT scores or IQ tests or in hiring at Facebook/Google/Amazon/Microsoft equivalent to being a racist?

That's not what that comment argued.

Yikes. All he did was present some data (with references) that you dispute. You banned him for that? You seem to be the one who isn't interested in an open-minded and open-hearted discussion about the truth.

I made it clear why I banned him: repeated creation of accounts to violate the HN guidelines with, a.k.a. serial trolling.

Ycombinator themselves are acting reverse racist by preferring to hire minority groups, so these rules are very hypocritical. Just saying.

And banning for a post that provides evidence is ridiculous. You should apply for a government job in Turkey.

> reverse racist

Racism implies an overarching ideology. A true racist would say something like, "White employees are better than black employees."

Y Combinator has never suggested that minorities are categorically better employees. Preferring to hire from minority groups has many benefits unrelated to race: increased diversity boosts profits[1], for starters. There's a huge difference between racism (trying to keep minorities out of your office) than what you're calling "reverse racism" (trying to have a diverse office).

1. https://www.sciencedaily.com/releases/2009/03/090331091252.h...

It's not just sleight of hand. Race isn't even a scientifically defined concept, and it can't currently be defined or measured with existing technology. Everyone is a wide mix of many races.

Genome-wide clustering algorithms that are run with no a priori assumptions about population structure end up discovering human population groups that match almost perfectly with self-identified "race".

The claim that race doesn't exist is simply false. We can talk about whether or not it's useful to make policy decisions based on race, but if we're respecting the data, we can't say that there are no coherent human population groupings smaller than homo sapiens or that these groups do not align with traditionally-understood "race".

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1196372/

It's incredibly frustrating to see well-intentioned people just flatly deny facts about the world. The mistrust and resentment this denial causes makes it much more difficult to get useful messages across.

Again, there is no scientific consensus on how to define race, and many scientists see it as an impossible or meaningless taxonomy[1]. That's the core problem. You've chosen to use a purely genetic definition.

My assertion is that everyone is a mixture, which is true. People exist on a continuum of race without discrete borders[2].

1. http://www.scientificamerican.com/article/race-is-a-social-c...

2. http://www.nature.com/ng/journal/v36/n11s/full/ng1435.html

The position on that continuum of race is just as correlated with the same metrics as the self-identified classification.

The issue 'race is correlated with X' is not materially affected by how exactly race (especially its borders) is defined and taxonomical problems - it's important in it's own regard, but in this regard, it's not the core problem, it's a red herring distracting from the core question of "does race affect X'; and that question is the same, with the same answers, no matter if in your analysis you choose to model race as discrete borders or as a continuum.

Hmm, yes those popular racial categories of "European/West Asians" and "Hispanic".

If your test can accurately group "Hispanics" which no-one actually claims is a race, then it doesn't prove that any other grouping it finds is a "race", however that is defined.

Assume for a second these tests are valid ways to prove what you are trying to prove[1].

First, you can't simply reslice the data and think you have shown anything valid. Simpson's paradox and all that. It could be that women do better than men in all but one race, for example, and that race "brings the average down" to the point where the overall gender scores look like they do. In fact, if you look at the wikipedia article on simpson's paradox, it gives the great example of .. a gender bias study :)

Whoever is creating these numbers by confounding variables and grouping data from other studies is just making a mess of the original data :(

(This is sadly not uncommon, and happens a ton in race and crime statistics, etc. The folks who make the original data often go to great pains to explain their methodologies, etc, and then some random org just throws it all in excel so they can produce charts that support whatever, and acts like anything they do to the data makes it still valid)

Second, ignoring that serious issue for a second, The only thing this gives you is "a snapshot of the world today". This is a test of actual current performance trying to predict success, not trying to predict potential performance. The SAT gives no data on what innate ability is, nor does it try to. That means it can change, and you can improve the numbers.

[1] i'm pretty sure the main correlation the SAT shoots for, and succeeds at, is predicting whether folks will be "successful the first year of college". There are a number of studies that show this. There are also newer, but less studies, that show other predictors may be just as good.

These numbers tell you where the bump is in the bell curve, but tell you nothing about who's out on the long tail. Learn to treat people as individuals, not statistics.

This list also says nothing about cause/effect. Do the scores represent innate aptitude, or rather the result of a lifetime of societal biases?

> result of a lifetime of societal biases.

Not just societal biases, but simple demographic realities that are not controlled for at all looking solely at scores on a test.

I would say at the moment there's still a feedback loop. Little girls somehow learn that they're expected to be not as good at math, and maybe they try less hard and give up more easily. Not to mention socio-economic backgrounds when you bring race into it.

What you are describing is stereotype threat. High-quality evidence for it is scarce to none; a recent meta-analysis: http://www.sciencedirect.com/science/article/pii/S0022440514....

The evidence for poverty affecting education performance, however, seems pretty high here. Example paper out of many: http://www.academicjournals.org/article/article1379765941_La...

And there's more too it, of course. I chose this link primarily because I found another observation very interesting: "In many studies, [the] mother’s education [level] had a more significant effect on children’s scores than income."

Equating inborn aptitude to SAT scores alone seems very short-sighted.

The interesting thing about the study you cite is that it only focuses on children and adolescents. You fail to mention that the majority of research has been carried out in college student populations and the findings are fairly robust. One resource for dozens of articles is the bibliography at reducingstereotypethreat.org; since you seem to be down on it I'll just link straight to their discussion of criticisms of stereotype threat [1].

However, stereotype threat is not the only mechanism that fits the poster's description. Most Americans, upon hearing that I'm a mathematician, say, "I hate math! and I suck at it." Then they stare at me hostilely or say, "You must be really smart." There's a little subset of 40-65 year old women, though, who say wistfully, "I loved math and it was my best subject... but... in high school they said I wouldn't need it so I couldn't take the calculus class, I had to take typing instead..." Tracking in high school is used extensively and dissuades people tracked into crappy high school STEM classes from pursuing STEM degrees in college (or getting in at all).

[1] http://www.reducingstereotypethreat.org/criticisms.html

That's not just stereotype threat - to give an example, there are people I know in their early 20s who were strongly discouraged from doing higher level maths in high school, and as a result, had a much harder university level maths experience.

Why do you think these numbers are fixed? They can change, if there exists competency and a desire to learn.

You are also ignoring the effects of changing environment which those numbers disregard. For example, we know blacks disproportionately come from low income backgrounds with higher crime and a "rough" home life, which can affect a childs ability to effectively study and learn. That's not because they're genetically black as you allude, that's because they grew up in that environment. Your data does not capture or ignores this crucial variable and you should stop pushing it.

Even if the data you're pointing to was accurate and wasn't influenced by upbringing, they represent averages. Not everyone in each of those groups is going to have an aptitude for maths, but enough will.

Also, aptitude for mathematics and aptitude for tech aren't directly correlated, though there is likely to be some overlap.

Wow the bigotry. What is wrong with you? These tests are created and scored by privileged white people.

Please don't respond to a bad comment by violating the HN guidelines yourself. That just makes this place worse.

... if so, then why do asians appear to dominate them?

Going by your extremely tenuous suggestion that genetics is an accurate predictor of math/engineering aptitude because of how people of certain ethnicities or sex perform on the SAT, we should have a distribution in tech that looks like this:

- Asian/Pacific Islander = 29.6%

- White = 26.4%

- Mexican American = 22.8%

- Black = 22.1%

- Men = 51.6%

- Women = 48.4%

In other words, roughly equal numbers of all of the demographics (specifically _not_ an overwhelming majority of White Men).

So even if your argument were valid it seriously undermines your point.

Not that it likely impacts your point much, but does this adjust for population sizes?

I totally misinterpreted the data; for some reason I assumed it was the number of students from each group that scored above some threshold, rather than the average score of that group.

My calculations are completely garbage, and I would withdraw the comment but it looks like I can't at this point.

Aren't you making a big assumption about linearity?

I think it's more that I totally misinterpreted the data. For some reason I assumed it was the number of students from each group that scored above some threshold, rather than the average score of that group.

My calculations are completely garbage, and I would withdraw the comment but it looks like I can't at this point.

And I bet this would be true, if women didn't gravitate moreso to the natural sciences.