Women and Minorities in Science (Warning: Basic calculus and statistics; political incorrectness)

http://www.lagriffedulion.f2s.com/women_and_minorities_in_science.htm

by _h4xr • 19 years ago

7 46 19 years ago

lagriffedulion.f2s.com

amalcon2• 19 years ago

This entire article appears to be based on the assumption that standardized tests can be used to determine anything useful.

I suspect everyone can see where that's going.

_h4xr14• 19 years ago

Well, here's the approximate conversion between SATs and IQ: http://www.gnxp.com/MT2/archives/002360.html . And here's an interesting scatterplot: http://en.wikipedia.org/wiki/Image:IQatWoN_GDP_IQ.png . Put those together, and...

Also, assume they don't determine anything useful, but that schools have been magically fooled into thinking they do: what part of the conclusion to this essay changes?

amalcon2• 19 years ago

1) IQ tests are also standardized tests. If they are somehow beyond question, why do schools just not use IQ tests?

2) Nothing. I just doubt that schools have actually been magically fooled into thinking they do: They actually get to do the regression analysis, and see that past grades are a better predictor than standardized tests. This has been well known for a long time. If applicants have been fooled, however, it allows the school to get useful effects from the test (even if the effect isn't predicting future performance).

axiom2• 19 years ago

That scatterplot comes from http://en.wikipedia.org/wiki/IQ_and_the_Wealth_of_Nations

Which by the way predicts that the average IQ of Ethiopia is 63. Note that an IQ of 70 is considered mentally retarded. Anyone see a problem with this?

Edit: my point is that it's absurd, and the study is nonesense!

npotter• 19 years ago

where's that signal to noise essay from pg? I've always viewed standardized tests as a way to approximate intelligence, but they aren't very good ones at that. A search will show a lot of discussions about why.

It's pretty silly to use scores that are partially determined by a given input to prove that some people have that input.

It's circular causality at best and blatant deception at worst.

npk2• 19 years ago

Bleh, so what. This guy claims to have measured something, but who knows if it's even interesting. At the end of the article, do you even understand what the point in reading it was?

I think what is interesting is tagging the article as "politically incorrect." I really hate this term. It forces people to think, "If I don't agree, or find this offensive, I'm small minded."

What makes this particular article reek of crackpotdom is that it offers no references to any other pertinent articles, yet it acts as if it is "scientific" you know, with all that math.

Don't bother reading this thing.

_h4xr14• 19 years ago

I hate the term 'politically correct', but you've got to admit that it drew more viewers (and votes) than some other adjectives I could use. And if you don't tag something like that, people whine.

pius2• 19 years ago

Agreed on all points. The whole "warning . . . politically incorrect" thing is not only lame, but smacks of a good ol' boy telling a ribald or xenophobic story to his buddies with a wink and a nod. Might as well have just said "We all get it, but you know how those chicks and minorities whine about every little thing . . . they just don't have any sense of humor."

Don't know what OP's intended tone was, but it certainly didn't come off well to me.

_h4xr14• 19 years ago

I apologize if it didn't come off well. What I wanted to avoid was a response like "That's racist!" I should probably ask doctors how they do that -- do they say something like "The racist disease, sickle-cell anemia, tends to strike people of African descent," or what?

I mean, the point of the article was that, given some data about tests, the correlation between tests and academic appointments, and some basic math, you can at least say "If a group is under-represented in the sciences, it is mostly because their test scores are low." Which means that either the test is inaccurate, and they're really underrepresented, or the test is accurate and they aren't. It's not about the research -- it's about showing where the prejudice has to be hiding if there's any prejudice to be found.

Tichy3• 19 years ago

Hm, in one sentence he claims that the gap shows up in all tests across all cultures, but then he goes on talking only about SAT scores. Some citations would have been in order... I don't think the SAT thing is very interesting: presumably (or possibly, at least) it is still a culture thing that women are less interested in maths, which would explain the gap by the end of high school.

Several years ago I took some interest in the Bell Curve and IQ tests. The IQ test people also claim their tests are culture insensitive, but that is definitely not true. Just think about the following kind of test: show 10 kanji symbols on a page of the book and on the next page ask the reader to identify them. Clearly people familiar with kanji will have big advantage. That might sound constructed, but there were times when IQ testers concluded that people from primitive societies can't think in 3d because they would build a drawing of an abstract box as a 2d structure. That's so obviously a culture thing... Anyway, not refutations of IQ testing, but enough to let one be sceptical.

_h4xr14• 19 years ago

"I don't think the SAT thing is very interesting: presumably (or possibly, at least) it is still a culture thing that women are less interested in maths, which would explain the gap by the end of high school."

That could be it. But I'd actually be more worried about students who were just "less interested" than students who were interested and tried really hard.

"The IQ test people also claim their tests are culture insensitive, but that is definitely not true. Just think about the following kind of test: show 10 kanji symbols on a page of the book and on the next page ask the reader to identify them."

Actually, what IQ testers found was that the culturally insensitive tests exacerbate group differences. Unless there's something cultural in being able to remember a sequence of numbers and recite it in reverse, or to have high reaction time, or to remember the orientation of a set of geometric shapes.

DaniFong10• 19 years ago

I used to tutor junior high and high school kids in physics. In junior high, nearly all the top students were girls. For some reason, in high school, there was this huge switch -- suddenly, most of the girls weren't interested in physics or math. It was weird, but I suspect that it's largely a matter of social pressures.

Luckily, I've had no direct experience with this. I skipped high school.

Tichy3• 19 years ago

There is definitely something cultural to being able to remember sequences of numbers in reverse: I am sure societies that don't have a concept of numbers would have a much harder time with it... As for the geometric shapes, remember the example of the africans building 2d models of the 3d drawing. It seems likely that there is a cultural component to seeing abstract shapes. The Kanji example was perhaps bad - OK, think about recognising kaukasian and chinese faces. They are just faces, no? But somehow it is hard for many kaukasians to tell chinese people apart, and presumably vice versa. I don't think that is genetic, it is trained. It makes me think twice about the cultural aspects of recognising "geometric shapes".

For completeness I want to mention that I don't doubt that IQ tests measure something. Like if they predict success in the military or whatever, they are probably right, ie the IQ63 guy from Ethopia (from another post) would probably have a much harder time in the US military than the IQ 100 American. It is just not clear that it measures "intelligence" per se. And perhaps for the Ethopian military, the results would be reversed. Ie try putting up "surviving in the Saharah without water" as an IQ test.

DaniFong10• 19 years ago

The article is bullshit. I didn't read anything after the quote "On all standardized exams, including IQ tests, men score higher than women on mathematical reasoning sections."

That's simply not true. Women are known to score higher (and have since the 60's) on tests of calculation, which is just one example.

To have such a sweeping fallacy stated as a general truth betrays a disgustingly prejudiced intent.

There are volumes more wrong with the whole essay, but really it's not worth my time. If you take any particular statement there, and dig a little deeper, you'll be sickened by what comes out.

_h4xr14• 19 years ago

"Women are known to score higher (and have since the 60's) on tests of calculation, which is just one example."

That's true. Usually, the breakdown is that women are much better at arithmetic and the like, but men are much better at more abstract problems. Which is going to make someone a better EE professor?

DaniFong10• 19 years ago

Well, we weren't talking about EE professors. But, if we were, we're talking about the top 1% of quantitative ability, in which case there's barely any statistical validity in what is written there at all.

Here's a good piece of data. The Center for Talented Youth at Johns Hopkins has been testing youth (~12 years old) across the country since 1972. One of the first things that they noticed was that of the students scoring above 700 on the math SAT, boys outnumbered girls 13 to 1. Okay, wow.

They have, of course, run the tests every year since then, and now it's down to a 3:1 ratio. This indicates a very, very strong cultural variable. The fact that this isn't even mentioned, again, betrays bias.

Tichy3• 19 years ago

Interesting, how is the "abstractness" measured? Isn't arithmetic rather abstract?

npk2• 19 years ago

Dear Dani -- i think most people on this board agree with you. I suggest you read: http://paulbuchheit.blogspot.com/2007/03/are-you-my-audience...

to remind yourself that you don't need to expend much effort on this issue :)

jsnx3• 19 years ago

I know a few smart women. Even if there are important statistical differences between men and women in the way of analytical ability, we still have to take individual men and women on their own merits.

It's too bad that colleges are unfairly admitting sub-par women, but then again, that's probably not as bad a thing as admitting all those god damn business majors...

pius2• 19 years ago

I'm not particularly impressed by the research but, then again, I'm black . . . I'm probably not bright enough to understand all this "basic calculus" and other "smart" talk.

viergroupie• 19 years ago

Well, the distribution of asian SAT scores looks vaguely bimodal, so let's validate our preconceived notions by arbitrarily stuffing two gaussians inside. Very methodical.

DaniFong10• 19 years ago

With enough Gaussians, we can say anything!

Women and Minorities in Science (Warning: Basic calculus and statistics; political incorrectness)

Related Stories