Comment by yummyfajitas | Hacker News Reader

Comment by yummyfajitas | Hacker News Reader

Parent Full thread

Some bloggers in the United States persist in blaming these outcomes on the ethnic diversity of the United States (ignoring the ethnic diversity of Singapore and other countries that outperform the United States).

Which bloggers?

Incidentally, I think "ethnic diversity" is a red herring. Any bloggers who talk about "ethnic diversity" are merely trying to couch their conversation in PC language to avoid ad-hominem accusations of racism.

The issue is not "ethnic diversity". The issue is that certain ethnicities underperform. Specifically, African Americans and Hispanics (30% of the US, 0% of Singapore) underperform. Nonhispanic whites (about 65% of the US and close to 0% of Singapore) achieve mid-level performance. Asians (about 4% of the US and close to 100% of Singapore) overperform.

These effects are HUGE in comparison to inter-country effects. Asian Americans score about 10 pts below Singapore. The EU15 scores about 50 pts below Asian Americans, and a couple of points above All Americans.

http://super-economy.blogspot.com/2011/01/how-well-do-above-...

http://super-economy.blogspot.com/2010/12/amazing-truth-abou...

It is simply incorrect to pretend that ethnic gaps in education do not exist, or to pretend that they do not explain a large portion of the gap between the US and the rest of the [edit: first] world.

Also, it would be helpful if you were a little more specific in your citations. Citing a gigantic PISA reports is much less useful than citing a specific table or figure. I'd love to learn more, but I don't have time to read the whole thing, and I have no idea which parts you are referring to.

[edit: wanted to clarify that I don't think the gaps between the US and poor locations, e.g., rural inland China or India, are primarily due to ethnicity. But gaps between wealthy first world countries do seem well explained by such factors.]

Unfortunately, you can't disaggregate in the way suggested above. Blacks and Hispanics are also far less wealthy than whites in this country (which has a larger impact than even income). Once you control for wealth a lot of the gap disappears.

Ethnic gaps may exist, but they probably are symptomatic of gaps in wealth more than anything else. But who konws. There are so many other factors at play for people of darker skin in general in society that I'd hesitate to subscribe to one theory.

With that said, it does point to the fact that US public schools are probably better than most believe them to be.

Once you control for wealth a lot of the gap disappears.

I haven't looked at it extensively, but I don't believe this is true.

This study shows that SES (Socio Economic Status) doesn't explain much of the gap.

http://www.umich.edu/~rdytolrn/pathwaysconference/presentati...

This study (sorry, can't find a non-paywall version) shows that even holding income constant, blacks do worse than whites at the college level. In particular, low income whites perform as well as high income blacks.

http://www.jstor.org/pss/2963200

Also, even if you could explain the Black/white income gap via wealth, racism or other environmental factors, how would you explain the White/Asian gap?

http://en.wikipedia.org/wiki/File:Personal_income_race.png

And how would you explain that in spite of wildly different environmental factors (compare Singapore to Texas), there is a relatively small gap between Asian Americans and Asians? And similarly, there is only a small gap between white Americans and white Europeans?

(By the way, I'm not trying to claim there are no other factors at play. I'm just pointing out that race does seem to be a biggie.)

The study you note focuses on income. There's another branch of research that focuses on wealth that shows that wealth is more important than income when it comes to things like college attendance.

See: http://collegepuzzle.stanford.edu/?p=1590

"“The Differential Impact of Wealth vs. Income in the College-Going Process” finds that wealth and income affect the college choice process differently, with wealth consistently being more significant in predicting who enrolls in college, and the type of college they attend – even after controlling for student differences in academic achievement, habitus, social capital, and cultural capital."

More research clearly needs to be done here, but from what I've seen its clear that wealth looks to dominate income.

And don't get me wrong, I'm not saying there aren't other differences. In fact I suspect there are societal and cultural differences that will manifest in one way or another. If you spend a lot of time in Asian households in the US, you'll see they often resemble, culturally, canonical Asian homes more than "US" homes.

And there's data about the success of African immigrants to the US. Who often don't come wealthy (although some are), obviously are Black, but still do well. So there's clearly a culture issue here too, but I don't think we can really tease this apart without more data.

You could be right about wealth, though I'd be surprised - it would be really weird if wealth effects in the US caused Asian Americans to perform almost exactly as well in school as Asians, and white Americans to perform almost exactly as well as white Europeans.

But this doesn't change my original point - US schools do only slightly worse than Singaporean schools at educating people with an Asian-style home environment, Asian genes, or whatever else being Asian is a proxy for. And cultural factors in black America may cause black Americans to perform worse than whites and African immigrants. Whatever the underlying factor is, it has nothing to do with the school system. That's the only point I'm trying to make.

Whatever the underlying factor is, it has nothing to do with the school system.

This cannot possibly be true. School system are artificial constructions created by men to serve men. They are not given laws of nature. The only thing we can tell from the above is that the current school system does not serve well black Americans. School system should adapt to whatever the "cultural factors" of the populace they serve.

Whether this should be through prolonging schooling hours; providing stronger and more persistent emphasis in personal hard work; correcting a harmful self-image; getting rid of institutional racism; using different instruction techniques; or any other method that may addressed the "underlying factor" can only be discovered through research, experimentation and a willingness to change and adapt.

It is easy to find examples of a schooling that adapts to the unique circumstances of some children, for example: intense language training for children from immigrant families, schools at hospitals or schools that follow a circus around the world (http://www.guardian.co.uk/education/2007/jan/26/schools.uk).

It may well be the case that one or both of US and Singaporean schools are worse at educating blacks/whites/hispanics than hypothetical school system you can dream up.

It is irrelevant when considering the question of whether Singaporean schools do a better job than US schools. For all we know, Singaporean schools might be worse than US schools - i.e., they could education Asians equally well, but blacks/whites/hispanics vastly worse [1]. But it wouldn't matter since Singapore doesn't have any blacks/whites/hispanics.

[1] You actually see this effect (Simpson's paradox) when comparing US states. Texas is better at educating blacks, whites and hispanics than Wisconsin, but appears worse when averaging over the entire population.

http://iowahawk.typepad.com/iowahawk/2011/03/longhorns-17-ba...

I recall reading about a study that showed that self-image (gender and race) contributed to performance on a test.

Yeah, more to the point, there's about 500 million asians in rural inland China who will fail the hell out of a math exam (or often a literacy exam). Meanwhile, there's a few million (citation needed) asians in America who share the trait of having parents with the smarts/resources to move halfway across the world, along with a few million in Singapore living in a first world city-state and another billion or so in developed China / Japan / South Korea / Taiwan / etc.

Looking across those examples, parental income and access to good schools seems to be a more powerful indicator than ethnicity. It just so happens that the 2 are correlated in America.

Meanwhile, there's a few million (citation needed) asians in America who share the trait of having parents with the smarts/resources to move halfway across the world,...

...or at least an employer willing to ship them across the ocean in return for indentured servitude as manual laborers in the 1850-1900 era. And it's not so much parents as great great great grandparents.

Asians got to the US the same way as Italians and Irish, and pretty much at the same time.

I haven't conducted a scientific study on the matter but 85% of the Asians I've met are 1st or at most 2nd generation, with another 10% recent immigrants, and not many descendants of the guys who built the railroad. Maybe it's a little different out west.

The reason for this is the Immigration Act of 1965:

http://en.wikipedia.org/wiki/Immigration_and_Nationality_Act...

Especially look at the section called "Long-Term Results". I haven't studied this personally, but my history professor in college claimed that this act is essentially responsible for the fact that America has the ethnic diversity it does today.

Basically, before 1965 only a very small number of non-white people were allowed to come to the United States. (Partly because of immigration restrictions, partly because of the Gentleman's Agreement of 1907 in which Japan agreed to restrict immigration for us.) After 1965, the quotas were much more equal for different regions of the world.

Reply on news.ycombinator.com