Hacker News Reader

Comment by ktizo | Hacker News Reader

Ok, the underlying idea seems to be that it is highly beneficial to have segregated schools based on ability. However, if we look at the global ratings for education, this does not seem to be the case, otherwise countries such as Denmark and Finland would not be doing so well.

That said, I think that the largest effects would seem to be having high educational standards required to be a teacher, coupled to a high degree of autonomy to do their jobs and a massive public budget for resources across all levels of education, from kindergarten, right the way through adulthood. If you have that covered then you can get good results and have streaming based on ability, if you like.

View on HN

The scales used on global measures of education are scored in a way that makes a few low scores a lot more influential than a few high scores (since the average score on the test is around 70%, it takes more than 3 100% scores to balance out one 0%). So countries which are more homogenous (so that their upper end's scores don't get capped at 100%) and don't have as big of a low class do well, such as Japan, Korea, and Finland.

If the tests were structured so that the average score was 10 out of 100, one genius who scored 100 would be worth 9 people who scored zero.

I think this is a more realistic way to score education; would you rather have the median of the population go from grade level 7 algebra, to grade level 7.1? or would you rather have 20% more utter geniuses in your society? Both have value, obviously, but breakthroughs are made at the edges, and only need to be made once.

All this is to say that we should focus on where the returns are greatest.

In terms of intellect, I'd think that the returns to society from state schooling are greatest from educating the bottom 95%, as the geniuses tend to be more auto-didactic anyway and usually need more in the way of social support and access to academics rather than traditional school teaching. A society that reserves high quality education only for those children already considered to be intelligent, is a pretty stupid one in my opinion.

Also, high quality education is not the same as teaching advanced courses, although obviously advanced courses should be taught to a high quality. High quality education is about the ability to bring standards up across the board, not just the production of geniuses, which educators largely can't take credit for anyway.

The US outperforms Denmark educationally after adjusting for one of the bigger predictors of student quality, by roughly the same margin that Finland outperforms the US.

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

If we want to cherry-pick some more, Germany performs well (marginally better than the US and far better than Denmark) and does segregate schools based on ability.

I see, so you think the education problem in america is down to people who aren't descended from europeans dragging the side down.

You know, you could make exactly the same argument with income if you like and then argue that the figures would be higher if it wasn't for all the poor people.

I see, so you think the education problem in america is down to people who aren't descended from europeans dragging the side down.

It is my belief that the US is marginally better than Denmark at educating people of European descent. The data supports this belief.

I have no data on whether the US is better or worse than Denmark at educating people from other groups. If you have data on income I'd love to see it.

I don't know what you mean by "education problem". The main "education problem" in the US is cost, and this has little to do with the quality of the students.

The main "education problem" in the US is cost, and this has little to do with the quality of the students.

I partly agree with you there, the administration costs are atrocious. However, as a percentage of GDP you are low down the list on education spending even while you are near the top on spending per capita, so economically the US could afford to spend a lot more while at the same time what is being spent seems to be spent rather badly.

Personally I suspect spending a fair bit more, but firing three quarters of the management staff would be not a bad start. I also think that raising both the wages and required qualifications of teachers would be a good idea, with paid for education available to current teachers who don't yet meet those standards, and job reviews and some redundancies for those teachers who really can't hack it.

Teaching at a state school should be a very well paid, highly sought after, difficult to get into and thoroughly respected profession.

Again, the difference between the two populations "Danish" and "US minus non-whites" is that the latter neatly excludes a lot of society's poorest, whereas the former does not. It's not an apples-to-apples comparison.

You seem to be implying that the non-white poor in the US somehow lift whites out of poverty.

The objection I'd make is that unless non-white Danes are a small fraction, "white Americans" vs "all Danes" should be instead vs "white Danes" (that is, the motivation in slicing out a subpopulation of the US is not to exclude the poor, but to control in a coarse way for the influence of genetics, via the proxy "European" (really, "white").

The objection I'd make is that unless non-white Danes are a small fraction

They are a small fraction.

Non-whites in the US are heavily overrepresented amongst the poor. Excluding them excludes much of the poor from the US figures, which would raise the average performance - specifically given that the author is saying that it's the kind of homes you find in places of poverty with high crime that reduce PISA scores.

Also, claims of genetic intellectual differences between the races are dead as a dodo. No real genetic difference has been found, and if there was any, it is so small that it's utterly swamped by cultural issues. Claiming that you want to cleave off a different race to 'make the comparison fair' is tantamount to saying 'blacks are intellectually inferior to whites'.

Poverty, crime, and broken homes are presented as the thesis for poor performance in school, then the author removes a large amount of those people from the US results citing this rather racist "but, genetics!" idea.

When I moved to Sweden from the US, I was surprised to see that children born in Sweden, but with one parent from outside the country, were considered 'foreigners.' One of the books we read in Swedish class had a character who had a single grandparent from Chile. That contributed towards her dark eyes and complexion, so that another character asked "are you really Swedish?". I was astonished even, since my personal belief (being brought up in Miami during the 1970s and 1980s) is that nationality is not so stingy. My father and his parents were born in Canada, but Dad became a US citizen when he was a teen. I can't count the number of friends of mine whose parents are both from Cuba but consider themselves patriotic US citizens. Compare that to my dad's mother's mother, who forbade her daughter from dating my grandfather because she's a Norwegian Lutheran while he's a German Baptist. Those being very different cultures, you know.

These memories come to mind because the author, who is raised in the Swedish system, uses this same definition of 'foreign;' and it's a view I find rather un-American.

So when I read that essay, which excludes from consideration the rather large sub-population of people whose distant ancestors came to the US as slaves in the 1800s, who are by now fully part of the American system, with basically zero language or cultural ties to any ancestral lands, I can't help but think "so what?" I don't care if there's a sub-population which did better ... and why is race used in one place and immigration status used in another? Is the correlation with relative income more significant? It's easy to see how doing just one simple factor analysis isn't enough on its own to be anything more than suggestive.

BTW, the re-analysis on that blog shows that "the mean score of Americans with European ancestry is 524, compared to 506 in Europe, when first and second generation immigrants are excluded." Because the original analysis shows that Europe overall did better than the US, then I think it's equally reasonable to interpret the re-analysis as saying that Europe does a much better job of educating its first and second generation immigrants than the US does at educating its citizen with non-European ancestry.

In other words, since most of the excluded populations are black, the US seems much worse at educating citizens of African descent than of European descent. I would like that to change, so that all children have access to a good education, as well as to social services to help with nutrition, health, housing, and other factors that strongly affect the education and well-being of children.

From your link:

I define immigrants here as those with a parent born outside the country, so it includes second generation immigrants.

then

In the case of America, 99% of the population originates from other countries, be they England, Italy, Sweden, India, Africa, Hong-Kong or Mexico.

So if you're saying that 99% of the US population is first- or second-generation, that means that amount of population have come here in the last two generations. The baby boomers are retiring now, so lets use them as an easy 'oldest' generation.

The US currently has 300 million people. So according to this guy, just before the boomers, there were only 3 million people.

Truly, it is a testament to the industriousness of the American public how with only 3 million people, they managed to send 16 million to serve in the military in WWII, and still have enough people left over to manufacture military hardware to supply the Allies.

The guy has a point about home environment being important, and that endemic poverty for blacks in the US is a problem, but you can't just carve those out of the results. For a start, those blacks are still American, born and bred for generations. Secondly, every country has sections of entrenched poor. In the US it's much easier to divide along race lines. In the UK, you have a ton of whites that come from equally crime-riddled, poor neighbourhoods, but they don't get excised from the tampered data.

The author starts out with a reasonable concept, but horribly mangles the philosophy and method to isolate the numbers.

So if you're saying that 99% of the US population is first- or second-generation...

No one besides you is saying this.

Incidentally, race is highly predictive even holding income reasonably close to constant. For example, Asian Americans tend to have income comparable to Europeans Americans, yet their test scores tend to be considerably higher (in the same range as other Asian nations).

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

Now, you are correct that given more granular data, a better comparison could be made. Do you have more granular data?

No one besides you is saying this.

In the context of the article, these two quotes are right next to each other. The author is saying this. He's trying to paint cultural homogeneity with the parent country, which is reasonable with first and second generation immigrants, but it's a sloppy way to exclude blacks from the American scores.

Let's be very clear here with another quote:

So similar to my comparison of GDP levels, let us compare Americans with European ancestry (about 65% of the U.S population, and not some sort of elite) ... We remove Asians, Mexicans, African-Americans and other countries that are best compared to their home nations.

The author is removing all blacks as they are 'best compared to their home nation' as he has defined them as 'immigrants'. (he's also exluding hispanics, apparently, despite a lot of them already living on land that was acquired by the US in the 19th century - how these are 'immigrants' is beyond me)

The thing is, blacks in the US don't live like their West African forebears, nor raise their kids like West Africans. They live like Americans and raise their kids like Americans. They are Americans and have been for multiple generations. To exclude them on 'American' scores is pretty racist.

Also, how can you really make a representative argument when you cut out one third of your subject pool? If the system only 'works' if you exclude them, then that means that the system is not working so well for that particular country.

Reply on news.ycombinator.com