I've always heard this explained as a sort of "selection bias". Since immigration to the US (particularly for university education) is often seen as desirable, the people who manage to pull it off tend to be above the mean. Do you feel that explanation rings false?
I think this is an explanation that could only be come up with by the descendants of those who have emigrated.
Thinking here in Scotland, the people who emigrated were not necessarily the most able or genetically superior somehow. Often, they were simply the most desperate. People who were cleared off their farms by landowners, people who had no other options available to them but to roll the dice and go abroad to Canada or Australia or the USA.
Most folk don't want to emigrate, certainly not in the 19th century. It is a last resort that you do if you are out of options. But perhaps the most capable and able have other options to take advantage of?
"Collective emigration is, therefore, the removal of a diseased and damaged part of our population. It is a relief to the rest of the population to be rid of this part."
I suggest your theory is probably very true for Asian Americans - the one's who came here are 'la creme do la creme'.
But not for Europeans. Europeans that came here were the poorest, the least educated, criminals, fringe religious types etc.. Europe was 100x more civilized than America during early history - why would anyone with any social status leave London in 1800 - to go and live a million miles away, a very, very hard, back-breaking life?
Well - those tenant farmers who could get cheap land and get out from under the thumb of their landlords etc..
But it's a good point.
Europeans in those statistics include all ethnicities living in Europe - e.g. about 1/3 of students in Germany are not ethnically German.
https://en.wikipedia.org/wiki/Academic_achievement_among_dif...
The "immigrant advantage" fades in three generations, at which point achievement by children reflects the US average. This is the influence, if you like, of American "ethnicity", which is often quite anti-intellectual.
The intro here is an interesting read: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3442927/
There is no need to bring up race or ethnicity to describe this effect of scores getting worse. Everything you attribute to being Black or Hispanic can more simply and more accurately be described by being poor. SAT score is most directly tied to family income, not race. http://blogs.wsj.com/economics/2014/10/07/sat-scores-and-inc...
Instead of worse scores being the result of changing ethnic composition, they are the result of changing income distribution, and the expansion of the lower class.
+ This is not true +
Even when normalized for income - there are still large variations in outcomes.
Again, I'm not really saying it's 'race' or 'genetic' or 'IQ' so be careful with your disdain :).
My friend - do you know the stereotype of the 'Asian who works hard in school because his parents compel him to' - well, it's not just a stereotype, it's true. Some cultures value education more than others. That will show up in the results.
Asian Americans are 600% over-represented in tech (i.e. 5% of population but usually 30% of tech companies). That's not some fluke - and it's not because 'Asians are super rich' - it's obviously a cultural preference. They are choosing STEM and tech for whatever reasons.
It doesn't make some people better or worse than others as human beings, it just means some variations in outcomes whenever you measure something.
This this is what actual 'diversity' really means, though ironically I think most 'pro diversity' advocates really intend to have a hyper-egalitarian situation wherein there is little diversity beyond skin tone :)
> Unfortunately, it's so sensitive few will want to talk about it - for fear that the general public equates educational outcomes to 'intelligence' and try to strongly correlate ethnicity + race to this, which would be fodder for racist/KKK types, which wouldn't really help the overall social situation in America.
Fortunately we have you.
If he did go looking for statistics, he would find that the effect he describes is better explained by the expanding lower class in america, regardless of race.
http://blogs.wsj.com/economics/2014/10/07/sat-scores-and-inc...
It's not ok to comment like this here.
Several of your recent posts have crossed the line into incivility. Addressing people as "buddy", telling them about the "fetish of [their] ideology", etc.: all this is patronizing and rude. Please don't do any more of it on Hacker News.
If you're going to engage in difficult topics, you have a responsibility to do so with extra respect, not less.
Because preeminence of top tier institutions (which are kind of global centres anyhow) - has absolutely nothing to do with teaching math to the commons.
Here's a hint:
+++ Americans don't suck at Math +++
There's a very un-PC but very large elephant in the room that people won't discuss.
+ European American and Asian American 'testing scores' are actually pretty good - and have been holding steady for a very long time. (Asians do a little better). Nothing has changed.
+ Latino American and African Americans fare poorly, but having been getting better since we've been measuring by standards (i.e. 1950's-1970's).
Here's the trick:
+ European Americans actually do better than Europeans - on average. + Asian Americans to better than Asians - on average. + Latino Americans do better than Central/South American Latinos + African Americans do better than Africans.
The key correlating factor here is 'ethnicity'. 'Ethnicity' is the broad, generalist predictor of educational outcomes. This definitely not 'race' and it's not even 'IQ' (those things are plausible but controversial) - it's a series of behaviours, social norms, examples, attitudes towards work, success, access to services, social networks, mentors, role models, etc. etc. etc..
Educational outcomes (and crime stats, income stats) break down along ethnic lines. In a manner of speaking - America can be thought of as 'four nations' - White, Black, Asian and Latino. Obviously - it's very crude and generalist, and policy based on this would probably be racist - nevertheless - you pretty much have to look at the data given this.
In the end: American test score results have more to do with the changing ethnic composition of the American population than they do anything else. Again: White people and Asians in America have performed consistently he same for decades. Teaching methods haven't changed much, students habits haven't changed much - so the outcome is naturally consistent.
More economic prosperity, access to services and different attitudes + deeper integration have meant Latino A. and African A.'s are doing a little bit better - but because there are so many more Americans of those groups - particularly Latino Americans - it changes the outcome of the 'average american test score'.
Analyzing educational results does not make sense until you break it down along ethnic lines. Once you do - it becomes crystal clear. It's the absolute #1 most important thing about the educational data that turns 'paradox' about educational investment (teaching has remained largely the same) and outcomes into 'perfect sense'.
Unfortunately, it's so sensitive few will want to talk about it - for fear that the general public equates educational outcomes to 'intelligence' and try to strongly correlate ethnicity + race to this, which would be fodder for racist/KKK types, which wouldn't really help the overall social situation in America.
Anyhow - America is actually doing pretty well overall.