March 7, 2014

How is the New SAT not going to help Asians the most?

From Unsilenced Science
The main change in SAT results since scores were boosted by recentering in 1995 is that Asian average scores have been exploding (the upward sloping yellow line above). This has generated a lot of resentment of Tiger Mothers among upper middle class white moms, but these feelings have to be recoded into an attack on white male affluent privilege.

David Coleman, author of the Common Core, was hired by the College Board to redo the SAT. We'll have to wait to 2016 see what he has up his sleeve, but judging from all the ideology he has presented this week in setting the state for his changes, it sure sounds like, unless he's utterly Machiavellian, they're just going to increase Asian supremacy on the SAT.

From the NYT:
Then and Now, a Test That Aims to Neutralize Advantages of the Privileged 
By RICHARD PÉREZ-PEÑA    MARCH 6, 2014

When the College Board announced Wednesday that it was overhauling the SAT in ways that would curb the advantages enjoyed by affluent students, it sounded a bit like the people who first designed and popularized the test decades ago. 
The similarities end there. Across more than eight decades, the SAT’s backers have held it out as a yardstick, albeit an imperfect one, for academic merit, but notions of what defines merit have changed profoundly. 
The test began in the 1920s supposedly as a gauge of intelligence, but in recent years has moved toward measuring whether high school students have learned what they should. The latest changes give the SAT a hard shove further in that direction, making it more like its competitor, the ACT, in redefining merit as less about cleverness, and more about curriculum mastery. 
David Coleman, president of the College Board, says he wants to democratize higher education — lowering barriers to admission and helping more people go to college. That would not have sat well with the person most responsible for popularizing the SAT, James Bryant Conant, president of Harvard from 1933 to 1953. 
Conant saw the test as a tool for identifying the most talented people outside Harvard’s usual pool of privileged applicants. He disliked previous assessments tied to the teaching of exclusive New England prep schools, said Nicholas Lemann, author of “The Big Test: The Secret History of the American Meritocracy.” 
“He specifically didn’t want a test of mastery of the high school curriculum — he wanted a test to tell you how smart the person was,” Mr. Lemann said. “He was haunted by the idea of a brilliant student ending up walking behind a mule and a plow because nobody knew how to find him.” 

And that's a bad idea because ... And Harvard shot itself in the foot by taking the lead in emphasizing the SAT in admissions as can be seen by Harvard's now tiny endowment and the current worthlessness of the Harvard brand name?
But the goal was not democratic. Conant’s aim was to identify a new elite based on brains rather than heredity, not to expand access to higher education.

Uh, the only effective way to expand access to Harvard is for Harvard to have larger freshman classes, which is something that elite universities in general hate. For example, over the quarter-century from 1986 to 2011, while America's population grew by 75 million, Harvard's entering class grew from 1722 to 1726. Changing the SAT doesn't change those numbers.
In tune with the times, Mr. Coleman does want to improve access, in part by making a more level playing field. 
To counter test preparation courses taken by more affluent students, he announced a partnership with the online Khan Academy to make preparation videos available free. And he will make it easier for low-income students to take the test and apply to colleges without charge.

Everybody loves the Khan Academy and promoting the application waivers harder is a good thing, too.
It has been understood for decades that people who grow up in families with wealth, education, access to good schools or all three have a leg up in testing, a fact that has often been used to attack claims about innate ability or merit. That will remain true, but supporters hope that planned changes in the test will reduce those advantages by tying the SAT more closely to the material that any college-bound senior should have learned in a common core curriculum across the nation. 
“It elevates the importance of hard work on a day-to-day basis,” said William R. Fitzsimmons, the longtime dean of admissions at Harvard.

And how is this going to do anything other than make Tiger Cubs even bigger winners? As Education Realist points out, this kind of thinking was test run in the University of California system ever since Proposition 209 outlawed racial quotas:
So de-emphasize those evil, racist tests that traditionally represent, in the typical progressive’s mind, a means of reinforcing the institutionalized hegemony of the white man’s values. Grades, in contrast, reflected the school’s values, the school’s priorities. So majority URM [Under Represented Minority] schools, both charters and inner city, can put whatever grades they like on classes that can be called whatever they want. UC officials made the change, along with Eligibility in the Local Context, so that majority URM schools could lie about their students’ academic abilities properly reflect the students’ diligence and abilities in subjects simply not valued by the institutional racists at the College Board. 
The problem is, alas, that UC admissions made changes to their policy based on the “demographic footprint” of tests, but they forgot about the demographic footprint of grades.
Namely: Asians, particularly recent immigrant Asians, kill whites on grades. The test score advantage is getting (suspiciously) worse, but the grade advantage is huge. 
That wasn’t part of the plan. Look, universities know the game as well as anyone: grades are a fraud. That’s why, until relatively recently, all universities weighted test scores as high or higher than grades.
     

42 comments:

Portlander said...

I think it's pretty clear they are dumbing down the test which will result in more compression on the top-end, a smaller racial gap, and thus more heuristic (read arbitrary) admissions decisions to get the racial make-up progressives in the colleges and universities desire.

It probably forces more whites into expensive, small liberal arts institutions to escape the teaming, unwashed affirmative action masses in the politically correct public universities. Our universities, like our fire and police dept's will reflect the demographic make-up of the broader society.

More debt, more indoctrination, fewer years in the labor force, it's win-win-win for the progressive framework.

Obviously, ORM's (over-represented majorities) will be the big losers. Back to their Gameboys and Playstations for the middle 66% of whites. Lousy upward mobility for the 12% at +1 to +2 std deviation. More lottery style rewards for the 2% at +2 std deviation playing the tech & finance rent-seeking game. Great country we live in.

Sean said...

There does not seem to be any problem with keeping Asians out of Harvard. It's already blatant.

vinteuil said...

"how is this going to do anything other than make Tiger Cubs even bigger winners?"

Well, indeed.

I really wonder about the David Coleman's of the world: totally clueless? or, as you put it, utterly Machiavellian?

On the one hand, it seems hard to believe that they could be so totally clueless that they can't see that "tiger cubs" will be the big winners here.

On the other hand, I just can't figure out what utterly Machiavellian scheme might be afoot.

TC said...

How would the Sailer/Murray SAT II solution work? Would a prospective liberal arts major have to take a content test in a quantitative field? Could you take unlimited tests, or just a set number? (Obviously, unlimited tests + cumulative admissions value per test = even more Asian pwnage.)

roundeye said...

So if grades are more important, how do we keep the Gulenist kids out of Harvard?

dearieme said...

“It elevates the importance of hard work on a day-to-day basis": why?

I did well at school. I did spells of sustained, intense, hard, academic work, spells reading whatever I fancied, spells running around on cricket and rugby pitches, spells building boats or earning money at the harbour, spells pondering and reflecting, spells watching Intellectual Continental Films, and spells idling or cycling in the countryside.Then I went to university and did very well indeed. The way I had chosen to behave at school turned out to be an excellent preparation for university.

"Grinding" would be a particularly bad way to spend your childhood if (i) you are clever, (ii) most of your classmates aren't, (iii) your teachers adjust their teaching to suit the dimmer brethren.

Hell you'll end up mastering fluff and rubbish. What a waste of youth. You want to improve schooling? Teach the bright children together and let the teachers respond to the class's ability. Do it again for the next quality down, and so on, so that the youngsters learn from each other and are faced with teaching at a pace, and a level of abstraction, that they can thrive on. Stop pretending that every child, if only he worked hard enough, can go to a high-calibre university and do well. Give the poor buggers a chance to do well at something within their competence. Do not try to turn everyone into a pale simulacrum of Harvard Man. It's a batty idea, essentially snobbish and deeply intellectually frivolous. We differ. Cope with it; use it for the better.

Luke Lea said...

I'm with dearime. Have a good time while you are young and the truth will out. Hell, I didn't even begin reading books until my second year of college. I was sort of a math and science jock though, mostly on my own.

nooffensebut said...

“The main change in SAT results since scores were boosted by recentering in 1995 is that Asian average scores have been exploding (the upward sloping yellow line above).”

I think Asian SAT improvement relative to whites goes back to about 1987. It does appear to be accelerating since 1995, especially on the math subtest, but I suspect that East Asians are experiencing a large, global Flynn effect. There is evidence of this in China. The recentering was supposed to decrease Asian and white advantage.

Perez-Pena said:
“It has been understood for decades that people who grow up in families with wealth, education, access to good schools or all three have a leg up in testing, a fact that has often been used to attack claims about innate ability or merit.”

If you follow that link to the Unsilenced Science, you will see that family income does not have a significant independent effect on SAT score. Something more on this is in the works. Race confounds the effect of school quality.

Education Realist said:
“The test score advantage is getting (suspiciously) worse, but the grade advantage is huge.”

I haven’t read anything by Education Realist that would lead me to believe he is something other than a crank. Without grades or test scores, what can we rely on to prove whites are the master race? Do we have to go back to an invention-list pissing contest?

Portlander said:
“I think it's pretty clear they are dumbing down the test which will result in more compression on the top-end, a smaller racial gap, and thus more heuristic (read arbitrary) admissions decisions to get the racial make-up progressives in the colleges and universities desire.”

Theoretically, the interactions of test changes and test score minimums and maximums should be to lower the Asian math advantage and increase the African-American verbal disadvantage because large numbers of Asians hit the math maximum and large numbers of African Americans hit the verbal minimum. However, this should have happened with the recentering for African Americans, and not only did it not, but the African-American verbal standard deviation didn’t even change much. I also doubt that the College Board would tolerate a dumbing down of their test, which would decrease its usefulness, especially with their base, the coastal elites. They are probably mostly doing this in response to ACT competition. Some of the changes sound most harmful to women. It bothers me that emphasizing canonical knowledge will punish nonconformists, but some knowledge, especially scientific knowledge deserves to be widely known. Even so, there is something to be said for specialization and diversity of perspective. If they can place more emphasis on practical knowledge, which is still vast, and maintain a bell curve, which I’m sure they want to do, they can achieve a sensible balance. This, like intelligent populism, is not contradictory. If they do nothing, they will decrease emphasis on the critical-reading subtest. Ever since the mid-1960s, critical-reading/verbal scores have been falling relative to math scores. The recentering erased that gap, and the gap immediately started growing again. Maybe the changes will make the verbal knowledge more resistant to time.

2Degrees said...

I accept HBD and my place within it, but the steady rise in Asian scores is suspicious. More information please.

Intelligence is largely innate, so differentials between races should remain fairly constant.

anony-mouse said...

'...Asian average scores have been exploding (the upward sloping yellow line above)...'

Its a good thing you run your own blog Steve. If it were someone else these days...

countenance said...

I took the SAT in June 1994, right after the end of my junior year in HS, the final SAT testing date before the late 1994 recentering. I got a 1350.

During my senior year, my HS guidance counselor showed me a conversion chart between the previous and "recentered" scoring formulas, based on the raw scores (i.e. total number of correct questions) on both the verbal and math parts of the test. If I would have waited one sitting, my 1350 would have magically become a 1440.

"But the college admissions boards all know about this recentering, so they'll know your 1350 then is a 1440 now, so you didn't lose anything by taking the test when you did," the guidance counselor told me.

"Okay, then, why bother recentering at all?" I asked him.

He feigned cluelessness, but the way we looked at each other, we both knew we knew the answer but we also knew that what we knew was too taboo to say openly.

I looked at the conversion chart, and I saw that in most instances, it meant increased scores, especially in the middle-upper ranges of number of questions answered correctly. Therefore, that the '94 recentering benefited Asians the most doesn't surprise me, because relatively above average IQ Asians can tiger cram their way into middle-upper scores.

Anonymous said...

Asian average scores have been exploding (the upward sloping yellow line above)

'Sloping', 'yellow'? Has anyone alerted the Federal Microaggression Taskforce yet?

JR said...

Dearieme is just describing how it actually was in the 50's and 60's
I know. I was there. And I still ended up with a Ph.D.

Education Realist said...

Crank? Good lord. I'm cranky. It's not the same thing. Besides, any idiot talking about whites being the master race has no business calling me a crank.

You're wrong, by the way, assuming the numbers in that link are accurate (and I provide cites). The acceleration of math scores began after the 2005 changes, because the math test got much easier and more amenable to coaching, which East Asians do 3x more than whites. The verbal scores, not the math scores, have seen a tremendous improvement since 1996, which is one thing I'm suspicious of. Both Asian and white verbal scores in the top tier have been increasing dramatically since the 2005 changes.

That said, standardized tests are still the best game in town. Grades are a fraud. Standardized tests are just being gamed by one population.

Generally, not in response to numbers freak:

The obvious interpretation is that the College Board wants to be more like the ACT.

But I can't figure out if they haven't thought it through, or if there's some element I'm unaware of or haven't considered.

1) 3x as many Asians take the SAT as the ACT--and I'm pretty sure that's in America alone. ACT's overseas market is minimal, while the SAT is huge in Asian markets. As I mention in that piece, Asians appear to be less dominant in the ACT, or at least aren't getting scores of the same percentile level--but I'm not enough of a numbers geek to be sure of that.

2) Asians are not.happy. with the SAT changes, last I checked.

3) Cheating is a huge problem with the SAT in Asian countries overseas, and a non-trivial problem here.

Multiple interpretations. I'm going to take the interesting one, the one that snags at me, but of course there are other possibilities:

In my opinion, a good number of Asian scores here and overseas are achieved by cheating (more, as a percentage, than any other race). Some additional percentage of dominance is achieved by number crunching For example, test prep companies obtain copies of the tests, do predictive analysis on the vocab used and when it will appear again, and sell a list of vocab words that have a high probability of showing up on the test. From what I can gather, the accuracy is not perfect, but considerably accurate some times (and completely inaccurate others). But the kids memorize, say, 200 words with a high probability of appearing on the test.

The ACT doesn't test vocabulary. It is reading intensive, which makes it very difficult to game for non-English speakers.

So if the SAT is going to be like the ACT, then the SAT is abandoning the Asian market, which is about the only area it's still dominant. How likely does that sound?

So it comes down to this: can *any* test be gamed? Remember, the Asians had to resort to cheating on the GRE---they couldn't do it by study alone. Can the ACT be gamed as thoroughly as the SAT has since they started making it easier?

Possibilities:

1. I'm wrong, the SAT can kiss off the Asian market.

2. I'm wrong, because the Asians will now just figure out how to game the ACT as well as the SAT.

3. I'm right, because the CB has been taken over by reformers utterly ignorant of the realities of the test prep market. Not just Coleman, but also Porter Magee and others actually believe the swill they spew.

The other thing to remember is that the SAT's customers are universities, not testers. It's kind of the reverse of the insurance industry. Hard to know if Coleman understands that--he doesn't seem to. But it's also possible that universities have been unhappy with the most recent changes. They can't be happy with the Asian overperformance, and not just from an admissions standpoint. They want Asian money, but they don't like being fooled. So maybe universities are pressuring for a test that can't be gamed.

I dunno. It's very interesting.

Anonymous said...

Because universities are the true consumers of these tests you need to think of their incentives. The SAT and ACT are already useless for distinguishing gradations of elite performance. When Harvard or Stanford want to admit someone just because they're super smart they look at things like USAMO and performance on international math olympiads, knowing full well that the gap between a 760 and 800 may be down to random chance for many. In contrast having a test that roughly ranks mid to upper students but blurs things a bit allows them to tailor their class (i.e. keep out Asians) while claiming that all their admits are "well qualified" -- a favorite newspeak term of the Ivies.

nooffensebut said...

“The acceleration of math scores began after the 2005 changes, because the math test got much easier and more amenable to coaching”

The Asian progress relative to whites grew faster after 2005 than from 1996 to 2005 on the math and critical reading tests, but Asian math scores increased faster from 1996 to 2005 than since 2005. My point was that Asians started gaining relative to whites on these tests in about 1987 with faster gains after 1995 than before.

“Asians appear to be less dominant in the ACT”

Asian composite ACT scores are 5.9% higher than those of whites. Asian composite SAT scores (M+CR+W) are 4.3% higher than those of whites.

Anonymous said...

We're witnessing a 'cut' in the distribution of Asian immigrants -- particularly the Chinese.

This is causing the Asians that make it over to America to experience an IQ shift MORE dramatic than the Ashkenazi Jewry that fled the Pale of Russia a century ago.

A similar shift has occurred in African populations that migrated. They are also shifted about one standard deviation above that of their homeland.

This has to be expected for any population migrating across the Atlantic under harsh conditions. Both the Africans and the Irish had horrific blood losses crossing over. And both suffered from slavery/serfdom across both sides of the oceans.

In the case of the mainland Chinese now arriving: politics intervenes. Beijing simply does not permit the dummies to flee.

Something like that happened a century ago with the Japanese. The Emperor flatly prohibited unmarried citizens the right to emigrate. Further, they had to be young, from the 'right places' -- and farmers -- only. (No samurai, no fishermen, no tradesmen, etc.) A staggering fraction came from Okinawa -- with the remainder coming from Kyushu and the Inland Sea. None came from Honshu.

Not surprisingly, Japanese Americans refer to the natives as Buddha heads. It's not a term of endearment. The DNA shift is strong enough that they can spot each other from one-hundred meters away.

In the not-so-long run, the premier status of Ashkenazi Jewry in American cultural life will not be able to hold against quite that many ultra-smart, ultra-hard working Chinese -- all of whom are imbued with cultural supremacist attitudes of the first magnitude. (Just ask Tokyo, Hanoi or Seoul.)

This culture-kampf will largely occur at least 40-years into the future.

&&&

It is universally true that emigrating sub-populations have a positive up-shift in IQ and other health metrics. In every case, this is due to the fact that the true dummies -- in every distribution -- are left back in the old home stomping grounds.

This shift is the PRIMARY reason why America of the 20th Century pulled away from the rest of the planet. (exception: Canada)

It applies to every ethnicity.

Of late, the separation is much reduced, as mass immigration sucks in even the less talented souls. We are now witnessing some reflux as these players are discovering that America is not a 'fit' for them. Just when no-one is looking, the economics are driving Mexicans back to the old homestead.

Because all of the above is taboo, you will not find it discussed in professional journals -- nor research along these lines being advanced.

Consequently, our academics are as politicized as any Commie or Nazi. No-one dares go off the reservation.

Anonymous said...

Let me go meta for a moment. Companies who need bridges built will still look to Caltech and the like where these nuances are moot.

If all the 'knowledge workers', or whatever they call bright young things these days, in Manhattan are super- improved-minority-hires, then: The system will self correct to find the right talent; or, the system will fail. The uncorrected system, (that of placing politically-correct boobies in key positions), is bound to collapse under what the Marxists called its own contradictions.

I guess eventually is a long time, but it can be shorter than you think. Especially in the the Internet Age. That brilliant reject fron the hinterland starts a blog. The ivy bots in the MSN keep doing what they do. The people vote with their feet.

Gilbert P.

Goat cheese gas said...

"David Coleman, president of the College Board, says he wants to democratize higher education — lowering barriers to admission and helping more people go to college. "

-Sure. The strategy of opening things up to everyone worked great in housing in the US during the 90s and 00s, what could possibly go wrong?

Seriously though, I have found some of these newer crops of college geniuses to be so dim, they have to get a calculator to figure out how much two 50c candy bars totals up to.

The way things are set up at many universities you have curves in classses, so if the class average on a test is 47 then scoring a 70 on the test, which in years past would be barely passing, will get you an A these days. A graduate student TA that I know gripes routinely about the students in his class- the average on the first test this semester was a 38, and the second test was a 56.

Its true what they say about college becoming what HS used to be, and everyone now having a college degree. Well, that's great CYA for the big lie of the age, pretending that everyone is equal, but in reality, University degrees used to have value, showing that you were knowlegeable about a subject and that you were reasonably bright and hard working. It was a benefit for you,in getting a job, and for the employer, to more easily discern who would be a good fit for the job.

Now that the degrees have been dumbed down, they lose their value (which ironically has become more expensive) and its harder for employers to separate the bright lights from the dim bulbs.

Its an obscene waste- college tuition inflation wastes a lot of money for the student and the society, and they both get something that is worth a lot less at the end.

Its an absolutely stupid thing to do to what had been a very beneficial thing for the country and its citizens, all for belief in and covering for, the big lie. That and fattening the pockets of college administrators, and creating intelligent sounding jobs for black academians "diversity officer", "dean of outreach".

It also cheapens and destroys the brand of places like Harvard, and US schools in general. US schools, particularly grad schools, have historically been very highly viewed in the world, but as they become dens of mediocrity, that goes away, destroying something of value for the US.

Anonymous said...

"Grades are a fraud".

That's unfortunate. My son's just been named in the Dean's List for his first term at a US university, and I was delighted - then I saw how many other students were on the list. Any idea what proportion of students get listed ? We don't have that in the UK.

Anonymous said...

1. They're just going to fuck up the system so the admissions officers can pick whoever they want.

2. Educationrealist - don't underestimate the ability of Asians to figure out how to cheat on tests. They have a long history of having 2 main life options: 1. be dirt poor 2. pass a test (by any means necessary) and be rich.



Education Realist said...

"Asian composite ACT scores are 5.9% higher than those of whites. Asian composite SAT scores (M+CR+W) are 4.3% higher than those of whites."

I did say "or aren't getting scores at the same percentile level"--that is, not as many high scores. However, I'm not even sure of that; it's the sort of thing you do, not me.

It's at the high end that the Chinese, Koreans, and Indians are playing. So composite or combined scores are useless, as are averages, as they include Vietnamese, Filipinos, Thais, Cambodians, Tongans, and other Asians. Besides, since 3x as many Asians take the SAT as the ACT, overall comparisons of averages don't make sense.

The way to figure out what's going on with the Chinese, Koreans, and Indians is to look at the top percentiles.

My point was that fewer Asians seem to be getting 32 or higher as get 700 or higher.

"don't underestimate the ability of Asians to figure out how to cheat on tests"

I'm not. Cheating is different from gaming. My suspicion is that the SAT can be more easily gamed than the ACT, that is a bright person can memorize certain responses and patterns and do well without the requisite underlying knowledge. The Chinese couldn't manage this with the GRE, which is why cheating was so rampant. So I'm thinking, possibly, that the ACT is harder to game, making cheating the only option.

Incidentally, I am not arguing that the Chinese/Koreans/Indians etc don't have "high IQs". They do. Lots and lots of kids can't achieve the level of memorization needed. It's more that many Asians with "high IQs" aren't smart in the way that whites think of smart. I am talking here about my direct observation, as well as many other tales of direct observation from colleges, as underperforming and cheating Asians is a huge problem.

Likewise, even though my sample size with blacks is much smaller, I have routinely run into blacks that have demonstrated abilities, abilities I personally see, that have trouble translating that to higher test scores.

None of this denies the statistical validity of IQ as a predictor. And of course, there are whites that profile just as both the blacks and Asians I describe. Normally, it would be a fringe issue. What's happening with Asians and their use of our colleges---their near destruction of our college access, with the help of the colleges themselves---makes it a more pressing issue to figure out.

Best answer I can think of is tests you can't prepare for--can't schedule, just called in one day hey, it's test time. And no, we won't tell you what subject until you open the test. That's a lot of work and money, though.

Anonymous said...

The First Law of Discrimodynamics states that valuable discrimination cannot be destroyed. Thus, to the extent that the SAT has lost its power as a separation function; it has been replaced by an alternative function, or the value of discrimination has declined.

Neil Templeton

Anonymous said...

In the not-so-long run, the premier status of Ashkenazi Jewry in American cultural life will not be able to hold against quite that many ultra-smart, ultra-hard working Chinese -- all of whom are imbued with cultural supremacist attitudes of the first magnitude. (Just ask Tokyo, Hanoi or Seoul.)

Just remember you evil white supremacists, this is none of your business, so don't go thinking it is.

Anonymous said...

This is causing the Asians that make it over to America to experience an IQ shift MORE dramatic than the Ashkenazi Jewry that fled the Pale of Russia a century ago.

The Ashkenazi Jews who emigrated to the U.S. between 1880 and 1920 came mostly from the lower strata of Ashkenazi society. The shtetl machers were more likely to stay put, and the high elite Ashkenazi families were barely represented among the immigrants until the 1930's. Also, Ashkenazi immigrants represented a significant fraction of the total Ashkenazi population, which isn't the case for Chinese and Indians immigrants relative to the billions of their co-nationals at home.

virgil xenophon said...

As a fossil age 69 who took his SATs in spring, 1962, I read such things as these proceedings with amazement. "Back in the day" as it were there WERE NO Kaplans, etc., or they were just in their infancy. In fact it was a badge of honor that one took only the PSAT and the test cold with just the information one had absorbed--only the "stupids" needed review/cram courses, etc. And no retakes to raise scores, either, a single shot being all that was thought to be needed. I guess my 1600 and Nat Merit Scholarship under the old system should garner me two scholarships today, right?

Of course I did have the advantage of growing up on a small mid-western college campus as the son of two professors and going to a Univ Lab school.

(BTW, I never took the ACT; taking SATs was quite prevalent throughout the Mid-West in my day, iirc, as it was the Southern schools that mainly relied on the ACT.)

Tony said...

Good grief it's not like hard work and aptitude are mutually exclusive. You don't say that Wayne Gretzky shot (no pun intended) to stardom just because he worked hard, which he did. He also obviously had talent, just like scores of lesser-known but excellent players in the NHL. Same for those who burn the midnight oil studying in high school and prepping for exams. I doubt you boost your score from 1000 to 1500 M+V; for most people in the very high range, they would've had great scores without the prep.

Tony said...

It's kind of funny to see those on the college message boards (and a fair number in the mass media) vilify so-called 'Tiger Moms'. Many if not most of these complainers, I suspect, send their kids to top suburban or private schools, which bestow on their kids an advantage, in terms of superior academics, counseling, and connections with college admissions officers that put the best test prep to shame. At any rate, those who can pay private tuition or afford to live in the wealthier school districts are the ones who can afford to pay for fancy test prep services. Even then, their kids can get 140 pts lower than a comparable Asian kid and be equally competitive at most Ivies according to a Princeton study.

Tony said...

What does it matter if the SAT helps Asians the most? Colleges will simply move up the cutoff score for Asians. A Princeton study a few years back showed that an Asian student had to get 140 pts higher, controlling for other factors, than a white student at that school to be equally competitive. Now, that figure is for Princeton, but given what Unz has showed about Asian proportions at Ivies staying unchanged over 2 decades despite their numbers going up, other Ivies probably use similar differentials.

Anonymous said...

@Tony

"At any rate, those who can pay private tuition or afford to live in the wealthier school districts are the ones who can afford to pay for fancy test prep services."

Not always. Some of those people, especially those who send their kids to (e.g.) Catholic schools, may be scrimping everything they can just to do that.

And even in most of the wealthier school districts, a percentage of the population will be below the poverty level, and the population will range across the wealth distribution. At least outside of the extremely gentrified districts on the coasts.

sunbeam said...

An Anonymous wrote:

"Something like that happened a century ago with the Japanese. The Emperor flatly prohibited unmarried citizens the right to emigrate. Further, they had to be young, from the 'right places' -- and farmers -- only. (No samurai, no fishermen, no tradesmen, etc.) A staggering fraction came from Okinawa -- with the remainder coming from Kyushu and the Inland Sea. None came from Honshu.

Not surprisingly, Japanese Americans refer to the natives as Buddha heads. It's not a term of endearment. The DNA shift is strong enough that they can spot each other from one-hundred meters away."

Okay, I'll bite what exactly are you trying to say there? That Americans of Japanese ancestry that have been in America for generations (with no admixture of non-Japanese genes) are physically different from "normal" Japanese?

Not sure what you are saying here.

"In the not-so-long run, the premier status of Ashkenazi Jewry in American cultural life will not be able to hold against quite that many ultra-smart, ultra-hard working Chinese -- all of whom are imbued with cultural supremacist attitudes of the first magnitude. (Just ask Tokyo, Hanoi or Seoul.) "

I kind of think Ashekenazi Jews are going to lose a lot of niches to Asians, particularly Chinese over the next few years myself.

Ashkenazi may be smarter, but they don't seem to work as hard at academics as Asians. And the academic route is the way Ashkenazi have been rolling since WWII. I mean when is the last time you met a Jewish tradesman, mechanic, or anyone who works with their hands?

As far as "cultural supremacy" goes, my observation is that Asians assimilate pretty quickly into American culture. They seem to have a pretty strong materialist bent, and well this is the inferno for something like that.

Of course past trends are no guarantee of future results. It's not hard to imagine Chinese Americans, for example, leveraging Chinese ancestry to profit from Chinese preferring to deal with them, as well as connections they may have in the old country.

jeremiahjohnbalay said...

Am I the first to point out that the goal of testing everybody for the same learned knowledge sounds a bit orwellian?

Anonymous said...

"Okay, I'll bite what exactly are you trying to say there? That Americans of Japanese ancestry that have been in America for generations (with no admixture of non-Japanese genes) are physically different from "normal" Japanese?"

The islands of Japan have broken up the DNA chain in exactly the same way that the Hawaiian Islands have broken up the Polynesians and the genetic variations that Darwin spotted in the Galápagos Islands.

And so it is that Hawaiians can easily spot which island a fellow came from. Those from Molokai really stand out. They hail from a single family line going back centuries, if you believe the DNA.

Likewise, the Japanese -- to this very day -- can spot Okinawans -- discriminated against, BTW. This bloodline, trivial in the Home Islands constitutes over 25% of the DNA on Oahu.

The Japanese from the Inland Sea also have quite a shift. Some of their kin fled an eruption about 1,000 years ago and transited the Pacific to land in Ecuador. (!) While flamingly Japanese in blood, they speak both a local dialect and, for commercial purposes, 'Ecuadorian.'

(None of the Spanish American ex-colonies speaks Castilian Spanish. Our visiting Spanish exchange students found that it took a full ninety days (of soap operas) to learn the Western lingo. The shift is MUCH more profound than found anywhere in the Anglo-sphere. The shift goes way beyond just slang.)

And those Japanese Ecuadorians look EXACTLY like my old business partner. All of them. You have a 'founder population' and refluxed DNA.

Consequently, when a native Japanese businessman drops in from Tokyo (Honshu) he stands out. This facial difference is totally lost on most non-Japanese.

You can take it to the bank that the Honshu guy feels superior to the yokels. You could cut the arrogance with a knife.

Consider yourself blessed if you've never witnessed it.

Naturally, they are resented, and have picked up the sobriquet: Buddha head.

The Japanese in the Islands don't have a high opinion of Japanese Americans born and raised on the mainland -- anywhere. Indeed, there is more than a little concern if their own sons and daughters spend too much college time on the mainland. (Cultural contamination)

This tick goes triple back in Tokyo. If any executive spends more than ninety-days overseas, particularly in America, he's at severe risk for his career. I know of no Japanese CEO who has that stain on his resume.

This tick means that corporate Japan has a brutally difficult time getting talent to leave Tokyo. You can read the literature on the topic. Much has been written.

As for the IQ up-shift. Yes, it's there. Honshu Japanese are constantly surprised at how clever the Oahu Japanese Americans are. This goes double for Okinawan Japanese Americans. It is normally the case that they just don't do business with them. Outsiders are normally oblivious to any and all of this.

Something like this is only seen in Europeans that used to be 'potted' like plants in their old homesteads. Sicilians come immediately to mind. Like the Japanese, Sicilians just did not DARE travel far from their villages. No jobs were to be had, no girls, just plenty of trouble. The result is that they had to migrate to America to have any chance at all.

As with all other ethnic migrants, those that made the crossing were smarter than the average Sicilian. The 'hicks from the sticks' became overnight successes in the New World.

The IQ up shift is also seen in the Irish. Many studies place the native Irish at 95 ish. American Irish norm towards 102 ish.

German Americans -- almost to a man -- came from the poverty-belt of Germany: the Black Forest in southwest Germany. They were -- like the southern Japanese (Okinawans), and the southern Italians (Sicilians) considered to be the 'hicks in the sticks' -- and dull, plodding beasts -- draft humans, if you will.

The IQ shift must be profound, for German Americans consistently test 100-103.

BurplesonAFB said...

>"The way things are set up at many universities you have curves in classses, so if the class average on a test is 47 then scoring a 70 on the test, which in years past would be barely passing, will get you an A these days. A graduate student TA that I know gripes routinely about the students in his class- the average on the first test this semester was a 38, and the second test was a 56."

It's a good system when admissions isn't going for the biggest possible tent.

Ask questions hard enough that you have to have a real solid grip of the subject to get it right. 45% averages on tests allow top performers to really distinguish themselves. The high school way of giving everyone an 85 and making the entire upper half compete to make 0 mistakes on the easy questions is a bad way to test.

Veracitor said...

Fascinating, Captain: Nat'l Assoc. of Scholars leader Peter Wood explains how Coleman's new SAT is intended to validate and complement his Common Core standards and further the dumbing-down of American secondary and tertiary education.

Anonymous said...

I suspect the improvement is Asian SAT scores has to do with their changing demographics. Don't we have more Indians and Chinese students now?

For all the focus on Chinese students, it's really the Indians who learn English as their first language, who are really dominating academically. The Chinese students struggle with English, since they are years (up to a decade for some) behind and have to catch up. The Indian kids may not have the reputation of math geniuses, but they're no slouches, and they do dominate verbal achievements. Just watch the Spelling Bee.

There isn't an equivalent public figure and stereotype a la Amy Chua, for Indian-Americans. But look around Silicon Valley and the politicians and CEOs, they're Indian-Americans who are fluent in English.

Anonymous said...

If David Coleman is ex-McKinsey, he should just boil the SAT down to one, 60-minute test....it is what McKinsey does in the hiring process for (BA/BS or non-trad grad-degree hires...I believe MBA's are not included).

The PST is very well designed and is compressed enough in regards to time to measure 'g' very well. Hell just take the PST verbatim and make that the SAT.

The CIA DI-test is slightly longer (90 minute multiple choice and 90 minute written analysis) that is likewise effective.

TC said...

The undemocratic Conant believed that inheritance taxes should be 100%.

M said...

unsilencedsilence I think Asian SAT improvement relative to whites goes back to about 1987. It does appear to be accelerating since 1995, especially on the math subtest, but I suspect that East Asians are experiencing a large, global Flynn effect. There is evidence of this in China. The recentering was supposed to decrease Asian and white advantage.

Interestingly, as I am aware a longitudinal meta-analysis on Taiwan and the Flynn Effect show that they seem to have stopped showing gains before Northwestern European countries do.

http://jntnu.ntnu.edu.tw/pub/PaperContent.aspx?cid=150&ItemId=1298&loc=en
Increased IQ for Taiwanese Children from 1997 to 2007: Flynn Effect Investigated

"Increased IQs over the intervening 10.5 years were 2.5 points, yielding an estimated growth of 0.24 IQ points per year. While America and other developing countries were still reporting an increased rate of 0.30 IQ points per year, increased IQ for Taiwanese children seemed slower than what the Flynn effect would have indicated. Current findings have matched recent reports from Scandinavian nations that the increase in IQ is slowing."

Likewise China - http://lesacreduprintemps19.files.wordpress.com/2013/08/china-flynn-effect.pdf

"Increases (for China) are of similar magnitude to those of approximately 2.18 IQ points a decade for the United States for the years 1948–2002 calculated by Flynn (2012)."

Not sure what the evidence is for a "super dooper Flynn" effect for East Asians. Their education improvements continue, but that's not IQ. It correlates to IQ, but...

It seems an assumption based on the economic backwardness of many Asian nations and their current growth.

Like others ITCT, I think the disproportionate Asian American SAT growth effect primarily represents changes in composition of the Asian-American population, to a more selected stream of migrants.

Ryan said...

Since many people seem to be asking, I think educationrealist's take on the recent rise in Asian SAT scores is that the population has reached a critical mass which enables effective cheating.

Anonymous said...

"Asian composite ACT scores are 5.9% higher than those of whites. Asian composite SAT scores (M+CR+W) are 4.3% higher than those of whites."

This is a mathematically nonsensical argument because the tests have different scales.

On the ACT, a sd is about 23% of the mean while on the SAT the sd is about 10% of the mean. Thus on the ACT the asian mean is .25 sd higher than whites while on the SAT it is .45 sd higher. They appear to have almost twice as much relative advantage on SAT. And this is with inclusion of the SAT essay portion of the SAT, where the asian edge is probably smaller...

Veracitor said...

Coleman's plan to get rid of all but the least interesting math questions will greatly reduce the SAT's predictive validity (i.e., ability to predict college performance and therefore the SAT's usefulness to admissions officers), and Coleman knows that because College Board statisticians know it and have explained it to all and sundry.