I read the NY Times because of the quality of the journalism. While the paper is left-leaning, the quality overcomes the bias. However the editorials are a completely different matter.

Today's editorial is regarding the NY exam schools (Stuyvesant et al), which admit students based upon performance of an admission exam. The study materials for the exam are widely available and free. However, the "mix" of the students is not what some groups would like. So do those groups try to get students to work harder so they can perform better on the exam? Nope. This being 21st century America, they instead decided to sue.

http://www.nytimes.com/2012/11/12/o...-the-federal-radar.html?ref=opinion&_r=0

The editorial mentions that the Department of Education decided to investigate the exam schools. And the NY Times editorial hopes that the investigation will force the exam schools to modify their admission criteria.

Does merit matter anymore in the US?

PS: Several years ago, I was part of a group that was invited to visit the New York Times. Jill Abramson gave us a lengthy presentation. Afterwards, the reaction from my group (business executives from several industries) was that she had an ivory tower mentality, out of touch with how the real world actually worked. Jill is now executive editor, and the New York times now unfortunately reflects her view.

It says


If such a complaint is well-founded, then sueing doesn't seem an unreasonable response. Have you looked at this "devastating analysis"?

What do you when measures of intellectual merit, such as the Scholastic Aptitude Test (now just the "SAT"), IQ tests, and the SHSAT (the test used for admission to NYC public schools) exhibit large group differences? I say accept the reality that there *are* large group differences and stop mandating equal results by group (as NCLB does). Most people want to avoid the question.

The summary of the complaint and the full complaint are available here:

http://www.naacpldf.org/press-relea...and-center-law-and-social-justice-medgar

http://www.naacpldf.org/files/case_issue/Specialized%20High%20Schools%20Complaint.pdf

The gist of the complaint seems to be that there is no "equality of outcome", and therefore the test must be invalid. I don't buy into the assertion that the difference is due to pricey test prep in richer households. I remember reading somewhere that over 40% of Stuyvesant students receive free or reduced priced lunches.

In other posts, I have mentioned that we have not accelerated our kids, in part because we are fortunate enough to live in a town with an outstanding public school system. We do not live in New York, but schools like Stuy offer the opportunity for highly intelligent children to learn together in a single school. And they do this without the cost of an expensive private school education, or having to live in expensive towns that are closely associated with most of the excellent public schools.

Apple. Fall. Tree.

I think there would be under-representation without test prep, for the reason I mentioned earlier, but the article

http://www.nytimes.com/2012/10/27/e...ons-test-highlights-a-racial-divide.html
For Asians, School Tests Are Vital Steppingstones
By KYLE SPENCER
New York Times
October 26, 2012

suggests that test prep priced at levels that immigrant Asians can afford does contribute to Asian over-representation at the NYC schools. I don't see hard work being rewarded as a problem.

Seems like this is another one of those very complex questions that's being reduced to black and white nuclear-option alternatives.

Some individuals are smarter than others. That's just how it is. But that doesn't mean that aggressive test prep doesn't give wealthier kids an edge. I've seen lots of stories about people who spend thousands of dollars over the course of years on prepping their kids for these exams. Where does that leave you if you can't pony up the cash? There is no way my family could have afforded that kind test prep for me.

IMO, any entrance exam to a public high school should be given on even ground. Outcomes will differ, but everyone should start from an equal footing. In other words, they should be using an exam that people can't prep for.

But more importantly, the real problem is that New York City needs more schools for gifted kids. As the system stands right now, it pits people against one another with predictable results: the parents of hyper-prepped, stressed out upper middle class teenagers are outcompeting the parents of bright low income kids who see themselves as having no recourse but to sue. Each side has its own nuclear option. What they really need is to designate more schools are being for gifted students. It's not like they'd have to build new ones. They'd just have to do a bit of shuffling.

Val,

I want to focus on one part of your fine post:


I imagine that would be quite difficult to do. Something that is easier and is likely to address part of the problem is to simply adjust for household income.

This is a bit off (my own!) topic, but I am currently reading Charles Murray's Real Education. What I have read so far doesn't address group differences, but it agrees with you that NCLB has been a disaster at both ends of the IQ distribution.

I think it's wrong to effectively punish children for their parents being productive. And think of the logistics -- should an application to a public high school really require something like the parents' tax returns? To truly estimate "privilege" from household income you would need the income history since birth.

Smart people earn more, and since intelligence is highly heritable, rich people have children who are on average smarter than poor kids. Any regression of test scores on income is picking up this effect as well as the ability of affluent parents to provide a better environment. How do you separate these two effects?

Murray has deliberately avoided the topic in his books after co-authoring The Bell Curve, so that people would consider his other ideas.

Difficult, yes, but possible if there's a will to do it. They'd have to start by offering a standard curriculum city-wide in the lower grades and providing schools or programs for gifted kids in each neighborhood, and they'd have to provide seats for every kid who scores in, say, the top two or three percent for that neighborhood. Just by guaranteeing seats, you ease some of the competition strain. You might also ease other strains, too.

Then they'd have to be very secretive about the contents of the exams. CTY takes this approach with its SCAT exam, and so do the people who make the Miller Analogies test. You can get enough information about each exam to understand the format of the test, but nothing else, really. They don't really release previous exams, and this cuts down A LOT on how much you can prep for these tests. I believe that it's generally agreed that you can't really prepare for an IQ test either (apart from learning tricks to remember number sequences?).

But again, I think what they really need to do is provide a seat for any child who qualifies. Not to mention that they also need to pay attention to the ones who were close, but didn't qualify.

The standard IQ tests such as the WISC and Stanford-Binet exhibit large group differences, and the NAACP lawsuit is prompted by large group differences in admissions to NYC exam schools. So your suggestion above to use g-loaded tests that are difficult to prepare for -- effectively IQ tests -- would not solve the "problem".

That was only part of my suggested solution. The other part was to establish schools or programs for gifted students in each neighborhood. This would also cure problems related to commuting long distances to get to one of a very few schools.

Regardless of whether or not you believe that IQ varies among different ethnic groups, it's irrefutable that there's always going to be a top 2% or 3% in any group. Kids in this group need their own instruction compared to the group they're in, regardless of the scores of some other group on the far side of a gigantic city.

The real problem in NYC is that a large percentage of the regular public schools are apparently completely terrible and are failing to actually teach children. This leave parents with only a few options - private school (most cannot afford), a charter school (way, way more students than spots) or try and get your kid into a gifted magnet.

Until the 'regular' school system starts doing it job, the 'problem' in NYC will continue.

Why is no one responding to this? It seems like an important criticism to me. I don't know what this test is like, but if what it's looking at is, for instance, skills that can be easily prepped for if you have the bucks, but that are quickly forgotten and don't predict anything meaningful...well, I'd say that's problematic. Wouldn't you?

Whatever can be said about the SAT, it's been shown to predict college success pretty well, IIRC.

This is true, but I don't know what it has to do with this particular issue.

You still should't be tossing the people with IQs (or whatever intelligence-learning thingy you use) of 120 into the same place as people with IQs of 150.

It's dumb to deal with ethnicities and races because they don't really exist as bright lines.

We need to be looking to individual families.

Again, Apple. Fall. Tree.

When I'm dealing with families with a solid history of mental retardation, I'm much more likely to take a case of their brother, cousin, etc., because I know that there's a good chance that I will get mental retardation out of *that* case without too much effort.

Not because of their race, but because of their family tree.

(And if not that, I can at least score some really low adaptive function. Winning! $$$$)

I had wonderful SAT scores and eventually gave up in college, resulting in C's, D's, and F's.

We really need to start giving scholarships to people who would actually benefit from them.

Some of this might have to do with coping and social skills deficits, which was apparently my problem.

Sample SHSAT exams appear in the Specialized High Schools Student Handbook http://schools.nyc.gov/NR/rdonlyres/1F19F679-EAAB-4372-A6C7-E89E951E8C6C/0/201213SHSHandbook.pdf . The SHSAT does not look so different from the SAT.

Also, we're talking about NYC, which is, as far as I can tell, a gordian knot.

Agree. To my mind, the part that might be devastating, if true, is "has not been shown to be a good predictor of student performance in high school". From a brief look at the complaint, it isn't clear whether the claim is that no evaluation has been done of the relationship between test scores and performance, or that they haven't given lower scorers a chance to show that they'd succeed at the school. The latter wouldn't be devastating in my view. The former would: I would expect the school to have collected statistics about the relationship between the scores *of those admitted* and their performance at school. If the test is an appropriate one, I would expect there to be a clear positive correlation at least towards the bottom of the scale of marks: those admitted with the lowest scores should tend to do a bit worse than those admitted with slightly higher scores, etc. (I put i this way because you might expect ceiling effects to obscure this at the top.) If this isn't the case, then indeed, it would seem reasonable to question the appropriateness of the entrance test.

But the interesting general issue is: this seems to basically be an achievement test, not an aptitude or IQ test. So, even if for the sake of argument we assume that aptitude is fixed at birth, better educated students are going to do better. "Better educated" is going to include going to better middle schools, having more supportive parents, and yes, having been tutored (whether "to the test" or "to enrich and improve the student's mind"). On average, that's going to lead to a skewed racial mix: in current circumstances, no way round that. People asking for a test that can't be tutored for are asking for (the impossible, and) admission by IQ test. But at some point, what makes for a coherent class that can all be taught at the same level isn't just aptitude, it's some combination of aptitude and achievement. So it wouldn't work just to replace this test by an IQ test (whose questions are miraculously kept secret so noone can prepare for them :-); that would weaken the school.

It's essentially the same problem that we have in the UK where the admission is to elite universities and the disadvantaged/advantaged groups are those who have been to state/private schools. There is pressure to "level the playing field" by discounting the better education that the private school students have (on average!) had up to that point, but this ignores the fact that it would have to cause either a dumbing-down of the universities' courses, or else mass failure of the students admitted who can't cope with them.

Wouldn't this force all affluent, upper middle class, and even middle class families to have their kids test prepped? If you're going to ding my kid's scores because you assume I must have test prepped her, I might as well test prep her to make-up for the lowering factor that was implemented.

-S.F.

I disagree, though I wasn't phrasing my ideas optimally when I wrote "can't be prepped for." I should have said, "Serious prep isn't a realistic option." As I mentioned, the Miller Analogies test and CTY's SCAT are relatively prep-resistant. This is because the test makers don't really release previous versions of the test. You can purchase access to three former Miller Analogies tests, but you can't save or print and they also don't tell you how the scoring system works. AFAIK, no SCAT tests are available. CTY just provides a page or two of practice questions. Likewise, the entrance exam for Thomas Jefferson is in this group (see this link). To the best of my knowledge, there was a pretty strong correlation between test scores and performance at Thomas Jefferson before they changed the admissions process. Today, 65% of the admissions decision is based on recommendations and writing samples. I don't know if they've studied correlations between test scores and other factors in their admissions process, but the remediation rate skyrocketed after they changed the process. But the number of minority students didn't change.

But the real problem is that focusing on the the admissions process ignores the root of the issue, which is that New York City 1) doesn't provide enough seats for gifted students and 2) NYC has some really crappy schools.

What large city with similar demographics does not? Bad students make bad schools. Public schools that are required to accept all students in a bad area and that are mandated to keep almost everyone in school will not be very good. I support charter schools and exam schools, but they operate under different rules, and there need to be schools to serve the rest of the population.

A thread "Are gifted education programs a waste of money?" http://giftedissues.davidsongifted..../all/Are_gifted_education_programs_.html discussed such research.

I see mithawk and Bostonian have got their polititroll hats on today.

At least Bostonian isn't being subtle about his social Darwinism this time. That bit about wealth being related to intelligence is HILARIOUS. It's almost as if he's never heard about gifted underachievement. If only there were a place on the internet where he could talk to experts on the subject, and individuals with real-world experience...

As for mithawk, I particularly enjoyed the bit about Stuyvesant being available to families who can't afford to live in an outrageously expensive city like, for example, New York City.

Enjoy your false reality.

From the Wikipedia article on "intelligence quotient" http://en.wikipedia.org/wiki/Intelligence_quotient :

"The American Psychological Association's 1995 report Intelligence: Knowns and Unknowns stated that IQ scores accounted for (explained variance) about quarter of the social status variance and one-sixth of the income variance. Statistical controls for parental SES eliminate about a quarter of this predictive power. Psychometric intelligence appears as only one of a great many factors that influence social outcomes."

One-sixth of income variance means that the correlation of IQ and income is about 0.4, since 0.4^2 = 0.16 which is about 1/6.
A 0.4 correlation is not negligible.

It doesn't equal 1, either.

Also, see correlation fallacy.

Finally, this: Psychometric intelligence appears as only one of a great many factors that influence social outcomes.

Why are we concerned about social/income outcomes here?

I can raise my income by 100% by moving to NYC or DC while simultaneously degrading my quality of life by 50%.

I'm not, but it seems to be important to Bostonian for some reason.

I can't speak for NYC except by reputation, but I have spent some time around DC, and based on my experience I'd say you're probably lowballing that 50%.

I was called away to Japan on a business trip and am just catching up. Normally I would have let this thread die except for Dude's comment:

Discrimination makes me upset. If that counts as being a polititroll, then sign me up for a leadership position. Why doesn't discrimination make you upset? Never mind, on second thought, I am really not interested in your smug answers.


The reality of NYC is that the majority of people are middle to lower class and some of them strive for good schools. The fact that the same income would be middle class in a different city is rather beside the point. First, people are resistant to moving, and second, the same job (e.g. laundromat owner, bus driver) might not be middle in another city due to lower pay.

Maybe next time, let the thread die by not posting it in the first place.

My commute is 10 minutes.

And that's in heavy traffic.

I can walk my kids to school (and normally do) and I also have no snow.

I grew up in PA and spent time in both NYC and DC growing up (in fact, I specifically avoided working in BigLaw there), so I know for a fact that I'm not lowballing it.

I really don't like cold weather.

I took a 40% pay cut to escape billing hours in a law firm (and to escape the snow).

I think we agree here, but something got lost in translation. When I say you're lowballing the 50%, I'm saying the quality of life degrades by more than 50%, it's actually worse than that.

Like you, I left a dense, expensive metro area (Southern CA) and took a 25% pay reduction to improve my family's quality of life in Hurricane Alley.

I be sure to consider your opinion for what I think it is worth.

Priceless.

I guess that I can't help but read that article with some inherent bias. I went to college (an ivy)with many students from Stuyvesant and Hunter, and many of them were middle and lower-middle class, including first-generation immigrants. To me, I always thought the exam school system was a great way to level the playing fields for the economically disadvantaged who couldn't afford the exorbitant private prep schools. Of course, the problem is really, as Val described, that there are not enough slots for all deserving students. By the time you get to these exclusive high schools, it really is less about your IQ, and more about the complete package of intelligence, achievement, discipline and dedication.