Something must be wrong

Paterson, N.J., a city of about 150,000 people, had exactly 19 high school students this year who managed to break a 1500 score on the 2400-scale SAT (that's a score of 1000 on the old 1600 scale, for us codgers).  The teacher-student ratio in the Paterson school district is about 13 to 1; annual spending per student is about $17,000.  Alertly discovering that something was badly wrong, school district officials explained that the local school system has been plagued by numerous restructurings in recent years, leading to confusion and dysfunction.  As a result, they've decided to take the obvious step and quit using the SAT to gauge performance.


Ymar Sakar said...

I was thinking 1500 was pretty good... until you mentioned the scale.

Grim said...

That is the obvious step, isn't it?

Cass said...

Wow. Just wow.

I will say that it's not just the schools. I don't buy that it's the school's responsibility (primarily, at least) to make sure students learn.

It's the parents' job first and foremost. We made sure our boys could read and write correctly and well, and we oversaw their curriculum every year and monitored their grades. If we hadn't done that, I don't think they would have done as well as they did.

I don't really understand delegating that responsibility to a school. Having tutored students at the college level, I'm very aware that even the best teachers and books don't do much good unless the student puts in the necessary time and effort.

And with reading in particular, you have to read challenging books outside of school to have the kind of vocabulary you need to do well on the SAT.

Gringo said...

And with reading in particular, you have to read challenging books outside of school to have the kind of vocabulary you need to do well on the SAT.

How many of those Paterson students read even ONE book out of school in their four years of high school? Not many, I wager.

Not long ago, there was a profile of a point guard who led his school to the NCAA basketball championship. As a measure of how far he had come, the article mentioned that for the first time in his life, he had recently completed reading a book cover to cover. That this happened to be a book on sports is fine- read what you are interested in.

Though in sympathy to this "student"/athlete, for years he had a very busy schedule involving sports. Big time school athletics can take up a lot of time. I have read estimates of up to 30-40 hours, an estimate which doesn't include classes and homework. Just the sport.

Anyone involved in a big time college sport who takes and completes a "regular" student curriculum, one devoid of "sports studies" and the like, merits a great deal of respect, considering the time his sport involves.

Grim said...

My cousin who is dyslexic got all the way through college without reading a book. She's really smart and works hard, of course, but it was possible to do it.

J Melcher said...

Okay, but let's not let the journalist/reporter by without mocking the innumeracy demonstrated by this supposed adult.

19 passing out of how many students (X) attempting the test. (Dunno, not reported, perhaps not even asked.) Students attempting out of how many eligible (N) -- seniors, or juniors, or both? (Dunno.) 19 out of X Compared to ratio (X/N) in high schools in cities of similar size in the same state? (Dunno.) How this compares to the same, and nearby, schools' ACT results, if any (Dunno.)

To be fair to the reporter, the school district staff itself (in the district where I served as a trustee one term) is prone to serve up "naked numbers" of this sort, devoid of all context, unless deliberately and aggressively challenged. But a so-called journalist should be in the habit of asking such questions and making such challenges, and one who does not suggests he or she doesn't understand numbers at all.

Gringo said...

J Melcher
Okay, but let's not let the journalist/reporter by without mocking the innumeracy demonstrated by this supposed adult.
Yes, the report would have been better with more numbers, so that "bad" could have been better translated into "how bad." But that number is BAD.

To be fair to the reporter, the school district staff itself (in the district where I served as a trustee one term) is prone to serve up "naked numbers" of this sort, devoid of all context, unless deliberately and aggressively challenged.
I have found very little number of students data on the Paterson Public Schools website. As corroboration of that, the Wikipedia page on Paterson Public Schools gets its number of students data from the National Center for Educational Statistics. [footnote 9]

From downloading the Excel file at that site, we find that there are 5788 students in Grades 9-12 in Paterson Public Schools. I will estimate 800-1200 graduates.

Regarding fudged education statistics, some years ago I looked at enrollment data for some urban high schools in TX. The senior class would be about 60% of the size of the freshman class, yet the dropout rate would be reported as 2-5% per year. This was widespread.

Cass said...

I am a very fast reader, but my husband (who is, in many ways, far smarter than I am) is a very slow reader. I suspect that part of this is personality driven - he is nothing If not thorough, but his reading comprehension is excellent and he tends to retain details of anything he reads.

I literally inhaled books as a child, and even as a new mother I read 3-4 novels a week (OK, this isn't challenging fare, but it's more challenging than watching televised dramas).

Most kids today don't read: reading requires concentration, whereas the Internet actively inhibits focus. I was rather amused to watch my 7 year old nephew retreat to his room recently to read Pokémon training manuals :p

I hadn't seen him take quite that much interest in reading by himself, but he's obsessed with Pokémon. If that's what it takes....

My other son and his wife already read 4-5 books a day to their 4 month old baby daughter. I love it!

Assistant Village Idiot said...

Because I had exceptionally good scores in 1970, I went for years pontificating about what people should do to increase scores - what parents should do, kids should do, what preps should be sought. I kept it up through my first two sons taking their SAT's through 2002. They likewise had superior scores. I was an authority.

I was completely wrong.

SAT scores are largely genetic, and we need to stop pretending that parents, schools, great books, high expectations, reading styles, breakfasts, or charters will move the dial more than a few points.

They just don't. Anecdote and narrative don't change that. The evidence is accumulating year over year.

And yes, the SAT is an IQ test.

Anonymous said...

I woud have the school system audited, because I am willing to bet that a substantial fraction of the per capita spending is NOT spent on teaching the children. I would ask specific questions about what math topics and grammar topics are addressed. This is only one step, of course.

"SAT scores are largely genetic" -- this misconstrues the nature of the SAT test. The SAT test measures achievement, and people who are not exposed to the subject matter of the test will do poorly, regardless of their native intelligence.

I was an Air Force brat who attend multiple school systems before I was tested for the SAT as part of a suburban San Antonio, TX student population. I was a Semi-Finalist, along with two other people at my school, both of whom had a similar (peripatetic) history. The people who grew up in the local system and were good students scored in the 1200 range, and I was at 1600. The population was large enough (and included kids smart enough) to yield at least two semifinalists.

In that Texas high school, I took dumbed-down physics and math classes, because the classes I would have qualified for in the Air Force high school were not offered.

When I went to college, the the top grades in my chemistry classes were dominated by people who were geniuses and those normal people who had taken calculus in high school. I was too lazy to make up the difference in schooling, and settled for Bs.

The Texas high school did have good english teachers, and I scored substantially higher on my english portion of the SAT than on the math, although I outscored very talented math students on the math portion, as well.


Ymar Sakar said...

If parents refuse to do as they are told, their children will be either taken by CPS or put under stricter regulation by the state.

Gringo said...

SAT scores are largely genetic

My personal experience with inherent abilities came in my first two years of high school. Before entering high school my favorite subject by far was history. I did OK in math, but I was indifferent to it- I went through the motions. Early in high school I had a bad math teacher with a good math text to teach myself, and a bad history teacher. Result: I developed a loathing for history and a liking for math. Conclusion: experience with a bad teacher can give a student a clue to where his inherent abilities and interests lie. From experience with bad teachers, I found out that my inherent abilities and interests tended more towards math than towards history.

While SAT scores are largely genetic, I suspect that given two individuals with identical or near identical IQ scores, one of whom reads a lot and the other who hardly reads at all, the one who reads a lot will have a higher SAT Verbal score.

I taught a year at a black- hispanic "poverty" school.Many or even most of the students were hostile to the idea of learning. As I saw it, their bad test scores reflected this hostility. There is a definite negative "nurture" component to the nature/nurture argument in such schools.

State test scores at the school gradually improved over the years, as the school made progress in neutralizing the negative aspect of the nurture side.

I once read an article comparing the study behaviors in Calculus classes at UC Berkeley of Black and Asian students. Blacks studied alone. Asians studied in groups. Blacks didn't study as hard as Asians. There was a definite "nurture" aspect to the different Calculus class outcomes.

Trivial fact re SAT. In my high school class of around 160, I can recall 9 with SAT Math scores of 750 or above. And this was before the "recentering." In 2005, of around 150,000 blacks who took the SAT, there were about 250 who scored 750 or above. Not encouraging news.

Texan99 said...

AVI makes a good point, though. Whether or not the SAT is purely an IQ test, it's not really an achievement test, either. If the Paterson school district were ditching the SAT because they'd found a better achievement test to use that measured the acquisition of knowledge and skills more than innate intelligence, I wouldn't have a problem. In this case, though, they clearly are ditching the test because their students can't score high enough on it.

And what in the heck is going on in Paterson that the IQs of its high school students are so low? Some kind of brain-flight? A 1500/2400 (or old-fashioned 1000/1600) score is not way out there on the long tail of the distribution curve; it's about state-school level. Above average, but not brainiac territory.

Cass said...

I think IQ gets conflated with a lot of other factors that are impossible to tease out.

My IQ is supposedly in the top 1%. I got very high verbal PSAT and SAT scores (not surprising, as I literally inhaled books growing up) and a very "meh" PSAT math score (490 before recentering). Just an additional few months of math and my math SAT rose to 620 - not in the stratosphere, but clearly not as bad as 490 either.

I'm pretty sure my IQ didn't spike in those few months :p

Likewise, standardized tests are often claimed to measure intelligence rather than achievement but I think that's BS too. My kids' scores (especially in math) were all over the map and I definitely noticed a strong correlation between being in a challenging school and better scores the next year.

After home schooling for only 1 year, both their scores went through the roof. 1 year of Saxon math in another school improved one son's scores from below the 70th percentile to almost 90th (and he is not terribly good at math, nor was he ever interested, nor did he put forth much effort) :p This is the son most like his Mom.

I have also taken IQ tests before and after a few practice questions and seen a definite increase in my scores. So I can't believe this is all down to IQ, no matter what "Science" tells us.

I'm pretty sure there's a strong correlation between having smart parents (though we're told that IQ isn't heritable!) but it's one factor among many.

And I think the smarter you are, the more you get out of school and/or tutoring. It may be at the bottom of the scale that school doesn't help as much, just as voice lessons won't help someone with a lousy voice become a coloratura opera singer, but might well be the defining thing that allows a talented singer to excel.

Cass said...

One more thought: some people enjoy taking tests and test well. I do, though I don't tend to test well on problems that require a lot of thought or are very difficult. They make me anxious, or I get bored and b/c the stakes aren't high, don't try hard.

Others are terrified of tests and consistently test below their real knowledge/ability. This is well documented in girls and women.

Tests are very imperfect instruments.

Texan99 said...

Even an aptitude test like the SAT is responsive to practice effects. The verbal portion relies heavily on vocabulary, which of course is learned and not inherent. Still, it's a far cry from what I would call an achievement test, which focuses on whether specific batches of material have been learned. Colleges don't use the SAT to discover which students have completed certain material, they use it to judge general aptitude for college material, which is pretty close to IQ.

It's true, though, that a dose of studying habits and test-taking familiarity is thrown in. Even on the math part, which is nearly pure abstraction, all but the wildest sort of natural geniuses would have to be reasonably familiar with the style of problems to have much hope of scoring well.

But really: 19 kids out of a town of 150,000 could hit 1000/2400? The Wash. Times reports that the 19 Paterson students work out to 3.2% of their class, down from 4.3% last year. Nationwide, a 1500/2400 score bumps you just slight over the 50th percentile. Paterson’s cumulative mean SAT score this year was 1120 (747 on the old 1600 scale), which if I'm reading the SAT site correctly is about the 9th percentile nationwide.

In my public high school, with a class of about 800, there were 3 kids to my certain knowledge who broke 1500 out of 1600 (2250 on the 2400 scale), and there may have been more who didn't speak up. I can't imagine how many hit at least 1000/1600 (= 1500/2400).

As for the heritability of IQ, there obviously is a very, very strong genetic component, though equally obviously no trait, and especially no complex trait, is perfectly and predictably heritable. Studies of identical twins separated at birth make this point beyond any reasonable ground of dispute.

Texan99 said...

Correction: "But really: 19 kids out of a town of 150,000 could hit 1500/2400?" not "1000/2400." Two scales messing with me! Do they change these things around deliberately to keep anyone from making invidious comparisons?

Anonymous said...

"Do they change these things around deliberately to keep anyone from making invidious comparisons?"


I have learned to notice and translate numbers given in new scales, and the comparison always turns out to be invidious.

For example, I was reading a front page story about local air pollution in an area I know is relatively clean (because the onshore wind crosses the Pacific before it hits land), and I saw some alarming numbers, so I sat down and read the article, wondering where the hell the pollution originated.

The numbers were reported in parts per billion, instead of parts per million. The numbers were actually good. The newspaper wanted to alarm people, so it used bigger numbers, knowing full well that most people would assume they were on the scale they are used to seeing, parts per million.

There was a similar scam worked with our unemployment figures: the reporting was changed to reflect only the people who were unemployed AND receiving benefits. Once a person lost the benefits, the person simply wasn't counted. This scam was caught, which is why numerous sources began to report the official unemployment rate and also the real unemployment rate.

PSAT scores and SAT scores had been dropping for years, before they were re-jiggered. I refuse to believe the kids got dumber over this interval, but I know for a fact the curriculum changed.


Gringo said...

The book City Kid is about a teacher's success with a hard to reach student in a NJ city that sounds to me like Paterson. This is from four decades ago, and sounds as if it could have been written about Paterson today- with the exception that back then more of the students were white.

Gringo said...

Here is my guess on why Paterson scores so low on the SAT. It is heavily immigrant, which makes me think of bilingual education. ”Bilingual” Advanced Google Search at Paterson Public Schools website: 875 hits.

Bilingual education should be a bridge- no more than a year or two. I suspect that students in Paterson are stuck in bilingual education for years and years, which will put them at a disadvantage for the SATs. And more.

Years ago I read about baseball great Manny Ramirez, then with the Indians, taking English classes because he didn't speak English very well. He was 25 or so at the time, and had been in the US for around 13 years. And he didn't speak English well, courtesy of years of "bilingual"[Spanish monolingual] education. Anyone who comes to the US at age 12 should have a good oral command of English within several years.

For WW1 the Army gave its recruits a batter of aptitude tests, which concluded that all those first or second generation doughboys were lacking in intelligence, which was probably due to not as much experience with English. And that was generations before bilingual education. Which also shows why one should be skeptical about aptitude testing of ESL students.

Nicholas Darkwater said...

For Gringo:

Bilingual adjective unable to speak English.

True, bilingual education typically shields children (who have a natural aptitude for learning language beyond the ability of adults) from learning English.