Looks like a copy of the chart available here:

http://economix.blogs.nytimes.com/2009/08/27/sat-scores-and-...

That post created a lot of debate a couple of months ago which was summarized here:

http://www.marginalrevolution.com/marginalrevolution/2009/08...

The chart on that second page appears to show that "higher parental income predicts higher child income but only for biological children and not for adoptees."

The primary source for the self-reported SAT and income band correlations,

http://professionals.collegeboard.com/profdownload/cbs-2009-...

the annual College Board report for class of 2009, is linked to from

http://news.ycombinator.com/item?id=908779

along with links to other SAT data reports.

The only wierd thing to me is that the strength of the relationship appears to gradually weaken as income gets higher (which would make sense), but then sort of jumps up at $200,000. That's surprising, I would have thought the highest would have been around the upper middle class - comfortable enough to be able to dedicate oneself to school, but poor enough to still be hungry.

I'd like to see the relationship between extremely high SAT scores (e.g., 99th percentile) and family income. I suspect middle to upper-middle class students would proportionately outnumber the rich.

Maybe. My guess is there is a correlation between {income and emphasis on education} and {emphasis on education and SAT scores}. It looks like you are saying at some score the SAT becomes something of an IQ test and IQs are more evenly distributed. I'm not sure that is the case.

No. I think SAT score would be correlated with IQ, but not too strongly on its own (many not particularly brilliant people get high scores by preparing excessively). I think measured IQ and family income would, together, be quite good predictors of SAT performance. But what leads me to believe that also leads me to believe that predictive strength would tail off as family income (or IQ) got extremely high. E.g., for very smart people, the SAT is probably not a challenge, so their score wouldn't benefit much from 10 more IQ points. Similarly, very rich people can't use their money to do much for their kids SAT score than those making $150k / year can.

I think people that make a lot of money tell their children to focus on the important aspects of education. And prepping for the SAT has a much better payoff than most extracurricular activities.

I think the jump is because of the cut-off. If the X were allowed to continue smoothly past 200k, the scores might have followed the same curve all the way up.

But presumably average score for each group is used. I wouldn't think the difference in average score between the 200k group, and the 220k+ group would be great enough to cause that (and I don't think there'd be too many data points the farther you went out to have much influence on the average).

Unless they were using some type of weighted average for each interval and then skewed the $200k group by adding its weighted average in with the intervals beyond it, then averaging that. Which wouldn't surprise me given the appearance of the chart.

It's actually possible to predict someones SAT scores with decent accuracy by knowing nothing other than their zipcode.

Not saying that correlation == causation, but damn.

Comes down to the fact that IQ has a high correlation to your offspring's IQ (although regression to the mean happens there) and to your income (although incremental increases in IQs > 130 make no difference to income, the professor effect). That being said, it is clear to see why if your parents make more money you get better SATs. Of course there are other reasons too, like more stable families, more attention (because higher earners have less kids), more access to books, more mathematically inclined parents that can help you on homework, but in general, these things all come back to the parent's IQ.

There is a fairly logical chain of events here, and I'll agree that while not perfect causation this is interesting.

Higher SAT scores generally mean better college education (on average) with more scholarships (enabling less student-loan slavery post-graduation), more grad school, better jobs, etc.

People that got an 800 on their SAT (out of 1600 in my day) weren't really on the fast-track anywhere except mowing lawns. The kids pushing closer to 1400-1600 were generally headed to decent schools.

Wait what?

I took the chart to mean that if your parents make tons of money, your SAT score is going to be better, not that if your SAT score was good, you will make tons of money. Not sure where exactly that chart is coming from so I dunno what context the axes are in.

I interpreted it that way too, x axis is normally the independent variable

I think you've got it right. Mostly because the writing section was added in 2005. So that would mean that anyone who did better than 560 on the writing is making $200,000 a year at the age of 21. And as a 21 year old who did do much better than 560 on the writing section (as well as the other two sections) I for one am not making $200k+...

Yeah, you could interpret it the other way if you wanted to be contrarian. E.g. A rich family is a family with good genes. But that seems like it would be too categorical for a 15% difference (maybe, what's the min SAT score?).

What makes it striking though is that the average(I assume?) test score isn't higher than 580 at the >$200k level. Granted, I haven't done the SAT's in 20 years, but that's certainly not genius territory there.

It leads me to think that the carefully cynical hypothesis is the more correct one. :-) 15% of your SAT score probably measures your _proximity_ to wealth.

What % of the population takes the SAT at all? Could one argue some sort of survivor bias since we don't include income for non-SAT takers? (sorry, I'm a stats noob so I honestly don't know)

It seems only about a third of the US population takes SAT.

I couldn't get the actual size of the 18-year-old cohort from US Census [1], but from various age groups it could be estimated to be ~4-5M, while SAT was taken by ~1.5M people in 2009 [2].

This would also roughly fit with 27% of the population which got bachelor degree or higher [3].

[1] http://factfinder.census.gov/servlet/STTable?_bm=y&-qr_n...

[2] http://professionals.collegeboard.com/profdownload/cbs-2009-...

[3] http://factfinder.census.gov/servlet/STTable?_bm=y&-geo_...

The submitted graph only shows mean SAT score (central tendency) for each test-taker's self-reported income band, and doesn't show the full score variance in each income band. See also

http://en.wikipedia.org/wiki/File:1995-SAT-Income2.png

(from a Wikipedia article that reports 1995 score correlations) for another relationship in the data that doesn't get reported as often by College Board.

SAT is full of cultural bias. The classic example from years ago was a vocab question including yacht; inner-city kids may never have heard of a yacht, so scored badly. So imagine my amusement when my son saw yacht-racing terms on the GRE last year.

Sure, but that doesn't exactly explain the difference in performance between adopted and non-adopted children. Whether we like it or not, people are different, and many of those differences are genetic.

That chart was totally made in Excel. I'm just saying ...

I thought the same, which caused me to click on the link they used as reference, which turned out not to work... <insert eyebrow raised emoticon here>