Hacker News Reader

Comment by aantix | Hacker News Reader

Why would "they" argue against it?

"Socioeconomic status (SES) and SAT scores are positively correlated: Students from higher income backgrounds generally achieve higher scores, and “21.2% of variance in SAT scores is shared with SES, as measured here as a composite of mother’s education, father’s education, and parental income.” The researchers note that the “source of the SAT-SES relationship is likely due to some combination of educational opportunity, school quality, peer effects and other social factors.”

https://journalistsresource.org/economics/ses-socioeconomic-...

View on HN

Presumably because high-SES folks can do even better if they remove standardized testing. I.e., it can be true that "high-SES correlates with high-SAT" but it does not follow that removing SAT requirements will result in more equal outcomes. It's very likely that SAT requirements level the playing field considerably if not completely. It's probably quite a lot of work (and cost of education) for those high-SES folks to get such high SAT scores, so by removing those SAT requirements they can be admitted more easily/cheaply.

Do you have any citations for this? I would think that removing SAT requirements would be done alongside an explicit push to admit more lower-SES students. Anecdotal but where I grew up it was never much work for students to get high SAT scores, most people I knew scored relatively high even before preparing.

> It's very likely that SAT requirements level the playing field considerably if not completely. It's probably quite a lot of work (and cost of education) for those high-SES folks to get such high SAT scores, so by removing those SAT requirements they can be admitted more easily/cheaply.

Those are the only claims I made, and no, I don't have citations for them. Mostly just intuition that the requirements existed to ensure merit rather than nepotism carried the day, so undoing those requirements is likely to have the inverse effect.

> Anecdotal but where I grew up it was never much work for students to get high SAT scores, most people I knew scored relatively high even before preparing.

"Getting into prestigious schools" high? If you're doing that effortlessly then you should bottle your local water and sell it, because there's something in it. :)

>"Getting into prestigious schools" high? If you're doing that effortlessly then you should bottle your local water and sell it, because there's something in it. :)

Yeah, I think it's pretty standard for certain Bay Area public high schools. If I remember correctly the school average was around 1470/1600 (though after prep). For reference Google says the average at Harvard is 1510. Probably has to do more with the parents than the water though.

Doesn't that corroborate the hypothesis that wealth begets higher SAT scores? Access to the quality education available in the Bay Area requires wealth (in general, you have to be pretty wealthy to live in the Bay Area--of course there are some exceptions which prove the rule)?

I'd say it's consistent with that hypothesis, but not that it corroborates it. An alternative hypothesis is that the Bay Area draws in smart folks (who are then wealthy, because Bay Area salaries are outrageous), and smart folks have smart kids. In other words, while wealth is correlated with higher SAT scores it doesn't "beget" them.

To tease that apart you'd need a counterfactual universe where those same Bay Area people weren't wealthy but still had kids. So, uh, good luck finding that...

One datapoint that at least gestures in "correlates" but doesn't "beget" is that half of the students at Stuyvesant High School, the most prestigious math/science school in New York with an admission based on a test akin to the SAT, are economically disadvantaged[0]. If wealth truly "begot" high scores on these tests, you'd probably not see that.

(It also demonstrates that Stuyvesant does a great job making the test accessible: there's free study materials, test times on weekdays and weekends, free shuttles to test site, etc. I wish that was the takeaway here with the SAT discussion: help low income folks prep for and take the test, don't eliminate the test altogether.)

[0] https://data.nysed.gov/enrollment.php?year=2018&instid=80000...

I guess the way I saw it was that the wealth itself begets high SAT scores (through the channel of paying for a good education, which would be invested in regardless of SAT requirements). Removing SAT requirements doesn’t help most wealthy students because they were likely going to do well on the SAT anyway, if anything it hurts them as it’s one advantage they have over others that wouldn’t be considered.

As anyone who worked through high school can attest, not having thr time or money for clubs, volunteering and activism looks poorly in admissions.

> Socioeconomic status (SES) and SAT scores are positively correlated.

Yes and socioeconomic status and being able to write a good essay, come accross as charming in an interview, etc. are even more correlated, which means that people with a high socioeconomic status still benefit from removing SAT scores.

Hard to take seriously any researchers who don't list genetic inheritance of intelligence as one of the likely causes of the SAT-SES relationship.

If they brought it up, they might not be employed as researchers for long. So your heuristic will end up selecting for cranks.

Better a crank scientist than a stable religious official. Because that's what it's called when there are possibilities you are not allowed to examine.

Are you aware of any research supporting genetic inheritance of intelligence. The psychologists I know have told me that there is very little evidence to support that.

See:

* https://en.wikipedia.org/wiki/Heritability_of_IQ

* https://onlinelibrary.wiley.com/doi/abs/10.1002/neu.10160

* https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4270739/

* https://www.cambridge.org/core/product/identifier/S183242741...

From that wiki article:

Twin studies of adult individuals have found a heritability of IQ between 57% and 73% with the most recent studies showing heritability for IQ as high as 80%. IQ goes from being weakly correlated with genetics for children, to being strongly correlated with genetics for late teens and adults. The heritability of IQ increases with age and reaches an asymptote at 18–20 years of age and continues at that level well into adulthood.

> The authors acknowledge that an enriched social, educational, economic and intellectual environment can and does raise intelligence scores, but they insist that even when due allowance is made in terms of statistical comparisons of all other factors and measuring their relative weights, intelligence still seems to be strongly influenced by the genes of one's forebears.

https://www.nytimes.com/1994/10/16/books/what-is-intelligenc...

https://www.nytimes3xbfgragh.onion/1994/10/16/books/what-is-...

There is quite a bit of evidence suggesting genetic inheritability of intelligence. For a very simple example, we have twin studies that show a greater correlation in the intelligence of identical twins than fraternal twins. (The correlations of identical twins still aren't 100%, though - intelligence seems to include a random or "non-shared environment" component) Here's Wikipedia's take: https://en.wikipedia.org/wiki/Heritability_of_IQ

You don't think that smart people are more likely to have smart children? Do you think that tall people tend to have tall children?

Why didn't you say so! I mean, if your psychologists said so then it must be

The alternative hypothesis is completely stupid. Chimps and humans have different intelligence. Are you saying that this isn't genetic?

I am barely paraphrasing, but they want admissions based on skin color. dark skinned people step to the front of the line while light skin people face lottery admissions.

Getting rid of the SAT serves two purposes:

1. Pulling up the drawbridge for smart kids from disadvantaged backgrounds, thus making room for less talented children of the wealthy.

2. Getting rid of objective, falsifiable assessments so they can select instead for ideology and party loyalty.

> select instead for ideology and party loyalty.

Already being done, but not yet at the student level: https://reason.com/2020/02/03/university-of-california-diver...

Not officially, at least. The kind of organization that requires a political pledge as condition of employment, was not bias-free the day before the requirement became official.

> source of the SAT-SES relationship is likely due to some combination of educational opportunity, school quality, peer effects and other social factors.

Parent's attitude toward education is also extremely important.

Is it normal to read books for fun? Is using "big words" acceptable at the dinner table? Is asking questions encouraged?

> Is it normal to read books for fun? Is using "big words" acceptable at the dinner table? Is asking questions encouraged?

My wife once taught a kid whose father kept brusquely "correcting" them any time they they used saw, as in the simple past tense of to see, correctly. The father insisted on seen, and not in a "that's how we talk here" way, but in a "that's how you talk period" way. I'm sure that wasn't the only way he was harming (or trying to harm, anyway) his kid's prospects.

Of course they were poor, too. :-(

"I seen a tree" is less ambiguous than "I saw a tree"

This is quite possibly the funniest thing I have ever read on hacker news.

Reply on news.ycombinator.com