Amazon Asking Cities for Their High Schoolers' SAT Scores | Hacker News Reader

Amazon Asking Cities for Their High Schoolers' SAT Scores

markets.businessinsider.com

80 pointsmeri_dian56 comments

The legality or the morality of this aside, I doubt this is being used as a metric for directly hiring those taking the SAT.

My gut is telling me this is a metric to evaluate the education system for families to move there (or remain there).

Having recently moved, education, was a huge metric – but it is tough to determine. SAT scores are much better than the state testing which vary from state to state. SAT scores gives you the outcome of K-12 experience.

edit - typos

That doens't make sense. Cities typically have lower test scores than the wealthy upper class suburbs that surround them.

>And SAT scores gives you the outcome of K-12.

If you account for many other factors.

The percentage of students that take the SAT in a given area is often overlooked, but it varies wildly by state.

You also have to correct for racial/cultural, language, and income bias in SAT scores.

  You also have to account for racial/cultural,
  language, and income bias

Whatever backlash Amazon faces for asking for overall average SAT scores, it's nothing compared to the backlash they'd get if they'd asked for average SAT scores for white and asian kids from married two-parent families with a six-figure household income, english as a first language and both parents college graduates.

They probably wouldn't need to ask for this. They can just combine public data sets with the SAT scores to come up with a rough, adjusted estimate.

But that's assuming they'd even want to correct for any of the aforementioned biases. I doubt they care because their employees wouldn't care. "Good school" in some places often also means "white school" because of the nature of income disparities, but you never hear parents asking, "But when you say 'good,' are you adjusting for race?"

The gap between SAT/ACT is really important. In my state, SAT scores are actually really high - but it's only because almost everyone takes the ACT and the SAT is only taken by elite people.

Your comment is very on point. The situation where more elite students take certain tests makes it much harder to understand skills and achievements that come from different schools.

Unpopular truth... but most companies (and people) looking for places to move to will, in reality, factor those biases into their decision rather than correct them out.

Who honestly ever says, "Oh, this location is fantastic... as long as you close your eyes and imagine a completely different population living there!"? There are many moral considerations to argue, but raw outcomes do usually prevail.

Except I wouldn't call that a bias. Bias is prejudice, and in this case would mean something like hiring people for certain characteristics that are irrelevant w.r.t. to the job. Even where ethnicity and race are concerned, it doesn't necessarily imply bias. If, say, a black social organization only hired black people, it wouldn't necessarily be biased. Why on earth would they hire, say, white people?

If you were starting a company and you wanted to attract people that value education and want the best for their children and see SAT scores as an indicator of quality, then yeah, you'd do exactly something like this. There's nothing immoral about it, unless you want to debate the effect Amazon will have on the town. What's immoral is the weird guilt and self-flagellation involved in thinking there's anything wrong with it. Who gains exactly from plopping Amazon HQ2 in some shitty town?

If there's anything I worry about, it's the growth of Amazon and its bed fellows.

> My gut is telling me this is a metric to evaluate the education system for families to move there (or remain there).

Exactly. Their employees aren't going to want to move somewhere with bad school systems, even if SAT scores are largely just a measure of home prices. This doesn't strike me as being especially controversial.

Well they should cross Atlanta off their list then, Georgia is 31/50 for schools in the US.

Toronto is looking better now! We even have free Catholic schools!

Georgia is far more than just Atlanta. How does Atlanta itself rank?

It was tough finding a solid answer, but everywhere I looked, it was distinctly not in the top 10 or 20.

Yay for large numbers of first-generation Chinese immigrants!

I don't think Toronto stands a chance. Americans would not move to Toronto because 1) Wages are much lower in Toronto and I don't see Amazon paying twice the prevailing wage. 2) Americans would have to pay taxes in both Canada and America.

> Americans would have to pay taxes in both Canada and America.

You deduct foreign tax payments on foreign earned income against your US tax bill. My assumption is that Canada has higher or comparable tax rates to the US which makes this a non-issue.

If the taxes are higher it makes it an issue again. Only if they are the same or lower in Canada would this be a nonissue.

It's a very selective metric. Only college-bound kids take the SAT.

Not in Georgia. They pretty much try to get everyone to take it here.

Yes. Most people who point to low standardized test scores in Southern states, and snicker about sterotypes, are probably unaware that the region is unusually insistent on high school students taking either the SAT or ACT.

Many Southern states require one or the other as a requirement for high school graduation, period, whether you're planning for college or not. And many others that do not require it pressure all students to take it by default:

https://magoosh.com/hs/act/2017/states-that-require-the-act-...

So of course your statewide averages will be lower than the national average, when the testing population in other regions is limited to college-prep students only.

Yes, but virtually all of Amazon's white-collar employees have college degrees, and it's a good bet they expect their children to get them too.

More problematic for Toronto: Canadians don't usually take the SAT unless they are aiming at US colleges. Most don't.

Meanwhile I'm wondering why Amazon is optimizing for people with families, since it seems to be their MO to hire young people who don't know better than to work 60-80 hours a week.

Because those young people generally tell themselves that they want to settle down and start a family, so school districts seem important.

Is there any evidence that, once you adjust for parental income, ESL status, etc., SAT scores tell you anything about K-12 quality?

My mom was telling me yesterday that a big part of the political agenda in South Carolina revolves around education. Apparently, they're finding it very difficult to encourage companies to locate firms there because it's so difficult to find local talent. It costs far more to relocate people than it does to find someone local.

So lots of legislative activity around education policies.

Umm...are high schools authorized to share that kind of information with a giant retailer? What is to stop Amazon, or anyone else who asks, from simply using that data to target its advertising towards children?

Most of the data is already public: https://blog.prepscholar.com/sat-scores-by-high-school-and-d...

I'm assuming they're asking for city-wide averages (they say "statistics" in the article), not individual results.

It's actually public information, at least at a state level. Extrapolating that to the largest city in the state wouldn't be a hard process, and so I don't think it's concerning information if anonymized and aggregated.

Also, as a Bostonian, I really hope this will edge us over Atlanta and whatnot, as MA has the highest standardized test scores in the nation. Been waiting for ages for this city to become more of a tech center.

ISTM housing costs are the big barrier for Boston? Tech companies already know the area has plenty of talent.

Expensive housing can be a difficult problem, since lots of vested interests are opposed to any effective solution.

I think the reason the Boston is not more of a tech center is that MA law supports no compete clauses as condition for employment. CA has high housing prices where the tech industry is located. However, CA does not support no compete clauses for condition for employment.

There are definitely a lot of factors at play (education, etc. etc.), including a preferential attachment model where cities with established startups and funding opportunities will attract new startups.

You've really hit the nail on the head, however, as that no-compete clause is, while perhaps even enticing to large companies, very restrictive to new startups. We seem to disproportionately have ideas about and go on some entrepreneurial mission in an area we already know intimately, usually related to our current employment. No-compete clauses destroy that entire basket of opportunities.

That seems like a reason that no state besides California would have lots of startups? We're moving the goal posts a bit, however, since thread parent from Boston cares about tech jobs which is not the same as startups.

Are there specific products you would be advertised to based on your SAT score?

SAT Prep guides, among other things? AP Exam study materials? Books on getting into your desired college?

But that's before you take your SATs, not afterwards.

If a school has comparatively high SAT scores, it stands to reason that the kids do SAT prep. Which means even the juniors and seniors are done with theirs, the freshmen and sophomores will be buying materials for the next year.

I'd imagine it's more content curation (i.e. prime video recommendations) and advertising products the kids would want to the parents in that area.

This vein of advertisement is governed by strict 'truth in advertising' standards, but not much else. Amazon has something for everyone and therefore doesn't need to worry about false claims - they just need to know what to put in front of you.

I think most (but definitely at least some) of the departments of education at the state level release average SAT scores by high school/district annually.

MA: https://www.bizjournals.com/boston/news/2018/02/15/here-are-...

TX: https://www.bizjournals.com/dallas/news/2018/02/26/these-tex...

NJ: http://www.nj.com/education/2018/01/the_50_nj_high_schools_w...

IL: http://www.chicagotribune.com/suburbs/ct-school-report-card-...

IN: https://www.doe.in.gov/sites/default/files/accountability/sa...

VA: http://www.nbc12.com/story/33261679/look-sat-scores-show-wid...

CA: http://data1.cde.ca.gov/dataquest/

Teacher here.

Whether or not schools are authorized to share information like that, companies ask all the time. I get tons of junk mail sent directly to me at school (as does every other teacher) asking for student names, DOB, address, etc. or for me to give the students the forms directly for them to fill out and then mail back.

It’s usually for very sketchy “scholarship” or “study abroad” opportunities (from companies and organizations no one has ever heard of and the return address is some random PO box in Deleware.)

Sometimes, however, it’s from straight up marketing companies.

The vast majority of teachers throw these things straight in the trash but some teachers will fill out any form they are given...

Maybe they're trying to strategize on SAT Prep book marketing /s.

Jokes aside, SAT score is not a good indicator of "success" in terms of salary at least in my personal circle. Some of my friends who got perfect scores on their SATs are all at sub-standard jobs, getting paid significantly less than the average salary. Of course, this is purely based on personal anecdote but attending top-tier schools seem to have a stronger correlation.

Top tier schools are a proxy for people who score high on SATs, and/or have wealthy networks that can get them in. Many of the companies that pay the most only recruit at specific schools for this very reason, the high hurdles of getting into the school itself give them candidates that have higher chances of being successful hires.

SAT is not a good indicator of success for a specific individual, but when sifting through thousands of people, it seemingly is a useful metric.

Do you feel that SAT scores have a negative correlation to "success"?

A quick Google for “correlation between sat and salary” reveals a pretty strong correlation. I wouldn’t take your personal observation as meaning anything here.

Haha, Amazon asked me for my SATs way back when too. What dopes.

Amazon kinda already has a metric for this based on what areas are buying prep books.

So, Boston wins!

They've probably already chosen their finalists for the next round and are looking for a metric to justify their choice.

Per [1] the list is currently Atlanta, Austin, Boston, Chicago, Dallas, Denver, Washington DC, and Columbus, Ohio.

Boston has a tech workforce that has a lot in common with SillyValley so it seems like it would be a good culture fit and for a company like Amazon the benefits to being located in the same city and sharing the same workforce as the federal government are innumerable.

Denver is SV-lite. If that's what they were looking for there's a lot of other West coast options that would be on the list.

Dallas and Colombus seem to be the wildcards. They don't have the advantages or workforce or access that Boston or DC do but they're cheap and Amazon might be betting that its size and lack of competition there would let it dominate local politics. They might also be betting that they can tap into the "I'll move but not to CA" demographic" and get a good 5-10yr run as a monopoly on local talent until the tech scene there grows.

Austin, Chicago and Atlanta seem like the middle ground between Boston/DC and Dallas/Columbus.

I don't know enough about local economics in all those places to know how sorting by SAT scores would turn out.

I'm betting DC and Boston are gonna make the next cut. I don't think Denver will make the cut. the other four I have no idea.

[1]http://www.businessinsider.com/amazon-reveals-hq2-candidates...

Edit: Not sure why I'm being down-voted here. Is my analysis wrong or is the suggestion that there exists people in tech who don't want to be part of the SV tech scene just that abhorrent?

edit2 If I wanted to evaluate the pros and cons of all 20 cities listed in [1] I'd write a real article about it and post it on HN for internet points. I just evaluated the ones the author highlighted at the top.

Dallas has a sizeable tech workforce, but they're primarily telecom/embedded engineers instead of the webapp engineers you'll find in SV. It also has a major international airport that's halfway between the two coasts of the US.

Dallas also has a sizable amount of really unsexy business process automation-type development jobs - it's the original home if both EDS and Perot Systems, for instance, and DXC and NTT DATA still have sizable facilities there.

Dallas isn't the #1 tech job location or anything, but there's a decent workforce and Amazon would be a fairly unique offering in the Dallas area.

Lots of data centers too...

Not that those jobs are great... but...

Data center jobs are nice for the relatively small number of employees that find employment in them...

As far as I know, Toronto Canada is still in the running.

Is there an update to that list where you are excluding 12 other cities?

I think Denver is out simply because it's on mountain time. I'd think they'd want a location on EST. It also suffers from a huge divide in talent between Denver and Boulder, which IMO are just far enough apart to make commuting between the two unbearable (even though a lot of people still do).

I think their choice will depend on which business units might be housed in the second headquarter. Atlanta has seen an influx of new corporate headquarters and has a major distribution center already which could help its chance at being selected. The downsides are the local education metrics will differ wildly depending on which area they decide to settle in. The northern section of Atlanta does better than the southern and downtown section, but neither perform to the standards of private schools which is where most affluent residents send their children.

Reply on news.ycombinator.com