Comment by chewbacha | Hacker News Reader

Comment by chewbacha | Hacker News Reader

Parent Full thread

I have another spin on it. Of those demographics, asians tend to be very recent migrants (https://www.pewresearch.org/fact-tank/2021/04/09/asian-ameri...) and migrants tend to be very hard-working within the first generation of migration (https://obamawhitehouse.archives.gov/blog/2012/07/12/ten-way...). Some of this could be selection bias (those who choose to move are those who are motivated to seek a better life).

I did get a little confused from how we went from Asians doing well in Michigan means the PRC is going to dominate the engineering sector. There seems to be an implication that these are connected. Its my understanding that these demographic data would represent South, SE, and East Asia as part of the same demographic.

Also, his top observation: > In the top band there are many more males than females.

Seems like it could be within a margin of error (3 v 5%). Also, there are more males in the bottom band than females (18 vs 14%). This seems spurious at best.

> 2. The Asian kids are hitting the ceiling on this test. > 3. There are very few students from under-represented groups who score in the top band.

I also suspect that taking his next two observations together imply that he might have a bias against out-groups and in favor of his own in-group. Especially considering much of the topics in the comments section focus on the upcoming SCOTUS trial over affirmative action.

> I did get a little confused from how we went from Asians doing well in Michigan means the PRC is going to dominate the engineering sector. There seems to be an implication that these are connected. Its my understanding that these demographic data would represent South, SE, and East Asia as part of the same demographic.

No, I think his observations about Asians doing well on the SAT in Michigan is unrelated to his point about China. He's saying that since China generally gets high scores on international math tests, and they have a greater population, there's just going to be many more qualified STEM grads than in the US. It's a higher proportion of talent multiplied by a higher population.

This makes sense intuitively. But I'm unsure how reliable these international tests are. Is the test being given at an elite school in Shanghai or in a small rural village? Is it representative of all of China? Do higher scores translate to more innovation down the line? Are high scorers likely to also come up with creative ideas for solving problems? Lots of questions.

if it's any help, Gaokao is a much, much harder test than SAT, and there's only 1 test in a year, so an average Chinese (or Korean, Japanese, Vietnamese,... as they all have the same thing) in general would do much better than your average American in SAT

> Seems like it could be within a margin of error (3 v 5%).

There is no margin of error when your sample is the entire population.

There is if you are missing more than 10% and trying to make extrapolations. Also, there is variations between individuals between tests. You’d need to make multiple samples to reduce noise from the test itself.

So if you're worried about it, do the math. My napkin calc puts the margin at around +-0.3, enough to easily separate 3% from 5%.

So, you expect the exact same results next year? If the same students all retook the SAT with different questions would we get the exact same breakdown +/-0.3%? I doubt it. These data are for this specific measurement and must be tempered in their extrapolation into other areas. I don’t think you can declare them significant.

You must not believe any statistics if you think sampling 85% of a 110k population can't resolve a score like this to 1%. Your original question, which I have been addressing, was about generalization to the population (i.e. what the margin of error quantifies) not nebulous "other areas."

I would expect a re-test to have slightly higher averages overall because all the kids now have experience with the class of questions, whereas only some did the 1st time.

Not at all, I trust statistics a great deal. I don't have statistics on the reliability of the SAT nor how reproducible it is, so I have no way to know based on the numbers presented how noisy the distributions ought to be. You can imagine that some band of test-takers is on the bubble and can fall above or below the "top-tier" line on any given day. You are asserting that this is less than 1% of the population, I'd argue it's likely higher than that. But I don't believe I have the data to present either way.

To address the population; if the "population" you are referring to is only the seniors in Michigan on _that_ day then it's might be pretty accurate. Looks like the population of Michigan is around 10 million (20% of that is under 18), meaning that this would be extrapolated much more to address just the population of Michigan, not to mention what it means if we apply it to the entirety of the united states. Any population that we apply this to that isn't the sample audience at that specific time will incur some error.

At each of these higher levels of population abstractions we have orders of magnitude higher error margins, which would quickly dwarf the ppt difference between males and females in a single test.

Reply on news.ycombinator.com