Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

"even if you only counted the scores of black students"

This is the sort of thing I can never really swallow when reading stuff about education rankings in the states. There's a lot of what seems to be really casual\accidental racism along the lines of the quoted text above as well as things like "controlling for race" and excuses pinning the US's low scores on being "more multicultural". Perhaps this is not the intent but it's sort of painful to read.



It's not "casual racism." It's simply a recognition of the fact that blacks in the U.S. face unique challenges that are not just economic, but social. Those challenges, while important, have effects that can obscure whatever it is that you're actually trying to study.

If you're trying to evaluate school quality, it makes no sense not to control for race and socioeconomics. It wasn't the schools that enslaved blacks, segregated them into ghettos, then subjected them to disparate treatment in a violent drug war. And schools won't be the solution to those things either. This isn't an "excuse" it's a recognition of reality. Education by itself cannot be America's salvation for its past sins.

Failing to control for race and sociology-economics leads you to the wrong conclusions. If you look at the test scores of students in Chicago, ignoring race, you might be tempted to conclude that Chicago has a problem with its schools. Maybe they need more funding or better teachers. That's the wrong conclusion. The reality is that 90% of kids in Chicago public schools are black or hispanic, and almost 90% are low-income. The schools are in areas of strong gang presence, and gang activity has infiltrated every one of the schools. The gangs have replaced parents and teachers as authority figures and the bedrock of social order has crumbled. Schools, no matter how well funded, and teachers, no matter how qualified, aren't going to fix those root causes.

My wife taught in a school on Indian land in eastern Washington. She had an awful experience, because she realized that these kids didn't need teachers with degrees from good schools, but fathers who weren't alcoholics, or indeed fathers in the first place. Schools and education cannot replace these things.


So then why not control for low income or poor social conditions? Why make the connection from black\hispanic to "bad"?


First, because there is more to it than just income. My wife grew up in a town of about 3,000 people in Iowa. The town had poor people and rich people, but was ethnically homogenous (98% white), and everyone went to the same schools, same churches, their kids played together, etc. I've been to a similar-sized town in Illinois that is ethnically mixed (about half hispanic, half white). The hispanics in that town are poorer than the whites, as a group, go to their own church, don't speak English at home, their kids play with other hispanic kids and not so much with the white kids, though they all go to school together. Then you have places like Chicago, where a neighborhood might be 95% low income hispanic, where kids only play with other low income hispanics and go to a school that is 90%+ low income hispanic.

A child with the same family income is going to nonetheless have a dramatically different experience in the three cases. They will have dramatically different experiences not because "hispanic = bad" but because the segregation is bad, and race is a powerful driver of segregation in American communities even today.

I agree that it would be better to break things down into the relevant elements, instead of using race as a crude proxy. Adjust for whether parents speak English at home, adjust for educational status of parents, adjust for level of gang activity in the community, adjust for whether father figures are present, adjust for level of social integration between economic classes within the community, etc. Considering all of those things, together, would be better than using race as a proxy. But we don't keep the data in a way that would allow us to measure things that way. We keep track of racial demographics, but not things like the degree to which kids of different economic backgrounds play in mixed instead of segregated groups.


Well for a start, there's a very common [far]-right wing argument that Finns and Chinese are probably innately smarter due to not having many native black or hispanic citizens. You can either ignore it, or you can demolish it pointing out that black kids in certain states actually do better than the very, very white Finnish population....


There's also a lot of what seems to be really casual/accidental accusations of racism in contexts where it clearly doesn't exist (i.e. recognizing and accounting for racial differences isn't racist). That different racial groups score differently on IQ tests, for example, is a fact that can be stated without any judgement of value, or, indeed, without any attempt to explain its causes (which are numerous, I'm sure.)


A true fact stated in a certain way can lead to false impressions e.g. Group A is poorer than Group B, or Group A have less years of education than Group B. If however, Group A is on average younger than Group B, then the differences may disappear. Is is still "true" to state those "facts" without the extra context?


You'd obviously want to control for age in exactly the same way as the comment that started this thread mentioned controlling for race. That's exactly what we're talking about. I'd hope nobody would call you ageist for doing so.


I don't really have any issue with the original comment, but it was someone rushing to provide context for the apparently misleading fact that "Finland does better than the US" so the thread itself shows that facts do not stand alone, and that people care about that context when it's the US as a cultural whole that is low-ranked, in fact it's basically a cliche in these discussions.




Consider applying for YC's Winter 2027 batch! Applications are open till November 2.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: