4 ms·

I look at this through the lens of the San Francisco public schools. When I went to the magnet high school there, Lowell, they had explicit race based standards

I look at this through the lens of the San Francisco public schools. When I went to the magnet high school there, Lowell, they had explicit race based standards. We all got scores out of 70 based on our standardized tests and middle school grades. The differences were pretty stark. Chinese kids needed a 68, which could mean they got just one B. White kids needed a 62, black and latino even lower. So I totally get why Chinese families were screaming. The standards for them were way higher than for other races. But also just as jarring, the middle schools that fed this system were incredibly biased. My middle school, Hoover, had a segregation system that split the school in half. The honors half had at most one black kid per grade. We even had a different lunch period. The honors half was being groomed for schools like Lowell or at least to look at High School as an on ramp to college. It's not like we didn't have a lot of black students. We actually bussed in two, maybe four busloads of black kids from another neighborhood. But only one of those was ever admitted to honors. I just can't believe that there were no kids on those busses that were smart. I knew a lot of them through sports and they were plenty smart, together, hard working. If this level of segregation can happen in San Francisco, I can't imagine how bad it is in the rest of the country. It's complicated, and you can have your own takeaway, but mine is that merit often defined wrong. I'm on the lookout for people who did more with less, or even did equal or near to it with less. To me, that is more a measure of merit than top line measures like test scores.

hot_gril 4y ago
> It's complicated, and you can have your own takeaway, but mine is that merit often defined wrong. I'm on the lookout for people who did more with less, or even did equal or near to it with less. To me, that is more a measure of merit than top line measures like test scores. I agree with this, and the problem is how many schools look directly at race for that. This reminds me of predictive policing ML models that take race as a factor. Yes it gets more accurate results than if it didn't use that information, but that doesn't make it fair.
- cool_dude85 4y ago
  So giving Black students a slightly higher chance to get into Harvard, all other things being equal, reminds you of sending the cops to overpolice neighborhoods Black people live in? I guess they're the same in a way. Doing bad things reminds me of doing good things because things are being done in either case.
  - hot_gril 4y ago
    > giving Black students a slightly higher chance to get into Harvard Giving Asian students a significantly lower chance to get into Harvard, because they're considered "more privileged" based on race alone.
- godelski 4y ago
  > This reminds me of predictive policing ML models that take race as a factor. In statistical systems you often have to take in race as a factor to remove it as one. You can't debias a system if you can't determine the bias. There are many things that highly correlate with race: a notorious one being zip code (byproduct of red lining). But there's also a lot of non-obvious ones and considering historical data is used, the machine will find those correlations. So only with that data can you then analyze and tweak your model to remove the racial biases.
  - hot_gril 4y ago
    The system developer has to look at race to de-bias a system, but the system shouldn't take race as input, unless it's somehow able to de-bias itself. In this example, I'd wonder why ZIP code is even an input.
    - godelski 4y ago
      So I'd consider that part of the system. You could also train your model to debias by using a cost function that incorporates race and minimizes the KL divergence between different ethnicities. (Obviously it would be best to use interpretable models and/or causal ones) A Mahalanobis distance might also be good here. But when training your model you might want to PCA it and if race isn't an included factor then you can't see if it is a main contributing component or not. You should be training your model such that the racial variable contributes near nothing to the model's outcome (easier said than done). > In this example, I'd wonder why ZIP code is even an input. The school you go to is strongly correlated with zip code. So... if any of your schools are considered on your college app (hint: they are) then you have extracted the zip code (which again, points back to red lining[0]). In the crime example, the zip code is going to also correlate with the frequency and type of crime as crime (especially when broken down into types of crime) is not homogeneously distributed across a country/state/city (to be honest, not even a zip code! But the more you zoom in the more accurate you get. Aggregation bias is a pain). This stuff is a pet peeve of mine. Our tiny human brains weren't designed to figure out how all these things are connected. That's why we use data science and statistics. It is why data science and statistics is so fucking hard too! Even more so when you get into causal statistics. Contributions may frequently be obvious post hoc, but they rarely are a priori (which is why "novelty" is so absurd). The world is extremely complex and we have long chains of causality that we have to work with. The problem here is our brains are designed to simplify things. To take first or second order approximations. The problem with this is that these approximations are not sufficient for navigating the major issues we face in the modern world. These low order approximations often lead us in the wrong direction rather than being a small step in the right direction. If you're taking a gradient descent in a complex feature space (thousands of features/dimensions) then approximating it in 2D will get you nowhere near an equilibria. There's probably dozens, or hundreds, of features that are highly significant. All before we get into the mess of coupling and confounders. This is a problem because: 1) we fight over things we're not even qualified to have an opinion on; 2) the actions we take are frequently determined by popularity with a hint of analysis and the popularity is determined by low order modeling; 3) because we have a limited view of our actions we can easily convince ourselves that low order approximations are good because they work locally (spacial and temporal) but not over the global solution space (humans are notoriously terrible at long term thinking/planning and this too is highly correlated with income levels/life stability). Our modern world is highly dependent upon specialization but we have a culture that tries to convince us that expertise in an extremely narrow domain generalizes really well. It doesn't and that is destroying us. This isn't just a "you" problem, so you're not being targeted. These issues are highly common on HN and anywhere nerds congregate because that cultural pressure convinced us that our self-worth is dependent upon the generalization of our knowledge. Einstein nor Feynman said you don't understand something unless you can explain it to a simpleton/your grandma/a child. Most stuff you can't because domain expertise is required for even basic understanding. So I'm trying to put social pressure to solve this problem and I hope others will join. tldr: The world is fucking complex, treat it that way [0] One of the major issues with red lining (there are a lot) is that where you live determines the quality of education you get. School funding is often related to housing taxes. Cheaper houses? Less funding. Less funding? Lower education. Lower education? Increased difficulty in climbing the socioeconomic ladder. Which correlates with crime rates. Which correlates with a lot of things. These downstream effects aren't actually dependent upon race but if you correlate an upstream effect with race then the downstream ones do too.
naniwaduni 4y ago
> If this level of segregation can happen in San Francisco, I can't imagine how bad it is in the rest of the country. The forward conclusion is that the rest of the country must be unimaginably worse than SF—which is an easy failure of imagination. The backward conclusion is that what SF is doing is unhelpful or actively counterproductive—which is not inconsistent with track record.
whimsicalism 4y ago
At least since 2007, that sort of explicit race based discrimination undertaken voluntarily is not legal in public school districts, see PICS v. Seattle.
- rahimnathwani 4y ago
  What about this, which has been on SFUSD's web site for the past 12 months? Tutoring for African-American SFUSD Students Citywide Posted on September 23, 2021 The Citywide Tutorial Program is partnering with after school programs across San Francisco to offer homework help and tutoring for African-American students enrolled in SFUSD schools, kindergarten through 12th grade. SFUSD Access and Equity Department and SFUSD State and Federal Programs Department provide a SFUSD credentialed teacher to tutor African-American students at each location, with a virtual option: https://web.archive.org/web/20220819030847/https://www.sfusd.edu/announcements/2021-09-23-tutoring-african-american-sfusd-students-citywide https://web.archive.org/web/20220819030847/https://www.sfusd...
  - whimsicalism 4y ago
    Good question, I do not know the answer.
  - frumper 4y ago
    This sounds like it violates prop 209 unless either the program predates 1996, or anyone can get tutoring despite it's specific program name.

hot_gril 4y ago

> It's complicated, and you can have your own takeaway, but mine is that merit often defined wrong. I'm on the lookout for people who did more with less, or even did equal or near to it with less. To me, that is more a measure of merit than top line measures like test scores. I agree with this, and the problem is how many schools look directly at race for that. This reminds me of predictive policing ML models that take race as a factor. Yes it gets more accurate results than if it didn't use that information, but that doesn't make it fair.

cool_dude85 4y ago
So giving Black students a slightly higher chance to get into Harvard, all other things being equal, reminds you of sending the cops to overpolice neighborhoods Black people live in? I guess they're the same in a way. Doing bad things reminds me of doing good things because things are being done in either case.
- hot_gril 4y ago
  > giving Black students a slightly higher chance to get into Harvard Giving Asian students a significantly lower chance to get into Harvard, because they're considered "more privileged" based on race alone.
godelski 4y ago
> This reminds me of predictive policing ML models that take race as a factor. In statistical systems you often have to take in race as a factor to remove it as one. You can't debias a system if you can't determine the bias. There are many things that highly correlate with race: a notorious one being zip code (byproduct of red lining). But there's also a lot of non-obvious ones and considering historical data is used, the machine will find those correlations. So only with that data can you then analyze and tweak your model to remove the racial biases.
- hot_gril 4y ago
  The system developer has to look at race to de-bias a system, but the system shouldn't take race as input, unless it's somehow able to de-bias itself. In this example, I'd wonder why ZIP code is even an input.
  - godelski 4y ago
    So I'd consider that part of the system. You could also train your model to debias by using a cost function that incorporates race and minimizes the KL divergence between different ethnicities. (Obviously it would be best to use interpretable models and/or causal ones) A Mahalanobis distance might also be good here. But when training your model you might want to PCA it and if race isn't an included factor then you can't see if it is a main contributing component or not. You should be training your model such that the racial variable contributes near nothing to the model's outcome (easier said than done). > In this example, I'd wonder why ZIP code is even an input. The school you go to is strongly correlated with zip code. So... if any of your schools are considered on your college app (hint: they are) then you have extracted the zip code (which again, points back to red lining[0]). In the crime example, the zip code is going to also correlate with the frequency and type of crime as crime (especially when broken down into types of crime) is not homogeneously distributed across a country/state/city (to be honest, not even a zip code! But the more you zoom in the more accurate you get. Aggregation bias is a pain). This stuff is a pet peeve of mine. Our tiny human brains weren't designed to figure out how all these things are connected. That's why we use data science and statistics. It is why data science and statistics is so fucking hard too! Even more so when you get into causal statistics. Contributions may frequently be obvious post hoc, but they rarely are a priori (which is why "novelty" is so absurd). The world is extremely complex and we have long chains of causality that we have to work with. The problem here is our brains are designed to simplify things. To take first or second order approximations. The problem with this is that these approximations are not sufficient for navigating the major issues we face in the modern world. These low order approximations often lead us in the wrong direction rather than being a small step in the right direction. If you're taking a gradient descent in a complex feature space (thousands of features/dimensions) then approximating it in 2D will get you nowhere near an equilibria. There's probably dozens, or hundreds, of features that are highly significant. All before we get into the mess of coupling and confounders. This is a problem because: 1) we fight over things we're not even qualified to have an opinion on; 2) the actions we take are frequently determined by popularity with a hint of analysis and the popularity is determined by low order modeling; 3) because we have a limited view of our actions we can easily convince ourselves that low order approximations are good because they work locally (spacial and temporal) but not over the global solution space (humans are notoriously terrible at long term thinking/planning and this too is highly correlated with income levels/life stability). Our modern world is highly dependent upon specialization but we have a culture that tries to convince us that expertise in an extremely narrow domain generalizes really well. It doesn't and that is destroying us. This isn't just a "you" problem, so you're not being targeted. These issues are highly common on HN and anywhere nerds congregate because that cultural pressure convinced us that our self-worth is dependent upon the generalization of our knowledge. Einstein nor Feynman said you don't understand something unless you can explain it to a simpleton/your grandma/a child. Most stuff you can't because domain expertise is required for even basic understanding. So I'm trying to put social pressure to solve this problem and I hope others will join. tldr: The world is fucking complex, treat it that way [0] One of the major issues with red lining (there are a lot) is that where you live determines the quality of education you get. School funding is often related to housing taxes. Cheaper houses? Less funding. Less funding? Lower education. Lower education? Increased difficulty in climbing the socioeconomic ladder. Which correlates with crime rates. Which correlates with a lot of things. These downstream effects aren't actually dependent upon race but if you correlate an upstream effect with race then the downstream ones do too.

naniwaduni 4y ago

> If this level of segregation can happen in San Francisco, I can't imagine how bad it is in the rest of the country. The forward conclusion is that the rest of the country must be unimaginably worse than SF—which is an easy failure of imagination. The backward conclusion is that what SF is doing is unhelpful or actively counterproductive—which is not inconsistent with track record.

whimsicalism 4y ago

At least since 2007, that sort of explicit race based discrimination undertaken voluntarily is not legal in public school districts, see PICS v. Seattle.

rahimnathwani 4y ago
What about this, which has been on SFUSD's web site for the past 12 months? Tutoring for African-American SFUSD Students Citywide Posted on September 23, 2021 The Citywide Tutorial Program is partnering with after school programs across San Francisco to offer homework help and tutoring for African-American students enrolled in SFUSD schools, kindergarten through 12th grade. SFUSD Access and Equity Department and SFUSD State and Federal Programs Department provide a SFUSD credentialed teacher to tutor African-American students at each location, with a virtual option: https://web.archive.org/web/20220819030847/https://www.sfusd.edu/announcements/2021-09-23-tutoring-african-american-sfusd-students-citywide https://web.archive.org/web/20220819030847/https://www.sfusd...
- whimsicalism 4y ago
  Good question, I do not know the answer.
- frumper 4y ago
  This sounds like it violates prop 209 unless either the program predates 1996, or anyone can get tutoring despite it's specific program name.