If Pittsburgh wasn't so artificially small (Penn Hills, Mount Lebanon: rental, crime rates, homes)

[eschaton]

Originally Posted by Goinback2011

What constitutes a 'good' school district? It's being surrounded by parents who motivate their children to achieve. People who live in poor neighborhoods, no matter their color, tend not to be those kind of parents.

I've read a lot about child development, and you're wrong. Parental motivation doesn't really matter that much in school performance. It seems like it does because of genetics. Parents who are above-average and personally driven tend to have above-average and driven kids. When you compare parental involvement and results for adopted children, you find that parents spend more time pushing their adopted kids to succeed, but they have outcomes no better than any random child off the street.

Peer groups do matter a lot however. Studies have shown the biggest reason for a change in student performance during a given school year is if their circle of friends changes - going from the brains to the slackers or vice versa. Students seek to emulate their peer group, and if most of their peer group cares about grades, they will also care about grades. This is undoubtedly another big part of the reason that suburban schools, non-religious private schools, and self-selected urban public schools (magnets, charters, etc) tend to have great results.

Originally Posted by Goinback2011

I lived for 20+ years in Florida, where there was a county school-district. Busing was still being (minimally) used to ship black students to mostly white schools and vice versa during the middle school years. Neighborhood schools were attended during elementary and high school. The worst achieving schools were in neighborhoods with low income parents with children who had behavioral and emotional problems that were evident by the first grade. No one wants these kids in their school. Sad but true. But they WERE, shipped to other schools for a variety of reasons. It meant that areas adjacent to the worst neighborhoods were required to deal with these children at the expense of their own kids' education. An inordinate amount of class time is spent just trying to get control of a classroom with cases like these. Education suffers. Parents move further out if they can. Net net, the bad neighborhood expands its boundaries.

I'm not aware of a single Florida county which is "Detroit" to the county line however.

FWIW, I'm actually not a fan of busing, except maybe at the elementary level. By middle-school, kids tend to self-segregate their social groups by race. So if you ship a bunch of inner-city minority kids to a suburban school, and they all hang out with each other and not the locals, most will continue to have the same peer group mentality that working hard is for suckers. I think it's better to have expanded public school choice through extensive magnet systems, where kids of all backgrounds can mix without any feeling like someone else "forced" them to accept their classmates.

Originally Posted by Goinback2011

This doesn't even touch the cost-shifting that occurs, where money is poured down the drain to try to make kids with emotional disturbances behave normally.

I tend to agree here that some of the kids are lost causes. That doesn't mean that I think that the poor neighborhoods bare sole fiscal responsibility for dealing with them however. We live in a wider society, and if they grow up to be criminals, its going to be everyone's problem.

Originally Posted by Goinback2011

I use Detroit as an example of a city where all but welfare cases, criminals and their case-workers
leave an area because of misguided government policy.

I don't think Detroit declined due to any misguided policy, unless you maybe suggest it focused to heavily on the automotive industry to the exclusion of all else. I think the levels of white flight reached a tipping point in the 1970s, after which point it's been essentially impossible for the city to change course.

To the degree a city can change course, it's largely not a matter of local government. In 1980, DC was 70% Black, and 27% white. It's now 51% black, and 39% white. It was not known for particularly innovative governance during this time period, or dynamic school reforms. It just gentrified due to high local demand, which was a private sector affair.

Originally Posted by Goinback2011

Were the Pittsburgh area to create an Allegheny County School District, I guarantee you, those living in places like Mt. Lebanon and Fox Chapel would move to another county. This would set the stage for the kind of income death spiral we saw in Detroit the last 40 years.

I think you're wrong for the most part. Mount Lebanon and Fox Chapel would still likely retain their old schools, even with merged administration, and city kids wouldn't be bussed in. Regardless, people pick those places for reasons other than the schools, and they'd still retain cache the same way parts of Squirrel Hill and Shadyside do. I'd say somewhere like Upper Saint Clair, which makes its name almost entirely on "good schools," is far more in danger, but it's not like it would become Penn Hills. There's only so much ghetto to go around in the county.

[eschaton]

Originally Posted by pman

while i don't disagree that decoupling residence from school is a good thing, I also don't think that people who are happy with the status quo are wrong. the right solution will likely somehow bridge the difference.

I think they're very wrong, because they equate their own kids' performance with the local school district, when they could be two variables which have very little relationship to one another.

Originally Posted by pman

anyway, to the point about parents, the question is how can a city accomodate both kids that are trouble and kids whose parents want the best? the district needs to kick out the bad kids but those kids need somewhere to go.

I think the magnets in Pittsburgh deal with this well, as the troublemakers can be kicked back into neighborhood schools very easily.

In general I wish the U.S. education system did more "streaming" in the manner that Europe does - developing good tests which determine who is gifted, college track, non-college track, and remedial early, and shunt them into different schools. Without mixed classrooms, teachers wouldn't have to slow down to deal with the few students in a class who aren't getting a lesson, and teachers with average or below-average kids can cater their lesson plan towards easier to grasp concepts.

[BrianTH]

Originally Posted by pman

while i don't disagree that decoupling residence from school is a good thing, I also don't think that people who are happy with the status quo are wrong. the right solution will likely somehow bridge the difference.

I don't know if I would say "wrong", but I do think some of their concerns about possible reforms are likely to be misplaced.

Nonetheless, I would agree there can be a legitimate conflict of interest between those who have already invested a lot on the status quo and those who have not (those who have not including future parents (and in fact others) who have not yet bought anywhere). Because on the whole the incumbents are in a better position to actually exert their will on the local political system, it is true you might need to somehow find a way to placate them if you wanted a purely local solution, but unfortunately it is not clear to me how you would be able to placate them.

wilkinsburg-that's not unique to wilkinsburg

What is specifically true about Wilkinsburg is that its enrollment has gotten so low it doesn't have enough students to fill out even a single high school. There are specific scale efficiency problems with trying to apply the traditional American school district model to a student population that low.

anyway, to the point about parents, the question is how can a city accomodate both kids that are trouble and kids whose parents want the best? the district needs to kick out the bad kids but those kids need somewhere to go.

The only possible answer is to prevent the accumulation of a high concentration of disadvantaged students in any one school.

[pman]

Originally Posted by eschaton

I think the magnets in Pittsburgh deal with this well, as the troublemakers can be kicked back into neighborhood schools very easily.

provided you have enough magnets, but they aren't really a solution since there are plenty of kids that won't be top tier that get stuck in awful schools. it's only a partial solution. of course, provided you have enough magnets, you can at least concentrate on elementary schools.

Originally Posted by eschaton

In general I wish the U.S. education system did more "streaming" in the manner that Europe does - developing good tests which determine who is gifted, college track, non-college track, and remedial early, and shunt them into different schools. Without mixed classrooms, teachers wouldn't have to slow down to deal with the few students in a class who aren't getting a lesson, and teachers with average or below-average kids can cater their lesson plan towards easier to grasp concepts.

I'm not sure I agree with such a track system nor do I think that's really addressing the heart of the problem. kids aren't being held back by the lack of a track system. specifically, plenty of kids could learn more than they do in bad districts if they weren't sharing classrooms with truly unruly people. the kind of kids that rape girls in hallways, assault teachers, disrupt the class room, etc, and simply get shifted around the district. think of it as the same as prisons, there's a different between a guy with possession and a guy who savagely beat another person for $50. anyway, it's not necessarily about ability at all. plus kids need to fail if they've failed.

Originally Posted by BrianTH

What is specifically true about Wilkinsburg is that its enrollment has gotten so low it doesn't have enough students to fill out even a single high school. There are specific scale efficiency problems with trying to apply the traditional American school district model to a student population that low.

that's one of demand (and if wilkinsburg wants to merge with pittsburgh, nothing would stop them right unless pittsburgh didn't want them). there are a lot of small districts in the state with plenty of demand as well.

Originally Posted by BrianTH

The only possible answer is to prevent the accumulation of a high concentration of disadvantaged students in any one school.

spreading them around could do the opposite an undermine schools that would otherwise be decent. it doesn't take many to really disrupt a classroom. I think these types of kids are really special needs kids and probably need a different approach.

[BrianTH]

Originally Posted by pman

(and if wilkinsburg wants to merge with pittsburgh, nothing would stop them right unless pittsburgh didn't want them).

That's two big "ifs". In any event, I was just citing the Wilkinsburg SD as an example where there likely would be efficiency gains from merger into a larger school district.

spreading them around could do the opposite an undermine schools that would otherwise be decent. it doesn't take many to really disrupt a classroom.

There is a good amount of research on this specific issue. There tend to be threshhold effects, and compounding effects, such that you will have more disciplinary problems and adverse effects in total if you concentrate disadvantaged students in a few schools, as opposed to having a few disadvantaged students in each school. That's basically for the reason another poster mentioned above: kids take an awful lot of their behavioral cues from the other kids around them, and if most of the kids in a particular setting are predisposed to better behavior, that will carry through to other kids as well.

In the Pittsburgh region, I would suggest a goal of having no school over 50% disadvantaged. Schools could be a lot less too--the overall percentage of disadvantaged students is about 34% (last I knew), so with some higher than 34% (but below 50%) you could have some considerably lower than 34% as well. The research supports around 50% being a bad threshhold to cross, so this would be a relatively modest but nonetheless important step toward improving outcomes in the overall system.

Now of course that is in the aggregate--switching from the current system to such a reformed system wouldn't likely be a pure Pareto improvement. Again, I think the concerns are often overblown--just to give an example, the Fox Chapel SD actually serves a decent percentage of disadvantaged students (a number consistent with the above-mentioned 50% goal if there were a lot more districts like it), and it is one of the best districts in the area. But that fear of a slight increase in problems in your own school, even if in the aggregate there would be less total problems, is part of the path dependency I noted previously.

I think these types of kids are really special needs kids and probably need a different approach.

It depends, of course, on the individual kid. Again, we know from an awful lot of research that many kids take their cues from the kids around them, and therefore if put into situations where a majority of the kids around them are reinforcing better behaviors, a large portion of these kids would themselves behave better.

Of course that won't be true of every kid--and in fact you can get sociopaths from any sort of background. But I think first you have to try to give every kid a truly fair chance to socialize and develop positive behaviors, and only after having such a chance should you consider any sort of deliberate segregation.

[eschaton]

Originally Posted by pman

provided you have enough magnets, but they aren't really a solution since there are plenty of kids that won't be top tier that get stuck in awful schools. it's only a partial solution. of course, provided you have enough magnets, you can at least concentrate on elementary schools.

I'm actually less sure about the effectiveness of magnets for elementary school for a few reasons. One, even "bad" neighborhood schools are seldom unsafe on the K-5 level. There may be disruptive kids, but there won't be a ton of bullying, fighting, drug issues, etc. In addition, it's pretty early on in the educational lives of students, and there's more than enough time to play catchup in a rigorous middle school/high school. I also like the theory of setting a level playing field to start with, so tests can determine which school kids can go to later. While I understand why Pittsburgh's magnet system was set up as lottery and not admission-based (in order to promote diversity), other cities like Baltimore have managed to keep their magnets diverse with admission-based selection.

Practically speaking in terms of Pittsburgh, while I have no issue using the magnet system, I can't help but wonder if some neighborhood schools would be better without them. For example, it's hard to believe if the magnets didn't exist and Highland Park/Morningside shared an elementary school, that the neighborhood school wouldn't be roughly on par with Minadeo. Ditto for a neighborhood school shared by Shadyside and Friendship/Bloomfield. Of course, these neighborhoods might not be as nice as they are today if they didn't have access to the magnet system, but it still intrigues me.

Originally Posted by pman

anyway, it's not necessarily about ability at all. plus kids need to fail if they've failed.

I don't think we should hold all kids to the same standards. Other countries recognize there is worth in both white-collar and blue-collar jobs, and offer a vocational track. We treat our working class folks here as if they are white-collar failures. Not everyone is cut out for college. Hell, not everyone is going to be cut out to graduate high school (although a lot more could in inner-city schools of course)

Originally Posted by pman

spreading them around could do the opposite an undermine schools that would otherwise be decent. it doesn't take many to really disrupt a classroom. I think these types of kids are really special needs kids and probably need a different approach.

Broadly agreed. Of course, concentrating "problem kids" has its own issues. I had a friend growing up who was essentially a "class clown" was sent to the "problem kid" school for a year. He ended up taking a lot of drugs as a result.

[eschaton]

Originally Posted by BrianTH

There is a good amount of research on this specific issue. There tend to be threshhold effects, and compounding effects, such that you will have more disciplinary problems and adverse effects in total if you concentrate disadvantaged students in a few schools, as opposed to having a few disadvantaged students in each school. That's basically for the reason another poster mentioned above: kids take an awful lot of their behavioral cues from the other kids around them, and if most of the kids in a particular setting are predisposed to better behavior, that will carry through to other kids as well.

I think this would be effective at the elementary level. Past this, not as much, because middle/high schools are large enough and diverse enough that people can find a niche to fit them rather than modify themselves to fit a niche. You'd need to make the kids virtual singletons to ensure that they didn't continue to acculturate to their own youth cohort rather than the school at large. This is also probably why black kids adopted by white families score relatively highly in elementary school, but drop to the median score of their race by high school.

Also, I'd be careful about mixing and matching "disadvantaged" and "problem." While the majority of kids at a place like Westinghouse are clearly "disadvantaged," and score terribly on standardized tests, I don't think you could argue the majority are problem students in the sense of naturally-born sociopaths.

The goal, I think, is to accurately identify the "incorrigible" and isolate them. The amount of sociopaths has been argued to be as low as 0.6% and as high as 5%, but either way, it's a relatively small number.

[pman]

Originally Posted by BrianTH

...
There is a good amount of research on this specific issue. There tend to be threshhold effects, and compounding effects, such that you will have more disciplinary problems and adverse effects in total if you concentrate disadvantaged students in a few schools, as opposed to having a few disadvantaged students in each school.

disadvantaged and problem can be two very different things. I'm referring to kids who can be very violent and uncontrollable. i take it you simply mean lower income.

Originally Posted by eschaton

Broadly agreed. Of course, concentrating "problem kids" has its own issues. I had a friend growing up who was essentially a "class clown" was sent to the "problem kid" school for a year. He ended up taking a lot of drugs as a result.

there is certainly an issue with how you define "problem kid" I suppose, what you mention is sort of the track effect.
regarding elementary, I think I was unclear. I meant by running magnets you could concentrate on improving elemtary schools not that you should run elementary magnets.

Originally Posted by eschaton

I don't think we should hold all kids to the same standards. Other countries recognize there is worth in both white-collar and blue-collar jobs, and offer a vocational track. We treat our working class folks here as if they are white-collar failures. Not everyone is cut out for college. Hell, not everyone is going to be cut out to graduate high school (although a lot more could in inner-city schools of course)

while I don't disagree, there isn't much merit in not holding kids to standards and simply moving them onto the next grade. that's how kids get into high school without being able to read or bothering to show up for much class. one problem I see is that most schools aren't teaching things you won't need working in a shop.

re wilkinsburg-I somewhat doubt they are representative of a lot of districts across the state. simply because most schools might not see much financial benefit doesn't mean that none would.

[BrianTH]

Originally Posted by eschaton

I think this would be effective at the elementary level. Past this, not as much, because middle/high schools are large enough and diverse enough that people can find a niche to fit them rather than modify themselves to fit a niche.

I'll have to double-check to confirm this, but I don't think the research I have seen was limited to K-5.

Assuming that is correct, I think one basic answer to your concern is that it is a possibility but by no means a certainty that a given disadvantaged kid will specifically seek out a "niche" made up only of other disadvantaged kids. Even if there is some correlation along those lines, as long as it isn't a perfect correlation what you should find is that as the percentage of non-disadvantaged kids in the school goes up, the average number of non-disadvantaged kids in a given disadvantaged kid's "niche" increases.

I'm also not sure it is true that peer effects are really limited to some particular "niche". Again, we can assume there may be some sort of disproportionate effect from those within the "niche" without assuming there is no effect from those without the "niche".

Finally, I think we could very likely improve these dynamics precisely by starting early and remaining consistent. It isn't hard to imagine, for example, that if you go to schools with are 80% disadvantaged from K-8, then a school which is 40% disadvantaged 9-12, you will get less benefit during 9-12 than if your K-8 had also only been 40% disadvantaged.

Also, I'd be careful about mixing and matching "disadvantaged" and "problem."

That is the exact opposite of what I am trying to do. What I am pointing out is that there is no reason to believe the number of "problems", or "problem students", is actually fixed, such that this is all just a big zero-sum game. Instead, with the right policies you can have fewer problems, and fewer students regularly causing problems.

But to achieve that end you need to have some way of identifying the conditions that are likely to cause more or less problems to occur, and for that purpose one thing you can do is look for, and alleviate, excessive concentrations of disadvantaged students.

While the majority of kids at a place like Westinghouse are clearly "disadvantaged," and score terribly on standardized tests, I don't think you could argue the majority are problem students in the sense of naturally-born sociopaths.

I think you need to consider at least three categories, and even this is a gross oversimplification. In the first category, you would have actual sociopaths, and I agree that is a low percentage. In the second category you would have non-sociopaths with good behavioral habits. The third category would be non-sociopaths with not-so-good behavioral habits.

We can agree one goal is to properly identify the actual sociopaths, and figure out how to deal with them. That might well mean some sort of segregation, medical treatment, or so on.

But another very important goal would be to end up with as many non-sociopaths as possible in the second category rather than the third category. In other words, we should be trying to help as many of the non-sociopaths as possible develop good behavioral habits.

What I am proposing is most directly aimed at that second goal. But it actually has some relevance to the first goal as well, because it may not always be so easy to tell the difference between the first and third categories. So the more people you can move from the third category to the second category, the less possibility of misplacing someone into the first category, and in fact the more attention and resources you can devote to the problem of properly making that assessment.

[BrianTH]

Originally Posted by pman

disadvantaged and problem can be two very different things. I'm referring to kids who can be very violent and uncontrollable. i take it you simply mean lower income.

See my last post, which I believe covers this ground fairly thoroughly, including how it is a motivating factor behind the proposal I have outlined.

re wilkinsburg-I somewhat doubt they are representative of a lot of districts across the state. simply because most schools might not see much financial benefit doesn't mean that none would.

I agree. As I stated before, "there are not necessarily huge cost savings on a countywide basis available from school district consolidation, although I again think in specific cases (as in my home school district, Wilkinsburg) there are likely a lot of available efficiencies (aka, wastes to be eliminated)."