[kenberg]

I continue to be interested in that study. I was mulling it over with my afternoon trip to Starbucks.

It occurred to me that the horizontal scale reflects some standard deviation. The numbers seem reasonable for that. But standard deviation of what? All Massachusetts students, I suppose. The center is at 0, as they say. The curves are not really symmetric, but maybe close enough so that the peak can be thought to occur at the median.

This would mean that the median score for the black kids was approximately the median for the state when they entered the program in sixth grade, but then went up by about half a standard deviation by 8th grade. That's impressive.

But it still leaves me wondering why there was little if any change in the white scores. They were put into the same high expectations format, coupled with longer school days and monitoring of teachers. It produced no result at all? I was born a skeptic about data, and I have not grown out of it. I just don't understand this.

[kenberg]

 barmar, on 2016-November-06, 13:41, said:

You might have to find the graph in the original paper

https://seii.mit.edu...anuary-2016.pdf

to find out how to interpret the details.

It's not the students who are already so good, but the schools that white students come from before enrolling in a charter. They'd already received a good education, so there's not as much room for improvement.

I was posting my second thoughts as you were writing this. I in fact did go to that paper.Thirty pages or so, fairly dense. It has graphs, some similar to the ones in the article but I could not find the exact graphs.

I can well imagine a lesser effect. If the change in the school environment is not as large, presumably the effect will be not as large. But there appears to be almost no effect. That's harder to understand.


And it's a bit ominous. I imagine the kids have to travel further to get to this charter school than if they just stayed put. And the hours in school are longer. If there is no effect, why would anyone do this?


Becky went to Lowell high school in San Francisco, public but you had to apply. She worked hard in middle school to get into it, and she took bus and BART daily to get there. It paid off. Me, I just walked, or sometimes hitchhiked, to the nearest one.

[hrothgar]

It appears as if the authors are excluding poorly performing charter schools from their analysis

Quote

We set out to study the effects of attendance at six charter high schools
in Boston. These schools generated the lottery-based estimates of charter
high school achievement effects reported in our earlier study ðAbdulkadirog˘lu
et al. 2011Þ, and they account for the bulk of charter high school
enrollment in Boston today.3 Two additional charter high schools serving
Boston students in the same period are now closed. One school that is still
open has poor records and appears unsuitable for a lottery-based analysis.

[awm]

 hrothgar, on 2016-November-06, 14:53, said:

It appears as if the authors are excluding poorly performing charter schools from their analysis

In general the results from charter schools are very mixed. Some seem to do very well, others very badly. I think any blanket statement as to whether charters are good or bad is over-generalizing. The study seems to be specifically about a certain type of charter school which sets very high expectations of students and has very regular observing/coaching of teachers in a classroom environment. It's believable that this type of school tends to perform well, while some of the charters which seem more about cutting costs (and possibly making a profit for the organizer) are worse than the public schools. Of course, deciding which are the "good kind" of charter schools can be somewhat arbitrary, but it seems reasonable to try to learn from the "best practices" while shutting down the charters that perform worse than the local public schools.

[kenberg]

 hrothgar, on 2016-November-06, 14:53, said:

It appears as if the authors are excluding poorly performing charter schools from their analysis

Yes, very much so. As I understand their claims(s), it is not at all that "Charter schools are great". Rather it is that "High expectations coupled with long hours and good teaching produces good results".

The NYT may have been a bit too quick to accept the results at face value. I might or might not read the referenced article but I hope that it all receives the careful analysis such claims deserve.

If we can really bring about results such as the study claims this would be great news. I hope it stands up.

[billw55]

Regarding charter schools. Even when they are not selective at admission - either for academics or behavior - they still may not be representative of public school students at large. I have read pieces before about charters that are aggressive in removing kids with behavior problems from the school; I suspect that this alone could show up as a performance improvement over the general population. And even failing all that, applicants for a lottery admission still come from parents that care about their kids education, which can make a big difference. Short version, most charters have some selection bias at work.

[hrothgar]

 billw55, on 2016-November-07, 07:17, said:

Regarding charter schools. Even when they are not selective at admission - either for academics or behavior - they still may not be representative of public school students at large. I have read pieces before about charters that are aggressive in removing kids with behavior problems from the school; I suspect that this alone could show up as a performance improvement over the general population. And even failing all that, applicants for a lottery admission still come from parents that care about their kids education, which can make a big difference. Short version, most charters have some selection bias at work.

The paper in question was using the Massachusetts Charter school lottery to control for some of this.

MA doesn't have enough charter schools to admit every applicant
They assign folks based on a lottery

The comparisons that are being done are between students who applied to join a charter school and got a spot versus students who applied to join a charter school and did not get a spot

[kenberg]

Yes, this is a strong part of the research. The random selection was from those who had applied. So it becomes possible to compare the group "applied and randomly selected" with the group "applied and randomly not selected". Clearly this helps. I am still concerned. Charter schools are a political issue and, this year, a ballot issue in Mass. I try to be neither cynical nor naive. I suppose the vote in Mass is tomorrow, but I hope this gets ongoing review and discussion.

[barmar]

John Oliver did a scathing report on charters schools a couple of months ago. It was basically about the way that many of them have been poorly managed, and that the free market is not the right way to do public education (similar to arguments against privatization of prisons, which gives incentives to increasing prison population).

https://www.washingt...-video-contest/

Despite that, I voted for more charter schools here in MA. It seems like if they're done right, they can be a big benefit.

[mike777]

 barmar, on 2016-November-07, 11:09, said:

John Oliver did a scathing report on charters schools a couple of months ago. It was basically about the way that many of them have been poorly managed, and that the free market is not the right way to do public education (similar to arguments against privatization of prisons, which gives incentives to increasing prison population).

https://www.washingt...-video-contest/

Despite that, I voted for more charter schools here in MA. It seems like if they're done right, they can be a big benefit.

My take on Charter schools is they offer the consumer more choices, more choices is a good thing. They present competition to non charter schools of all types, public and private. Competition is a good thing.

Unlike your local public school, charter schools are easier much easier to destroy and replace with something or ten somethings if they fail, that is a good thing. Trying to destroy and replace your local public school is very difficult.


Barmar as to your point about using incentives to increase populations, incentives can work and often work. Thus the debate about free college. If we want more students in schools or more crooks in prisons, incentives may work.

[kenberg]

 barmar, on 2016-November-07, 11:09, said:

John Oliver did a scathing report on charters schools a couple of months ago. It was basically about the way that many of them have been poorly managed, and that the free market is not the right way to do public education (similar to arguments against privatization of prisons, which gives incentives to increasing prison population).

https://www.washingt...-video-contest/

Despite that, I voted for more charter schools here in MA. It seems like if they're done right, they can be a big benefit.

I would probably vote the same way. I find the comparison with prisons amusing. Many students would agree, I am sure.

[matmat]

This thread is a fascinating read. Math is hard.

[Elianna]

 matmat, on 2016-November-07, 23:56, said:

This thread is a fascinating read. Math is hard.

Said the Physicist. Did I tell you I'm teaching Physics again?

[kenberg]

 matmat, on 2016-November-07, 23:56, said:

This thread is a fascinating read. Math is hard.

Back when Mattel was getting so much heat for this utterance by their doll, I thought of starting a "Barbie tells it like it is" movement.

[kenberg]

 Elianna, on 2016-November-08, 00:10, said:

Said the Physicist. Did I tell you I'm teaching Physics again?

A story from my youth illustrates the problems teachers have. I was going to put it up before but I must have thought better of it. The cure has worn off.

Billy was a friend in high school. We were not close, but I had a car, he had a car, and he would come over and we would work on cars. Someone told him Physics has something to do with cars so he signed up for the Physics class. So did I, but I was an avid reader of Scientific American, interested in neutrinos, positrons, all that jazz. I was 15, he was 15 or 16. Anyway, Billy would sit in class, sprawled out (he was large) at his desk, and every so often interrupt with "What does this have to do with cars?". I think he switched out of Physics, or the teacher switched him, after not long. He graduated from high school, I think. I lost touch with him but I heard he died in a car accident while street racing.

There might not be any real point to all of this except, perhaps, that teachers have challenges. If there was any way to to get Billy interested in Physics I don't know what it was. Neutrino detection (1956, I remember the excitement but I had to check the date) was still in the future for my 1954 class, but he was not interested in such things. Elapsed time for the quarter mile, yes. Neutrinos, no.

Good luck to you and the quarks.

[barmar]

 mike777, on 2016-November-07, 15:48, said:

Barmar as to your point about using incentives to increase populations, incentives can work and often work. Thus the debate about free college. If we want more students in schools or more crooks in prisons, incentives may work.

But they can also have negative consequences. Commercial prisons are also rewarded when there are more non-crooks (e.g. non-violent drug offenders) in prison, or when they stay longer than necessary. So they don't have an incentive to foster an environment where people will be released for good behavior.

The point is that the incentives are not necessarily consistent with the societal goals of these institutions.

[mike777]

 barmar, on 2016-November-08, 11:03, said:

But they can also have negative consequences. Commercial prisons are also rewarded when there are more non-crooks (e.g. non-violent drug offenders) in prison, or when they stay longer than necessary. So they don't have an incentive to foster an environment where people will be released for good behavior.

The point is that the incentives are not necessarily consistent with the societal goals of these institutions.

okj ok ok



We agree that incentives often work to achieve stated goals but not always. We agree that goals change or are confusing over time.

It looks like for this thread the goal was and is to increase the population of whatever.

[mike777]

 matmat, on 2016-November-07, 23:56, said:

This thread is a fascinating read. Math is hard.

Wow what a lazy, very lazy excuse.


You are paid money ...lots and lots of money to make math easier.....

I go to my 9th grade math class.


My teacher made it silly hard.....I WAS CONFUSED.


i REMEMER AT SOME POINT i TOLD HIM IT HE MAKES MATH TOO HARD TO CONFUSING....AT SOME POINT i TOOK OVER...i SAW THIS PROBLEM OVER AND OVER AGAIN.

sEE kENBERG WHO TEACHES WONDERFUL