« New site about prediction markets | Main | I loved this question, and answer »
The Storm
The storm ravaged the city’s architecture and infrastructure, took hundreds of lives, exiled hundreds of thousands of residents. But it also destroyed, or enabled the destruction of, the city’s public-school system—an outcome many New Orleanians saw as deliverance....The floodwaters, so the talk went, had washed this befouled slate clean—had offered, in a state official’s words, a “once-in-a-lifetime opportunity to reinvent public education.” In due course, that opportunity was taken:...Stripped of most of its domain and financing, the Orleans Parish School Board fired all 7,500 of its teachers and support staff, effectively breaking the teachers’ union. And the Bush administration stepped in with millions of dollars for the expansion of charter schools—publicly financed but independently run schools that answer to their own boards. The result was the fastest makeover of an urban school system in American history.
That's from The Atlantic just over a year ago. Guess what? It's working. The storm is coming.
Posted by Alex Tabarrok on May 13, 2008 at 07:30 AM in Economics, Education | Permalink
Comments
hmmmmm....interesting.
Posted by: sa at May 13, 2008 7:46:19 AM
Since lots of people--poor people especially--left New Orleans and haven't come back, it's hard to see these figures as pointing clearly to "quality of instruction" as the main cause.
Posted by: Alan Gunn at May 13, 2008 7:54:20 AM
Alan, that was my first thought too. But N.O. did re-elect Mayor Nagin; I'd say most of the under-educated are back.
Posted by: Tom at May 13, 2008 8:22:08 AM
@Tom: Nagin is republican I belive.
@Alan: My thought exactly: you can't speak of a random selection of students. In econometric terms: the estimate of the effect of 'breaking the teacher's unions' is severly biased.
Posted by: JSK at May 13, 2008 8:42:54 AM
@JSK: Mayor Nagin a Republican? I don't think so...http://en.wikipedia.org/wiki/Ray_Nagin.
Posted by: Dave Richardson at May 13, 2008 9:02:36 AM
This looks like it has the potential to be evidence in support of more school privatization and such. But unless you get a study of some sort to look at changes from demographic to demographic, you're going to be looking at a pretty biased sample.
Assuming when you looked at the remaining students, you generally found they were wealthier and "more white", you'd have to take a different approach. One way would be to track individual students and look how their test scores changed from year to year (or from 4th grade to 8th grade or whatever) and how the sudden conversion to charter schools effected this relative to some other control group of students in some other major southern city (Atlanta?).
Posted by: mraver at May 13, 2008 9:27:09 AM
At no time is the epistemological admonishment "correlation is not causation" more a propos. I strongly doubt that the New Orleanian's that fled and did not return were a representative, randomly chosen cross-section of public school students. I don't have the data, but one wonders how the exodus rearranged the demographic composition of the student body.
Posted by: Marshall at May 13, 2008 9:30:56 AM
I agree that the solution to improving education involves teachers, parents, and students. Here are my three solutions to improving education that would require very little money:
1) SEND YOUR CHILDREN TO SCHOOL!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Too many kids miss school for “education trips” to Disneyland. I am not saying that vacations during school should be banned, but some kids cannot handle the time missed. I even had one student stay home for the release of Halo III with parental support. What signal is that sending to the student?
2) Allow teachers to punish students. I am not saying corporal punishment, but something/anything to end the student-led anarchy in today’s schools.
3) Quit badmouthing teachers around your children. I an NOT saying that you cannot discuss your issues with the school, but too many kids show up with no respect for the school or its teachers. How do you expect your child to succeed in school if they have adopted your disdain for the teachers?
One more thing. Teachers do not get paid vacation. We work all year. Do lawyers only work when in front of a jury? Do teachers only work in front of students? Both have lots of planning to do a good job!
Posted by: Eric of PA at May 13, 2008 9:35:49 AM
Ditto to Alan. To Tom's point, there was a large exodus of lower class into cities like Baton Rouge and Houston. Many of the ones who were permanently displaced out of the city did not have the resources to return, and it is part of the story as to why the footprint of New Orleans is now smaller. I suspect there's a selection story in this.
Posted by: jason voorhees at May 13, 2008 9:40:24 AM
Alan,
Do you have any evidence that the people left in New Orleans who attend public school are richer than the people who attended public school there before the storm? My gut reaction is if anything the opposite -- those with the least resources are the least likely to have relocated.
Posted by: Josh at May 13, 2008 9:43:33 AM
While we await some unassailable evidence on school choice, can anybody tell me why school privatization (while subsidizing those who can't afford schooling, of course) isn't better than the status quo?
Posted by: josh at May 13, 2008 9:50:17 AM
Josh, as you point out, the data is not yet in on school choice, voucher systems, etc., so the answer to your question lies in the data from test cases. The problems for evaluating such programs lie in self selection. For instance, if there are some charter schools in a district, and parents have to apply to them, then you get a school filled with the kids of parents willing to put out the extra effort to apply and participate in such a school. So, even if the instruction were worse at such a school, test scores would likely be better. Econometric studies that properly control for all of these issues have been mixed, with no clear outcome.
Outside of the data, there are other more normative reasons to not jump at privatization. First, many would argue that privatized schools would allow the wealthy to have better schools by paying the mark-up over subsidy necessary for elite schools. There would then be less pressure for subsidy increases and we would see a greater level of inequality between schools. The notion there is that kids ought to not be subjected to weak schools just because they are born into poor families. The current system already punishes kids for that, especially in inner city situations. Lower income kids in exurban districts benefit greatly from the current public system, since there is intradistrict income mixing there and few kids in such areas go to private schools. Second, and a better reason in my opinion, is that private schools are more likely to blur the seperation of church and state. Many people do not want their tax money to go to schools that are sponsored by or sympathetic to religions that they don't approve of. The state could micro-manage exactly what schools teach, how they recruit, who they accept, etc., but that would end up being unpopular and a scape goat for any performance issues.
Posted by: liberalarts at May 13, 2008 10:12:14 AM
The Atlantic article gives a figure of 75% of NOLA students being eligible for free/reduced lunch in 2004.
Current numbers from Teach For America's site -- http://www.teachforamerica.org/corps/placement_regions/greater_new_orleans/schools.htm -- give 99% eligible in Orleans Parish, 66% in Jefferson, and 78% in St John. I'm not sure what the extent of the New Orleans school district is, but it is at least obvious that public school students in NOLA are not richer on average now than they were pre-Katrina.
Posted by: Josh at May 13, 2008 10:15:09 AM
@Josh
For a number of reasons that are, of course, best laid out in Milton Friedman's "Capitalism and Freedom." However, one of the basic reasons is that teachers are one of the only positions running in which effectiveness is not fiscally rewarded. Teachers are largely paid by the number of years they've been in the school system (and to a much, much lesser extent by the level of education they've attained). Moreover, the teachers themselves have relatively little control over the curriculum, and this leads to students often being taught things they don't need (see most of English), and abandoning things they do (sentence diagramming is awesome).
Posted by: Mdesus at May 13, 2008 10:16:19 AM
Josh - My sense was it was the least well off who were permanently displaced. I know, for instance, that Houston, Jackson Mississippi, Atlanta, Baton Rouge all received waves of lower class individuals. But I don't know of hard data, just anecdotes. But, the March CPS included a set of questions in 2005 on migration related to Katrina that would enable one to answer that easily.
Posted by: jason voorhees at May 13, 2008 10:18:20 AM
@liberalarts
First, many would argue that privatized schools would allow the wealthy to have better schools by paying the mark-up over subsidy necessary for elite schools. There would then be less pressure for subsidy increases and we would see a greater level of inequality between schools.
Because this isn't exactly what is happening in the current system. Yeah Exeter and Compton Central are really on par as institutions. Moreover, while urban schools tend to be worse this is not always so. In fact in the most densely populated parts of NYC the schools tend to be best. Really all the best schools in NYC have independent admissions committees. Is your argument that these problems (all those you named exist in endemic proportions) would be exacerbated? This I don't understand.
Posted by: mdesus at May 13, 2008 10:23:04 AM
I realize my previous comment did not have much to do with this post, but I figured that some teacher bashing will come up at some point, so I though I better get it in.
Another comment: I am a public school teacher who is started to agree with the ideas of privatizing school.
Posted by: Eric of PA at May 13, 2008 10:27:24 AM
mdesus -You clipped the part off where I said that the current system allows exactly what is argued may happen under privatization. And I put that because this is not really my argument but _an_ argument. Currently in cities, the well off go to very good schools (either private like Exeter or public like those in elite suburbs) and the poor mostly go to very bad public schools. Would privatized vouchers change that? In smaller cities and towns it is different (where I live), with income mixing occuring in fairly good public schools. In my kids schools, we have families on welfare all the way up to kids from strong 6 figure income families, kids who drop out of school at 16 and kids who go to Princeton. Were we to move to an all voucher system, there would probably be less mixing in the small town and exurban areas too. Pressure to control local property taxes would keep the voucher value low, with the better off finding it cheaper to pay the mark up for better private schools than to pay the higher property taxes to make all schools better funded via a higher voucher. In other words, the current system is redistributive, with the middle class who are in the public system willing to accept the redistribution in order to improve the schools for their own kids, who are grouped in with the poor kids in a single public school budget. I am not sure how this would shake out. Is redistribution good? That answer is normative, and thus not proveable but rather political. Is redistribution in a school system more justified than other forms of cash redistribution? Again, political and not proveable.
Posted by: liberalarts at May 13, 2008 10:52:10 AM
By the way, here is a related question. We typically compare our school results with those of other countries, and the outcome is usually that we are falling short. Most or all of the countries have public school systems, so what are they doing that we are not, and if it is our government control that weakens our outcomes, why isn't that a problem in other countries? In all other matters, we (the USA) usually suggest that their governments are more inefficient than ours (France, Italy, Japan, etc.), so why are schools flip flopped there? Do other countries have local control? Is it all socio-economic? I honestly don't know and would be interested in what other people have to say about that.
Posted by: liberalarts at May 13, 2008 10:57:28 AM
Here is an article that has many of the facts that are being speculated about. Before the storm there were 60,000 students vs 32,000 today and the return rate to Catholic school was double the return rate to the public schools. This would indicate that not only has the number of students has decreased, but the returning students have different demographics.
http://www.districtadministration.com/newssummary.aspx?news=yes&postid=17167
Posted by: joan at May 13, 2008 10:58:08 AM
we can already see the stratification in our society between rich kids whose families can afford to move to places where other rich families live, or send their kids to expensive private schools to make sure they go to school with only kids from rich families. Privatization and vouchers will merely accelerate this process.
Wealth also creates other potentials. Poor parents are forced to work, and so are not able to participate in the education of their children. In wealthier families the mom, dad, or sometimes both, are able to achieve a 'better work-life balance' and participate in the education of their kids. This fact alone creates enough incentive for private schools to use price discrimination to keep the poor working family kids out, and thus raise their test scores simply using demographics, and not better teachers.
@liberarts, my feeling is that if more conservative christians in the US actually lived in a place like the UK where religious institutions can receive public funding, as long as they adhere to a centrally administered national curriculum, they would be more open to the idea of this. As long as we only offer one-size-fits-all public schooling, this will be used by both conservative and liberal politicians as a hot-button issue for years to come,
Posted by: darin london at May 13, 2008 11:00:12 AM
I clicked through and found Teach for America, and Wendy Koop, the founder. They seems to work miracles.
Posted by: Matt at May 13, 2008 11:02:58 AM
mdeus,
First, English is something most students don't need? Last time I checked, people need to read and write fairly competently for most of their lives in order to be effective workers, citizens, etc. In fact, one could pretty persuasively make the case that more people need high-school level reading and writing skills more than they need high-school level chemistry, physics, or even math skills.
Second, like many libertarian advocates of so-called performance-based teacher pay, you seem to elide the whole question of how performance is measured. (And you also neglect to consider that a teacher's pre-tenure period is effectively that type of evaluation period. But back to measurement...)
How does one measure teacher performance? Student tests? Who designs the tests? What do the tests measure? Do we create a situation in which some states dumb down the tests, which we have now? Do we have some teachers simply teaching wrote memorization for the test and therefore possibly arriving at better scores than teachers who teach more creatively, but who put less stock in the test? What about a situation where one teacher only teaches to the test and another does that and then some? Would they both get paid the same if their respective classes score roughly the same on the test, even if one's students actually learn much more? Is an effective teacher one whose students meet "x" average score? Won't that just encourage teachers who want to make more to teach in high-income areas, since that's correlated with high achievement? Or will the testing track individual students' performance from year-to-year and reward teachers for improvement? What happens if a teacher inherits a class that already has high scores, leaving little room for improvement? What about the fact that since curricula, etc. differ from year-to-year, one needs to find some way to control for that so that teachers who teach "easy" years don't get unduly rewarded?
Or what about having principals or fellow teachers doing the evaluation? But what happens if the principal has a personal grudge against the teacher? Or the fellow teachers represent the teacher-in-question's overachieving nature? Does it just become a popularity contest?
As I think the above makes clear, there's good reason why teachers' unions have advocated for seniority based pay.
That being said, I'm not totally opposed to considering some type of performance pay, but only if it complements, not replaces, performance pay, and if it's something a lot more complex and nuanced than the naive test-based pay suggested by many proponents.
In any case, it's far from the panacea that many suggest.
Posted by: J at May 13, 2008 11:04:56 AM
In that second-to-last 'graph, I meant "...but only if it complements, not replaces, seniority pay..."
Posted by: J at May 13, 2008 11:08:45 AM
@liberarts
in the UK, the state sponsored school system is rigorously managed at the federal level. There is school choice based on the fact that you can send your kid to any school that you can physically deliver the kid to at the schools designated start time. With the trains and bus systems (not that great, but much better than in the US), this makes it much easier to shuttle kids around to different schools if the one in your city (or village, hamlet), isnt up to par. As I stated above, people can also choose to send their kid to a publicly funded Church of England, Catholic, Baptist, Secular, Islamic, Hindu, etc,. school, but these must meet rigorously enforced mandates and follow a national curriculum. This means that the christian school would need to teach evolution as part of the national science curriculum, but could offer electives for creation science as well for those children seeking to avoid the inconvenient facts that they are being forced to learn in the science class. The very wealthy can elect to send their kids to elite public (e.g. private) schools, such as Eton, which are not regulated like the state sponsored schools.
So, the fact that conservatives argue for local control, and vouchers as the solution to our problems despite evidence from every other country whose children perform better than ours is truly a conundrum.
Posted by: darin london at May 13, 2008 11:15:37 AM
@Darin, for a year I attended St. Michael's Anglican school in Highgate, London back when I was 11 in 1977. It was indeed a fine school. Religion was confined to singing "God Save the Queen" and the school song, "Ye Holy Angels Bright" once a week at an assembly. The kids there were less advanced in math and more advanced in writing skills than my U.S. public school. Another difference was that I won a bottle of whisky as a prize at a Guy Fawkes Day carnival at the school!
Posted by: liberalarts at May 13, 2008 11:28:56 AM
Detroit desperately needs a hurricane.
Posted by: save_the_rustbelt at May 13, 2008 11:53:01 AM
Many people miss the point:
It isn't a choice between public education and vouchers. The public education system is a complete failure in much of the urban U.S... citizens of Detroit, for example, either send their kids to private school (without any vouchers or government assistance), or their kids don't get an education. When you oppose vouchers, you are not saying that you would rather see poor kids better educated in other ways. You are saying that you don't want to fund the education of poor kids.
It is the people in rich white neighborhoods with good schools who oppose vouchers and charter schools, as they don't want their lily white schools "tainted" with inner city black people coming to their rich districts using vouchers.
And here is the REAL reason for public education:
To *PROMOTE* inequality. See, for most Americans, the largest if any investment/savings they have is their home. For most Americans, their home is their life savings. And an absolutely critical element in the value of their home is the quality of the local public schools compared to other schools.
If people could send their kids to any school they want, instead of being forced to attend the schools closest to them, education between municipalities would become more equal... and, as a result of that equality, the differences in property values would drop. People, whose life savings are their home, would lose a big chunk of their life savings. After all, there are some fantastic big homes for dirt cheap in places like Detroit, if the crappy public education system wasn't an issue.
The whole public education system is designed to shaft poor neighborhoods, to prop up property values in rich neighborhoods.
Posted by: Rex Rhino at May 13, 2008 12:01:17 PM
Matt, I'm a Teach for America teacher in a school that has 10 of us plus 2 alumni. Don't take the hype at face value.
Posted by: North at May 13, 2008 12:02:05 PM
J,
you point out many reasons why it's hard to assess teacher quality, and you're correct.
But it's also hard to assess, say, consultants' quality, yet I would not want to run a consulting company if I were forced to pay consultants based on seniority.
Posted by: LemmusLemmus at May 13, 2008 12:12:23 PM
@Rex_Rhino I am willing to pay higher taxes to support the education of poor kids along with my own. I am fairly well-off, and white. I also would would rather have my kids go to school with a diverse racial mix, as I myself benefited greatly from my own public school education with my Asian, Mexican, African, Indian, Native, and other American friends. Living in the UK made me realize the value of a public-funded education system that everyone values. Vouchers are not the solution to getting poor kids into the rich white private schools. These schools will merely raise their rates to ensure an optimal (to them) demographic makeup. What you will see instead is the emergence of a single, monopoly school over which you will have not a wit of influence. Call it the Walmart School, Inc.
Posted by: darin london at May 13, 2008 12:15:42 PM
@liberalarts
Ok so I don't know where you are coming up with the assumption that a full on voucher system would be more expensive per student (tax wise) then the current system. Redistribution only occurs in the isolated rural case that you described. I would argue (again like you without real stats) that this is not the norm. Rather the norm tends to be good rural schools are primarily composed of students from poor backgrounds, and good urban schools are the same. In your example you show a mixing. My own experience tells me this is not the case. My point is the public education system as it currently stands could only be called redistributive by those only willing to look at the relative dollars spent per pupil, and not at the educational product those pupils receive.
@J
You measure teacher performance in the same way you measure performance throughout the public market. That is with your feat. Students know even if the school (and especially the educative system as a whole) is unable to judge. I'm making the assumption that you are teacher yourself. Can you honestly say you don't know tenured teachers who haven't "checked out?" As to English I was referring to the way its taught, and how it tends to neglect writing and grammar. I mean why on earth should highschool kids have to read Dickens, Homer, shakespeare? I liked them, but you can't learn to write by reading him. Shouldn't English classes be concentrated in a more modern vernacular then the archaic realm currently favors? I mean essentially the public education system (the private one to a lesser extent) gave up on grammar as a meaningful field. This is especially exacerbated by the content we make the kids read.
@ Darin
Your assertion that vouchers will accelerate the problem borders on ridiculous. Vouchers would mean rich families wouldn't have to move to wealthy suburbs just for the schools. With vouchers they could send their kids to whatever school district they wanted (and could reasonably get to). If anything I believe the opposite effect would happen.
Posted by: mdesus at May 13, 2008 12:15:43 PM
J,
It is absurd to say that it is too difficult to measure the performance of teachers and so we should give up. How do you measure performance at ANY job? You can't just give the employee a test to see how well they performed on the job. And yet we don't seem to have much trouble figuring out who are the good workers and who can't cut it. I imagine the same is largely true for teachers. Students know who the bad teachers are. Parents know. Everybody knows.
Posted by: Cliff at May 13, 2008 12:16:08 PM
1) The NY Times article refers to 2008 test scores improving over 2007. Hurricane Katrina took place in 2005. Why are so many commenters referring to demographic changes in NOLA before and after the hurricane? Those changes are irrelevant.
2) "Many people do not want their tax money to go to schools that are sponsored by or sympathetic to religions that they don't approve of."
And yet they are currently forced to pay taxes for secular schools, which itself is a religious bent that many don't approve of.
3) "We typically compare our school results with those of other countries, and the outcome is usually that we are falling short. Most or all of the countries have public school systems, so what are they doing that we are not, and if it is our government control that weakens our outcomes, why isn't that a problem in other countries?"
Could the problem be that what our schools teach well isn't measured by international standardized tests? No doubt our students aren't as literate or numerate as we would like them to be, but I would venture they do learn how to work in teams and how to think creatively, which systems in other countries may not do as well.
Posted by: Thelonious_Nick at May 13, 2008 12:17:10 PM
I think it might aid the discussion if the following three questions were being kept apart:
1. Should schools be publicly funded?
2. If so, should schools in richer neighbourhoods get more money?
3. If so, should parents be allowed to send their children to whichever school they like?
My understanding is that the three go together in the US, but this need not be so.
Posted by: LemmusLemmus at May 13, 2008 12:23:58 PM
@mdesus,
"Ok so I don't know where you are coming up with the assumption that a full on voucher system would be more expensive per student (tax wise) then the current system."
What I reported is that a voucher system may reduce the pressure to fund the system, hence it may be less well funded. And again, that is not really my argument, just one that I have heard.
@Thelonious,
"2) "Many people do not want their tax money to go to schools that are sponsored by or sympathetic to religions that they don't approve of."
And yet they are currently forced to pay taxes for secular schools, which itself is a religious bent that many don't approve of."
What I mean here is that many would not like their money to support Madras-style Islamic schools or fundamentalist christian schools that teach that homosexuality is an abomination, etc. Public schools simply ignore
Posted by: liberalarts at May 13, 2008 12:36:06 PM
@mdesus,
"Ok so I don't know where you are coming up with the assumption that a full on voucher system would be more expensive per student (tax wise) then the current system."
What I reported is that a voucher system may reduce the pressure to fund the system, hence it may be less well funded. And again, that is not really my argument, just one that I have heard.
@Thelonious,
"2) "Many people do not want their tax money to go to schools that are sponsored by or sympathetic to religions that they don't approve of."
And yet they are currently forced to pay taxes for secular schools, which itself is a religious bent that many don't approve of."
What I mean here is that many would not like their money to support Madras-style Islamic schools or fundamentalist christian schools that teach that homosexuality is an abomination, etc. Public schools simply ignore
Posted by: liberalarts at May 13, 2008 12:36:10 PM
What you will see instead is the emergence of a single, monopoly school over which you will have not a wit of influence. Call it the Walmart School, Inc.
Um, that is what we have already. The trouble with the arguments of the "schools must be run by the government" zealots is that your worst case scenario of how vouchers will turn out isn't any worse than the situation already is.
The current government-run system is already racially segregated!
The current government-run system is a monopoly!
The current government-run system doesn't give me any say!
The current government-run system creates a huge inequality in education!
And the rich still send their kids to private schools, or to "public" schools that are de-facto private because of the extremely high costs of property and exclusiveness of the neighborhood.
There is no possible way vouchers could make things worse for the poor and the working class. The upper-middle-class and rich, who have half a million or a million plus in their home, and whose property values are propped up by education system inequalities - yeah, those people might be pissed.
Posted by: Rex Rhino at May 13, 2008 1:02:26 PM
More likely explanation: In addition to bad teachers, Katrina swept away bad students.
Posted by: skeptic at May 13, 2008 1:13:37 PM
In Japan, 25% of high schools are private and not free (but almost all lower school students attend public free schools).
There are four types of school in England. All schools can receive funding from the local government. Community schools are owned & operated by the local government. Foundation and Voluntary aided schools are owned & operated by a private foundation. Voluntary controlled are owned by a private foundation but operated by the local government. About 10% of pupils attend privately schools with government grant funding.
In Denmark, 12% of pupils are attending privately owned and operated schools with government grant funding.
In France, 86% of primary schools and 33% of secondary schools are private, but under contract with the state which provides teacher funding.
In Germany, 1% of primary education students attend grant-aided private schools, but 10% of those in Gymnasien attend grant-aided private schools.
In the US, 10% of pupils attend private schools. 1.8% of pupils attend "charter schools" with government grant funding.
Posted by: Mr. Econotarian at May 13, 2008 1:17:13 PM
Back to the fundamental question, here are some data from the 2007 Census for New Orleans.
http://www.brookings.edu/reports/2007/07katrinafreysinger.aspx
Takeaway:
"B. New Orleans’ 2006 post-storm population was smaller, older, more educated, less poor, with fewer renters, and fewer households with children than was recorded in Census 2000."
And:
"C. Compared with “stayers” in the city of New
Orleans, out-migrants were younger, poorer,
more likely to be black, and more likely to
have children."
There is some danger in extrapolating from aggregate characteristics to school enrollment patterns, but this appears to be the most comprehensive data available, since NO school district is not required to report demographic and achievement data for state and federal accountability purposes.
From the data, it appears as though the demographic makeup of the suburbs was largely undisturbed. This would provide a nice research design for analyzing the true effects of the variables extolled by school-choice advocates.
Posted by: Marshall at May 13, 2008 1:34:57 PM
I strongly suspect that even a radical overhaul of the educational funding system will create only a miniscule improvement in any metric we'd care to examine. You can't help kids in spite of their parents or in spite of their social environments. Kids that are interested in learning can do so quite well almost anywhere in the US. We are upset that more kids don't take advantage of the system, that we haven't yet created a system that forces kids to do what's good for them in spite of poor home conditions and in spite of what all their friends are doing.
There's little any public policy can do to address the situation unless you want to send in the government to take these kids out of harmful environments.
Posted by: JasonL at May 13, 2008 1:40:37 PM
What I am arguing is that vouchers will make the overall educational output of our system worse off, and, thus, make our society as a whole worse off. I will actually cede to @Rex that vouchers would be a short term benefit to some poor families over the next 10 years, as the transition from a multi-tiered to a two-tiered education system will take some time to occur. Why do you think the Republicans have been pushing the issue?! They were hoping to use the issue to get a few more people to vote for them against their economic interest, along with the conservative christians, and tax haters. In the long-term, we will all suffer together as the cost of sending our children to the 'right school' becomes prohibitively expensive, while the diversity of schools offering 'alternative education' proliferates. Meanwhile, the sale of government services to the private sector at fire-sale prices will continue unabated. And what little political influence we have over the education of our children will be lost.
Posted by: darin london at May 13, 2008 1:59:29 PM
In New Orleans, poor people lived below sea level, rich people above sea level, so the poor got hit much, much harder by Katrina. When I visited New Orleans in 2007, the parts above sea level looked like a theme park full of stylish scions of old money families while the parts below sea level looked like a depopulated post-apocalyptic wasteland.
So, that appears to be simplest solution for raising school achievement scores -- drive out poor people.
Posted by: Steve Sailer at May 13, 2008 2:00:38 PM
While we await some unassailable evidence on school choice, can anybody tell me why Government schools are better than privatization (while subsidizing those who can't afford schooling, of course)?
BTW I think this emphasis on test scores is wrong becuase I think that they are not testing the right things. To me the schools are more off in what they teach than how well they teach.
Posted by: Floccina at May 13, 2008 2:02:42 PM
LemmusLemmus makes some good points.
The thing that voucher-proponents have correct is that students often get trapped in areas with poor schools and have no ability to leave.
It does not logically follow, however, that all of our problems will be solved with the introduction of vouchers alone. The selection problems that many have mentioned will certainly occur, leaving students without effective parent-advocates to wallow in the (now even worse) public schools.
Fundamentally, the current American public educational system is based on socioeconomic and racial segregation, because it is based on the housing market, housing policies, and the resultant property taxes.
This, however, is not an indictment of public education vs. private. It's an indictment of the U.S.'s particular public education system.
Studies that show great improvement based on school choice programs or "better, more energetic" teachers (i.e. Teach for America) are mixed at best, primarily because it's difficult to properly account for selection issues.
Moreover, many such studies are often based on per pupil expenditure, but they often don't parse our just how that money is being spent (instruction vs. other needs), making it seem that areas that use large amounts of money for non-instructional purposes that suburban schools generally don't require -- i.e. urban (infrastructure costs, securty), rural (transportation), etc. -- are actually spending more money per pupil on instruction than the are. Plus, some school districts are simply broke. (Witness Detroit's employment of legions of long-term substitutes who are cheaper than real teachers and don't really teach, but who do prop up the student-to-teacher ratio numbers).
So, the ideal solution certainly has to involve some equalization of funding and, in the long run, instructor and facility quality. To do that may involve spending more money in needier areas (to attract good teachers willing to go into lousy environments, etc.), which is generally the opposite of what happens now, despite a few notable exceptions that are often incorrectly held up as the rule.
A perfect implementation of the above would make school choice unnecessary. That's unlikely, obviously, and there's no reason to believe that real change might not involve both. (There's something to be said, however, for the benefits of making sure that high- and low-achieving students are alway mixed, even if it smacks of social engineering).
American education does need fundamental change, but it's not going to happen by de-funding the public schools or by breaking the teacher's unions. (I've always found the latter to be particularly funny, since the proponents of that saw we need better teachers, which involves attracting more people who could do other things, which involves making teaching more attractive and better paying, which is precisely what the teacher's unions have done. If teaching isn't an attractive job now, imagine what it would be with no unions, lower pay, and lower job security.)
Finally, JasonL hits on some fundamental points, even if he's maybe a little too pessimistic. Much of what determines student achievement happens outside of school. Lousy schools could be better and that would help many, but it wouldn't solve everything. Ultimately, larger issues of inequality seep into the schools and can't easily be undone by the educational system -- no matter what politicians like to claim
Posted by: J at May 13, 2008 2:17:27 PM
mdesus and Cliff,
Essentially, you're both just articulating a pretty basic anti-union -- anti-employee input, workplace democracy, or whatever-you-want-to-call-it -- attitude.
What does Cliff's comment -- "Students know who the bad teachers are. Parents know. Everybody knows." -- mean exactly? What about students who like easy classes. (Most students today?) They'd surely give poor ratings to the difficult math or history teacher. Who knows, they're parents might (probably would?) take their child's side. Let's even assume the school has a good principal who disagrees and thinks the teacher is great. Who wins that debate? Does the teacher get fired? A raise? It would depend on who was making the decision.
It's one thing to say that someone in authority -- principal, PTA, school board, whatever -- has the ultimate say over who is a good teacher or not, regardless of the consequences, but that's not really advocating meritocracy or "performance pay." It's just shifting power from teachers to someone else. Period.
Simply because bosses in most business environments have the right to fire non-union employees at will -- even for petty or wrongheaded reasons -- doesn't mean that it's the right model for that workplace, or that it would make schools better. Education isn't as easily quantifiable as other things.
Assuming that union-based seniority isn't perfect, either, (it isn't), the choice is then between a system that gives employees more protection from arbitrary power and one that gives them less. Certainly when it comes to schools, I'd prefer the former.
Once again, if the goal is to recruit better people to be teachers, undermining their security and autonomy (two of the few clear benefits to teaching today) isn't one of the answers.
Posted by: J at May 13, 2008 2:29:54 PM
I second J on this. I spoke with someone about charter schools in my area. The person didnt like them because they were a teacher who initially went to work for one of them and found that it equated to more stress and job pressure with less benefits, much less security, and only slightly better pay (for now, though wait until the public school teachers get a 7% raise next year while the charter school teachers get 1%). Her position was obviously biased, but only in a way which reinforces what J is stating. Charter schools are designed to undermine the unions, and make teachers subject to the same pay and benefit regimes as walmart employees. How are we going to attract the best and brightest this way?
Posted by: darin london at May 13, 2008 2:50:05 PM
I can honestly say that it has been the observation of every teacher I have known who has been at it for 20+ years that a combination of bad parenting and administrative capitulation to unreasonable parent demands (typically in the form of assigning higher grades to undeserving students) has virtually destroyed the public school system. My father, a former teacher, now retired, likes to tell a story from his childhood of a teacher calling home to say he wasn't doing his work (an assessment with which he disagreed) and coming home to be disciplined with a belt by his father, no questions asked. As he put it, "I doubt there are any parents like that left."
Posted by: Rock On at May 13, 2008 2:50:52 PM
"By the way, here is a related question. We typically compare our school results with those of other countries, and the outcome is usually that we are falling short. Most or all of the countries have public school systems, so what are they doing that we are not, and if it is our government control that weakens our outcomes, why isn't that a problem in other countries?"
In many other countries, the students are separated according to ability early in their academic careers and moved into different tracks: High School, trade school, etc. As a result, many of these tests compare the brightest Japanese or German students with all of the American students. Those who quote the difference in test scores are not compares like things.
It is the same way with exchange students. Some people marvel at how much brighter the exchange student is over the American students. This is because dumb students are rarely exchange students
Posted by: Eric of PA at May 13, 2008 3:09:32 PM
What you will see instead is the emergence of a single, monopoly school over which you will have not a wit of influence.Errr, are you kidding? Because people have a lot of influence when they're not being forced to pay money, but instead can simply use their money at other schools.
The last time I checked, the public school system is a textbook example of a monopoly. And without coincidence, it has the same problems as other monopolies.
I always find funny that liberals come off as anti-choice when it comes to schooling.
How are we going to attract the best and brightest this way?I could simply say, "if we wanted the best and brightest," we'd pay more for them (with vouchers and a free market), but that'd be too simple and quite frankly that doesn't highlight the problems in wanting to pay more for every teacher.
See Akerlof's Market for Lemons. I'm not going to explain it because I trust that you know it. Note that the same situation is applicable in hiring teachers. Think about it. How do you know that the person you are hiring is a good teacher? You don't!
So schools have to use signaling to make sure they get the best... but there's actually no difference in performance between people who have MA's and BA's, despite that the public school system often pays people with MA's substantially more. I skimmed through an NBER report on it, but I can't find it as of now.
See, just because you know how to, oh, derive Schrödinger's equation from the Feynman Path Integral, it doesn't mean that you can intuitively explain Newton's Laws of Motion in a way that an eighth grader could easily understand and relate to, and then run a grade school class efficiently. And even if you can, it certainly doesn't mean that you can do it better than anybody else. There are plenty of intelligent people who make bad teachers, many times because they "just get it" and can't teach it to students who are extremely slow learners. So was it this blog or Freakonomics where the post writer hired the guest speaker and that speaker delivered one of if not the best lecture the writer had ever seen?
Paying more only attracts more fodder that needs to be sorted through and quite frankly may be chosen. Plus, who's to say that the "best and brightest" deserve to shine by teaching students? Bright people are just as useful doing things that pay more money, you know. Generally, I'd say that's why they get paid more money to do those things!
Posted by: Daniel Reeves at May 13, 2008 3:59:45 PM
I love how the article leaves out all of the important statistics.
Keep in mind I haven't yet accessed the original source data, so my assumptions first:
assumptions regarding data:
The school population of 32,000 is evenly distributed 1-12 - meaning 2700 per cohort. The 2016 and 2012 New Orleans student cohorts therefore together contain roughly 5500 kids.
"More than half failed this year" so my estimate of the failure rate is 55%. If it was 60%, the wording of the article would have been "six in ten", and if it was 50%, the wording would have been "half". A failure rate of 55% indicates roughly 3000 of the hypothesized population failed, and approximately 2500 passed.
Another assumption: that the relative passing proportions in cohorts 2011 and 2015 were equal, and that cohorts 2012 and 2016 were of equal size to cohorts 2011 and 2015, meaning that the cited 12% and 4% growths in passing rates could be combined in the equation:
1.12X+1.04X=2500
yielding an estimate of the numbers of students passing in 2007 of ~2300. Roughly 200 more kids passed the exam in 2008 than in 2007. The probability that a cohort 2011 and 2015 student passed was ~42%, and the probability that a cohort 2012 and 2016 passed was ~45.5.
Given the following assumptions:
1) together, the 2008 fourth graders and eighth graders (cohorts 2012 and 2016) belong to the same overall population as the combined 2007 4th and eighth graders (cohorts 2011 and 2015).
2) That cohort 2012 is sampled from the same population as cohort 2011, and that cohort 2016 is sampled from the same population as cohort 2015.
3) Cohorts 2010, 2011, 2014, and 2014 come from the same population, so estimates of the population probability of passing can be compared.
3) That the probability of a student passing the exam can be modelled as a binomial probability distribution, and that the observed proportions can serve as estimators of the actual probability.
4) The tests were invariant between 2007 and 2008.
More in a bit.
Posted by: The New York City Math Teacher at May 13, 2008 4:08:53 PM
"In many other countries, the students are separated according to ability early in their academic careers and moved into different tracks: High School, trade school, etc. As a result, many of these tests compare the brightest Japanese or German students with all of the American students."
As far as Germany in the PISA and TIMMS studies is concerned, this is incorrect. That was a representative sample of classes in all tracks.
By the way, some people advocate tracking as a remedy for the problems of the American school system. That's sort of funny because Germany didn't do too hot in the PISA studies and lots of people blame that on tracking. They point to Finland (which seems to come out as the winner pretty much all of the time), which has a one-track system.
Posted by: LemmusLemmus at May 13, 2008 4:12:00 PM
Check out this better article from the Times-Picayune:
Relevant quotes:
'Although substantial changes in the schools and student populations during the past two years make direct comparisons or decisive conclusions difficult, Recovery School District Superintendent Paul Vallas said the scores represent "consistent growth across the system." He credits smaller class sizes and an energetic pool of new teachers.'
'...Vallas acknowledged, however, that apples-to-apples comparisons with last year are difficult because of the large influx of new students in district-operated schools... Given the instability New Orleans children faced in the two years after the storm and the turmoil as the Recovery District opened in the fall of 2006, some test score gains were to be expected.'
More evidence, please.
Posted by: Algernon at May 13, 2008 4:20:56 PM
@J
I think it's safe to say that the teachers union is detrimental to public school education. Other than that I disagree entirely with your sentiment. Especially the idea that students prefer easy teachers. Students prefer good teachers. I think you miss interpret my stance as being opposed to things rather than for something other than the status quo.
@Daniel
Take a basic economics course bud it would help. Paying more (fiscally or in social recognition) is the way every industry in the world attracts talent. Want better people? Pay them more, and the better people will come.
@Everyone
How is it possible that on an economics blog run by two libertarians that (what I'm assuming) the regular readers differ so greatly from both economic theory and libertarian ideals? I had just assumed everyone on this site agreed unions, in todays market, where citizen worker rights are largely guaranteed, are bad and stifle competition. Strange...
Posted by: mdesus at May 13, 2008 4:20:59 PM
From the Times article... "Classes are smaller, many of the teachers are youthful imports brought in by groups like Teach for America, principals have been reshuffled or removed, school-hours remedial programs have been intensified, and after-school programs to help students increased... Mr. Vallas attributed many of the improvements in testing to the new teachers. 'The biggest contributing factor was the quality of the instructors,' he said."
Smaller classes. Young, energetic teachers. More remedial programs. More after-school programs. Can't goverment do this? If the answer is no, why not? If the answer is yes, why hasn't it?
Posted by: Cougian at May 13, 2008 4:55:41 PM
@The New York City Math Teacher, RE: point #3:
Your choice of probability distributions seems wrong. Testing a group of kids is not sampling from the same random variable (Bernoulli) many times but is instead sampling from many different (independent?) RV's once. This seems to lend itself more to a Gaussian.
@mdesus:
Not everyone! :-p
Posted by: Matt at May 13, 2008 5:42:06 PM
Not sure if this has been mentioned, but could it be possible that the extra dollars flooded (pun not intended) into New Orleans after Katrina by the Bush Administration be the reason for the improvement? School funding, or per pupil spending, is the largest determinant of student outcomes, and subsequently, test scores. The UK's charter school reform in the 1980s had a one time rise in test scores, which research suggests was due to the firing of problem teachers and additional funding to the newly chartered schools and afterwards those charter schools remained about on track with other public schools.
Posted by: Axis at May 13, 2008 6:00:43 PM
The "Teach for America" program isn't driven by high minded altruism from the students it recruits. These kids come from elite programs, they're not going to be career teachers on a scale you can implement on any large scale. Just another status game to demonstrate to employers that they are willing to toil for long hours in adverse environments before making big bucks at BIGLAW.
Posted by: Billare at May 13, 2008 6:23:30 PM
I'd like to add a few experiences from the Netherlands, since we have a system that includes what you would call 'vouchers' and 'tracking'.
Primary and secondary education is partially organized by municipalities, and partially by private groups: usually religions or people with quasi-experimental schooling ideas. Both get the same amount of government money per pupil, pupils from immigrant backgrounds will get you a bit a more. Curricula are state-determined, but at primary school level there is a lot of freedom left.
But schools are prohibited to demand money on top, they are either fully subsidized by the government or fully paid by parents. As a result, privately financed schools are very rare. This has wide support, from basically all layers of society. ( By the way, schooling is mandatory. Permission for home schooling is hard to get)
Inner city schools with high percentages of immigrant children have a bad reputation, although I really can't judge how they would compare to American inner city schools.
Tracking is a major part in Dutch secondary schooling, and starts from age 12. Its net effects are very hard to judge. From a social mobility point of view for example: On the one hand, it leads to a separation where children of well to do and well educated parents are much, much more likely to end up in the higher levels of secondary schools.
On the other hand, good test results and good marks will move children from all social levels in the higher levels, and once in these levels, moving on to good tertiary education is expected from everyone. For smart children with parents with little education, this is a large benefit, as they will start university from a very similar starting point as other students.
Posted by: greatzamfir at May 13, 2008 6:28:04 PM
Matt - you're absolutely right, the real measured variables don't follow Bernouillian distributions, but we don't have real access to the real measured scores. The only data available is multinomial relative frequency data of test score category in percents.
But the magnitude of whatever performance gains can be estimated, and it's not large. 200 extra passes in 5500 - ~3.8% - is in the noise of all the other variance a big city school district can provide, not to mention potential variability in testing.
I'd actually want to do a chi-square goodness-of-fit on the multinomial probability distribution of the five categories of performance on the tests, but the only data actually available is relative frequency in percents, so back to estimation again.
, and perhaps afterwards do an ANOVA on the t
Posted by: The New York City Math Teacher at May 13, 2008 7:07:56 PM
And the Bush administration stepped in with millions of dollars ... Guess what? It's working.
So, Alex, are you now firmly in favor of federal takeover of schools? Or just funding?
Posted by: Eric H at May 13, 2008 9:58:18 PM
I'm always amazed that so many people will predict dire social outcomes for vouchers, and use this as an explanation for why we mustn't use them even to replace, say, the nightmarishly bad DC public school system. Because hey, what bad effects could shorting thousands of poor black kids out of a decent education possibly have?
Posted by: albatross at May 13, 2008 10:51:39 PM
J: Once again, if the goal is to recruit better people to be teachers, undermining their security and autonomy (two of the few clear benefits to teaching today) isn't one of the answers.
But that seems to contradict the more recent party line pronouncement from the teachers' unions that you can't evaluate teachers based on the performance of the students. In fact this principal has been written into law in New York State.
Since the current educational theory among our esteemed pedagogues (at least when they are negotiating labor contracts) is that teaching inputs have no effect on educational outputs, there does not seem to be any point in recruiting better people to be teachers. In fact, teachers' unions seem to be arguing just the opposite. Their rabid opposition to performance pay incentives and support for small class room size seems to argue that the most cost effective schools would have maybe a dozen kids in each class room taught by a minimum wage teacher.
Posted by: Mark in Texas at May 14, 2008 6:26:34 AM
Mark In Texas, i see no problem with believing that good teachers matter, while simultaneously believing that their quality is hard to measure, and that the measurement itself would distort education to produce worse results. I am not sure if it is true, but I do not think it contradicts itself.
If the results of the measurements are only a weak proxy for quality, then it is possible that the improvements from selection effects are more than compensated for by the quality decrease from teach-to-the-test behavior.
Posted by: Great Zamfir at May 14, 2008 7:30:52 AM
@Zamfir:
What is the evidence that teaching to the test produces a decrease in quality?
Posted by: Cyrus at May 14, 2008 8:18:06 AM
"quality decrease from teach-to-the-test behavior. "
Actually understanding English and Math is a negative in some places?
Posted by: Tom at May 14, 2008 9:00:37 AM
I am not saying teaching to the test is necessarily bad,it can go both ways and I really don't know.
But IF you think it is bad, as the teachers apparently do, then it is absolutely possible to oppose performance-based pay, even if you believe it would raise the average quality of teachers. Especially if you think, as they apparently do, that performance measurements will be weak proxies for actual quality, in which case performance-based pay will not raise teacher quality by much.
And note that these two opinions go together: if you think that teaching to the test is good, because you think that the tests do cover the most important things to teach, then you will also believe that performance-based pay will be effective. And vice versa.
Posted by: Great Zamfir at May 14, 2008 9:53:25 AM
So far the best argument that I have seen for vouchers has been that it ca not be any worse than what we, in America, already have, so why not let private companies profit off of it, they might make things better. Some of us believe that the economics of mass education preclude the generation of sufficient profit for private firms in any way that would bring about the improvements that all of us are wanting to see. We then point to evidence from other countries whose students do better than ours across the board. These countries generally do not have the decentralized, privately funded, non-unionized system that others argue would improve our school system. We wonder what is wrong with this evidence?
Posted by: darin london at May 14, 2008 9:56:06 AM
Teach this:
http://www.hodakvalue.com/blog/2007/03/teach_for_america_reality_and.html
Posted by: M. Hodak at May 14, 2008 10:09:05 AM
@Cyrus: Good question. Let's set them a test, then we can find out.
Posted by: Sam B at May 14, 2008 11:07:25 AM
One thing that sticks out by its absence in these debates is the MASSIVE differences in student cultures between socio-economic groups (even beyond race, although particularly in the inner-city these tend to blur together as criteria). Like a number of commenters have mentioned, public education naturally tends to segregate communities from each other. Some districts have bussing, and at least back in my hometown of Charlottesville, it was possible if you could find transportation to swap schools for a little bit of a premium yearly to try and get to one of the better districts, at least at the high school level, but mostly people are forced into one particular set of schools based solely on geography.
However, this strays from my original point; when comparing the United States and our test scores, one major thing is the vast range of "diversity" that our school system embraces, and the regional differences, as well as the completely different mindset of Americans towards education in general. We are widely divided as social groups as to our attitudes; some say it is self-improvement, others feel that is unnecessary for whatever reason...
Posted by: Neal at May 14, 2008 7:27:49 PM
Billare, I dare you to come to my professional development meeting tomorrow and say that to us in person.
I'm really sorry to have to tell [almost] everyone on this thread that you have no idea what you're talking about. For that reason, I encourage you to stay out of the next conversation about public schooling you run across.
Posted by: North at May 14, 2008 8:55:49 PM
Billare, I dare you to come to my professional development meeting tomorrow and say that to us in person.
I'm really sorry to have to tell [almost] everyone on this thread that you have no idea what you're talking about. For that reason, I encourage you to stay out of the next conversation about public schooling you run across.
Posted by: North at May 14, 2008 8:56:10 PM
