Charter High Schools Increase Earnings and Educational Achievement

by on February 3, 2014 at 7:25 am in Uncategorized | Permalink

Private and charter schools appear to have significant but modest effects on test scores but much larger effects on educational attainment and even on long-run earnings. A new working paper from Booker, Sass, Gill and Zimmer and associated brief from Mathematica Policy Research finds that charter schools raise high school graduation, college enrollment and college persistence rates by ~7 to 13%. Moreover, the income of former charter school students when measured at 23-25 years old is 12.7% higher than similar students. Similar in this context is measured by students who were in charter schools in grade 8 but who then switched to a traditional high school–in many ways this is a conservative comparison group since any non-random switchers would presumably switch to a better school (other controls are also included).

The effect of charters on graduation rates is consistent with a larger literature finding that Catholic schools increase graduation rates (e.g. here and here). I am also not surprised that charters increase earnings but the earnings gain is surprisingly large; especially so when we consider that the gain appears just as large among charter and non-charter students both of whom attended college (i.e. the gain is not just through the college attendance effect).

I wouldn’t bet on the size of the earnings effect just yet but what we are learning from this and related research, such as Chetty et al. on teachers, is that better schools and better teachers appear to have a significant and beneficial long-run impact that is not fully captured by higher test scores.

As I said in Launching, one of the factors that makes me optimistic about education in the United States is that it remains relatively decentralized and open to experimentation and evolution.


1 prior_approval February 3, 2014 at 7:58 am

‘…in many ways this is a conservative comparison group since any non-random switchers would presumably switch to a better school’

Like when a parent loses their job, for example, and the family moves elsewhere.

Best satirical site on the web.

2 Ninth February 3, 2014 at 9:42 am

Um, I believe those would be random switchers…

Best boneheaded comment of the post.

3 LemmusLemmus February 3, 2014 at 11:00 am

Random? Really?

(Also, I have the same suspicion as Roger Sweeney below.)

4 KLO February 3, 2014 at 9:42 am

It is as if the authors do not know that 1 in 4 kids who starts the year at a low performing school does not finish the year at that same school and that the “switchers” group includes most of the lowest performing students in any school. Indeed, you could dramatically improve the quality of a school by allowing students to enroll only at the beginning of every other year. This has been adopted as a best practice among the highest performing charter schools.

5 Roger Sweeny February 3, 2014 at 10:43 am

Actually, many non-random switchers will be kids who find the charter too hard or too much work. Most of the charters that improve student performance have longer school days and longer school years and work their students harder. It is hardly a surprise when kids who deliberately switch out of that are less successful later in life.

6 Strick February 3, 2014 at 8:16 am

I’m not sure this is really measuring teacher or school performance. While there are certainly bad teachers and bad schools, if you talk to educators at school neither at the top nor the bottom of the curve where student performance is mixed, they’ll tell you that they see the crucial difference is the interest a student’s parents take in their work. Students whose parents respond to changes in grades or behavior issues invariably perform better shortly afterward. They’re not always the most gifted students, but they get feed back and encouragement.

So the charter boost might simply be a case of parents who care getting their students as a group to better schools. The real boost might come from a combination of factors: an achievement feedback loop (success breeds success), peer pressure as the students around them begin to expect to succeed, and the values the parents instill in their children in the process of sticking to it and overcoming obstacles.

7 Elwin February 3, 2014 at 8:43 am

Exactly. If you don’t control for parental involvement in the student’s education you are missing a huge factor.

8 Seb February 3, 2014 at 8:48 am

That is the point of using the control group of students who were also at charters in grade 8.

9 steve February 3, 2014 at 8:57 am

So what they did is select out for parents who were committed to keeping there kids at charter schools vs parents who found it too much of a hassle?


10 NPW February 3, 2014 at 9:53 am

Why does it really matter? To say charter schools attract a higher ratio of engaged parents than public schools do, seems to be somewhat trivial in this context.

11 dead serious February 3, 2014 at 10:09 am

Does it, if you’re going to claim that “charter schools raise graduation rates”?

Seems more plausible that parents who care about their kids’ education – the kind who try a new school setting when the current one isn’t working for their kid – make sure their kids graduate high school.

12 NPW February 3, 2014 at 12:46 pm

Don’t they raise graduation rates simply by being an option?

Unless Alex is arguing that there shouldn’t be non-charter schools, which if he is my apologies, isn’t it a good thing that option B is raising grad rates?

Does it matter if charter schools are “better”? They work for some kids it seems. Perhaps the statics in comparison fail to account for parental involvement. I don’t care, and I don’t see how it matters. Unless someone is campaining for ending non-charter schools, this is just one schooling option that appears to be working.

13 JosieB February 3, 2014 at 3:33 pm

I think we underestimate parents’ concerns about their children’s education. Newark, NJ public schools have been wretched for many years (although an energetic new superintendent is trying to make wholesale improvements.) In Newark, 8,000 students are enrolled in charter schools, and another 10,000 are on charter school waiting lists. No doubt some of the charters are better than others, but the important thing may be the message charter parents give their children — I chose this charter school for you because I believe it is a good school where you will learn — is powerful. The alternate message — you have to go to this public school because it’s in our neighborhood — does not convey parental faith in the school (and why should it, given the schools’ records?) Clearly, at least in Newark, many parents want to send the former message.

14 Rahul February 3, 2014 at 8:19 am

Do Catholic schools still increase graduation rates when controlling for things like family income, parental education, etc.?

15 Seb February 3, 2014 at 8:47 am

Yes, no econ paper would be publishable without controlling in some way for those elements.

16 Urso February 3, 2014 at 10:11 am

Did they control for Catholicism? (not a joke)

17 LemmusLemmus February 3, 2014 at 10:15 am

I believe I’ve read a paper that actually does this. No link, though. Might have been coauthored by Christopher Winship.

18 Marie February 3, 2014 at 11:00 am

Oddly, there’s a ton of nonCatholics or not practicing Catholics in many parochial schools these days.

19 Urso February 3, 2014 at 12:00 pm

Always has been true in my experience – heck I went to high school with a few Hindus (who are rare as hen’s teeth around here).

20 NPW February 3, 2014 at 12:48 pm

I have devout muslim friends. They’re sending their two sons to a Catholic school.

21 Emily February 3, 2014 at 12:01 pm

You have to have some controls but not necessarily good ones. The income variable here is just whether the student got free or reduced-price lunch and there isn’t a parental education variable. I’m sure they would have used a more granular income variable and a parental ed one if they had it available, but they’re probably working from some administrative data set that doesn’t have that stuff.

22 ummm February 3, 2014 at 8:55 am

Nothing too special about attending college. Most will gladly take our money. completing , on the other hand,…

23 steve February 3, 2014 at 8:56 am

Doesn’t look like they control very well for selection bias. Also, just lie a lot of other studies, they mostly look at the characteristics of the students, and very little at the parents and extended family, which are probably more important in many ways.


24 NPW February 3, 2014 at 9:50 am

Like everything else in econ, there are too many interconnected factors for a definitive answer. However, even if the only thing charter schools do is sort engaged for disengaged parents, isn’t it useful to point to that as a factor in educational outcomes? As opposed to more money for teacher’s unions? If, for instance, grouping children who have parents who are actively engaged in their kid’s education in an environment with rules that are actually enforced produces better results than more funding for bureaucrats, then perhaps the idea that, on average, parents and their kids should be held responsible for their education is an idea that holds merit. And perhaps throwing more money at the Dept of Education isn’t the answer.

Selection bias could be the very point. Unless we are comfortable with requiring a reproductive license and a crèche system, I don’t see divergent outcomes changing. I’m willing to accept that there will be failures in the system, but I’m not willing to accept that because some of us are losers we all have to be to make it “fair”.

25 Marie February 3, 2014 at 10:21 am

“However, even if the only thing charter schools do is sort engaged for disengaged parents, isn’t it useful to point to that as a factor in educational outcomes? As opposed to more money for teacher’s unions?”

So yes.
But I’d make an edit, and say charter school sort for allowing parents to be engaged.
Public schools spend a lot of time undermining parent involvement and then complain that they can’t be held accountable for teaching kids whose parents don’t care about their education.
The answer is not to “hold parents more accountable”, it’s to stop disincentivizing and punishing parents who naturally want to be accountable.

26 NPW February 3, 2014 at 12:30 pm


Let’s assume charter schools have better outcomes because they have more engaged parents and that public schools would have more parents engaged if they encouraged parents to be involved. How specifically would you encourage parents to be involved? Or at least not undermine and disincentivizing them?

Would you, if you had children and the means to do so, put them in a public or a charter school? And why?

I have a three year old, and I wonder what we will do when she is kindergarten age. Neither my wife or I have particularly fond memories of education prior to college, and grade school education has become an topic of interest for us.

27 Marie February 3, 2014 at 1:12 pm

I was a teacher and sub teacher in two states and we tried three different schools in two communities before moving to home education, which works for us. So my opinion may not be useful to you.

Most parents want to be involved in their kids’ school day, although not all are able to be. The schools simply need to allow it, but because it is messy to have a hundred bosses looking over your shoulder while you’re doing your work most de facto discourage it.

How can schools reverse that? Hard. In brief, some successful strategies I’ve seen are

1. Magnets and charters that don’t use much busing — parents are then at campus usually twice a day.
2. Programs like a 20 minute start of day session where parents read with their kids, all over the hallways and classrooms, school wide.
3. Smaller schools with more local control (no guarantee).
4. Lower administration ratio (top heavy staff tends to folks with more degrees and less experience, and higher salaries so more to lose, so that encourages the self-protective attitude that parents shouldn’t be involved because they don’t have degrees in early childhood development).
5. Local hiring — staff already have connections with the community.

Biggest litmus test, though — before your kid is enrolled, ask to be able to drop in on two or three random days and sit at the back of several classrooms for the full day. If you’re denied and told to come when all the other parents come, at kindergarten roundup, you’ve got a pretty good clue that they consider parents adversaries instead of allies, distractions and disruptions instead of helps.

If you have the luxury, get a substitute teaching license and spend some time sub teaching in the district you are favoring. No better inside view than that!

28 dead serious February 3, 2014 at 10:28 am

Your conclusion is that we should stop throwing money at teachers unions because possibly ineffective but rather throw money at charter schools with possibly same outcome?


I’m not agreeing or disagreeing that charter schools are a good thing – in general I think it’s good to have different approaches to education as what works for one kid won’t necessarily work for another.

However, this isn’t clear evidence – to me, anyway – that charter schools in and of themselves have better outcomes.

29 Jay February 3, 2014 at 11:58 am

On average don’t charter schools cost significantly less per student? So yes would be my answer to your question dead serious.

30 dead serious February 3, 2014 at 1:08 pm

I generally prefer not to throw money at anything not proven effective, but I guess I’m in the minority.

And to NPW below, if you’re not throwing money into the public ed system, your other option is the charter route which, if this self-selecting study is the only evidence in its favor, isn’t exactly ironclad.

Again, I’m not opposed to charter schools, but the comparisons to its performance vs the charter universe are not apples to apples comparisons.

31 NPW February 3, 2014 at 1:52 pm

dead serious,

“if you’re not throwing money into the public ed system, your other option is the charter route”

This is incorrect, and if you read what I wrote you would have seen that I don’t view the options as binary. The solution does not necessarily have to be more funding for anyone. I didn’t suggest any more funding for the charter route, in fact, I specifically suggested a non-monetary solution.

“I generally prefer not to throw money at anything not proven effective, but I guess I’m in the minority.”

If you think that throwing money at non-charter public schools is a good thing, you most certainly do believe in throwing money at something not proven effective.

“comparisons to its performance vs the charter universe are not apples to apples comparisons”

And again, like I said in the first post econ cannot provide apples to apples, but there are useful observations from original post from Alex.

32 Brandon February 6, 2014 at 5:39 pm

not when you control for factors such as special education. Charters don’t take the higher-cost students.

33 NPW February 3, 2014 at 12:21 pm

dead serious,

I went back through what I wrote. Didn’t see the part where I arrived at the conclusion to throw money at charter schools.

34 Jan February 3, 2014 at 9:03 am

More research is always better, but a couple pieces of info from my sister’s experience having taught in both public ed and at a charter.

Most of the “problem” kids that go into charters are eventually pushed out to public schools. That could be due to behavior problems, lack of parental engagement, excessive absences, lack of school infrastructure to accommodate certain disabilities, etc.

Many of the charter students with learning disabilities that aren’t severe enough to kick out are never formally designated as disabled. This is because the schools are by and large run by for-profit companies (at least in Michigan, not sure about elsewhere) and they simply do not want to pay for the services these kids are required by law to receive. They are usually pushed through the system, even if they are not thriving. I do see the analysis attempted to control for learning disabilities, though I think the differences in prevalence and severity of the disabilities are often not well-incorporated into these papers.

35 Rahul February 3, 2014 at 9:17 am

Does the “services these kids are required by law” apply to Charter schools too or only public schools? If it does, wouldn’t it make sense to fix a higher per student funding rate for every such disability student?

36 Marie February 3, 2014 at 9:59 am

A U.S. charter school is a public school. They are within the public school system and fully funded by public funds, so, yes, they are entirely required to follow all the laws about services to kids with disabilities of all kinds.

37 Jan February 3, 2014 at 10:04 am

But unfortunately they don’t always follow those laws and especially in poorer communities, the parents aren’t totally up to speed on their rights or vigilant in demanding those services.

38 Marie February 3, 2014 at 11:03 am

The charters we have here are usually grassroots, and almost never corporate run, so I can’t speak to them dodging the law, I don’t doubt it.

I have seen plenty of dodging of the law in public non-charter schools, though. It seems to be the norm, rather than the exception, that a parent of a kid with almost any disability needs to fight for the provisions the law requires, and even then sometimes doesn’t receive them.

39 Jan February 3, 2014 at 11:30 am

I’m sure it happens in public schools, too, and I don’t envy the parents who have to fight for their kids to get basic services.

I was curious so I googled it and Michigan apparently leads the country in for-profit charter schools. I think the incentive to spend less and return a profit is what complicates it at some charter schools.

40 Rahul February 3, 2014 at 1:28 pm

According to Wikipedia the term “learning disability” covers problems not severe enough to warrant an official clinical diagnosis. Is this really right? How flexible is the term?

41 Marie February 3, 2014 at 8:14 pm

Oh, that’s a can of worms.

It is not a diagnostic term in itself, to my knowledge.

I could give you a million examples of the vagaries of the system, there are many levels of accommodation for differences that range from true medical disabilities to things that I think are best described as learning differences, but differences that are so hard to address in an inflexible classroom as to be impeding to learning.

42 Rahul February 3, 2014 at 11:14 pm

@Marie: Thanks. In which case I don’t blame the charter schools for sometimes not paying for services demanded by parents. Anywhere that there’s a vague diagnosis guideline there’s bound to be disagreements between parents & schools over what needs a special service & what doesn’t.

43 Marie February 4, 2014 at 9:01 am


Oh, that’s a huge part of it in all the public schools.

This is laced with my personal opinion, but human beings vary. One room with thirty kids is going to have thirty different issues in learning, and those will be on a spectrum.

If you have a very structured, rote style learning, most kids can adapt to that, even if they don’t flourish under it.

But when you have the flavor of the minute in teaching styles and curriculum, there’s no consistency in behavioral expectations or learning atmosphere, these differences are going to make for chaos and some kids can’t learn.

So you get IEP, which I think are now ILPs, all these different individual plans that are opt out instead of opt in in the details, and are largely about CYA for the school when the kid can’t flourish.

To be fair, I think Jan was talking about disabilities anyone would recognize, that take one on one care, etc. But in my very, very limited experience in a far different state than hers, it seems to me that non-charters are just as likely or more to fight accommodations (or just neglect them until they are forced to provide them) as charters. Charters, being smaller, are generally more flexible, and they get as much money per pupil (in our state) as the other schools do. The post from Mr. Sewell belong seems to back me up on that.

44 Jan February 3, 2014 at 10:02 am

I am not an expert, but at least in Michigan I think there seems to be an incentive to say kids are too disabled and must be bussed to a special ed campus or that they aren’t disabled. You’re right there should be a higher payment for kids with special ed requirements. But I’m y thinking the schools must be paid an unadjusted rate for the total number of pupils (that is supposed to include the funds for special services, or the schools don’t see that bump in money until the next year, making finances for the immediate year the primary concern.

45 Marie February 3, 2014 at 11:05 am

I don’t think they even have separate campuses of that kind out here. I can’t see parents putting up with something like that.

Michigan does not, hope you don’t mind my saying so, seem like a pleasant place to live if you aren’t rich and healthy.

46 Jan February 3, 2014 at 11:25 am

Ha. Well there is a budget surplus in Michigan! It’s all great since the legislature cut their way to freedom and clamped down on unions. I don’t live there anymore, but seriously it has been a rough few years.

They definitely bus some disabled kids to special schools after they try the charter. As I understand, it is most common for students that are low functioning with autism spectrum disorders. Sometimes their parents or even their docs think they can cut it in regular school and suggest they make a go of it. If the school doesn’t provide a great deal of expensive special services for the student it usually doesn’t work out. Often the administrators “strongly suggest” they head straight to the publicly-run special ed campus. I’d actually be surprised if other states don’t have similar arrangements — after all, there are some severely disabled students who can be much better accommodated outside a regular school, where the services are really targeted for similar students — but maybe I’m wrong.

47 Thomas Sewell February 3, 2014 at 3:40 pm

States vary a lot because of funding formulas, etc…, but I ran a Charter School that had twice the state average in percent of special education students (and twice the state average in gifted as well, we really flattened the bell curve) and the biggest issue we ran into was that the State cut off the “extra” funding for SPED students after 20% of enrollment unless you were a school specifically designed to handle SPED only.

Now, we did individualized instruction and mentoring for all of our students, with the ability to progress at their own pace, which explains our student distribution, but I never saw anyone in our State trying to push students out to other schools. Assuming your Charter School can handle the students, the money from additional students was always preferred over trying to use student population to influence test scores. At the end of the day, test scores just aren’t as important as community reputation and making parents happy with what you’re doing with their specific students.

48 Marie February 3, 2014 at 8:15 pm


49 S February 3, 2014 at 9:08 am

Are we talking about self selected treatments?

50 Marie February 3, 2014 at 10:08 am

1. Yes, it makes a huge difference that the parents of the kids in charter schools are more involved in education — usually, and a big hurdle, the parents have to drive the kids to the charter school. Funny that this should be a big deal, but it really is.

However, keep in mind that many of these are parents who wanted to be involved in their neighborhood schools, the schools often selected these parents out, not the other way around. So even if this is a huge factor, it’s informative — public schools (despite their rhetoric) don’t want parents paying attention, and that makes a huge difference in outcomes.

2. Charter schools normally have particular features, they will charter as a school that teaches Latin, or a vocational focus, or concentrate on reading, or STEM, etc. Whatever the exact mission in, what you see in a charter is that the curriculum and focus of the teaching has a target. In most public schools, the only target is the testing. So if the tests are not a good marker of adult success, any school that inherently focuses less on the testing (even if they incidentally do as well or better on it) is going to provide a better rate of adult success overall.

3. Home school families run into this all the time, if they are required (most are) to standardize test. You have goals for your kids’ learning that do not match up with the testing goals, often (more with common core). But you need to make sure they can do well on the tests, also. So you put, say, 30% of your effort into teaching to testing, and 70% to teaching to what you think is important for them to know as adults. Some of that 70% helps with testing, also, so you get decent or good testing results but pretty good real world results, also. But a school might put 30% of effort into local priorities, and 70% into teaching to the test. Winds up with acceptable test scores (or if it doesn’t, you bump it to 20/80) but doesn’t do so good on long term outcomes.

51 Rahul February 3, 2014 at 10:42 am

Re. #3 Why do people not like standardized tests? Unless the test is so stupidly written to test stuff that you think is irrelevant? Are they? I’d love to see a typical standardized school test.

Besides, how long can you keep your kid insulated from a standardized test? Eventually he will have to clear the SAT or some other similar test.

52 Marie February 3, 2014 at 11:13 am

Yes, that’s why most parents use them (our state requires evaluation for home school to make sure you really are schooling your kid, not just sleeping in — you can use either standardized tests or a qualified evaluator). Most pick the tests because it’s important to be good at test taking once you hit college, and in other areas, too. You can be brilliant and capable but if you can’t take a test you’re at a severe disadvantage.

But much of the standardized test is not useful for evaluation, much less for “teaching to”. For example, my oldest consistently scores in about the 90th percentile in spelling on those tests, but in writing she can’t spell worth beans. She needs to learn to edit and rewrite because of her spelling issues, but the test would never tell me that.

Another example, we took the ITBS one year, and it had an entire section on using reference material — e.g. using a dictionary. My kids scored middling on that because it didn’t test a skill, it tested whether those things had been taught yet. They basically scored average even though we had not gone over it. Now, we only have so much time in the day, and I’d rather spend it teaching other things than how to use a dictionary when, frankly, my kids just click spell check.

This sort of thing.

Common Core testing is a big controversy in part because it intentionally tests different things than old fashioned tests do. Some think it’s about time, others think it’s a faddish disaster in the making (you can find tons of horror stories), but the fact is that the idea that these tests match up with the skills you need to be successful in life, or even in academics, is by not means a sure thing.
You can, actually, get standardized tests. Seton testing can send you current tests, Christian Liberty Press can send you old ones, you can actually take one online for $25, but if you get the paper tests you have to send them back. It’s interesting.

53 Rahul February 4, 2014 at 6:10 am

I just feel that if the same person both teaches & tests always there’s often problems. It’s challenging but rewarding to teach a student material when you know someone else is going to test him on stuff and you sort of know what but not exactly.

54 Marie February 4, 2014 at 9:13 am

I agree, it’s essential to have an outside yard stick, and it’s an issue with home school where you often evaluate informally and it’s easy to assume a kid knows something because you’ve talked about it.

That actually is a problem with the standardized tests, though, in schools, the teachers are well informed about the content and teach to the test, which sucks up instructional time and skews the results. The tests are testing the school and the teacher, not the student — they are about evaluating whether the teacher is doing her job.

So too with home school, the point of the test is not to evaluate my kid, or to identify issues, but to see if her teacher is doing her job. It’s a check on me, not her.

So they are useful for safety nets, but they aren’t useful in the same way that, say, the SAT or placement tests might be.

I agree on the challenge, and there’s always a secret guilty fun in having your kid test and seeing how he does, whether you home school or not. The kids often like testing, too, it can be like a game or puzzle.

The problem is when you get to something like the Common Core. If you see the criticisms, the test evaluates based on understanding and process, not conclusion. So, essentially, you get math tests where the child is scored on whether she follows a path, not on whether she gets to the right place. So in my house, my kids do their Saxon math and when they take a test, if they get the answer right, they get points for it. But when you teach to the Common Core, a kid doesn’t get full points if he doesn’t use the right process to get the answer. In some cases I understand a kid might do a complex problem and come to the exact right conclusion, but still get the problem wrong because he can’t describe how he got the right answer.

This is a fundamental difference in outlook that testing can force on a situation. What we will have to do at home is take the hit on the testing scores (which is fine since we have wiggle room for that) in order to not get off track on what we feel is important (getting the right answer). But a classroom teacher can’t do that.

55 Rahul February 4, 2014 at 11:11 am

The way you described Common Core Math testing it sounds pretty horrid. Do you have a link to the sort of testing you describe? Where they focus on a particular method & not an answer?

56 Marie February 4, 2014 at 12:32 pm

Here’s an article, I’m not behind it entirely because part of the premise seems to be that it’s too “hard”, which is not the point, but if you click on the example test for first grade (PDF) you’ll see it’s . . . . . weird.

It’s hard to parse out accurate info on CC right now because it’s so controversial you’re likely to get only defense and offense.

57 Rahul February 5, 2014 at 12:36 am

Mary, That test is absolutely ghastly! It’s all aimed at arcane method than result. I really hope this thing doesn’t pass.

58 mulp February 3, 2014 at 10:38 am

Does this mean the best place to be raising kids are
New Orleans: 79% charter
Detroit: 51%
DC: 43%
Flint, MI: 36%
Kansas City, MO: 36%
Gary, IN: 35%

As I recall looking at the data for the DC voucher program that was the Republican Congress lottery for a small number of lucky DC poor families, the persistence in the charter and private schools depended mostly on household stability. The biggest reason kids dropped out of the program was the family moved and reasons like transportation or a better public school or leaving DC ended the attendance at the school that the parents had to work at to get their kid into. Few of the students wanted the school their parents chose, or thought the school was better – the parent’s involvement was what was key to getting the kids into a charter or private school, plus luck.

So, unless the study controls for:
family stability
changing residency
parental effort in getting kids to a different school
parental ability to get kids to a different school on daily basis

then my guess is the study is basically saying that parents with the economic means and the willingness to personally sacrifice for their kids is the key to kids getting into higher ed and thus a better life.

59 We live in interesting times February 3, 2014 at 11:06 am

After the disastrous Kansas City, MO, public school experiment from around 1986-2001, could charters be worse?

60 We live in interesting times February 3, 2014 at 11:08 am

And how many teachers in public school systems across the country would send their children to private school?

61 zbicyclist February 3, 2014 at 11:45 am

Actually, it’s pretty common for teachers in public school systems to send their children to private/parochial schools. One explanation for that should be familiar to economists: public schools pay better. It may be quite rational to send your children to St. Scholastica, but to pay for that (and a higher standard of living overall) teach in the public system.

62 We live in interesting times February 3, 2014 at 12:10 pm

Considering all the money that flows into the DC area, and the colleges surrounding that area, their public schools should be better.

63 Marie February 3, 2014 at 12:21 pm

But also, lots of former public school teachers send their kids to private schools or home school.

64 Rahul February 4, 2014 at 11:20 am


Say you dislike the way public schools go about teaching or the peer group etc. Even then, isn’t it easier to co-ordinate with a bunch of like minded parents, pool resources & hire tutors etc. that you approve of and will follow a style of teaching you like? Isn’t there utility to specialization of knowledge (e.g. can I teach both High School Physics & History decently? ) or having one teacher teach say eight kids together instead of just one etc.? Or do you not find enough like minded parents? Is the co-ordination problem too onerous?

I’m trying to understand why people home school. Avoiding the mess of a public school sounds reasonable but unless you are really remote geographically or your concept of education is really so unique isn’t there some economy of scale possible?

65 Marie February 4, 2014 at 12:23 pm

Easy answer — government regulation.

There are a ton of regulations concerning what gets to call itself a school in the U.S.

But if you home school, most states can’t deny that you have a right to individually teach your own kid.
What you describe is exactly what most home school parents do — they teach what they can and wish to themselves. They use materials from professionals when they are needed. But then they hire out to tutors and organizations certain subjects and areas where that’s the best way to teach, either on a one on one or group basis.

As an example, we have a home school co-op. There are 70 families that meet, so about 150 kids. The parents are present at all times, so it does not break state rules about hiring someone to teach your kid that doesn’t have a license, and the facility is not a school so they don’t have to follow state regs on schools. We just all work together. But last year one of the moms with a science degree taught high school biology with lab and several of the teens that took it passed the CLEP for that, which gives you college credit with many universities. There are classes in literature, science, sign language, Spanish, as well as fun classes like choir, cooking, etc.

Home schooling is the route, in the U.S., to the exact kind of thing you describe. And to a large degree, this answers your question of why people home school.

66 The Anti-Gnostic February 3, 2014 at 12:01 pm

It’s the students, not the schools.

Everybody knows this which is why we pay $100,000 more for houses where our kids will have lots of white/Asian friends, but it’s impolite to mention so they throw out the old ‘parent involvement in children’s education’ canard. That sort of talk goes nowhere as well, as even the most obtuse progressives realize we don’t want a bunch of lawsuits 30 years down the road by millions of children forcibly separated from their biological parents.
So we continue to chase these fads and do backflips around very obvious phenomena in the name of equality-of-inputs-yields-equality-of-outcomes. Things like kicking out the very small minority of violent, pathologically disruptive students, ensuring basic nutrition, recognizing (dare I even say it) that not more than 20% of high schoolers in aggregate should consider college–are ignored.

67 Marie February 3, 2014 at 12:31 pm

I can’t speak to where any one person starts, or the whole phenom of segregation by school choice, but having taught in a majority minority school I can tell you that no matter whether the school is the only problem, the school is a huge problem.

We had a ton of smart, good kids, with families that cared a lot about whether they learned, and the school stuffed them down. Absolutely crushed them under heel. Maybe part of why white parents send their kids to schools with other white kids is because the teachers and admin don’t feel like they can get away with the garbage they do other places. Well, with as much garbage.

68 John B. in NE February 4, 2014 at 1:22 pm

Marie, can you expand on that? How did the school crush kids? Is it something that outside forces like the local board or local politicians or local laws could help? Personally what I’ve seen of that sort is tolerated bullying (tolerated by the teachers as well as the other students) but I’m sure worse things can happen.

69 Marie February 4, 2014 at 7:56 pm

General school atmospheres can be very dehumanizing. I substitute taught for awhile and posted horror stories about student cruelty towards each other and the Vichy administrations that tolerated or gained by it.

The situation I was referring to was on the border, and there were so many predatory groups trying to gain by using the students. The worst was the “soft bigotry of low expectations” type of stuff — e.g. selling candy and soda to the kids to raise money for a big party at eighth grade graduation, because we all know that these kids will never get to celebrate a high school graduation, right? Fortunately, this was a Mexican American bunch and they didn’t care about the “message”, they just enjoyed the party.

Then there was the advanced math program that was cancelled — since so many kids had English second, language based paths were harder to excel at, so many were able to really stand out in math. Nope, not allowed, shut it down.

Then there was the guy who breezed into town and convinced the school to sponsor his anti-gang programs, since he was a former gang member from California. Of course, as could have been foreseen, he took a bunch of wannabes and gave them aspirations to be as bad as he was, and he leveraged this into a position in local politics. Meanwhile, not long after the fake gangs got real enough that one middle school student murdered another.

The crushing was incidental, these kids were a means to an end — their own inflated salaries or ambitions, desire for a low accountability job, whatever. If the kid made an effort to excel, most of the time that meant more work and a distraction from the adults’ goals, so it was discouraged or left to dry up and blow away.

70 NPW February 3, 2014 at 12:36 pm

“we pay $100,000 more for houses”

I’m pretty sure I was involved in that part of my child’s education……Just sayin’

71 Jason February 3, 2014 at 12:07 pm

Correlation does not determine causation.

72 Govco February 3, 2014 at 12:23 pm


73 LemmusLemmus February 3, 2014 at 1:10 pm

I was considering writing that exact same comment (when my screen showed Jason’s comment but not yours).

74 notanidiot February 3, 2014 at 4:40 pm

Get back to Reddit.

75 ano February 3, 2014 at 1:15 pm

Finally some evidence that can be cited to go along with economists’ “mental-model-of-the-world” based faith in school choice and competition.

Comments on this entry are closed.

Previous post:

Next post: