There was a nice study of pre-K in Tennessee. People applied for pre-K , more than could be accommodated, so some were admitted (randomly) and some were not. Then the kids admitted were compared with the control group, those not admitted. This nicely controls for any systematic differences among volunteers, which often exist. By third grade the pre-K kids were doing worse, although I think not hugely worse. So pre-K is pointless (big surprise, considering that shared parental environment doesn’t affect smarts)- not that this or anything else is going to change James Heckman’s mind.
But then, I don’t think that most proponents of this kind of intervention have ever cared a whole lot about efficacy, because if they had, they would have run statistically powerful studies of the effects of pilot programs. I mean, it’s not as if statistical methods are a secret. If they were serious they would stop spending money on approaches that have been seen to fail and try to spend it on something that worked, or might work – but I don’t see that happening.
West Hunter 
I’m willing to believe the null effects after three years, but there were several methodological problems with the study. It was based on joiners, first of all (consent to the study happened after randomization), which means that you can’t be sure that unobserved characteristics are equalized.* More problematically, the “negative effects” happen between the third and fourth wave of data collection, which is also when they appear to be losing lots more treatment sample members than control members (maybe the parents who are happy with how their kids are doing are less likely to answer the survey?) and it’s not clear how they’re adjusting their statistical significance (which is only <0.10, not 0.05) for multiple comparisons, of which they have a ton given how many outcomes they collect.
*They do an extensive exercise in equalizing baseline characteristics. But this is exactly the problem with the Chetty et al. studies that show huge teacher effects on later outcomes in samples that are also balanced on observables; the further you get from baseline, the more- not less- unobserved, primarily hereditary characteristics matter. So whenever you see quasi-experimental studies (which despite the lottery, this ends up being due to the joiner issue) with weird or interesting or big long-term effects relative to short-term effects, I’m increasingly convinced that this is just the result of unobserved hereditary variation explaining a larger portion of variance over time. This also explains fade-out, of course, which many education researchers have to create somewhat involved explanations in order to explain, because they want to believe that conditional on test scores at a given point in time, all kids are the same.
“If they were serious they would stop spending money on approaches that have been seen to fail and try to spend it on something that worked, or might work – but I don’t see that happening.”
And that is the real puzzle. We can assume that the people who continue to “teach” these programs do so because they are paid to, and they defend them because they are paid more than they think they could get elsewhere. But what about the people who dream them up? Why are they unwilling to draw the perfectly obvious conclusion that they are wasting resources?
I am thinking the reason is, that they are committed to the idea — no, to the moral certainty — that there are no educational shortcomings that cannot be fixed. And since the evidence is clearly against that position, they are willing to ignore that evidence. But why?
Heckman accepts that long-term cognitive effects are zero. He just thinks there are a lot of other impacts. I think he’s probably overstating the case, but isn’t 100% wrong. A society with well-run Pre-K is going to be a nicer place to grow up in than a society without, on balance, and even if we think that behavioral traits are mainly genetic in origin, their expression depends on how nice the place you live is.
Heckman’s biggest problem is that he understates how much things have changed. The early Head Start studies (which he doesn’t trumpet, but weren’t that badly done despite being quasi-experimental) showed big effects in large part because they were often conducted in counties in the Mississippi Delta which where kids were still getting TB and a lot of nasty parasites. The counterfactual was a lot worse, and the ability of the program to improve nutrition and provide basic health care to the kids could completely plausibly make a big difference. Even Perry and Abecedarian (Heckman’s preferred programs) were conducted in much different societies, particularly for poor Black kids, than our own.
For all I know, Heckman has the sign wrong. I don’t know that a society with a well-run pre-K would be better and neither do you: but we do know that it would cost a lot of money. Arguably, having less money is bad.
As for expression depending upon how ‘nice’ the place you live – what is ‘nice’?
So why not let kids have time to be kids? Why do they need more structured government time? They have enough of that.
The argument about pre-K is not about a mandatory program. Even Oklahoma’s universal program isn’t mandatory. So if parents have a great idea about how to let their kids be kids they can do it.
But let’s not pretend that the modal participant in these programs, if they aren’t going to school, is going to be taking apart engines like Danny the Champion of the World or reading ten books a day about World War II like Greg’s kids. They’re going to be staring at a screen, and driving their parents crazy, and rarely going outside or playing with other kids.
Who cares if they are taking apart train engines? Kids play in the sand. Walk in the woods, go to the park, read ‘Guess How Much I love You’ and ‘The Little Engine that Could’. Or have them read to them or whatever. I raised my first set of kids across the street from a homeschooling family. All five of the kids are normal, healthy adults, most with kids of their own all with jobs.
This is about failed programs that spend billions I don’t know maybe trillions of dollars in the last several decades and have squat to show for it except programming children to be good little drones for the first few years of life.
If you want to program your kids Skinner. Go ahead and put them in the box. It’s useless and expensive but at least you won’t have to suffer the consequences.
Here’ how we make this work. Anyone who advocates an educational program has to put in a ‘chit’ worth twelve years of his life. If the program works or at least does no harm he can get the chit back. If it is bull feathers and actually wastes money time and energy and shows negative results he can sit in his program for the next twelve years of his life.
Some people point to the Finnish practice of kids starting school later than in the US and say that that is the way to go.
It probably is. For Finns. People will soon look back fondly on the Finns, and Swedes, and other self-destructing nations.
Charles Murray has a line somewhere that kids aren’t a science fair project; if you assume that long term effects are zero (and I’d disagree with you and JayMan that outcomes 20 years down the road are the only ones that matter), you’re left with your own moral and aesthetic intuitions about what you want for and from your kids. As a society, “niceness” is obviously susceptible to the disagreement over those moral and aesthetic intuitions, but most people can agree what terrible is. And schools, even in bad neighborhoods in our country, are generally not terrible places. It may not be an ideal of intellectual engagement, but that’s not usually within the opportunity set anyways, and it’s better organized, more peaceable and pleasant, than most other places in the kids’ neighborhoods or most aspects of their lives.
What’s your alternative?
The simplest alternative is to not implement a policy when the evidence suggest it has no real positive effect. In fact, before spending a few trillion, it could be argued that you need really strong evidence. Considering that many somewhat similar programs have been tried to no particular effect over the past fifty years or so, I’d say my skepticism level is pretty high. But as for the notion that this record of failure means that the fucking default is that you should get to implement your latest dream (What’s your alternative?) – isn’t that just a little bit overconfident ? Some might find it positively irritating.
And, as in many other cases in recent American history, I’d like to see prominent advocates suffer in some way when their scheme goes wrong. They need to have some skin in the game. Right now, failure seems to mean that at worst you’ll be hired as an expert on the subject. What if we try this and it does nothing useful? What price should James Heckman pay? It is not as if he has applied best practice to his research on this question, done the best any reasonable person could. He ignores all biological effects, and I can tell you, the people he has working the problem for him do not impress.
What are you talking about? There is good research, that shows that 4 year olds learn stuff when you teach it to them. The much-hailed (by critics of pre-K) impact study of Head Start had a very mild service contrast — one year of Head Start versus two, pretty much, and it showed that the kids learned a fair amount during that year, which they later forgot. (Insofar as we can detect those later impacts, which gets much harder the further you get from random assignment.)
The controversy over long and medium term effects isn’t a conspiracy– it’s genuinely unclear if these programs have medium-term effects. I said before that I thought that the service contrast was higher in the 60s than it is now, but it also seems likely that if you did away with Head Start tomorrow, you’d make all the other early childhood and daycare programs in the area quite a bit worse, which RCTs aren’t going to tell you much about.
I read a lot of interviews with Head Start parents, and for an awful lot of them I have a feeling that their kid’s pre-school is their main source of community and support.
I think you radically underestimate the crappiness of lower-class life in America if you think that most pre-K programs are something worth getting angry about.
Pre-K is not really the problem. Head Start, although useless, is not really the problem.
People like you are the problem.
Why, pray tell, am I the problem? I’ve read your blog completely, I’ve read at least one of your papers quite closely, I’ve read many of the papers JayMan cites, and nothing I’ve said on this page contradicts any of your or their main findings. As I indicated below, I respect your and Henry’s work, and am interested in how to reconcile it with the rest of the world as it exists. Why am I the goddamn problem?
I’ve been hearing this crap my entire life and I have no more patience for it. Or for its advocates.
Maybe I’m misunderstanding you.
At least one of Heckman’s former students is trying to incorporate GWAS and genetics into the model. This paper is about obesity, but he doesn’t hide that he thinks it applies to cognitive skills as well.
https://769262cf-a-62cb3a1a-s-sites.googlegroups.com/site/pietrobiroli/files/PietroBiroli_JMP.pdf?attachauth=ANoY7crqu8En7PF7qJIk3EUsY1bKzgctyw-L9vbSXfcLuSIpf8GdjZhlQSh38R8cTGvud1VfclbpxkfsiM0GNDR6bRwhg7oL7gDNJGTm1h_lZ6qsNo3-wLvLaSd7JWZuNDMc4owLpZOzfNFxx0rMWuMFHsQKei8nA6Us1PXim6OB3dBeZOWT0dHK4R2nTGilvEKBjS5paypvdfVHop6iBxeW41cHOhfTrbFbE4eT0iLZPQlSsOBDZVc%3D&attredirects=1
“schools, even in bad neighborhoods in our country, are generally not terrible places”
Wrong.
“[schools are] better organized, more peaceable and pleasant, than most other places in the kids’ neighborhoods or most aspects of their lives.”
Also wrong.
Read Gatto’s “Underground History of American Education”, (available free online) for evidence that not only are schools terrible, a crippling waste of human potential of no benefit to the students, they are so by design.
I approved this just because we needed some fresh nonsense, rather than the same old nonsense.
If Heckman can cause a huge national program to be put into place based on his research, one which once in existence will be impossible to dismantle, that sounds pretty good for his prestige and future prospects. And universal pre-K is the kind of program you can easily imagine catching on–it provides an effective subsidy to lots of voters, and justifies building up new bureaucracies and hiring lots of people. So maybe the whole inept use of statistics is more of a feature than a bug.
I’d say that you don’t understand the motivations and goals of the originators. If their actions indicate that they view the outcomes as successes, despite not having any benefit to the educational achievement of children, then it shows they were never really concerned about that.
Doesn’t it mean they just ignore evidence? Like pretty much every advocate group ever to exist.
I think the big effects for some of the pilot programs come from crime reduction rather than education benefits. They likely are spurious effects that won’t scale but they did find benefits.
If you measure enough outcomes, you will always find some that move in the right direction.
“So pre-K is pointless (big surprise, considering that shared parental environment doesn’t affect smarts)”
One of my bigger challenges is getting certain people to admit that for groups observed in developed and semi-developed countries (including East Asia and Eastern Europe), the shared environment impact on IQ, behavioral traits, and most all outcomes is zero (comments seen here among other examples).
This is despite the enormous evidence on the matter (which requires a bit of intelligence to interpret, which I guess is one of my challenges).
Of course, one hold up is that there is a non-zero shared environment component on educational attainment. But that doesn’t seem to carry over to anything of interest (like income, for example); indeed, a twin control study from Denmark (Bingley, Christensen, and Markwardt (2015) in case the link one day dies) found that MZ twin differences in educational attainment don’t correspond to differences in income. I.e., education is entirely signalling. I try to tell people who cares if there is a non-zero C on education – it doesn’t seem to “mean” anything, in the long-run, but…
Even if you believe that broad psychological traits are genetically determined, the mapping function between the joint distribution of those traits and the outcomes we care about is dependent on culture, the quality of public services, and economic institutions. So if you torch all the schools, it might make little difference to who ends up on top, but it still matters to how well everyone lives.
@Jacob:
Where do culture and institutions come from?
But to your point, yes, no one is suggesting we dismantle the structures of society.
I know your answer to that question, that culture is an emergent phenomenon of the biology of the people who make it up. I’m sure that’s partially correct.
Lots of people suggest we dismantle the structures of society all the time, this is the internet. Even among serious people like our distinguished host, there’s no clear agreement about where the defining structures of our society end and where meddling and social engineering begin.
@Jacob:
“Even among serious people like our distinguished host, there’s no clear agreement about where the defining structures of our society end and where meddling and social engineering begin.”
That’s because there is no such line.
When you’re discussing the system under examination, it’s essential to keep in mind that you (and all of us) are a part of it.
By the way, that was a brilliant answer to the question of where does culture come from. The Chick approves:
Reminds me of a study showing that E2 component of IQ had ~no effects either in a twin control study. Perhaps it is just noise (measurement error), or maybe it is hollow (does not measure GCA) and the non-GCA variance is not causally efficacious. N=289 twin pairs.
Nedelec, J. L., Schwartz, J. A., Connolly, E. J., & Beaver, K. M. (2012). Exploring the association between IQ and differential life outcomes: Results from a longitudinal sample of monozygotic twins. Temas em Psicologia, 20(1), 31-43.
I thought these programs existed so parents could feel better about going to work all day. And the extra work force should boost the economy, or something like that. Am I way off base here?
I think they exist because communist utopias, from Plato to Thomas More, and ever since, have normally had the children being raised communally, instead of by the family. There is something about that idea that just appeals to some people.
It’s like a disease for which there is no known useful treatment. Quackery of some sort or the other will flourish indefinitely until something is found that actually works.
The point for the proponents of pre-K is to get the kids under their control at an earlier age so they can more effectively (they hope) propagandize them. Got to influence those tender young minds to have the ‘correct beliefs’ about all the important subjects, like racism, and sexism, and global warming, and…! Best to do that long before the beginning of rationality.
So, it doesn’t matter, objectively, if pre-K has the advertised effect on grades or not.
What makes you think that there is any systematic plan behind this at all? Or behind damn-near anything in public life?
Back in 1990, the Democrats supported drawing overwhelmingly black Congressional districts in the south, because a district needed a high black fraction for a black congressman to win, especially in off-year elections, when black turnout slumps. Of course this means that the Democratic Party wins fewer seats by bigger margins (and screwier candidates), the reverse of gerrymandering. Suicidal politics. The Republicans, in a strange moment of semi-intelligence, said ‘OK’.
There is no Inner Party.
There are plenty of attempts at Inner Party. Just not effective, intelligent ones that work. People think they are outwitting their opposition all the time. Occasionally, it’s true, but seldom for long.
Yup.
The Law of Unintended Consequences strikes again.
The good thing about being clueless is you’re bound to make some good decisions by pure chance.
We wind up with people in Congress who think Guam is about to tip over.
“There is no Inner Party”
Almost certainly true. The question is, is this fact reassuring, scary, or maybe a little of both?
Is it better to be ruled by Professor Moriarty or the Three Stooges?
Creating a bunch of gummint or subsidized pre-schools creates numerous jobs for middle-class do-gooders and the deserving poor, who will join unions and vote Democrat. There is also the hope that keeping kids away from their feckless parents most of the day will somehow make them better. Not much different from the old Indian boarding schools.
if they let the kids play and don’t try too much formal teaching, they might have a good time. That’s worth something. The question is, how much?
The kind of people who run these programs took Stats 101 and got through it on memorization and rote application of rules, and almost certainly don’t even know what a p-value even means. And that is the extent of their knowledge of the topic. It is a magical box.
I’m not sure I know what Greg is saying here (about statistical power); Pre-K has been studied by statistically capable people for several decades now, and there are lots and lots of smart statisticians studying it now, not just Heckman and his Heckmen. I could project that he’s saying that if you created a bunch of programs, each with a small, relatively cheap to implement intervention, but with outcomes that are plausibly easy to influence and to detect, you could iteratively improve early childhood programs in a way that large-scale evaluations of expensive programs won’t do. But RCTs are very hard to implement and collect data for, and every quasi-experimental evaluation is hobbled by the fact that there are a million other things going on in schools at any given time (so attributing any difference to the impact of the intervention is a fool’s gambit), so there is a reason that evaluations tend towards the large study studying a large intervention.
There is no doubt that everyone in the Federal Government who is running early childhood programs, and most of the people in State Government doing so, understand the basics of statistics, and the people running evaluations definitely do so.
” in the Federal Government who is running early childhood programs”?
Who in the federal government is running these? There is some verbiage in http://www2.ed.gov/about/inits/ed/earlylearning/index.html
but no federal funding for pre-k yet, in spite of presidential verbosity.
http://www.acf.hhs.gov/programs/opre
People who are in charge of government thingies are like any other managers. I work with these people all day. If their department or title has something about “education” or “development” in it, then they feel they are required to keep on looking like they are trying to solve the problem. Their methods of doing so usually involves going to conferences with other people like themselves with hired speakers; or going and speaking at places themselves, to raise awareness of What Is Being Done; the political struggles at all levels also occupy much of their time and require consultation with similar others. They don’t do research in any formal sense, though they will pass out lots of surveys. What the researchers in the field research, I don’t know. I don’t see them. The longstanding belief is that even a blind pig will find an occasional acorn. I have no personal confirmation that this is true.
It’s not just social sciences. It’s what many managers do. They have to look like they are solving the problem. Actually solving it is secondary.
The old rule is that all educational experiments work. That’s because they draw on the enthusiasms of atypical teachers, parents, and pupils plus, I suppose, biased researchers. The experiments then all fail when extended to bog standard schools. The experimental design here is intended to rule out unrepresentative parents and children (though whether it does ….) but is there no effect even from dedicated (presumably) teachers?
My view is that if uncountable experiments over the years have failed to show any improvement that is both substantial and sustained, stop. Go away and think and read for a decade, for heaven’s sake.
Heckman (also Goolsbie, I think) cites lowered rates of criminality, etc. in later life as evidence favoring pre-K –in the absence of any showing of lasting academic improvement. This is a con job. Lowered rates of criminality can’t be attributed to academic improvement because there was none. But the studies were designed with only two groups by people who hoped to show academic improvement. Consequently the studies did not include treatments intended to compare intervention effects not caused by academic improvement. You might get a similar reduction of criminality by setting up scout troops, little league teams, or sending the kids to Sunday School. No argument for pre-K here. Doubt that Heckman wants to know.
Heckman says small dosages of M&M candies completely eliminate the black-white IQ gap, look for a big jump in student performance in schools across the country next week.
I have to throw in with Jacob here, and anyone who thinks I believe preschool affects academic outcomes can read this. Even the “gold standard” studies like Perry show that the treated group only had a 33% chance of being arrested five times by their 40th birthday, while the control group had a 1 in 2 chance. That’s nothing to get worked up about. You will never hear me pushing for preschool as an end to the achievement gap, or even a means of improving academic performance.
I am often asked why I bother to teach, if I believe that many of the students I teach don’t have the intellectual capacity to understand the material. I wrote an answer here:
“Suppose that in six months my weakest kids’ test scores are identical to the kids who doodled or slept through a boring lecture on the same material. Assume this lesson does nothing to increase their intrinsic motivation to learn math. Assume that some of the kids end up working the night shift at 7-11. Understand that I do make these assumptions.
Are the kids in my class better off for the experience? Was there value in the lesson itself, in the culmination of all those worksheets that gave them the basis to take on the challenge, in the success of their math in that moment? Is it worth educating kids if they don’t increase their abilities?”
I believe the answer is yes (and qualify the yes in the next paragraph slightly).
Likewise, suppose you have two kids raised by welfare moms who are barely functional. One goes to preschool, spends time with teachers who provide snacks, activities, engagement, evidence that people work for a living, evidence that people in the outside world care about him. The other stays at home in an utter shithole drinking Fanta in front of a TV all day when he isn’t watching his mom get beat up by her boyfriend/pimp or playing in traffic, eating dry Cheerios towards the end of the month when the foodstamps run out.
Five years later, they both have the same test scores. But one of them has memories of a caring environment and fun.
Anyone who wants to argue against preschool for disadvantaged kids is welcome to do so. There are valid arguments. But money spent taking kids out of crappy homes with a cramped, limited cultural environment isn’t automatically money wasted.
edited to add: I’m well aware that the people pushing preschool aren’t using this argument. But Jacob made that distinction very clearly. I read him as pushing back against the assertions that preschool is a pure waste of time, period, given these outcomes. No. Preschool to fix the achievement gap is wasted. Preschool itself for poor kids to help them feel safer, cared for, more engaged by a wider world? Different topic.
“Preschool itself for poor kids to help them feel safer, cared for, more engaged by a wider world?”
If this is a valuable goal, and it might very well be, one would think that a program specifically created and designed to achieve this goal would be both more effective and more economical than one that produced this outcome as an unintentional by-product of what it is supposed to achieve. I mean, you can open a can of beans with a machete, but you’re much better off with a can opener, and much less likely to cut your foot off in the process.
It’s one of the weakest excuses for picking my pocket that I’ve ever heard. There is only so much money, so many resources: a real bleeding heart would want to do something that worked – of course by that definition there aren’t many bleeding hearts. Norman Borlaug, I guess.
And, by the way, ‘poor’ must be the wrong word, because lots of people who were kids in the Depression (like my Dad) felt cared for and safe, although maybe hungry. How about ‘fucked up’ ? Surely it would be more accurate. And just a few years later, many from that generation were overly engaged with the wider world, too (flak over Kassel in my Dad’s case).
“a real bleeding heart would want to do something that worked”
Well, if “worked” means “raising test scores”, nothing “works”. If “worked” means “taking kids out of bad situations and giving them structure” then preschool “works” provided the kids are fed, engaged, and cared for while in school. Preschool “not working” would be kids in abusive situations. The stats for poor kids in daycare are bad on that measure. Stats for Head Start compared to daycare are much better.
Look, it’s fine to argue against preschool. And I completely agree that if the goal is improved academics, preschool is unsuccessful. I also agree that if someone argued for preschool purely on the grounds that leaving kids in an environment where everyone thinks living off the government dime is a great idea, he’d be shot down as a racist.
“And, by the way, ‘poor’ must be the wrong word, because lots of people who were kids in the Depression (like my Dad) felt cared for and safe, although maybe hungry. ”
No, “poor” isn’t the wrong word. These days, “poor children” almost always carries the attribute “low IQ parents”, which creates a host of other bad outcomes. And you know this, of course.
You’re definitely a one-hammer guy.
“an environment where everyone thinks living off the government dime is a great idea” – like preschool?
You can’t even honestly tell me that pre-K reduces the chance of the kid knocking over a liquor store later in life. Whereas I, for less money, could easily reduce that probability.
Lots of people attribute the low performance to the poverty – the people who run the show. That’s why calling things by their right names matters.
One hammer – obviously not. I just think you’re dead wrong on this. But let’s make a deal – you can have all the pre-K you want, as long as YOU pay for it.
About ‘nothing’ working: I am not very optimistic, but I don’t know that nothing can work. It seems to me that we keep trying the same thing over and over again – face time with liberal arts majors – and are surprised when it doesn’t work again. Out of the space of possible tries, seems to me that we haven’t tried more than a teeny fraction.
Ironically, I taught Borlaug’s great-grandson, and it was not evidence for the persistence of behavioral traits across multiple generations.
Well, that’s a small sample. I was reading something in my maternal great-grand-father James Gregory’s obituary – how he couldn’t suffer fools gladly. And this was written by an old and close friend ! ( who happened to be my great-great-grandfather on my Dad’s side – Civil War buddies and pseudo-cousins).
“If this is a valuable goal, and it might very well be, one would think that a program specifically created and designed to achieve this goal would be both more effective and more economical than one that produced this outcome as an unintentional by-product of what it is supposed to achieve.”
Well, as I mentioned above, the stats on daycare–the economical version– for working moms are terrible. In fact, if you view Head Start as just a means to get mothers working and their kids in daycare, it’s much more successful than it is at producing better academic outcomes.
So if you want to improve academic outcomes, nothing works. If you want to get welfare moms into a job simply to force them to at least try to be functional, then preschool does as well as leaving the kids at home with mom, and better than daycare. At least, the last I’ve read that was the case.
I’d certainly rather spend the money on paying women not to have children until they’ve reach 22. That’d be even cheaper.
But the value of doing that depends on how that affects the kids lives, right? I mean, if pushing the mom into a crappy job and the kid into government run daycare leads the kid to do better in life, it might be worthwhile. If we can’t see any effect, then why bother?
Your reason seems to be that the kids will have a nicer life. And I’m sure that’s true for the kids with the worst home lives. But can we at least measure that somehow? How do we even know the pre-K kids are having a better life on average? Plenty of kids hate school now, and if it’s doing them no good, it’s hard to see why we’re subjecting them to it.
“If we can’t see any effect, then why bother?”
With the caveat that all effects aren’t academic, that’s fair. My primary purpose was to observe that not all effects are academic.
“Ironically, I taught Borlaug’s great-grandson, and it was not evidence for the persistence of behavioral traits across multiple generations.”
Well, my Dad was a decorated war hero. Regression to the mean strikes again, I guess…
Well, I for one am happy to see some here advocating for the most expensive baby-sitting service in the world – one which allows its patrons to feel loved, cared for, and smarter, too, as we will undoubtedly continue to lie to them about the effects of baby-sitting on their learning curves.
Head Start is around 7,500 per child. This is less than half of what a place like Baltimore spends on K-12. There are more expensive public pre-K programs for disabled kids. These are, I think mandated by IDEA, but I don’t know the details.
The policy margin in most of these cases is between means-tested (Head Start) and universal (eg, Oklahoma’s) programs, not between a means-tested program and no program at all.
If you want to save money, Ed Realist did an excellent series on educational policies that would actually save money:
None of this will happen, of course. There is a much greater chance of the White House being put onto a truck and moved to Sioux City, Iowa by the next administration than of anyone repealing IDEA or restricting K-12 to citizens only.
Jacob,
I’ve read Education Realist for many years. I wouldn’t trust ER to budget a dog pound.
Jacob,
The proper comparison for the cost of pre-K programs is not with K-12, but with a baby-sitting clinic. After all, that’s what you and Education Realist seem to think pre-K programs should be, something to help with the quality of life by allowing single mothers to work and to provide a safe, nurturing environment for pre-K tots who might otherwise watch TV all day.
$7,500 per child is not a bargain for that service.
In other words, what you’re saying is that you prefer to limit the debate about these programs. Let’s not talk about whether they’re needed. Let’s just talk about how we will determine funding for them.
I’m not interested in what you think the “policy margin” is in “most of these cases.” I’m interested in what’s good for the country. And if that means expanding the debate to questions you seem to think ought to be ignored because they aren’t on the “policy margin,” so be it.
I agree. It is easy to think of current programs or policies that, only a few years ago, were considered to be not just ‘outside the policy margin’ but floridly insane. Of course, they often were insane, but that pattern shows that we’re not straitjacketed by the ‘policy window’.
” I wouldn’t trust ER to budget a dog pound.”
I think a dog pound has a tough budget. To truly insult my financial skills, you should declare me incapable of budgeting a billionaire’s bimbo suite.
“You can’t even honestly tell me that pre-K reduces the chance of the kid knocking over a liquor store later in life.”
Good lord. That’s kind of the point. I’m AGREEING WITH YOU. Learn what that reads like. It’s not that complicated.
Look, my support for pre-K is tepid at best, and would always be means tested. I’m not even arguing for it here. I was merely taken aback at the nastiness shown Jacob for pointing out that preschool isn’t necessarily a bad thing if you have the right objectives.
“Lots of people attribute the low performance to the poverty – the people who run the show. ”
Yes, but I’m not talking to them. I was talking to you. But you’re so blinded by whatever bug crawled up your ass you don’t seem capable of realizing it.
“About ‘nothing’ working: I am not very optimistic, but I don’t know that nothing can work. ”
Well, then you’re the one determined to waste money, so stop yelling at everyone else. As you and liberals define it, nothing can work.
Jacob wants to do something that has no practical payoff and costs a lot of money. The hell with that. I’ve seen enough of that. And the bit about ‘what’s your alternative ?” was infuriating – like we have to go along with every such piece of nonsense. Admittedly we often do…
I’m talking to more than you. Capische? Calling the people in question ‘poor’ is inaccurate.
I don’t know that nothing can work – neither do you. But then I’m considering possibilities that you don’t know anything about – something other than face time with liberal arts graduates. If any of those possibilities looked like panning out, I’d maybe be for a small test, which wouldn’t bankrupt the country. You wouldn’t want to implement it generally unless A. it was known to work and B. reasonably cheap – costs matter.
” The hell with that. ”
That’s fine. Seriously. I think it’s a legitimate disagreement. But you saw his question as a mandate. I saw it as a question about the well-being of low income kids with low IQ parents.
My answer to “what’s your alternative?” is “spend the money persuading low income young women to avoid parenthood”.
” Calling the people in question ‘poor’ is inaccurate.”
No, it’s not. If you want to address the wider audience, then they are unquestionably arguing in favor of universal preschool to a) improve the performance of low income black and Hispanic kids by b) putting them among higher income white and Asian kids. That’s why the push is for universal preschool. It’s universal, but they want to help poor kids.
This is an idiotic idea. But the kids that they–and I–are thinking of are the low income kids. No one is worrying about middle income black and Hispanic kids. Them because they don’t dare admit that poor white kids outscore middle class blacks and Hispanics. Me because they aren’t living in conditions I think are a problem, whether or not they cause poor test scores.
“I don’t know that nothing can work – neither do you.”
Yeah, I pretty much do, if it’s about early childhood interventions. So do you. I think we could do more to teach kids to use what they have, but that’s going to be a high school thing.
ER’s convoluted reasoning in (tepid) support of spending beaucoup bucks on pre-K is apparently – I say “apparently,” because who really knows? – that “doing nothing” is as expensive as any untested grand scheme.
ER’s support for this notion that spending a little money on a few well-tested pilot projects or even doing nothing is as “wasteful” as what’s being implemented or proposed on the “policy margin” is … oh, how shall I put this?… not exactly sound financially.
But then if you follow ER down that rabbit hole you’ll discover all kinds of “education realist” inversions that don’t make sense on what we here call planet earth.
” that “doing nothing” is as expensive as any untested grand scheme.”
I said that? I think about the only thing I said about money was “money spent keeping kids out of a hellhole isn’t automatically money wasted”, or something like that.
I am mildly pushing back against “Pre-k is utterly wasted because it doesn’t do any good”, to restate is “pre-k is a waste of time for raising academic outcomes, but whether or not we should fund preschool for other reasons is a different conversation”. The result of that conversation being unknown, and open to many different attacks.
ER,
Rather than say a thousand words which have no meaning and leave readers trying to guess what you’re trying to say, why don’t you state plainly why Americans should pay for universal pre-K (or for whatever variant of limited national pre-K you believe is in the country’s good)?
What are Americans getting? What is the price? Why is it worth the price?
Those are the simple questions you dance around. Because the moment you are forced to speak plainly about what you want to do and what it’s going to cost, you’re going to say something absurd. You know it, and I know it. That’s why you dress your posts up with so much useless verbiage.
I think ER has made clear that their support is for something like Head Start (means-tested, and at a current cost of 7500 per low-income kid.)
Not to sound like a…liberal…but the sum cost of Head Start over its five decades is less than the sunk costs of the F-35 (go Lockheed!) We’re not talking Medicare, Social Security Disability, Iraq War, or even SNAP levels of expenditure here. (Incidentally, this is probably because the actual program model is people with associates’ degrees rather than liberal arts’ graduates as Greg keeps saying.)
Jacob,
ER made nothing clear.
What is the purpose of the program? Baby sitting for poor women so that they might raise the nation’s fertility rate by having even more children? Because if so, I don’t think you and ER have fully thought through the consequences.
So your level of budget sophistication is that if some program doesn’t cost as much as a military boondoggle, it gets your appreciative thumbs-up.
I don’t mind that you sound like a liberal, Jacob, I just wish you sounded smart.
France and the Nordics adopted universal early childhood programs to boost native-born birth rates. It worked…a little. Increased TFR by maybe 0.3. It’s one reason French elites aren’t so jazzed about migrants as Germans are/were. It worked better than Singapore trying to pay parents to have kids, though there’s not a lot of room in Singapore so that may be why.
In the U.S., though, ER is probably right that early childhood programs, even ones conceived as open to middle class parents, aren’t intended as natalist, let alone eugenic. In DC or NYC they might be about trying to convince middle class parents not to leave when they have kids, I suppose. In Oklahoma, the state having extra oil money and wanting to spend it.
Where are you getting this from?
Sorry, but limited success of natalist politics of France and Nordics is well-known (not only for first-generation immigrants). Not a victory, but better than Germany.
Dmitriy,
So you don’t know, either.
Let me rephrase: Please find me some reliable source which says the differences between French and German elite attitudes toward migrants is due to the “limited success” in universal early childhood programs that have slightly boosted French native born rates relative to German native born rates?
It sounds like a bunch of just-so hooey to me, but I wait to be enlightened.
The only source that I’ve recently read on the subject is Jonathan Last’s What to Expect When No One is Expecting: America’s Coming Demographic Disaster. I don’t consider this popular treatment definitive, but the author does mention France’s natalist policy, and what he has to say about it does not inspire any confidence that America should be copying France.
First, Last points out that it is recent French immigrants who are driving France’s high TFR. These French immigrants are more fecund than the native white population (3x), which suggests it’s more about culture and less about policy.
Second, Last claims that France has been promoting pro-natalist policies for a century, and the country still has a below-replacement level TFR.
Third, Last says a large body of research by both economists and other social scientists suggests France’s pro-natalist policies have had “no effect at all.” (He also says that some research claims it has had a marginal effect.)
So please, Dmitriy, spare me your “… [the] limited success of natalist politics of France and Nordics is well-known”, when what you really mean is that you heard about read about in The Economist once or twice. What source do you have that supports Jacob’s claim that French and German elites have differing attitudes toward migrants based on France’s limited success with universal early childhood programs?
Because I’ll put it to you plain and simple: I don’t think you and Jacob have the slightest clue what you’re talking about.
I made two claims, and I think you are objecting more strenuously to the second.
A) This report argues that a difference of 0.4 can be attributed to Nordic family policy-
Click to access 58-20073.pdf
This meta-analysis, a more critical take on natalist policy, suggests a long run increase in TFR by 4 percent for a 25 percent increase in spending- which is given increasing spending on natalist policies by factors of over 2 as these countries did, is consistent with an increase of about 0.2.
http://www.demographic-research.org/Volumes/Vol19/10/
B) The second claim, that French elites are less gung-ho about current migration because of higher fertility (and more importantly perhaps, viewing higher fertility as the result of the triumph of policy, ie something to pat themselves on the back about) was just my impression from reading recent interviews and coverage of the migrant crisis. Some French minister of something is always saying, “well the Germans have their terrible demographics but we at least have solved -that- problem.” (If your criticism of the efficacy of natalist policy is that it’s nonsense that only readers of the Economist would believe- well these people read the Economist.)
Oh, great.
Most of “we’re doomed” authors never take into account the dynamics of demography. If you use it, you’ll find out, that “not-the-first-generation” immigrants are not much different in TFR from aboriginal population. Even more, they get the “urbanisation shock”, which already stucked (nearly all) European population, futher reducing TFR. So, oversimpiplification of processes (“It all is from wellfarequeen Africans and Arabs”) don’t work.
And at last. Read it one more time.
I support the point that France is doing better than Germany. There exist NO modern urbanised populations, which managed to rise theirs FR over 2.0 for more than a couple of years. A limited success is better than none.
Anyway, it will be Amish, who will retain Earth in the long term 🙂
Or I might not find that out, actually.
Dmitriy,
You interjected yourself into an argument that apparently you didn’t understand.
Jacob thinks that French elites are more suspicious than German elites about the benefits of accepting more migrants because France has universal pre-K and Germany does not.
I asked for proof of that claim. None was provided.
Now it appears you have you own discussion going on in your head, and I’m sure it’s quite entertaining. But unless it bears upon the question of how universal pre-K makes French elites more suspicious of migrants than German elites, I’m not interested in participating in it.
You’re waging a war against a position, which exists only in your imagination. The thesis is not “Pre-K -> Less suspicion of migrants” (which you oppose), but “Pre-K -> Higher fertility of aboriginals -> Less suspicion of migrants”. Life shows, that some support of the first part (“”Pre-K -> Higher fertility of aboriginals”) exists. That’s the point that I do make.
Dmitriy,
The position in my imagination would have put up a better fight.
Greg has limited patience for his blog being used to beat on dunderheads, unless he is the one doing the beating, so I will make this my last comment.
You and a few other posters here have an exaggerated sense of what can be accomplished by expanding or universalizing pre-K. Perhaps this is because most of you are also vague about what it can accomplish. Some seem to want to make kids happier; some seem to want to ease the burden on single mothers; some want to increase the native fertility rate.
All of these potential benefits are either vague or marginal, and possibly nonexistent. But the costs in pursuing them are certain. Making you defend their merits with small pilot programs would at least have the benefit of forcing you to drop your vagueness about what you want to accomplish and putting you thought the gauntlet of a tougher empirical standard before allowing you to start spending public money by the truck load.
OK. I understand your point. What is your political affilation, by the way?
Once more. As was shown (some pages up) there’s improvement of native fertility. It’s not great. It costs great (with increase of the selling price of “25 billions per year for the USA” by 100%, we’re going to something like 50 billions per year), but really – that’s nothing compared to the already established Welfare State.
Some studies show, that there’re some improvement of children’s quality. Education works for large groups.
Summing it all, pre-K is going to (possibly) affect future generations (for better), not improve the quality of life for some old people who already have spent their lives and are trying to suck the society dry before they die.
Finally, if you’re for democracy, than I just don’t get, how’re you going to defend your point “not a dime for kindergardens” with most of your electors having children or being “do-gooders” ;).
Lots of popular programs are stupid, and their popularity does not decrease their stupidity.
Jacob,
You made a single claim that was built on a causal sequence, and you have provided absolutely no support for it. I can find nothing at all in your sources comparing German and French elite attitudes toward migrants, and partly assigning the differences in those elite attitudes to universal pre-K. So I’ll just assume you made it up.
And damn you for making me read that dreck. Pushing my way through Björklund was like listening to a three-year-old bang on a piano. I had to stop for prayer and rest at the end of every page.
Apart from the fact that your sources don’t support your specific claim about French and German differences, they’re so filled with caveats and backtracking that they don’t even support your general claim about fertility rates. Björklund says the evidence for policy affecting the spacing of children is strong, but the policy effect on fertility rates is “more uncertain” and has “more reservations.” He then paradoxically goes on to say that the available evidence still suggests a “substantial” impact on fertility.
Hoem’s meta-analysis even has this funny line, “Western society would have been quite different if economists were equally timid in offering their opinions about the usefulness of economic policies.” Stop being so timid, demographers ! Learn from the economists and don’t let your ignorance get in the way of policy-making ! Words to live by, Jacob, if you earn your bread by convincing people you have answers that you really don’t have.
This is the kind of reading that gets you up in the morning safe in the knowledge you have some coherent and empirical view of how the world works?
A straw in the wind. Columnist Tim Lott has been persuaded by Robert Plomin that heredity accounts for two thirds of intelligence. The interesting thing is that he writes about it in the opinion section of The Guardian , where such views are normally considered heresy. In consequence, Lott holds, “we have to find a way of setting ‘intelligence’ on an equal level with all the other qualities that we collectively value in human beings. Or perhaps we just should carry on not talking about it.”
I begin to foresee the Left coming clean about the true intelligence of its client groups. How about “The BBC is an equal opportunity employer and will not discriminate against any employee or applicant on the basis of age, colour, intellectual ability, gender, national origin, race, religion or sexual orientation” etc.? Or “The Prime Minister reaffirmed his personal opposition to brainism in any shape or form”? It would make life simpler all round.
With all due respect (and I mean that – it’s obvious that the pro-Headstart people in this discussion are intelligent, concerned individuals, and not at all raving loony leftists) it does look like that some folks have adopted the attitude that Headstart and Pre-K should be continued at all costs, and are casting around for reasons to keep it going, other than the reasons for which it was started, which all admit have been an utter failure. Pre-K cuts crime! It provides self-esteem! It limits immigration! It makes your breath smell fresh, and gets your laundry whiter than white! I mean, really, guys, come on. If you think that this country needs a daycare system for kids with stupid parents, let’s just admit that we need it, and design one. If I said that the F-35 was a failure as a jet fighter and could do none of the things it is supposed to do, but made a dandy roost for endangered birds, would you take this as a good reason to continue it? Couldn’t we just make some birdhouses, and save a few billion? Anton Checkov once said that, if a medicine was supposed to cure many diseases, it probably cured none of them. To an outside observer, that’s what it looks like is happening here.
I think there’s a line between small-c conservatism and panglossianism, and I’m trying to find it.
“it does look like that some folks have adopted the attitude that Headstart and Pre-K should be continued at all costs”
That doesn’t describe my position–but then, under no circumstances would I be considered pro-Headstart. I’m simply someone who doesn’t want the pendulum to swing too far.
In fact, that’s why I’m entered this conversation, because people were treating Jacob’s position as if it was pro-HS. And in joining in, now I’m characterized as pro-HS. In fact, I stated my position here:
“I am mildly pushing back against ‘Pre-k is utterly wasted because it doesn’t do any good’, to restate as ‘pre-k is a waste of time for raising academic outcomes, but whether or not we should fund preschool for other reasons is a different conversation’. The result of that conversation being unknown, and open to many different attacks.”
My positions:
1) I absolutely oppose universal preschool.
2) I would push back on Headstart funding that was justified to improve academic outcomes. My opposition is soft; that is, of all the things we waste money on, this is not a hot button.
3) I agree with Pincher that if we just give incompetent women a place to park their kids, it’s not going to discourage them from having more. On the other hand, they aren’t discouraged now. We’ve run to the end of welfare reform, it seems. So using preschool as a way to basically force women into some sort of unappealing activity like work, even in if it’s picking up trash, would probably be win win.
4) Greg says he’s not convinced something can’t improve academic outcomes. I think it’s still open at the HS level, but I’m far more convinced than he is that nothing we can do at the preschool level will improve academic outcomes.
I think the most important issue here is being ignored by most commenters: should pre-K, and Kindergarten, be mandatory? The correct answer is no. Should they be optional and paid for by public funding? The answer is, it depends.