Lindsey Burke and Salim Furth write,
New studies of large-scale preschool programs in Quebec and Tennessee show that vastly expanding access to free or subsidized preschool may worsen behavioral and emotional outcomes.
You may have seen it claimed elsewhere that pre-K has been proven to work. Read the whole article. I am reminded of David Weinberger’s observation that nowadays for every fact there is an equal and opposite fact.
Meanwhile, long live the null hypothesis.
It is amazing how they state these things as fact and settled science.
Even the Heckman stuff I interpreted as the benefits going mostly to the furthest behind kids. It is not a stretch to this result from even the miniscule evidence they hang their hats on.
Harry Harlow’s findings replicated with human infants.
The problem these days is that we can’t even agree on what the facts are.
I think the problem is that one side can run complete BS up the flagpole and if people don’t immediately salute they just move to a different flagpole before the facts can even be collected. The real scandal is that side also claims the mantle of being the party of science if you don’t accept something off of insufficient facts.
Actually, reading these has made a left center voter turn against universal pre-school and most likely young children benefit most from living in married couple supported one income. (Either wife or husband stay home raising young children.) Of course, with falling working class wages how do we economically support that in the long run or is the economic gains worth slightly less educated kids?
However, are there any pre-school studies on partial pre-school? In my experience, kids benefited from limited pre-school (say 4 – 8 hours a week) versus all day sessions. Simply put, pre-school for 3 – 5 year olds, has a lot of diminishing returns to the children.
It is almost a miracle, but why were you for it?
Universal Pre-School sounds good on paper and as I noted the kids I have known (including my three kids) benefited from the experience. And almost every Kindergarten teacher I know has stated they know which kids had pre-school and which ones did not. However after reading the studies, it is wise to let the stats do the talking.
I also thought about pre-school experience that maybe limited pre-school and 40 hour pre-school are not necessarily the same thing. 4 year kids might really benefit from a weekly 4 hour session or twice a week 3 hour sessions to learn the basics of alphabet, numbers and learning to behave in group settings is all that necessary. (And after 5 hours there is diminishing returns on the pre-school and kids would benefit with their time elsewhere.)
“almost every Kindergarten teacher I know has stated they know which kids had pre-school and which ones did not. ”
That does not mean it is necessarily good.
I have thus theory that, at least for some kids, good boys esoecially, traditional classroom schooling causes some mild psychological dissociation. Just one of my pet theories.
The best pre-school experience that any of my kids had was 3 hours/day 4 days a week. I do not think more than that has any chance at all of being incrementally helpful for kids and considering that maybe none of it is helpful I feel pretty confident in this view. A problem is that such a schedule is not as convenient as many parents want. Many see this as a child care function and expect 7 or 8 hours a day every day of the work week. This is a ludicrous amount of school for a 3 or 4 year old and I have my doubts it is very valuable for 5 year olds either.
The median income will always afford the median home. It is more of a collective action problem. If every wife worked for in home consumption rather than in the taxed economy the median home might be smaller and the median meal have less meat or fish and the median air conditioner be set a little higher in summer and heat lower in winter, median shoes a little cheaper but we could certainly afford it.
A large tax credit for kids under 6 and then killing child care deductions I think would give parents more options. I still have one in this age group but he’ll be out before such a law could be implemented. Say $5000 a year credit in addition to the standard deduction. I do not like the idea that funds will be made available for families provided they select a particular manner of raising their family. Better to simply hand over some money and let families figure out how to use it. The day care industry will object, but I don’t see that as a bad thing.
Helping kids is a collateral benefit if it happens since the motivation is mostly to help out adults of two categories: parents who get the equivalent of free childcare for part of the day and teachers’ unions who stand to pick up membership. Educational results are really unimportant, but advocates don’t yet feel confident enough to just sell their top motivations on this. If this was part of an overall plan to reform education and make it pre-K-10 for example instead of K-12 I think we would have heard about that by now.
Anecdotes from those I know who work in early childhood education line up with what this study reports. In fact, the more time a child spends in such programs, usually the worse their behavior is.
Of course, this was in private schools, and in such settings the it seemed behavior seemed worst among those households with the highest incomes. It was explained to me that in such cases both parents were working, often at high-powered jobs who were basically willing to pay top-dollar for someone *else* to raise their kids.
When it comes to raising well-adjusted adults, education should be seen as a supplement to parenting, but it is increasingly seen as a replacement.
That is fantastic!!! Long live non-linear systems and applied-science. So here we have three large scale and careful studies that completely contradict one another, or more accurately, contradict the assumptions the Michigan study researchers expressed in their original conclusions/recommendations. My favorite part is that this is the best (only?) evidence I’ve seen that contradicts Judith Harris’ Nurture Assumption (parents/teachers have zero impact on personality/life-outcomes).
Of course, Harris says that her theory only applies in non-abusive environments and that might be the key. A new hypothesis might be that pre-school aged children are highly susceptible to traumatic experiences and that each of the three studied pre-school systems inadvertently introduced or reduced traumatic experiences somehow.
The bridge failed again and the social scientists here all revert back to lazy/dismissive explanations of how this is all unknowable. Like heck its unknowable, the only failure is our inability to appreciate how hard it is to understand complex non-linear systems. Its a long schlep trying to figure this stuff out and these results are just outstanding and may be the basis of a major insight. Either something in the experiment design/execution was wrong or there is an alternative set of one or more models that fit the results of all three studies. Wow. Cool 🙂
The Quebec study looks at day care, not pre-school. Nice if you are interested in day care, but not sure why you would lump it in when talking about early education. (Actually, I do know. Just kind of sad to see people do this.) Also, would be interesting to see these studies get carried out further. They pretty much confirm, so far, the earlier studies that are being criticized. Those studies also showed a loss of effects very early when it came to test scores. It was the effects seen in adulthood that were most interesting.
Probably because there’s really no substantive difference between pre-school and daycare for 3-4 year olds.
Thus is why they want kids when they are just about ready for traditional school.
Steve, you speak as if 3k would be like 5k instead of 3k being like 3 yo daycare. Why?
On top of that you attribute the common sense position to be some kind of dishonesty. Why?
Maybe if there were actual studies that were good to hang thus on we’d use those. But until we have actual studies looking at this thing they want to force everyone to do all of an instant, we’ll use these.
Would the null hypothesis be consistent with a theory that there is some optimal approach for each individual student, but that applying a top down approach on all students increases the probability that on average, some benefit/some don’t and the net effect is nothing.
Is it consistent with any approach having basically no net difference on a given student?
With both?
I’m simultaneously very confident that for a given student, there is some “optimal” set of conditions while highly doubtful that applying student A’s optimal approach will be optimal for student A’s classmates. I wonder how much of the “null” effect is just the failure of a “non-precision” approach.