Child Care and Early Intervention (Redux)

I somehow neglected to write a post on Jennifer Buckingham’s new CIS childcare report when it came out last week.* The report surveys the evidence on child care, and concludes that we have very little high-quality evidence on whether formal childcare is good or bad for kids. Much more articulately than my recent ramblings on the topic, Jen points out that the really robust evidence relates to randomised trials of early intervention programs for very poor kids, but it would be silly to extrapolate from those to regular childcare.

• American studies regularly cited to support the argument that child care is widely beneficial include the High Scope Perry Preschool Project, the Abecedarian Project, Project CARE, Head Start and Early Head Start. Each of these studies involved children from low-income or disadvantaged families, who were given a combination of centre-based child care and home visits and, in some cases, health and parenting services. The results achieved were significant but they cannot be expected to be replicated with the broader population.

• Studies that have involved a more representative population provide a less conclusive and more cautionary picture of the effects of child care. Some, including the US National Institutes of Child Health and Development (NICHD) study, have found risks associated with early child care. Australian research is relatively scarce but is equally mixed, and the effect sizes have also been relatively small. Much research has focused on the quality of child care and has concluded, unsurprisingly, that high-quality child care is better than low-quality child care, but has not shown that any quality of child care is superior to parental care.

• This paper concludes that there is insufficient evidence to believe that, in general, even high-quality formal child care in the early years is either beneficial or harmful to children in the long term. The oft-claimed developmental, social and economic impacts are by no means guaranteed.

These findings accord with one of the nicest economics papers on the topic – Janet Currie’s 2001 JEP paper, which stated:

These findings suggest that the payoff to early intervention is greatest for the most disadvantaged children.

and concludes:

The available evidence sheds less light on the wisdom of establishing a universal public preschool program. Such a program would be costly and would provide a large child care subsidy to many middle- and upper-income families, rather than targeting benefits primarily towards the neediest children. However, such a program might enjoy greater popular support than one targeted only to needy children.

* Disclosure: I commented on a draft of Jen’s report.

This entry was posted in Economics of Education, Economics of the Family. Bookmark the permalink.

10 Responses to Child Care and Early Intervention (Redux)

  1. Patrick says:

    I agree with that conclusion. One possible benefit from a more universal scheme (perhaps the only one) arises if the universal scheme can increase ‘mixing’ between socioeconomic groups- but French experience suggests that it is very hard to provide for genuine mixing.

  2. derrida derider says:

    Umm, absence of evidence is not evidence of absence unless someone has bothered looking for the evidence. We know early intervention for disadvantaged kids has a big payoff, and we know there’s little evidence either way of the effects of early chidlcare on more advantaged kids, mainly because such evidence is much harder to get.

    But how do we get from that to “universal childcare is a Bad Thing”? Especially as the merits or demerits of universal childcare aren’t usually argued for on the basis of the outcones for the kids but the outcomes for the mum.

  3. Leon says:

    Perhaps I’m wrong, but arguments for universal early education (and universal education generally) seem to be about equity of outcomes for those being educated, rather than equity of outcomes for their parents/mothers.

    But that aside, universal early education has a huge cost – both a fiscal cost, and a cost in the extra incentive to spend less time with one’s children. Perhaps that comes back to the outcomes for mothers issue – although it’s seems to conflict with the logic of maternity awards, etc.

  4. Leon says:

    (Sorry, previous post incomplete)

    The cost is why you should have evidence of non-absence, i.e. evidence of effect (including the outcomes for mothers) before implementing the policy. Any intervention is a bad thing until proven good, because it has definite costs and uncertain benefits.

  5. conrad says:

    DD — I think you need to reconceputualize childcare in terms of the quality of the childcare and individual differences of the child, rather than just low and high SES.
    I think there is a fair bit of evidence to show that if you go from average to excellent childcare (at least early on), it doesn’t make much difference (Hernstein & Murray use this as a rather poor argument to conclude IQ is everthing in their controversial book). This makes developmental sense, because it shows that kids will learn well under innumerate different situations. No doubt a big benefit is got just from kids playing with other kids (it would be interesting to know how bad “bad” would have to be before you start losing such benefits). Thus providing some basic childcare to all might be worthwhile, even if it isn’t very good.
    In terms of individual differences, my impression from people doing stuff on kids with literarcy is that the smarter the kids, the bigger the benefit. So if the kids are smart but have had problems for whatever reason, you’ll get a bigger bang for you buck. Of course, it probably isn’t very ethical to stop dumb kids entering the same programs, even if it doesn’t help them to the same extent.

  6. backroom girl says:

    I’m still not convinced that any study that tells you that the average effect of x is y (statistically significant) is really of any practical help to anyone trying to make a decision about what is best for their own lives and those of their family. This is because it is highly unlikely that an average difference of y arises from everyone’s outcomes changing by y. Or am I missing something?

    I don’t know whether, in the end, if I had stayed at home instead of going back to work (full-time) and having a proper career, my children would have turned out to be world-beaters. (Seeing as how they are both female, though, they would probably have had to stay at home to raise the next generation of hopefully male world-beaters.)

    As it is, they are both doing OK and I have had a career and am looking forward to a well-funded retirement. As far as I can tell, most studies don’t add up the outcomes of the mother and the children to work out whether the family as a whole is better off. Most just look at the kids’ outcomes and continue to treat mothers as just an input to the process, rather than a person with her own outcomes to worry about.

  7. conrad says:

    BG: There’s lots of programs which are useful — it just depends on what you want out of them. If your kids couldn’t read or do maths well, for example, you’d want to know if plan X would solve this problem, or at least have some meaningful benefit. Perhaps some people can have a happy life being illiterate, but almost all people will have an even happier one being literate. Alternatively, whether getting 1 extra IQ point by using your life savings to pay for some fancy get-ahead program is another story.

    A lot of the problems with these programs is just the advertising and to whom they are targeted. There’s lots of data floating around about individual differences, and for some areas, there are now tests that allow some predictions as to what sort of program is likely to have the most efficacy. If I can identify certain speech/language processing difficulties in a 3 year old, for example (of which there are different types), I could probably try and devise something that might help based on that information.

    I think the problem is that the main group of people that know about this sort of stuff (and have the stuff to do the testing, for that matter) are either academics who you’ll never be able to ask in case you need to, and even if you could, probably wouldn’t help you (I’ll add myself category, just to be honest), people trying to make money out of these programs, who have no interest in telling you the truth (or running expensive programs that work well), or government beuracrats, who also often don’t appear to know anything more than the sales people tell them, and don’t appear to be able to evaluate what often amounts to a large body of complex literature. If you didn’t live in Australia, you also might be able to find more educational/child psychologists that knew about educational problems, although they arn’t plentiful anywhere I know.

    Given this, I don’t think the problem is lack of need or lack of programs that might work. If you looked at any population, you’d probably find 5-10% of kids that could significantly benefit from something useful (as in my example of reading and doing maths) that no-one is ever going give them (nor arguable about whether it would be useful having).

  8. christine says:

    BG: every time I read one of these studies (esp the one that suggests daycare is not good for kids of highly educated mothers, but of course pretty much doesn’t bother correlating with dads’ education), I worry about my own choices, but the studies truly don’t tell you anything useful on that. The thing is, what’s good for my particular situation is not the issue with funding universal childcare – it’s whether there’s enough average (social) benefit to the kids that it’s worth funding, which is one reason the studies focus on it (also because they just can’t identify the effects on every kid).

    You’re right about the net benefits of mum working not being really considered much – another eg, the ‘option’ value of having worked meaning that you can find a job in the case of divorce/breadwinner hubby dying.

    From an individual perspective, my bottom line take on the research here is really, we can’t find much effect of any particular part of child raising – what matters is the whole package. So whatever decision you make is probably just fine.

  9. conrad says:

    “we can’t find much effect of any particular part of child raising”

    Like I said before. Yes you can. You can find markers for particular things and henceforth fix them or not. Thirty years of cognitive neuroscience and hundreds of years of psychology has taught us a lot.

  10. backroom girl says:


    Don’t worry, I’m not really concerned that I personally have made wrong decisions about what to do with my life – I’m certainly not going to beat myself up about the fact that my kids might have ended up with an average .5 of an additional year of education, or something, if I’d worked part-time instead of full-time. I’m a fairly strong believer (probably without an scientific basis) that if mothers are broadly happy with the choices they have made, whatever those choices, the kids will probably be OK. On the other hand, if Mum is stressed out and/or constantly worried about the effect of her choices on her kids, that will probably help to screw them up.

    I can also understand that from the point of view of government, all that matters might be the average impact of some program when it comes down to thinking about whether to fund it or not. But the real issue for me, in all these areas, is that the things that work for one person are not the same as what works for someone else. Some kids thrive in formal child care and others do not, regardless of whether the average outcome is positive or negative. But in the end, government usually prefers to implement one-size-fits-all solutions, rather than offering a range of options along with assistance for people to find out what will suit them (and their kids) best.

Comments are closed.