A good friend of mine works for the Fra/ser Inst/itute (a sort of Canadian Cato), and reading this I became curious about their take on the Quebec initiative. Big surprise: not in favor. They said that on a survey of parents, institutional day care comes in dead last in terms of preferred forms of child care, &c. &c.
Oh, I don't mean to suggest they're reliable, just that I was curious what they had to say. The claim is about the preferences of Canadian parents generally, not specifically those in Quebec, supporting the notion that if you give parents money directly, they would do something other than put the kid in institutional daycare.
There's a little bit of libertarian language in that NYT column: "The big lesson from Quebec is that parents really do need more support, but they need the kind of support that allows them to choose what is best for their family. "
It seems like they're saying that daycare in general doesn't really cause problems, it's bad daycare that does. High teacher/child ratios, limited availability of half-day programs, etc.
When my mom went back to work after my younger brother was born, she started working at home and also hired someone to watch us for 4 or 5 hours in the middle of the day. It was a nice setup that allowed her to have a job without sacrificing time with the kids. Obviously not everyone can work at home, but allowing more flexible arrangements would still be helpful.
Not sure about how to deal with the quality issue, but given the benefits of having top-notch child care, spending a ton of money might be worth it.
No time to look at this research right now (will do so later, when I'm supposed to be working), but here in the UK, the Equalities Review has recently published results showing that children from all socio-economic groups benefit cognitively from high quality daycare, and that the more impoverished the family, the more the child benefits. And the American scholar Waldfogel has published something for the Centre for the Analysis of Social Exclusion (London School of Economics) presenting very strong evidence that while poor quality daycare causes problems, good quality daycare leads to better outcomes, particularly for deprived children. And evidence from Sweden and Denmark, where a very high percentage of kids are in daycare, shows excellent outcomes as well.
One thing to watch out for in the debate that will ensue from this sort of finding is the idea that staying at home with mummy is always the best thing for a baby. Up to age one, that certainly appears to be the case, but the evidence is overwhelming that high quality daycare benefits kids cognitively and developmentally after that. There is some evidence that kids in daycare have slightly more behavioural problems than those who stay at home, but that strikes me as both natural - they're hanging out with other kids, so are going to get into more mischief than kids who are home alone with mum - and a damn fine trade-off for better cognitive development. It's also important to remember that high quality daycare is not meant to be a subsitute for parental love and affection, but a complement to it. In Sweden and Denmark, for instance, the approach is pedagogical, rather than "replacement mummy". Esping-Andersen is especially good on the Scandinavian model.
(Apologies for sounding like a complete bore in this comment, btw. Am slightly obsessed by this topic.)
There are a lot of good studies that show that high quality child care (abcedarian, perry pre school etc) can make a real difference over time. Head Start, however, has a less clear cut record. In part thats because the quality of head start varies considerably and is generally not as good as it could be. One of the things that worries me about this debate in the US is that advocates use the results of the high quality programs as a lever to get legislatures to act. Then, as a result of the political process, we get programs that aren't high quality. This is a big problem.
As to the Canadian study, are they looking at individual kids over time, or at the population before and after day care for some was instituted?
The argument seems to be: Since Québéc instituted subsidized daycare, more kids in Québéc (not in daycare) are rowdy.
I wonder -- having only skimmed the study -- whether the report takes into account increased awareness and diagnosis of hyperactive children. Anecdotally, it seems that far more kids have ADD or learning disabilities than when I was in public schools, but that's often attributed to pushy school districts, newer treatments, and anxious parents.
Oh, 7 was me. Apropos of nothing, I'd also like to ban the phrase 'raised by daycare' from the public debate, enforced by calabat. No one I know who uses daycare decided to plop out a kid and give it to the daycare workers to raise while they got on with their swinging martini-and-stiletto lifestyle.
"it compares children in Quebec before the program to children in Quebec after the program."
Haven't read the actual paper yet, but if this is true, it sounds like BS- obviously you're going to have more low-income kids in daycare after the program, who have more problems in general.
There's quite a bit of research in the States, too, that shows that good daycare is beneficial.
The real problem with the whole daycare debate is that it always gets set up as institutional daycare vs. mom staying home, like those are the only two options. But most of the parents I know have used lots of different childcare arrangements; I know mine depended a lot on what PK's age was at the time, what we could afford, and how many hours we needed.
When Keegan went through daycare, it was a university-funded research program that priced on an income-based sliding scale. When my wife and I divorced, that was based on just one of our incomes. Now that Noah's in daycare, and was born out of the date range to enter the program Keegan was in, we're paying market costs for it. He's in one of the more affordable ones, in an area where cost of living isn't that high, and we're still dropping $900 a month on it.
It's absolutely killing our monthly budget and we're both making decent middle-class salaries. I can't imagine how working-class parents could even begin to afford it.
My understanding is that head start teachers are paid crapola, and that there are thus not loads of highly qualified ones. The European research shows that child outcomes are strongly correlated with educational levels of daycare teachers. What this means in the big picture is that, in order to provide universal high quality care, governments need to acknowledge that they're going to have to spend a fat wodge of cash on it every year, and need to have the balls to remember that by doing so they're being pound-, rather than penny-, wise. (Most recent results from Perry say that gov't saves $12 to every $1 spent.)
A major issue in the UK is that the government is expanding childcare places, but is focusing on quantity alone, rather than quantity and quality. Most daycare workers are paid less than supermarket checkout people, so the quality of care tends to be poor, and staff turnover high. As someone implies above, even if the US does expand daycare provision, they're likely to make the same mistake, but even worse.
12: Or how flexible even a daycare arrangement may be. Friends of mine enrolled their son in a daycare while they are finishing their PhDs. He's there three days a week, and the rest of the time, he has essentially two stay-at-home parents. Does he get more or less parental attention than the stereotypical dad's-at-work, mom's-at-home? Damned if I know.
To 11, given immigration etc, that doesn't seem like a great study design to me.
To 14, exactly. There is a great study of Oklahoma (Oklahoma!?!) preschool. They use a center based model, but the teachers are all certified and employees of the school district. That also means that the training and professional development are aligned with what's necessary for success in school. This gets into the battle over whether head start should be an education program or a child wellness/development program though.
a key point to the study, as I understood the article, was the timing of the day care. For infants and toddlers, it was problematic. For pre-schoolers and above, it was actually seen as beneficial. So maybe it's more of a timing issue than anything else.
Huh. Neat. Michael Baker taught me labo(u)r economics in grad school. Milligan is a former student of his too. I haven't read the study (getting home way too late for that). But they don't do shoddy, or kneejerk-ideological work. I'll be interested to see their methodology.
The study seems to have handled the issues as well as is econometrically possible with this data. In the actual paper (not the summary linked in the post) they're very up front about some limitations - that is results, while the first for a universal day care program, may not generalize to other populations. In particular, they only study households with two married parents because of confounding changes in policies affecting single mothers. This means they miss any benefits of this program to single mothers, but they argue that the program mainly benefited middle and high income parents because there was a lot of day care assistance for low income parents. They do additional analysis to rule out some alternative explanations of their findings.
The most stable results seem to be that scores of anxiety and motor/social skills fall for children under this universal day care program, and mothers are significantly more likely to suffer from depression. Oh, and there's a drop in measures of the quality of parents' interactions with their kids (not shocking to anyone who's stood around a playground with lots of parents with nannies).
One thing that bothers me is how the authors deal with that depression result for mothers. When economists encounter results that threaten their most cherished beliefs, they're quick to point out that people tend to act in their own best interest, and that we may not be observing other reasons for their behavior. Yet they don't point out that, still these mothers are choosing to enter the work force and so they must be getting significant financial and/or personal benefits from it that may outweigh the risk of depression.
In particular, Quebec seems to have had significantly lower work force participation by women before the program and jumped to higher than average (for Canada) after the program. These mothers may have been entering workplaces that with more gender problems and fewer family-friendly policies than they would face in other cases. This could be another adjustment cost of the program, which would make it a temporary rather than a permanent effect.
BTW, this study is quasi-experimental, in the sense that the subjects sorted themselves into treatment groups (that is, Quebec and the other provinces got to decide whether or not to adopt universal day care). This always introduces the potential for unobserved bias. The authors try to address a number of potential ones, including using that control group of older kids who didn't have the opportunity offered by universal day care, but can't rule out all of them.
There are some great examples of problems that arise with this sort of design (by necessity the most common in the social sciences) in the science column of today's WSJ. I'm not linking because it's behind a subscription wall, but the column goes through a bunch of health and psych results on the bad effects of divorce, smoking during pregnancy, and teen pregnancy, that all disappear in twin studies that account for the genetic predisposition that may lead to both the treatment (divorce) and the outcomes (depression, health or behavior problems).
Hrm. I have to say that I am systematically suspicious of twin studies given the difficulty of actually coming up with any significant population of separated twins to study.
This:
One thing that bothers me is how the authors deal with that depression result for mothers. When economists encounter results that threaten their most cherished beliefs, they're quick to point out that people tend to act in their own best interest, and that we may not be observing other reasons for their behavior. Yet they don't point out that, still these mothers are choosing to enter the work force and so they must be getting significant financial and/or personal benefits from it that may outweigh the risk of depression.
A good friend of mine works for the Fra/ser Inst/itute (a sort of Canadian Cato), and reading this I became curious about their take on the Quebec initiative. Big surprise: not in favor. They said that on a survey of parents, institutional day care comes in dead last in terms of preferred forms of child care, &c. &c.
Posted by ac | Link to this comment | 06-14-06 9:14 AM
Hmm. I wonder if part of what's going on is that the Quebec daycare particularly stinks.
Posted by LizardBreath | Link to this comment | 06-14-06 9:24 AM
Oh, I don't mean to suggest they're reliable, just that I was curious what they had to say. The claim is about the preferences of Canadian parents generally, not specifically those in Quebec, supporting the notion that if you give parents money directly, they would do something other than put the kid in institutional daycare.
There's a little bit of libertarian language in that NYT column: "The big lesson from Quebec is that parents really do need more support, but they need the kind of support that allows them to choose what is best for their family. "
Posted by ac | Link to this comment | 06-14-06 9:30 AM
It seems like they're saying that daycare in general doesn't really cause problems, it's bad daycare that does. High teacher/child ratios, limited availability of half-day programs, etc.
When my mom went back to work after my younger brother was born, she started working at home and also hired someone to watch us for 4 or 5 hours in the middle of the day. It was a nice setup that allowed her to have a job without sacrificing time with the kids. Obviously not everyone can work at home, but allowing more flexible arrangements would still be helpful.
Not sure about how to deal with the quality issue, but given the benefits of having top-notch child care, spending a ton of money might be worth it.
Posted by Matt F | Link to this comment | 06-14-06 9:31 AM
No time to look at this research right now (will do so later, when I'm supposed to be working), but here in the UK, the Equalities Review has recently published results showing that children from all socio-economic groups benefit cognitively from high quality daycare, and that the more impoverished the family, the more the child benefits. And the American scholar Waldfogel has published something for the Centre for the Analysis of Social Exclusion (London School of Economics) presenting very strong evidence that while poor quality daycare causes problems, good quality daycare leads to better outcomes, particularly for deprived children. And evidence from Sweden and Denmark, where a very high percentage of kids are in daycare, shows excellent outcomes as well.
One thing to watch out for in the debate that will ensue from this sort of finding is the idea that staying at home with mummy is always the best thing for a baby. Up to age one, that certainly appears to be the case, but the evidence is overwhelming that high quality daycare benefits kids cognitively and developmentally after that. There is some evidence that kids in daycare have slightly more behavioural problems than those who stay at home, but that strikes me as both natural - they're hanging out with other kids, so are going to get into more mischief than kids who are home alone with mum - and a damn fine trade-off for better cognitive development. It's also important to remember that high quality daycare is not meant to be a subsitute for parental love and affection, but a complement to it. In Sweden and Denmark, for instance, the approach is pedagogical, rather than "replacement mummy". Esping-Andersen is especially good on the Scandinavian model.
(Apologies for sounding like a complete bore in this comment, btw. Am slightly obsessed by this topic.)
Posted by reuben | Link to this comment | 06-14-06 9:49 AM
There are a lot of good studies that show that high quality child care (abcedarian, perry pre school etc) can make a real difference over time. Head Start, however, has a less clear cut record. In part thats because the quality of head start varies considerably and is generally not as good as it could be. One of the things that worries me about this debate in the US is that advocates use the results of the high quality programs as a lever to get legislatures to act. Then, as a result of the political process, we get programs that aren't high quality. This is a big problem.
As to the Canadian study, are they looking at individual kids over time, or at the population before and after day care for some was instituted?
Posted by benton | Link to this comment | 06-14-06 9:51 AM
The argument seems to be: Since Québéc instituted subsidized daycare, more kids in Québéc (not in daycare) are rowdy.
I wonder -- having only skimmed the study -- whether the report takes into account increased awareness and diagnosis of hyperactive children. Anecdotally, it seems that far more kids have ADD or learning disabilities than when I was in public schools, but that's often attributed to pushy school districts, newer treatments, and anxious parents.
(Did Québéc schools eliminate gym classes, too?)
Posted by Anonymous | Link to this comment | 06-14-06 9:52 AM
Oh, 7 was me. Apropos of nothing, I'd also like to ban the phrase 'raised by daycare' from the public debate, enforced by calabat. No one I know who uses daycare decided to plop out a kid and give it to the daycare workers to raise while they got on with their swinging martini-and-stiletto lifestyle.
Posted by Cala | Link to this comment | 06-14-06 9:55 AM
The latter.
Posted by LizardBreath | Link to this comment | 06-14-06 9:55 AM
"it compares children in Quebec before the program to children in Quebec after the program."
Haven't read the actual paper yet, but if this is true, it sounds like BS- obviously you're going to have more low-income kids in daycare after the program, who have more problems in general.
Posted by SP | Link to this comment | 06-14-06 9:56 AM
Sorry, that was to benton's 6. The study compared Quebec before and after the program, rather than any cohort of kids over time.
Posted by LizardBreath | Link to this comment | 06-14-06 9:56 AM
There's quite a bit of research in the States, too, that shows that good daycare is beneficial.
The real problem with the whole daycare debate is that it always gets set up as institutional daycare vs. mom staying home, like those are the only two options. But most of the parents I know have used lots of different childcare arrangements; I know mine depended a lot on what PK's age was at the time, what we could afford, and how many hours we needed.
Posted by bitchphd | Link to this comment | 06-14-06 9:58 AM
When Keegan went through daycare, it was a university-funded research program that priced on an income-based sliding scale. When my wife and I divorced, that was based on just one of our incomes. Now that Noah's in daycare, and was born out of the date range to enter the program Keegan was in, we're paying market costs for it. He's in one of the more affordable ones, in an area where cost of living isn't that high, and we're still dropping $900 a month on it.
It's absolutely killing our monthly budget and we're both making decent middle-class salaries. I can't imagine how working-class parents could even begin to afford it.
Posted by apostropher | Link to this comment | 06-14-06 10:06 AM
Head Start, however, has a less clear cut record
My understanding is that head start teachers are paid crapola, and that there are thus not loads of highly qualified ones. The European research shows that child outcomes are strongly correlated with educational levels of daycare teachers. What this means in the big picture is that, in order to provide universal high quality care, governments need to acknowledge that they're going to have to spend a fat wodge of cash on it every year, and need to have the balls to remember that by doing so they're being pound-, rather than penny-, wise. (Most recent results from Perry say that gov't saves $12 to every $1 spent.)
A major issue in the UK is that the government is expanding childcare places, but is focusing on quantity alone, rather than quantity and quality. Most daycare workers are paid less than supermarket checkout people, so the quality of care tends to be poor, and staff turnover high. As someone implies above, even if the US does expand daycare provision, they're likely to make the same mistake, but even worse.
Posted by reuben | Link to this comment | 06-14-06 10:17 AM
12: Or how flexible even a daycare arrangement may be. Friends of mine enrolled their son in a daycare while they are finishing their PhDs. He's there three days a week, and the rest of the time, he has essentially two stay-at-home parents. Does he get more or less parental attention than the stereotypical dad's-at-work, mom's-at-home? Damned if I know.
Posted by Cala | Link to this comment | 06-14-06 10:21 AM
To 11, given immigration etc, that doesn't seem like a great study design to me.
To 14, exactly. There is a great study of Oklahoma (Oklahoma!?!) preschool. They use a center based model, but the teachers are all certified and employees of the school district. That also means that the training and professional development are aligned with what's necessary for success in school. This gets into the battle over whether head start should be an education program or a child wellness/development program though.
http://www.crocus.georgetown.edu/reports/executive_summary_11_04.pdf
Posted by benton | Link to this comment | 06-14-06 10:36 AM
Oklahoma!?!
Shocking, I know.
Posted by Tarrou | Link to this comment | 06-14-06 11:17 AM
Re 17 -- I kid. But it is interesting to see Oklahoma do this while other, wealthier and more service oriented states trying to do things less well.
Posted by benton | Link to this comment | 06-14-06 11:23 AM
Oh neat, another day care study. I have yet to see one that doesn't support the following hypotheses:
1. If the choice is between good parental care and shitty day care, parental care is better.
2. If the choice is between good day care and shitty parental care, day care is better.
3. If both day care and parental care are good, kids do fine however they spend their time.
4. If both day care and parental care are shitty, kids don't do as well.
5. If both day care and parental care are mediocre, results vary depending on individual circumstances.
Can I have my grant now?
Posted by DaveL | Link to this comment | 06-14-06 12:27 PM
a key point to the study, as I understood the article, was the timing of the day care. For infants and toddlers, it was problematic. For pre-schoolers and above, it was actually seen as beneficial. So maybe it's more of a timing issue than anything else.
Posted by tweedledopey | Link to this comment | 06-14-06 3:56 PM
Huh. Neat. Michael Baker taught me labo(u)r economics in grad school. Milligan is a former student of his too. I haven't read the study (getting home way too late for that). But they don't do shoddy, or kneejerk-ideological work. I'll be interested to see their methodology.
Posted by Carl | Link to this comment | 06-14-06 10:24 PM
For infants and toddlers, it was problematic. For pre-schoolers and above, it was actually seen as beneficial.
No! You don't say. What a shocker!
Posted by bitchphd | Link to this comment | 06-14-06 10:45 PM
The study seems to have handled the issues as well as is econometrically possible with this data. In the actual paper (not the summary linked in the post) they're very up front about some limitations - that is results, while the first for a universal day care program, may not generalize to other populations. In particular, they only study households with two married parents because of confounding changes in policies affecting single mothers. This means they miss any benefits of this program to single mothers, but they argue that the program mainly benefited middle and high income parents because there was a lot of day care assistance for low income parents. They do additional analysis to rule out some alternative explanations of their findings.
The most stable results seem to be that scores of anxiety and motor/social skills fall for children under this universal day care program, and mothers are significantly more likely to suffer from depression. Oh, and there's a drop in measures of the quality of parents' interactions with their kids (not shocking to anyone who's stood around a playground with lots of parents with nannies).
One thing that bothers me is how the authors deal with that depression result for mothers. When economists encounter results that threaten their most cherished beliefs, they're quick to point out that people tend to act in their own best interest, and that we may not be observing other reasons for their behavior. Yet they don't point out that, still these mothers are choosing to enter the work force and so they must be getting significant financial and/or personal benefits from it that may outweigh the risk of depression.
In particular, Quebec seems to have had significantly lower work force participation by women before the program and jumped to higher than average (for Canada) after the program. These mothers may have been entering workplaces that with more gender problems and fewer family-friendly policies than they would face in other cases. This could be another adjustment cost of the program, which would make it a temporary rather than a permanent effect.
Posted by cw | Link to this comment | 06-16-06 8:38 AM
BTW, this study is quasi-experimental, in the sense that the subjects sorted themselves into treatment groups (that is, Quebec and the other provinces got to decide whether or not to adopt universal day care). This always introduces the potential for unobserved bias. The authors try to address a number of potential ones, including using that control group of older kids who didn't have the opportunity offered by universal day care, but can't rule out all of them.
There are some great examples of problems that arise with this sort of design (by necessity the most common in the social sciences) in the science column of today's WSJ. I'm not linking because it's behind a subscription wall, but the column goes through a bunch of health and psych results on the bad effects of divorce, smoking during pregnancy, and teen pregnancy, that all disappear in twin studies that account for the genetic predisposition that may lead to both the treatment (divorce) and the outcomes (depression, health or behavior problems).
Posted by cw | Link to this comment | 06-16-06 9:16 AM
Hrm. I have to say that I am systematically suspicious of twin studies given the difficulty of actually coming up with any significant population of separated twins to study.
This:
One thing that bothers me is how the authors deal with that depression result for mothers. When economists encounter results that threaten their most cherished beliefs, they're quick to point out that people tend to act in their own best interest, and that we may not be observing other reasons for their behavior. Yet they don't point out that, still these mothers are choosing to enter the work force and so they must be getting significant financial and/or personal benefits from it that may outweigh the risk of depression.
I thought, was an excellent point.
Posted by LizardBreath | Link to this comment | 06-16-06 9:23 AM