National Post

Page URL:

Tuesday, October 19, 1999

Daycare statistics don't add up
David Frum
National Post

Does daycare make kids smarter? How very convenient it would be if it were true! Now the ever-obliging folks at Statistics Canada have actually produced a report that purports to prove that children who spend time in care perform better in the first grade of school than children who stay at home. So you see, there's no need to feel guilty about working late at the office: Your children are actually better off without you.

Or are they? In recent years, Statistics Canada has manifested a disturbing tendency to twist and jumble data on the subjects of marriage and child-rearing in suspiciously predictable directions. The daycare-makes-you-smart report is one of the worst examples yet of this disturbing trend.

The report, From Home to School, by Garth Lipps and Jackie Yiptong-Avila, uses data from the National Longitudinal Survey of Youth, a grand study of the lives of 20,000 children under age 13 that began in 1995-1996. Mr. Lipps' and Ms. Avila's assessment of this rich trove of information is flawed in three alarming ways. First, their report fails to offer a clear definition of what they mean by "child care." All that the authors offer by way of definition is the fuzzy observation that "nursery schools, kindergartens, mom and tot programs, play groups, structured and unstructured daycare programs are all popular options that parents have for their young children in Canada." So, of course, they are -- but that's no justification for lumping them all together when assessing their impact on children's cognitive development. Five hours per week of mom-and-tot play is a very different thing from 50 hours a week of institutional daycare, and no important question can be answered by aggregating the two together.

The report's second flaw is very nearly as serious. Suppose you read a report in the newspaper claiming that children whose parents drive sports utility vehicles do better in school than children whose parents take the bus to work. Would you rush out to borrow the money for an SUV in order to boost your kids' academic attainments? Or would you figure that SUV-ownership is very likely a proxy for something more relevant?

Educated and affluent parents are much more likely to send their children to preschool programs than their less educated, less affluent counterparts. Children whose mothers had university degrees were four times as likely to attend preschool of some sort than children whose mothers had dropped out of high school. So when Mr. Lipps and Ms. Avila find that 26% of kids who spent some time in non-maternal care are described by their teachers as being "near the top" of their class, while only 16% of at-home kids are ranked "near the top," you have to wonder: Are they really describing the effects of preschool? Or the effects of having a well-educated mother?

And (flaw number three) the already dubious difference between the at-home and the in-care kids is even more doubtful than Mr. Lipps and Ms. Avila acknowledge. Comparatively few Canadian children reach the first grade without experiencing at least one of the disparate early-childhood programs that the report bizarrely lumped together. The National Longitudinal Survey of Youth' s sample of at-home kids is therefore considerably smaller than its sample of in-care kids.

Small samples are much less reliable than large samples. How much less reliable? In a breathtaking footnote, Mr. Lipps and Ms. Avila warn, "Coefficients of variation are between 16.4% and 33.3% suggesting that these estimates should be used with caution." Translated into plain English, that footnote means that the number of at-home kids "near the top" of their first-grade class may not be 16%. It may in fact be as high as 21.3%. In other words, the big gap in achievement the report headlines may barely exist at all.

This is not to knock preschool. There is strong evidence that children from economically disadvantaged backgrounds benefit from enrichment programs outside the home. But it is disingenuous to the point of dishonesty for researchers to blur their findings in such a way as to imply that, because a three-year-old from a poor home benefits from a couple of hours a day of block-play and story-reading at the Y, non-poor parents should therefore feel they're doing their kid a favour when they consign him or her to eight hours a day at the local child-storage facility.

Copyright Southam Inc.