Parents on a guilt trip after reading last week's widely reported news that nonmaternal child care and education cause aggressive behavior in children as late as sixth grade should relax a little. That's because reaching such a sweeping conclusion from one limited study needlessly exacerbates the worries of working moms and dads who have few good alternatives for affordable child care outside the home.
Reporting on studies such as this is no simple task. Yet it is incumbent upon those whose job it is to interpret social science for a lay audience to take into account the broader body of scientific knowledge as well as the strengths and weaknesses of the methodology employed by the researchers. When that does not occur, there is the danger the reporting will produce more heat than light - and in this case, unnecessary alarm among working parents.
The Study of Early Child Care and Youth Development, sponsored by the National Institute of Child Health and Human Development, tracks more than 1,300 children from a variety of child care settings as they progress through school.
One finding of the latest report from that study was that the more time children spent in center-based care before kindergarten, the more likely their sixth-grade teachers were to report problem behaviors such as getting in fights, disobeying and arguing. Other research shows that may or may not be the case.
We don't know for certain because the national institute's study is not well designed to reach such a broad conclusion. Unlike a randomized trial (the gold standard for this kind of research), this study cannot really tell us whether child care centers caused the problem behaviors.
Other factors might have caused the result they found, and the authors acknowledge as much in their report. Some in the media neglected to report this important context.
Another serious, technical issue is that the study conducted eight separate analyses to test for effects of participation in center care on three cognitive and five social-emotional outcomes. The only finding that had significant results was for one type of behavior problem, and the result was incredibly small. Thus, if real, the negative effect is highly limited and does not extend to other measures, such as getting along with others and self-control. Other research has demonstrated positive effects on social-emotional development in better-quality centers.
One danger of statistical studies such as this is that the sheer number of comparisons made increases the chance that one small finding can occur purely by chance. In fact, after making an appropriate adjustment for multiple comparisons, this effect on problem behavior is not reliably different from zero.
Finally, it seems to have gone unnoticed that quite a lot is known about how to produce better outcomes for children in centers. National randomized trials of Early Head Start and Head Start, as well as smaller studies of strong, educationally effective preschool centers, have found meaningful positive effects on children's social development and behavior. There is no hint of this in the national institute study - probably because the institute studied typical child care, much of which does not live up to basic quality standards.
So parents should relax. Chances are, their children will not be noticeably worse-behaved because they attend a center rather than some other form of child care.
The key is to find a center that meets the quality standards that ensure kids are getting the best education and social-emotional development possible.
Instead of scaring parents about their child care choices, we would be better served by asking why policymakers aren't doing more to provide the kinds of quality programs that could improve children's learning and development.
W. Steven Barnett is a professor of education economics at Rutgers University and director of the National Institute for Early Education Research in New Brunswick, N.J. His e-mail is email@example.com.