Wednesday, October 04, 2006

When Moms Work, Kids Get Fat



This is the headline of a Slate post by Tim Harford. The next line says "An incendiary new explanation for childhood obesity." I'm sure that Harford planned it that way, to get a lot of hits. The relevant piece in a long article about quite different explanations for childhood obesity is this:

That is all very interesting, but it does beg the question of why things are getting worse. Another trio of economists—Patricia Anderson, Kristin Butcher, and Phillip Levine—has suggested that two-income families may be producing the problem. They find that children are fatter if their mothers work longer hours. This is true even within families: The sibling who spent more time as a latchkey child will tend to be the fatter one, perhaps because the mother is less able to supervise outdoor play or has less time to cook and therefore buys more fast food. Unfortunately for working mothers who are already struck by guilt, the effects are pretty substantial. A mere 10 hours at work raises the chance of childhood obesity by 1.3 percentage points, which is about 10 percent.

Let's not go quite that fast. I looked at the original research summary, and what it states is that weeks worked by the mother had no correlation with childhood obesity. So when moms work, kids don't get fat. Get it? What the study did demonstrate was that childhood obesity was positively associated with mothers who work long hours per week, and only then for the educated white mothers, and that is the percentage Harford chooses to cite in his article. There was no relationship between childhood obesity and working hours for the black or Hispanic mothers. It might be worthwhile to dig out the sample numbers here. How many white and educated mothers did the study include?

Of course what this study and the Harford article really demonstrate is the accepted sex role division within families. Note that the researchers didn't even bother to study what impact fathers working might have on the children's obesity. Because women are supposed to be solely in charge of the family members' health and well-being. So we don't know if the highly educated women who worked intensive hours also had highly educated male partners who also worked intensive hours or if they had male partners who were at home and were supposed to check the children's diet. And we don't actually know if the children were "latchkey" children or if someone else was supervising them.

I really must make one explicit correction here, too, about this Harford sentence:

A mere 10 hours at work raises the chance of childhood obesity by 1.3 percentage points, which is about 10 percent.

This is just not true. Here is the actual quote from the study:

Their results suggest that a 10-hour increase in average hours worked per week will increase the overall probability a child is overweight by 0.5 to 1 percentage point. In the probit models, the point estimate on hours per week is always positive and increases with income quartile. For mothers in the highest socioeconomic status, the results indicate that a 10-hour increase in average hours worked per week since a child' birth increases the likelihood that the child will be overweight by 1.3 percentage points.

This is a very different thing from the Harford quote. The bolds are mine.
---
Initial link via feministing.com.