Record: Michelle Cardel et al., “Home-Schooled Children are Thinner, Leaner, and Report Better Diets Relative to Traditionally Schooled Children” in Pediatric Obesity 22, no. 2 (February, 2014): 497-503. Abstract Here.
Summary: In this piece 11 authors compare the diets of 47 home schooled children in the Birmingham, AL area with 48 demographically similar children from the same region who attend public schools.
Going into the project the authors wondered, given that both homeschooling and childhood obesity have grown markedly since the 1970s, if homeschooled children might be more prone to obesity. Previous studies of homeschoolers’ fitness levels have found them to be somewhat lower than those of children attending formal schools (see on this Welk, et. al, Pediatric Exercise Science 16:4 (2004) and Long, et al, Pediatric Exercise Science 22:1 (2010)).
To test this hunch the researchers recruited 95 children aged 7-12, all white, all living in Birmingham, Alabama. Half (47) were homeschooled and half (48) attended public schools. As socio-economic status is well-known to be an important variable in obesity rates, the study included an assessment of social class, which found that the two groups were almost equal (the public school group on average was a tiny bit richer).
After securing the sample the researchers gathered extensive data on their subjects. Each subject wore an MTI ActiGraph Accelerometer for seven days to measure physical activity levels. Researchers checked in on each student on average twice a day to record all food and drink intakes using standard protocols. When they analyzed all of this data the researchers were surprised that their original hunch had been completely wrong.
Homeschoolers had lower BMI and less body fat on average than public schoolers. While both homeschoolers and public schoolers were consuming more calories than they should be, public schoolers were consuming even more than homeschoolers (about 120 calories a day more).
When they broke down the data to particular meals, they found that the biggest difference between the two groups was lunch. Outside of lunch homeschooler and public schooler diet was pretty similar (public schoolers actually ate about 20 calories less), but at lunch the public schoolers consumed more calories, sugar, sodium, potassium, and calcium. Lunch alone accounted for the extra 120 calories public schoolers consumed a day.
Public schoolers were a little bit more active than homeschoolers, but not enough to be statistically significant.
Though the design and execution of this study is strong, the way it was written up and titled is misleading. Given the title one would think that huge differences between home and public schoolers had been uncovered in terms of diet and exercise. But that’s not really the case. 120 calories is not very much, and I have a hunch what the cause is. Note that calcium and potassium are higher among public schoolers. Why? What do kids at school drink for lunch? Milk. When given the choice, what kind of milk do the kids drink? Flavored milks like chocolate, strawberry, and (for shame), vanilla. 8 ounces of 2% chocolate milk contains 190 calories. There’s your difference.
Given that the differences in caloric intake between the two groups are not very big, there were a few demographic distinctions that the researchers didn’t discuss that could have perhaps influenced the data showing higher body fat and BMI for public schoolers. Only 42% of the homeschool group was female, while 56% of the public school group was. Also, though the SES rates on average were similar, the public school group had almost twice as many children who qualified for free lunch. The fact that a sample of 48 has a couple more poor kids and a couple more females could explain much of the difference. It would have been nice to see the raw data on BMI and body fat for all 95 children to see if the differences are due to a small handful of kids or are spread out evenly throughout the sample.
So this study needs to be more modest in its claims. It also, in my view, needs a qualitative check. Rather than just reporting averaged figures, it would be nice to get actual observations of what the kids in these two groups look like, how active they are, how much they are eating and how they are eating it. While there might be a statistical difference on paper, how significant is it really? Do homeschooled and public schooled kids look all that different to the naked eye? Do their families seem to have different food and exercise attitudes and habits?
If there’s any real message in this article, to me it’s that schools should abolish flavored milk and whatever else might be contributing to higher rates of sodium and sugar in school diets. And of course that is precisely what schools have been working toward doing in recent years through initiatives like the Department of Agriculture’s Team Nutrition Program and many state programs along the same lines.
Milton Gaither, Messiah College