Saturday, February 12, 2011

Oh boy! Stat Baby is going to be a a fatty! Healthday and Babycenter are AT IT AGAIN!!!

Thanks to all of you for your understanding during a rather long absence since my last posting. I've been devoting time to my employment outside of the home (there, no one feels badly about that term I hope).

Monday morning, bright and early - my weekly "Scare the mother and make her feel guilty" email from my friends at Babycenter arrives right on schedule.  The news to smack me down as I try to prepare for meetings with my biggest and most important client was,
"The more moms work, the heavier their kids get, says study."

Part I: The headline and the article

Any mother who works (I am assuming they mean paid employment in or outside the home) should assume that the more she does, the heavier her child will be. Taken to it's logical conclusion, Stat Baby should be the size of a house by now.

Now, let's see what the article says the study says. This is, of course, likely to be different from what the study ACTUALLY says. Ah the perils of allowing undereducated writers report on things that they can't understand because their reading comprehension and awareness of the scientific method are non-existent.

The study didn't say much because a study author has to use OTHER data to justify her "conclusion."
Surprisingly, there was no evidence that the increase in BMI was linked to more TV viewing, a decrease in physical activity, or more time spent unsupervised.The researchers concluded that it may be changes in children's eating and sleeping patterns (factors that were not included in the data) that account for the BMI changes. "While we weren't able to identify any specific environmental factors, it's clear from other research that nutrition and sleep are important," she said. "So, one possible policy implication is to do more to help working parents find quick and easy ways to prepare healthy foods."
Eating and sleeping patterns were not included in the dataset but the conclusion it that eating and sleeping patterns account for the change. Shut the front door. These people are a classic case of a hypothesis desperately trying to find data (not the other way around).

The quote from the article says that the data showed no relationship between the time spent viewing tv or decreased activity. Yet the researcher says in the interview,
The effect was even greater among children in fifth and sixth grades. "It is possible that because fifth and sixth graders generally have more independence and less adult supervision over their time use and food choices than third graders, maternal employment precipitates poorer food choices and more sedentary activity," the authors wrote.
You can't have it both ways. The data showed no evidence surrounding decreased activity (is that not what sedentary means?), yet the researcher insists on using this to explain the difference. How about spurious correlation?

Mommy works. Mommy wants to do the best for her child. Mommy sees headline that says she's making her child fatter by the minute each time she takes a conference call.  Mommy CLICKS.  Babycenter has a unique pageview and traffic to click on the ads on their site. But in the back of her head, Mommy doubts. Mommy worries. Mommy is one step closer to giving it all up.

Not me. I say, "This author is full of it. I have work to do."

Part II: What other factors WERE NOT examined?

  1. Did Daddy work? If so, what were his hours like? Nope.
  2. What was the income level of the family? Nope.
  3. What was the mother's highest level of education attained? Nope.
  4. What type of food was being served? Nope
This all happens "below the fold." Hide the reason for skepticism and increase the guilt quotient. Try accessing this sheer and utter nonsense on a mobile device. You see the headline, scan first paragraph and move on with your day.

Part III: Where I get to indulge my fascination with the concept of "average" child.

About the danger of averages:

The study says for the AVERAGE child at grade three there is in increase in weight of about 1.5 lbs over what is "expected" at that age. DING, DING, DING...BS alarm bell.
What is an average child, anyway? What is the distribution surrounding the mean in this case? Do boys and girls vary in the timing of growth cycles at this age? The average is a dangerous concept. The MEDIAN height of a child of 8 years is 129 centimeters according to the growth chart provided here. The 5th percentile is 119 centimeters. The 95th percentile is 149 centimeters. So the vast majority of 8 year-old-girls are between 45 and 54 inches tall. That's 9 inches or 2/3 of a ft. That's not really a narrow window and many children are not at average.  However, if my population has 100 children in it and 75 of them weigh 60 lbs and 25 weigh 15 lbs, the average child weighs 48.75 lbs and looks nothing like ANY child in the population.

I have an acquaintance who wanted to live in a city, her husband wanted to live on a rural farm. They live in a McMansion, in a suburb. They are both miserable. Good compromise are rarely rarely found dead center  of two extremes.
Tweet