An archetypal example of spectacularly shitty “research” and even worse reporting

I recently hinted at the fact that surveys are not scientific studies, and should not be treated as sources of useful data when it comes to certain areas of research. I now have a “study” that exemplifies this perfectly.

It’s one thing to ask people yes-or-no questions that are hard to fuck up without conscious effort, e.g., “Who do you plan to vote for for president next week?” It’s a far sketchier matter to rely on self-reporting and self-recall when it comes to something not nearly so black-and-white, such as food consumption.

Therefore, “research” findings that the author of this article suggests are surprising are not startling in the least, and it’s not difficult to understand why.

“Overweight Teens Typically Eat Less Than Normal-Weight Peers”

Okay, right off the bat, the editor who wrote this headline should be force-fed several pounds of stir-fried Crisco and then forced to run a few miles, because even though the research the article below it describes is fraught with frivolity, its authors make so such definitive claim, though they give the appearance of doing so. In the study abstract, the authors bluntly state that “increased energy intake in early childhood is related to the onset of obesity, but other mechanisms, such as differences in energy expenditure, may contribute more to maintaining obese/overweight status through adolescence.” In other words, more sedentary kids tend to weigh more. Now that’s fucking groundbreaking.

Then there is the article itself.

As expected, young children who are overweight consume more calories each day than do their thinner peers, a new study says.

But, in a decidedly unexpected finding, the researchers also discovered that older overweight children may actually consume fewer calories daily compared to their healthy-weight counterparts.

Wait. Now we've scaled back from "typically" eat less to "may" eat less? Subtle, motherfuckers, subtle.

“The message for society and parents is: Don’t assume that a child who’s overweight is overeating. Obesity isn’t just a simple matter of eating more,” said study author Asheley Cockrell Skinner, an assistant professor of pediatrics at the University of North Carolina, in Chapel Hill. “Be sympathetic. Overweight children reported eating fewer calories, and to lose weight, these kids have to eat even less. It’s probably even harder for them to lose weight than we give them credit for.”

Here the lead researcher seems to be back-pedaling from the statements appearing in the study abstract, perhaps having been cajoled into saying what the reporter wanted her to say — it’s not as if any reason to trust this publication, or at least this article, hasn’t already been shot to hell.

The study included dietary information from nearly 13,000 children between the ages of 1 and 17. The information came from U.S. National Health and Nutrition Examination Survey, which was conducted from 2001 to 2008. The population included in this study is representative of the U.S. population.

OK so far, if only because there was nothing in this to credibly lie about. But then the whole train quickly detours into Shitsville:

“The food-consumption data were collected on two separate days, and children and their parents were asked to recall what the child had eaten in the last 24 hours and how much they ate of any particular food. The researchers had a number of representative measuring devices to try to get the best approximation of portion size that they could.”

And further down:

Cockrell Skinner and her colleagues believe that activity may play a big role in these findings.

“Overweight children tend to be less active,” she said, but added that other factors are likely at play. “The body has complex reactions to how much you eat, when you eat and your activity levels. And, we just don’t completely understand these reactions yet,” she explained.

This is just a hot mess, although I do like the name “Cockrell Skinner” and wish it were not just a surname but a whole name, as it conjures up images of a tobacco-chewing member of the Georgia House of Representatives or something.

Let’s look at the potential sources of error here:

* Food intake wasn’t merely self-reported, but reported second-hand, at least in part.
* Portions were not weighed; instead, their sizes were estimated.
* Food intake was tracked for a grand total of two days.

This alone renders the “study” virtually worthless. Other research has definitely shown that people tend to underestimate their food intake and that overweight people do so to a greater extent than normal-weight people, although this is probably related to their tending to choose larger meals in the first place, as underestimation of calorie content rises in concert with portion sizes. So this particular type of surveying is virtually assured to be filthy with noisy bias, and the researchers have to have known this.

But the kicker is the downplaying of physical activity, as if this is just an afterthought in all of this. It’s as though the researchers had done a study of distance runners in Peoria, Illinois and distance runners in Iten Province, Kenya, and found that the average marathon time was 47 minutes faster in the latter group despite similar sample sizes and BMIs and marathon courses and so on, and stated, “Well, we don’t know what the fuck that’s all about, because even though the Kenyans averaged about 112 miles a week for the six months before their marathon and the Peorians 47, there are a lot of factors at work here.”

To be fair to the researchers, they make many useful and helpful comments as well, but these of course have been pushed to the bottom by the clowns at the US News & World Distort (yeah, I made that up just now), since no one wants to read common-sense stuff that’s not controversial. After all, controversy gets people going, especially those with strong biases of their own.

  1. #1 by Warren Adams-Ockrassa on November 16, 2012 - 2:37 pm

    Sigh.

Follow

Get every new post delivered to your Inbox.

%d bloggers like this: