Lies, Statistics, and “Childhood Obesity”

You Forgot Your BullshitSeveral memes have been making the social media rounds based on a New York Times headline: “Obesity Dropped 43% Among Young Children In Decade, Study Finds”   When you look up the study that’s referenced, the conclusion states “Overall, there have been no significant changes in obesity prevalence in youth or adults between 2003-2004 and 2011-2012”

So just what in the Sam Hill happened here? It’s something that my favorite reporting methods professor in college used to call “fun with statistics” and it most often occurs when someone (a researcher, reporter, company, or *cough* First Lady *cough*) wants a particular outcome and so they just finesse the numbers to get it.

In this case, what the researchers found was that (to quote the NYT article) “About 8 percent of 2- to 5-yearolds were obese in 2012, down from 14 percent in 2004.”   (In actuality the study states right there in the abstract that it was 13.9% to 8.4%  which is a delta of 5.5% rather than 6% but why would accuracy be important here?)  So how did a 6% (actually 5.5% but who’s counting) difference become a 43% change in the New York Times headline?  Through the magic of absolute vs. relative change!

Absolute change is the difference in the item being studied at two different times:

[% of fat 2-5 year olds in 2012] – [% of fat 2-5 year olds in 2004]  = Absolute change

.08 – .14 = -.06

Relative change is the absolute change calculated as a percentage of the indicator in the first time period.

([% of fat 2-5 year olds in 2012]  – [% of fat 2-5 year olds in 2004])  / [% of fat 2-5 year olds in 2004] * 100%

(.08 – .14) / .14 * 100 = -42.85% (they rounded to -43%)

Often when someone is trying to make something seem like a bigger deal than it is, they will use relative change rather than absolute change. And this doesn’t even take into account the margin of error, number and effect of dropouts etc. That’s all the formulas in this post,  I promise (though if you want to see further discussion of the fuzzy math, there’s a good piece in Forbes, though trigger warning for obesity epi-panic language). Something that is worth noting is that for all of this talk about OMGDEATHFAT increasing so much, this study actually found no significant change since 2004, but that’s a subject for another blog.

This is certainly not the first time that this has happened.  Allergan has used “Fun with Statistics” to try to sell lapbands. We’ve even seen this kind of reporting be not only crappy, but actually dangerous. So why am I talking about this now?

First of all, because it’s making the rounds again and every time I see it, it pisses me off. Also since focusing on the weight of children has been a pet project of Michelle Obama, who’s time at the White House is winding down, I expect that we may see a lot more attempts to suggest that focusing on the weight of children – and telling fat kids that we’re all trying to create a world where nobody who looks like them exists – has been a fantastic idea, despite the evidence.

This kind of reporting drives the discussion and policy – including policies that body shame children as a substitute for actually supporting their health – so it’s important that we know that it’s always reader beware, that we can and should ask questions, and that we can and should hold the people doing this reporting accountable.

Like the blog?  More Cool Stuff!

Book and Dance Class Sale!  I’m on a journey to complete an IRONMAN triathlon, and I’m having a sale on all my books, DVDs, and digital downloads to help pay for it. You get books and dance classes, I get spandex clothes and bike parts. Everybody wins! If you want, you can check it out here!

Like my work?  Want to help me keep doing it? Become a Member! For ten bucks a month you can support size diversity activism, help keep the blog ad free, and get deals from size positive businesses as a thank you.  Click here for details

Book Me!  I’d love to speak to your organization. You can get more information on topics, previous engagements and reviews here or just e-mail me at ragen at danceswithfat dot org!

I’m training for an IRONMAN! You can follow my journey at

A movie about my time as a dancer is in active development, you can follow the progress on Facebook!

If you are uncomfortable with my offering things for sale on this site, you are invited to check out this post.

19 thoughts on “Lies, Statistics, and “Childhood Obesity”

    1. Yes, it’s an amazing post! Just wondering, how do you insert pictures in comments? 🙂

        1. Thanks! 🙂 For some reason my comments aren’t showing up on many posts here…and it doesn’t say “awaiting moderation” 😦

          1. This may be a technical problem; I’d advise you to send a private message to Ragen. If you are friends with her on Facebook, that’s a great way to send her a message. There may be other ways, but that is the one I use. Good luck! 🙂

          2. Weird, I looked through and your comments have been approved and when I click through I can see them in the threads. Can you e-mail me the text of one of them that isn’t showing up (ragen at danceswithfat dot org) and I’ll try to figure out what’s up.



  1. Michelle Obama’s time in the White House is ending, However, if Hillary ends up (back) there, expect even more anti-fat-kids crap from that building, based on a speech I heard her deliver to the American Academy of Pediatrics last year.

  2. Do you have a phd, ragen? I’m calling you dr chastain anyways, because that was some high-level work to take down the fuzzy math in that article. Such a shame that you need to be able to do so much math just to see what we already know. The diet industry will stop at nothing to make us pay to hate ourselves!

  3. I am fat. My mom is fat. My late father was fat. Their parents were not fat; they survived the Depression.

    So, when I had kids in the early 80s, I did everything our pediatrician suggested: We ate lots of fresh veggies; kept fast food to a minimum (couldn’t afford it anyway), provided them with bicycles and sports toys and encouraged their use; there was no internet or video games, and tv was minimized. They were healthy, medium-sized kids.

    Now that they’re in their 30s? They’re fat. And they’re still healthy. And they’re great.

    1. If one were to ask a doctor today what the cause of obesity in the Middle Ages and Antiquity was, they would not be able to resort to video games or eating too much since there was very little leisure time in the past, and most people consistently ate less than even their BMR.

      I was born in the 1980s, and I am fat, similar to your kids.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.