June 19, 2012

It’s been a while, i know. So something dramatic must have happened. Multiple dramatic things have happened, but one in particular drove me to dust off my WP login. This article on obesity has been rapidly making the rounds (i’ve seen it linked no less than a dozen times since yesterday afternoon). It’s got lots of interesting information in it and easily quotable lines. But in virtually every single cite i’ve seen of it so far (on Facebook, news articles, etc.), the same finding has been quoted as the most striking punchline of the article, but has, in every case misrepresented what that finding actually is. The line from the article that matters is this (from the abstract):

North America has 6% of the world population but 34% of biomass due to obesity.

The quotes of that finding i’ve seen thus far however have inevitably been framed as:

North America has 6% of the world population but 34% of biomass.

A seemingly innocuous edit, but resulting in a complete misrepresentation of what the study actually found. What the analysis shows is that if you limit the analysis only to account for the obesity in the world (i.e., we look only at the “excess” weight) on the planet, N. America accounts for 34% of that amount. Not 34% of all biomass. In order for the citation as it’s making the rounds to be true, the average N. American would have to weigh 371 kg (or roughly 818 pounds; calculations stem from Table 3 in the paper). Now part of the problem is the way things are written as most people aren’t used to having to do a little simple math to digest their research blurbs, but sometimes it’s completely necessary to be sure you aren’t completely lead astray.

Now, i’m not trying to diminish the findings in the article, as i think they have done a good job of demonstrating how weight is unevenly distributed across the globe. Just not in the way people are saying they’ve said it is.


Neal Caren on citation patterns in sociology

June 2, 2012

What articles and books have sociologists been citing a lot recently?  See the list compiled by Neal Caren.  He also made a cool network diagram showing which articles are cited together.