Fat = contagion?
Jul. 27th, 2007 09:33 amThere's not much I can say about this latest wave of "fat friends make you fat" mania, other than to point out that the researchers who concocted this "study" fairly tortured their data into submission. Methodology HURTS, baby.
For starters, they didn't actually follow people over the course of years - that takes forethought and money and we don't bother with that anymore. They took handwritten tracking sheets from Framingham Heart Study interviews conducted between 1971 and 2003 (only seven interviews per person over the course of that 32-yr stretch) on which the interviewees were asked to list *one* person they might be in contact with in a few years.
From this list the authors created "38,611 different people connections" (people connections, I like that) and plugged that mess of data into a virtual network imaging program to create pretty 3-dimentional designs based on algorithms. Note: different algorithm = different result. Then in order to wring any kind of statical connections from it all, "These authors created models and ran countless simulations, with so many assumptions and complex selections and elimination of variables, that no one could hope to unravel it."
Even then, they couldn't get good odds - so instead of reporting results were null they used the time-honored technical trick of comparing them to each other. The final analysis? Obesity was associated less with genetics, geographical proximity, being married, etc. than with... wait for it... SAME SEX MALE FRIENDS.
Since I'm sure 99% of the press this report is getting is aimed at women, why is nobody (except the the few, the proud, and the scientifically literate) pointing out that same sex female and opposite sex friendships had no bearing whatsoever on social BMI, no matter how much data cosmetology went on...?
Well, I guess I did have something to say about it after all. Anthropologists should, particularly medical anthropologists. This framing of fat as contagion, as impurity, as danger, is a cultural issue that we need to address - preferably sometime before the advent of fat internment camps.
For starters, they didn't actually follow people over the course of years - that takes forethought and money and we don't bother with that anymore. They took handwritten tracking sheets from Framingham Heart Study interviews conducted between 1971 and 2003 (only seven interviews per person over the course of that 32-yr stretch) on which the interviewees were asked to list *one* person they might be in contact with in a few years.
From this list the authors created "38,611 different people connections" (people connections, I like that) and plugged that mess of data into a virtual network imaging program to create pretty 3-dimentional designs based on algorithms. Note: different algorithm = different result. Then in order to wring any kind of statical connections from it all, "These authors created models and ran countless simulations, with so many assumptions and complex selections and elimination of variables, that no one could hope to unravel it."
Even then, they couldn't get good odds - so instead of reporting results were null they used the time-honored technical trick of comparing them to each other. The final analysis? Obesity was associated less with genetics, geographical proximity, being married, etc. than with... wait for it... SAME SEX MALE FRIENDS.
Since I'm sure 99% of the press this report is getting is aimed at women, why is nobody (except the the few, the proud, and the scientifically literate) pointing out that same sex female and opposite sex friendships had no bearing whatsoever on social BMI, no matter how much data cosmetology went on...?
Well, I guess I did have something to say about it after all. Anthropologists should, particularly medical anthropologists. This framing of fat as contagion, as impurity, as danger, is a cultural issue that we need to address - preferably sometime before the advent of fat internment camps.
no subject
Date: 2007-07-27 02:30 pm (UTC)no subject
Date: 2007-07-27 02:32 pm (UTC)no subject
Date: 2007-07-27 02:50 pm (UTC)Thanks for providing a more intelligent response.
no subject
Date: 2007-07-27 03:11 pm (UTC)I just posted the links to the "few, the proud and the scientifically literate" you provided (much thanks). I wish I could pay to have you come down here and guest lecture in my classes this fall!!!!!
no subject
Date: 2007-07-27 04:15 pm (UTC)no subject
Date: 2007-07-27 05:31 pm (UTC)Do your students sniff glue? Because it can't just be about age, there's a 15 year old on my LJ fl that saw right through the bullshit of this article.
I wish I could pay to have you come down here and guest lecture in my classes this fall!!!!!
Hmmm, is there a student organization that would at least foot the bill for
no subject
Date: 2007-07-27 09:34 pm (UTC)2. do they give the same credence to all the other pseudo-science that gets published in the popular media? Because take your pick, you can swing a dead cat and hit at least a dozen a week.
3. This is a chance to teach people how to THINK not to be a herd of mooing cattle, accepting every craptastic story that is put out by the same people who follow Paris Hilton's every bowel movement.
Sheesh.
no subject
Date: 2007-07-27 03:45 pm (UTC)*fumes*
Prick them do they not bleed offe thee ill humours?
Date: 2007-07-27 04:28 pm (UTC)Jesus Christ...should I drop them a line and ask the creators/authors of this study if they are aware that diseases are NOT caused by ill humors, or for that matter, The Devil?
This is a fantastically well thought out and articulate reply. (Sheesh, when I get mad at stuff like this I get all Emo and start going on about my feelings!)
no subject
Date: 2007-07-28 05:06 am (UTC)And I agree with durgablue, the thinking behind this is so "middle ages" that it's incredible. Reminds me of the flat-earthness of the Answers in Genesis people.
no subject
Date: 2007-07-28 05:56 am (UTC)