What do you enjoy reading the most here on my blog?

Search My Blog

Wednesday, December 05, 2007

Correlation and Causation

Sweet Jesus, I am so incredibly pissed off over seeing yet another failure of a researcher producing yet another failure of an article based on some half-assed finding! Okay, so the person writing the article isn't exactly a "researcher" - more likely a journalist who was abysmal at every subject in school other than lying to the bullies to avoid getting beaten up or forging doctor's notes to get out of gym throughout high school.

Either way, a complete dipshit.

The article is about a study that showed a correlation between people who said they drink diet sodas and the subjects being overweight. That's seriously it. 41% of respondents who said they consume diet soda were overweight.

And the journalist immediately blurts out things like: "One recent study has shown that people who drink diet soda still have a 41 percent chance of being overweight."

Right off the fucking bat, with the ONE piece of real actual honest-to-goodness fucking SCIENTIFIC DATA and it's just slathered in make-up and given a skimpy bikini and made into whatever fashion of whore the journalist cares to make it out to be. They're actually implying causation IN THE CORRELATION!

It's not that those who consume diet soda have "a chance" to be overweight - it's simply that this percentage WERE overweight. One has absolutely nothing to do with the other, unless you feel like conducting THAT experiment to test THAT hypothesis!

Sure, they then go into some blather about "halo effect" and how you think you're being better to your body with certain consumables and it leads to overcompensation down the road for daily calories as a result. Again, nothing to do with the 41%. You can't just staple a fact to a scientific result and expect the resulting blood pool to be a testament to how the fact is represented by the lies you've made up about the data.

Not once did they ponder the notion that the obesity came first. Not once did they randomly speculate that instead of the consumption of diet sodas causing overweight participants that it MIGHT be the case of the participants being overweight causing them to consume diet sodas! I'm not saying that's the case - because I know better than to assume that correlation is grounds for causation - but I'm throwing out a tat for their tit in the grand scheme of Devil's Advocacy.

I happen to be overweight. And I happen to try and consume more diet soda as a result of being overweight. To hopefully see the result of being LESS overweight.

No matter what some bogus idiot decides to claim about the mangled remains of a once-adequately-scientific study.

(Judge the article for yourself!)

No comments: