Reporters Don't Understand Statistics, Film at 11

[chuckro]

http://www.cnn.com/2007/HEALTH/conditions/07/24/diet.sodas/

"People who drink one or more soft drinks a day have a more than 50 percent higher risk of developing the heart disease precursor metabolic syndrome than people who drink less than one soda a day, a new study has found. And it didn't matter if it was a regular soda or a diet soda."

Correlation does not imply causation! Gah!

Gee, I wonder if people who drink a lot of soda, regular or diet, are also likely to not exercise, eat poorly, or be sedentary? As opposed to the health freaks who drink water instead, etc etc.

Why are people so goddamn stupid?

Comments

[ivy03.livejournal.com]

Also, the distinction they're making is between one and two or more cans, not between one and none, and the correlation is with a syndrome that itself is correlated with heart disease, and the actual percent increase in risk isn't that much.

In other words--this study means nothing! Or, alternatively, it was a slow news day.

[dushai.livejournal.com]

There is a statistically significant correlation between waking up with a headache and going to sleep the night before with your shoes on.

Clearly this means that sleeping with shoes on causes headaches the next morning!

... or maybe both imply you passed out drunk the night before... nah...