Saturday, June 5, 2010

Correlation does not imply causation: The Cholesterol example.

Correlation does not imply causation

Such a simple statement... Yet AMERICA's S.A.D. Diet is completely broken because of a couple simple instances of Assuming Causation based on a flawed and limited Correlation study.

I felt this deserved a post all on its own, since this flaw is why we believe so many things we shouldn't and should be one of the first things you check when concerned about a study.

True Science has a "control" when the control is uncontrolled... the science is flawed.

The more firemen fighting a fire, the bigger the fire is going to be.Therefore firemen cause fire.

We use that same flawed logic to explain Dietary Fat, Cholesterol, Diabetes, and a number of other health issues.
Yet the Firemen are there to help.

Cholesterol is a great example of this.

This is a forum post with great explination of whats going on, with references and more reading

If you havent yet, I Strongly Strongly suggest getting this book:

It is NOT a Diet Book. It is a Review of Science gone wrong, where it went wrong, and why its wrong.
Good Calories, Bad Calories

The Great Cholesterol Con: The Truth About What Really Causes Heart Disease and How to Avoid It

This one is also a good read if Cholesterol is a concern, but the bulletpoints are covered for the layman quite well by Griff in the Marks Daily Apple Forum post that I linked above.


Post a Comment