|
I would also add small lesson on statistics or probability. When people talk about smoking and cancer, for example, they often claim that their grandmother smoked for 90 years and did get lung cancer; therefore smoking must not cause cancer. Well, there is a difference between a sample and a population. Smoking does increase the RISK for certain diseases but it does not mean that everyone who smokes will get cancer. Just as not everyone who eats bacon a lot will die of heart disease.
Statistical significance does not always equal biological significance. I had a stat professor who jokingly referred to a company he wanted to form called "Small P-values are Us", meaning, I guess, that he could get you a statistically significant finding for whatever you wanted. I've often been skeptical of claims in the media that substance X (say, cell phones) causes cancer. Frankly there is just not enough evidence to say so. The studies that have claimed so are deeply flawed. You can't really say what causes any one case of cancer.
I stand by the statement that extraordinary claims require extraordinary proof. I used this in reference to a post criticizing James Randi for the standards he uses in his million-dollar prize contest. Those standards are not onerous and the fact that no one who enters the contest meets the standards shows that paranormal claims do not stand up to the standards of scientific proof.
|