PZ has a post about Steve Jobs and woo. In it he summarizes the conclusions everyone on the side of science has been giving. However, he does make one error:
So the final result is that real science kept him alive and healthy as long as possible, and that an early flirtation with ‘alternative’ medicine might have contributed somewhat to lowering the odds of survival, but that what killed him is cancer. And cancer is a bastard.
No, there is no “might” about it. There is a direct correlation between when one treats cancer and how long one is likely to survive. By looking at ‘alternatives’, Jobs’ odds of survival lowered. Think about Huntington’s Disease. It’s a neurodegenerative disease which is passed on genetically. In the average situation where this disease is involved in a family, one parent has a single defective gene whereas the other parent is fine. Any child those two people have has a 50% chance of getting the disease. Those are the odds. Period. Even if the child gets tested and is found to not have the disease, the odds of contracting it will still always be 50% (as just described).
When we talk about the odds of this or the stats on that, we are not referring to a single individual. Even if Jobs’ time of survival remained the same – or even, against all the evidence, it increased because of the woo – his odds were absolutely decreased. Odds refer to the numbers we have on either a sample or population. The only way a person can change his odds is by doing something which has statistical significance. Jobs’, for instance, did just that by engaging in woo instead of treating his cancer.
PZ’s statement is no better than when Bill O’Reilly cited a single poll about atheists and then claimed a trend was evidenced.