2 ms·
>"Affirming the consequent is a deductive fallacy. It is not a fallacy in probabilistic reasoning" Nevertheless, people commit that fallacy all the time in epi
by nonbel 8y ago
>"Affirming the consequent is a deductive fallacy. It is not a fallacy in probabilistic reasoning"
Nevertheless, people commit that fallacy all the time in epidemiology. They say stuff like "if smoking caused cancer, then people who smoked would get cancer more often". Then if there is a (statistically significant) positive correlation between cancer and smoking, they conclude smoking causes cancer.
That is quite obviously affirming the consequent. I explain the (completely different and correct) probabilistic way to think about this situation using Bayes' rule in my post.
It has nothing to do with arbitrary cut-offs for telling us whether a correlation exists or not. Instead it requires comparing the relative performance of the various explanations (at least the top few candidates since extremely unlikely ones can be dropped from the sum in the denominator).
- sanxiyn 8y agoI don't think anyone serious about epidemiology says stuffs like that. On the other hand, it is true many people interpret what epidemiology says as such. Saying smoking is a risk factor of cancer is very different from smoking causes cancer. As you said, you must consider alternative explanations. On the other hand, people can disagree about what alternative explanations are "extremely unlikely" and "can be dropped", that's what I meant by prior. For example, nuclear test causes cancer, is probably extremely unlikely. Smoking being a risk factor of cancer is an evidence of smoking causing cancer. Being evidence of causal relation is completely different from being sufficient to prove causal relation.
- nonbel 8y ago>"I don't think anyone serious about epidemiology says stuffs like that." Pick a paper, I bet they do.