Thursday, December 15, 2011

How can you argue against confirmation bias?

Apart from pointing out that they are committing the confirmation bias fallacy, which is the easy part.|||Odds are good that if the person you're arguing with doesn't have a clue that s/he's fallen into confirmation bias, then you're not going to have much luck talking them out of it. (Think creationists, homeopaths, etc.)





People, even the most educated people, are NOT good at avoiding confirmation bias, which is the reason behind the very strict methodology involved in scientific experiments. A classic format built to avoid confirmation bias is the double blind placebo controlled study. Not even the scientists running the experiment know who is getting the real drug and who is getting the placebo, just to avoid an unconscious influence they might make on the subject, or in their interpretation, due to pre-conceived expectation regarding the action of a particular drug.





Are you referring to a specific line of argument? Unfortunately, the understanding of most people is entrenched in context and as such whatever argument you make would probably be most effective coming from a context that is understandable and "close to home" for the individual you are arguing with. Knowing more information regarding the context of the confirmation bias would be helpful.|||Unless you are engaged in some process of formally deconstructing a logical argument, pointing out that someone has used a "fallacy" is pretentious and pointless.





You can't argue against it. You can point it out, you can offer counter examples, but if a person is prepared to accept entirely anecdotal evidence of which they are the judge of what is and isn't evidence, then you have nothing to talk about.|||The best way to argue it is by constructing a religiously-neutral example so people can grasp the general principle. But for that to work, you have to get the person to sit still long enough to listen to the example.|||To whom are you referring?





The Catholic sacrament of Confirmation?





Or, the process of confirmation of nominees to political office?|||Point out how they could do a controlled experiment.

No comments:

Post a Comment