Beliefs, prejudices and data

Posted on | May 31, 2012 | No Comments

When you see a bunch of things moving around in a flock on a pond in the distance, making quacking sounds, and occasionally becoming airborne, you do not ask for DNA evidence to determine that they are ducks, because you already have a lot of background information about swimming ducks. If someone told you they were in fact not ducks but electronic toys, you would—quite reasonably—demand a lot more evidence because this clashes with your beliefs about ducks.

If you were told these were not ducks but toys you would demand more evidence.

George Monbiot, a British journalist, wrote a scathing article   a while ago on “how the greens were misled.”  He described how claims by Christopher Busby, an anti-nuclear campaigner, were shot down by the Welsh Cancer Intelligence and Surveillance Unit. Busby had claimed that there were leukemia cancer clusters along the north Welsh coast. The Welsh Cancer Intelligence and Surveillance Unit studied his claims and published an article.     Amongst their findings was that the data Busby used was incorrect, having used 138 cases of cancer versus the Unit’s 48. They were “completely unable to confirm the results of Busby.”  Their summary is blunt: “No evidence has been found from this study to support Busby’s hypothesis.”

Monbiot says about this that

One of the most widespread human weaknesses is our readiness to accept claims that fit our beliefs and reject those that clash with them. We demand impossible standards of proof when confronted with something we don’t want to hear, but will believe any old cobblers if it confirms our prejudices.

Well, yes, it can be a weakness if people ignore all data and only go on dogmatic belief, ignoring all other information.  But it is a logical fallacy in our reasoning to think that a new measurement unequivocally either supports or refutes a claim. Whether it does or not depends on the background information of the person who hears the claim, and the probability that person assigns to being misled by those who take the measurement and make the claim. It may support the claim for one person and refute it for another if the probabilities they assign to being misled differ. We have blogged about this here.

It is perfectly rational to demand more evidence when something contradicts a previously held belief.  It is only when people assign a zero probability to being misled that measurements can unequivocally either support or refute a claim. And this is rare.  So it is not a human weakness to demand more evidence when something contradicts previously held beliefs. It is in fact the only rational way of making inferences.

Comments