September 17, 2004

Bayesian votes, classical deaths

In the comments, a reader asks:

what is it about americans that make them think a terrorist attack is so imminent? (i like to think that they're just stupid, but that would be
politically incorrect or something)

Indeed, it puts me in mind of the old George Carlin quip "Think of how stupid the average person is, and realize half of them are stupider than that."

But more seriously, George Stigler pointed out once "that which is regular, is not stupid." In other words, when many people behave in the same way it's worth taking seriously.

What appears to be "stupid" behavior can often be attributed to bounded rationality, incomplete information, cultural norms, and the like. In other words, people generally act rationally, but within a framework we don't necessarily understand.

On this specific issue, "why do people think a terrorist attack is imminent?" it's likely that we have a very widespread example of people over-estimating the probability of infrequent events, especially when the outcome is pretty horrible.

So, it's strange indeed that the presidential election is not turning on who will be toughest on firearms and road accidents, since these are both areas in which neither Bush nor Kerry are proposing much substantive to deliver a "safer America."

Our votes may be Bayesian, but our deaths will largely be classical ...

Posted by robe0419 at September 17, 2004 7:49 PM